THE JOYFUL REDUCTION OF UNCERTAINTY

Open Thread 113.5

This is the twice-weekly hidden open thread. Due to the recent schedule change, this thread is not culture-war-free, but please remember the three day moratorium on politicizing tragedies. You can also talk at the SSC subreddit or the SSC Discord server.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

Leave a Reply

965 Responses to Open Thread 113.5

  1. skef says:

    This seems like a reinaugural-Sunday-hidden-CW-thread-appropriate subject:

    One of the things you often hear about evangelical support for Trump is that it is focused on Supreme Court nominations. There has just been a recent one. That political segment cares most about abortion, or at least claims that it does.

    For many decades the Republican coalition has been an awkward mix of social conservatism and tax-cut proponents (frequently mislabeled “fiscal conservatives”). By and large the tax cut side has done well out of this deal (in comparison to what would have happened if the other party were in power, at least) and the social conservative side has done poorly. It could be argued that the gradual ramping-up of anti-left rhetoric over the years was in part an attempt to paper over this reality. It could also be argued that Trump represents the close proximity to the limit of this arrangement. The Republicans have continued on the taxcuts-over-all path under Trump, but the whole message is in disarray. Hence, for example, the rash of retirements.

    In the midst of all this, the conventional wisdom on this board (I’m not going to scrounge for links, people who disagree can just say so) is that Roe is quite likely to remain enough intact so as not to much change the political landscape.

    So:

    1) How do people see this happening with Roe? Is the theory that Roberts, a Catholic, will “do an Obamacare” in an attempt to keep the court out of things? What is he going to say at dinner for the next twenty years?

    Or is the idea that there is some implicit or explicit back-room deal in place? Maybe the fiscal folks know that Roe going away would be bad politically (more on this below), and one or more appointees knows to keep it in place on as technical grounds as possible. On this view: Do you really think Trump is organized enough to ensure this? That he would ensure his appointees conformed to long-term Republican political needs? Do you think he cares at all about abortion law at this point?

    2) Say that Roe is overturned. Assuming the justices don’t discover a right to life, that will throw the issue firmly and loudly to congress and the state legislatures. And although it’s sort of been there for a while in “chipping away” mode, they haven’t been the psychological focus.

    So how aren’t we sliding towards a new political reality in which an issue that many women care about more than they often care to say, and many men care about less than they tend to say, is suddenly a primary electoral driver for some number of years? Note that I’m not making any claims about abortion availability; I personally expect that that wouldn’t change much. I’m saying that many women will now go to the polls thinking “whether I can get an abortion partly depends on my vote” in a way they have not been thinking that, or at least not in election-driving numbers. In different words, if Roe falls couldn’t that have at least as profound an effect as the Johnson “there goes the South” civil-rights era?

    tl;dr: I think people are underplaying the significance of U.S. abortion politics.

    • Furslid says:

      I think abortion has been a main driver in US politics. Abortion being legal has been keeping some of the Republican voters in the fold. I’m thinking of my parents here. They are Catholic, and feel that they have to vote Republican because they have to oppose abortion. If abortion wasn’t an issue, they could very well vote democrat for other reasons. This could break the Republican party.

      Abortion being made illegal would probably not last. Any abortion law that was acceptable to the general public would have exceptions for the health of the mother. This could be a formality where anyone can get an abortion, and getting a health of the mother exception is as easy as getting a medical marijuana card is in CA or as easy as Scott says getting a prescription for certain psych meds is. If it isn’t, then there will be cases where a doctor wants to give permission for an abortion, but can’t. Even a normal pregnancy isn’t risk free. Within a short time we would have cases where a woman couldn’t get an abortion and had major adverse health effects of her pregnancy as national news. I don’t think any abortion ban could survive that. An exception for rape could be even worse.

      • skef says:

        Abortion being made illegal would probably not last.

        As I alluded to, I think the political implications of tossing Roe are much more substantial than the abortion-availability implications.

        I’m not sure I follow what you say about your parents, though. Presumably if the Republicans continue to be the anti-abortion party, it’s all the more important for voters who oppose abortion to vote Republican — because the place it stops is in the state legislatures, with possible pressure (either way) from the federal level.

        If you look at the political numbers they’re fairly close, which could argue for a wash. But there is a gender bias, and I think women who think they might need an abortion are far more motivated on this question. (I also think there is some hypocrisy built into those numbers based on how long the issue has been off the table, but trying to argue for that would be a lot of trouble with questionable payoff.)

        • SamChevre says:

          Women in my observation are more extreme on abortion than men, on average–not more in favor. The core of the pro-life movement seems to me to be women, and the core of the pro-choice movement certainly is.

          • AG says:

            This seems to imply that there’s room for Democrats to make up the numbers of women lost through a gain in men. As Deiseach point out below about their refusal to allow moderates on the issue within their party, they could tradeoff with allowing gun moderate Democrats, which they currently still do.

      • skef says:

        Any abortion law that was acceptable to the general public would have exceptions for the health of the mother. This could be a formality where anyone can get an abortion, and getting a health of the mother exception is as easy as getting a medical marijuana card is in CA or as easy as Scott says getting a prescription for certain psych meds is.

        Right, so let’s consider this from a political angle. Abortion is legal now due to Roe. If Roe falls, states will have a variety of laws on their books. Maybe some state-level politicians will be able to get away with saying “Hey, it was already here! What can we do? Gridlock!” but I doubt that will give much cover.

        So now the party controlling the legislature gets to decide on abortion availability. In most places now, a Republican primary candidate who does not oppose abortion cannot win. But more importantly, the issue is so polarized that even the rape and health-of-the-mother exceptions are either (more rarely) off the table or (less rarely) candidates hedge and try not to answer.

        So one question is how fast that can change. Because if it doesn’t change then many, many Republican candidates in the general election can be cast in the light of these very motivating and unpopular views. After decades of all the rhetoric being irrelevant, this stuff will be on the table. My sense is that most women want to decide whether or not to bring a rapist’s baby to term. And that choice is salient before they’ve been raped, and therefore salient to most women.

        • MrApophenia says:

          Also, don’t forget that a number of states have already passed full-fledged abortion bans, set to take effect of and when Roe is overturned. Others never repealed their old bans, and they would immediately resume.

          Abortion access disappears instantly for a huge portion of the country immediately upon Roe being overturned, with no further action needed.

          • skef says:

            See my first paragraph.

            As I said, I doubt that the electorate in those states will see the issue as suddenly settled as a political issue, as much as abortion foes may try to steer things that way.

          • albatross11 says:

            I think skef’s take on this is probably right: If Roe is overturned and abortion becomes a live political issue at the state level, I expect that favors Democrats over Republicans. One reason is that when actual abortion bans are on the table (rather than symbolic abortion bans that everyone knows the courts will overturn), that will probably motivate more people who want abortion legal to come to the polls in many states. Another reason is that when abortion is already banned locally (in places where the pro-life voters have a commanding majority) and it’s no longer a federal issue, the Republicans will lose some of their hold on a subset of voters for whom abortion is the most important issue.

          • Abortion access disappears instantly for a huge portion of the country immediately upon Roe being overturned, with no further action needed.

            Not as long as abortion is fully legal in other states not too far away.

          • Hoopyfreud says:

            @DavidFriedman

            “Not too far” is relative. An abortion may well be inaccessible for people if getting one takes more than a day in another state – accommodations and transportation can be quite expensive. Moreso if those nearby states have mandatory waiting periods.

            This isn’t to say that access to abortions should be free, but that there are quite reasonable objections to a state banning, say, the practice of oncology that also apply here (from an accessibility standpoint, not a moral one).

          • @Hoopyfreud:

            There are surely objections. But I was responding to the claim that abortion would disappear instantly for those people, which is a very considerable exaggeration.

          • Hoopyfreud says:

            @David

            I was responding to your “not too far away” rider more than anything else; there will surely be people – many of them – for whom “a nearby state” is too far away. If the Southeast banned abortion in a bloc and none defected, I assume you’ll agree that abortion would become “inaccessible” to many Floridians. To those who cannot afford a long trip to get an abortion, this has nearly the same effect whether it’s Mississippi to Pennsylvania or literally just Florida.

      • Deiseach says:

        If abortion wasn’t an issue, they could very well vote democrat for other reasons.

        As an outsider looking in, I think that’s the interesting part of American politics. There definitely was an exchange of voters between the two parties over various platforms, if the “Reagan Democrats” moved rightward in the 80s I do think abortion was part of it and I think a lot of traditional Democrats or would-be Democrats might move back if abortion as a sacred value hadn’t been made such a central plank of party policy.

        For instance, there’s this story about a Democrat going for governor in South Dakota:

        But on a sunny Saturday, shaking hands and nudging his wheelchair up the parade route, came Billie Sutton, a 34-year-old state senator and onetime rodeo rider who is making a surprisingly competitive run for governor against South Dakota’s four-term Republican congresswoman, Kristi Noem.

        Mr. Sutton is running as an anti-abortion conservative Democrat with cowboy cred and a stirring life story. His supporters think he can show Democrats how to start rebuilding the party in socially conservative states where the ag-heavy economy rises and falls with rain cycles and soybean prices.

        …To South Dakota’s handful of liberals, it sometimes feels like he is running as an anti-Democrat. He is pro-gun, anti-abortion and says he opposes a state income tax. He has a Republican running mate. He deflects when asked about social issues that divide Democrats from most South Dakotans, like whether he would have voted to confirm Justice Brett M. Kavanaugh to the Supreme Court.

        And this other story (see the headline: “Is it Possible to be an Anti-Abortion Democrat?”) about a member of the Democratic Party in Missouri trying to get pro-life wording into the party platform in 2016 after heavy defeats in state elections (along the lines of language used in the past by the Democrat party), hoping to appeal to conservative Democrats and maybe some Republicans who would be wavering, and getting first tentative approval and then definite pushback:

        Ms. Barry thought her plank might help the party reclaim some districts that seemed hopelessly lost to Republicans.

        She worried that the Democratic Party had moved too far left on abortion. Gone were the days when the party, under President Bill Clinton, called for abortion to be “safe, legal and rare.” She also noticed fellow Democrats showing contempt for her when they learned her stance on abortion.

        She recalled a cocktail party conversation with a woman who asked why she was not seeking help from Naral during her run for State Senate in 2008.

        “I said, ‘Oh boy, you know I don’t think that would work,’” Ms. Barry replied. When she explained that she opposed abortion, the woman “looked at me like I had the plague. She had this horrible look on her face of just disgust and she walked away from me.”

        On June 30, when dozens of Democratic State Committee members gathered in a university conference room in Jefferson City to vote on the new platform, Ms. Barry nervously introduced her plank. It said that the party recognized “the diversity of views” on abortion and “we welcome into our ranks all Missourians who may hold different positions on this issue.”

        To her surprise, it passed.

        Shortly after the vote, Pamela Merritt’s cellphone started pinging with texts. An abortion rights activist who was a member of the platform committee, Ms. Merritt had not attended the vote, but started hearing about it almost immediately from angry friends demanding an explanation.

        “My stomach dropped,” said Ms. Merritt, who had agreed to join the committee after the party’s steep losses of 2016, thinking she needed to do more than criticize from the sidelines.

        In her view, Missouri Democrats needed more progressive politics, not less.

        “I don’t understand Democrats who quote Truman and F.D.R. and then act like they are terrified to run as an actual Democrat,” said Ms. Merritt, 45, who lives in St. Louis. “You have to believe in something in order for somebody to believe in you. You can’t be such a watered-down thing.”

        The fight over abortion in the party, she said, epitomized that. So she sprang into action, talking on Facebook and Twitter with hundreds of angry progressives, some of whom were threatening to stop their donations, calling her fellow committee members, and ultimately the party’s chairman.

        “I felt horrified that someone would associate me with that bizarre, regressive anti-woman language,” she said.

        The party was trying to placate people who opposed abortion at the very moment that abortion was most under threat, Ms. Merritt said. Days before the vote, Supreme Court Justice Anthony Kennedy had announced his retirement and the court had backed anti-abortion pregnancy centers. Missouri, one of the most restrictive states in the country, is now down to one abortion clinic.

        “The last thing we needed was for that language to linger,” she said of the plank. “It was a foul stench that needed to be addressed sooner rather than later.”

        So that looks like that for the immediate future as far as getting the support of Planned Parenthood and NARAL for Democrat candidates.

        • Deiseach says:

          And while I’m banging the drum, this GetReligion piece on “Blue Dog” Democrats in the Tennessee Senate race, have they any chance at all in the modern party, what’s so special about East Tennessee, and Al Gore’s evolution of attitudes on abortion once he moved out of Tennessee onto the national stage:

          Here’s the top of a 1999 Weekly Standard piece on how Gore moved left — when he prepared to leave Tennessee.

          Guess which presidential candidate wrote the following: “I have consistently opposed federal funding of abortions. In my opinion, it is wrong to spend federal funds for what is arguably the taking of a human life. It is my deep personal conviction that abortion is wrong. I hope that some day we will see a drop in the outrageously large numbers of abortions which currently take place. . . . I share your belief that innocent human life must be protected, and I am committed to furthering this goal.” …

          Al Gore wrote those words in July 1987, on the eve of his first bid for the White House. … The record shows that he began his political career in the 1970s as a consistent pro-lifer; his conversion to pro-choice began with his 1984 Senate bid, and his enthusiasm for abortion rights has only intensified. The simplest explanation for the switch: Ambition trumped principle.

          Oh, I think that’s a bit tough on Al; when the electorate ate up “I’m pro-life”, he was all pro-life; when a different electorate preferred “personally opposed but”, that was how he went, and when it was “abortion is a right”, there he went as well. Just like the vast majority of other politicians. Ambition is always there, principles are always negotiable.

    • Thomas Jørgensen says:

      Most western nations legalized abortion via the legislature – Decades ago.

      I sincerely doubt the proposition that the US as a whole is significantly more conservative on this than 1977 Europe.

      There is a big coalition that, as long as Roe is in effect, can grandstand about opposing abortion without having to contend with any of the horror stories that would result from abortion actually being illegal. So, yes, actually repealing Roe might blow up in the face of the right.
      Or, maybe, if the result is patch-work enough, they can get away with it – no horror stories for the press to print if every state that outlaws it just ends up with abortion clinics 5 meters past the state line, nothing to see here, not our problem. (Ireland kept abortion restricted until very recently, in large part because, well, overwhelmingly anyone that needed one just went to the UK)

      • johan_larson says:

        If Roe fell, I think the big change would be the treatment of second-semester abortion. Lots of states would prohibit it or add various restrictions. My impression (though it has been a while since I looked into this) is that first-semester elective abortions would continue to be legal in most places in the country.

        • Deiseach says:

          If Roe fell, I think the big change would be the treatment of second-semester abortion.

          My very cynical take on that (thanks to the Irish abortion campaign* which Thomas Jørgensen mentioned) is that the various activist groups would just find a few photogenic and appealing women with sob-stories about how they totally absolutely needed an abortion in the second trimester and if these repressive restrictive laws are enforced they will die, they will! Plus a couple of Real Doctors to explain why it would be absolutely totally necessary to carry out these very few, very rare, very scarce procedures after any limitation date put into law. (How if these are going to be very few etc. cases, simultaneously thousands of women will die if they don’t get them, nobody bothers to explain but it’s going for convincing the public via emotional appeal not logic).

          It’s how they swung third-trimester abortion after all.

          *We’ve only just had the referendum passed, the goddamn legislation isn’t even written yet, and we’ve already had the activist groups marching about how the limits are too restrictive.

      • HeelBearCub says:

        I sincerely doubt the proposition that the US as a whole is significantly more conservative on this than 1977 Europe.

        Regardless of the the population being more conservative, the set of state legislatures is much more conservative, currently. If Roe were to fall as a SCOTUS ruling, you would see bills making abortion illegal start to circulate in nearly every Republican controlled legislature, essentially immediately. I would say 10 to 20 states would pass those laws.

        There are currently multiple state laws on the books, passed within the last few years, that criminalize abortion past 3 months. Theoretically they are not legal under Roe and Casey. Many states have almost no providers of abortion at all, as legislatures and governors make providers jump through more and more hoops (my wife’s office stopped providing them).

        People who are confident that the state legislatures aren’t willing to end legalized abortion are not paying much attention to recent trends.

        • Thomas Jørgensen says:

          I do not doubt that, I just think it carries a significant chance of resulting in those legislatures loosing the next election really, really badly. Currently, being anti-abortion is in large part zero-consequence virtue signalling.

          If the courts get out of the way, then yes, the legislatures will probably carry through on their very public decades long stances. Then get utterly bloody massacred in the next election.

          Parallel: Gun control. Gun control is also mostly set by the courts, and that allows democrats to grandstand without having to send the cops out to peoples houses to confiscate guns or any of the other things logically required by the rhetorical stances.
          Do you think it would work out well for the democrats, electorally, if the courts went “You know what, the second amendment is about the national guard and only the national guard” and they then actually implemented super-strict gun-laws?

          • HeelBearCub says:

            Currently, being anti-abortion is in large part zero-consequence virtue signalling.

            Again, I’m giving you evidence that this is not the case. Abortion providers are being driven from the field, leading to the absence of available providers in large geographic regions.

            The government of Kentucky is currently attempting to close the sole remaining provider in that state.

            How is that “zero-consequence virtue signalling”?

          • Thomas Jørgensen says:

            The attempt that got slapped down by the courts in incredibly harsh language this September? That attempt? Yes, it kind of is.

            Hilariously, the states attorney explicitly made the argument that trying to prevent abortions in Kentucky was meaningless symbolic politics given that there are abortion clinics in the neighboring states, map of travel distances and everything. Still got slapped down.

          • HeelBearCub says:

            So, only managing to narrow the availability of an abortion provider to a single location in a state that is fully 2.5 times the size of The Netherlands is zero consequence?

            I think you are entrenched in defending a hyperbolic position that doesn’t hold merit. It’s like pointing at activists who supported same sex marriage and saying “all they managed to do was strike down sodomy laws, they don’t really want gay marriage”.

          • sandoratthezoo says:

            @HeelBearCub

            I think there’s a coherent position that says that anti-abortion minorities can sneak in restrictions that may in fact be pretty effective precisely because “everyone knows that abortion is legal.” Like for a low-information voter, they say, “Look, I know that all this stuff is just people fussing around the edges of a settled issue, so I don’t need to worry about whether closing a particular provider is or is not legitimate stuff or sneaking anti-abortion.”

            If there are headlines on every paper saying, “YUP, THEY CAN TOTES BAN ABORTION,” then, in this view, people would be all like, “Wait, what center is being closed where?”

            Being pro-abortion is an ideological preference for the overwhelming majority of people. They do not after all, need abortions right now. They don’t keep track of where they’d have to go if they did need abortions. With Roe, they can be assured that they have ideologically won.

            So I think there’s a colorable argument that it would work out the way that Thomas is suggesting. I haven’t dug deepl

          • sandoratthezoo says:

            Hmm, sorry, misclicked and now can’t edit the previous comment for some reason. I cut off as I was about to say:

            “I haven’t dug deeply enough to figure out whether I agree with that argument, I just don’t think it’s obviously ridiculous.”

          • HeelBearCub says:

            @sandoratthezoo:
            That’s a reasonable-ish position, but I don’t think it matches the available evidence. The anti-abortion measures mostly aren’t being passed as small riders on must pass legislation (where the minority of a party can extract concessions). They are frequently stand alone bills that get nearly unilateral support from the Republican representatives.

            You can argue that anti-abortion conservatives represent only a rump of support in the Republican party, but that hardly matters, as the party has found that rump crucial to its electoral chances, and that rump has an outsized effect on the primaries (which determine not just the positions of the nominees but have an effect on the actual beliefs of the selected nominee as well).

            It’s a mistake to think that the anti-abortion right is not sincerely represented by the elected Republican representatives. Skef may be right that their electoral chances will dim if they manage to become “the dog that finally catches the car”, but in the interim they are quite likely to sincerely try to catch that car.

          • sandoratthezoo says:

            To be clear, I’m suggesting that the voters ignore restrictions on abortion now, because “they know abortion is legal,” not that legislators do.

      • albatross11 says:

        ISTR that most European nations have legal abortion, but with more restrictions than US law has thanks to Roe. For example, I think late-term abortions are illegal a lot of places.

      • Conrad Honcho says:

        I’m not sure the “horror stories” are persuasive. Are you persuaded by the horror stories of legalized abortion?

      • Aging Loser says:

        It seems to me that many generally destructive Progressive “reforms” have been driven by horror-stories — for example, the dismantling of marriage was driven by consternation over rare cases in which men really did frequently hit their wives, and the dismantling of sex-based specialization was driven by consternation over rare cases in which women who really did long to achieve great things in the sciences, medicine, etc., and were really capable of doing so were miserable because they weren’t able to do these great things that they longed to do and were capable of.

        • HeelBearCub says:

          Poe in action?

          • Plumber says:

            @HeelBearCub

            “Poe in action?”

            If @Aging Loser isn’t sincere than the “act” has been for a long while.

          • Aapje says:

            @HeelBearCub

            You need to keep in mind that for much of history, very few men had the opportunity to achieve great things in the sciences, medicine, etc. So for most women the result of dismantling of sex-based specialization would not be a decent chance to achieve great things (either).

            This is also revealed in the support for suffrage and feminism, which was almost exclusively upper class. Those women were jealous of the lives of their male peers, but not the lower and (lower) middle class women.

            This only changed relatively recently, with many more people getting the opportunity to study. My observation is that very many people judge the past by treating it as if it was far more like today than it was.

          • Aging Loser says:

            Plumber,

            Yes, I do feel what I say. I don’t see what’s so weird about my attitude — it’s what pretty much everyone felt until World War One destroyed the world.

            A sad thing about the dismantling of marriage and sex-based specialization is that we’re now cut off from the ways of thinking and feeling that are presupposed by almost all of the great art and poetry that has ever been produced.

            Yes, there are female superheroes in the comicbook-like story-poems of Ariosto and Tasso — but these are superheroes, and those story-poems seem to me to have been written for a teenage audience (Tasso’s superhero Rinaldo is 18 years old).

            I think that a commenter here — but maybe it was elsewhere online — wrote a long time ago that “you can’t put toothpaste back into the tube.” This seems to me to be a wise and true observation. But I wish some sorrow were regularly expressed over what we’ve lost.

            Aapje,

            It is also the case today, and will always be the case, that very few men or women long to do or are capable of doing great things. Which is what makes the spectacle of both young men and young women who would have drawn some satisfaction from the performance of traditional social roles wasting their youthful years sitting in classrooms looking at their phones so sad.

            (For the record — Deiseach is my favorite online writer.)

          • My observation is that very many people judge the past by treating it as if it was far more like today than it was.

            And the future as if it will be far more like today than is likely.

            I still remember an ex-cabinet member commencement speaker talking about putting the key in a car’s ignition and turning it, and nothing happening because the world had run out of oil.

            And people worry about the effect of our nuclear waste ten thousand years from now.

        • Hoopyfreud says:

          “Rare” is vague enough that it could mean “less-than-universal,” so I’ll agree and note that I think that the tradeoff is worthwhile.

      • gdanning says:

        There was a study out several years ago that stated that, at the time, the US had the most liberal abortion law in the world, outside places like China that encouraged it as population control. And FWIW, this article claims that the US is currently only one of seven countries in the world which allows abortion at all after 20 weeks. https://www.irishtimes.com/opinion/abortion-in-the-us-actually-its-laws-are-among-the-world-s-most-permissive-1.3459718 So, the end of Roe might well amount to a substantial restriction in the abortion availability in the US, if legislatures converge on the global norm.

      • Most western nations legalized abortion via the legislature – Decades ago.

        All the remaining [European] states make abortion legal on request or for social and economic reasons during the first trimester. When it comes to later-term abortions, there are very few with laws as liberal as those of the United States.

        It sounds as though most of Europe has more restrictive abortion law than the U.S. at present, not less, and as though, absent Roe, some U.S. states would be more restrictive than most European states, some less.

        • Eugene Dawn says:

          You also have to compare restrictions on first-term abortions: waiting periods and parental consent and stuff like that. My impression is that Europe (or at least France, which is the usual comparison for some reason) trades off having essentially no restrictions on first time abortions with stricter restrictions on later-term ones. I’m not sure it’s obvious how to measure what is more restrictive: a number of less onerous restrictions all the way through, or complete permissiveness right up to the point of a total ban.

          I have also heard, though don’t know how widely in Europe this applies, that some European states make sure abortion is funded by their healthcare services and that there are abortion providers available in otherwise underserved areas, making it easier for people to obtain abortions in practice, something not captured in the laws or regulations.

          • HeelBearCub says:

            I would think all the EU comparitors have universal healthcare, and abortion is just another service. EU countries tend to make going to a care provider easier in other ways as well (I don’t think you will get fired because you needed to go to the doctor, which occurs not infrequently in the US. You will get paid time. etc.).

          • brmic says:

            some European states make sure abortion is funded by their healthcare services and that there are abortion providers available in otherwise underserved areas,

            That’s a huge factor. Restrictions on 2nd trimester are explictly defended by pointing to the availability of free and local first trimester care. Except for rare exceptions, every woman that wants one can get an abortion during the first trimester, so in turn it’s acceptable for society to then turn around and say, sorry, you had your opportunity to abort and didn’t take it, now you’re stuck with that decision.

          • Douglas Knight says:

            My impression is that Europe (or at least France, which is the usual comparison for some reason) trades off having essentially no restrictions on first time abortions with stricter restrictions on later-term ones.

            France is the usual comparison because it’s an outlier. Half of EU countries have waiting periods for first term abortions, including Germany.

          • Aapje says:

            Dutch law is:

            – allowed up to 24 weeks (due to viability), although doctors commonly limit it to 22 weeks, due to difficulty in guessing the age

            – 5 day waiting period after a doctor has determined the pregnancy

            – mandatory talk with a doctor who has to determine whether the decision was made without duress and after careful consideration

            – there has to be an ’emergency situation,’ which is not defined by law. So it depends on the interpretation of the doctor. In practice there are doctors who accept any reason, including not liking the gender. A woman can also simply give a more acceptable reason, regardless of the actual reason. So this doesn’t seem to be a hurdle for a determined person.

            – all insurers have to pay for it

            – abortions are allowed for a young girl without permission by her parents if she is deemed capable of making the choice. Then the parents don’t have to be notified.

            – Late abortions are not legal, but not prosecuted when certain conditions are met (this is similar to how euthanasia is ‘permitted’ in my country).

    • Deiseach says:

      I’m saying that many women will now go to the polls thinking “whether I can get an abortion partly depends on my vote” in a way they have not been thinking that, or at least not in election-driving numbers.

      I honestly don’t see a change there, given the “muh reproductive rights!” scare-mongering I’ve been seeing all over the place forever. Unless you think that Trump is going to announce in the morning “Right guys, getting serious on this one, gonna overturn abortion”, I see nothing new going on, apart from the usual: the Republicans giving a nod and a wink on abortion, the Democrats using this as one of the issues to scare their supporters to the polls, and neither side genuinely expecting any kind of change in the status quo.

      • HeelBearCub says:

        Do actually passed laws count as “a nod and a wink”? Do you know the US state legislative history on this issue over the last 20 years?

        • Deiseach says:

          HeelBearCub, if the proposal is that “Republicans campaign and get elected on promises to overthrow Roe vs Wade, Democrats squeal about Roe vs Wade getting overthrown (see the recent “don’t confirm Kavanaugh, he’ll overturn Roe vs Wade” stuff)”, then you tell me why is Roe vs Wade not overthrown in the past twenty years of all this promising by one side and fearmongering by the other?

          • HeelBearCub says:

            Because, Roe and Casey are rulings handed down by the Supreme Court, which has lifetime appointments and only nine members. What this has to do with what the legislatures will do if those rulings fall is unclear to me.

            As you noted, politicians are respondent to their constituents. If their constituents support the proposition that abortion is evil and should be eliminated, they will vote for that. There are currently 7 states that have but a single abortion provider. The history over the last 20+ years is extremely hostile to providers in certain states. There is actual legislative and executive activity that can be examined. This is not a hypothetical.

          • Eponymous says:

            then you tell me why is Roe vs Wade not overthrown in the past twenty years of all this promising by one side and fearmongering by the other?

            20? Try 40!

            The reason is simple: Reagan and Bush Sr. nominated three justices who voted to uphold Roe in 1992.

            So you’re right, it sure looks like they didn’t care enough about the issue to make sure of their nominees. Hard to imagine democrats nominating 3(!) justices who wouldn’t strike down the one ruling 40% of their base cares most about.

          • Eugene Dawn says:

            So you’re right, it sure looks like they didn’t care enough about the issue to make sure of their nominees.

            It’s important to recall that at least one of Reagan’s nominees, who likely would have voted against Roe, was voted down, and so he was forced to pick someone who was more of a compromise. So it’s not necessarily that he didn’t care, but that he didn’t have the political power to appoint his preferred judge.

          • Eponymous says:

            That is a good point. I’m not sufficiently familiar with the politics of the time to say whether someone more solid on abortion could have been chosen for any of the three spots. But it’s hard to imagine this really represented their best effort.

          • Eugene Dawn says:

            Bush probably not; Souter was picked at least in part because of intra-New Hampshire politics (Bush’s Chief of Staff was former NH governor), although some of the other part was that his lack of paper trail was seen as an asset post-Bork.

            Looking at the Republicans who voted against Bork, I think a big part of this is that some partisan sorting hadn’t finished yet: Bob Packwood was a Republican from Oregon who introduced the first abortion legalization bill in the early ’70s. Lowell Weicker of Connecticut was a notable critic of the religious right who wanted federal funding for abortions in the case of rape and incest. Arlen Spector of Pennsylvania claimed to personally oppose abortion, but “supported a woman’s right to choose” and he eventually crossed the aisle to become a Democrat.

            Reagan was constrained by the existence of people like that in his Republican caucus, and so I think it’s fair to say he made the best effort he could given the circumstances. Whether circumstances have changed enough in the last thirty years is to some extent a matter of judgment, but there were only two Republicans in the caucus who said abortion rights were important to them during the Kavanaugh confirmation, which suggests that anti-abortion presidents have more freedom to maneuver now than they did in Reagan’s day.

      • Tatterdemalion says:

        I don’t think “scare-mongering” is a fair or accurate characterisation of that piece, and I also share Scott’s distaste for demonstrating that people are stupid by having them use the word “muh”.

        Yes, people on the left have been warning for decades that we’re just one or two conservatives on the SCOTUS away from abortion bans becoming widespread. That’s because it’s been true all that time, but until now there have always been just enough justices to uphold abortion rights.

        If people have been warning for generations that if we let the dikes decay the sea will get in, and it never has, that doesn’t mean we don’t need the dikes any more.

    • Brad says:

      1) How do people see this happening with Roe? Is the theory that Roberts, a Catholic, will “do an Obamacare” in an attempt to keep the court out of things? What is he going to say at dinner for the next twenty years?

      My guess is that Roberts will do everything he can to give the states flexibility in limiting abortion without explicitly overruling Planned Parenthood v Casey. I agree that this is a much trickier thing to pull off if he can’t get anyone else to sign on and has to do it in a series of solo concurrences.

    • SamChevre says:

      One of the things you often hear about evangelical support for Trump is that it is focused on Supreme Court nominations. There has just been a recent one. That political segment cares most about abortion..

      I’d agree with the first half, but disagree with the second half.

      Yes, a lot of Evangelicals*, Catholics, and others care a great deal about abortion, and so about Roe–but that’s no longer the heart of the concerns with the Supreme Court. The heart is “how much will the courts serve as a weapon, and how much will they restrain the other branches, in the culture wars generally.” The courts have been a huge weapon for the secularist side in the culture wars; having them be basically neutral rather than wildly partisan, and restrain rather than enable the rest of the government, would be a huge win. If you want the core of the reason that Evangelicals fear the Supreme Court, you need to look at Obergefell (where the court threw out a policy which had strong historical precedent, that Evangelicals had mobilized to have passed and gotten passed by popular vote in most states) and Bob Jones (where the Court agreed that the government may require all non-profits to serve the government’s interests to maintain their tax exemption–which is theoretically general, but has only been used against conservatives). And then look back and note that when it was convenient for anti-conservative publishers to libel conservatives, the standards for punishable libel suddenly changed (Sullivan). When US-style bicameralism meant conservatives voters controlled some state legislatures that they wouldn’t under proportional representation, the court made up out of whole cloth (Baker vs Carr) a requirement that state senates must be one man one vote (like the US House of Representatives), rather than by political unit (like the US Senate, and most state senates until that point).

      The central concerns of Evangelical/Catholic voters are “religious liberty” (more usefully, “the right of free association as applied to religious groups”), the right to maintain their own institutions, and the ability to contest public policy without the courts continually re-writing the rules to eliminate their wins.

      *A wildly ambiguous term which I don’t prefer, largely because no one can agree who it includes.

      • dodrian says:

        I think SamChevre hits the nail on the head right here. Shortly before Kavanaugh’s confirmation, a conservative Christian satire site published an article Democrats Shudder At Idea Of Having To Legislate Through Congress Should Supreme Court Lean Right

        It’s satire, so of course it’s misrepresentative and hyperbole, but I think it shows the emotional reasons for the supreme court is currently such an important issue for the right.

      • Tatterdemalion says:

        I think you’re probably projecting your own concerns onto others here – I suspect that the fraction of American voters who have even heard of Baker v. Carr or New York Times Co. v. Sullivan is tiny, and fewer still will object to them (although I suppose there might be a tradition of conservative concern over the SCOTUS that traces back to them even if the roots have since been lost).

        Conversely, while I totally agree that a lot of conservative voters do know and care about Obergefell v. Hodges, they cared about the SCOTUS a lot even before then, so at most it can have amplified or illustrated their concerns.

        I think that Roe v. Wade really is pretty central here.

        • Conrad Honcho says:

          No, I think Sam is mostly right. They may not know the specific names of each case, but there is a broad complaint of “the left has used the courts to impose their agenda in ways they never could via legislation and we really want that to end.” Abortion is definitely one of these things, but Republican voters are broadly aware of this phenomenon.

          • albatross11 says:

            Actually, I suspect abortion is *not* one of those things nearly so much as the other stuff. Put it to a vote, and abortion probably ends up legal in the first trimester with some hoop-jumping required for later-term abortions, at least in most of the US. (In the rest of the US, abortions will be available with the added cost of a bus ticket and a room in the local Motel 6 for a day.)

          • Joseph Greenwood says:

            I also agree with Sam. This was definitely a sentiment in my conservative family growing up.

          • hls2003 says:

            Another way to look at it is comparing the number of evangelical pro-life protesters (some but not many) versus evangelical private religious school attendees (a lot). It may be hypocritical or a failing or what-have-you on the part of evangelicals, but while they deplore the concept of abortion, they get really riled up at the prospect of culture wars affecting how their kids get taught. It’s closer to home and there are more of them. In what I’ve seen, abortion is often seen as emblematic of leftward pressure tactics in the culture wars, rather than the sine qua non of the culture wars in and of itself.

          • Chevalier Mal Fet says:

            Chiming into add that my own perception and that of my righty friends is that Sam is correct. I wouldn’t say that Roe is unimportant, but it’s only a part of a much larger perception that left runs to the courts whenever it can’t win at the ballot box, and that getting the right Supreme Court judges is mostly a matter of defense.

          • AG says:

            In the same way that the Republicans have to look at their own Garland conduct to see why the Justice confirmation process is so horrid now, there’s also room to examine how the left was pushed to relying on the judiciary because they were denied avenues elsewhere. Or are you going to argue that Brown vs. Board wasn’t necessary?

      • Douglas Knight says:

        When US-style bicameralism meant conservatives voters controlled some state legislatures that they wouldn’t under proportional representation, the court made up out of whole cloth (Baker vs Carr) a requirement that state senates must be one man one vote (like the US House of Representatives), rather than by political unit (like the US Senate, and most state senates until that point).

        That was Reynolds v. Sims. Baker v. Carr was about whether the feds could enforce the blatant violation of the TN constitution.

        • SamChevre says:

          My mistake-you are correct: Baker v. Carr put state-level districting under Federal control and articulated the one-man one-vote standard, but it was Reynolds v. Sims that applied that test to State Senates.

          • Douglas Knight says:

            No, really, Baker v. Carr didn’t articulate such a standard. The TN constitution did. (Douglas’s concurrence did, though.)

      • skef says:

        I’d agree with the first half, but disagree with the second half.

        If you’re going to “disagree” using a quote, could you please not delete the part of a sentence that makes by implication a point very close to the one you make explicitly?

    • Walter says:

      I definitely agree that abortion is super significant. I dunno who you are talking to who doesn’t think so though. I thought everyone agrees that it is the main plank in both parties platforms.

      • Conrad Honcho says:

        It’s not like everyone’s a single-issue voter on abortion, though. I mean, I care, but abortion probably doesn’t make my top 10 issues.

      • skef says:

        I’m responding to posts in past threads here, and in particular what I take to be minimization of the significance to politics of voters taking the issue to have moved from the courts back to the legislatures (regardless of how blurry the actual line has been and will be).

    • Conrad Honcho says:

      I really don’t think they’ll find a “right to life” that constrains citizens. I could see them saying that government funding of abortions is unconstitutional because it deprives someone of life without due process of law. But murder isn’t illegal because it’s unconstitutional. It’s illegal because states have passed laws regulating the circumstances under which citizens may or may not kill each other.

      As for what Trump is doing…I don’t think Trump really cares about abortion. He wants judges that are going to read what the constitution says and apply that. So, he wants textualists, no “living document” stuff. However, I think the natural conclusion of that is an end to Casey/Roe because there is no right to an abortion in the plain language of the constitution. If you want to add one, fine, amend the constitution, and then those same texualists will say “yup, here’s the right to an abortion, right here in 28th amendment, ‘The right of citizens to seek and obtain medical care including abortions shall not be infringed.'”

      And yes, I do think some state legislatures will absolutely vote to ban abortion. Utah, Mississippi, sure. But then California will make them government funded and mandatory. Laboratories of democracy and all.

      • RalMirrorAd says:

        What was the rationale behind ‘right to an abortion’ anyway? It seems so ad-hoc to me.

        Certainly there’s no right to control one’s own body if the selective service means anything. Also there’s the ‘we can do anything’ [commerce] clause.

        • Conrad Honcho says:

          Other people can explain it better than I can. It was something about the right to an abortion emanating from the penumbra that the 4th and 14th amendments imply a right to privacy. Right to be secure in your person -> right to privacy -> government can’t regulate what you do with your abortion doctor. But I don’t think this makes much sense anyway since the government heavily regulates your doctors and what they can do with/to you. This particular SHALL NOT BE INFRINGED only seems to apply to abortions.

          • albatross11 says:

            Yeah, antisodomy laws were still constitutional for several decades after Roe, laws against illegal drug use and euthenasia are *still* constitutional.

          • Lillian says:

            More’s the pity, i do genuinely wish we had a serious right to privacy that was upheld across the board. Instead, the right to privacy is just an ad-hoc justification for the right to an abortion and little else.

          • meh says:

            1. Why doesn’t the same argument apply for assisted suicide?

            2. If a state did hypothetically decriminalize all murder, is there any section of the constitution that would effectively criminalize it at a federal level?

          • Evan Þ says:

            @meh:

            1) Who knows. One factor is probably that suicide has traditionally been against the law, and “assisted suicide” is considered murder. Note that in England, the victim’s consent still isn’t considered a general defense to assault (to the disconsolation of BDSM enthusiasts).

            2) Consider the Fourteenth Amendment: “No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law.” This currently provides the ground for civil suits against murderers for depriving the victim of his civil rights.

          • Lillian says:

            Note that in England, the victim’s consent still isn’t considered a general defense to assault (to the disconsolation of BDSM enthusiasts).

            Consent is considered a defence against a charge of assault provided that the assault in question happened in the context of a boxing or rugby match. It’s unclear to me under what principle the law protects two people beating each other up when they call it boxing, but not when they call it BDSM.

          • BBA says:

            In the US, boxing matches (and UFC, etc.) take place under the auspices of state athletic commissions. Since a governmental agency approves the rules and conditions of the bout, the pugilists are permitted to engage in what would otherwise be illegal acts. There’s one case I know of where a boxer using doctored gloves was convicted of assault for severely injuring his opponent.

            But in the UK, the British Boxing Board of Control is a private organization with no official status. The legal reasoning why their fights are allowed and others are forbidden looks like a handwave to me.

          • HeelBearCub says:

            @BBA:
            The boxing commissions seem like something of a non-sequitur to me. You can open a boxing (or other martial art) gym and all that is really needed is that people sign some waiver forms.

          • dick says:

            Consensual fist-fights are legal in some US states (e.g. WA) but not others (e.g. OR). In Washington you can even ask the police to referee.

          • Mark V Anderson says:

            @evan

            Consider the Fourteenth Amendment: “No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law.” This currently provides the ground for civil suits against murderers for depriving the victim of his civil rights.

            I know very little about the rationale for civil suits, but this makes no sense. The 14th Amendment says the state may not deprive anyone of life without due process of law. How does this give anyone the right to sue individuals? I am not personally constrained by the 14th Amendment.

          • Lillian says:

            @BBA: Last i checked, sparring matches not under the auspices of any athletic commission are completely legal both in the US and in Britain. As HBC says, two guys can just get in a ring at a gym and slug it out without needing to get the commission involved. Hell i’m pretty sure it’s legal to hold an informal boxing match without a gym or ring, and it’s definitely legal to play rugby and tackle football in public parks.

            As for the guy convicted of assault for wearing altered gloves, well of course he was. Billy Collins Jr. consented to fight Louis Resto under the understanding that they would both abide by the established rules of the ring. Resto voided that consent by maliciously breaking those rules and causing crippling injury Collins.

        • Hoopyfreud says:

          Aside from the penumbral emanations…

          The argument of Casey is, in part, that the state cannot have an interest in the survival of a non-viable fetus, since the mother is the sole agent on who the fetus’ survival can depend.

          This argument would presumably not hold if the state had a compelling interest in people giving birth (as opposed to fighting wars [the draft], not getting addicted to drugs [which I disagree with], or dying [euthanasia]). If the argument is accepted as valid, the response must be to argue that the state does have an interest in people birthing more babies, which is… not supported by society at large or the history of common law. It’s not an especially strong argument, but I think it holds together.

          • RalMirrorAd says:

            If that is indeed the logic used then it seems like the ability to extract fetuses and grow them in laboratories would invalidate the notion that the fetus is solely dependent on the mother.

            But again I ponder what effect this kind of reasoning would have on the rest of laws governing medicine.

          • Hoopyfreud says:

            @Ral

            I mean, there aren’t really any other cases where someone is “beyond the reach” of the state, except for things like cryogenics – and I already don’t think defrosting someone is a crime, except insofar as someone’s frozen corpse is property. Maybe conjoined siblings, but there’s a WHOLE lot written about them, and they interact with the law in very strange ways. Also maybe vegetables on life support, in which case their right to life is similarly abridged – it’s left to their family to decide when to kill their body. I think this one can still be charged as murder, though.

            E: also, yes, if the age of fetal viabiliy became the date of conception, under current doctrine, all states would be allowed to ban abortion.

          • AG says:

            If the ability to extract fetuses and grow them in laboratories became available, women wouldn’t need to resort to abortion anymore (provided that the process isn’t prohibitively expensive). No pro-choice activist anywhere is saying the such technologies are bad.

            Rather, the hypocrisy is on the side of pro-lifers who nonetheless use in-vitro and flush all the unselected fertilized embryos.

          • Rather, the hypocrisy is on the side of pro-lifers who nonetheless use in-vitro and flush all the unselected fertilized embryos

            .

            Which is presumably why they get called “pre-embryos.”

            My guess is that some pro-lifers avoid in vitro just the reason you suggest, but I don’t actually know.

          • If the ability to extract fetuses and grow them in laboratories became available, women wouldn’t need to resort to abortion anymore (provided that the process isn’t prohibitively expensive). No pro-choice activist anywhere is saying the such technologies are bad.

            Children can be born quite premature and survive. Suppose, for the sake of argument, that it works at six months. Would pro-lifers approve of a law requiring any woman in her third semester who wanted an abortion to instead have a caesarian, with the fetus transferred to a premie hospital? That’s the nearest real world equivalent at present.

          • LewisT says:

            @AG, @DavidFriedman

            Many Christians do avoid in vitro fertilization for that very reason.

          • Hoopyfreud says:

            @David

            This appears to have happened at least once in Ireland, but the risks associated with such a procedure are not small, either for the mother or for the baby. It looks like you gain a few percent on infant mortality and quite a bit on maternal morbidity, but currently c-sections are only really performed where there are complications anyway, so there are other factors at play.

            That said, premature babies only become likely to survive around 7 months, at which point the vast majority of abortions take place because of unsurvivable birth defects in the fetus. So, y’know, you don’t gain much from it.

      • Walter says:

        I agree that Trump doesn’t care about abortion, but I don’t think he cares about the judges being strict constitutionalists or whatever either. I think he is just holding up his end of the ‘We’ll vote for you if you help us stop the dems from slaughtering thousands of infants every year’ bargain that rep politicians have to sign with the ladies of the pro life movement. He needs them for re election, so he comes through with justices on request.

        • Jaskologist says:

          In the “Trump: Crazy, or Crazy Like a Fox?” debate, his stance on judges is what clinches the latter for me.

          Evangelicals were willing to hold their noses and vote Trump on the idea that what really matters now is control of the court, and however deficient his personal morality, Trump would deliver judges who would protect religious liberty and the unborn and Hillary wouldn’t.

          Trump probably didn’t care about judges much, but he knew that a big chunk of the voters he needed did, so he made that deal. During the campaign, he provided a list of judges he would nominate from and promised that anybody would be Federalist Society approved.

          Since that time he’ll tweet all kinds of craziness, but he has never gone off-message on judicial nominations. The process of announcing new Supreme Court picks has been accompanied with reality-tv style theatrics and fake-outs, but even in that, the theatrics have never come close to implying that he might betray the base. It’s always of the form “will it be Candidate A who you’ll like or Candidate B who you’ll also like? Wait, I just got a tip that Candidate C (you’ll like him) is flying to DC!”

          He made a deal and he’s kept his end of the deal.

          • meh says:

            During the campaign, he provided a list of judges he would nominate from

            Judge K was not on that list.

          • One person told me that Kavanaugh was not on the list the Federalist Society originally submitted to Trump, was added at his request. I don’t know if that is true, or how one would know if it is true–the person who told me that was not in a position to have inside information.

            A possible reason why it might be true is that the Federalist Society is a libertarian/conservative group, and although Kavanaugh looks pretty good from the conservative point of view he doesn’t look all that good from the libertarian. So I can believe that the Federalist Society would be willing to recommend him if asked but wouldn’t rate him high enough to propose him.

            But whether it’s true I don’t know.

          • HeelBearCub says:

            It’s absolutely true that Kavanaugh was not on the original list and was added after Trump took office.

            So, the Federalist society clearly did not originally propose him. Why 5 names were added, well that would depend on inside knowledge that people will dispute even if reported.

          • Brad says:

            When I was in law school during the GWB administration, the federalist society seemed pretty open to anyone not on the left. It was the nature of the beast at my particular school that there were more libertarians than conservatives, but they brought in speakers from both sides of the coalition.

          • So, the Federalist society clearly did not originally propose him.

            That isn’t clear to me, although it might be true. It sounds from that story as though there is a list of judges Trump is considering, not that the list contains all judges that the Federalist Society has proposed.

            If so, it’s possible that Kavanaugh was on the Federalist Society list, not on the subset that Trump had earlier published, added to that later.

            From the article:

            Kavanaugh was left off the original list of 21 in part because his jurisdiction is Washington, D.C., and Trump was elected as an outsider.

            That suggests that the reason Kavanaugh wasn’t on the earlier list was Trump’s preferences, not the Federalist Society’s.

          • Jaskologist says:

            To my point, though, the base did not see Kavanaugh as a betrayal; even the people who would have preferred e.g. Barrett, didn’t think Kavanaugh was a bad choice. The very broad consensus on the right was that K was still very much the kind of judge they wanted, and they weren’t just saying that because they were being unthinking Trump partisans.

          • That may have been true of conservatives. My impressions is that there was some negative response from libertarians, based on his views of a unitary executive and similar issues.

        • Baeraad says:

          My personal belief is that he doesn’t care about political bargaining either, at least not in this case. Feminists care about abortion rights. Therefore, attacking abortion rights annoys feminists. Therefore, Trump is totally going to do it.

          Which is admittedly the only reason to attack abortion rights that I’m actually somewhat sympathetic to, even if I don’t consider it nearly enough of one… :p

    • John Schilling says:

      1) How do people see this happening with Roe? Is the theory that Roberts, a Catholic, will “do an Obamacare” in an attempt to keep the court out of things? What is he going to say at dinner for the next twenty years?

      I don’t think that Justice Roberts’ religion needs to be brought into this. I would have hoped we were done with that sort of thing when we elected JFK and he turned out not to be the Pope’s brainwashed stooge after all, but, OK, here we are.

      About 40% of American Catholics believe that abortion is morally acceptable, and that’s probably higher among the sort of cosmopolitan upper-class white Catholics who go to Harvard. Add in the Catholics who believe in separation of Church and State and that criminal sanctions for abortion are properly rendered unto Casesar per something some guy said once upon a time. The odds that Roberts will actually believe himself morally obligated to do the Pope’s bidding in this manner, or that he will fear Hellfire and damnation if he doesn’t, is exceedingly small. And considering the general American tradition of not haranguing people over politics or religion at the dinner table, and he’s probably safe on that front as well.

      Honestly, he’d probably take a lot more heat from family and friends for upending forty years of judicial precedent than for failing to criminalize abortion.

      • Plumber says:

        Before Roe v. Wade Catholics were much more likely to be Democrats, now they’re no more likely (or less likely) than the general population, though they were also more likely to be union members and live in cities in the past as well so the change may be part of that.

      • skef says:

        I don’t equate and did not mean to associate Catholicism with control by the pope. My implication was that on a cultural level (“dinner”) being a 1) prominent Catholic who is 2) not against abortion is difficult in the U.S right now.

        A contemporary Catholic who votes his conscience to preserve a right to abortion is possible. It’s also unlikely.

        • John Schilling says:

          I was trying to cover both possible factors, and I don’t agree that “at a cultural level” being a prominent Catholic who is not against abortion is at all difficult. I know a number of Catholics who were strongly and openly pro-Hillary and still are prominently anti-Trump and get no apparent pushback over their supporting the pro-abortion candidate vs the appoint-conservative-justices candidate. It does not come up at the dinner table, even at the proverbial Thanksgiving dinner, which I have attended. And being a more prominent Catholic isn’t going to change that, because the extra attention Roberts gets from his prominence aren’t going to be family, friends, or his parish priest (they all knew him before he was prominent) but strangers and political partisans whose religious alignment is irrelevant.

          I think you are working from a flawed model of American Catholic culture here, or at least the subset applicable to inside-the-beltway legal circles. Justice Roberts may face being harangued over dinner if he makes the “wrong” ruling on abortion, but only if he goes to the wrong sort of restaurants and only because that’s a thing in the New Washington – and being a Protestant or Jew wouldn’t save him from that.

          • HeelBearCub says:

            “Catholic” is not nearly a significant enough descriptor. The US Catholic Church covers congregations that look like everything from (only slightly hyperbolically) Unitarians to the most fire and brimstone Baptists. If you are a conservative Catholic, you are against abortion, full stop. “Artificial” birth control, full stop. My grandfather was one of these (with a healthy dose of paranoid style conspiracy theories to boot). I have a conservative Catholic friend who referred to Sandra Fluke as “The Whore of Babylon”.

            I have no idea what kind of congregation John Roberts comes out of, but the idea that Catholics in the US don’t care about abortion is just plain wrong. You can be Catholic and not care too much about abortion, there are lots of them, but that is the kind of thing that gets people to call you a “Cafeteria Catholic”.

          • John Schilling says:

            I have no idea what kind of congregation John Roberts comes out of, but the idea that Catholics in the US don’t care about abortion is just plain wrong.

            The idea that Catholics can’t be trusted to faithfully exercise the duties of judicial office because they care so very much about abortion, is even wronger.

            And “Catholic” is doing absolutely nothing useful here. Protestants have approximately the same range of belief and commitment, and approximately all practicing US politicians are nominally either Catholics or Protestants. With a handful of Jews, particularly in the judiciary, but see above. Chief Justice Roberts has a long professional track record as a judge that anyone who is interested in can study at their leisure; tossing in an offhand reference to his religion is worse than useless in this context.

          • skef says:

            The idea that Catholics can’t be trusted to faithfully exercise the duties of judicial office because they care so very much about abortion, is even wronger.

            How many conservatives think Roe was decided correctly? Roe is cited as the paradigm case of judicial overreach. Your premise seems to be that saying Roberts is likely to vote to get rid of Roe besmirches Roberts. I and I think HeelBearCub are saying that every reason to vote that way and little reason not to. Liberals will hate him for doing so; there is no neutral standard he would violate such that we are insulting him by saying he is likely to do so.

          • John Schilling says:

            Your premise seems to be that saying Roberts is likely to vote to get rid of Roe besmirches Roberts.

            Saying Roberts is likely to get rid of Roe, because he is Catholic, besmirches Roberts. In this case, without cause or evidence. It’s pretty much the same damn thing as the all-too-common construction, “[X], a Jew, does [nefarious or suspicious thing Y]”.

          • skef says:

            It’s pretty much the same damn thing as the all-too-common construction, “[X], a Jew, does [nefarious or suspicious thing Y]”.

            Only if the act is nefarious or suspicious, John. In this case, both the church that Roberts belongs to and the party of the president that nominated Roberts take strong, official positions against abortion. The U.S is very polarized on this issue and Roberts is very much a fixture in one of the poles.

          • John Schilling says:

            The churches that every US politician belong to, or close enough as makes no difference, take strong official positions on abortion. Except for the ones that are decentralized enough that they can’t have official positions, but that’s a distinction without a difference since as you note in your very next post, “I also recognize that contemporary U.S. Catholics are polarized in much the way the U.S. population is generally”

            There are no atheists worth mentioning in US politics, nor Unitarians that I know of. Everyone who wants to be elected or appointed to high office, nominally signs on to one or another Catholic, Orthodox, Protestant, or Jewish denominations that at least nominally says that Abortion is Bad, Mmmkay?, and then in the United States turns a blind eye to the fact that individual members of that denomination then go off and believe whatever they want and abort whatever they want. Catholicism is no different, in any way that matters, and it does not matter that they’ve got some guy in Rome who makes very very official pronouncements for his American followers to obey or ignore as they please.

            The polarization you describe is real. The divide between Catholic and Protestant (and the odd Jew), is basically orthogonal to that polarization. And saying “This guy we have to be extra suspicious of, because he’s Catholic and you know how they are, it would be different if he were a decent Protestant sort”, I’m really not seeing the difference between that and plain old-fashioned bigotry.

          • brmic says:

            FWIW, I think the ‘a Catholic’ was an unimportant aside, but John Schilling is right that functionally it’s a smear. If the argument is supposed to be that his religion (of whatever denomination) keeps him from doing his job, that argument needs to be made on the basis of past job performance. If that argument can’t be made, his religion has nothing to with the matter and shouldn’t be brought into it.
            This is uncontroversial on the left when the subject is the ethnicity or religion of a criminal suspect in a case that does not involve ethnicity or religion.

          • skef says:

            John, would you also object if the topic were “evangelicals” and abortion, or would that not be a problem?

            Anyway, I’m fine with this. Plenty of slurs are expressed about me on a daily basis and my response is to be vaguely grumpy about some of the people who say them when I happen to hear about it. You can’t change the world. If what I said constitutes a slur in contemporary America, I said it and I am still now not particularly worried about having said it. Feel free to feel grumpy about me — you already had reasons. Or feel and do more than that and fulfill your destiny as a social justice warrior.

          • skef says:

            If the argument is supposed to be that his religion (of whatever denomination) keeps him from doing his job

            I am not approaching these questions from a conservative position. With respect to this view:

            A. Jurists should decide cases according to textual interpretation and (when applicable) common law precedent while keeping their own life experience out of the process as much as possible.

            Would you agree that this is, overall, a conservative view, frequently used to criticize liberal jurists? And that those jurists don’t particularly tend to object that the critic is mistaken and that they do keep their lives out of their decisions?

            I am not committed to that view. Stop trying to bash me over the head with it.

          • Paul Zrimsek says:

            A. Jurists should decide cases according to textual interpretation and (when applicable) common law precedent while keeping their own life experience out of the process as much as possible.

            Then you’ve solved your own riddle. That conservative view is what Roberts, a conservative, will say at dinner for the next 20 years.

          • skef says:

            “common law” modifies “precedent”. I don’t believe conservative jurists generally consider interpretive errors plus time to equal common law.

            +1 for attempted snappiness.

          • Paul Zrimsek says:

            The “keeping their own life experience out of the process” part is all Roberts would need at the dinner table; the rest of it gets hashed out at the office.

          • skef says:

            That’s easier when you don’t have to write down a summary of why you did what you did at the office that then gets published and widely reported on. Sure, you can say “I don’t want to talk about work”, but you already did.

          • Paul Zrimsek says:

            Then it just turns into the joke where they refer to jokes by number instead of telling them: all Roberts has to say at dinner is “587 U.S. 1149, at 173”. (If it were me in that situation I might hold up a printout of the meme with the picture of Freddie Mercury– DIDN’T EVEN LIKE FAT BOTTOMED GIRLS. STILL DID HIS JOB.– but Roberts doesn’t strike me as that kind of guy.)

          • skef says:

            “What I would like to know is how it wound up that after we finally got what appeared to be a majority of responsible justices, every woman in this country still has a right to have her baby murdered.”

            “587 U.S. 1149, at 173.”

            “Ha ha! You are the very soul of wit, sir.”

        • Plumber says:

          @skef

          “…A contemporary Catholic who votes his conscience to preserve a right to abortion is possible. It’s also unlikely”

          Both Nancy Pelosi and Paul Ryan are Catholics. 

           From Forming Consciences for Faithful Citizenship: A Call to Political Responsibility from the Catholic Bishops of the United States (edited for length, emphasis theirs): 

          “…we bishops do not intend to tell Catholics for whom or against whom to vote. Our purpose is to help Catholics form their consciences in accordance with God’s truth. We recognize that the responsibility to make choices in political life rests with each individual in light of a properly formed conscience, and that participation goes well beyond casting a vote in a particular election…

          ….The Church’s teaching is clear that a good end does not justify an immoral means. As we all seek to advance the common good-by defending the inviolable sanctity of human life from the moment of conception until natural death, by promoting religious freedom, by defending marriage, by feeding the hungry and housing the homeless, by welcoming the immigrant and protecting the environment-it is important to recognize that not all possible courses of action are morally acceptable. We have a responsibility to discern carefully which public policies are morally sound. Catholics may choose different ways to respond to compelling social problems, but we cannot differ on our moral obligation to help build a more just and peaceful world through morally acceptable means, so that the weak and vulnerable are protected and human rights and dignity are defended…

          ….The right to life implies and is linked to other human rights-to the basic goods that every human person needs to live and thrive. All the life issues are connected, for erosion of respect for the life of any individual or group in society necessarily diminishes respect for all life. The moral imperative to respond to the needs of our neighbors-basic needs such as food, shelter, health care, education, and meaningful work-is universally binding on our consciences and may be legitimately fulfilled by a variety of means…

          …When all candidates hold a position that promotes an intrinsically evil act, the conscientious voter faces a dilemma. The voter may decide to take the extraordinary step of not voting for any candidate or, after careful deliberation, may decide to vote for the candidate deemed less likely to advance such a morally flawed position and more likely to pursue other authentic human goods.

          In making these decisions, it is essential for Catholics to be guided by a well-formed conscience that recognizes that all issues do not carry the same moral weight and that the moral obligation to oppose policies promoting intrinsically evil acts has a special claim on our consciences and our actions. These decisions should take into account a candidate’s commitments, character, integrity, and ability to influence a given issue. In the end, this is a decision to be made by each Catholic guided by a conscience formed by Catholic moral teaching….

          ….In reference to solidarity, a special emphasis must be given to the Church’s preferential option for the poor. While the common good embraces all, those who are weak, vulnerable, and most in need deserve preferential concern. A basic moral test for any society is how it treats those who are most vulnerable. In a society marred by deepening disparities between rich and poor…”

          Yeah, I don’t see how you may follow all that advice and vote for any candidate who is in either the Democratic or Republican parties except under the “may decide to vote for the candidate deemed less likely to advance such a morally flawed position and more likely to pursue other authentic human goods” clause. 

          There is a very small political party that looks like it tries to be more for what the Bishops suggested but as far as I know no member of that party has been elected to any office.

          • skef says:

            Both Nancy Pelosi and Paul Ryan are Catholics.

            Nancy Pelosi is a political liberal. Paul Ryan is a libertarian Catholic who, as a libertarian, you might think would be an exceptional case but in fact supports overturning judicial abortion rights.

            I recognize that there are many liberal Catholics. I also recognize that contemporary U.S. Catholics are polarized in much the way the U.S. population is generally.

      • Don P. says:

        @John Schilling:

        Everyone who wants to be elected or appointed to high office, nominally signs on to one or another Catholic, Orthodox, Protestant, or Jewish denominations that at least nominally says that Abortion is Bad, Mmmkay?, and then in the United States turns a blind eye to the fact that individual members of that denomination then go off and believe whatever they want and abort whatever they want.

        It should be noted that this is not the position of Reform Judaism, per WP:

        More generally, the “Reform perspective on abortion can be described as follows: Abortion is an extremely difficult choice faced by a woman. In all circumstances, it should be her decision whether or not to terminate a pregnancy, backed up by those whom she trusts (physician, therapist, partner, etc.). This decision should not be taken lightly (abortion should never be used for birth control purposes) and can have life-long ramifications. However, any decision should be left up to the woman within whose body the fetus is growing.

        So they’re not taking it lightly, but being a Reform Jewish politician doesn’t entail the sort of look-the-other-way stance you describe.

    • John Schilling says:

      So how aren’t we sliding towards a new political reality in which an issue that many women care about more than they often care to say, and many men care about less than they tend to say, is suddenly a primary electoral driver for some number of years?

      If the Supremes do throw abortion back to the states, doesn’t that make it an electoral driver at the state level? That might not be a bad thing, given how crooked and dysfunctional state government can get and (not coincidentally) how little people seem to care.

      As HBC has pointed out numerous times already, there are states standing ready with abortion laws – some even technically on the books right now – to deal with this. And as far as I can tell, every one of them allows first-trimester abortion in the case of rape, incest, or danger to the mother’s health, or something along those lines. That points towards abortion following the path of “medical” marijuana use, where there are always going to be doctors willing to rubber-stamp “Alice needs an abortion because reasons” paperwork and few people are going to want to make a fuss over that. And of course half the states of the union will have outright legal elective abortion regardles (probably about the same ones that will have legal recreational marijuana in ten years).

      So, one possibility is that nobody does make a real fuss, abortion remains de facto legal except that in some states the pointless paperwork changes form. That probably doesn’t make for a major political issue once the new rules sink in.

      The other possibility is that some states do push back against doctors rubber-stamping abortion paperwork, in which case those rules and policies become a live political issue that women are likely to care about. But a local issue, because it would be in an arena the Supreme Court has hypothetically just carved out for local jurisdiction.

      • HeelBearCub says:

        And as far as I can tell, every one of them allows first-trimester abortion in the case of rape, incest, or danger to the mother’s health, or something along those lines.

        But you can’t evaluate these laws properly unless you look at them as relate to Roe and Casey. Those laws were written specifically to challenge Roe and Casey given a specific supreme court bench.

        In a world where the SCOTUS ruling on those laws indicates willingness to move even farther away from Roe, you will see even more restrictive laws forwarded. In a world where one of the liberal justices has to leave the court before Trump leaves office, moving the court even farther right, we are even more likely to see test laws passed.

        • John Schilling says:

          Lacking your psychic powers of mind-reading and precognition, I can only note that this isn’t what state legislators are actually doing or even specifically proposing, there is not even close to majority support for it in any state, and the optics of forcing women to die in childbirth would be horrible even by red-state standards. I don’t think I am going too far out on a limb by saying that, even with Roe/Casey off the table, Red State abortion laws will still include exceptions for e.g. medical necessity. Really, does anyone anywhere still pass abortion laws without that get-out-of-horrible-press-free clause?

          The enforcement of which, or lack thereof, may be a significant political issue if Roe & Casey go away.

          • HeelBearCub says:

            I think I misread your initial comment, that or I am not understanding the current one.

            Are you proposing that they may outlaw abortion, but allow it only in cases of “rape, incest or health/life of the mother”?

            If so, I agree that I don’t see a bill passing that outlaws abortion without those provisions. My confusion stems from your insertion of “first trimester”. I don’t see that as being particularly relevant to the previous exceptions.

            But you seemed to also be saying that the wouldn’t outlaw first trimester abortions.

          • John Schilling says:

            Sorry about the confusion. I meant to indicate that the worst plausible case is that the rape/incest/health exceptions themselves apply only in the first trimester, and that second trimester abortion is right out regardless of the reason. Which in practice means medically necessary late abortions go one state over, out of sight out of mind.

            If the practical effect of Roe’s fall is that half the states have laws that say women can only have abortions in the first trimester and only with a doctor’s note that is as easy to get as a CA medical-marijuana card, I expect there will be a vocal minority trying to make an issue out of both of those limitations but most women will have higher priorities.

            If some state tries to crack down on abortion-prescribing MDs the way the Feds (intermittently) do with opioids, then there would be problems.

          • HeelBearCub says:

            only with a doctor’s note that is as easy to get as a CA medical-marijuana card

            I think this part is highly unlikely to be anything like universal. The Republican party has been perfectly willing to do a great deal at the executive level to highly discourage abortion providers. As I said, my wife’s office no longer provides abortions already because of this.

          • ana53294 says:

            Really, does anyone anywhere still pass abortion laws without that get-out-of-horrible-press-free clause?

            El Salvador does. There was a case in the press a couple of years ago where there was a woman who had a foetus that would not survive longer than a couple of hours, and she would herself die in childbirth, because of a medical condition. They locked he up in a hospital. And although most Catholics supported allowing her to live, they thought that her case was an entry-way drug to allow more abortions, so they didn’t allow her to have the abortion.

            They also prosecute women for miscarriages.

            This is not to say that the US will become like that; I am just pointing out that such cases do actually exist.

      • skef says:

        If the Supremes do throw abortion back to the states, doesn’t that make it an electoral driver at the state level?

        For at least a couple years it makes it an electoral driver for whoever manages to make hay about it. Neither side in the electorate is particularly rational when it comes to the federal/state split, and there are plenty of levers to pull at the federal level.

    • hls2003 says:

      There are a lot of elements in response to your point (1). Count me as one that thinks outright overturning of Casey (eliminating the “right” to abortion) is very unlikely with the current Court, or indeed even with a Trump replacement of, e.g., Ginsburg.

      There are multiple mechanisms at play in that calculus. Let’s take the current Court composition. Gnashing of teeth aside, the replacement of Kennedy with Kavanaugh effectively only moved the deciding median Justice from Kennedy to Roberts (assuming arguendo that Kavanaugh is more textualist / conservative than Roberts). There is not that much ideological space in between those two, meaning the Court as a whole did not move that much. Moreover, Roberts’ “it’s a tax” maneuver in the Obamacare cases strongly suggests that as Chief Justice, he cares a great deal about the Court’s legitimacy and public perception. It is hard to make sense of his position there as anything other than trying to keep the Court’s head down in the midst of a political food fight. He has also tended to opine in favor of the status quo and avoiding the Court being the agent of major social upheaval. If you assume that both (ideologically) judicial restraint and stare decisis (as opposed to, e.g., very vigilant textualism) and (pragmatically) the Court’s political legitimacy are key values for Roberts, then it follows he would be inclined to leave Casey alone as an affirmation of precedents that have now stood for nearly half a century, and also inclined to avoid apoplexy by the liberal political wing, members of which are already threatening everything from Court-packing to nullification. Given that most people are very loss-averse, in practice this means more-or-less kowtowing to the left on the issue, both because their dismay would be more intense, and their dominance over cultural platforms means it would be much louder.

      (Incidentally, I think this is also the simple answer to your Question 2. Loss aversion means liberals will scream more about a perceived loss of their existing position than conservatives will scream about a failure to implement their improved position. That by itself, I think, without further imputations of motive of “what they secretly think” amongst men or women, would result in greater political activism by the left in the event of overruling Casey than would result from a status quo decision defeating conservatives).

      If Roberts wants to prevent Casey from being overruled – not nibbling at the edges, but as you postulate throwing the issue entirely back to the states – he is perfectly positioned to do so. First, grant of certiorari requires four votes. You would think this would empower the (presumptive) conservative four, but it does not. It empowers Roberts. Neither the conservative nor the liberal wings would dare vote to grant cert without an assurance from the swing vote that the outcome will be as preferred. The merest hint by Roberts that he will not join a conservative (or liberal) majority will cause that faction to refuse cert entirely – better to live to fight another day than to have your issue squarely confronted and squarely rejected. In addition to his enormous-but-crude leverage as (threatened or actual) fifth vote, because his is the CJ, Roberts can influence issues at the margins also by voting last and selectively, and by the (threatened or actual) assignment of opinions. If he wants to tinker at the margins on something, he can reserve the opinion for himself for a majority, or threaten to assign it to a firebrand to coax recalcitrant opposition on board to a more middle-of-the-road consensus opinion.

      Certainly some of this is mind-reading Roberts, but it seems plausible based on Roberts’ prior actions with regard to hotly contested political footballs.

      In addition, for what it’s worth, I think the proposition that Kavanaugh is more likely to join a conservative majority to overturn Casey is highly suspect. His jurisprudence does not read to me like a man on a conservative mission (whether the all-out war by the left has changed him – either embittering or cowing him – remains to be seen). He seems like a pretty middle-of-the-road conservative more in the Roberts mold than that of a Scalia or Thomas. He seems an establishmentarian, not a firebrand. If that is the case, then even a Ginsburg replacement would only move the median Justice to Kavanaugh, which might not be much different than present. And if our guess about Roberts’ motivations is right, Roberts could still probably avoid an outright overrule of Casey either by joining the majority and taking the opinion and watering it down (perhaps a concurrence with Kavanaugh), or by sweet-talking / horse-trading Kavanaugh. If Kavanaugh turns out to be a Scalia, and so do each of Alito and Gorsuch (Thomas would do it in a heartbeat I think, the others there’s less track record), and also Trump’s hypothetical replacement of Ginsburg or Breyer, then that’s the only way I could see it happening. That seems like an unlikely chain of events. And even then, remember the “Switch in Time” aspect, where even supposed firebrands can get cold feet if confronted with the prospect of an electoral wipeout followed by a political undoing of their decision. I think even a solidly textualist majority of five (excluding Roberts) would be cautious enough not to simply write “F___ you, Casey” on a piece of paper at the earliest opportunity.

      • skef says:

        If Roberts wants to prevent Casey from being overruled – not nibbling at the edges, but as you postulate throwing the issue entirely back to the states – he is perfectly positioned to do so. First, grant of certiorari requires four votes. You would think this would empower the (presumptive) conservative four, but it does not. It empowers Roberts.

        How, specifically, does this help if one of the conservative circuits tees up a new decision that tosses Roe? That is, how does this get finessed if to preserve the status quo a decision must be overturned?

        • hls2003 says:

          That is both very unlikely, and not at all a problem for Roberts. With few exceptions (e.g. the Ninth Circuit on almost anything to do with Trump’s executive orders), the circuit courts will not directly contravene then-existing Supreme Court precedent like Casey. They lack the authority to do so. Even if they go ahead and say “We decline to follow Casey, finding it outdated,” that tees it up perfectly for Roberts to vote with the liberals and write a narrow opinion grounded in the principle that circuit courts cannot ignore Supreme Court precedent.

    • I’m saying that many women will now go to the polls thinking “whether I can get an abortion partly depends on my vote”

      Are you talking about polls for state elections? The Supreme Court changes very slowly. If a conservative majority is enough to reverse Roe, which is possible although not what I expect, then electing Democrats in federal elections won’t change that for a long time, unless the Democrats are willing to expand the size of the court. If Trump gets to appoint one more justice, a very long time.

      In which case your point only applies to state regulation of abortion. And given how inexpensive transport is nowadays, having a particular state ban abortions only makes them a little less convenient to women in that state so long as other states not too far away still permit them.

      • skef says:

        On a practical level, since Roe abortion has mostly been in the judicial rather than legislative arena. And yet it’s been a huge political issue. I am arguing that it will remain a huge political issue but that the dynamics of the politics will change suddenly and significantly.

    • palimpsest says:

      I think the “SCOTUS won’t overturn Roe v Wade” take is overplayed. Maybe Roberts would rather duck the issue, but it takes 4 votes to grant cert so the other Republican members can force the question on him. Elite Republicans think Roe v Wade is bad, and additionally tend to think it is a major the cause of the politicization of the court. So even if he is concerned with the institutional reputation of the court I still think he will reverse. Obergefell is a different story. Despite Robert’s vociferous dissent in the case, gay marriage is an issue elite Republicans would really prefer to never to have to talk about again. I expect the lack of political enthusiasm on the issue will also hold at the court.

      Reversing Roe v. Wade may hurt Republicans in the short run. People are more motivated by losing than by winning. Also, purely anecdotally, it seems to me that Democrats are talking about abortion more, and Republicans less, than they used to, indicating that both sides estimate that the issue favors Democrats somewhat. Polls on the issue maybe lean towards Democrats but it depends hugely on the exact question asked so its hard to tell. In the long run if Democrats do indeed have the advantage, it’s more likely to play itself out in a leftward shift for Republicans on the issue, rather than some sort of permanent electoral advantage for Democrats.

      • Mr. Doolittle says:

        Also, purely anecdotally, it seems to me that Democrats are talking about abortion more, and Republicans less, than they used to, indicating that both sides estimate that the issue favors Democrats somewhat

        I have a longstanding theory that for most of the leadership of either party, they care very little about the wedge issues (CW stuff) personally, but hammer it when out of power in order to get their base going. Once in power, they don’t want to actually solve the issue for two reasons. One, they will lose the wedge issue for future elections. Two, they will deal with all the negative consequences of having made the change.

        Since the Democrats have no federal authority at the moment, my theory would posit that they would be hammering this issue as hard as they can, while the Republicans would be quietly downplaying it – which is what we see. When the Democrats are in power, they aren’t trying to pass constitutional amendments to enshrine abortion or whatever either.

        • Mr. Doolittle says:

          Incidentally, this is also why I think the filibuster has survived for as long as it has. When a party wins an election and has a small but significant (maybe 55/45) advantage in the Senate, their constituents are going to want to see action on the wedge issues that drove them to the polls.

          It’s much better for the party in power to be able to say “Sorry, we really want to do [major wedge issue], but we don’t have the votes!” It works as a rallying cry to help retain/gain votes in the next election, and saves them having to actually take action. Without the filibuster, that would not at all be a plausible explanation for why they didn’t follow through on a major plank of their ticket.

  2. gbear605 says:

    Effective altruism proposal: deep analysis of relatively obscure medical knowledge and reporting on it. For example, Gwern’s writings on vitamin D have gotten me and other people I know to start taking supplements, which should increase his readers’ number of QALY by a significant amount compared to the cost of writing the article about vitamin D. I haven’t done the calculations to compare it to other common EA tasks, such as donating to Against Malaria, but it seems likely to be comparable.

  3. Sniffnoy says:

    Hi benwave! I wanted to finally get back to that discussion about the foundations of leftism that I never got back to, if you might happen to be reading this. The other day I thought of a nice simple way to illustrate what I was saying, so…

    (Oh hm now that I actually pull that up I guess I should answer the questions you asked. I’ll do that at the bottom.)

    Let’s consider the following two 2-player games. In both, our two players will be presented with two options; let’s call them A and B. In game 1, the payoff for (A,A) is (3,3), for (A,B) is (0,5), for (B,A) is (5,0), and for (B,B) is (1,1). In game 2, the payoff for (A,A) is (3,3), for (A,B) is (2,2), for (B,A) is (2,2), and for (B,B) is (1,1).

    Let’s make some notes about these games. First, they’re both symmetric — interchanging the roles of the two players leaves the game unchanged. Secondly, they both have a unique Nash equlibirium, which, moreover, is composed of pure strategies. As such, in both cases, resulting Nash equilibrium lies on the diagonal; the symmetry forces this.

    But the equilibria of these two games are different. Game 1 is of course just the Prisoner’s Dilemma; the unique equilbrium is both players choosing B and both getting a payout of 1. In game 2, on the other hand, the two players always get the same payout — their interests truly are perfectly aligned — so the unique equilibrium is both players choosing A and both getting a payout of 3.

    The thing to note here is that this occurs even though both games satisfy the symmetry conditions above and both games have the same diagonal. This illustrates how your reasoning earlier was mistaken. Your reasoning was basically, well, the game is symmetric in all the players (an assumption I’ll grant for the sake of argument); so its equilibrium must like on the diagonal (correct, assuming that there is indeed a unique equilibrium (which we might also insist be pure, depending on how restrictive we are being with the term “diagonal”)); so we only need to look at the diagonal to determine what that equilibrium is. And of course it’s that last step that is wrong, the above two games providing a concrete example — even though we know the equilibrium must lie on the diagonal, just what it is is still affected by what goes on outside the diagonal.

    This is because, as I mentioned earlier, even if both players know that ultimately the equilibrium will be on the diagonal, it still remains the case that either player can threaten to play something different than the other players, and the other players have to account for this in their strategy. You have to look off the diagonal to determine where on the diagonal is safe.

    And so, as mentioned, you cannot validly infer, from the fact that a game is symmetric, that its players will automatically, with no mechanism of coordination or enforcement, “act as a class” simply out of self-interest. A player may expect some simple diagonal cooperation, but to play it safe, they have to account for the possibility that anyone may defect at any time — throwing things off of the diagonal. The off-diagonal cannot be ignored; “acting as a class” is not something that can be derived from a simple assumption of rationality — it is a substantial additional assumption.

    I’m hoping that example makes clearer now what I was saying earlier?

    Meanwhile returning to the question you asked…

    And now, considering this dynamic as a given backdrop, what you have been discussing in the previous posts is a level on top of that: coordinated action of some group to change this game – not secretive, as in a literal conspiracy, but also something more than individual-level responses to incentives. Do I have that right?

    Um, I guess so? I’m not sure I’d call it “changing the game” although I guess that is essentially what’s happening. Just… any mechanism of coordination. Anything that lets you escape Nash equilibrium.

    And what I’m hoping I’ve made clear above is, that yes, under generic conditions, you do need some such mechanism of coordination in order to get coordination in general. Without special conditions, it does not happen just due to uncoordinated following individual incentives; and importantly, this remains true even under your assumption of symmetry, because the off-diagonal matters.

    • benwave says:

      Well now, let me just file this down under pleasant surprises and things I never expected to be continued! I’m lucky I did check this open thread, I don’t always. Thank you for your continued participation.

      Let us suppose then, for the sake of argument, that all the workers and all the capitalists in the nation of Hypothetica are in various employment relationships with each other, representing a Nash equilibrium. Resources of capital and labour are deployed such that nobody can become better off by changing the wages they offer or the job offer they choose to accept. A conspiracy of all capitalists could conceivably increase their profit by all lowering their wages simultaneously, but it won’t happen without collective action because a defector who kept their wages higher would attract more workers and market share, thus more profit (the (0,5) payout). So far we are in agreement.

      In neighbouring country of Let’sSaylia, there are significant numbers of unemployed workers. Here, the reward to capitalists of lowering wages remains, but the reward of raising wages does not. Here, without any mechanism for coordination, the effect of the individual incentives of capitalists is to push wages always downwards. (In reality I guess you would hit another upwards pushing mechanism where reduced income resulted in worse productivity from workers due to hunger, poor health, etc. and wages would stabilise around that point, but I would not consider this an acceptable result, I would hope you would agree.)

      I think I disagree with your 2-person game model for the following reason – In Let’sSaylia capitalists play an analogue of your game 2 above where they each benefit but workers suffer in the process. To me this presents a clear situation where it takes no mechanism of coordination to produce this bad result for workers, and I would propose that there have been plenty of times and places in our history where this has in fact played out.

      [I won’t be able to provide any further replies for about eight hours, I am going to bed]

      • 10240 says:

        [I haven’t read the original discussion, if that matters.] In your Let’sSaylia, non-coordinating capitalists also have an incentive to scale up their businesses in order to increase their profits, which involves hiring more workers, until there is no significant unemployment (at least among workers who can actually do a job adequately). If a (labor) market is competitive both among suppliers (workers) and among buyers (capitalists), and there is a surplus at the current prices (wages), then prices decrease not until 0, but until the price where demand equals supply and the market is cleared.

        • benwave says:

          I agree that this is true, conditional on the market for whatever they’re selling being able to absorb more of it, and on the capitalists having sufficient extra capital to effectively employ additional labour power. (and I guess on there not being some other barrier to their expansion such as laws, access to raw materials/technology etc)

          Let’sSaylia is obviously an unstable situation, and it will be stabilised in some way, of which your suggestion is one. Other ways involve deaths, emigration, crime, and innovation. In most modern societies, welfare provides a damping effect which might itself be enough to stabilise the situation, or might slow the progress of one of the other ways.

          • 10240 says:

            I haven’t done calculations, but I’m pretty sure that even if the supply of capital and natural resources is fixed, the equilibrium is one with more labor-intensive and less capital- and resource-intensive production methods where the entire labor supply is utilized, rather than one with unemployment and near-zero wages (even with hypothetical workers who are capable of working without eating). Of course such a restructuring of the economy may take a while.

            But in a free and competitive labor market, the situation with high unemployment and above-market-clearing wages is an anomaly that only arises if market conditions (supply and demand for natural resources, capital, labor or products) change suddenly. Usually these conditions change gradually, which allows for the prevailing wages to stay approximately market-clearing, while any restructuring of production methods also happens gradually.

            In practice, even free labor markets are not perfect markets, but we don’t observe the “race to the bottom” situation either.

          • benwave says:

            I’m interested why you believe that the equilibrium should be that one.

            I also think it’s very important to consider off-equilibrium situations (crises, in Marxist parlance). As an example, I’ve seen fairly often charts showing that wages as a proportion of GDP or some such still haven’t recovered to levels before the 2008 market crash, although unemployment has dropped back to low levels.

          • 10240 says:

            At many companies, humans and machines (i.e. capital) can substitute each other to some extent. That is, if you have a given amount of capital, you can produce more goods if you employ more workers. As you have observed, as long as there is high unemployment, it’s cheap to hire more workers, and you can even decrease the wages of current workers. If, say, there is 10% unemployment, you hire 11% more people, you can sell more products, and if you spend the same amount on wages in total as before (or even a bit more), you make more profit.

            In other words, the demand curve for labor is usually not horizontal but it has a significant downward slope, so if a it’s profitable to hire given number of workers at present wages but there is unemployment, the equilibrium is employing more workers at slightly lower wages.

            Crises will indeed cause some wage drop temporarily. However, I’m not sure that the lower ratio of wages to GDP is a result of the crisis. Perhaps the amount of capital in the economy is slowly growing, and in some sense capital contributes more to production relative to labor than before (though the two are hard to separate), so it reaps a bigger share of the fruits.

      • Sniffnoy says:

        [I won’t be able to provide any further replies for about eight hours, I am going to bed]

        I’m kind of generally low on time these days, so, uh, this reply is coming considerably later than that. 😛

        Anyway, I think you’re missing what the fundamental point is that I’m trying to make. I’m not making a claim about what world we’re living in — I’m no economist! You think I know how to answer empirical questions? 😛 (Also, uh, 10240 already appears to have started an argument on that point. 😛 ) My point is about the more fundamental question of how we think about these things in the first place.

        Let’s go back to the previous discussion. I asked, paraphrasing, why do leftists posit coordinated action of capital against labor, but not of people named Sam against people named John? You responded, again paraphrasing, that it’s due to the symmetry that exists in the former situation but not the latter, and, heavily paraphrasing, that since the symmetry constrains the equilibrium to lie on the diagonal, it is rational to act in a superrational manner, resulting in class-action without any mechanism of coordination.

        My point is that this last part is not correct; superrationality is not, in fact, ordinary rationality, and you do have to look off the diagonal to determine the best play, even as symmetry constrains the equilibrium to the diagonal. That is to say, if it is the case that the incentives that exist lead to such automatic coordination, then that would be a contingent, empirical fact about the world we live in; it is not something that holds generically, it is not something that can be just deduced or inferred without looking.

        Saying “we happen to live in the game where this is true” is very different from saying “this is true of every game with certain features”. Like, if leftism is just making claims about the game we happen to live in, then, why, it’s basically just standard economics, just with some unusual empirical assumptions! But it’s not, is it? It really is something fundamentally different, isn’t it?

        Or to put it another way: If you agree that it is correct to analyze these two games in terms of their Nash equilibria, then it would seem to me that you have basically conceded the argument.

        This sort of thing is why I keep saying that leftism’s foundations are rotten. Even if they make a number of empirically true statements, the whole way of thinking, of relating these statements to one another and generalizing them to other situations, it all looks like a giant ball of confusions and bad thinking to me.

        So what’s the conclusion here? Because I’m pretty sure it’s not that leftism is actually just standard economics but with unusual empirical assumptions.

        • skef says:

          [After reading this I scanned back through the thread here and the more relevant parts of the linked thread and am responding based on that understanding.]

          The first most basic thing to say is that someone in the Marxist tradition is unlikely to go to game theory as a first step. That doesn’t argue against anything you’ve said, but it’s worth keeping in mind that isn’t committed to looking at these questions through that lens.

          There are two argumentative strategies that you’re raising and I only have a rough idea of how they are linked. One is a particular point about equilibria, the other is an attempt at negative argument by exhaustion: What could make the difference that would tend to make members sharing this property collude when other properties don’t tend to lead to collusion? I take it that the point about equilibria is meant to remove large swaths of potential explanations by modelling them as games and showing how they would fail as a class. [Is that about right?]

          In response to that approach, I will suggest a potential explanation that could support the (more theoretically tending) leftist view. Assume a game with:

          1) Imperfect information, such that players can only approximate each other’s motives. (This could be because of a lack of information, or simply the cognitive inability of the players to grasp the whole even with the information available.)

          2) The individual players have drastically different kinds of power available to them, in terms of the moves they can make.

          I don’t have strong intuitions about how to model such a game formally, and one can easily imagine coming up with different specific games that might go differently when modeled. But one thought that immediately occurs is that you would expect all players, strong and weak, to worry more about the motives and potential actions of the strong players, as those are the likely sources of greatest benefit and harm.

          This is itself a kind symmetry breaking and therefore alignment. And I think it is fair to say that the prototypical Marxist “classes” are primarily characterized in terms of just these kinds of power differences. A poor aristocrat may be an irony, but a poor capitalist in the Marxist sense is confused.

          It is an additional step to propose that wariness of a power differential will tend to push the more powerful players into alliances. Well, you don’t even need alliances, necessarily. It could be enough for individual strong players to tend to leave each other alone and deal with weaker ones. It will really depend on 1) what the powers are and 2) specifically what the players, strong or weak, can confidently know given the assumed limitations.

          So I guess the most basic question of joining your approach with this response is: what can game theory tell us about trends in this family of games?

          • A poor aristocrat may be an irony, but a poor capitalist in the Marxist sense is confused.

            I haven’t been following the thread closely, but this struck me as interesting in terms of how different people use words.

            Suppose someone lives entirely on the income from his capital but he doesn’t have very much capital and as a result is poorer than most workers. I would still describe him as a capitalist–a poor capitalist. Would a Marxist?

          • benwave says:

            David,

            I agree with you. I would describe that person as a capitalist. (Although I don’t think that person would contribute very much to the overall power relations acting between workers and capitalists.)

          • skef says:

            Suppose someone lives entirely on the income from his capital but he doesn’t have very much capital and as a result is poorer than most workers. I would still describe him as a capitalist–a poor capitalist. Would a Marxist?

            We would first need to clarify what “doesn’t have much capital” means.

            Someone could not have much capital in the sense that the investment income they earn without dipping into the capital is lower than most workers. That could still be quite a bit of capital. Assuming a 10% rate of return, with reinvestment to keep up with inflation, it could be enough to live on at the same standard for ten years without working. Even if the investment income is half the typical worker’s salary, the person could live like a standard worker for around five years without working.

            The former could be true regardless of how liquid the capital is. Perhaps the investor is stuck in this position because he cannot pull his money out. Still, he holds assets of substantial value.

            If “doesn’t have much capital” means the person’s assets are also meager as compared with a worker’s assets, it’s arguable whether the investor is capitalist. Marxists tend to be very explicit about how their principles arise from material conditions. A situation in which someone can reliably invest with a 60% return, for example, has different enough material conditions that they would call for a different analysis. And a person who coincidentally invests with a 60% return and lives off the investment may not be a true capitalist.

          • skef says:

            Anyway, the more challenging test case would be a worker who has a large savings and chooses not to invest it. So not a “poor capitalist” but a “rich worker”.

            But as long as there are few such workers there doesn’t seem to be an obvious issue. If most workers were rich, then you have different material conditions, people with large amounts of money would have less leverage, and there may not be any proper capitalists in such a system.

          • Sniffnoy says:

            The first most basic thing to say is that someone in the Marxist tradition is unlikely to go to game theory as a first step.

            Yeah… that’s the problem! As best I can tell, they don’t have any coherent mathematical model of such situations at all, just lots of vagueness. But why would you not want to use game theory? Even if you don’t agree with the use of Nash equilibria — and there certainly are good reasons to be skeptical of Nash equilbria! — just the simple description is clearly a good thing, regardless of what solution concept you want to use.

            I mean, I’ve ranted about this here before, but the short version is, healthy fields steal. If there is a generally useful tool and your whole field is ignoring it, that is a serious red flag that you are in fact part of a stagant program of research.

            There are two argumentative strategies that you’re raising and I only have a rough idea of how they are linked. One is a particular point about equilibria, the other is an attempt at negative argument by exhaustion: What could make the difference that would tend to make members sharing this property collude when other properties don’t tend to lead to collusion? I take it that the point about equilibria is meant to remove large swaths of potential explanations by modelling them as games and showing how they would fail as a class. [Is that about right?]

            The overall background point I’m trying to make is that leftist thinking is basically fuzzy and that when you try to formalize it you realize there’s nothing there. Of course, that doesn’t mean it’s impossible to come up with formal models that will replicate their conclusions. But such a steelman would be substantially different internally. It couldn’t replicate the internal structure of leftist thought because the internal structure is nonsense and conflations.

            And so of course I’m attempting to demonstrate that with an example, which is the whole notion of how they think about classes and acting-as-a-class. It’s all conflations, conflating individuals with groups. The uncharitable explanation for this is that, duh, people normally think in groups; you have to learn to think in terms of individuals. And people tend to think in particular sorts of groups because that’s the way we’re wired (that’s maybe not a good description but I’m trying to be brief).

            It’s this latter bit that led to my first point: Why capital against labor but not Sams against Johns? The real reason people come to the former conclusion but not the latter, I think — and yes, this is Bulverism, I know, but it’s what I think is the actual reason — deals with how people think about groups, and how people think about simplicity. (Note that in some cases that how people think about simplicity actually creates real alignments of interest. But these rae alignments of interest that would not exist if people did not think about simplicity in the way that they do, and so if you try to explain these alignments of interest without using this fact, you’re not going to get it right.)

            But benwave said he had an actual explanation! And this explanation did not involve human ideas of simplicity, but did involve a certain symmetry. Which is then what led to my argument that this symmetry does not, in fact, lead to the conclusion he said it does.

            Does that make clearer the structure of what I was saying?

            In response to that approach, I will suggest a potential explanation that could support the (more theoretically tending) leftist view. Assume a game with:

            […]

            I don’t have strong intuitions about how to model such a game formally, and one can easily imagine coming up with different specific games that might go differently when modeled.

            I mean, sure, you could do that. But see what I said above — such a formalized leftism wouldn’t be leftism as it currently exists at all.

            This is itself a kind symmetry breaking and therefore alignment.

            I don’t think you can make that inference. I don’t think that’s valid. Again, review the discussion above. Of course, there we weren’t talking about symmetry breaking, but rather symmetry. Because you see benwave said, the relation of each capitalist to each laborer is roughly the same. This gives us symmetries by permuting the capitalists and symmetries by permuting the laborers. But, as I pointed out above, symmetry does not imply any sort of alignment. Now you’re talking about symmetry breaking rather than symmetry, but it seems to me you’re still making basically this argument — the alignment seems to come not from the breaking of the symmetry, after all, but rather from what symmetry still remains. Or is that mistaken?

            Again: Steelmans of leftism, based on keeping the conclusions but tossing out the internal structure, may have interesting things to say. But the internal structure that currently exists is nonsense.

        • benwave says:

          Well, the first thing I would say is that leftism as you call it is emphatically Not something fundamentally different to economics. Marx was an economist in his time. Many academics in modern times who call themselves Marxists (such as David Harvey) or get called Marxists (such as Pikkety) are studying the economy. I’m not quite sure where the idea that Marxism is not economics came from.

          Secondly, I think I am beginning to understand that we have pretty fundamentally different views of theory vs experiment. I’m definitely an empiricist first – if a model doesn’t describe/predict some aspect of the world I would tend not to value that model highly. It sounds to me as though you are asking for a convincing explanation from within your model as to how they would result in one or both of: situations that leftists complain about; or capitalists in aggregate acting to worsen conditions for workers without an explicit coordination method.

          I haven’t studied game theory. I don’t think I’m claiming: “that since the symmetry constrains the equilibrium to lie on the diagonal, it is rational to act in a superrational manner, resulting in class-action without any mechanism of coordination”, but I don’t fully understand your meaning, probably. If another Marxist out there with better knowledge of game theory than me would like to step in, that would be great. But otherwise, I think you may have to just stoop to my level to help me follow the argument.

          In my last post, I tried to follow your game by relating it to a labour market. Can you walk me through your game model, as applied to that?

          • Sniffnoy says:

            [This is going to be a bit out of order for convenience, hope you don’t mind]

            Well, the first thing I would say is that leftism as you call it is emphatically Not something fundamentally different to economics. Marx was an economist in his time. Many academics in modern times who call themselves Marxists (such as David Harvey) or get called Marxists (such as Pikkety) are studying the economy. I’m not quite sure where the idea that Marxism is not economics came from.

            I didn’t say it wasn’t economics (which is obviously false for the reason you point out); I said it wasn’t standard economics, which it isn’t, any more than Austrian economics is. I guess the more standard term is “mainstream economics” rather than “standard economics”.

            This is not of course to assert the correctness of mainstream economics (that would be, uh, quite a strong claim) — merely to say that, from what I’ve seen of it, it seems to at least have that minimal clarity of thought that means it could be correct. Whereas leftist economics looks like confusion after confusion to me. Not to mention that it basically looks like a stagnant research program to me; see my reply to skef above, about how healthy fields steal.

            It sounds to me as though you are asking for a convincing explanation from within your model as to how they would result in one or both of: situations that leftists complain about; or capitalists in aggregate acting to worsen conditions for workers without an explicit coordination method.

            Well, specifically the latter, although I think “in my model” is too strong an assumption here. I’m not asking for something that explains it in my model — I’m just asking for something that explains it in a model. An actual model, that is. Without conflations. Of course naturally I started describing things in terms of game theory, because when you sit down to describe incentives in multi-agent situations, that is what you get. That’s all that game theory is. Assuming of course you actually manage to keep everything clear.

            And so, having said that, to jump back a bit–

            Secondly, I think I am beginning to understand that we have pretty fundamentally different views of theory vs experiment. I’m definitely an empiricist first – if a model doesn’t describe/predict some aspect of the world I would tend not to value that model highly.

            The point I’m trying to make here is that the “leftist model” is not a model. It is not mathematical, it is not formal, it does not have an ontology; it does not exhibit basic clarity of thought. It is full of confusions and meaningless assertions.

            Like I said above — I wasn’t asking for an explanation in my model; just an explanation free of confusions such as individual-group conflations.

            In my last post, I tried to follow your game by relating it to a labour market. Can you walk me through your game model, as applied to that?

            Well, my point isn’t about labor markets. My point is about how your earlier explanation of coordination had an invalid step in the argument. So I don’t really see how I’d do that.

            I haven’t studied game theory.

            Neither have I! In the sense that I’ve never, like, taken a class or read a textbook on the subject. But nothing I’m saying is complicated; all you need to understand what I’m saying is just some basic math and stuff you can learn on Wikipedia (well, not necessarily Wikipedia specifically, but you get the point).

            (Whereas if you actually don’t know that much… well, see above and also my reply to skef above as well…)

            I don’t think I’m claiming: “that since the symmetry constrains the equilibrium to lie on the diagonal, it is rational to act in a superrational manner, resulting in class-action without any mechanism of coordination”, but I don’t fully understand your meaning, probably. If another Marxist out there with better knowledge of game theory than me would like to step in, that would be great. But otherwise, I think you may have to just stoop to my level to help me follow the argument.

            Well, I mean, I had to steelman what you were saying a bit to make it make sense. It would be helpful here to know what part of the above you’re stuck on, but, OK, sure, let’s walk through this.

            So: We’re discussing symmetric games; the examples above are two-player games, let’s stick to those. “Symmetric” here then means that interchanging the two players results in the same game — that is to say, if player 1 choosing A and player 2 choosing B results in payoff x for player 1 and payoff y for player 2, then player 1 choosing B and player 2 choosing A results in payoff y for player 1 and payoff x for player 2.

            If a game is symmetric, then anything else that is determined from it, any functions of it, must of course also respect that symmetry, if applicable. For instance, its set of Nash equilibria. So, for a symmetric game, if (A,B) is a Nash equilibrium (where here I’m allowing A and B to be mixed strategies, not just pure strategies), then (B,A) must also be a Nash equilibrium. (If you’re unfamiliar, “pure strategy” means making a choice deterministically, i.e., the pure strategies can be identified with the particular options; “mixed strategy” means making a choice by some random method (not necessarily uniform).) In particular, if there is only one Nash equlibrium (A,B), then since (B,A) is also a Nash equilibrium, we conclude that A=B; so it’s of the form (A,A), i.e., it’s on the diagonal of possible mixed strategies. (Again, if you are not familiar, in mathematics, in the Cartesian product X x X, the set {(x,x): x in X} is referred to as the “diagonal”.)

            Let’s suppose we go further and suppose that not only is there a unique Nash equlibrium, but that further it’s composed of pure strategies. So it’s of the form (A,A), where A is a pure strategy, a particular option. Meaning it’s not just on the diagonal in the loose sense above; it’s actually on the diagonal in terms of the possible options. Two players who play it will necessarily pick the same option. (Whereas say, if it were composed of mixed strategies, say A consists of picking a_1 with probability 1/2 and picking a_2 with probability 1/2, then if two players actually played this strategy, one might randomly pick a_1 while the other randomly picked a_2.)

            So, to go back to the original argument, you wrote:

            Capitalists as a whole influence Workers as a whole because workers all have some variation on the same relationship with capitalists and vice versa.

            Right, the context here was that you were trying to justify the idea that the class of capitalists acts as a class, and the class of workers acts as a class, just naturally due to incentives, with no mechanism of coordination necessary beyond that. Right? “Capitalists as a whole influence Workers as a whole.” I’m not misreading you there, am I?

            OK. Now, what’s the reason you give for this? “Workers all have some variation on the same relationship with capitalists and vice versa.” In other words: There is an approximate symmetry here. That is to say, interchanging two capitalists, or two workers, leaves the game pretty much unchanged.

            So, it would seem that you are claiming that if we have a set of players such that the game is symmetric with respect to permuting the players in that set, then that set of players implicitly coordinates — i.e., the incentives of the game alone, with no external mechanism of coordination, push the players in that set to coordinate with one another.

            What do we mean by coordinate with one another? Well, in the weak sense, just to escape the Nash equlibrium. But in the strong sense, to act as a class. That is to say, not just to all make the same choice, but to do so in a way that takes advantage of the fact that they are all choosing together. In other words, to look only at the diagonal and then choose the best option on that diagonal, ignoring everything outside. Which is what Douglas Hofstadter called “superrationality”.

            Thus, it seems to me that you were implicitly arguing the following: When a game is symmetric among a certain set of its players, then that set of players is incentivized to implicitly coordinate; that is to say, that if we look at the resulting rational behavior for them, and compare to the superrational behavior for them, we will find that these are the same; and so in particular, focusing on just those players — focusing on just the capitalists, in the example above; we can abstract out the role of the other players, the laborers, and imagine that the game is symmetric among all the remaining players — the off-diagonal outcomes will not affect the how those players will rationally behave, because they don’t affect how they would superrationally behave. But, this is false, as the examples above demonstrate, and so the inference is invalid. Symmetry does not necessarily lead to implicit coordination.

            On rereading, it does look like some of what I said was putting words in your mouth, so, sorry about that. Specifically, it doensn’t look you at any point explicitly make the argument that the reason that symmetry implies implicit coordination is because symmetry means the equilibrium is on the diagonal so therefore you can ignore what’s off the diagonal; you didn’t make any argument for why symmetry implies implicit coordination at all, you just asserted it. I just assumed that the argument above was what you were basically thinking, because it’s an easy mistake to make, but I guess you didn’t actually say that (you didn’t actually provide any intermediate step there, you just went from premise directly to conclusion). Nonetheless, the basic point stands: Symmetry does not imply implicit coordination.

            And so your argument for why capitalists would act as a class, or labor would act as a class, is not in fact correct. “Capitalists as a whole influence Workers as a whole” simply does not follow from “workers all have some variation on the same relationship with capitalists and vice versa.”

            Does that now make clear what I am saying?

            (But of course the real broader point I’m making here — go see once again my reply to skef above — is that, with just a little bit of clarity of thought, you should have noticed this. You don’t have to study game theory to notice the hole in the argument — you just have to think clearly. Sure, knowing a tiny bit of game theory (again, I have not seriously studied game theory!) allows one to expound on the hole in laborious detail; but just a bit of clarity of thought should be enough to say, “Hold on, I don’t think that follows. How are you making that deduction? I mean yeah it sounds plausible, certainly sounds like it should be true, but you haven’t provided any actual argument. This sounds a bit like a part/whole conflation, which causes me to worry that it’s actually not true.” You see what I’m saying?)

          • benwave says:

            Ooookay, I Think I am starting to understand you well enough to get the argument. We sure have done a lot of talking past each other and misunderstanding each other! This is good. It looks like you have some very specific definitions of coordination and symmetry which I wasn’t aware of. I think that when I was claiming that capitalists as a whole influence workers as a whole, you understood me to be making a claim that something off of a Nash equilibrium was occurring, but I never intended to make such a claim (If I did, I apologise, I spoke wrongly).

            I claimed that Capitalists as a whole influence Workers as a whole. I still think that is true. In my original reply of this thread I laid out an example of how under a given (and not uncommon) circumstance, individual incentives on capitalists drive wages down for the whole class of workers. Now that I know your definition of coordination, I think maybe you would not consider this to be coordination? But this is the kind of result I was meaning. You’re right that this result does not lie off of a Nash equilibrium, but I don’t consider it to be desirable, fair, or optimal. Leftists would consider it something that is necessary to resist, and historically we have done so by pushing for collective action among the working class (such as unionising).

            As regards your claim about the fundamental building blocks in decision theory, I still don’t think the models you are using are a good basis for the relationship between capitalists and workers. I don’t argue that any of the things you claim follow from that model don’t follow from your model, I just don’t think that model is a sufficient representation of our world – I think it makes some initial assumptions which restrict it beyond that. I haven’t successfully communicated that to you.

            Specifically, the examples you provide are about games that are the same if the players are switched around – what I think is your definition of symmetric. Games between capitalists and workers are emphatically Not the same if they players are switched. A worker is incapable (the vast majority of the time) of even playing the move “pay somebody $30,000 to buy their labour power for a year”, (Note that the vast majority of capitalists however, Are capable of playing the move “sell your labour power for a year for $30,000” – so that’s a big break of symmetry there)

          • Sniffnoy says:

            Ooookay, I Think I am starting to understand you well enough to get the argument. We sure have done a lot of talking past each other and misunderstanding each other!

            Yeah… sorry for kind of going off in that last post. Thanks for responding so reasonably!

            I think that when I was claiming that capitalists as a whole influence workers as a whole, you understood me to be making a claim that something off of a Nash equilibrium was occurring, but I never intended to make such a claim (If I did, I apologise, I spoke wrongly).

            Well, not exactly — see my post above. Normally I’d say that a decent way to detect coordination… but you’re talking about implicit coordination — that is to say, the parties don’t coordinate (and so don’t escape Nash equilibrium) but end up making the same decisions as if they’d coordinated anyway. Hence I operationalized it as “Nash equilbrium = superrational equilbrium”, as discussed above.

            As regards your claim about the fundamental building blocks in decision theory, I still don’t think the models you are using are a good basis for the relationship between capitalists and workers. I don’t argue that any of the things you claim follow from that model don’t follow from your model, I just don’t think that model is a sufficient representation of our world – I think it makes some initial assumptions which restrict it beyond that. I haven’t successfully communicated that to you.

            But, see, none of that is my point. I’m not talking about how labor markets work because I don’t know how labor markets work. I do, however, know a thing or two about good and bad reasoning, and precise vs fuzzy thinking.

            My point here is, again, not about labor markets but about reasoning and foundations. It’s about the question, how can the idea of action-as-a-class be justified?

            Fundamentally, possible answers to this can be grouped into two types. It can be assumed as a result of empirical observation (I am including in this simplifying assumptions that are then then afterward justified wth “well, it seems to work”); or it can be deduced or inferred from other facts. I’ll just call this assumption vs. deduction, for simplicity.

            Now if you just want to claim that this occurs as an empirical fact, then, hey, whatever, I’m not going to argue with that, because like I said, I’m no economist. My issue is not with the claim that this occurs, but with the claim that its occurrence can be easily deduced from other obvious or widely agreed-upon principles.

            If leftists generally laid out the justification for action-as-a-class as, well, it seems to work empirically, so we assume it on that basis, I wouldn’t have such a problem. Nothing wrong with making assumptions, as long as you’re explicit about it! But when instead it’s presented as something that should be obviously true, that you could figure out for yourself without detailed observations of labor markets just from general principles — well, what that looks like that is some bad reasoning, and so what I conclude is that leftism is built on unsound foundations.

            And this is just one example! I haven’t even touched all the fricking map-territory confusions (although that’s not actually leftism proper, but seems to go along with it for some reason). My experience is that leftism just seems to be full of arguments that demonstrate horrifically fuzzy thinking.

            Who cares if a few conclusions are right? You want to draw conclusions beyond what you already know, you need good thinking, and that means precise thinking. When I say foundations, that’s what I mean. Not a few particular facts about labor markets, but the fundamental way of reasoning.

            And my point isn’t just that the argument you made is wrong — it’s that, based on my experience, if you went to a bunch of leftists and made that argument, none of them would notice the error (or they’d offer even worse arguments — seriously, yours is the best I’ve seen by quite a bit). Maybe they’re perfectly careful thinkers in their own domain but haven’t grasped that good thinking is domain-general. Who can say? But I just see such constant fuzzy thinking that, well…

            Specifically, the examples you provide are about games that are the same if the players are switched around – what I think is your definition of symmetric. Games between capitalists and workers are emphatically Not the same if they players are switched. A worker is incapable (the vast majority of the time) of even playing the move “pay somebody $30,000 to buy their labour power for a year”, (Note that the vast majority of capitalists however, Are capable of playing the move “sell your labour power for a year for $30,000” – so that’s a big break of symmetry there)

            Right, I’m not claiming that the capitalists-and-workers is symmetric in all the players. Rather, I’m starting from your claim that it’s symmetric in the capitalists and separately symmetric in the workers. Because that was what you used — you said, in effect, that since it’s symmetric in the capitalists, the capitalists can implicitly coordinate (and presumably so can the workers — it’s not clear where that symmetry gets broken, so that capitalists can implicitly coordinate but workers can’t and have to explicitly coordinate, but let’s put that aside for now). This doesn’t follow, of course, but I thought it would be easier to demonstrate that it doesn’t follow with just a 2-player game symmetric in all the players. So, as mentioned above, I abstracted out the role of the workers, considering isntead a game between two capitalists, with the role of the workers implicit in the payoff matrix. You see? Sure, we could demonstrate the same thing in a context closer to your original claim, but there’s just no need; the reasoning’s still wrong either way. Indeed, since my example has maximal symmetry, it is in some sense a better demonstration, since it demonstrates that even full symmetry is not sufficient to deduce implicit coordination, let alone the partial symmetry that actually exists.

            I claimed that Capitalists as a whole influence Workers as a whole. I still think that is true. In my original reply of this thread I laid out an example of how under a given (and not uncommon) circumstance, individual incentives on capitalists drive wages down for the whole class of workers. Now that I know your definition of coordination, I think maybe you would not consider this to be coordination? But this is the kind of result I was meaning. You’re right that this result does not lie off of a Nash equilibrium, but I don’t consider it to be desirable, fair, or optimal.

            Well, again, actually getting it off a Nash equilibrium would presumably be explicit coordination. For implicit coordination, well, see above — we can say implicit coordination is occurring if no-coordination and coordination lead to the same results. Is that the case here? You tell me.

            But notice now that you’ve made your argument contingent not just on the symmetry you originally described, but rather on particular features of the labor market beyond that. Like, you seem to be implicitly agreeing here that your original symmetry-based argument was insufficient, is that right? (Especially given what you write afterwards, about how apparently laborers don’t implicitly act as a class and have to explicitly coordinate instead.)

            But that means that action-as-a-class isn’t some necessary phenomenon that can be deduced from first principles under general conditions. Rather it’s dependent on what the labor market looks like. Which casts things in a very different light! Well — I basically already made this point above; I don’t need to belabor it.

  4. skef says:

    Evolution in action: Remaining specimens of rare animal selected to look poorly matted into video, making hunters assume they are a hoax and stay home.

    (Let me explicitly cancel the implication: I’m not making a claim about the production of this video.)

  5. sty_silver says:

    Democrats seem to have about an 8% congressional edge over Republicans right now. The chance for them to win the house after the coming elections, according to PredictIt, is 65%.

    These two facts together must seem absolutely astounding to anyone unfamiliar with US politics. In a real democracy, if one party is leading by 8% nationwide, one would naively expect at least a 99% chance here. Perhaps 95% with randomly drawn districts rather than proportional allocation. Is anything unfair or misleading about putting it this way? And am I correct that the sole reason for this is Gerrymandering? And if yes, is there any justification for why this is okay?

    I’ve never heard it put this way, and I feel a strange disconnect in general about this issue. If the left camp talks about the right, it seems to be lumped together with other things that evil republicans do, as if policy or value differences are comparable to substantially subverting democracy. On the other hand, I’ve never heard any justification from the right. The only one I could think of is ‘both sides do it. we’re just more competent’. Which I guess might be true.

    • skef says:

      Even with no Gerrymandering you can see that kind of gap with any region-based voting system. A lot of the extra 8% might be lumped into cities, for example, and some voting regions might have large cities and others might lack any. There are aspects of the U.S. system built to be biased to area- rather than population-based representation that existed before Gerrymandering was (much of?) a thing.

      On the broader “is it okay” question, there is enough disagreement on what factors should count (e.g. the extent to which voting should be regional at all) as to make the current situation de facto meta-okay. There are enough variables to make a definitive argument difficult.

      • albatross11 says:

        This is a feature of voting districts, I think. The dynamic to understand here is that every vote past 50% + 1 is wasted. So you can imagine some weird contrived situation where every voting district but one is 50%+1 Republicans, the last voting district is 100% Democrats. The Republicans win all but one seat in the House, but the Democrats have a majority of voters overall.

        • HeelBearCub says:

          So you can imagine some weird contrived situation

          It may be weird and contrived, but it’s not hypothetical. Roughly, this is what gerrymandering is, by definition. States like NC and PA have huge Republican majorities in congressional seats, but are split close to 50-50 in partisan advantage, with the majority vote share for Congress overall frequently going to Democrats.

          • gbdub says:

            To be a bit more precise, the point of gerrymandering is to maximize this effect, but it could also come about naturally even if you drew logical looking districts. A simple urban / rural split might have deep blue cities while the rural areas are purple.

            A state where the districts were drawn so that every one had precisely the same partisan split as the state as a whole would have to be itself pretty gerrymandered.

          • HeelBearCub says:

            @gbdub:

            A state where the districts were drawn so that every one had precisely the same partisan split as the state as a whole would have to be itself pretty gerrymandered.

            I don’t think anyone is arguing for this, and it seems like something of a red-herring. No one is denying that voters with similar interests tend to also cluster geographically.

            Nonetheless, the examples we see of redistricting reform tend to show increases in numbers of competitive seats. For example, in CA:

            Although the commission was expressly prohibited from drawing maps based on political outcomes, the maps drawn by the commission nonetheless improved competitiveness by returning community-focused districts, roughly tripling the number of districts considered competitive.

            I believe that the CA commission used as their starting point a map drawn via algorithm which sought to draw districts that followed straight lines or natural boundaries as much as possible to result in districts of equal population size. Not sure if I can dig up a citation for that, though.

          • Garrett says:

            For reference, a comparison of the old and new congressional maps.

          • HeelBearCub says:

            I’d love to see those PA maps overlayed with the metro area limits indicated, or population density, or something.

          • gbdub says:

            @HeelBearCub

            I don’t think anyone is arguing for this

            Never said they were. Point was just that even if we were to completely undo gerrymandering, one party is still almost definitely going to have some more “wasted votes” than the other, as long as we have districts and Ds/Rs are not evenly dispersed geographically.

            Thus “more wasted votes” is not necessarily “unfair”. The OP asked if the “sole” reason was gerrymandering, and I think the answer is “it’s worse because of gerrymandering, but there are real demographic trends that will always cause some degree of mismatch between nationwide voter totals and party representation in a district-based FPTP system”

        • sty_silver says:

          Gerrymandering works through districts; the principle you just explained is the whole point of Gerrymandering. So it’s definitely about districts, but that doesn’t answer whether or not it’s about Gerrymandering.

    • beepboopbopbeepboop1 says:

      Gerrymandering is overstated:
      1. Intuitively, there shouldn’t be any partisan gerrymandering effect, since this contradicts the incentive of incumbents to gerrymander for an incumbent advantage.
      2. Empirically, there does not seem to be any partisan gerrymandering effect.

      The GOP’s advantage comes down to 2 things:
      1. Structural advantage. Young people and people of color cluster themselves in cities and thereby reduce their liklihood of being in swing districts.
      2. Overestimating the GOP’s chances. Predictit has been hugely rightwing ever since Trump voters won money in 2016. For this reason, Predictit’s numbers have a GOP-favorable polling error baked into its assumptions. If you look to 538, where I assume you are getting the 8% number, the Democrats have a much better chance.

      A more novel argument that gerrymandering is probably ok most places (not totally confident in this one though): since local governments are the ones districting in many states, it can often be seen as a simple extension of local popular will.

      • skef says:

        Intuitively, there shouldn’t be any partisan gerrymandering effect, since this contradicts the incentive of incumbents to gerrymander for an incumbent advantage.

        At the federal level “incumbents” don’t draw their own districts, the state legislatures draw them.

        At the state level the majority party typically gets to determine the map. That leads to stronger incumbency for the other party (as voters for that party are packed as tightly into those districts as possible) and sometimes stronger incumbency for influential members. But the parties don’t automatically make every state-level district safe; they want some to be more even so as to make a majority more probable.

        Or at least these are reasonable “intuitive” considerations.

      • Chalid says:

        The conclusion that there is no partisan gerrymandering effect is the sort of thing that people come to different conclusions on based on their methodology, and shouldn’t be baldly stated as fact.

        For example, the first hit off google for “gerrymandering house net effect” for me is this which finds a significant pro-Republican effect, cites some other papers finding pro-Republican effects, and argues that the paper you linked is flawed because “the logistic-regression approach of Chen and Cottrell is insensitive to the main technique of partisan gerrymandering.” Also, on geographic advantage, “the authors find a high correlation between states with single-party redistricting control and large numbers of extra seats, thereby suggesting that geography fails to account for a significant portion of the asymmetry.”

        I myself don’t have the time or energy to get acquainted with this literature (I did not read the paper I linked, just glanced through it for highlights and to check for obvious crazy), but I’m content with the conventional wisdom that all these funny-looking districts are being drawn for mostly partisan reasons, and the people doing the drawing in 2010 were mostly partisan Republicans, so gerrymandering favors Republicans.

      • Scott Alexander says:

        “Structural advantage. Young people and people of color cluster themselves in cities and thereby reduce their liklihood of being in swing districts.”

        I’ve never understood this. As partisan segregation increases, why should it place Democrats in non-swing districts and Republicans in swing ones?

        Do cities for some reason have fewer districts than their population would imply? Or are rural areas where Republicans usually win more evenly balanced between parties than the stereotypes would suggest, and Democratic cities are very unusual in being places where people of one party are heavily concentrated?

        • johan_larson says:

          There is a general trend of people migrating from the conservative country to the liberal cities. But redistricting is not instantaneous; it typically happens every five or ten years. Redistricting rules also tend to allow for a certain level of inequality of representation before they trigger. This means the districts that are emptying out (conservative areas) have districts that represent smaller-than-average numbers, which those that are filling up (liberal areas) represent larger-than-average numbers. And since there is one seat per district, conservatives have a persistent structural advantage.

        • skef says:

          Do cities for some reason have fewer districts than their population would imply? Or are rural Republican-leaning areas more evenly balanced than the stereotypes would suggest, and Democratic cities are the only place where people of one party are heavily concentrated?

          It would be best to do this with data, but from past memory of related discussions:

          1) In the U.S. rural rural tends to lean pretty conservative.

          2) But at this point only a small percentage of people live “in the middle of nowhere”.

          3) Suburbs are mixed.

          4) “Towns” are politically more like suburbs than anything else.

          So most of voters in the large districts that don’t include cities are living in towns, making the districts less conservative-centric than cities.

        • A Definite Beta Guy says:

          Or are rural areas where Republicans usually win more evenly balanced between parties than the stereotypes would suggest, and Democratic cities are very unusual in being places where people of one party are heavily concentrated?

          Probably.
          538 took a look at this in the past, you can see their tool here:
          https://projects.fivethirtyeight.com/redistricting-maps/
          “Compact” districts normally result in a much more GOP-leaning house than a proportionally partisan redistricting. The “current” shows the GOP gaining 10 seats relative to “compact,” but it also shows the Dems gaining 10 seats relative to “compact.” This is an obvious indication that Dems are also gerry-mandering aggressively where they have the ability.

        • John Schilling says:

          I’ve never understood this. As partisan segregation increases, why should it place Democrats in non-swing districts and Republicans in swing ones?

          Geography. Barring extreme gerrymandering, an urban congressional district is going to be nothing but city and suburb, whereas a “rural” congressional district will probably have a small city or large town embedded in it somewhere. Possibly a college town or a bedroom community for a nearby large city.

          Simplistically, assume that Americans split 54-46 Democrat-Republican because all urban dwellers are Democrat, all rural dwellers are Republican, and the US is 54% Urban(*). But assume that 20% of the urban population lives in small cities that are embedded in rural districts, so 43.2% big-city urban and 10.8% rural-embedded urban.

          Dividing 435 House districts evenly by population, that gives 188 big-city districts that vote 100% Democrat, and 247 mostly-rural districts that vote 77% Republican and (thanks to the embedded small cities) 23% Democrat. The Democrats have an 8% net edge in votes, but throw away 20% of those votes in districts they can never win, the Republicans don’t, so the Republicans always win the House 247-188. No gerrymandering required, though it will happen anyway.

          Reality isn’t quite that simple, obviously, with real live Republicans in big cities and Democrats in the wilds. But geography does at least marginally favor the rural side of the rural-urban gap.

          (*) Census department says 83%, but they count any town over 5,000 population as “urban”, so shift that threshold as needed to fit the simplistic model.

        • Erusian says:

          I’ve never understood this. As partisan segregation increases, why should it place Democrats in non-swing districts and Republicans in swing ones?

          Do cities for some reason have fewer districts than their population would imply? Or are rural areas where Republicans usually win more evenly balanced between parties than the stereotypes would suggest, and Democratic cities are very unusual in being places where people of one party are heavily concentrated?

          I’ve actually done data work on this. The reason is the last thing you said. The Republicans tend to preside over more politically diverse districts than Democrats. The cities are unusual in being so uniform. For example, Pelosi’s worst election saw her getting 80.9% of the vote. Paul Ryan’s best election saw him get 68.2%.

          The Democrats are effectively packing themselves into districts of likeminded people while Republicans are not. This has a couple of effects that help Republicans win swing districts. Republicans are likely to have a population edge in swing districts. They are more likely to understand moderates and liberals than Democrats are to understand moderates and conservatives. They also tend to be more moderate.

          • Hoopyfreud says:

            To nitpick…

            The people packing themselves into homogeneous districts are Democrats. If the inverse were true, there would be no homogeneous districts for the Republicans to have a majority in.

            (The inverse is somewhat true, but less-true than my statement, IIRC)

          • Erusian says:

            Isn’t that what I said? I’m not sure what your nitpick is.

          • Hoopyfreud says:

            I think

            The Democrats are effectively packing themselves into districts of likeminded people

            Implies that Democrats are the party of packing, when it’s more that a subset of Democrats pack.

            It’s a very small nit to pick.

          • Erusian says:

            That’s true. It’s definitely not that Democrats have some ideological commitment to packing. It’s just that cities are filled by groups with certain homogenous traits (wealth, professional careers, low preference for lawns,* etc) that strongly imply Democratic preference.

            *This is not a joke and is a very strong correlate of supporting the Democrats.

          • LewisT says:

            @Erusian

            This is not a joke and is a very strong correlate of supporting the Democrats.

            Preference for lawns is a new one to me. Do you know of any place I can read up on that phenomenon?

          • Nornagest says:

            Preference for lawns is a new one to me. Do you know of any place I can read up on that phenomenon?

            New Urbanists get really upset about lawns. But there are only like twelve of them, so it doesn’t matter very much.

      • sty_silver says:

        I’m generally going to disbelieve every claim that prediction markets are systematically biased in such a crude way on popular markets. This should be impossible by design; if what you said were true, then everyone who sees this would be able to make money off of it.

        It’s possible that I’m overestimating PredictIt here, but pointing out that 538 disagrees isn’t enough. my default explanation is that 538 is wrong and PredictIt is right.

        • John Schilling says:

          I’m generally going to disbelieve every claim that prediction markets are systematically biased in such a crude way on popular markets. This should be impossible by design; if what you said were true, then everyone who sees this would be able to make money off of it.

          The amount of money that can be made from actual prediction markets is capped at a fairly low level, and the value many people implicitly lace on believing in the future victory of their favored causes is not small. This isn’t a stock or commodities market, with billions of dollars available to whomever most ruthlessly purges their thoughtstream of error, this is still a hobby and prone to wishful thinking.

        • Douglas Knight says:

          Accuracy isn’t everything. Even if prediction markets were the best source of knowledge, there is no prediction market on the aggregate partisan edge. If we are worried about how the voting edge relates to congressional edge, we should get both numbers from the same source, even if that isn’t the best source. In fact, even if there were a market predicting votes, we probably should still prefer 538 because what you are asking about is the model turning the one number into the other number. 538 might have a worse model than the markets, but the market’s model is implicit and opaque.

          [But there is something much more accurate than prediction markets: the past. If you want to know what’s going on, study 2016 and 2014.]

        • massivefocusedinaction says:

          Keep in mind that Predictit regularly has arbitrage opportunities that are beyond large comparted with other markets (last time I looked selling all the house predictions makes something like 6% or more after fees) in less than a month.

        • albatross11 says:

          I wonder if anyone finds it worthwhile to spend some of their advertising/influence budget on prediction markets.

        • A Definite Beta Guy says:

          Predictit ratings do not seem sensible in this case. Here are 4 GOP seats that are currently “lean Dem.” If the GOP retains the house, you’d expect the GOP to retain some of these seats:

          IA 01 Blum 23
          IL 06 Roskam 36
          KS 03 Yoder 18
          VA 10 Comstock 18

          I am looking at a bunch of other open seats that are Lean Dem – Current R and they are all pretty likely to flip D on Predictit.

          Most of the “tossup” seats also seem “correctly” rated on Predictit.

          But this doesn’t seem to add up to a 33% chance of winning. The GOP needs to win the majority of the tossup races and they need to win some of the Lean D seats to retain a majority.

          • Nornagest says:

            What do those numbers mean?

          • A Definite Beta Guy says:

            My understanding is that it is the purchase price of a share that pays out $1 if the event occurs. So you buy a Roskam-Retains-Seat at 36 cents, which pays out $1 if Roskam retains his seat.

            From what I understand, Predictit takes 10% of your winnings and charges 5% to withdraw any sum of money? So you are putting down 36 cents on Roskam to win 88.6 cents. 538’s odds for Roskam is even-money (and they are pessimistic for the GOP on the whole).

      • Douglas Knight says:

        Overestimating the GOP’s chances. Predictit has been hugely rightwing ever since Trump voters won money in 2016.

        I have always advocated quoting betfair over predictit, though mainly for noise, not bias. (and quoting both is good, at least for American topics)
        But in this case, betfair agrees with predictit.

      • meh says:

        So then why is there so much gerrymandering? Are lawmakers stupid?

        • John Schilling says:

          Gerrymandering is overrated, but it’s not useless and it’s really cheap. If you control the legislature during a redistricting, which we more or less have to do every 10 years, why wouldn’t you take advantage while you could?

          • meh says:

            this is losing the thread a bit, no? (understandable in a deep thread, but these seem to be new claims from the OP?)

    • Chalid says:

      Both sides do it, the Republicans just had a chance to do it more recently – district boundaries for today were decided in 2010 which was a Republican wave year, so lots of state legislatures were heavily Republican then.

      There’s a decent chance the situation will be reversed in 2020 and we’ll have a Democratic gerrymander, at which point the Supreme Court will suddenly find some reason to declare aggressive gerrymandering unconstitutional.

      • sty_silver says:

        I had an intuitive model where gerrymandering is a slow process, that the party in power slowly pulls it into their direction rather than just setting a new standard regardless of how it was before. But that wasn’t really based on anything. Your post implies that redrawing goes very quickly. How sure are you that this is true?

        The use case that differentiates between both is if party A is in power for 20 years and then party B is in power for 4 years. If what you said is true, then at the end of all 24 years, it things would look gerrymandered in favor of party B.

        • HeelBearCub says:

          NC flipped legislative control in 2010, after the redistricting, the map change gave NC 9 or 10 Republicans and 4 or 3 Democrats. The previous maps had majority flipping back and forth with four 7-6 results in the previous five elections.

          ETA: The important thing here is the decennial census and the resulting redsitricting. Redsitricting is the mechanism for the gerrymander, so the effect of the gerrymander is felt immediately after that (rather than gradually).

        • Chalid says:

          Everyone has to redistrict right after the census, because that’s when we measure population. The census is every ten years. Additional redistrictings do occur but are relatively rare.

          • acymetric says:

            I’m not sure that is true. I think it is fairly common for districts to be redrawn any time a state legislature’s power balance flips to the other party.

            My impression may be skewed by living in North Carolina.

          • HeelBearCub says:

            @acymetric:
            Full control of the NC legislature flipped in 2010 for first time in a century and hasn’t flipped back, so I’m not sure where you are getting that impression?

          • acymetric says:

            @HeelBearCub

            I maybe wasn’t clear, they were separate points.

            1) That redistricting is more common than Chalid suggests, particularly after a power shift in the legislature

            2) Because redistricting has been going on more or less in perpetuity in North Carolina for the last 8 years (by way of courts rejecting various redraws during that time), maybe I think it is a bigger/more common issue than it really is nationally.

          • HeelBearCub says:

            Redistricting because of court rulings (or previously because of rulings by the Justice department in regards to pre-clearance for Southern states) means that redistricting “happens” more frequently than just once per ten years.

            But I think it is quite rare for redistricting to occur on a simple flip of control, although, apparently it occurred in Texas and Georgia in the mid 2000s.

      • meh says:

        extreme gerrymandering was possible in 2010 in ways it wasn’t before, because of increased capabilities of data and voting models.

        As for both sides do it.. is there any way to quantify how much each side does it?

    • Thomas Jørgensen says:

      There is no justification. The republican party has to an astonishing degree thrown over any and all respect for democratic norms – it is not just the gerrymander, it is also the various efforts at vote suppression, which are both incredibly overt (The Indian tribes do not use street addresses and vote democratic? Lets make having a street address a voting requirement! That shit is just blatant) , and sometimes more subtle – for example, the custom of putting elections on a workday and then not allocating enough poll stations to urban areas so that voting takes 4 hours of queuing. Yhea, that really promotes participation.

    • JulieK says:

      You get this result because the US has a lot of districts where Democrats have a heavy advantage, and a lot of districts where Republicans have a slight advantage. For the most part, I don’t think this is the result of gerrymandering; it’s just how the population developed. (E.g., cities tend to be heavily D, while suburbs tend to be somewhat R.)

      • HeelBearCub says:

        When the legislature is drawing maps that split college campuses into two districts, it’s not the result of “how the population developed”.

        When a legislator says they will “draw the maps to give a partisan advantage to 10 Republicans and three Democrats, because I do not believe it’s possible to draw a map with 11 Republicans and two Democrats,” it’s not the result of how the the population developed.

        • woah77 says:

          I think what Sen. Jeff Jackson said on the topic is probably terribly apt here.

          “When the Republicans were back in the minority, they wanted to end gerrymandering, and my side always took those bills they filed and threw them in the trash can because we never thought we’d be out of power,” Jackson said. “Then as soon as it switched and Democrats were in the minority, the first thing we said was, ‘Hey, how about independent redistricting?’ And the first thing Republicans said was, ‘How about epic payback?’”

          • HeelBearCub says:

            Yes, gerrymandering is essentially a non-partisan issue. The Gerrymander itself is named for a candidate of a now defunct party dating to 1812.

            The Southern states were one party Democratic rule states for a very long time, sometimes through aggressive use of every political trick in the book. NC is (roughly) the last Southern state to flip to (roughly) one party Republican rule.

            However modern mapping technology allows current gerrymanders to be even more outsized in their effect. Google maps didn’t even debut until 2005, which should give you some sense of what is possible now that wasn’t as possible before.

          • meh says:

            apt more than just here! it is a beautiful example of simultaneous hypocrisy. it is amusing how often you hear cries of ‘hypocrisy’ from a side, when they are engaging in the exact same hypocrisy (just with orders reversed).

            hopefully mistake theory can help move us forward.

          • m.alex.matt says:

            Yes, gerrymandering is essentially a non-partisan issue. The Gerrymander itself is named for a candidate of a now defunct party dating to 1812.

            Not defunct! The Democratic Party is a one-part descendant of the Democratic-Republican Party. Democratic Party is one of the oldest political parties in the world for this reason.

          • HeelBearCub says:

            Interesting. Wikipedia doesn’t characterize it as you do, the party having splintered and then a new, Democratic, party forming. Either way, politicians selecting voters, rather than the other way round, has been with us for a long time.

          • Mark V Anderson says:

            The Democratic Party is a one-part descendant of the Democratic-Republican Party.

            I don’t think this is true. Andrew Jackson is usually considered the founder of the Democratic Party. My understanding is that the Democratic-Republican party became pretty much the only viable party in the 1810’s and 1820’s (age of Good Feelings), and it wasn’t until Jackson that a new clear partisan split appeared again. I don’t think most historians would consider Jackson’s party a successor of the Democratic-Republican Party, just because it came after, or that the name is similar. My knowledge is limited, however, so feel free to trash my comments.

          • Nornagest says:

            I don’t think this is true. Andrew Jackson is usually considered the founder of the Democratic Party. My understanding is that the Democratic-Republican party became pretty much the only viable party in the 1810’s and 1820’s (age of Good Feelings), and it wasn’t until Jackson that a new clear partisan split appeared again.

            You’re right, but so is the person you’re quoting. Jacksonian Democrats were one branch of a split in the Democratic-Republican Party of the early 1800s; the other branch thrashed around for a while and threw off a bunch of short-lived and unsuccessful parties (National Republican, Anti-Masonic) before congealing into the Whigs.

      • meh says:

        When computer mapping software is being used, it’s not ‘just how the population developed’.

        When it’s being done intentionally, it’s not ‘just how the population developed’

        https://en.wikipedia.org/wiki/REDMAP
        https://www.theatlantic.com/politics/archive/2017/10/gerrymandering-technology-redmap-2020/543888/
        https://www.pbs.org/newshour/politics/ap-analysis-shows-gerrymandering-benefited-gop-2016

    • Brad says:

      The root of the problem is single member districts. IIRC the constitution allows for multi-member districts in the House but federal law currently prohibits it.

      • Plumber says:

        That law is relatively recent.

        As late as 1967 Hawaii and New Mexico still had “at large” congressional districts.

        • Brad says:

          Like a lot of bad laws and legal precedents in the US it was an unfortunate but necessary response to massive resistance that stuck around. Maybe we should have just reinstituted military reconstruction.

    • Deiseach says:

      And am I correct that the sole reason for this is Gerrymandering?

      I think it’s what skef said: a lot of Democratic support is packed into cities. Whether you elect Representative Dee M. O’Crat by fifty thousand or fifteen million votes, you are still only electing the one representative in a first-past-the-post system. Where the tug of war comes in about redrafting electoral districts is that no party wants to lose votes, so both parties have an incentive not to rock the boat too much (if Party Tweedledum gerrymanders so it gets extra congresscritters in State A, Party Tweedledee will gerrymander so it gets extra congresscritters in State B to balance out).

      • HeelBearCub says:

        This just makes the gerrymander easier to pull off. In the absence of redrawing district lines, we could say that population trends would have this effect. But we do re-draw district lines, with the heaviest effect coming immediately after lines are re-drawn. The population trends only serve to entrench the effect. They don’t create it.

      • sty_silver says:

        But pointing out that democrats are disproportionately located in cities only answers the question if you introduce the further claim that, in cities, there tend to be more citizens per district. Otherwise, it would at worst increase variance, which could change the probability a bit, but not that much.

        • SamChevre says:

          I think that even in districts all the same size, the disproportionate concentration of voters for one party hurts in any district system.

          Imagine 5 districts, each with 100 voters. Two are heavily D–90% of votes are for Democrats. Three are moderately Republican–60% of votes are for Republicans.

          In that case, the population is 60% Democrats, but the office-holders are 60% Republicans.

          A large part of the gerrymandering argument is over when this sort of arrangement of districts counts as a gerrymander, versus reflecting a goal that districts have some coherence.

          • HeelBearCub says:

            That answer begs the question.

            You have assumed into being a map with a specific distribution, but that map isn’t a natural consequence of anything.

          • sty_silver says:

            As HBC said, the question is how likely is that situation to happen? Without a specific biased mechanism (like the migration to liberal cities some have mentioned) it’s very unlikely to significantly favor either party. Disproportionate population density itself doesn’t do that.

          • SamChevre says:

            HBC and sty_silver

            It’s not disproportionate density: it’s disproportionate concentration. All that’s required is that D-leaning areas are more-heavily-D than R-leaning areas are R; the population density can be identical.

          • HeelBearCub says:

            @SamChevre:
            No that still begs the question, because “area” doesn’t mean anything absent a specific map of districts.

            Areas that are lower population density are currently more Republican. If we combined all of non-metro areas into one big district, and added in the suburbs where necessary, we could easily draw maps that segregated the Republicans.

            I think the mistake here is to think that actually land area matters very much in a demographic landscape which varies widely in population density.

          • SamChevre says:

            @ HeelBearCub
            … we could easily draw maps that segregated the Republicans.

            This might be the case, but I do not think so: here’s why (put a different way).

            (I think) the floor for Democratic support is much much higher than the floor for Republican support, across the smallest unit of voting space. This means that it’s very hard/nearly impossible to draw plausible maps that crowd Republicans as tightly as Democrats.

            In other words, there are plenty of adjacent voting precincts that vote 100% Democratic, and all adjacent voting precincts are above 90% Democratic: there are very few such precincts for Republicans.

            This would be worth confirming, but a few representative states match this pattern. Look at the Atlas of Redistricting with Compact districts following county borders where possible; that’s generally favorable to Republicans relative to a strictly proportional split. (I checked North Carolina, Tennessee, Illinois, Ohio, Pennsylvania, Massachusetts, and California; only CA and MA were more D using compact-counties than proportional.)

          • cryptoshill says:

            I mean – I think there’s strong evidence that the larger your cities network effect is, the more likely it is to be heavily liberal. I don’t have a ton of research on this but I have an intuition that there are Substantive Facts about city life versus rural/suburban life that make this to be so.

          • idontknow131647093 says:

            Samchevre is correct.

            If you require continuity in districts, the heaviest D district will be heavier D than the heaviest R. In 2012 the most Republican Congressman got something like 80%, the most Democratic D got 90+.

            There are entire precincts in my city where Trump and Romney got 0 votes. Its basically the geographic version of liberal arts colleges becoming devoid of conservative voices.

            And, indeed we used to draw much more “objective” districts way back when. They used to be historical like using rivers or county lines. It wasn’t until the Warren court created the 1-man-1-vote doctrine that redistricting even became a consistent thing.

    • A Definite Beta Guy says:

      As has been stated, the PredictIt odds are more conservative leaning than they should be. 538 shows a Dem chance of 85%.
      538 also shows the Dems having a 8.5% advantage. If you assume the same 3% 3rd party vote as in 2016, you get the following %:
      Dem: 52.75%
      Rep: 44.25%
      3rd: 3%

      Translating into the following seats:
      Dem: 234 (54%)
      Rep: 201 (46%)

      These numbers taken by themselves are misleading because the GOP definitely has over-representation over the last few cycles, but in this particular cycle the winner-take-all nature of the contests will produce something close to the vote-share (at least according to the forecasts so far).

    • dodrian says:

      This will inevitably happen in a FPTP system. If one party wins 51% of the vote in 50 districts, and the other party wins 90% of the vote in 50 districts, both parties will have an equal number of seats, but the second party will have much broader national support. Gerrymandering can exacerbate this effect, but it isn’t by necessity the cause.

      In the UK 2015 election, the Conservative party garnered 36% of the national vote, but 51% of parliamentary seats. It’s even been argued that the conservatives were at a disadvantage in that election because of population changes since the last constituency boundary changes.

      The disadvantages to this system are obvious (underrepresented parties), but it does have two big advantages (in my mind). Firstly, it ties a politician to their district, ensuring that every district has someone who must listen to local issues, and as a plus even the leader of a party has to fight an election battle. Secondly it makes it much harder for an extremist party to gain seats, as though they might get a few percent of the vote nationwide (enough for a few seats), they can’t do that in a single district, resulting in no policy influence (contrast with most European legislative bodies, which use proportional representation and have to deal with fringe parties).

      • sty_silver says:

        It only inevitably happens in a FPTP system if you introduce an additional mechanism by which districts are systematically drawn to cause this effect in one direction. Otherwise, you would have the effect pop up occasionally, but in both directions, and it would equal out in expectation. It would increase variance (I think), so that would push any probability a bit towards 50%, but not much. It doesn’t explain a probability of 65% (of Democrats winning the house). I don’t think it would explain 90%, either.

        • dodrian says:

          I think others have demonstrated that this systematic bias can be produced by sensible districting policies, such as rural districts, rural districts including a small city, and urban districts.

          As a further example that this can greatly affect an election – again in the UK 2015 election the Scottish Nationalist Party received 4.7% of the nationwide vote, but took 8.6% of the seats in parliament — almost double the representation they polled at. That’s because their support is entirely north of the England/Scotland border, ie, it’s naturally a biased distribution.

          Imagine an alternative universe where Scotland has a closer population to that of England – say 40% of the UK total. In Scotland the Pro-Unicorn party is doing really well, and expected to mop up 80% of the vote, but they only have 30% support south of the border. The rest of the vote in each region is taken by the Anti-Unicorn party. In total they are polling at .4 * .8 + .3 * .6, or 50% support nationwide. From this you would guess that the chances of the Pro-Unicorn Party winning the most seats in Parliament is about 50%, but in reality the actual results of the election will be about 99% in favor of the Anti-Unicorn party – the Pro-Unicorn party would have to win all the Scottish seats (which they can do handily), but they’ve also have to win a large number of non-Scottish seats (which they simply can’t do).

          It’s a bit more complicated than that in real life of course, but demographics that do correlate with which party you vote for (age, occupation, education), and they also correlate with where you live.

          Note, I’m not arguing that gerrymandering doesn’t occur in the US, just that you can get a result like the one you mentioned – where one party is polling at +8 points nationwide, but because of how seats are awarded in a FPTP system they only have a 65% chance of winning.

      • Thomas Jørgensen says:

        Question. Are you really so sure that is a good thing?

        Yes, europe gets a few communists and fascists in their parliaments, but it also means the pirate party gets to fight back against the intellectual property lobby in a venue where people have to listen, and hell, even the commies at least are staunchly against caving to the banking lobby.

        • Ketil says:

          I’m pretty sure that having fringe parties with at least nominal influence is a good thing, and googling for extremism the other day, I found several references claiming that the presence of populist right parties reduced extremism.

          If you look to Scandinavia, Sweden has bigger problems with right-wing extremism than either Denmark or Norway, and this might be in part due to the availability of populist parties serving as a lightning rod (especially) for anti-immigration concerns. Also, I think there is a greater willingness to discuss controversial issues in public, while Swedish press has had a stronger policy of no-platforming.

          I was at a panel debate on extremism yesterday (hence the googling), and one take-home message for me (but not stated explicitly) was that extremists – both religious and political – lack strong ties to society, they feel unwelcome and marginalized – which breeds resentment.

          Perhaps the danger of fringe parties is that they might gain a pivot position between traditional blocks, giving them more influence for their particular cause than they deserve. This only works to the extent that the larger parties are willing to compromise, and in practice, actual policy tends to be more symbolic than effective.

          • azhdahak says:

            If you look to Scandinavia, Sweden has bigger problems with right-wing extremism than either Denmark or Norway, and this might be in part due to the availability of populist parties serving as a lightning rod (especially) for anti-immigration concerns.

            This might also have something to do with the existence of the Progress Party and Danish People’s Party, which AFAIK don’t have a clear parallel in Sweden. Popular concerns that lack serious electoral vehicles are probably more likely to motivate people to extremism than popular concerns that have them.

            Another possible factor is differences in the size and composition of the immigrant population.

            Were either of these mentioned at the panel?

      • AlphaGamma says:

        Some European countries have fringe-exclusion mechanisms. For instance, Germany (which uses MMP so has some deputies elected in constituencies and others from lists to ensure proportionality) only allows parties with less than 5% of the vote nationally to enter the Bundestag if they have won three constituencies or represent a national minority (Danes, Frisians, Sorbs or Romani). To be honest, I think the German electoral system is the best one I’m aware of.

    • Slicer says:

      I would take all polling with a massive grain of salt at this stage, considering the amount of disdain Republicans have for it.

      Consider:

      – The vast, vast majority of people do not respond to polls (enormous non-response rate), and this has only gotten worse in recent decades; therefore, you are sampling from “people who answer their phones for people they don’t know” – which is less and less representative of the general population
      – A growing tendency on the right is to disdain polls and the people who conduct them – this began in mid-2016 and was validated by the 2016 election, particularly in supposedly safe blue states such as Michigan and Wisconsin
      – If people choosing not to talk to polling organizations are skewed by party (enforced systemic bias), this is going to make nonsense out of the results very quickly
      – If even a few people on the right are intentionally lying to pollsters because they hate them, this is not going to make things better

      GIGO.

      • Conrad Honcho says:

        I agree with this. If you go on right-leaning forums you will find people bragging about how they hung up on a pollster, or cursed them out, or lied to them. A growing proportion of right-wingers do not believe the polls exist to report the truth but to persuade or demoralize, and they have no interest in taking part and legitimizing this propaganda. I’m not saying everyone does this, but I personally would not talk to a political pollster.

        I’ve never seen Nate Silver, etc, comment on this. So if they’re not even aware that an awful lot of Republicans are lizardmen, then their polls are going to be meaningless. We’ll find out what the country thinks when the results come in next week.

        • albatross11 says:

          Interestingly, there’s another side to the hanging-up-on-pollsters thing: The number of spam calls I get on my landline phone has gotten up to like 19/20 calls at this point. We maintain it as a spamtrap and a way to give people a number that will eventually reach us (when we check messages once a week). A pollster will never reach us unless I happen to pick up the phone, and even then, I’m likely to hang up because my assumption is that anyone calling me up is trying to waste my time or scam me or both.

        • mdet says:

          Why wouldn’t you talk to a pollster? Do you think that there’s something to the “polls don’t exist to report the truth, but to persuade or demoralize”? Because right-wingers collectively not-responding to polls for fear of bias is going to be a very self-fulfilling prophecy, and sabotaging our ability to collect basic public opinion data seems pretty not-worth.

          I rarely answer the phone, but I do try to respond to polls whenever I find them / they find me.

          • Conrad Honcho says:

            Because they’re not done in good faith to begin with. Push polls, misleading questions, arbitrary weightings applied after the data is collected (“We’re just going to assume a D+9 electorate in Texas, oh look, Ted Cruz is finished his voters should all just stay home!”), intentionally misinterpreted results. I would much prefer “no data” to “bad data.”

            We have a method of collecting basic public opinion data: elections. Let’s do that instead of the perpetual poll flogging.

          • albatross11 says:

            I think there are organizations that genuinely try to get the truth with polling–the Pew Center, for example. And polls help you understand how people feel about issues that don’t ever get on the ballot. For example, the reason we know something about how people feel about abortion is from polling data.

        • Scott Alexander says:

          If this were true, wouldn’t you expect polls by Fox or some other conservative pollster to look very different from mainstream polls?

          • Eugene Dawn says:

            Also, if you’re just going to make up the numbers, why even call?

          • Conrad Honcho says:

            No, because many conservatives don’t want to talk to Fox News pollsters either.

            Also, Fox News is not a monolith. There’s plenty of neocons and #NeverTrumpers, and they’re every bit as much a propaganda outlet for Rupert Murdoch as WaPo is a propaganda outlet for Jeff Bezos. Rupert Murdoch’s interests are probably not my interests.

          • Brad says:

            ^^
            If the votes come in next week roughly in line with the polling will you update away from polls are useless?

          • Conrad Honcho says:

            If the results come up in line with the polling, yes.

            If they do not, and Republicans overperform, will you acknowledge that, given their inaccuracies, political polling is misused in media reports?

          • mdet says:

            Not Brad, but just from reading the two Nate Silver articles Dan L posted I think we can say both “political polling is misused in media reports” and “polls are useful sources of information, if you know how to interpret them”. Silver argues that the average error on a single election poll is ~6 pct points, that averaging polls can get you much better accuracy (though he never says how much), that partisan bias in polling happens but isn’t consistent, and that many pundits mischaracterize poll results by leaving these facts out.

            You might be right when you say Nate Silver has some incorrect premises that cause him to miss certain trends, but given that his main job is being a poll aggregator / predictor, not a partisan pundit, I trust that he’ll find a way to correct his blind spots soon enough.

          • Brad says:

            If they do not, and Republicans overperform, will you acknowledge that, given their inaccuracies, political polling is misused in media reports?

            If the polls are in accurate, I’ll update away from them being accurate in the future. The last election, I updated that direction. The election before that I updated in the opposite direction (after Nate silver called almost every election perfectly.) I don’t see what THE MEDIA has to do this one way or the other, except inasmuch as they run their own polls.

          • Dan L says:

            Fox is actually a pretty damn good pollster, I could see a standard Neutral v. Conservative split resulting in them getting attacked from the Right. But if you’re telling me Rasmussen is cooking the books against Republicans, then we’re in crazy land. Cherry-picking polls is bad, but it’s well-within the Overton window – it’s the blanket denunciation without attempting something better that rankles.

            Also, bear in mind that fully 50% of expected outcomes result in Republicans overperforming. They keep the House, then there’ll be some explaining to do.

            (Of course, even a mild Democratic overperformance* is an outsized problem for invisible-conservative theories. The expected update is asymmetric, but the beauty of it is that everyone gets to think everyone else is going to be updating in their direction.)

            ((*I’m interested in hearing why special election results in the past two years don’t count.))

          • Brad says:

            ^^
            Pretty thin polling as I understand it, but I don’t follow super-closely.

          • Dan L says:

            Long story short: margins of error were reasonably large on any given result, but the aggregate is compelling and the signal is really not subtle. Total Democratic swing was a double-digit number that arguably didn’t start with ‘1’.

            (Disclaimer: I fully expect next week’s results to be less dramatic, as turnout advantages are mitigated in regular elections.)

          • Dan L says:

            @ mdet:

            I think we can say both “political polling is misused in media reports” and “polls are useful sources of information, if you know how to interpret them”.

            Succinctly put. It’s easy to lie with statistics, but it’s easier to lie without them.

        • Dan L says:

          I’ve never seen Nate Silver, etc, comment on this.

          I’d like you to elaborate on this in particular – are you saying that you’re dismissing his coverage because you’ve never seen him comment on systemic bias or likely voter models, or is this a more narrow objection regarding this particular cycle’s flavor of the Shy Tory voter?

          • For what it’s worth, my guess is that the polls are wrong, with no opinion on which direction they are wrong in. My basis is that the political situation at present is sufficiently unusual that the techniques they have developed for predicting the vote are not very reliable.

            If they are wrong in one direction, the Republicans gain seats in the Senate and hold onto the house. If they are wrong in the other, the Democrats get a substantial majority in the House and may take the Senate as well.

            If the pollsters are right, the Republicans retain their senate majority, perhaps one seat more or less (50/50 is a majority because Pence breaks ties), the Democrats get a small majority in the House.

          • Conrad Honcho says:

            I read that whole thing. Nowhere in there does Nate recognize that conservatives/Republicans are ignoring pollsters or intentionally feeding them bad data.

            This is not “Shy Tory.” This is “Contemptuous Tory.”

          • Suppose you wanted to mislead pollsters in order to serve your political cause, how do you do it? A Republican can make the Democrats look stronger than they are by not answering the poll or claiming to be a Democrat, which might make Democrats not bother to get out the vote or donate money to candidates. But it also might make them feel like winners and so make voters want to be on their side, while persuading Republicans that it isn’t worth voting.

            The only way you can make the Republicans look stronger is by saying you are going to vote when you aren’t going to bother, but someone who isn’t going to bother to vote probably isn’t going to bother to try to game the polling either.

          • Dan L says:

            @ DavidFriedman:

            For what it’s worth, my guess is that the polls are wrong, with no opinion on which direction they are wrong in. My basis is that the political situation at present is sufficiently unusual that the techniques they have developed for predicting the vote are not very reliable.

            Polls are always “wrong” for some trivial definition of the term, a meaningful prediction needs to discuss confidence intervals or the like. Alternatively, historical comparisons can be interesting: do you think this election will see errors greater than those in 2016? 2014? Some other year, which might be interesting to compare?

            @ Conrad Honcho:

            This is not “Shy Tory.” This is “Contemptuous Tory.”

            Funny you should phrase it like that. How would you feel about a hypothetical article written in the immediate wake of the 2016 election that uses returns to demonstrate that greater response errors strongly correlated with better election-day results for Trump, and does so in order to explicitly refute the idea of “shyness” among Trump supporters? Pre-register your objections now – does it not count if such an article proposes a different mechanism, even if said mechanism had clear empirical support?

            Because if you’re declaring that the crucible of legitimacy isn’t quality of analysis but rather comes down to whether or not an outlet agrees with and promotes your specific explanation for a recognized phenomenon, the burden of proof on you gets much higher. Show me that controlling for “Contempt” resolves biases that aren’t already accounted for in voter models. Show me that this research has multiple data points by multiple outlets, and that there are people (niche is ok!) actively using it to repeated success. Show me that it takes a willful effort to disregard such truths, or an otherwise blinding bias actively harming those that deny it.

            Don’t show me a handful of anecdotal forum posts.

          • Conrad Honcho says:

            I’m not talking about the 2016 election. I didn’t see “I hang up on pollsters” rhetoric until this cycle. We’ll have to wait and see what happens next week.

            I will give you another anecdote, though. We’re building a new house and yesterday my wife was talking with our mortgage broker, and the impact of the election on interest rates came up. The broker probed a little around the edges until my wife admitted we were Trump supporters at which point the broker breathed a sigh of relief and said she could talk freely. Discussing the upcoming election, unprompted, she told my wife she will not talk to pollsters.

            I do not think Democrats have similar anecdotes.

          • Dan L says:

            @ Conrad Honcho:

            I’m not talking about the 2016 election. I didn’t see “I hang up on pollsters” rhetoric until this cycle.

            You’re in a dilemma: either said shift happened early enough and data from the past two years is good enough that you ought to be able to point to its traces, or you’ve suffered a failure of media literacy by expecting 538 to be in the business of pre-emptively second-guessing pollster methodology based on anecdote.

            We’ll have to wait and see what happens next week.

            What will we have to see? You’ve added enough details to your theory that you’ll need a clear signal to support it over a host of plausible explanations. “Strong Republican performance” is… unlikely. “Republican overperformance v. polling” has a ton of moving parts. “Republican overperformance v. polling correlated with strong Republican sentiment” would be a good candidate, but it predates your theorized shift.

            If your theory can’t be translated into anything empirical, it starts to sound a lot more like typical arguments-as-soldiers rhetoric.

            I do not think Democrats have similar anecdotes.

            Sounds like a faulty representativeness heuristic – check the nonresponse base rates, there aren’t enough Republicans in the world to account for all of it. Alternatively, if you prefer anecdotes you could probably get your fill from a brief conversation with an actual pollster. There’s plenty of anti-media sentiment on the Left, it just hasn’t been as incorporated into the mainstream.

      • sty_silver says:

        I’m very skeptical of this explanation. Has it ever happened before that polls were significantly wrong on the nation-wide popular vote? Because if not, then you need districts as a further explanation for why Democrats have 65% (or 85%) and not 99%.

        • Slicer says:

          Depends on what you mean by “polls”, really. Do you mean the average of polls on RCP (some of which are 10 percentage points off one way or the other)? RCP’s average is consistently 3 or 4 points off in midterm elections.

          Also, about districts- we simply don’t have a national party vote in the United States the way other Western countries do it. All voting is done at the district level, and candidates make themselves palatable to the constituents in that particular district. This isn’t to say that gerrymandering doesn’t exist- it’s to say that the results that we could expect when talking about a national voting preference probably aren’t the same as the votes that are actually cast for district-level candidates.

        • Plumber says:

          @sty_silver

          “…Has it ever happened before that polls were significantly wrong on the nation-wide popular vote?…”

          Yes.

        • John Schilling says:

          This could do with an explanation of how one monetizes marginally superior public opinion polling, to the tune of a billion dollars and starting with nothing more than a median SSC reader’s resources.

          • Dan L says:

            to the tune of a billion dollars and starting with nothing more than a median SSC reader’s resources.

            This feels like a bad threshold for determining if an Economic Argument is compelling.

            But to reveal my playbook a bit, a sufficiently risk-tolerant actor can leverage even a marginal advantage into significant returns easily enough if the prediction is against conventional wisdom. Think futures trading on a timed drop and rebound – there was a perfect opportunity in 2016, and anyone who claims special insight without having a fresh pile of money has demonstrably failed to put their money where there mouth is.

            A more sane strategy looking for limited exposure can stick to the prediction markets, where aggressive arbitraging and extensive hedging can get you a decent take. Show me an average return >20% per election cycle and I’ll buy the first round, we can swap notes.

        • 10240 says:

          You could only make money from a new model (e.g. by betting) if others don’t already know it. Maybe better models than raw opinion poll results are already known (or, rather, speculated) and that’s part of why there is a discrepancy between polls and prediction markets.

    • 10240 says:

      Besides reasons others wrote, other causes of uncertainty are last week swings in public opinion, people lying to pollsters or refusing to answer (in a way that’s correlated with their party preference), and perhaps other sources of non-random bias in opinion polls. (In Italian elections, I recall there have sometimes been like 10% differences between exit polls and what some party actually got, even though exit polls are sometimes said to be more accurate than other opinion polls.)

    • rlms says:

      Polls have errors, and support levels can change. I’ve done a bit of election forecasting for fun in the past year (the sequel to GJO) and was apparently reasonably good at it, and even for a presidential election I wouldn’t go 95% based on an 8% lead with a week to go.

      • sty_silver says:

        The presidential election has the electoral college, so that’s an additional element, too. Would you go 95% for the popular vote?

        • rlms says:

          Nope. <5% of an upset is really pretty low. The past 20 US presidential elections included the upset wins of both Truman and Reagan, which involved fairly large differences in the ultimate popular vote from fairly recent polls — maybe not quite 8% with a week to go but not too far off. Also, Clinton won in 1992 by a significantly smaller margin than expected thanks to Perot outperforming expectations, and then there's Trump. I was also imagining the 8% presidential election as being the next one, which I expect will have higher probability of upset than average in recent history (for instance it won't be an easy reelection of a popular incumbent).

    • meh says:

      538 predicts democrats need a 5.5 point popular vote margin to take the house.
      https://projects.fivethirtyeight.com/2018-midterm-election-forecast/house/?ex_cid=rrpromo

      (I will not be defending 538s prediction any further on this thread, but feel free to discuss amongst yourselves)

  6. Aging Loser says:

    Does everyone already know about this difficulty with the “Act-Utilitarian” proposal “The Right Action is the One that Results in the Highest Average Level of Worldwide Happiness in the Long Run”?:

    No single action results in an average level of worldwide/long-run happiness. No single action even results in any level of happiness during the following day for the person/s directly acted-upon. For example, if I tell someone he’s fat he might at first feel sad then tell a friend about what I’d said and the friend might say “Hey, why don’t you start going to the gym with me!” and then he’ll start going the next day and feel great about it – as a result of (at least) three actions – mine, his friend’s, and his own.

    So, all that can be said is that a world with a certain average level of happiness in it which includes the contemplated action has such-and-such a likelihood of occurring.

    But now, suppose you can do A or B. A world with A in it has a 5% likelihood of being super-happy (+10) and a 95% likelihood of being utterly miserable (-10). A world with B in it has a 100% likelihood of being merely tedious and stressful (-1). (Action A might be “lead a Utopian Revolution”; Action B might be “retire to the countryside and tend your garden”.) Wouldn’t there have to be a Non-Utilitarian rule governing whether to do the Risky Thing or the Safe Thing?

    • skef says:

      No single action results in an average level of worldwide/long-run happiness. No single action even results in any level of happiness during the following day for the person/s directly acted-upon. For example, if I tell someone he’s fat he might at first feel sad then tell a friend about what I’d said and the friend might say “Hey, why don’t you start going to the gym with me!” and then he’ll start going the next day and feel great about it – as a result of (at least) three actions – mine, his friend’s, and his own.

      On the timing issue: I think the idea is supposed to be level of worldwide happiness averaged over the long run. On that view the feeling great the next day balances with the feeling sad just after.

      On the more general question:

      So, all that can be said is that a world with a certain average level of happiness in it which includes the contemplated action has such-and-such a likelihood of occurring.

      Not on, for example, the B theory of time. If the whole structure is available, you can judge the rightness of the individual action by tracing the causal implications forward to see what it led to.

      Which is an unnecessarily obscure way of pointing out that for all you say act-utilitarianism can be a sound metaphysical theory of rightness. The concerns you raise are mostly epistemic: How can I judge the right thing to do if only have a vague idea of what else will happen? (There are also similar moral concerns: What about the person who does something likely to work out terribly that just happens to work out well? Isn’t that act still wrong?) But these questions don’t settle the case against utilitarianism; they’re terms in the debate over the nature of “rightness”.

      Wouldn’t there have to be a Non-Utilitarian rule governing whether to do the Risky Thing or the Safe Thing?

      Why do there have to be any rules?

      • Aging Loser says:

        skef,
        Whether in the short or in the long run (unless the short run is extremely short), no single action has any effect on anyone’s happiness. The longer the run, the greater the quantity of other actions that might or might not be performed by other people that come into play. I gave a very short-run example. As the run gets longer the problem just becomes more and more obvious.

        Obviously my telling Jack “You’re fat” doesn’t determine whether or not Jack’s friend Bob will console Jack by inviting him to the gym, so that Jack ends up happier within a day than he would have been if I’d never said this to him.

        With regard to your last question, “Why do there have to be any rules” — the point is that Utilitarianism can’t deal with the question of whether to act in a risky or playing-it-safe way, so Utilitarianism isn’t an adequate moral theory.

        This seems to me to be such an obvious problem with Utilitarianism that I’m surprised I’ve never seen any discussion of it anywhere, but since I’m not very well-read on the topic I thought maybe someone else here had seen it discussed somewhere or maybe had thought of it.

        It’s really two related problems: (1) no single action increases or decreases average happiness or even a single person’s happiness; (2) risk vs. playing-it-safe. The link is that (1) requires probabilistic thinking, leading to (2) because the happiest world might only be possible if you do A, but A might be much more likely than B to lead into a miserable world.

        • Peffern says:

          Anecdata: when I did debate team, this was our go-to Util-basher and everyone either deployed it, or prepared against it. Especially with lay judges.

          • Aging Loser says:

            Peffern,

            So both points, and their relationship, are common knowledge among high school or college debaters and are especially employed in the presence of “lay judges” whom you perhaps regarded as ignorant and naive?

            That’s interesting, because I don’t remember ever having encountered these points in any of the discussions of various moral theories that I’ve read, or among any of the professional “philosophy”-teachers with whom I’ve ever conversed (although my conversations with them haven’t been very extensive because I’m a filthy adjunct.)

            I suspect that in fact you debate-teamers merely said things along the lines of “Yeah, well how do we know what other people are going to do later, if we do that?” and “Yeah, that might have the best results but it’s more likely to screw things up completely.” These are of course things that ordinary people say quite frequently. The question is whether the general principles involved have ever been presented as an objection to Utilitarianism (of the “The Right Act is the One That Will Result In Highest Level of Average Happiness” sort) in general.

            Moreover, if the objection that I’ve summarized is such common knowledge among debate-teamers, and is prepared for by everyone, it would have been easy enough for you to reply to it. That would have been helpful; then I’d be able to summarize this reply for my “students” on Sunday morning.

          • AG says:

            @Peffern: Pfft, that’s why all the good affs play it both ways and include both util and deontology cards in the 1ac. Though apparently Khalilzhad is outdated and there’s probably a new util-nuclear-war card now. (pours one out for Bearden 2000)

        • Hoopyfreud says:

          Utilitarianism can’t deal with the question of whether to act in a risky or playing-it-safe way, so Utilitarianism isn’t an adequate moral theory.

          I don’t see this as a failing; utilitarianism can’t tell me which flavor of pie to buy (assuming I’ll enjoy both equally, and the end-effects of producing the pies are identical. For the sake of argument).

          In practice, the probability distributions probably come into play to tell you what your expected outcome is, but in a truly binary sense, I think a good utilitarian would consider them equally morally weighted.

        • skef says:

          With regard to your last question, “Why do there have to be any rules” — the point is that Utilitarianism can’t deal with the question of whether to act in a risky or playing-it-safe way, so Utilitarianism isn’t an adequate moral theory.

          The universe isn’t required to make Rightness something that can be reasoned out in the moment. Rightness does not have to align with blamelessness, for example*. To assume either does not quite beg the question for deontology, but it comes close.

          * Although they wouldn’t say it this way, many people intuitively think that you can deduce rightness (or not-wrongness) from blamelessness; that the center of any moral framework has to be on the moral character of the actor. But why is this a necessary aspect of morality?

          • Aging Loser says:

            Skef and Hoopyfreud,

            A moral theory must be able to fill in the blank in the sentence “The Right Action is the one that …” in a way that keeps making sense even after you’ve thought about its applications from various angles.

            You seem to agree that this sentence cannot be successfully completed with the words “… results in the highest average level of happiness in the long-run,” for the reason that I gave.

          • Hoopyfreud says:

            @Aging Loser

            You’re arguing that a moral theory must be absolutely prescriptive. I disagree. I think that it’s fine for a moral theory to say, “these two options are equally good.” Something something something [terrible misinterpretation of Godel]. No moral system is absolutely prescriptive, unless you believe in Divine Command and have a red telephone to God.

        • Michael Handy says:

          If I read you right, your problems are basically “People don’t have full knowledge of every state of the universe, and cannot perfectly predict how their actions will propagate beyond the immediate future.” Which is true for any single act, but the Act Utilitarian will say to do the best act given your limited data set.

          That is subtly different from Rule Utilitarianism, in that the Act Utilitarian evaluates based on the probability of the Act being effective, which the Rule Utilitarian would continue to perform according to the rule even if he strongly believed it would lead to a bad outcome in this particular case.

          Yes, that leads to probabilistic reasoning, but any decision made ever in the entire universe devolves to probabilistic reasoning at some point (and Deontological and Virtue theories also have to work out if something is allowed/virtuous under limited data as well. What happens if two acts are equally virtuous? Or one leads to a single greatly virtuous act and the other encourages the cultivation of the virtue more modestly over the long term.)

          The second question is essentially “How do I choose between two equal but radically different outcomes” Which is an open problem in decision theory most commonly solved by throwing a coin. Because the predicted outcomes are EQUAL. And you’ll probably get a chance to make another decision.

          I’d like to point out, though, that probabilities don’t have error bars. They are precise. It is both vanishingly rare that you will have two acts exactly equal in utility, and that your utility function will have absolutely nothing to say about the effect of risk on happiness. So from a practical perspective you will never encounter this scenario.

    • Lillian says:

      But now, suppose you can do A or B. A world with A in it has a 5% likelihood of being super-happy (+10) and a 95% likelihood of being utterly miserable (-10). A world with B in it has a 100% likelihood of being merely tedious and stressful (-1). (Action A might be “lead a Utopian Revolution”; Action B might be “retire to the countryside and tend your garden”.) Wouldn’t there have to be a Non-Utilitarian rule governing whether to do the Risky Thing or the Safe Thing?

      You don’t need a non-utilitarian rule to figure out what to do. Action A has an expected value of -9, Action B has an expected value of -1. Given that -1 is greater than -9, it should be manifestly obvious that utility is maximized by picking B. The true difficult is coming up with such numbers for real world scenarios, but if you do have the numbers then maximizing for any given term is simple math.

  7. johan_larson says:

    You have a very particular superpower. If you look at someone directly and push with your mind, they orgasm. (Obviously, this only works on those who are capable of orgasm in the first place.) Your mission, should you choose to accept it, is to use this power to fight crime.

    • Furslid says:

      Let’s get the really stupidly obvious answer out of the way. Use it on criminals and then punch them. Orgasms are distracting, and I doubt people can coherently defend themselves while cumming.

      • johan_larson says:

        The other way to go is to use your power to make money and pay others to do the anti-crime legwork. Set yourself up as a prostitute catering to those who have a lot of trouble getting off. Charge what the market will bear. I bet there are some older men who due to age and stress and whatnot can’t get off by normal means, and would pay princely sums to hoist the flag one more time.

      • Jaskologist says:

        But that’s not systematic change.

        I attend the rallies of the soft-on-crime politicians I don’t like, and use my power during the keynote speech.

    • The Nybbler says:

      Aww, man, this sounds much more fun to use for petty evil. Boring meetings, political speeches, that sort of thing.

      It could be used as a form of torture for extracting information from criminals. This is evil too, but we can allow for anti-heros. Slightly less evilly an undercover crime-fighter (or spy) could use it to suborn criminals; if regular old sex-based methods work, a super-powered one ought to work even better.

    • Aapje says:

      @johan_larson

      Set up a cheap prostitution service that visits high crime neighborhoods and gives orgasms to adolescents. They will likely be less sexually frustrated and less prone to sex crimes (and perhaps other crimes as well).

      Alternative: Get the president addicted to this and extort him to legalize drugs and implement other measures that lower crime.

      PS. Does this work remotely (when videocalling with someone)?

      • AG says:

        What about getting criminals addicted to this and extort them to give up their lives of crime to continue receiving the service?

      • johan_larson says:

        Does this work remotely (when videocalling with someone)?

        The power does not work over digital video or analog TV. It does work through sunglasses, panes of window glass and simple mirrors. Your call on whether it works through more complicated optics such as binoculars or telescopic sights.

      • John Schilling says:

        Set up a cheap prostitution service that visits high crime neighborhoods and gives orgasms to adolescents. They will likely be less sexually frustrated and less prone to sex crimes

        I believe most adolescents already have the technology to provide their own orgasms without anyone else’s help, and so I believe that you are misunderstanding what is meant by the term “sexually frustrated”.

        • idontknow131647093 says:

          Indeed. The real question is how widespread you can make the effect. Can you make enough adolescent boys orgasm enough against their will that you reduce the teenage pregnancy rate?

          That might have an effect on crime.

        • Aapje says:

          @John Schilling

          I assume that this orgasm would be more pleasant, otherwise this is a pretty useless superpower that can only make the life of a group of severely handicapped better (who cannot masturbate). Those are not going to be career criminals.

          • John Schilling says:

            I should think that the ability to briefly incapacitate people, and in some contexts to publicly humiliate them, would have practical applications that might be useful in crime-fighting. But if you are going to add features not proposed by the OP, then perhaps that would expand the scope of action.

      • Kestrellius says:

        Wait, do we have to reduce the number of destructive/immoral acts committed by people, or can we simply make it so that there are fewer crimes being committed?

        Because if it’s the latter, then Aapje’s presidential-addiction plan suggests a pretty effective technique.

  8. Vermillion says:

    SSC Rumble Recruitment

    Contestant signup: https://goo.gl/forms/WqwCLp1EFbICPn2A3

    Power submission (open to all): https://goo.gl/forms/uOJ8m0zoQ8wTQh333

    Like I said last thread I’m going to be running an email based, OT updated, game of superhero combat where players (and non-combatants) can dream up whatever power they like and then use said power to pound each other into jelly. Enough people seemed interested (n>2) so here’s the kickoff. First off I’ll explain the rules in brief, then post links to sign up and to submit a power.

    The full rules can be found here: https://kevan.org/rumble.cgi?genre=hero&mode=rules. Every contestant starts the game with 100 Energy representing their strength, resolve, and pulchritude. The first thing you use energy for is to secretly bid on the powers you want. Powers to bid on are chosen by the players and everyone can nominate 2 for inclusion from the master list, which you’ll have access to. You only spend energy on powers that you win (unless the power says otherwise) and however much you bid on that power is that power’s strength.

    Examples:
    Name: High Powered Energy Shield. Type: Defense
    Burn X: Add 10X to your Defense.

    Name: Extra Arms. Type: Offense
    You may only assign one Attack per round, but a copy of that Attack is made against every other opponent.

    Name: Mimic. Type: Special
    Spend X: You may use any single Power in play whose Power Strength is X, this round.

    After bidding, play will proceed in rounds, you’ll email me to allocate energy to attacks, defense and whatever powers you’d like to activate. Once everyone has finished allocations then attacks will resolve simultaneously and successful attacks reduce energy in future rounds. Once everyone’s energy levels are updated, the next round begins with the allocation phase. If a hero’s energy is reduced to 0 or below they are knocked out and removed from the game.

    Logistics:
    Signups and power submissions will be open for the next week, until the beginning of OT 114. During this time powers submitted can be discussed and modified but after that they’ll be locked in. There’s no limit to the number of powers you can submit by the way. Contestants will have until OT 114.25 to publicly nominate what powers to bid on, which is when I’ll send out the secret ballots. Basically every OT I’ll start a thread to post the results of the previous round unless I get responses from everyone early, and then we might have 2 rounds in one OT. Anyone who doesn’t email me their allocation by then will spend all their energy on defense.

    • helloo says:

      I’m going to submit
      All for One “Combined effect of all other powers that are being used”
      Incompleteness “All powers that are not being used in All for One”
      Rubber and Glue “Copy target power except double its cost and effect” (Target: All for One)

      😛

    • RDNinja says:

      Sounds interesting! I actually recently set up a Discord server specifically for the SSC community to set up and recruit for online RPGs. You can plug it there too: https://discord.gg/EKvwH6p

    • Jake says:

      Looks awesome, any chance you could make the results sheet of the power submission form public, so we can see what other people have come up with?

      • Vermillion says:

        Fair point, although I got quite a backlog to review atm. Honestly super pleased with the response on there so far.

        Anyone who’d like to view the work in progress can do so here: https://bit.ly/2D8zH6T. Also good for anyone who’d like to follow like, a power theme or something.

        Edit: Reviewed the first 10 submissions, you can argue with my choices and edits until submissions close next Sunday. Please keep discussion in the google doc in the tab marked ‘discussion’. I’ve put notes there for all the powers I’ve reviewed so far.

        • helloo says:

          Do you mind slightly less logic breaking powers that still mess with the mechanics and procedures?

          Ex:
          Power through Obsurity – Deal X Attack to target where X = 50 – Power Strength (how much the winning bet was).
          yfilluN – Disable target activated power with Power Strength greater than this power.
          Kirby – You gain all powers of a player that died by your attack or powers.

  9. Adam says:

    Hi Scott!

    Just wanted to make sure you saw my offer last week to help with the email subscriptions problem. Link to my comment on that Open Thread:

    https://slatestarcodex.com/2018/10/21/ot113-opentekonter-thread/#comment-681627

    TLDR: I work in support for the company that makes Jetpack, the plugin that sends your subscription emails, so I’m pretty much perfectly situated to lend a hand if you need it. My work email is in my profile here on the site.

    Thanks!

  10. bean says:

    Naval Gazing is now a year old!

    Also, I’ve just published my take on the proposed “Space Force”. The short version is that it sets up organizational incentives which are likely to lead to neglect of the parts of space that are valuable in favor of stuff that looks cool but isn’t necessary.

    Also, I plan to finally end posting regular links here. It’s been a year. I will still post occasional links, when I think I have something particularly interesting, but maybe just in the whole-number OTs.

    • Deiseach says:

      The short version is that it sets up organizational incentives which are likely to lead to neglect of the parts of space that are valuable in favor of stuff that looks cool but isn’t necessary.

      Any chance that by concentrating first on the stuff that looks cool, this will then capture public interest once again and help justify the boring uncool but valuable stuff?

      • bean says:

        Probably not, because those aren’t really complimentary. The boring uncool bits are things like communications and recon satellites, while I refer to the cool stuff as Orbital Laser Battlestations. If public enthusiasm for OLBs results in the Space Force getting more money, Space Force leadership doesn’t have much incentive to pour it into better communications for the terrestrial forces. They do have every reason to build more OLBs.

        • woah77 says:

          Which is what I would have really wanted anyways. I mean, with enough OLBs do you even need those communications for the terrestrial forces?

          • albatross11 says:

            Only if you need your military to be able to do something besides destroy targets from orbit.

          • woah77 says:

            But I mean with truly enough OLBs, there shouldn’t be any targets, right?

          • bean says:

            I can’t totally disagree with you on this. Really good orbital lasers are incredibly powerful, and change the ground combat game a lot, at least in hot war terms. And when I think the technology is ready for useful OLBs I’ll be an enthusiastic supporter of the Space Force. But we’re not even close to there yet, and we’ll get better value from other ways of applying military force for the foreseeable future.

          • woah77 says:

            That’s entirely fair, Bean. Until we have the Terawatt lasers required to shoot through atmosphere and do significant damage to terrestrial targets, the OLBs aren’t especially helpful. But that’s why we need to throw lots of money at that problem, isn’t it?

          • bean says:

            That’s entirely fair, Bean. Until we have the Terawatt lasers required to shoot through atmosphere and do significant damage to terrestrial targets, the OLBs aren’t especially helpful. But that’s why we need to throw lots of money at that problem, isn’t it?

            I’m not even opposed to throwing reasonable amounts of money at the problem. What I’m afraid of is that we’ll throw a bunch of money at high-power orbital lasers (terawatts, though? What are you planning?) and neglect our comsats. And then lose a war with China because we can’t talk to our carriers and the orbital lasers aren’t ready.

          • woah77 says:

            Ah, well… with enough OLB they might be able to double as communication satellites, thereby giving us two purposes for three times the cost.

          • albatross11 says:

            Would this use IPoD? (IP over Demographics)

          • woah77 says:

            I wish I could answer that, but if I were capable of doing so I’d be certainly required to covertly remove anyone who observed the answer.

          • Deiseach says:

            Only if you need your military to be able to do something besides destroy targets from orbit.

            Why ever would I want anything else?

          • bean says:

            @woah77

            You could theoretically dual-role the OLB as a comsat, although it’s going to be in the wrong orbit to be really good at that. If you have a phased-array laser, it would be a really good optical comsat, but it would have to chose between doing that and being a weapon, which tends to make people nervous.

            @Deiseach

            Maybe you’re trying to stop someone smuggling weapons, but don’t want to just sink all ships. You need boots on the deck to inspect. The orbital laser is great for making the ship want to stop, but not particularly useful for checking that the crates marked “agricultural equipment” aren’t full of guns.

          • James C says:

            I’m struggling to think of a single current US military commitment where a strategic laser would be better than terrestrial forces. Seriously, the US already has the power to blow up anything they dislike with about six hours notice, and look how effective that has proved. Strategic space weapons would be fantastic against another army in the field, but when was the last time you saw the US fighting one of those?

          • cryptoshill says:

            The US hasn’t had to fight a real force in the field because – well, the Alien Gods of the TLAM Strike can just choose that you don’t have a right to exist anymore. The problem is that if you stop spending money on that kind of thing – your ability to not fight huge conventional wars is reduced.

            That said – I liked the idea of putting all the US Space assets under one organizational roof, because there was a lot of functional duplication between the services for managing comsats. If we remove some % bureaucracy in exchange for paying x% more to get OLBs sooner than we really needed them – there might be a net budgetary savings.

          • bean says:

            If we remove some % bureaucracy in exchange for paying x% more to get OLBs sooner than we really needed them – there might be a net budgetary savings.

            I really doubt that comsats are enough of the military budget for that math to be remotely viable.

          • Why all this talk about lasers? When I was last watching this conversation, some decades back, it was smart crowbars–a smaller version of Heinlein’s rock throwing in The Moon is a Harsh Mistress.

          • Protagoras says:

            @DavidFriedman, Depending on the design of the lasers, it’s possible that with a good power source they may not have to worry about ammo. Also, while it would be insanely difficult to intercept smart crowbars, it’s not absolutely inconceivable. You might, for example, be able to vaporize one with a laser and let the hot metallic gas disperse in the atmosphere. On the other hand, it’s impossible to see a laser coming.

          • John Schilling says:

            “Smart crowbars” run into the problem that anything moving that fast, that low, will be surrounded by a plasma sheath dense and hot enough to block most any plausible source of terminal guidance data. So most likely blind, deaf, and dumb crowbars.

            If you’re looking to do WWII-style strategic bombardment from space, and if you can deliver the requisite mass to and from orbit, that works fine. For anything requiring precision, we may have to wait on OLBs.

          • bean says:

            @David

            First, I should remind everyone that lasers were just an example, not a prediction. In fact, thinking it over now, I my actual prediction is that Space Force would start talking up the benefits of Rods from God, and try to get the “smart crowbars” built. I get a minimum of 5 kg for tungsten and 14 kg for iron from my notes, and John does a good job of pointing out the guidance problems. But it’s a really expensive way of destroying targets, and not nearly as responsive as you might think, unless you have a lot of satellites.

            @John

            I do have notes about various tricky ways to get signals through the plasma, and we shouldn’t discount somewhat bigger rods that slow down enough to stop having a plasma sheathe, then guide the rest of the way down. Less efficient, certainly, and probably not a good idea unless you’re invading another planet, but I could definitely see a Space Force trying to sell them.

        • Deiseach says:

          Well gosh darn it, bean, now I want my Space Force Orbital Laser Battlestation! 😀

          woah77, that’s a platform I’d vote for: more money for terawatt lasers research!

    • Conrad Honcho says:

      Thanks bean, that was a good read. Do you think a Space Force could be useful if they only did OLBs and left the other branches to maintain their own satellites like how the navy still operates their own planes?

      • bean says:

        What OLBs? In the context of the immediate future, the OLB is a thing that Space Force is trying to buy because it’s cooler than existing satellites, but that doesn’t actually do much. I’m not sure why you’d set up a Space Force in that context. Right now, the USAF manages pretty much all military satellites, even though the USN is a bigger user of comm bandwidth. (Communications cables are not really viable for ships.)

        • Conrad Honcho says:

          What, in your estimation, is going to be the purview of the Space Force in its initial incarnation? Are you speculating that they’re going to be given the task of managing satellites and are then going to Distracted Boyfriend meme it, rubbernecking at the new hotness OLBs?

          • bean says:

            Exactly. I haven’t seen a coherent vision for the Space Force yet (probably because there isn’t one), but I assume it would take ownership of all US military space-related assets, including the satellites, the ground stations, and all the tracking stuff. But nobody has explained why that needs to be centralized. The USAF had the strategic nuclear mission as its defining purpose when we stood it up in 1947, and the previous three decades had been filled with prophets of air power declaring that it was the Way of the Future. That hasn’t really happened with space in the same way.

            Add in the fact that the OLBs don’t exist, and you have a recipe for wasting money.

          • Le Maistre Chat says:

            The USAF had the strategic nuclear mission as its defining purpose when we stood it up in 1947,

            Why did you stand her up in 1947? She maybe never got over it. ;_;

          • Dan L says:

            @ bean:

            I haven’t seen a coherent vision for the Space Force yet (probably because there isn’t one), but I assume it would take ownership of all US military space-related assets, including the satellites, the ground stations, and all the tracking stuff. But nobody has explained why that needs to be centralized

            I’m sympathetic to the misalignment concerns gbdub brings up below, but in the short run I find I agree with your argument here. I have difficulty imagining a world where we won’t want to have a mature Space Force several decades down the line, but now seems very premature.

            There’d been more talk about creating a Space Corps under DoAF, which I could see as an effective move in a few years… but I can’t see a compelling argument for why now and why a full service branch, besides naked politics.

          • cassander says:

            @danl

            a space corp under the DAF is more or less what’s under discussion, and that’s part of the problem. It gets you all of the problems of a separate service, but without the benefit of getting away from the air force’s pilot centric culture and concerns. And they’re not even giving them the ICBMs, which for me is the bare minimum you need to make a space force minimally helpful at this point in the game.

          • Dan L says:

            @ cassander

            My impression is that the current administration’s intent is (was?) to create a whole new service branch. Whereas a subordinate Corps wouldn’t have nearly as many complications, but would still have some power to prevent the AF from pilferingneglecting the Navy’s satellite money.

          • cassander says:

            @Dan L

            The terms here get a little…trinitarian. the of department of the Navy, has two services (the navy and marine corps) in a single department. The last I heard of the proposed plan was to apply that model to the space force within the department of the air force. This strikes me as close enough to neglect the navy, but far enough to create most of the problems of independence.

    • gbdub says:

      Good post, but one thing I was hoping you would address is whether the current system is inefficient / underserving certain elements. For example, you note that the USAF runs most satellites but the Navy is a bigger user – is the USAF underserving the Navy needs the way that it allegedly underserves close air support and air transport (i.e. functions that support the Army rather than fighter jocks)? You talk about the Fleet Air Arm problem in the UK, but is our current system, with much of the “Fleet Space Arm” owned mostly by the USAF, that much different? Is that a problem that a dedicated service could improve?

      The best arguments (which might still be bad arguments) for a Space Force would be:
      1) There is currently unnecessarily duplicated, but not mutually compatible, capability with each armed force (and several 3 letter orgs) owning some space assets. A Space Force might reduce this redundancy, and/or make redundancy useful by unifying standards (note that this would be a reason if the satellites are not in fact easily shared, but the various services might already be doing a good job of sharing, I honestly don’t have a great grasp)
      2) Some aspects of “space warfare” are not being adequately addressed because they aren’t the core function of any of the other services (this was part of the logic for spinning off the AF in the first place, that the Army wasn’t going to adequately handle strategic air power).

      You don’t seem to address 1 at all, and you casually dismiss 2 by saying the only thing in that box could be giant orbiting lasers. But I think there are current (or close to current) technologies that might fall into that bucket – space debris mitigation and anti-satellite warfare (offense and defense, kinetic and cyber) are two I can think of offhand. Enough to justify a whole service? Maybe not, but “giant orbital lasers” might be weakmanning a bit.

      EDIT: Also launch vehicles. Kind of crazy the AF let that languish as much as they did, to the point where our only liquid engine options were 40 year old tech from one company or even older tech from Russia (better soon, but only because a couple billionaires got bored). Our ability to make big solid motors is basically down to one company.

      • bean says:

        I have doubts that 1 is a big option that is going to save lots of money. Jointness has been really big since the 80s, and any differences in current systems are going to be based on actual different needs between the different users. And while there are some areas that could probably do with better funding, the answer is to look at how much money they actually need, not to create a service that will have every incentive to exaggerate how important those things are, and minimize the really valuable stuff.

        As far as I can tell, the US basically lost the capability to make a sensible launch vehicle some time in the 70s, and is just now recovering it. I don’t have a good explanation for this.

        • John Schilling says:

          As far as I can tell, the US basically lost the capability to make a sensible launch vehicle some time in the 70s, and is just now recovering it. I don’t have a good explanation for this.

          That would be the Space Shuttle. Nobody in the 1970s had the ability to make a launch vehicle as omnifantastic as the Space Shuttle was supposed to be, so that program failed to do anything well. Except public relations and politics, which it did well enough that for the next three decades anyone trying to get their launch vehicle front and center in NASA’s plans and budget requests was highly incentivized to advertise it as “Shuttle II” and promise to repeat as many of the Space Shuttle’s mistakes as possible.

          Preferably using the same facilities in the same states and congressional districts, because of political loss aversion.

          • Dan L says:

            Except public relations and politics, which it did well enough that for the next three decades anyone trying to get their launch vehicle front and center in NASA’s plans and budget requests was highly incentivized to advertise it as “Shuttle II” and promise to repeat as many of the Space Shuttle’s mistakes as possible.

            To this day, the worst-best-named design proposal I’ve seen has to be “Deep Space Exploration Shuttle”. Never said that out loud in the marketing meetings, huh?

  11. philarete says:

    I’d like to gather some recommendations for popular math books. I enjoyed both of John Derbyshire’s books, “Unknown Quantity” and “Prime Obsession”, so something else along those lines would be great. I like getting the history along with the mathematics.

    As to the difficulty level, my mathematics skill is probably college sophomore level. I can handle vector calculus and some really basic linear algebra.

    • skef says:

      Mathematicians seem to dislike “Everything and More”, but I found it to be an enjoyable read.

    • Tatterdemalion says:

      The book that first got me hooked on maths was “Mathematics: The New Golden Age” by Keith Devlin, which gives a fun overview of lots of areas.

      https://bmos.ukmt.org.uk/home/bmo.shtml has a collection of fun recreational maths problems which don’t require much background knowledge.

      Ian Stewart has written lots of pop-maths, much of it good.

    • Gossage Vardebedian says:

      I too have enjoyed “Prime Obsession” several times, and “Unknown Quantity” was also good, though not of the same quality, I thought.

      There are well-known books on i, e and pi that are good, though not terribly challenging.

      “Fractals, Chaos, Power Laws” by Manfred Schroeder is very interesting.

      I really liked “Gamma” by Julian Havil. Pretty mathy compared to the others.

      “Four Colors Suffice” by Robin Wilson is a decent book on the four-color problem, which is noteworthy for having been proved by exhaustive computation.

      “The Millenium Problems” by Keith Devlin covers seven remaining Hilbert Problems (as of a few years ago).

      Simon Singh wrote an excellent, “30,00-foot view” book on Fermat’s Last Theorem after it was solved, and then a good book on codes.

    • Eugene Dawn says:

      I liked Euler’s Gem a lot. I don’t recall how much history there was, so you may be disappointed on that account, but it’s a great book that shows how underneath a simple but beautiful piece of mathematics there lurks something much deeper with ties to many different areas of mathematics.

    • rubberduck says:

      I enjoyed “Zero: The Biography of a Dangerous Idea” by Charles Seife. It’s about the history of the number zero. It’s got quite a bit of philosophy and history though so if you’re looking for straight math then this isn’t it.

    • littskad says:

      Paul Nahin has written a lot of really interesting books along these lines. “Digital Dice”, “Will You Be Alive Ten Years From Now?”, and “Duelling Idiots” have a lot of nice probability problems. “Chases and Escapes” and “When Least is Best” have nice analysis problems. “In Praise of Simple Physics” and “Mrs Perkin’s Electric Quilt” are about mathematical physics.

      T. W. Korner’s “The Pleasures of Counting” is a really good book about the use of math in gaining insights in real historical problems.

      Raymond Smullyan’s many popular books are fun looks at mathematical logic.

      Martin Gardner’s many books collecting his Mathematical Games columns from Scientific American are all good reads.

      Ian Stewart has a bunch of similar books of his columns, which were a bit more mathematically sophisticated.

      Courant and Robbin’s “What is Mathematics?” has a lot of fun problems and mathematical ideas in it.

      W. W. Sawyer’s “Prelude to Mathematics” is a wonderful book that gives an idea of how modern mathematicians think.

    • I liked “Mathematics: A Very Short Introduction” written by Timothy Gowers. It does a very good job of giving a taste of what modern mathematics is like. However, it’s more difficult than most popular books. I think it’s not so bad, and based on what you say about your mathematical experience I think you’ll do fine, but I have a lot of mathematical experience so you should take this with a grain of salt.

      I also second the recommendation for Martin Gardner and Raymond Smullyan.

  12. SamChevre says:

    Let’s talk about corporate taxes.

    I’m looking for the top-level comments to note both goals and a proposed structure. And then we can critique the proposals, and see whether it is different goals, different ideas of the results, or what that makes them different.

    • SamChevre says:

      My goals: taxes that are indifferent to organizational form, are easy to administer and pay, and do not provide incentives that encourage large organizations.

      My proposal: tax business income when earned and NOT distributed, at the maximum individual tax rate. Count business income on a GAAP basis for businesses with GAAP books. Count income for multi-nationals as the proportion of total enterprise income that US sales are of enterprise sales. And allow business tax losses to be sold if the business closes, and they can be demonstrated to represent real cash losses. Tax capital gains at $0–but since all retained income has already been taxed, capital gains are likely to be low on average.

      • SamChevre says:

        Clearly, one thing that I intended, and thought was implicitly clear, was not.

        In this system, dividends count as income to the owner, and are taxed as income, but are not taxed as income at the corporate level. Only income not distributed as dividends is taxed at the corporate level.

    • baconbits9 says:

      Zero corporate level taxes.

      Done and done.

      • SamChevre says:

        Can you expand? Is all income attributed to individuals when earned? Or is income not taxed until distributed?

        • baconbits9 says:

          I’m personally for no taxes at any level.

          If we are going to tax then consumption and use taxes are far better than income taxes, and if we are going to have income taxes then individual taxes are for better than group level taxes.

      • idontknow131647093 says:

        This. A corporate tax is just an income tax mixed with a capital gains tax that dis-incentivizes re-investment at the corporate level.

        If you think investment is too high and investors make too much money, you raise the capital gains tax, if you think CEOs make too much you increase the marginal rates on high incomes. The corporate tax is just dishonest laundering.

        • SamChevre says:

          I partly agree with your beginning, but disagree with the ending.

          I want to “discourage” re-investment at the corporate level. (“Discourage” relative to present, not overall.) My problem with “no corporate tax” is that is makes re-investment out of corporate funds hugely tax-advantaged relative to investing out of personal funds, and I dislike the resulting tendency of fewer and larger businesses. I also think the principal-agent problems inherent int he corporate form grow with business size.

          • idontknow131647093 says:

            How can you want to discourage re-investment at the corporate level relative to modern times when it is at historic lows? The only major people heavily investing are Amazon and startups. The rest are doing buybacks of stock.

            Look, I’m no fan of capital gains taxes either, I think investment is ridiculously low right now and we need more savings so that people can make big investments and have the thought that those will pay off. But its silly for a corporation to pay a profits tax before distributing that to its investors so they can pay a capital gains tax. That’s just double taxation.

          • Eric Rall says:

            One idea I’ve had along those lines (discouraging retained earnings without double-taxation) would be to equalize tax treatment of equity with debt.

            When a corporation issues bonds and pays interest on them, the interest is a deductible expense for the corporation (i.e. they pay no taxes on the portion of operating profits that they use to pay out bond interest), but the bondholder pays taxes on their interest payments as ordinary income.

            For stock, the parallel tax treatment would be for dividends to be fully deductible as expenses, but taxable as ordinary income to the shareholder.

            Under this scheme, retained earnings would still be taxed ~1.5 times (once as corporate income, and again at a reduced capital gains rate as the value of the retained earnings is reflected in stock valuations), but earnings paid out to investors as dividends would only be taxed once as individual income.

        • Brad says:

          This plan effectively eliminates the existing taxation of non-US domiciled owners of US companies as well as taxation of foreign owners and employees of foreign companies that do business in the United States.

        • arlie says:

          CEOs seem to make most of their income via equity awards. That’s one reason they’ll do anything to goose the stock higher, with particular emphasis on stock buybacks.

          So to reduce CEO income, you either change the terms on which they receive stock awards (please!) or you take measures to reduce stock price gains, affecting all stock holders.

          • idontknow131647093 says:

            I mean they can take them in different ways. One is stock options which is easy enough to tax. You tax as income the initial price and capital gains on the gains.

            The other is bonuses based on stock price, which you just tax as income.

            There is nothing complex other than people being ignorant of the code.

          • Brad says:

            One thing that certainly needs to change is the economically ignorant notion that an option issued at FMV has no value.

    • ing says:

      My goals: we should start by taxing in proportion to negative externalities generated. I optimistically believe that total negative externalities generated are less than the total revenue our nation needs.

      Tax pollution. Tax waste that goes to landfills. Tax the use of raw materials.
      I suspect that it’s a lame cliche at this point to endorse “georgist land taxes”, but I’d enjoy having a conversation about why these are a bad idea, so let’s do that too.
      Aggressively tax any product that is bad for people (alcohol, tobacco).
      Tax for workplace accidents. (And that’s in addition to workplace safety laws, not a replacement.)

      I think Bay Area companies get taxed on traffic generated.

      I’d like to tax for bad memetics (gangsta rap, for example or Grand Theft Auto) but I think it would be too tricky to implement in practice.

      • Nornagest says:

        I’d like to tax for bad memetics (gangsta rap, for example or Grand Theft Auto) but I think it would be too tricky to implement in practice.

        Gangsta rap and Grand Theft Auto are protest music and satire, respectively, and those are the kinds of things you probably don’t want to tax. A lot of people enjoy them both on the object level, though, for the same reason that there’s no such thing as an anti-war movie: people go into media looking for emotional highs, and carnage, however unglamorous, is pretty good at delivering them. (GTA was designed with this in mind from the get-go; gangsta rap wasn’t, but it found the market for it on, like, day two.)

        • cryptoshill says:

          Gangsta rap hasn’t been protest music for at least a decade, probably more than that.

          • dick says:

            “Fuck the Police” was 1988 IIRC, and Gangsta Rap hasn’t really been a thing in the original form since the late 90s. Separate from that, having a tax on “bad memetics” sounds absurdly unimplementable but I’m open to hearing more I guess. Who decides which memes are the bad ones?

          • ing2 says:

            I concede that it’s absurdly unimplementable. : )

      • Tax the use of raw materials.

        Why?

        The user already bears the future opportunity cost, since once you use the raw material you can’t use it again. No externality there.

        • ing2 says:

          Um, like fresh water, for example. Maybe there’s a situation where anyone can go run a pipe into the river and use the water to irrigate their crops. But if too many people do it, then everyone runs out of river. The government should set a tax on the use of fresh water to make sure that it doesn’t get depleted.

          (I don’t actually know what the controls are on who can get fresh water from a river, except apparently California needs more of them.)

          Other resources which are cheap to extract but can become depleted include fish, oil, lumber, metals, and possibly soil quality. Some of these are less renewable than others and would probably need to be taxed differently.

          • Nornagest says:

            You’re reinventing the wheel. In most of the US there are two ways to get water when you’re off a city grid. You can buy fresh water from irrigation projects, in which case you’re paying at the pump, so to speak, or you can pump it out from a well if you have the groundwater rights. Irrigation water is a lot cheaper than city water, but it’s non-potable, and the details of how much you can buy are managed by a fiendishly complicated scheme of water rights. Wells are free except for setup costs and electrical power (which can be substantial), but you can’t usually pull up as much water with them, it’s often very hard water (which complicates using it for farming), and you’re paying property taxes on the land your groundwater rights came with. A lot of rurals have both an irrigation hookup and a small well, and use the well for cooking, drinking, and showers and their irrigation for gardens, fields, and livestock.

            There’s an argument to be made that irrigation water in parts of the West is too cheap or poorly allocated, and there’s a separate argument that the groundwater rights structure doesn’t do a good job of capturing the externalities involved, but it’s not like every farmer in California is digging a pirate canal to the Sacramento River and getting it for free.

          • CatCube says:

            There is an extensive body of water rights law in the American West. This is an extremely intricate body of law, but it rests on “prior appropriation,” where the first person to draw water from a particular watercourse gains the right to draw that quantity or to sell those rights to somebody else, and holds this right against users who begin to draw water from the watercourse later.

          • benwave says:

            Surely in these cases a cap and trade for extraction rights would work better? You ensure in that case that there’s never an unsustainable amount of the resource being extracted (legally, anyway) and the amount which Is extracted finds its true value in the market.

          • Paul Zrimsek says:

            ing2 is still lumping together the extraction of resources from a commons (which produces an externality) with the extraction of resources from one’s own property. I often see the idea of externalities get abused this way.

          • ing says:

            @Nornagest, re: fiendishly complicated scheme of water rights;
            @CatCube, re: “prior appropriation”:
            I hadn’t known this and I think it’s really interesting, but I also think it’s a terrible system which sounds a lot like rent-seeking, and we should try to replace as much of it as we can with taxation.

            @benwave, good point! If we try to control supply-and-demand by setting a tax rate, sometimes we’ll overestimate or underestimate and get bad results. But I think part of my goal was to raise money through taxation, and another part of my goal was to avoid clowny situations where individuals “owned” the “rights” to big chunks of natural resources, and cap-and-trade doesn’t really solve those problems.

            Maybe some sort of an auction, where every year the government auctions off the rights to certain amounts of fresh water? (or to catch certain amounts of fish, or to pump certain amounts of oil?)

          • ing says:

            @Paul, I actually take issue with the phrase “one’s own property” as applied to natural resources, or even to land ownership. I don’t think governments should be allowed to permanently sell these things to individuals, though they might rent out the rights to use them.

            I understand we can’t get to that situation from the situation we’re currently in. But this is a hypothetical exercise where we describe a good tax structure for an ideal nation, and in my ideal nation there is nobody who can describe a river as “their own property”.

          • Other resources which are cheap to extract but can become depleted include fish, oil, lumber, metals, and possibly soil quality.

            You are missing the distinction between resources that are private property and common pool resources. Water for wells is a common pool resource. Metal ore, lumber, and soil quality are not. In each of those cases, it is in the interest of the owner to include in his calculation of cost the fact that he is using up a valuable resource that belongs to him.

          • Paul Zrimsek says:

            @ing: Whether or not you like the present system doesn’t really matter for purposes of determining what’s an externality. What matters is that under the present system, costs resulting from depletion of a privately owned resource are borne by the owner and so are not externalities.

          • Nornagest says:

            I hadn’t known this and I think it’s really interesting, but I also think it’s a terrible system which sounds a lot like rent-seeking, and we should try to replace as much of it as we can with taxation.

            It’s not a great system, but taxing it isn’t a good system either. The basic constraints here are that we’ve got a lot of stupendously productive agricultural land in California, from which we want to get turnips, and a limited and highly variable supply of water going into it from a whole bunch of different places. From a policy perspective, the game is to convert as much of that water into turnips as possible without jeopardizing water supply to cities, flood control, the riparian habitats we’ve decided we need to preserve, and a half-dozen other things.

            Now, if we pick a fixed tax rate for those water sources that looks like it’ll give us about the right ratios on average, it works in average years — but in the first dry year the Central Valley farmers will suck everything dry, because the water they’re drawing is undervalued at the rate we’ve set, and in the first wet year all the farmers will quite reasonably moan for eight months about how they can’t afford to water their fields but look, there’s all that water that no one bought, just flowing into the ocean doing nothing. Making the tax rate variable fixes this, but only as long as the legislature (or whoever sets the rate) correctly estimates the demand for water. I expect they’ll make their first serious mistake about two hours after the law goes into effect.

            Taxes are not magic. Reform of systems like this is not simple. Aside from the above, there are a whole bunch of jurisdictional disputes we need to grapple with, among many other problems.

          • 10240 says:

            The “variable rates” version should really be a bidding system where the highest bidders get the resource at the cutoff rate.

          • ing2 says:

            @David, @Paul:
            I wrote: “I don’t think governments should be allowed to permanently sell [natural resources] to individuals, though they might rent out the rights to use them.”

            You both replied, telling me that things like metal ore were privately owned resources.

            Did you not read my message?

          • Guy in TN says:

            @DavidFriedman

            In each of those cases, it is in the interest of the owner to include in his calculation of cost the fact that he is using up a valuable resource that belongs to him.

            Yes, but it is not in his interest to calculate the negative effects harvesting these resources has on people who are not the resource’s owner.

            If you cut a tree, that’s a tree that I cannot cut down in the future, regardless of who owns the tree. My utility-position is made worse off by you cutting the tree, even if you are the person who owns the tree. This is because “externalities” are not defined by the legal fiat of ownership, but by whether someone is put in a worse position than they were before.

            If you think that’s crazy, consider this: if it wasn’t true, we could “solve” the problem of externalities by putting all resources under single state ownership.

            If you understand why that doesn’t work, then it should be clear why saying “the resources are his own property, so depleting them doesn’t cause externalities” also doesn’t hold.

          • Paul Zrimsek says:

            If you cut a tree, that’s a tree that I cannot cut down in the future, regardless of who owns the tree. My utility-position is made worse off by you cutting the tree, even if you are the person who owns the tree.

            “Worse off” implies a change, which is exactly what there isn’t in this situation as far as you’re concerned. All I’ve done is change it from a tree you can’t cut down because it doesn’t belong to you, to a tree you can’t cut down because it isn’t standing. The only person who’s lost the option is me; the idea of externality was invented precisely to distinguish costs like this one, which are borne by the decision-maker, from those which are borne by others.

          • Guy in TN says:

            All I’ve done is change it from a tree you can’t cut down because it doesn’t belong to you, to a tree you can’t cut down because it isn’t standing.

            The legal ownership of the tree can be changed. Where before, access to the tree was in my possibility-space, now it is not. This is why my position is made worse-off by the tree removal, regardless of who currently owns the tree.

          • Guy in TN says:

            @Paul Zrimsek
            An example:

            Let’s say I’m a carpenter who relies on lumber imported from China. The Chinese timber company decides to no longer cut trees, and instead turns the forest into a nature preserve. This causes my supply of wood to be cut off, which leads to me going out of business.

            By your logic: Since the Chinese timber company owns the forest, their decision to stop selling timber did not cause an externality. The tree doesn’t belong to me either way, so I can’t say I’m in a worse position now that they have switched to being a nature preserve.

            Is this right? It looks like a textbook case of an externality IMO.

          • If you cut a tree, that’s a tree that I cannot cut down in the future, regardless of who owns the tree.

            If I eat some of the ice cream in my freezer, that’s ice cream you cannot eat in the future. That isn’t an externality.

          • 10240 says:

            The Chinese timber company creates a positive externality when it sells timber, and no externality when it doesn’t. (When in sells timber, the timber company itself creates a positive externality, but the system of the timber company and you creates no externality. For tax/subsidy purposes, such a situation counts as zero externality.)

          • Guy in TN says:

            @David Friedman
            Why not? You’ve made my position worse off. Imposing a cost on me that I didn’t incur.

            Where before, access to that ice cream was in the possibility-space, now it is not.

            That you’ve made me worse off using the normal channels of legal ownership is irrelevant. If it was relevant, we would have to say that in a society with single state ownership, the state bulldozing a the home a person lives in, and causing them to be homeless, causes no externalities on the former home-dweller. Since its not that person’s house anyway, legally speaking.

          • Paul Zrimsek says:

            It would broaden the idea of externality to the point of uselessness to extend it to the loss of privately held things which could be transferred to others in the future: there are few things of which that could not be said. The most you can accomplish is to confuse people who are used to seeing “externality” used according to its usual definition.

          • Guy in TN says:

            @10240

            Whether you frame it as a “cost” or a “loss of benefits”, the decision by the Chinese company makes the carpenters position worse off, and is not incorporated into their decision-making.

            I admit to being technically incorrect in this example, by describing it as negative externality, instead of as a loss of a positive externality.

          • 10240 says:

            Any company that relies on the supply of some resource should prepare for the possibility that the price of the resource increases; it has no right to assume that it won’t. There is a variety of ways it can prepare: estimate how much increase can be expected at worst given the market conditions, make long-term supply contracts at a fixed price, buy commodity futures, make plans how it can substitute the resource if possible, make plans how to make money from other activities if its present business model is rendered unprofitable etc.

          • meh says:

            These seem to be supply/demand questions not externalities. Though I suppose if interstate commerce can be defined to mean anything so can externalities.

            Anywho if you are a carpenter, and limited lumber supply causes prices to go up, you will pass this on to your consumers, with a net zero effect to you. If prices go up though, new suppliers will arise.

            I’m sure you can argue around any toy example to turn it into a theoretical externality, but I don’t think there is much there.

          • Guy in TN says:

            @Paul Zrimsek

            It would broaden the idea of externality to the point of uselessness to extend it to the loss of privately held things which could be transferred to others in the future: there are few things of which that could not be said. The most you can accomplish is to confuse people who are used to seeing “externality” used according to its usual definition.

            It is the usual definition, I’m just not applying it in the way you typically see it.

            Let’s back up. The reason we are talking about externalities, is that someone proposed taxing the use of natural resources. The counter argument was that using natural resources doesn’t effect anyone else, because they are private property.

            If my argument is true, that using private property does effect other people, that doesn’t mean that the concept of externalities is useless. On the contrary, my argument for taxation of natural resources relies on the concept of externalities.

            I would say that if you understand that all human actions cause externalities, your analysis hasn’t become confused, but clarified.

          • CatCube says:

            I’m not clear on the concept of “taxing” the use of natural resources, as opposed to “selling” them. The attempted distinction seems totally incoherent.

            What, exactly, is this supposed to look like? Right now, if I own a piece of property with timber on it, I can sell the timber. Are you implying that purchasing a piece of property with trees standing on it means that you literally cannot do anything with it? Almost any use of real property, other than simply sitting outside the property boundary and staring at it, would require harvesting at least some of the trees. Hell, even hunting would seem to be verboten, since that’s removing a “natural resource.” That’s a rather fascinating concept of “ownership.”

          • sentientbeings says:

            @Guy in TN

            It is the usual definition, I’m just not applying it in the way you typically see it.

            It seems to me that you are not using the usual definition at all. How might someone convince you of that? People have already said most of what I would have.

            If there is an example that comes close to your usage, I think it would be positional goods, since you could have clearly-defined and non-violated property boundaries but still impose an externality on others. Your usage is much broader than that, because you are asserting an externality even when there is no information pathway linking events.

          • idontknow131647093 says:

            I have to say guyintn that I agree with other people that you are not using “externality” in its traditional sense, and even if you are hewing to a specific dictionary definition, your application indeed makes the word useless.

            For the word externality to be useful, in either the lumber or ice cream the loss in value to 3rd parties must be different than the loss to the property owner. Thus, the loss of the ability to harvest the tree is the same for the property owner and other 3rd parties, so it is not an externality. However if harvesting the tree causes someone else’s property to flood then that is an externality. Similarly, if David eating his ice cream means he and you both cannot eat it in the future, its not an externality because the cost is the same to both. But if him eating it causes him to have a ferocious poop that ends up causing your house to back up with sewage, that is an externality.

          • Guy in TN says:

            @CatCube

            Are you implying that purchasing a piece of property with trees standing on it means that you literally cannot do anything with it?

            Not at all. I am countering the argument that there should be zero regulation, based of the incorrect belief that usage of a resource affects no one but the owner. The logical conclusion of “all actions cause externalities” isn’t that no one should do anything, but rather arguments for regulating resource use can’t be waved away with “this doesn’t affect you”.

          • Guy in TN says:

            @sentientbeings

            Your usage is much broader than that, because you are asserting an externality even when there is no information pathway linking events.

            Could you explain this a little more? It seems to me there is clear pathway from to someone cutting a tree, to me gaining access to that tree at some point in the future and finding it cut. Many other externalities (such as pollution) have a lag time from when the action takes place, to when it affects third parties.

            @idontknow131647093

            For the word externality to be useful, in either the lumber or ice cream the loss in value to 3rd parties must be different than the loss to the property owner.

            I don’t understand this at all. Are you saying that an imposed cost on third party is not an externality, as long as the original actor incurs the same cost to himself?

            However if harvesting the tree causes someone else’s property to flood then that is an externality.

            With the implication being, that if it caused his own property to flood, in addition to other people’s property, it would no longer be an negative externality?

            I’m not an economist. But if this is true, my mind will be blown, and I’ll need to come up with a new word for “action that incurs a loss on a third party”.

            [edit: I’m pretty sure the loss isn’t the same in your examples, because:
            Person A: (-) ice cream is no longer available
            while person B: (-)ice cream is no longer available (+) got to eat ice cream

            So the net losses/gains are not the same for each party]

          • idontknow131647093 says:

            I don’t understand this at all. Are you saying that an imposed cost on third party is not an externality, as long as the original actor incurs the same cost to himself?

            More or less, yes. That is the purpose of inventing the idea of externalities.

            With the implication being, that if it caused his own property to flood, in addition to other people’s property, it would no longer be an negative externality?

            No, with the implication being that the flooding of his own property was a cost of harvesting the tree, and the flooding of other people’s properties is the externality.

            I’m not an economist. But if this is true, my mind will be blown, and I’ll need to come up with a new word for “action that incurs a loss on a third party”.

            You should just use externality because it is the correct word.

            [edit: I’m pretty sure the loss isn’t the same in your examples, because:
            Person A: (-) ice cream is no longer available
            while person B: (-)ice cream is no longer available (+) got to eat ice cream.
            So the net losses/gains are not the same for each party]

            Indeed, but we are not talking about net when we discuss externalities, what we are talking about is things that should be calculated as the “total cost” of doing something. The cost of any person A, B or C eating the ice cream is that no one, not A, B, C, D, etc can eat the ice cream. Under your formulation basically all consumptions of resources would have infinite externalities. For all normal purposes, the eating of the ice cream has no externalities because it doesn’t have a cost that is not also internalized by the ice cream owner (which is the newfound lack of ice cream).

            The purpose of the concept is basically to say, “well yes this activity would give property owner A a gigantic benefit, but it will have small negative effects on B, C, & D that he doesn’t care about.” Then you the next step is forcing the property holder to internalize the externalities he has caused, which ensures that only things that are actually efficient are done.

            The classic case is pollution. I have my tire fired electricity plant that makes me $1000/day, but it pollutes the surrounding town. The pollution is the externality, the fact that a tire I burn today cannot be burned tomorrow by me is an internalized cost.

          • Guy in TN says:

            @idontknow131647093
            Let’s do the tedious definitional debate first: Your first and third statements are in direct contradiction. In your third statement you agreed that an externality is “a loss incurred by a third party”. In your first statement, you said that a loss incurred by a third party is not an “externality”, under certain conditions.

            Anyway.

            For all normal purposes, the eating of the ice cream has no externalities because it doesn’t have a cost that is not also internalized by the ice cream owner (which is the newfound lack of ice cream).

            You eating the ice cream incurs a cost on me, by removing eating the ice cream from my possibility space. Just like the in the example of the Chinese logging company, they remove access to their timber from the possibility space. The carpenter’s position is made worse-off than he was before.

            What part of this logic do you object to? I feel like I’ve spelled this out many-times over in this thread.

            well yes this activity would give property owner A a gigantic benefit, but it will have small negative effects on B, C, & D that he doesn’t care about

            Yes. Exactly. The Chinese logging company doesn’t care about the losses incurred on the carpenter. The ice cream-eater doesn’t care about the losses incurred on non-ice cream-eaters who got access to the ice cream removed from their possibility space. My examples fit this perfectly.

            That my argument shows externalities are everywhere is a feature, not a bug.

          • idontknow131647093 says:

            You eating the ice cream incurs a cost on me, by removing eating the ice cream from my possibility space.

            It is not an externality because it imposes the same cost on me, which I internalize.

          • Guy in TN says:

            So is there a better word then for “action that imposes a cost on a third party”?

          • Guy in TN says:

            Indeed, but we are not talking about net when we discuss externalities, what we are talking about is things that should be calculated as the “total cost” of doing something.

            How does calculating the net gains/losses differ from calculating the “total cost”, in this scenario?

          • Guy in TN says:

            The classic case is pollution. I have my tire fired electricity plant that makes me $1000/day, but it pollutes the surrounding town. The pollution is the externality,

            Wait, doesn’t the pollution impose the same cost on everyone, including the polluter? How can you call it an externality, then?

            (by your definition, of course)

          • IrishDude says:

            @Guy in TN

            I am countering the argument that there should be zero regulation, based of the incorrect belief that usage of a resource affects no one but the owner.

            You seem to have an expansive view of regulation. Putting aside externalities for the moment and addressing direct interactions, let’s suppose Joe really likes Mary and asks her out but she rejects him. This makes Joe sad, perhaps even really sad to the point he becomes a recluse for a year.

            Would you say that Mary has imposed costs on Joe? If so, would you note that since Mary has imposed costs on Joe with her rejection, that her choice to reject Joe should be regulated by the state? Does it matter how much suffering Joe goes through because of the rejection, and should the state be allowed to intervene against Mary’s decision to reject Joe at some level of suffering for Joe?

          • liate says:

            I think that this argument is as much about different ideas of what rights are included in property rights as it is about externalities. Yes, eating ice cream, or cutting down a tree, that you own means that other people can’t have a chance of having that ice cream or tree, but getting them would generally require purchasing it from you; even death doesn’t necessarily mean anyone else has a chance, you could set up an entity in your will to prevent other people from using it. No-one else having a chance to get it is fundamentally a part of owning it; the fact that there generally is a chance of you eventually getting it is just because people tend not to be that possessive and/or eventually die.

            The problem with putting that property right in with externalities is that makes less clear why externalities are bad; if doing basically anything with something effects the object in such a way that could be thought of as an externality, then the actually-problematic externalities seem less a problem, because “effecting other people” is no longer an obvious-seeming bad thing.

          • idontknow131647093 says:

            Wait, doesn’t the pollution impose the same cost on everyone, including the polluter? How can you call it an externality, then?

            No, because it imposes the cost of polluting on himself (an internal cost) but then also imposes separate costs on each other person polluted. Just like the flood scenario. If you flood your own property that is an internal cost, if you flood dozens of properties, all the properties you flood that aren’t your own are externalities.

            If you are really concerned about having a word for the inability for everyone in the world not being able to cut down a tree, just call it a “cost”. A cost of cutting down a tree is that it is now cut down.

          • Guy in TN says:

            @IrishDude

            Would you say that Mary has imposed costs on Joe? If so, would you note that since Mary has imposed costs on Joe with her rejection, that her choice to reject Joe should be regulated by the state?

            Mary does impose a cost on Joe (in the form of deteriorated mental health), but I don’t think romantic relationships should be regulated by the state. There are good reasons not to try to control this aspect of human behavior, but “it doesn’t hurt anyone” is not one of them. After all, the act of regulation imposes costs as well, which have to be considered (for example, the suffering imposed on Mary by making her stay).

          • Guy in TN says:

            @idontknow131647093

            No, because it imposes the cost of polluting on himself (an internal cost) but then also imposes separate costs on each other person polluted.

            So, just like in my tree and ice cream scenarios. Where cutting the tree imposes a cost not just on one person who would have liked the tree for themselves, but on everyone who would have wanted the tree.

            So why are my examples not externalities, again?

          • sentientbeings says:

            @Guy in TN

            I wrote a long reply to your reply, but after checking back a while later I no longer see it posted, and I don’t feel like re-writing it at the moment. I might later.

          • Thegnskald says:

            “Externality” is economics jargon; the concept of “cost” as used there is likewise jargon. Arguing based on common definitions is only going to confuse people.

            The shortest possible version, however, is that, given that the property on which the tree is private, and given that it transfers, the value of the tree is priced into the value of the property. The cost falls on the owner who cuts it down by reducing the value of the land by the cost of the tree, basically.

            In practice things don’t work out quite exactly so neatly. But the version of “externality” Guy is grasping towards here isn’t a helpful one, which should have stopped this conversation before it began. Literally, the idea being reached for -is- cost. That is the word we use to describe that concept.

            An externality is a specific kind of cost in which the cost is imposed on a third party. Not a cost which a third party can potentially purchase. Insurance payouts aren’t externalities, for instance, because the cost has been sold in that case, not imposed.

            We might have a conversation if the tree were part of the commons, but that is a substantively different conversation, and gets into much thornier issues. (For example, property ownership in general might be treated as an externality, as the cost of ownership is mostly imposed on limitations of the rights/actions of others.)

            ETA:

            Which is to say, I guess, that Guy in TN is trying to argue for the externalities of private property from within the framework of private property, which doesn’t work.

            The proper externality being gestured at is the one that happened before private property, when everybody was forbidden from cutting the tree down in the first place.

            What is neglected in this treatment, however, is all the positive externalities that private property gave rise to. It may seem, in the moment, injust that I may not pluck an apple from a tree and eat it – and indeed, that inability is an externality imposed upon me. However, the existence of the apple orchard itself is predicated on the existence of private property.

            A lot of arguments against private property improperly ignore the positive externalities. To a first order approximation, the cost imposed by private property is the inability to live a hunter-gatherer lifestyle; counting the cost of not being able to use things that wouldn’t have existed isn’t property accounting.

          • Guy in TN says:

            @liate

            if doing basically anything with something effects the object in such a way that could be thought of as an externality, then the actually-problematic externalities seem less a problem, because “effecting other people” is no longer an obvious-seeming bad thing.

            The word “negative externality” was not intended to serve as a moral or political call-to-action, but a dry description of a cost imposed on a third party. There is no cut-off in magnitude for where you don’t call it an “externality” anymore.

            “The cost imposed by this action are very small” is a different argument from “The cost imposed by this action are zero”, because in the first argument we have to analyze trade-offs (i.e., what if the costs, though small, are still less than the benefits?), while in the second argument such considerations are unnecessary.

          • Guy in TN says:

            @Thegnskald

            Literally, the idea being reached for -is- cost. That is the word we use to describe that concept.

            The concept I am reaching for is “cost imposed on a third party”. Not just the free floating concept of “cost”, but specifically the kind that is imposed on other people.

            Maybe I should just taboo the term “negative externality” and replace it with “cost imposed on a third party”, since it seems to be causing so much trouble.

            given that the property on which the tree is private, and given that it transfers, the value of the tree is priced into the value of the property. The cost falls on the owner who cuts it down by reducing the value of the land by the cost of the tree, basically.

            This reads as a non sequitur to me. Can you explain how the second sentence follows from the first? How does the price of land in China negate costs imposed on the carpenter, in my timber example?

          • idontknow131647093 says:

            The concept I am reaching for is “cost imposed on a third party”. Not just the free floating concept of “cost”, but specifically the kind that is imposed on other people.

            The problem is you are not discussing costs imposed on third parties. You are discussing costs imposed on every party, including the property holder. The ice cream is eaten for everyone, the tree is cut down for everyone, etc. That is just a universal cost. An externality is specific because it is a cost different from the cost to the property owner.

          • Thegnskald says:

            Guy –

            First, ALL costs are infinite in reach. Name a cost, and I can tell you how it extends to a third party. So the stipulation you are trying to raise is imaginary.

            You have already conceded the China example, so it feels a bit disingenuous to raise it again, but in view of the challenge above:

            China Tree Farm cuts down a tree. Carpenter Today gets lower prices on lumber; Carpenter Tomorrow pays higher prices. Reverse if they don’t cut down the tree.

            Your definition is unlimited; the only “limit” is how imaginative you can get in extrapolation. And since we are only counting negative externalities, and the chain of events is infinite, every choice has infinite cost.

          • Guy in TN says:

            @Thegnskald

            What is neglected in this treatment, however, is all the positive externalities that private property gave rise to. It may seem, in the moment, injust that I may not pluck an apple from a tree and eat it – and indeed, that inability is an externality imposed upon me. However, the existence of the apple orchard itself is predicated on the existence of private property.

            I agree that the creation of legal property caused externalities, some positive and some negative. My position is that they are mostly positive on net, and I support legal property, despite the negative externalities it creates.

            However, not every instance of legal property, at all times, is net-positive. Costs and benefits must be analyzed, and trade-offs considered. The utility of property is more complex than the simple treatment David Friedman proposed that started off this debate.

            A lot of arguments against private property improperly ignore the positive externalities.

            I feel the same way about taxation and the state. Yes, they impose costs, but the we’ve got to factor in the vast utility-gains via positive-externalities they create. Just like in arguments against private property, arguments against taxation can’t be so simple as “but they impose costs”.

          • Guy in TN says:

            @idontknow131647093
            You say that my examples don’t qualify because:

            The problem is you are not discussing costs imposed on third parties. You are discussing costs imposed on every party, including the property holder.

            And earlier, in response to me asking you to clarify whether pollution that also effects the polluter would qualify as an externality, you agreed, and said:

            it imposes the cost of polluting on himself (an internal cost) but then also imposes separate costs on each other person polluted…

            Your statements are in direct contradiction. Am I misunderstanding you?

          • Guy in TN says:

            @Thegnskald

            First, ALL costs are infinite in reach. Name a cost, and I can tell you how it extends to a third party. So the stipulation you are trying to raise is imaginary.

            I mean, I’m glad you understand that all actions impose costs on third parties, but I don’t think this understanding is widely shared. Certainly not to the point where we can just shorthand “costs including imposed costs on third parties” to just “costs”. Remember, the OP that started this conversation was David Friedman saying:

            The user already bears the future opportunity cost

            Which would be incoherent under this shorthand.

            You have already conceded the China example, so it feels a bit disingenuous to raise it again. but in view of the challenge above:

            China Tree Farm cuts down a tree. Carpenter Today gets lower prices on lumber; Carpenter Tomorrow pays higher prices. Reverse if they don’t cut down the tree.

            I didn’t concede the example, I said it didn’t matter if you framed it as a “cost” or a “reduction in benefit”. This is because they have the same effect in terms of the carpenter’s utility.

            I don’t understand your explanation for how the price of land in China, means that the timber company is not imposing a cost (or, benefit-reduction) on the carpenter. [edit: we can change from the China example, to cutting a tree on your own property, to make things less complicated, if you’d like]

          • Thegnskald says:

            Guy –

            I disagree with either one of your axioms, or the way you have interpreted that axiom. Closing a soup kitchen you have run for twenty years isn’t equivalent to taking soup out of the mouths of the homeless who were previously fed.

            If I give a homeless man a bowl of soup, that is a good act. If I give him another bowl of soup the following day, that is also a good act.

            It doesn’t cease to be a good act over the course of twenty years; neither does it become an evil act cease.

            It is important to distinguish these sorts of things when we are analyzing situations, because the implications seriously matter. I think of a church in, IIRC, New York City, which shut down a soup kitchen after many years – and the response wasn’t “They did a great thing for many years”, but a snarling accusation of taking something away people were depending on. Given that this accusation wasn’t leveled at any of the many organizations, religious or otherwise, who never opened a kitchen in the first place… well, it doesn’t exactly create good social incentives going forward. “If you do a good thing for long enough, you no longer receive social credit for doing it, and instead begin accruing social debt against the day you stop” isn’t conducive to doing good.

            Likewise, with this example.

            But setting aside the “Drop in benefits” argument, and focusing on the core issue, if you consider only the costs, you lose sight of what is going on.

            Let’s take the simplest possible example: Cake.

            Do you eat it or save it for tomorrow?

            We can, for moral purposes, treat you-tomorrow as a separate moral entity from you-today. So, you-today eating the cake imposes a special-case externality on you-tomorrow, for the problem construction in which we treat future-you as a distinct entity. How do you justify this?

            There isn’t a single answer. One possible answer, however, is that we cannot actually consider externalities on future entities meaningful, since their existence is dependent on that externality. That is, the consciousness that exists tomorrow isn’t the same in the universe in which you eat the cake, as in the universe where you don’t. Since these are distinct entities, we cannot meaningfully owe one of these entities for something it could never have attained in the first place.

          • IrishDude says:

            @Guy in TN

            but I don’t think romantic relationships should be regulated by the state.

            Why not? I agree, but my response is that Mary has the right to choose which romantic relationships she wants to involve or not involve herself in, without regard to the costs imposed on others. I think utility matters too, but only secondarily, and to the extent it matters, I think individuals are in the best position to determine their own utility and should be able to make their own judgments on these choices without 3rd party imposition, given the limited information 3rd parties have on the utility function for each individual.

            Is your justification primarily based on utility calculations? If so, how do you make these calculations on Mary’s behalf? If not, what non-utility justification do you use for not wanting the state to regulate Mary’s romantic decisions?

          • Guy in TN says:

            @Thegnskald

            Its a little hair-splitty, but I agree that “doing nothing”, or the cessation of an activity, cannot be called a negative externality, but only the reduction in a positive externality.

            I call this hair-splitty, because even in the Chinese timber example, while the company ceasing to sell timber is the elimination of a positive externality, their decision to turn it into a nature preserve is the creation of a negative externality, because it is an action that reduces the carpenter’s utility (as opposed to merely the cessation of an action).

            One possible answer, however, is that we cannot actually consider externalities on future entities meaningful, since their existence is dependent on that externality.

            Struggling to understand what this means. Are you saying we can’t consider negative externalities that have a lag-time, from when the action takes place to when it effects people? Wouldn’t this ruin the classic pollution example- e.g., I dump toxic stuff in the drinking supply, the next day a baby is born and grows up drinks the polluted water, “well, future entities are owed nothing”?

          • Guy in TN says:

            @IrishDude

            Is your justification primarily based on utility calculations? If so, how do you make these calculations on Mary’s behalf?

            Yes, my perspective is all based on maximizing utility- I am a utilitarian. In this case, regulating romantic behavior would probably cause more utility-loss than utility-gain for a multitude of reasons, so I am against it.

            There are other areas in interpersonal relationships though, where regulating behavior would probably cause more utility gain, such as in cases of physical abuse.

            We compare utility all the time- humans are quite good at it, and it comes naturally. No one says “ah, but how can you say the utility he gains from beating his wife, is less than the utility she loses from being beaten?” Interpersonal utility is difficult to quantify, yes, but we can intuit in broad stokes, based on our shared humanity.

          • IrishDude says:

            @Guy in TN

            In this case, regulating romantic behavior would probably cause more utility-loss than utility-gain for a multitude of reasons, so I am against it.

            I’d be interested in seeing you expand on this. Can you discuss these multitude of reasons, and why they apply to decisions on romantic relationships but not other decisions people make?

            Also, how do you think 3rd party observers can determine other people’s utility function?

            Interpersonal utility is difficult to quantify, yes, but we can intuit in broad stokes, based on our shared humanity.

            I’ve been married close to a decade, and couldn’t tell you how much my wife values an apple versus an orange at any given moment, other than the revealed preference of watching what decision my wife makes, which varies by the day.

          • 10240 says:

            Its a little hair-splitty, but I agree that “doing nothing”, or the cessation of an activity, cannot be called a negative externality, but only the reduction in a positive externality.

            I agree. But then if you make some ice cream, put it in the freezer, then eat it next June, you haven’t made anyone worse off than if you didn’t make ice cream in the first place, so the former shouldn’t be considered “imposing a cost on others” any more than the latter. If you buy ice cream from a vendor, that’s a bit more complicated, but even then, the set of the vendor and you doesn’t make anyone any worse than if the vendor doesn’t make that ice cream, and you don’t eat ice cream.

          • Guy in TN says:

            @IrishDude

            I’d be interested in seeing you expand on this. Can you discuss these multitude of reasons, and why they apply to decisions on romantic relationships but not other decisions people make?

            Sure, I’ll outline some reasons:
            1. The harm of a bad relationship most often mental, not physical, so its more difficult to observe than other harmful human behaviors. This is a difference from regulating for example, physical assault.
            2. The individual utility of a relationship is more complex and fine-tuned for each individual than the utility of, for example, lack of food or shelter. So laws regulating it are going to have drastically different utility-impacts for different people, making it difficult to try to determine what the most utility-gaining laws would be.
            3. A person who enters a bad relationship bears the vast majority of the costs on themselves, so there isn’t much incentive to try to regulate it for the larger social good, because the utility-decreasing society-wide side-effects of regulating it likely out-weight the gains. This differs from human behaviors where the social utility gained from regulating the behavior might be greater than the utility-loss (drunk driving, hard drug use, ect.)

            If the problems with the relationship doesn’t match these criteria well, then I would say they do need to be regulated, but these are more rare. For example, physically abusive relationships (for 1), pedophilia (for 2), or incest (for 3).

            Also, how do you think 3rd party observers can determine other people’s utility function?

            At the most basic level, human emotion likely evolved to convey utility. I think all humans intuitively understand that crying=utility loss, smiling=utility increase. These two emotions, happy and sad, are so closely linked to utility that we colloquially use them as synonyms with “utility-gain” and “utility-loss”, and no one bats an eye. And later, speech fills the same role, in conveying desires.

            Determining whether an action increases or decreases a person’s utility, including comparing utility between people, isn’t difficult to do in broad stokes. We evolved to have to mental tools to do just that.

            You can’t rely solely on revealed preferences to determine utility, because you can’t be sure that was a person is “revealing” is actually their preference. For example: A prisoner “reveals” that he is in prison. He says he wants to leave. He’s crying and begging for you to let him out. But “revealed preferences” tells us that actions speak louder than words, so since the prisoner remains in prison, that must be his “preference”. This seems inadequate.

            That all of humanity is constrained in their life choices, by forces both natural and man-made, should give pause to any theory that relies on decisions being freely chosen in its underlying framework.

          • Guy in TN says:

            @10240
            All resources have a mix of both value added by labor, and value added by capital. And as you suggest, it makes less sense to regulate usage of resources that are largely labor-derived (as opposed to natural-resource derived), since the net social benefits of encouraging people to bring those resources into usefulness, are probably greater than the net social harms of the people consuming their relatively small natural-resource based components. (Apologies for the mouthful of a sentence.)

            For example, regulating writing computer programs is more likely to have net-negative utility (via discouraging people from bringing these useful programs into existence), than regulating who gets to use a town’s fresh water supply (which largely exists on its own).

            Still, no resource can fully escape its nature-derived components: even computer programs rely on silicon.

          • 10240 says:

            You are right here. I’m personally I don’t really see private ownership of natural resources (especially through the homesteading principle) as justified. Also, taxing natural resources has the nice property that it can be non-distortionary as long as it doesn’t exceed the profits that can be derived from the resource, unlike most other taxes. Ideally I would support a system where land is owned by government/society and rented out to the highest bidder. (Except perhaps in towns as long as they take up a small percentage of all land. Owning a house but not the land under it could lead to weird situations, and ownership of buildings is a good idea as most of their value comes from human labor. *)

            That said, a program of land confiscation, or imposing excessive tax on land that greatly reduces its value, would erode confidence in property rights a lot, as people wouldn’t trust that other forms of property won’t be affected (unless it becomes common knowledge that everyone has suddenly become a Georgist). It would also be unfair as it would make people who have invested in land worse off than those who have invested in other assets. It would be even more questionable if the land or natural resource was at some point bought from government for market value (i.e. the present value of all expected future profits), rather than homesteaded.

            At the same time, land and other natural resources are a small percentage of all capital in today’s economy, so such policies would do more harm than good in the system we have. If we colonized a new continent (or planet), we could try another system.

            * The high value of land in cities with good economic opportunities, especially where building is restricted, is an even more complex situation as it comes from neither labor (directly) nor nature.

          • IrishDude says:

            @Guy in TN

            The individual utility of a relationship is more complex and fine-tuned for each individual than the utility of, for example, lack of food or shelter. So laws regulating it are going to have drastically different utility-impacts for different people, making it difficult to try to determine what the most utility-gaining laws would be.

            I think you underestimate how complex non-relationship resource allocation decisions are. As Hayek notes in ‘The Use of Knowledge in Society‘:
            “What is the problem we wish to solve when we try to construct a rational economic order? On certain familiar assumptions the answer is simple enough.
            If we possess all the relevant information,
            if we can start out from a given system of preferences, and
            if we command complete knowledge of available means, the problem which remains is purely one of logic. That is, the answer to the question of what is the best use of the available means is implicit in our assumptions. The conditions which the solution of this optimum problem must satisfy have been fully worked out and can be stated best in mathematical form: put at their briefest, they are that the marginal rates of substitution between any two commodities or factors must be the same in all their different uses.

            This, however, is emphatically not the economic problem which society faces. And the economic calculus which we have developed to solve this logical problem, though an important step toward the solution of the economic problem of society, does not yet provide an answer to it. The reason for this is that the “data” from which the economic calculus starts are never for the whole society “given” to a single mind which could work out the implications and can never be so given.

            The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. The economic problem of society is thus not merely a problem of how to allocate “given” resources—if “given” is taken to mean given to a single mind which deliberately solves the problem set by these “data.” It is rather a problem of how to secure the best use of resources known to any of the members of society, for ends whose relative importance only these individuals know. Or, to put it briefly, it is a problem of the utilization of knowledge which is not given to anyone in its totality. “

            Knowledge of resource availability and preferences is highly local and always changing and attempts to centrally plan economic decisions can’t take that local knowledge properly into account.

            Determining whether an action increases or decreases a person’s utility, including comparing utility between people, isn’t difficult to do in broad stokes. We evolved to have to mental tools to do just that.

            I think it’s very difficult in broad strokes (depending on how you define broad strokes). Imagine designating someone else to make all decisions for you, even someone that knows you very well and loves you. They would decide where you live, what education you pursue, what job to take, what entertainment to watch and food to consume, when to go to bed and when to wake up, what consumer goods to buy and when, how much to save and invest or spend, what friends to make, etc. Even assuming they had your best interest at heart, how well do you think your utility would be compared to you making your own decisions?

          • Guy in TN says:

            Knowledge of resource availability and preferences is highly local and always changing and attempts to centrally plan economic decisions can’t take that local knowledge properly into account.

            Its true that knowledge of the preferences and situation of other people is imperfect. But there’s no need to design a system that pretends like we know nothing, when we know more than nothing. Its a false dilemma to have to choose between a “rational economic order” where all actions are planned by a central information-clearing organization, and a totally decentralized system.

            (Further, I don’t think a completely decentralized system is even theoretically possible. Given scarce resources, humans are inevitably going to have to come in contact with each other to try to figure out who gets to use what. We can use coin-flips, or we can try to use our rationality, including the messy bits where we try to maximize utility.)

            I think it’s very difficult in broad strokes. Imagine designating someone else to make all decisions for you,

            When I said its easy to do in “broad strokes”, I meant rather the opposite of “its possible to determine how all decisions effect another person’s utility”. Instead, I meant its possible to infer utility changes for very basic situations. You can infer a person with a bleeding wound is probably having a utility-loss. Or that someone who laughing and smiling because they are eating a delicious food, is having a utility-gain.

            More subtle or high-variance questions such as “how will watching this movie effect this person’s utility” are the sort we definitely can’t infer well.

        • idontknow131647093 says:

          And earlier, in response to me asking you to clarify whether pollution that also effects the polluter would qualify as an externality, you agreed, and said:

          Your statements are in direct contradiction. Am I misunderstanding you?

          Yes because you don’t understand that there are 2 costs we are discussing.

          Cost 1: Polluter polluting himself (internal cost) > This is the same cost you are talking about when you are saying no one else can cut down the tree or eat the ice cream.
          Cost 2: Polluter polluting everyone else (externality)

          • Guy in TN says:

            Cost 1: Polluter polluting himself (internal cost) > This is the same cost you are talking about when you are saying no one else can cut down the tree or eat the ice cream.
            Cost 2: Polluter polluting everyone else (externality)

            Yes, but there are two costs in the tree-cutting example as well.

            Cost 1: The cost to the property owner in future use of the tree (internal cost)

            Cost 2: The loss to non-property owners in future use of the tree (externality)

          • idontknow131647093 says:

            In your example cost 1 and 2 are the same. You are just phrasing them differently to create an artificial difference.

          • Guy in TN says:

            The difference between Person A and Person B is not artificial.

          • idontknow131647093 says:

            Yes it is because the cost is not to person A or person B in your example, it is a cost to the world. The entire world cannot harvest the tree anymore.

            If the fact is that person A, who owns the tree does not value the tree as highly as person B person B can purchase the tree, thereby eliminating the potential loss. In the case that person B cannot purchase the tree because of various reasons those things that get in the way are called transaction costs.

          • Guy in TN says:

            Yes it is because the cost is not to person A or person B in your example, it is a cost to the world. The entire world cannot harvest the tree anymore.

            Yes, its a cost on the entire world. Exactly what I’ve been saying. Hence, its an externality.

            If the fact is that person A, who owns the tree does not value the tree as highly as person B person B can purchase the tree, thereby eliminating the potential loss.

            Changing owners doesn’t eliminate the externality, it just shifts who it falls on. If A sells to B, then A experiences an externality when the tree is cut.

          • Guy in TN says:

            That we can create a contest where the highest bidder gets to avoid the externality (a market), doesn’t mean that everyone who isn’t the highest bidder suffers no loss.

            I doubt you would agree to that, if applied to the pollution example.

          • idontknow131647093 says:

            No he doesn’t because he has realized its value in it by selling it.

            Look at it this way: If an activity is efficient (which I will stipulate is theoretical, but such is all kinds of economic discussions), the benefits to the property holder exceed the costs to the owner himself + all summed externalities.

            So, I am owner A who owns land with tree 1 and I am the man who will get the greatest benefit possible out of the tree because I am a master craftsman and I create 10 amazing chairs from this tree. I can sell these chairs for $100,000. Lets say the next best use of the tree was only $50,000. I have created $50,000 for the world. However, lets also say chopping down the tree caused $100 of runoff damage to 10 people. We are down to $49k.

            Under your system no use could be efficient. That is because even though I am generating $50k more than any other user of the tree, there are 100 people who could generate $50k from the tree, so magically this is a -$4.9 million transaction (even ignoring the other billions of people).

            One of these calculations is an externality. The other is a nullity.

          • Guy in TN says:

            Under your system no use could be efficient. That is because even though I am generating $50k more than any other user of the tree, there are 100 people who could generate $50k from the tree, so magically this is a -$4.9 million transaction

            The utility of potentially gaining access to the tree isn’t the same as the utility of actually accessing the tree. The price of lottery tickets is less than the lottery payout. Very often, I would guess the utility of “potential access”, even multiplied by every other person in the world, is less than the utility gained from the owner harvesting it.

            That one person wins a utility gain, doesn’t mean he isn’t imposing a utility-loss on everyone else, still.

            the benefits to the property holder exceed the costs to the owner himself + all summed externalities.

            So you are saying yes, it is an externality, but the benefits to the property owner must be greater than the losses to the people he is imposing the externality on, otherwise the people would just buy the right to cut the tree themselves.

            In other words, Coase theorem. Which is utterly inapplicable to this discussion, due to the fact that dollars≠utility.

          • idontknow131647093 says:

            Actually, the Coase Theorem is very applicable to externalities.

            One of the big revelations of Coase is that if transaction costs are 0, those being subject to the externality will prevent it from happening if it is inefficient.

            One of the revelations of Arthur Pigou is that because transaction costs never are zero, sometimes a Pigouvian tax is the best way to deal with externalities.

          • Guy in TN says:

            Its applicable if we are strictly talking about economic value. But if we are talking about utility (and I thought we were- see my first post in this thread), it is useless. Because dollars are not interchangeable with utility.

            And I mean, if you really agree with Coase theorem, you also can’t say that taxing the natural resources is inefficient. Because if it was, then the people harmed by the tax would just buy the right to tax from the state.

            That no one buys this right, is evidence that no one values it as much as the state does. So the state retaining the right to tax is the most efficient outcome.

          • idontknow131647093 says:

            Oh I disagree 100%. People are always bidding to control the state. The problem is you cant control the state only for your own benefit without losing control.

            I think this discussion has reached its natural endpoint because you simply don’t accept my definitions which I think are standard and necessary for making sense of the world.

          • Guy in TN says:

            I knew “utils=dollars” was at the bottom of this. It always is.

          • Guy in TN says:

            Does your definition of “externality” relate only to changes in economic value, and not to changes in utility? That may be where we’ve started talking past eachother.

          • CatCube says:

            @Guy in TN

            “Externality” is an economic term–it was invented to refer to situations where an actor got some sort of economic benefit to himself by imposing a cost on others. As referred to above, an electricity plant that pollutes. The internalized costs of the plant are all of the stuff the plant owner had to pay for to turn the turbines. The person selling coal to the plant loses the opportunity cost of using the coal himself or selling it to somebody else, but in recompense he gets the money from the plant owner. The external costs are the stuff that the plant owner needs to do to turn the turbines that imposes an economic cost somebody else–here, the air pollution ruins other peoples’ property, interferes with their breathing, etc.

            I don’t know that there’s a coherent way to use the term in the way that you’re doing. If you’re concerned about something that doesn’t have an economic value, then why use the economic term?

            I had a longer post that yesterday that I ended up not posting because it was duplicative of points that others made, but I’ll ask the central question here (using the tree example):

            You’re saying that there’s an externality on you (and everybody else) from somebody cutting down a tree; that is, they’ve deprived you of some money in some way by somebody cutting down a tree they own. What flow of money are you expecting to make you not deprived? If the answer is “I’m not talking about money,” then “externality” isn’t the right word.

          • Guy in TN says:

            @Catcube
            I will make note to be more precise with my language next time. I wish there was a word for “action that imposes a utility-reduction on a third party”.

          • @Guy in TN:

            I suggest “utility externality.”

            I haven’t been following most of this thread, because it looked as though explaining why I thought your arguments were wrong would involve explaining a bunch of basic price theory, and pointing you at my webbed text didn’t seem either courteous or useful.

            But the claim that a pecuniary externality, which is properly ignored in economic efficiency arguments, might be a net externality (positive or negative) in utility terms seems to me entirely noncontroversial.

            I become a doctor. That pushes the price of a doctor’s visit down by a cent. That is a negative externality for all the other doctors, exactly balanced by a positive externality for their patients–the standard pecuniary externality.

            But it might be the case that the doctors have a lower marginal utility of income than the patients, in which case it is a positive externality measured in utiles rather than dollars.

            If you are curious about why economists do their analysis in dollars rather than utiles, the discussion in the chapter on efficiency in my Price Theory might be helpful. The most relevant part starts with the subtitle “PARETIAN AND MARSHALLIAN EFFICIENCY.”

          • Guy in TN says:

            Thanks for the input, David. I remember us having a discussion that mirrored much of this back in the spring.

            I need to be careful not to be so sloppy, in mixing economic and ethical terminology.

      • helloo says:

        I’m not sure if we should be controlling for ALL externality –
        One edge example of negative externality is pecuniary externality, or market changes from an action.

        It’s not all straightforwards though-
        Do you consider advertisement to be a negative externality towards other brands of a product due to the substitution effect?
        What about the food price increase when a lot of corn went to produce ethanol?

        • 10240 says:

          IMO no. Here is an (I think) exact way to define it:
          If a (legal or natural) person A takes some action(s), and that leaves no one (except possibly A) worse off than no action, then there is no negative externality.
          If a set of people A, B, C… take some action(s) that are voluntary on the part of all of them, and no one (except possibly some people in the set) are worse off than if no action was taken, there is still no negative externality.

          It could be assumed that Burger King creates a negative externality on McDonald’s when it takes away some of the latter’s customers. But take the set of persons which consists of Burger King and its customers. Burger King’s customers voluntarily choose to eat at Burger King, and that doesn’t make McDonald’s any worse off than if they choose to eat at neither restaurant chain, something they would have every right to do. We are supposed to compare the situation where they eat at Burger King with the situation where no action is taken (i.e. they eat at neither chain), not with one where they take a different, voluntary action (e.g. they eat at McDonald’s). It’s irrelevant if Burger King’s advertising makes people choose it over McDonald’s as long as it’s a voluntary choice on their part.

          Food and ethanol is the same story with sellers and buyers flipped: the set of corn producers, people who produce ethanol from corn, and ethanol buyers creates no negative externality, as they don’t make anyone worse off than if they didn’t produce corn for any purpose.

        • helloo says:

          Sorry, I didn’t mean to make it a definition matchup-

          Pecuniary externality is very much defined as an externality but even then, definitions often note it’s not a “real” or technological externality. https://encyclopedia.thefreedictionary.com/pecuniary+externality

          The question is that should we even count it as something to be considered especially for the question of taxation.
          I thought it was obvious that most people would not think so, thus the technical definition of an externality not that important.

          So the question then becomes at what point of causation should we discuss cost or benefit.

          • 10240 says:

            I agree that pecuniary externalities shouldn’t be taxed (I didn’t know the term), while “real” externalities should. I accept it if pecuniary externalities are usually defined as externalities, I was just trying to come up with a more narrow definition of “real” externality that are the only ones that should be taxed, and argue why pecuniary externalities shouldn’t be considered a wrong against the person they affect.

            Another precise way to define pecuniary externality (that is not “real”, or at least it’s fair game so not taxable) is that it’s mediated by one’s influencing the voluntary decisions of others. (E.g. I have the right to freely decide if I eat at McDonald’s or not, and nobody should be taxed for influencing my decision in a non-coercive way.)

            While I agree that most people wouldn’t want to tax pecuniary externalities in a systematic way, many people take issue with cases of pecuniary externalities time to time (e.g. protectionism), and use them as a justification for regulation, while I disagree with such justification.

            A more general issue with defining externalities is what it means that “A hurts B”. I would define it as “A does something that makes B worse off than if A didn’t exist (or at least didn’t interact with B even indirectly). I guess most people would agree with this definition when stated explicitly. Yet, in practice, many people also use it to mean “A makes B worse off than before”. (E.g. when a company lays off workers, people often say that it “hurts” workers, even though it doesn’t make them worse off than if it didn’t exist, just worse off than they are used to.) This issue interacts with the issue of pecuniary externalities, e.g. corn ethanol producers do make food buyers worse off than if they never existed, but the set of corn ethanol producers and corn producers don’t, they just food buyers worse off than if corn ethanol was not produced.

    • Plumber says:

      Raise corporate taxes a lot, give tax deductions for wages paid to American citizens up to the median wage (so no extra deductions for wages paid over the median wage), thus a tax incentive to have four employees are $50,000 per year rather than one at $200,000.

      Hopefully to increase employment and to bring back the great compression.

      • RalMirrorAd says:

        Are corporate taxes not already levied on *net* income rather than gross? How would the deduction work? For a company like, say, Walmart, the sum total of their salaries capped at 50K annually would likely exceed that company’s net income by a significant amount.

        It seems like a simpler solution would be to just have a more graduated individual income tax rate and evaluate inequality on a post-tax post-transfer basis.

        • SamChevre says:

          The tax deduction could be equal to some small fraction of salaries. I’ve thought of a similar system, with credit for the owner’s proportion of Social Security tax–for the same reason, to incent more lower-paid workers.

      • Brad says:

        About 6 million people looked for work in the last four weeks but are not working. That’s in a country of 325 million people. Just how low do you think that number should be?

        • John Schilling says:

          I see what you did there. So do too many other people for you to get away with it any more. How many people are not working, and did not look for work in the last four month because after looking for work the month before that and the month before that and the month before that and the month before that and the month before that and the month before that, they finally got the message that our society doesn’t want them to be working no matter how much they want to work?

          None of the official unemployment figures, actually measure what we mean when we colloquially say “unemployment”. Sometimes the right thing to do is suck it up and use the imperfect measure because it’s the best you’ve got, but in this case the disconnect is large enough that I’m not sure this works any more. We can use our intuitive sense for how bad a problem unemployment is, and we can see about trying to actually measure unemployment in the future, but we are severely handicapped in discussing the subject quantitatively at present because the people we trusted to gather data for us in the pas got lazy and botched the job.

          • meh says:

            participation rate?

          • The Nybbler says:

            How many people are not working, and did not look for work in the last four month because after looking for work the month before that and the month before that and the month before that and the month before that and the month before that and the month before that, they finally got the message that our society doesn’t want them to be working no matter how much they want to work?

            383,000. Or 1.6 million if you count all “marginally attached” members of the workforce and not just “discouraged workers”, which seems to fit your description.

          • Brad says:

            Oh come off it. To be counted in U3 you need to have undertaken one active job search activity in the past four weeks. An active job search activity can be done in well under an hour and from a cell phone or computer. To be counted in U6 you only need to have undertaken such activity in the past year. Less than one hour per year.

            “Society doesn’t want them to be working”, give me a break. They want to be working in the same sense that I want to win the lottery (N.B. I don’t buy tickets.)

          • A Definite Beta Guy says:

            I think at the macroeconomic level, the biggest gripe with unemployment numbers is low wage inflation. But my Twitter said that just crossed 3% last quarter or something, so that strongly indicates a strong economy.

            It seems like it is a strong economy because we are having more difficulty hiring people at all positions. I can tell you it is a LOT better than 2009.

          • HeelBearCub says:

            Real wage growth is continuing to be anemic. I don’t know that we are seeing any real signs of a hot job market.

          • Aapje says:

            @Brad

            Being rejected a lot is psychologically painful if not damaging. You can see people looking for acceptance in many ways, in society, because acceptance is an important need for most people. You might want to develop a bit more empathy for people who get looked down upon by society for not having work and then get turned down again and again. Studies suggest that being unemployed is literally sickening.

            Looking for jobs is also different from playing the lottery. When playing the lottery, paying for the tickets is part of playing the lottery and something that you have to keep doing. That is not the case for looking for work. Once you get hired, you don’t have to keep applying with very high risk of rejection to keep your job. So it makes a lot more sense to argue that someone is very willing to work, but can’t bring himself to look for work anymore, than to argue that someone is very willing to play the lottery, but can’t bring himself to buy a ticket.

            Imagine that I present someone with the opportunity to save their partner from death, but only in return for killing their children. If they turn me down, is the logical conclusion that they didn’t care about the life of their partner or should be entertain the possibility that the sacrifice we ask may be very high?

            PS. Note that this sacrifice is also much higher for some than for others and it’s easy to underestimate this for others, when you yourself are highly desirable as an employee.

            @HeelBearCub

            Perhaps a large number of jobs are marginal in how valuable they are to the employers, so people do get employed, but if the supply of workers is restricted, employers rather leave the position unfilled than to raise the salary.

          • John Schilling says:

            Oh come off it. To be counted in U3 you need to have undertaken one active job search activity in the past four weeks. An active job search activity can be done in well under an hour and from a cell phone or computer.

            Why would any reasonable person ever do that?

            If an unemployed person believes that the probability of success in one hour of job-hunting is sufficient to justify devoting that hour, they are going to spend many hours hunting for a job. The same logic that says to spend one hour on a job search, applies to the next hour and the hour after that and then an hour sometime tomorrow and so on, until they find a job or until they are spending enough time on it that the marginal value of a unit of time has shifted substantially.

            If an unemployed person believes that the probability of success in one hour of job-hunting is not sufficient to justify devoting that hour, then they are going to spend zero hours looking for a job. None whatsoever. And rationally so. There is no point at which it is rational to spend “under an hour” doing a cold job search and then giving up.

            Yes, they can move from one statistical category to another with an hour’s work, but what’s in it for them? Once their eligibility for short-term unemployment insurance has expired, there is no benefit to meeting the formal requirement for U3. And if they have decided on long-term disability as their optimal strategy, then getting caught looking for (or worse, finding) a job, is actively harmful to them.

            We’ve been talking elsewhere about how political polling is becoming increasingly unreliable as people refuse to answer their phones and talk to pollsters for five minutes. And here you are, expecting us to trust statistics on the basis that it only takes an hour’s effort for each unemployed person to arrange to be sorted into the most virtuous but otherwise unrewarding statistical bin?

          • Plumber says:

            @HeelBearCub @

            Real wage growth is continuing to be anemic….”

            Thanks for the link.

          • Dan L says:

            @ John Schilling:

            We’ve been talking elsewhere about how political polling is becoming increasingly unreliable as people refuse to answer their phones and talk to pollsters for five minutes. And here you are, expecting us to trust statistics on the basis that it only takes an hour’s effort for each unemployed person to arrange to be sorted into the most virtuous but otherwise unrewarding statistical bin?

            Point of order: there is a long and fraught inferential chain between “response rates are at all-time lows” and “polling is especially unreliable”. I’ll echo others here – Disaffected Workers seems to be exactly the category you’re speaking of, it’s been tracked for half a century, and I don’t see any good reason to believe it’s an order of magnitude larger than reported.

          • John Schilling says:

            A: It’s going to be somewhere between “discouraged workers” and “marginally attached”, because believing you’re never going to find a decent job is a big factor in deciding to become your senile mother’s live-in caretaker. Or just start making babies yourself. And given the current incentives, we ought to be including some unknown number of “long-term disabled” as well.

            B: I don’t think this number is likely to be literally an order of magnitude off, but the errors could still be substantial – particularly since we’re not sure where in the range of 506,000 to 1.5 million the BLS even thinks the number should be.

            C: It doesn’t matter how well we track discouraged workers or marginally attached workers if we bury those numbers in the fine print and only talk about the nominal “actively seeking work” unemployment rate as evidence that there is no problem. Which is what Brad did to start this digression.

            Whenever someone uses just the “actively seeking work” number as evidence that unemployment is a small problem and/or is getting better, then there needs to be pushback. Every time, because it’s wrong every time. This time, I decided to take that on. Next time, maybe I’ll see what sort of nuance and clarity you provide.

          • Brad says:

            there needs to be pushback

            Does completely unconvincing pushback count?

            U3 is a perfectly fine statistic to count people that actually want to work but aren’t. Your “society doesn’t want them to be working” is stridently nonsensical.

          • John Schilling says:

            Your “society doesn’t want them to be working” is stridently nonsensical.

            You think it is nonsense that there are a significant number of Americans who sincerely believe that society doesn’t want them to be working, or something along those lines?

            It makes no difference whether their belief is true, and I have made no regard in that respect. If they exist, then they won’t show up in U3 whether their belief is true or not. And we can disagree as to how many of them there are, but I’m pretty sure they exist and I’m pretty sure that’s not nonsense.

            Now go estimate their numbers, and show your work.

          • Brad says:

            Surely if the rate of mental illness is increasing in the US that’s an issue. It’s just one that has nothing at all to do with the conversation at hand.

          • Dan L says:

            @ John Schilling:

            The conjunction of my availability and interest is thin enough that I wouldn’t count on me being a regular source of pushback. But if I was going to sketch out a hierarchy of statistical virtue, then a key difference between valuable criticism and noise would be the extent to which an argument engages with existing data. That’s relevant here, because as far as this goes –

            If they exist, then they won’t show up in U3 whether their belief is true or not. And we can disagree as to how many of them there are, but I’m pretty sure they exist and I’m pretty sure that’s not nonsense.

            Now go estimate their numbers, and show your work.

            – there’ve already been several suggestions in this thread. Discouraged workers, marginally attached, and total labor force participation rates are all real metrics. The uncertainty of their use in this thread appears to be definitional, a result of not knowing what group you’re gesturing at, rather than error in the data itself. (And as far as methodological errors go, the first few paragraphs of this come to mind.)

            I can see the trend line of one of those three being an effective argument for the case you’re making. But making that argument requires indulging in a little mistake theory.

        • Plumber says:

          @Brad

          “…Just how low do you think that number should be?”

          I’d like the unemployment rate low enough that the upward pressure on wages was enough that men’s median hourly adjusted for inflation wages again reach their 1973 peak, if women’s could reach that peak as well that would be a nice bonus, and I’d like the labor share of GNP to again reach the 1970 peak.

          • Brad says:

            Didn’t we already discuss wages vs compensation? Why are you back to banging on about wages?

          • Plumber says:

            @Brad

            Didn’t we already discuss wages vs compensation? Why are you back to banging on about wages?

            How does more money going to “compensation” (the medical-industrial complex) negate the fact that men’s hourly wages adjusted for inflation are lower and more importantly the share of GNP going to labor is lower?

          • Mr. Doolittle says:

            Plumber,

            Although it’s nice when an economy is doing well enough for wages to rise, we have to be aware of when the economic conditions start pushing certain fields outside of a profitable return and they cut back or even shut down.

            If total compensation has not declined (and it seems that you already agree with Brad on that), then higher wages means an increasing share of a companies profits becoming an expense – going to the employee. That GNP in the hands of labor is lower does not mean that a companies revenue spent on that employee is lower, so at the individual firm level, that’s irrelevant.

            So if an increasing amount of revenue (otherwise profits) going to the employee means that the margin gets too small, or especially if it becomes negative, then the company will hire fewer people and begin to slow down the economy. The right balance is quite tricky, but it cannot be true that higher wages are always the better result. Some industries are better insulated than others, of course.

          • Brad says:

            Because it amounts to double counting medical inflation. That’s already built in to the inflation number. The fact of medical inflation doesn’t change one bit of it is paid by employers on employee’s behalf or paid by employees out of their cash wages.

            Also, I have no idea why you think specifically men’s wages are the critical issue. I probably don’t want to know.

          • 10240 says:

            Also, I have no idea why you think specifically men’s wages are the critical issue. I probably don’t want to know.

            I would guess because more women work full-time now, and they work in various high-income professions they rarely did before, which implies that we should expect their wages to have increased even if employees’ situation vis-a-vis capitalists (or even absolutely) has generally deteriorated. Whereas men’s work hours and job choices didn’t undergo quite as much change, so their salaries should more accurately reflect any changes in employees’ fortunes.

          • Plumber says:

            @Brad,

            I draw the comparison with men’s wages from the early 1970’s to have an ‘apples-to-apples” comparison, but you missed

            more importantly the share of GNP going to labor is lower

            and it’s instructive to see where income is going instead, as well as the well documented explosion in CEO pay and “stock options”, there’s: Financial “services”, which have grown both in their share of the total economy, and in their share of the total profits in the U.S.A.

    • Hoopyfreud says:

      Framework: the ideal role of a corporate tax is to eliminate tax-free “pseudo-income” for the owner(s). Money not used for individuals’ personal benefit should be untaxed. I’m generally skeptical of income tax, but I’ll assume that we’re working from the assumption that income taxes won’t be eliminated and that the tax burden on the rich should be commensurate with their increase in wealth. I also want to avoid disincentivizing co-ops and other businesses with locally-distributed ownership models, because I like them a lot.

      Proposal: 0% tax on revenues for corporations, and a tax on capital gains for corporations (including land, equipment (lmao like this appreciates), and investments) exactly equal to the personal rate per owner (basically, make the owners pay for it, weighted by ownership %). Other Pigouvian taxes also equal to the personal rate per owner (if we actually implement them). Money spent (not saved, or, in the case of charities, donated) that is not used to pay salary or benefits or purchase capital is taxed as income to the owners.

      The last stipulation is the one I think has perverse incentives baked into it, but I’m not enough of a finance nerd to see them, and I’m almost at work. Someone please tell me what’s wrong with this.

      • SamChevre says:

        I think there’s a problem with the last stipulation: what you don’t want is the corporation serving as a giant tax-deferral mechanism, and this seems like you could just save cash in the corporation, and it would be untaxed until paid out to the owners.

        This would make the company like an unlimited Roth IRA for its owners.

        • baconbits9 says:

          And what is the problem here?

        • Hoopyfreud says:

          This is where capital gains kicks in; the owners are still taxed on the corp’s investment income. If they just store cash in the corp, running expenses and inflation will whittle away at it until it’s withdrawn, at which point it’s a tax-deferred bank account that they have withdrawal limits on; I’m OK with this.

          Re: your other point on durable goods and land – I’m fine with a depeciation schedule for goods (which also rewards companies that take good care of their equipment). I don’t like the idea of holding land as an investment, and I think land value assessments and property tax increases should be more frequent, uncapped, and more open/less prone to corruption, but that’s another topic.

          Also, re: expense – this hopefully won’t be more expensive than declaring investment income already is.

      • SamChevre says:

        My critique of the proposal: taxing the increase in value of durable goods and land will be very tricky to administer. How much is a 10-year-old milling machine worth?

        It seems like compliance would be very expensive. Would your goals be equally-well-met by just having a depreciation schedule for the durable goods, and letting the local property tax handle land and buildings?

    • mrjeremyfade says:

      It will be interesting to see if there are any really good ideas. Corporate taxes are problematic in many ways. I see some problems that crop up over and over.

      Different treatment of the same transactions across legal jurisdictions.
      Incentives created to redefine income or whatever is being taxed.
      Resources devoted to tax minimization.
      There are edges between corporations and non corporations that can be controlled by the same people.
      I’m less sure about this, but my feeling is that some of the very smartest people in investment banking and law work on tax and generally out think the people who write the rules and enforce them. I’m not sure this game is that valuable socially.

      As a general principle, lower rates and a broader base will be less distorting.

      So if I play your game, I might very tentatively suggest a goal of raising tax revenue with minimal distortion. And it will be a very low rate revenue tax. Revenue broadly defined, GAAP like, cash basis perhaps for smaller firms.

      But I don’t know, I think it could radically discourage corporate spending. Maybe the transition cost would not be worth the increased efficiency.

    • massivefocusedinaction says:

      0% corporate income tax, it’s far to easy to shift income internationally for a corporate income tax to work effectively.

      Replace it with a tax on land, intellectual property, and financial markets.

      Land tax should be designed to be as Georgian as possible (my ideal is owner publicly lists the value which is the tax basis and represents a legal offer to sell).

      Intellectual property is an annual tax that rises exponentially, with the exponents and coefficients selected to produce a very low cost for 10-20 years, rapidly rising costs to make up the balance of 30 years and then ultra high increases after that (the idea would be to make a valuable property like Star Wars or Pokémon uneconomic to retain copyright by 40 years). That way cheap stuff is in the public domain in 20 years, and everything is in the public domain in 40 or less, and the good stuff pays most of the cost.

      Financial markets get an annual levy based on market value (similar to but bigger than the SEC’s filing fees), and maybe a tiny transaction tax (goal would be for this to be on par with current exchange fees).

      All three are intended to tax what appear to be the main US competitive advantages (ultra high income consumer markets, intellectual property, and financial markets) without regard to the corporate parent’s location.

      • ana53294 says:

        Would you also tax trademarks exponentially?

        The thing with IP is, it is very easy to shift property of IP abroad.

        • vV_Vv says:

          The thing with IP is, it is very easy to shift property of IP abroad.

          Then it’s not protected at home anymore.

        • massivefocusedinaction says:

          I’m not sure about trademarks, they’d need more protection than the 40 year goal of copyrighted works, but I’m not sure they deserve permanent protection (if they remain in use).

          The goal of my suggested change is to tie access to American markets to accepting American legal treatment, so a foreign company could keep their IP under whatever restricted regime they wish, but as soon as they sell in the US, they follow US laws as though the work was created in the US. Their foreign copyrights aren’t very meaningful if US citizens can freely use the content sold in the US.

      • RalMirrorAd says:

        IP strikes me as tricky since with certain obvious exceptions like maybe star wars a legal argument can always be made that a product is being incrementally changed and therefore isn’t the same product and so shouldn’t have the patent renewed [and so is never taxed] This becomes an issue if the product is different enough from the original to escape taxation forever but similar enough that consumers don’t benefit from products falling into the public domain because would-be producers fear copyright lawfare.

        Not an expert on this kind of law so I could be worrying over nothing.

        • vV_Vv says:

          I suppose the copyright for each individual item (character, location, named thing) starts when its first mentioned.

        • massivefocusedinaction says:

          I’m no IP lawyer, but I believe the incremental changes become a new product that receives its own protection, but don’t protect the earlier works. For example, the character, Sherlock Holmes, became freely usable when A Study in Scarlet’s copyright protection ended, not when The Case-Book of Sherlock Holmes left copyright many years later.

          I have no issue with others using public domain concepts and getting a new copyright that protects their work, but doesn’t impact other derivative uses of the original.

          In the Star Wars example, if Episode IV left protection, Han, Leia, Luke, and Darth Vader and concepts like the Force would be freely usable and the whole film could be distributed and shown freely, but Empire and later films would remain protected until their future terms ended.

    • Mark V Anderson says:

      I think we should get rid of corporate taxes altogether, because it is too distortionary and complicated.

      All business income should be passed through to the shareholders. We already do this for partnerships (and S corporations). IT is true that this will result in a whole lot of pass-throughs for some corporations which have millions of shareholders. But this is still simpler than having a separate tax for corporations and individuals. I think only individuals should pay tax, not artificial entities. How do you decide what is the corporate rate? Having a corporate tax just complicates understanding who is paying the tax.

      • SamChevre says:

        I agree. My proposal is an attempt to implement this is an easy-to-administer way. I wonder if you think my proposal actually would achieve this goal, though.

        • Mark V Anderson says:

          @Sam.
          Your latest comment above is to tax corporations on their profits that they do not distribute as dividends. This essentially means that dividends are deductible to the corporation. This does make sense, since interest expense is already deductible, and it is less distortionary to make debt capital and equity capital tax neutral. And I think that we have far too much debt in the US, which IMO was one of the major causes of the Great Recession, so making equity capital tax neutral to debt capital is a good idea.

          But my preference is to do away with corporate taxes altogether, since it is impossible to determine which persons are really paying the tax. Your method does not do that.

      • Brad says:

        Again, this just gives up on a lot of foreign revenue (foreign owners of domestic companies and foreign companies that do business in the US). It would be one thing if people would acknowledge this and say they are okay with it, but people seem to want to implicitly claim that it is mostly just a simplification, when it’s a big change to the tax base.

        • SamChevre says:

          I’m OK with it under my plan, but it’s much more limited under my plan than the alternatives–under my plan, only the dividends are untaxed, and foreign companies that do business in the US pay tax on a proportion of earnings proportionate to US sales unless paid out as dividends.

          • Brad says:

            I think your tax plan is pretty decent, though while I like forcing companies to report the same things to shareholders they do to the government, I do worry about handing off tax law to a private organization.

            But why not withhold taxes at some rate from dividends going to entities that can’t show they file US tax returns?

        • Mark V Anderson says:

          Again, this just gives up on a lot of foreign revenue (foreign owners of domestic companies and foreign companies that do business in the US).

          Again? This is the first time I’ve heard this objection. I hadn’t really thought of this issue, but I don’t think it is particularly difficult to resolve. One could require corporations to pay the taxes on behalf of foreign shareholders that are not expected to file returns. This could be set up as an add-on to our already existing system of foreign withholding. The only new part is requiring corps to pay out of their own funds instead of part of the payment to the foreigners. I would like to get rid of corporate taxes because I don’t think artificial entities should pay tax; but I don’t consider that an issue when the corps are paying on behalf of individuals.

          By the way, Brad, elsewhere you mention withholding tax on dividends to foreigners. You do realize we already do exactly that? Are you interpreting Sam as saying we should stop? I think he is saying that corps shouldn’t pay tax on the amount they distribute BECAUSE the recipient of the dividends also pays taxes. This is totally consistent with our current foreign withholding regime.

          • Brad says:

            Again? This is the first time I’ve heard this objection.

            Sorry, I made a comment to the same effect in the last thread and it seemed to get ignored. It wasn’t to you though, so I shouldn’t have said that here.

            One could require corporations to pay the taxes on behalf of foreign shareholders that are not expected to file returns. This could be set up as an add-on to our already existing system of foreign withholding. The only new part is requiring corps to pay out of their own funds instead of part of the payment to the foreigners. I would like to get rid of corporate taxes because I don’t think artificial entities should pay tax; but I don’t consider that an issue when the corps are paying on behalf of individuals.

            At the maximum individual rate?

            I think that could work. I don’t have any especial attachment to corporate taxation qua corporate taxation, but I’m also not sure that this system is especially simpler. In order to calculate the attributed passthrough wouldn’t essentially the same exercise as corporations go through today need to happen?

            Also, how do you envision taxing non-domiciled corporations? Would every company with US owners need to calculate passthroughs for their US shareholders? Would foreign corporations owe any tax for themselves or their foreign shareholders if they had profit in the US.

            Are you interpreting Sam as saying we should stop?

            Yes, that’s how I read him. Could be wrong.

          • Mark V Anderson says:

            At the maximum individual rate?

            I don’t think it needs to be that. Currently most withholding is at 15%; I think that would mostly work. Yes, that is well below the highest rate, but I think the assumption is that even those at the highest rate would have lots of deductions, so it might well equalize. In any case, part of my ideas for simplification of our tax system is to have only one rate of tax, so most wage earners don’t have to file a return at all. SO under that method, which rate would not be an issue.

            I’m also not sure that this system is especially simpler.

            I think it’d be simpler to have just one kind of income tax, not two. Also, I think it is more fair to bill everyone at the same percent. It is impossible to even know what individuals are paying for with the corporate tax, because of widely disparate tax incidence methodologies.

            Also, how do you envision taxing non-domiciled corporations?

            We currently tax foreign corps that do business in the US. This would be no change from that.

        • 10240 says:

          I don’t think much of that tax is actually incident on foreigners.

          Capital is easy to move between countries, so the rate of return (after corporate tax but before dividend or capital gains tax) for a given risk must be the same everywhere at equilibrium (otherwise capital would move to places with higher returns until the rates of return equalize). Similarly, the (wholesale) market price of a good that can be easily transported between countries is similar everywhere.

          Now consider a small economy. It’s policies don’t significantly change the rates of return on the international capital markets, nor the worldwide market prices of goods. Thus, if you impose a corporate tax, that doesn’t significantly decrease return on capital in that country (after corporate tax but before dividend/capital gains tax), since it must be equal to the international rate both before and after the tax is imposed (after things come to an equilibrium, with some capital leaving the country). Likewise, if a (foreign or local-owned) company sells its products to foreign customers, it can’t pass the new tax on to them, as they won’t pay more than the market price. Thus the company will almost entirely pass the corporate tax on to its employees, and possibly to its customers if it sells its products in the country. A small country can’t impose a tax on foreigners and use that to significantly increase government revenues at the expense of foreigners (except if it holds a significant percentage of the world production of some good); the tax will almost entirely be incident on its residents, and it can be replaced with an income or sales tax. It can tax domestic capitalists though (whether they invest in the country or abroad) through dividend and capital gains taxes, as many people won’t move to the country with the lowest taxes.

          The US is a big economy, so taxing foreign companies may actually direct some foreign money into its budget, but much less than you might think. Foreign companies operating in the US are a small fraction of the world economy. Taxing them slightly reduces worldwide return on capital, and it may slightly increase the international market price of some goods they produce. The amount of money that actually flows from foreign companies in the US to the US government is at most as much as the decrease in returns of these companies, and the increase in their income from foreign sales. I think only a small fraction of the tax is actually incident on foreign companies in the US or their foreign customers, the rest is incident on Americans. Other effects from decreased returns and increased prices of some goods include transferring some money from capitalists to workers worldwide, as well as from consumers of certain goods to producers, but generally not from foreigners to Americans.

          DavidFriedman can probably link some books that discuss it in more detail, and correct me if I’m wrong.

  13. ing says:

    I’ve recently completed this writeup of my favorite D&D oneshot adventure:
    https://sites.google.com/view/the-orb-of-storms/home
    What do you think?

    • dndnrsn says:

      My immediate thought is that your definitions of “simulationist” and “narrativist” don’t really match the meaning of those terms found in GNS theory (although they probably match the “colloquial” meaning better – one of the problems with GNS theory seems to be that the terms don’t mean what most people guess they would mean).

      I disagree with your philosophy of GMing, for the same reason I disagree with John Wick’s GM advice (which often boils down to, use all sorts of tricks and cheating and so on to leverage the unique identification-with-character of RPGs to deliver a particular emotional experience): the average GM is, by definition, mediocre. Trying to give the players a good story, keep things fun, etc by “goosing” things, will as often (perhaps more often) than not lead to railroading. John Wick might be able to do this (he tells a lot of stories about what a superlative GM he is, at least, and his friends seem to agree) and maybe you are, but legitimizing it as a tool for general use is dangerous, for the same reason that a detective framing a criminal they know to be guilty but can’t prove it is dangerous. Maybe really good detectives only use this on people who are actually guilty, but if it becomes generally accepted GM advice, you’re going to have a lot of mediocre detectives framing a lot of innocent people. It’s a great way to turn an RPG into telling a story to the players, creating all sorts of bad incentives (if the players know they’re never going to face a combat they won’t win, and there’s a guarantee of victory because victory is fun and losing isn’t fun, they’ll get lazy and sloppy), and ultimately defeating the unique virtues of the format.

      I played in a long-running campaign with an at-best-mediocre newbie GM, and it became increasingly obvious that he was railroading us, rigging things so combats would never kill us, and ensuring that anything we did succeeded if it would advance his plot (anything that would mess with his plot was made impossible). By the end, we were phoning it in; it was obvious nothing we (the players) did mattered. Our PCs were the heroes, sure, but guaranteed victory isn’t really victory.

      Thoughts about the adventure:

      1. Brief intro explaining the backstory. Good. Long intros are bad, but just dropping the GM in is also bad, because it means they have to read the whole thing to figure out what’s going on.
      2. Good that it’s explicitly not read-aloud text. Read-aloud text is bad.
      3. Q&A is good.
      4. First encounter could maybe be better structured as information to the GM, to then be dealt out to the players; the way it’s formatted, a GM could accidentally surprise themselves.
      5. Why don’t people think werewolves are real? There’s elves and magic and stuff; “nobody believes in that fantastical creature” always seems a bit odd in fantastical worlds. Wouldn’t a ban on silver raise eyebrows? Why would anyone make silver weapons, if not for fear of werewolves?
      6. What happens if the PCs talk to the crazy guy and go off on a series of red herring wild goose chases?
      7. Related to my above disagreement on GMing philosophy: if you alter reality to keep the PCs from getting killed when they go off half-cocked and do something stupid, that’s guaranteeing that in the medium or long-term the players will reliably go off half-cocked, because a bad plan has no disadvantages compared to a good plan. Foolish PCs should die if that’s how things go down, because if you provide no punishment for being foolish, you’re guaranteed to have foolish PCs.
      8. “but anyone with religious skill feels the place as hollow.” Why? Religion skill is usually “you know about religious stuff” – I did religious studies in school, and I probably have proficiency in Knowledge (Religion) or whatever because of that, but I wouldn’t be able to walk into a church where there’s a secret temple of Baal in the basement and identify that something’s off. If it’s some specific cleric or paladin thing where they can feel bad mojo, sure, but you don’t want the PCs getting the idea that an academic skill is a Wrongness Detector.
      9. Why give the bad guys a good plan, unless the PCs take too long, in which case they have a bad plan? Give them a bad plan from the beginning, and add something to make it harder if the PCs take too long.
      10. The writing style is good at delivering information; I never found myself wondering “what is going on; who is that; when did that come in to the picture.” If you put this in .pdf format, I’d recommend double columns: much easier to read than single columns, for whatever reason.

      • ing2 says:

        Thanks for the feedback! Your criticism is insightful.

        I’ll admit that I haven’t actually read “GNS Theory”, so it’s probably bad for me to be borrowing their vocabulary without understanding it. Is there better terminology I could be using here?

        I do think you’re right that I’m optimizing for short-term fun at the expense of long-term campaign viability. I will think on this.

        • Hoopyfreud says:

          FWIW, I think this is perfectly reasonable for a one-shot, and I always talk to the players before running one to say, “look, I like doing improv GMing, but one-shots are fairly on-a-rail in terms of their expectations; if you don’t do what you’re expected to, we won’t reach a resolution tonight. Or ever.”

        • dndnrsn says:

          If I were trying to put words on it, I’d differentiate between a “let the dice fall where they may, the world is a real world” attitude and a “guided experience” attitude.

          If this is meant to be a one-shot, or an introductory adventure, I’d just make it clear to the players that the purpose is to get them accustomed to the system, and so on. Maybe even present it as a series of vignettes. The best “rigged” adventure I’ve ran was one where the action was mostly in flashbacks, where I’d tell the PCs what was going to happen, in broad terms – but that was specialized.

          In anything longer than that, I’d just stay away from the “the GM’s job is to make sure everyone has fun by rigging things” – in my experience, lacking a superlative GM (which most GMs are not; I am not a superlative GM), rigging things to make sure everyone has fun leads down a bad road.

          • Le Maistre Chat says:

            In anything longer than that, I’d just stay away from the “the GM’s job is to make sure everyone has fun by rigging things” – in my experience, lacking a superlative GM (which most GMs are not; I am not a superlative GM), rigging things to make sure everyone has fun leads down a bad road.

            In my late face-to-face 3.5 campaign, I was stuck doing this because one player, J, would ragequit over his character failing. I was unable to reach the point of saying “OK bye” due to not being able to recruit stable new players.
            This being 3.5, where the rules start to fall apart in the players’s disfavor if the sacred principle of Wealth By Level isn’t maintained (as a better player put it to me, “If you don’t drop enough treasure, any Challenge Rating is higher than you think it is because we don’t have the loot the CR system expects you to handle that monster with”), the economics of the world kept changing. First there were historically plausible treasure drops, but that was at least almost two orders of magnitude too low for WBL, so instead of enemies fighting encumbered with heavy sacks of gold, I started giving them the magic items within that budget that would be most useful to them… which led to J threatening to ragequit again if I didn’t edit the universe mid-combat so that only PC casters could craft Monk’s Belts. @_@

          • John Schilling says:

            You definitely have to either keep WBL or ignore CR in 3.5e/Pathfinder. But since there are already sound reasons to treat the CR system as loose guidelines at best, WBL can go take a flying leap as soon as you’ve got a better idea.

            Keeping your players lean and hungry is, IMO, almost always a better idea.

          • Hoopyfreud says:

            @John

            This dramatically exacerbates caster/martial disparity. BAB and most class features scale roughly equally to WBL number progression; WBL barely scales casters (comparatively) because spell progression outstrips its impact so badly.

            Also, throwing enemies at the party with (at least one of) antimagic, things that target fort or reflex saves, teleportation+grappling, or other “fuck you wizard” shenanigans doesn’t help. The fact that you’re trying to fight one player more than the others is transparent and frustrating, IME.

      • Deiseach says:

        I did religious studies in school, and I probably have proficiency in Knowledge (Religion) or whatever because of that, but I wouldn’t be able to walk into a church where there’s a secret temple of Baal in the basement and identify that something’s off.

        Ooooh, unless it was something like “Yeah, the iconography here is not right, somehow” or “They’re celebrating a what ritual? That’s nothing I’ve ever heard of!” or observing what on the face of it is a conventional ceremony but the rubrics are completely wrong, but agreed, that’s not the same thing as “hmm, my sense of the sacred is tingling, this place feels hollow”.

        • ing2 says:

          A principle of the Dresden Files books is that all humans can sense the presence of evil magic. Most people don’t recognize it as evil magic, but everyone can kind of get the feeling that something is wrong. I think the quote is: “the heebie jeebies are an evolved survival mechanism.”

          I’ve sort of integrated this into my worldview; I guess I’ve got sort of an implicit house rule that, if something sufficiently evil is happening sufficiently close to you, you’ll notice.

          I agree that this is not an official 5e rule, and I should revise my oneshot to be clearer there.

        • Jiro says:

          I think some of this is an artifact of the fact that you can’t have skills at too fine-grained a level in a game, because the more fine-grained you make the skills, the less efficient it is to buy them. You’d have to give people hundreds of points for skills and then let them buy “Christian religion”, “Western religions”, “Eastern religions”, “ancient religions”, “temples in ancient religions”, “carvings in temples in ancient religions” etc. separately. It’s too unwieldy to do that, so instead you have one “religion” skill which covers everything religious, even if nobody could actually know everything in the category.

      • Le Maistre Chat says:

        7. This is something I’ve been seeing playing D&D 5E: it’s almost baked into the rules that you can’t die for doing stupid things unless your party ignores your dying body for 3+ turns and you roll 1-9 on a 20-sider thrice.
        Right now I’m playing an Arkadian Moon Druid 5/Bear Totem Barbarian 3 in my best friend’s mythic Greek campaign, and I keep asking for deadlier encounters because my PC can tank for about 200 damage regardless of type before even taking advantage of the self-Cure while in animal form rule, while keeping squishier teammates alive by passing out Goodberries (a Level 1 spell) ahead of time.
        With only that modicum of planning, I’ve yet to see anything we can’t handle with me jumping into the biggest danger’s face with a cry of “he wanax Jenkins!

        • dndnrsn says:

          It seems to me like the intention in 3rd is to make it like CRPGs where party members generally don’t die unless the entire party goes down – either you manage to win the fight, and you’ve got a bunch of wounded PCs, or your last PC goes down, game over.

          It seems like this would create the problem of only having PCs die when there’s a TPK, though.

        • Nick says:

          In the Pathfinder Western I’m in, we nearly had a TPK last session. Our tank was down, our rogue couldn’t fight, I was grappled and almost out of health, and our bard was almost out of health and considering fleeing. If things had gone even a little worse, we’d all have died, with the possible exception of the bard. It was frustrating for story reasons, but gratifying in its own way to learn that our GM would throw serious encounters our way.

          also +1 for “he wanax Jenkins!” 😀

          • Le Maistre Chat says:

            It’s a very gratifying feeling, isn’t it?

            also +1 for “he wanax Jenkins!”

            Thanks. If I ever get to play in a medieval French campaign, it’ll be “Leroy [Charles/Louis]!”

        • Gazeboist says:

          To be fair, you’re playing one of the most durable builds available. The 5e Druid class is also pretty broken. Not quite to the degree of the 3.x Druid, but it’s more obvious because they got the balance right in other places.

          • Le Maistre Chat says:

            To be fair, you’re playing one of the most durable builds available. The 5e Druid class is also pretty broken.

            True. 90 extra hit points starting at Level 2 that you get back after a short rest are so hilariously broken that multiclassing into Barbarian for half damage from every type but psychic is almost a trap option. By the time the 180+72(human form) HP build activates at Level 5, you’re only 2 levels below where a straight Druid gets the ability to summon 8 Pixies, which is even more broken.
            For an even more broken strategy, get another player to make an Assassin Rogue (broken Damage Per Round option) and ask him to ride you to the enemy you’re tanking, then dismount out of range after attacking.

          • Jiro says:

            I have played a druid and haven’t found it broken at all.

            The key is that pretty much everything you can change into has a terrible armor class. The 90 extra hit points go away much faster than actually having 90 extra hit points would. Using them also requires that you commit to not casting any spells until you’re done, and many powerful creatures are huge in size and won’t fit in a dungeon corridor.

            You can try to make up for the armor class by using spells such as Barkskin, but they have concentration requirements and you’ll be getting into physical combat, so they won’t last.

    • Vorkon says:

      I just wanted to say that I definitely appreciated the bit about the conspiracy theorist thinking that the best way to kill werewolves was to gather everyone in the town square and vote on who was most likely to be a werewolf.

  14. albatross11 says:

    Glenn Greenwald had a really interesting tweet w.r.t. Bolsonaro’s win in Brazil, which I think applies very widely:

    When a ruling class fucks over enough of the population for a long enough period of time, they’re going to burn it down – out of desperation & anger – one way or the other. At some point, democracies are going to need to grapple with that or there will be Bolsonaros everywhere.

    • 10240 says:

      IMO this sort of sentiment is common, but it’s also vague and relatively useless because whether the ruling class is fucking over the population is extremely subjective (and even the term “ruling class” is loaded).

      • Le Maistre Chat says:

        Eh, it’s only loaded because the term “ruling class of a democracy” causes cognitive dissonance. There’s no denying its existence.

        • 10240 says:

          The “iron law of oligarchy” is some guy’s observation, not a law of nature. The term “oligarchy” is itself vague and loaded; it’s clear that politicians are a small number of people compared to the entire population, but if they need to do approximately what the people want in order to stay elected, is it an oligarchy?

          But, more importantly, when I say that a term is loaded, I don’t mean that what it describes doesn’t exist, but that it passes judgement on what it describes, and/or it contains implicit claims. (Compare: “follows one’s interests” is a neutral phrase; “selfish” means the same but it implies that this is wrong, which makes it a loaded word.) It’s clear that there are people who make decisions for the country (politicians, political advisors, high-level bureaucrats) but “ruling class” has stronger connotations that are, as you say, viewed negatively in a democracy. My impression is that the term is usually used with the implication that the ruling class is best viewed as a monolith, that differences between the mainstream parties are unimportant, and what matters is the interests of the “ruling class” as a whole, as contrasted with the interests of ordinary people. Perhaps also that the political class is sort of closed, at least in mainstream politics.

          Such claims may or may not be true, but they are clearly subjective, and predictions such as the one in the thread-starter comment hinge very much on how much of the population agrees with them.

    • Eugene Dawn says:

      I don’t want to dismiss the idea out of hand, but it really needs some definitions: who is the “ruling class”? What does it mean to vote to “burn it down”–is the implication that voters are deliberately voting for Bolsonaro in the expectation that he will destroy the country? And are we assuming that the voters who vote to burn it down are the ones who are fucked over, and not the ruling class, whoever that may be?
      Because my understanding is that support for Bolsonaro was concentrated among the middle and upper classes, and according to these polls support for Bolsonaro is basically monotonically increasing with education. That sounds to me like Bolsonaro’s support is stronger among those from whom the ruling class, whoever they may be, is drawn, than among those who are most likely to be fucked over.
      This isn’t necessarily inconsistent with Greenwald’s argument, but I think it really needs to be clarified before I’m convinced it’s an insightful point.

      • Le Maistre Chat says:

        my understanding is that support for Bolsonaro was concentrated among the middle and upper classes, and according to these polls support for Bolsonaro is basically monotonically increasing with education.

        That’s interesting, and makes the moniker “the Trump of the tropics” a huge misnomer.

        • Eugene Dawn says:

          I mean, I think Trump was more popular among people with higher incomes as well, though education is a pretty big difference. But don’t read too much into what I posted: with Trump partisanship, age, race, gender, geography and other factors mattered as much or more and I presume there’s similar factors that better predict Bolsonaro support than just income and education. The coalitions they mobilize might be similar along multiple axes, and differ only on education, for example.

      • j1000000 says:

        Yeah — I don’t use Twitter much anymore but went to Greenwald’s profile specifically today to see his reaction. I was surprised by the tweet albatross points out because I assumed he’d (pretty easily) frame this as the rich and powerful once again screwing over the poor, as I like you was under the impression that this was not a man popular among the impoverished.

      • dndnrsn says:

        Not speaking to the situation in Brazil, but more generally, right-wing populists appeal primarily to the lower-middle classes, don’t they? Eg, in Weimar Germany, the Nazis never had disproportionate support among the urban working class (who tended to back the SPD, with the KPD being a major vote-winner among the unemployed); their support was concentrated in the lower-middle classes.

        The Nazis are, of course, an extreme example, but anything that could broadly be called “right-wing populism” tends to appeal to people who have theirs, but feel that they might lose it, or are embittered because they feel they should be better off than they are (a significant chunk of the SS higher-ups, for example, were either junior officers or noncoms in WWI or were slightly too young to serve in WWI, got involved in Free Corps stuff, and then found themselves in a succession of not-especially-great jobs because of the bad years at the beginning and end of the 1920s – guys who would have had prospects were it not for having lost the war and some bad economic breaks).

        • idontknow131647093 says:

          Right wing populism is almost uniformly a reactionary sentiment. In this case it is in response to blatant corruption of the left wing regime and a failure to properly check it by the center-right faction.

          I can’t recall such a movement that wasn’t preceded by a left wing government that was perceived to be corrupt of incompetent (which isn’t wrong in this case) or an ascendant communist movement (the case of Weimar).

          • baconbits9 says:

            Wasn’t the Weimar republic run by an incompetent left wing government as well?

          • dndnrsn says:

            (Interesting historical note: the Nazis considered the mainstream conservatives to be “reactionaries” and set themselves against “reaction”; they considered themselves to be reactionaries)

            There was a lot more wrong in Weimar than vote increases for the KPD; the degree to which the communists can be described as “ascendant” is questionable.

          • HeelBearCub says:

            But, isn’t that that just the pendulum in action?

            Populism in general tends to be a reaction to previous failures. FDR is a reaction the failures of the laissez-faire conservatives of the 20s. Communism is a reaction to the the failings of the monarchy or ruling class…

            I’m not sure we need to really distinguish various forms of reactionary populism as being in reaction to anything specific ideologically.

          • Protagoras says:

            Wasn’t the Weimar republic run by an incompetent left wing government as well?

            Until 1930, the center-left had consistently led fragile coalitions (which included the center-right party), but after the 1930 election Hindenburg bypassed the Reichstag to appoint first the center-right Bruening and then the conservative von Papen. Also, things had generally seemed on an upward trajectory for Weimar before the great depression hit them. So that doesn’t seem like a very good description of the lead up to Hitler.

          • idontknow131647093 says:

            Populism in general tends to be a reaction to previous failures. FDR is a reaction the failures of the laissez-faire conservatives of the 20s. Communism is a reaction to the the failings of the monarchy or ruling class…

            I’d generally agree with your premise, but not your examples.

            Hoover who preceded FDR was a standard progressive and FDR campaigned not against lassez-faire but rather simply against Hoover in a purely partisan race one could compare to Bush-Clinton. If you zoomed in there were differences, but only of degree such that from a long view they really weren’t that different (from a populist POV, Hoover was much more isolationist than FDR). RFC was a proto-New Deal program as was the passage of Glass-Steagal and the ERCA. Probably most disastrous among Hoover’s policies was the Revenue Act of 1932 which raised taxes across the board.

            Similarly, the Communist revolution wasn’t really in reaction to a particularly oppressive Tsarist government, it was just a weak government because it was getting its ass kicked in a world war. The Whites initially seized control, but the Reds were actually a better organized set of leaders, with foreign allies, and being anti-war was always going to be more popular given how poorly the army was faring.

          • HeelBearCub says:

            In order for a reactionary movement to succeed, the opposition has to be weak. I don’t think that’s really a point against communism.

            Regardless of Hoover’s policies, the reaction is against the laissez-faire policies of Coolidge and the Republicans in general in the lead up to the crash and the Great Depression. He was a member of the Coolidge administration, he isn’t going to wash that stink off.

        • WarOnReasons says:

          their support was concentrated in the lower-middle classes

          I have encountered this claim many times, but never seen the statistical evidence behind it. Do you happen to know any?

          The university campus elections show that the NSDAP was disproportionally more popular among students (hardly lower-middle class group back then) than among general population. Also, looking at the 1933 election results map, I see that the NSDAP won ~60% of the vote in Eastern Prussia compared to ~30% of vote in Eastern Bavaria. Was Eastern Prussia really more lower-middle class than Bavaria? Or were other factors (such as religion and local history) much more important than class?

          • dndnrsn says:

            With regard to the statistics, I’m going by the first volume of Evans’ trilogy, which I highly recommend. I’ve read the same thing elsewhere, but that’s what I’ve read most recently.

            With regard to university students, there was disproportionate support for the NSDAP among students, but university students were a smaller % of the population than they are today in developed countries.

            With regard to East Prussia versus Bavaria, religion and local history were indeed important factors. Bavaria was dominated by Catholics, and the Catholic Centre Party (not solely a Bavarian party, but pretty strong there) had a middle-class constituency; it saw relatively little defection of voters to the NSDAP (but ultimately, once the Nazis were in power, they caved; the Nazis then moved to neutralize them). East Prussia, meanwhile, had been cut off from Germany by post-WWI territorial adjustments, had lost territory elsewhere in same, and so supported revanchism.

      • albatross11 says:

        My sense of this: Brazil has gone through a pretty big economic downturn (I think driven largely by the oil market), and has had several years of massive corruption scandals. Their previous president was impeached and removed from office; her predecessor is in prison (possibly because he was planning to run for re-election and is still pretty popular). A whole bunch of high-profile politicians have been prosecuted for corruption. Crime is a huge problem in Brazil, and I think that people in many big cities have to deal with really scary levels of crime and little police protection.

        With that as a backdrop, it’s pretty easy for me to see how the public decides to vote for someone who’s pretty awful in many ways, but who at least seems not to be corrupt, and has a plan for restoring order.

        I think Trump and Brexit are less-intense versions of a similar dynamic. The Iraq war and the 2008 financial meltdown and subsequent bailouts made our own ruling class look both incompetent and corrupt. Worsening social and economic outcomes for large parts of the country helped that along. My guess is that this is part of what created an opening for someone like Trump to win the election. To be clear about what I’m claiming: I think in a world without the loss of faith in the elites[1] that came from the foreign and domestic policy clusterfucks of the 00s, we would not have seen Trump being successful–his remarkable gifts wouldn’t have been sufficient to overcome the impression that he wasn’t a serious candidate. But once the serious candidates had lost most of their credibility, in an election where the pre-ordained choice seemed to be a Clinton or a Bush, he had an opening.

        [1] This includes elites in academia, media, government, and politics. All came off badly in the Iraq War fever and worse in the financial meltdown, IMO.

        • Eugene Dawn says:

          Sure, I think the dynamic you describe is plausible, but it’s not obvious to me that it maps perfectly onto Glenn’s tweet.

          For example, you note Lula’s continuing popularity, and polling I glanced at while writing my earlier post seems to confirm that at least during the first round, when respondents were given the chance to pick Lula, he was far and away the front-runner. Even though he is one of the elites, currently sitting in prison! If Lula could have run, would Bolsonaro have won? And if not, what does this say about the thesis?

          Similarly, I think polling in the States showed that Obama would have beaten Trump handily had he been allowed to run for a third term, though obviously the same conditions would have held. In other words, Trump and Bolsonaro would both very plausibly have fared worse against elite politicians who were more directly responsible for the conditions that apparently pushed voters away from the status quo! This seems hard to reconcile with a simple reading of Glenn’s tweet.

          Now, I don’t want to push too hard here: there are obvious countervailing tendencies: incumbency matters, candidate quality (charisma etc) matters, and it’s certainly plausible that all else equal, an outsider, burn-it-all-down candidate stands to do better in such conditions than otherwise. But it seems to me they need a few things to go their way, such as drawing weaker candidates to run against; in Trump’s case his victory required the peculiar electoral system of the US (since he would have lost if it were a simple popular vote). In Trump’s case, I think his performance is pretty consistent with what fundamentals-driven models give you–in other words, he performed about as well as a generic Republican would have been expected to. Now, this still leaves us to explain how he won the primary, or why he didn’t drastically under-perform, but I’m not sure the evidence is great for a surge in enthusiasm for an outsider candidate among the general electorate.

          Finally, the implication in Glenn’s tweet is that it is the people being fucked over who revolt and vote for the outsider candidates. But as I said above, I don’t think that’s true, or at least it’s not so simple. Bolsonaro seems to have done best among richer folks and whiter folks in Brazil; I presume this group has more overlap with the “ruling class” referred to in Glenn’s tweet. In other words, Bolsonaro’s strongest constituency are those who presumably bear the most responsibility for the state of Brazil according to the formulation in the tweet. So it’s not obvious that people who resent the failure of the ruling classes and people who vote for outsider candidates overlap all that much. Now, they may not need to overlap that much to swing elections, but to the extent that my analysis above is accurate, I think it at least undercuts the general thrust of the tweet.

          • Eugene Dawn says:

            Maybe one more thing to say here: I think any discussion on this needs to distinguish between voting for the opposition, voting for an outsider candidate, and voting for a burn-it-down candidate.

            I think it’s pretty commonplace if the party in power does poorly, that increases the odds the opposition will win in the future. But this can’t be what Glenn is referring to because the opposition can and often will still be drawn from the ruling class.

            Next, in some instances of misrule, voters might tend to prefer candidates who aren’t associated with current elite politics at all. I think again this has some truth to it, though probably it needs a few more qualifications than the previous statement. But again, this doesn’t comport with what Glenn is talking about because outsider candidates might be ‘renewal’ candidates, not ‘burn-it-down’ candidates: Obama in 2008 is a plausible example of such a candidate, and I don’t think Glenn would classify him as a ‘burn-it-down’ candidate (though others might).

            Finally, if elite failure is spectacular enough, voters will want to vote for candidates who they understand to be explicitly awful along many axes but with the understanding that those are necessary character flaws to get the job done. This I am the most skeptical of. I think such candidates may do better by virtue of being a subset of ‘outsider’ candidates, but I’m not sure they do any better than that. As I say, outsider-Trump vs. ex-outsider-Obama seems like it would have been an easy romp to victory for Obama, which I take as evidence that there is nothing beyond ‘outsiderness’ that helps Trump.

            So why didn’t Brazil vote for a fresh-faced young reformer? One answer is that, though Glenn perceives Bolsonaro as a ‘burn it down’ candidate, his voters don’t: they just see him as a reformer. This I think is a plausible account for some of Trump’s success, but I don’t think fits well with Glenn’s explanation in his tweet–it’s not desperation and anger that lead to such a candidate doing well, it’s mis-identification of that candidate as a reformer when they’re really an arsonist.

          • albatross11 says:

            I suspect you’re right that Obama would have beaten Trump, but I’m not entirely certain–Trump beat a lot of very capable, seasoned professionals. He would have used entirely different attacks on Obama from the ones he used on Hillary, and they might have worked.

          • albatross11 says:

            I suspect very few people think of themselves as voting for a candidate so they can watch the world burn. More likely, they’re willing to go more extreme, the more the elite consensus in the country has visibly failed.

          • HeelBearCub says:

            He would have used entirely different attacks on Obama from the ones he used on Hillary, and they might have worked.

            I’m not seeing much, if any, deviation in tactics by Trump against his perceived opponents over the last 3+ years. This seems like an odd statement to me.

          • John Schilling says:

            I suspect you’re right that Obama would have beaten Trump, but I’m not entirely certain–Trump beat a lot of very capable, seasoned professionals.

            Right, but in a Trump/Obama election, he’d have been going up against one very capable, seasoned professional. That’s vastly harder. The only time Trump ever faced off against one very capable, seasoned professional politician, he barely pulled it off in spite of that politician being burdened with the “most hated person in America” title and not actually having much capability or seasoning in the election-winning subfield of politics.

            Obama was probably only the third most hated person in America at that point, and he was demonstrably very good at winning elections.

          • azhdahak says:

            in Trump’s case his victory required the peculiar electoral system of the US (since he would have lost if it were a simple popular vote)

            Nonsense. If we elected presidents by simple popular vote, candidates would campaign differently, and prospective voters would behave differently. If we’d switched to a simple popular vote at midnight on Election Day without telling one, Trump would have lost. If we’d had a simple popular vote all along, Trump would’ve run a different campaign, and we can’t predict the differences well enough to say what would’ve happened — especially since, beginning with the first presidential election decided by a simple popular vote, the entire political history of our country would’ve been different. No one has any idea what would’ve happened in a political context where Republican candidates campaigned in Massachusetts.

          • meh says:

            No one has any idea what would’ve happened

            Nonsense. We are quite capable of modeling, estimating, and predicting. We’ll never have certainty, but we have *some* idea.

          • John Schilling says:

            What’s your model for how Johnson voters would have split in a run-off?

            Because both Trump and Hillary would have lost in a simple popular vote. No nation on Earth hands out supreme executive power on the basis of some farcical mere-plurality popular vote, and for good reason. There’s always either a run-off or a counting of second-choice votes or an appeal to an entirely different set of electors

            I think we can safely model the Stein votes as going to Hillary and the McMullin votes going to Trump, but Johnson is the kingmaker in this scenario and his votes are genuinely tricky to model. And then there’s the question of differential turnout in the runoff or blank second-place spots in a ranked-preference ballot.

            If all you want to do is say that 48.18 > 46.19 therefore Hillary coulda/shoulda/woulda won, that’s almost as irrelevant as saying 247 > 188 or 191 > 165 therefore Trump properly won.

          • meh says:

            What’s your model for how Johnson voters would have split in a run-off?

            I don’t have one, but I’m certain if I proposed 2 different ones, you would think one was more accurate than the other.

            There was no election without Stein or Johnson on the ballot, yet you aren’t saying “no one has any idea” how they would have voted in a runoff.

          • Eugene Dawn says:

            If we’d had a simple popular vote all along, Trump would’ve run a different campaign, and we can’t predict the differences well enough to say what would’ve happened

            This is a fair point, but it doesn’t really help Greenwald’s argument, since it’s quite plausible the difference in Trump’s campaign would have been that he didn’t appeal to the anger and desperation that Glenn claims is the reason he won.

            My point isn’t to suggest that Trump couldn’t have won the popular vote, only that “people were angry and desperate and so lashed out and voted Trump” isn’t really a perfect fit for the data, though it’s still consistent with it.

            A similar point applies to Albatross’s argument:

            He would have used entirely different attacks on Obama from the ones he used on Hillary, and they might have worked.

            Yes, but in that counterfactual, the attacks on Obama might not fit as well with Glenn’s thesis. We just don’t have enough information to know. But we can say, in the world we live in, evidence indicated that Trump was less popular than Clinton, and dramatically less popular than Obama, and while that doesn’t kill Glenn’s argument, I think it’s at odds with it.

    • idontknow131647093 says:

      I think Greenwald’s priors are a bit of a problem here (as well as everyone else fearmongering about Bolsonaro). There appears to be a willful blindness to just how criminal, oppressive, and incompetent the previous regimes were.

      Almost every state in South America governs with an unearned sense of accomplishment. Wailing about “dog whistles” towards law and order from 1st World progressives need to be treated with skepticism. We are talking about a place that already was a de facto dictatorship, just one that happened to have some support from progressives, and now its a less powerful dictatorship with less western support.

      • Eugene Dawn says:

        Greenwald currently lives in Brazil, fwiw, so without defending anything else, I do think he has some sense of the previous regimes, and is partially exempt from the charge of being a 1st world progressive.

        • albatross11 says:

          He’s lived in Brazil for several years with a husband and some adopted kids and a bunch of dogs. So he may be wrong about Bolsonaro or Brazilian politics in general, but he’s definitely got skin in the game.

          • idontknow131647093 says:

            Greenwald’s domicile is not really that informative as to his worldview.

          • Eugene Dawn says:

            Then calling him a 1st world progressive, and implying that meant his views should be treated with skepticism was not a useful part of your original comment.

  15. Uribe says:

    Plastics.

    This idea occurred to me while watching Blue Planet. Basically every bit of energy in the ocean is used by one organism or another. Now there is all this plastic in the ocean. Isn’t it inevitable that organisms in the ocean and/or air will evolve that are able to metabolize all the plastic in the ocean? Is there anything special about plastic that makes this unlikely?

    • Le Maistre Chat says:

      We shall make our own aliens, like the crinoid Old Ones.

    • Basil Elton says:

      Something like this happened in the Carboniferous period. When the trees appeared, no bacteria of the time could decompose the wood fiber. They learned to do it eventually, but not before all this carbon giving the period it’s name was accumulated – and Carboniferous lasted 60 mln years. So we might be up for some really long wait here, and chances are all the plastic will be converted to paperclips long before that.

    • sty_silver says:

      I’m not sure whether or not there’s something special about plastic, but wouldn’t it take way too long for any evolution to happen to be helpful?

    • Machine Interface says:

      I recall reading that in fact, bacteria and fungi that feed on hydrocarbons have already been reported in many places, and it is theorized that they are a factor in why the measured content of plastic in the ocean seems to have stabilized for quite a while in spite of levels of human contamination having not sensibly diminished during that period.

      Examples I could find by quick googling include Acanivorax borkumensis, Ideonella sakaiensis and Aspergillus tubingensis.

    • Tatterdemalion says: