[Epistemic status: Low confidence. I have found numbers and stared at them until they made sense to me, but I have no education in this area. Tell me if I’m wrong.]
I.
There has recently been a lot of dumb fighting over who uses how much water in California, so I thought I would see if it made more sense as an infographic sort of thing:
Sources include Understanding Water Use In California, Inputs To Farm Production, California Water Usage In Crops, Urban Water Use Efficiency, Water Use In California, and Water: Who Uses How Much. There are some contradictions, probably caused by using sources from different years, and although I’m pretty confident this is right on an order of magnitude scale I’m not sure about a percentage point here or there. But that having been said:
On a state-sized level, people measure water in acre-feet, where an acre-foot is the amount of water needed to cover an area of one acre to a depth of one foot. California receives a total of 80 million acre-feet of water per year. Of those, 23 million are stuck in wild rivers (the hydrological phenomenon, not the theme park). These aren’t dammed and don’t have aqueducts to them so they can’t be used for other things. There has been a lot of misdirection over this recently, since having pristine wild rivers that fish swim in seems like an environmental cause, and so you can say that “environmentalists have locked up 23 million acre-feet of California water”. This is not a complete lie; if not for environmentalism, maybe some of these rivers would have been dammed up and added to the water system. But in practice you can’t dam every single river and most of these are way off in the middle of nowhere far away from the water-needing population. People’s ulterior motives shape whether or not they add these to the pot; I’ve put them in a different color blue to mark this.
Aside from that, another 14 million acre-feet are potentially usable, but deliberately diverted to environmental or recreational causes. These include 7.2 million for “recreational rivers”, apparently ones that people like to boat down, 1.6 million to preserve wetlands, and 5.6 million to preserve the Sacramento River Delta. According to environmentalists, this Sacramento River Delta water is non-negotiable, because if we stopped sending fresh water there the entire Sacramento River delta would turn salty and it would lead to some kind of catastrophe that would threaten our ability to get fresh water into the system at all.
34 million acre-feet of water are diverted to agriculture. The most water-expensive crop is alfalfa, which requires 5.3 million acre-feet a year. If you’re asking “Who the heck eats 5.3 million acre-feet of alfalfa?” the answer is “cows”. A bunch of other crops use about 2 million acre-feet each.
All urban water consumption totals 9 million acre-feet. Of those, 2.4 million are for commercial and industrial institutions, 3.8 million are for lawns, and 2.8 million are personal water use by average citizens in their houses. In case you’re wondering about this latter group, by my calculations all water faucets use 0.5 million, all toilets use 0.9 million, all showers use 0.5 million, leaks lose 0.3 million, and the remaining 0.6 million covers everything else – washing machines, dishwashers, et cetera.
Since numbers like these are hard to think about, it might be interesting to put them in a more intuitive form. The median California family earns $70,000 a year – let’s take a family just a little better-off than that who are making $80,000 so we can map it on nicely to California’s yearly water income of 80 million acre-feet.
The unusable 23 million acre-feet which go into wild rivers and never make it into the pot correspond to the unusable taxes the California family will have to pay. So our family is left with $57,000 post-tax income.
In this analogy, California is spending $14,000 on environment and recreation, $34,000 on agriculture, and $9,000 on all urban areas. All household uses – toilets, showers, faucets, etc – only add up to about $2,800 of their budget.
There is currently a water shortfall of about 6 million acre-feet per year, which is being sustained by exploiting non-renewable groundwater and other sources. This is the equivalent of our slightly-richer-than-average family having to borrow $6,000 from the bank each year to get by.
II.
Armed with this information, let’s see what we can make of some recent big news stories.
Apparently we are supposed to be worried about fracking depleting water in California. ThinkProgress reports that Despite Historic Drought, California Used 70 Million Gallons Of Water For Fracking Last Year. Similar concerns are raised by RT, Huffington Post, and even The New York Times. But 70 million gallons equals 214 acre-feet. Remember, alfalfa production uses 5.3 million acre feet. In our family-of-four analogy above, all the fracking in California costs them about a quarter. Worrying over fracking is like seeing an upper middle class family who are $6,000 in debt, and freaking out because one of their kids bought a gumball from a machine.
Apparently we are also supposed to be worried about Nestle bottling water in California. ABC News writes an article called Nestle Needs To Stop Bottling Water In Drought-Stricken California, Advocacy Group Says, about a group called the “Courage Campaign” who have gotten 135,000 signatures on a petition saying that Nestle needs to stop “bottling the scarce resource straight from the heart of California’s drought and selling it for profit.” Salon goes even further – their article is called Nestle’s Despicable Water Crisis Profiteering: How It’s Making A Killing While California Is Dying Of Thirst, and as always with this sort of thing Jezebel also has to get in on the action. But Nestle’s plant uses only 150 acre-feet, about one forty-thousandth the amount used to grow alfalfa, and the equivalent of about a dime to our family of four.
The Wall Street Journal says that farms are a scapegoat for the water crisis, because in fact the real culprits are environmentalists. They say that “A common claim is that agriculture consumes about 80% of ‘developed’ water supply, yet this excludes the half swiped off the top for environmental purposes.” But environmentalism only swipes half if you count among that half all of the wild rivers in the state – that is, every drop of water not collected, put in an aqueduct, and used to irrigate something is a “concession” to environmentalists. A more realistic figure for environmental causes is the 14 million acre-feet marked “Other Environmental” on the map above, and even that includes concessions to recreational boaters and to whatever catastrophe is supposed to happen if we can’t keep the Sacramento Delta working properly. It’s hard to calculate exactly how much of California’s water goes to environmental causes, but half is definitely an exaggeration.
Wired is concerned that the federal government is ordering California to spend 12,000 acre-feet of water to save six fish (h/t Alyssa Vance). Apparently these are endangered fish in some river who need to get out to the Pacific to breed, and the best way to help them do that is to fill up the river with 12,000 acre feet of water. That’s about $12 on our family’s budget, which works out to $2 per fish. I was going to say that I could totally see a family spending $2 on a fish, especially if it was one of those cool glow-in-the-dark fish I used to have when I was a kid, but then I remembered this was a metaphor and the family is actually the entire state budget of California but the six fish are still literally just six fish. Okay, yes, that seems a little much.
III.
Finally, Marginal Revolution and even some among the mysterious and endangered population of non-blog-having economists are talking about how really the system of price controls and subsidies in the water market is ridiculous and if we had a free market on water all of our problems would be solved. It looks to me like that’s probably right.
Consider: When I used to live in California, even before this recent drought I was being told to take fewer showers, to install low-flush toilets that were inconvenient and didn’t really work all that well, to limit my use of the washing machine and dishwasher, et cetera. It was actually pretty inconvenient. I assume all forty million residents of California were getting the same message, and that a lot of them would have liked to be able to pay for the right to take nice long relaxing showers.
But if all the savings from water rationing amounted to 20% of our residential water use, then that equals about 0.5 MAF, which is about 10% of the water used to irrigate alfalfa. The California alfalfa industry makes a total of $860 million worth of alfalfa hay per year. So if you calculate it out, a California resident who wants to spend her fair share of money to solve the water crisis without worrying about cutting back could do it by paying the alfalfa industry $2 to not grow $2 worth of alfalfa, thus saving as much water as if she very carefully rationed her own use.
If you were to offer California residents the opportunity to not have to go through the whole gigantic water-rationing rigamarole for $2 a head, I think even the poorest people in the state would be pretty excited about that. My mother just bought and installed a new water-saving toilet – which took quite a bit of her time and money – and furthermore, the government is going to give her a $125 rebate for doing so. Cutting water on the individual level is hard and expensive. But if instead of trying to save water ourselves, we just paid the alfalfa industry not to grow alfalfa, all the citizens of California could do their share for $2. If they also wanted to have a huge lush water-guzzling lawn, their payment to the alfalfa industry would skyrocket all the way to $5 per year.
In fact, though I am not at all sure here and I’ll want a real economist to double-check this, it seems to me if we wanted to buy out all alfalfa growers by paying them their usual yearly income to just sit around and not grow any alfalfa, that would cost $860 million per year and free up 5.3 million acre-feet, ie pretty much our entire shortfall of 6 million acre-feet, thus solving the drought. Sure, 860 million dollars sounds like a lot of money, but note that right now California newspapers have headlines like Billions In Water Spending Not Enough, Officials Say. Well, maybe that’s because you’re spending it on giving people $125 rebates for water-saving toilets, instead of buying out the alfalfa industry. I realize that paying people subsidies to misuse water to grow unprofitable crops, and then offering them countersubsidies to not take your first set of subsidies, is to say the least a very creative way to spend government money – but the point is it is better than what we’re doing now.
Just gotta say, I love that bit about the fish.
These details (along with the conclusions) just make the entire thing entertaining, even though it’s basically just statistics.
Are you implying, good sir, that statistics is not entertaining!?
Another situation in which the solution is absurdly simple, but tribalism renders impossible. You can rage at environmentalists all you like, but they own California. You can rage at farmers all you like, but “Farmers” are an american icon, and no one wants to stick it to Ma and Pa down on the farm (despite the fact that most of these are mega-corporation owned and illegal-immigrant worked). The only people you can stick it to is everyone else, or the 99% of the state that isn’t the problem, and people will do this, because society! So you get maximum inconvenience at maximum cost and minimum effectiveness.
I’m waiting for the article dinging golf courses and their evil, rich patrons as the culprits, which everyone will nod sagely at and do nothing.
Heh. Read Mencken on 1920s farmers.
I don’t think environmentalists are too big a part of the problem here, and I don’t think farmers are much of a national icon in California.
There are lots of ways to blame environmentalists. The ways that you mention in the article are, indeed, lousy. But it is legitimate to partly blame environmentalists for the failure of NAWAPA forty years ago.
NAWAPA sounds like another golden bullet, like nuclear power, that would have solved all our problem if it wasn’t for those damn environmentalists.
The zeitgeist within environmentalists at my university seems to be that water should be priced for full cost recovery to discourage inefficient use.
Then again this is probably weird, we have strong water management and environmental engineering departments.
At a wild guess, some nuanced policy is being quoted differently by different groups and/or in different contexts.
I haven’t studied this, but my off-hand environmental policy might be:
“To the poor, free. To the rich, taxed heavily and regulated.”
The main environmentalist component of the problem is their anti-marketism, in which they yell that water is sacred and needs to be controlled by the collective, which is not insignificant in forcing everyone towards the non-solutions and precluding the stuff that would actually work.
Without minimizing their culpability in that respect, however, it should be realized that their role there is largely acting as a political shield for the agricultural and urban SoCal special interests that have large monetary stakes in ensuring that water is NOT conserved by raising its price.
You are very quick to caricature people who disagree with you as airheaded hippie shamans, but you’re not so quick to explain exactly what positions environmentalists have taken that you disagree with.
Do words such as ‘yell’ rise to the level of fnords? In contexts like the above, I’d think so.
As someone who entirely understands where Irrelevant is coming from and has encountered quite a few environmentalists of exactly that type, I’d like to say I find it really frustrating precisely because I disagree with them on *almost nothing* in terms of… well, the sort of thing that makes you caricature opposing opinions, I guess? That’s hard to define, but “Tribe” works pretty well.
I’m still fairly Blue Tribe and agree with these sorts of people about concern for the environment, concern for the poor, concern about global warming, etc. It’s things like Sacred Values and the like that I disagree with.
Well, that’s part of the problem. Agriculture is a huge industry (we are not talking Ma and Pa Kent on their farm) and that’s also why I’m sceptical of “let the market sort it out”.
No industry wants to increase its costs – that eats into profitability. Paying the market value for water may indeed raise costs, and as the largest consumers, businesses such as agriculture will see costs shoot up. Passing prices onto the consumer only works so far because the end-purchasers of human foodstuffs are the supermarket chains which wield huge control over the price they’re willing to pay to processors (which in turn affects the price those processors – such as creameries and slaughterhouses – pay to primary producers, the farm gate price), and the export markets for soya and alfalfa are going to be equally reluctant about taking price rises for more expensive raw materials.
Which means massive industry pressure, as you’ve pointed out, for subsidies and for keeping prices for water down. I think environmentalists here may play the part of useful idiots; they can take the heat for “if it wasn’t for those blasted environmentalists, there would be no problem!”
+1
This is why we have amazing, cheap computers and cell phones, but overpriced healthcare and stagnant, overpriced education.
Just wait till having a computer with internet access becomes a “human right.” Maybe we can prevent the super-intelligent AI after all?
Computers, Cell-phones, etc. are relatively new technology, and unencumbered by restrictions on experimentation subjects. I do believe that the market would probably do a better job, but I’m not sure that’s a convincing argument.
You have a good point, but it’s worth remembering that the reason we have price controls and the hated “Farm Bill” is because unregulated agriculture is one of the best examples of an inefficient market.
In good years, farmers grow lots of crops, which drives down their market price, and thus their profits. As their profits go down, they plant even more crops to make up the shortfall, which drives the price down even more, creating a vicious cycle. Under the old farm subsidies created by FDR (and I’ll admit my memory of all this is a bit hazy), the government would buy the surplus grain from farmers in good years, and then store it (in what was called the “Ever-Normal Granary”). Then, during bad years, the government would release the surplus grain back onto the market. The system worked pretty well.
It was, of all people, Nixon who got Congress to change the system to direct payments to farmers–which is why we now have enormous surpluses of corn every year, and ridiculously cheap HFCS, and ethanol–it’s been a corn revolution.
Side point: I think Obama’s initiative to have the federal government become the direct lender of student loans will be as consequential as Nixon’s change to farmer payments, which is one of many reasons I think he’s the Democrats’ Nixon, and not their Reagan.
Farmers not making high profits, during a century in which technology enabled us to move from ~70% employed in agriculture to ~2% in agriculture, is an example of an *efficient* market. It takes incentives to persuade someone to uproot and leave the family profession. It takes a lot of incentives if you want your persuasion to have a 68/70 success rate.
Farmers not making *steady* profits, on the other hand, are what futures markets are for.
@ArgleBargleZarg
The reason we have farm subsidies is that farmers are good at lobbying. Farmers are a relatively small group that benefits a lot from farm subsidies, while non-farmers are a large group who are each hurt a little bit by the existence of these subsidies (though in sum the losses outweigh the benefits). Hence farmers have a strong incentive to support subsidies, while non-farmers for the most part don’t care about this issue.
Your story for why an unregulated agricultural market would be inefficient is unconvincing. When profits for farmers fall, that’s a sign that there are too many farmers. So the least efficient farmers being forceed to give up their farms and seek other employment is a good thing. But farm subsidies prevent or retard this salutary adjustment.
@roystgnr
It takes incentives to persuade someone to uproot and leave the family profession.
Not when the family profession is farming. It’s a lot harder to keep people in that particular family business.
@Joe Gunnarsson
It’s not “my story,” it was one of the central facts of life for agriculture up until the New Deal.
It may seem, from a theoretical standpoint, that unprofitable farms going out of business are a sign of an efficient market, but in the real world we have a nation to feed.
“We have a nation to feed, ergo this situation where farmers desperately produce more and more crops while driving the price ever downwards is somehow a bad thing.”
I wouldn’t call that an “environmentalist” value per se, but it’s sadly a “Liberal” value meaning that about 99% of environmentalists also hold it.
Obviously the whole issue could be solved if we just gave free drinking water (which as Scott noted is a *tiny* fraction of water usage; we could even throw in enough water for all household usage) to people, and charged money for other usages of water. I’m… honestly not sure what prevents this solution, besides 1. Agricultural and other business interests being against it and maybe 2. Rank-and-file Red Tribe (of the type that don’t understand economics well enough to know that some of their ideas are right) being against free drinking water?
Scott summed this up pretty well in
“A Something Sort Of Like Left-Libertarianism-ist Manifesto” – we have a partisan division where you can either not understand basic economics, or you can hate poor people.
I’m… honestly not sure what prevents this solution, besides 1. Agricultural and other business interests being against it and maybe 2. Rank-and-file Red Tribe (of the type that don’t understand economics well enough to know that some of their ideas are right) being against free drinking water?
Path dependence. Most of the water used by agriculture in California is delivered by some combination of State, U.S. Bureau of Reclamation, and U.S. Army Corps of Engineers projects. They all have their own rules about cost recovery and benefits, and the state of California can’t even (easily) change its own rules because there’s a quasi-property-right in the existing arrangements. And forget about getting two competing Federal departments to change their rules, especially just for California.
Right. California water is controlled as a whole neither by any central political body nor by the market.
Water would seem like the most fungible product imaginable, but in California it comes with a lot of property rights and customs and expectations and special relationships.
For example, in the 1910s, both Los Angeles and San Francisco secured separate water sources in the Sierra Nevadas. Los Angeles notoriously (“Chinatown”) dried up the Owen River, while San Francisco flooded the magnificent Hetch Hetchy Valley, which was second only to Yosemite Valley for giant granite cliffs. At the time, Hetch Hetchy was more controversial (the Sierra Club came into existence to fight filling Hetch Hetchy Valley with a reservoir), but since then Los Angeles’s acts have become more notorious.
In recent decades, Los Angeles has given back some of its property rights to Eastern Sierra water to restart the Owens River and keep salty Mono Lake topped off. In the 1980s, a mischievous Reagan Administration proposed tearing down Hetch Hetchy Dam and restoring the superb valley, but this plan has been blocked by San Francisco’s liberal Democratic politicians ever since under the idea that they desecrated Hetch Hetchy, they own it, and they ain’t giving it back. You could imagine Coasean solutions in which, say, the rest of the state and the country pays San Francisco to allow Hetch Hetchy to go back to nature so future generations can enjoy it the way millions have enjoyed visiting Yosemite Valley. But San Francisco’s politicians have been wary of re-opening this can of worms in any manner. Their forefathers struck fast and hard more than a century ago to assure the city of a source of water, and they don’t see much reason to revisit what San Francisco’s civic robber barons did.
Lots of different entities around California have long-running relationships with water sources much like San Francisco has with Hetch Hetchy. They more or less like things the way they are and aren’t excited about moving to more of an efficient market for water.
It takes incentives to persuade someone to uproot and leave the family profession.
Farm work is plenty of incentive to leave farming. So is cow shit.
Without minimizing their culpability in that respect, however, it should be realized that their role there is largely acting as a political shield for the agricultural and urban SoCal special interests that have large monetary stakes in ensuring that water is NOT conserved by raising its price.
I wouldn’t be surprised if such interests were cultivating (and buying ad space for) the less credible factions of, well, any social movement that is growing enough to threaten their stakes by a moderate, realistic stand that might actually have some success.
True or false: California farmers currently drawing down aquifers to grow almonds being shipped to China (and alfalfa being fed to dairy cattle to make high-protein yoghurt) are the genetic and/or cultural descendants of the Oklahoma farmers who plowed up the prairie and were shocked (SHOCKED I TELL YOU) when it all went wrong… then they had to move to California.
(I legitimately don’t know to what extent this is true, but it is an idea that is attractive in several ways.)
Not even false.
Not entirely sure what that means, but it is definitely true that in America there is a repeated pattern of
1) farming in a way that makes farming functionally an extractive industry and
2) ignoring that this is what’s occurring until the locally limiting resource starts to run low.
Whether or not the Okies are direct antecedents of present-day California farmers is an amusing adjunct to think about but not actually important.
Since this is Slate Star Codex, I couldn’t quite rule out that you actually thought that agricultural self-destructionism was caused by bad Oklahoma genes. “Bold” hypotheses of that sort are somewhat overvalued in here, in my experience.
In point of fact, I was myself born in California as a descendent of Oklahoma dust-bowl survivors.
The semi-non-fiction book “The Grapes of Wrath” by Steinbeck documents exactly the process you’re describing.
I personally think the problem is memetic, not genetic, but there *are* some memes that need a particular kind of brain to take root in properly.
Public choice theory: farmers just need to be a well-organised minority group.
Farmers are a well-organised minority group. We have to recognise that we’re talking about two types of farmers, or rather two types of farming; from a 2013 report farms in America seem to be dividing up between larger and small farms, with mid-sized farms declining: in other words, large agri-business commerical farming, and small, “gentleman farmer” operations where artisanal, organic, or simple “we fancied retiring to the country” farming goes on.
These statistics from 2004 demonstrate why agriculture in California will continue to be a thirsty business:
This table lists the states in order by total agricultural receipts. California is this nation’s most productive state followed by Texas and Iowa.
Total Agricultural Receipts 2004
1. California
Value ($1,000) 31,835,183
% of Total U.S. 13.20%
The second column of the table lists a number representing the dollar value of the product. This number is not the dollar value of the product. This number represents the dollar value of the product in thousands of dollars. For example, the number listed for the value of dairy products produced in California is 5,365,992. This number represents a dollar value of $5,365,992,000 (5,365,992 x 1,000): five billion, three hundred and sixty-five million, nine hundred and ninety-two thousand dollars.
As you can see from the table below, California ranks 1st among the states for total agricultural production and for total crop production. Only Texas surpasses California in the production of livestock and livestock products.
California’s crop production dollars total more than 2 1/2 times the dollars generated by livestock and livestock products.
California is the top producer of agricultural products in the nation. It produces a wide array of commodities but, its single most valuable category is dairy products. In the livestock group, production of cattle and calves is also important to the state, as is, to a lesser degree, the value of California eggs. Broilers can be counted among the state’s top livestock products but, because data are not available, it’s not possible to determine where they rank among the other commodities.
Among a broad range of crop products, California’s greenhouse and nursery products rank at the top of the list. Grapes, almonds, lettuce, and strawberries round out the top five most valuable California crops.
So, for some rough comparisons – in 2004 California had receipts of $32 billion dollars from agriculture. According to Wikipedia, Silicon Valley attracts one-third of the entire venture capital investment of the U.S.A. In 2011 this was 41% or $13 billion dollars.
Interestingly, public choice theory implies that the more important an industry is to a country’s welfare, the more likely it is to be not subsidised, but taxed.
Let’s say an economy is $1000 billion, and 80% of the income comes from alfalfa farming, and 20% from Hollywood movies (entirely made-up numbers). If the alfalfa farmers combine, and get the government to apply a 10% tax to the movie-makers, redistributed to the alfalfa farmers, and there are no transaction costs then alfalfa revenue goes up from $800 billion to $820 billion, a 2.5% increase, while moviemakers’ revenues fall 10% to $180 billion. But if the moviemarkers combine and get a 10% tax applied on the alfalfa industry, then they get a 40% increase in their revenue. There’s thus a lobbying mismatch.
(Obviously true public goods specific to the dominant industry are a different matter, and in my hypothetical 80% alfalfa economy, assuming a minimally-competent government, I’d expect more spending on public goods that support alfalfa farming, as those increase total tax revenue).
Why compare farming revenues (that’s what I assume you mean by “attracts receipts”) to venture capital investment. Why not compare it Silicon Valley revenue, which would be an order of magnitude higher.
Why not compare it Silicon Valley revenue, which would be an order of magnitude higher
I would, if I could find a figure for such. All I could get was “Silicon Valley attracts X% of venture capital” and some mention of the higher echelons of tech companies are white, male and well-remunerated, while the workers on the assembly lines are non-white, female, and poorly paid.
I mean, you can physically measure how many tonnes of wheat are sold, but what does Google (say) “sell”? It does generate revenue, but not as simply as “we grew so many tonnes of almonds, we sold so many tonnes at such and such a price”.
They don’t even need to be that if the electoral system gives a voting bonus to square kilometers.
Remember, “let the Sacramento delta salt up, whatever” is as much of a water savings as ending the alfalfa crop, and $860 million cheaper. I don’t know what the presumed crisis is, so I can’t say whether that usage is sensible environmentalism or not, but it seems quite clearly environmental.
Farmers aren’t as big of an icon in the populated cities on the coast, for sure, but they sure as hell are large lobbying icons in Sacramento.
I think we’re at the point where every problem faced by anyone, in any capacity, instantly becomes the field for another signaling game, destroying any possibility of solving anything.
Welcome to politics. You must be new here.
I know politics was mostly signaling. But now with the advent of instant social media communication, you can spend one hundred percent of your time and effort in pointless signaling. Before the Internet, you had to do something else eventually, and might end up being useful on accident.
FTFY.
Like this sort of bullshit doesn’t go on in China or Russia. Or the Qing Bureaucracy or Louis XIV’s court.
Bullshit goes on, but it’s a completely different kind of bullshit. In particular, political wrangling is more transparently a matter of buying off influential individuals, rather than signalling contests and propaganda war.
Yeah, so let’s get rid of politics altogether (or, at least, political authority).
Monarchy and explicit rule by elites do have an advantage though: people have no illusions that the king’s policies *aren’t* designed to intentionally enrich and empower him and his friends.
Onyami, is it possible your view of monarchy is tainted by a couple of centuries of populist propaganda? History is full of dynasties that thought about the future in terms of decades and generations. You’d tend to expect that on economic first principles – if a king did explicitly own the country, he would have far more incentive to husband its resources well than your typical president (or, more to the point) your typical voter does.
I am no monarchist (yet) but it does seem to me that in many ways democracy is the epitome of the tragedy of the commons.
And yet, somehow basic research with no obvious practical application still seems to get funded in democratic societies. That’s not a knockdown argument in favour of democracy, of course, but it is a datapoint that needs explaining.
My comment was meant to be in favor of monarchy relative to democracy. Though I’m not sure I’d go as far as Hoppe in actually preferring a monarch to a democracy, I agree with the argument that monarchs are, in some ways, better incentivized to good long-term shepherding of a territory’s resources. I also hate the democratic illusion that the government is “us.”
I am anti-government in general, but if one must have a government, enlightened monarchy is probably the best possible. Problem is, how to ensure the monarch is enlightened? And a very bad monarch can be a much bigger problem than a bad president or prime minister. The president can be impeached or, at least, voted out after 4 or 8 years.
I’ve seen some pro-monarchists argue that, historically, monarchy had a sort of safety valve in that nobles or even the royal family would assassinate, depose, or work around an especially bad monarch, but I know of too many historical examples of really bad monarchs who held real power for decades.
Though perhaps monarchs who claim to rule in the name of democracy and the people (Stalin, Mao, Kim Jong Il, etc.) are the worst of both worlds.
Thepenforests- As a fraction of total expenditures, or even as a fraction of funded research, funding for “basic research with no obvious practical application” is pretty small. I’m tempted to call it negligible, and as such I’m not sure there’s much that needs explaining. Or, to take another tack, we are still living with the memory of how basic research unexpectedly led to the atomic bomb in an astonishingly short time, which has led to a foolish inclination to suppose that all basic research can lead to amazing things in an astonishingly short time. It’s arguable, I think, that “basic research” is over-funded (even though I just called it negligible, heh) by our democratic government — it gets funded to the level it does precisely because no one is in charge. Take that as an argument for democracy if you like.
Onyomi- Apologies, I took your comment the wrong way. It is in fact describing exactly the effect I was getting at, but in more negative terms. I have also read the monarchists you speak of and I too am not (yet) convinced. But it’s interesting that you find the notion of moderation by the nobles and royal family implausible or inadequate but are optimistic about moderation by impeachment or election, both of which are orchestrated by our equivalent of the nobles and royal family.
@Dr Mist
Impeachment and election s both operate under explicit rules, the rule of law, Assassinations and court iintrigues don’t.
@onyomi
Court intrigues may be more explicitly about bullying, but they are more opaque in every other respect.
” Problem is, how to ensure the monarch is enlightened”
I’ll say. The lack of an enlightenment generating system….a timber straightener… is what prompts me to believe democracy is the best system out of the systems that actually are systems and not pipe dreams.
I think we’ve been at that point for a long time (I talk about this a lot on my Tumblr — I should make a real blog), and yet things still occasionally get solved, somehow. Which isn’t to say that tribal fighting is good, or anything, just that we’re not at some doomed historical low.
I don’t think things have changed. The people who do things besides signaling still do things besides signaling, and the people who don’t, never really did.
The only thing that’s changed is that it’s become easier to get a sense of what people outside your Monkeysphere are doing (and, perhaps more importantly, what the people IN your Monkeysphere are doing when you’re not around), so people who extrapolated the observed behaviors of their social circle onto humanity as a whole are going to assume that it’s humanity that has changed, not their information about humanity. Like, when Person X says something on their own social media feed that they wouldn’t say when you’re around, it’s easier to assume that they’ve suddenly changed, rather than that your model of reality was incomplete.
Welcome to democracy.
I generally think that worrying about unfriendly AI is silly, but you could make a pretty good case that the government itself is already a super-intelligence pursuing goals orthogonal at best to human flourishing. No idea how you beat it, either.
My optimistic perspective is that humanity is like the bird that cleans the teeth of the alligator that is government. Or perhaps an even closer symbiosis: government could not survive without humanity, and pace the anarchists I feel humanity would be even worse off without government. Perhaps humanity is the gut microbiota of government.
The better way to think of it is humanity as the body and government as the gut bacteria. A little bit is good, too much devours you from the inside.
What’s your justification for the super-intelligence?
And while normally I’m fairly skeptical about government, it does strike me that the governments of places like Switzerland, and Scandinavia, do, over the decades, do not that badly.
Yes, Scandinavian governments are awesome. Let’s get a king, an official language, and a state religion and see if we can replicate their success.
Who’s we? The UK has all of that bar the king, and the last is just a matter of time.
Pretty sure it’s not monarchy or religion that’s working so well in Scandinavia.
I’m going to have to use the “S” word. It’s social…democracy.
I chickened out 😉
Iirc, the uk’s only de jure official language is Welsh. English is merely a de facto official language.
Actually the Scandinavian countries rank highly on the Heritage Foundation’s Index of Economic Freedom, dragged down by the size of government measures.
(Yeah, yeah, but (a) the advocates of an idea are best-placed to define what that idea is, assuming they avoid the No True Scotsman Fallacy, and (b) ranking every country in the world numerically avoids the No True Scotsman Fallacy).
The Heritage Foundation’s economic freedom index is weird and arbitrary. It’s literally defined such that optimal “Government Spending” is always 0.
I get very tired of direct comparisons of Nordic countries to the United States. Sweden, for example, is a land area the size of California, much of it barely inhabited, with 9 million people of a roughly homogeneous racial, cultural, and linguistic heritage.
The United States is over 20x the size of Sweden, has over 30x as many people, and includes just about every imaginable climate, culture, race, religion, and industry.
So let’s just take what worked for Sweden and apply it unilaterally to the United States? I’m not saying anyone here is saying that, exactly, but that is often a strong undertone, if not an explicit suggestion in op-eds and the like on the topic.
@David Allen: I don’t see how that is arbitrary from a libertarian-type perspective.
But they publish the sub-indices so you can always strip that out and compare countries without it if you like. If you do, from memory the Nordic countries look a lot like the USA, or even more unregulated.
@onyomi: I gave an example of Switzerland, which doesn’t have ethnic homogeneity, nor linguistic homogeneity, nor a state religion, nor a king.
@Tracy,
I was talking specifically about comparisons of the US to Scandinavian countries, not responding directly to your comment. I would probably be happy if more people suggested we be like Switzerland (very non-interventionist, great financial privacy, everyone owns a gun), though if I were to continually offer Switzerland as a model for the US I think I would be rightly vulnerable to the same criticism I’m leveling here: which is, just because it worked for a tiny population surrounded by mountains doesn’t mean it can work here.
Switzerland (and the Scandinavian countries too, in fact), do prove one thing I find to be generally true: sovereign governments (i. e. not the government of Rhode Island) of smaller territories and populations tend to be much better, regardless of other factors, like diversity of population (see Singapore, for example, though I do think diversity of population provides a different set of challenges). Among other reasons, this is because it’s impossible to entertain dreams of autarchy, and, even more importantly, too much competition (it’s too easy to move if the government sucks).
In other words, if it’s smallness that makes Scandinavian governments good, then trying to apply their policies to a huge population is missing the point.
Imagine you go to interview a doctor with amazing success rates in order to find out his secret. He tells you “we only accept a very few patients and we spend a long time with each of them, crafting an individualized plan and following up frequently to adjust as necessary.” And you go back to your mega-hospital and say “we should really copy that guy’s office setup and intake forms!”
onyomi –
So let’s just take what worked for Sweden and apply it unilaterally to the United States?
It worked pretty well in Minnesota.
Don’t forget a mostly-homogeneous population interspersed with the hottest chicks stolen from half a continent during the Viking Age, and a thousand years of church pressure not to marry close relatives for a very loose definition of “close” that resulted in everyone being everyone else’s sixth cousin once removed.
I think it would be more easily classified as an alien intelligence; it is intelligent in that it is able engage in complex problem-solving. It is alien in that its desires are not our desires, and its ways not our ways. Distorting markets by piling subsidies on top of subsidies and then papering over the effects of those subsidies with more subsidies accomplishes the government’s goals pretty well. It does not accomplish the goal of using water in an efficient manner, but that’s not the point.
“Super” is admittedly debatable, but off-hand I’d say that the government is far better at making me do things than I am at making it do things, so in terms of being able to achieve desired goals, it seems to be the superior one.
This is only true in certain spheres of life. The government is spectacularly bad at doing things like making people change their attitudes towards others, stopping people from using drugs, etc. The vast power of the Soviet Union’s government was not able to get people to act in an economically productive manner.
Hmm…Maybe I’m just younger and less seaweary than you are, but it’s like we haven’t read the same post. Scott created a surface analysis of a deep situation and produced an incredibly simple solution–that does not involve blaming or “rage” or other forms of mindkilly tribalism at all. Further, other than a false sense of fairness, I really don’t see how the evidence points to blame at the environmentalists (it’s not like the 23 million MAF of cold sweet waters is gonna magically get damned…er, I mean dammed, if the environmentalists shut up).
Uncharitably, your comment sames slated to maximize wise nods and sage agreements by other rationalists looking to feel superior, and not to improve epistemic quality or deliver additional information.
Ding.
Much like any other blog that tackles anything remotely political, and has not become an one-sided echo chamber, every post is seen as a nail to be pounded by those with hammers. (Myself included, I’m sure.)
Although, I have a feeling that Scott’s post is not very far-seeing, given that CAs water problems are just at the beginning. So 40 years from now, when CA residents are still paying a billion 2015 dollars to “farmers” not grow crops that they are receiving federal money to grow, this will be held up by a future SSC as an example of the government mucking up the free market (rather than being regarded as the tragedy of the commons problem that it is).
See, nail meet hammer.
… as if Scott’s point was a policy proposal instead of a reductio ad absurdum.
Scott actually proposed a very pragmatic solution to the current problem. Some version that really bought out agricultural interests makes a great deal of sense in the short-term.
In the medium term, CA has to come to grips with the fact that the long term outlook is even drier, and therefore the old water rights no longer apply.
That two-pronged approach actually worked to ease NC out of tobacco without death spiraling the overall economy (although rural NC is not the same anymore). And my understanding of that system is that it was essentially private, so paying people to grow seems to have made sense for private concerns as well.
In my experience, the people who brag the most about how above-it-all and not-mind-killed and free-thinking and non-tribalistic and open-minded they are, usually turn out to be the most opinionated, dogmatic, tribalistic and mind-killed of us all.
Political opinionation is the Jungian Shadow archetype of the Rationalist, it seems.
What will the cows eat then? Won’t the incentive to produce alfalfa go up drastically as you start buying people out, so that you end up having to spend wayyy too much money?
Yeah, this is one of the things markets are really good at – you don’t have to figure out what the cows will eat, just stop distorting the market and the cows will eat whatever’s cheap.
repeat after me: markets are not magic.
Farmers will have to figure out what cows will eat.
first assumption: there IS a substitute for alfalfa (soy probably? i dont know much about farming)
second assumption: the substitute is NOT growing something else on the former alfalfa fields, makeing the farmers use the same amount of water while collecting no-alphalpha subsidies. You would have to oblige them to cut down on water usage. As XerxesPraelor said: the market works against us here, so you have to prohibit alphalpha use and pay compensation rather than “buying them out”.
Third assumption: While this will make some soy farmers (or whomever) very happy, Beef Farmers will have higher costs, probably alot. The market is flexible enough to compensate? What happens to the labour market, when all of californias alphalpha farmers and half of its cattle farmers suddenly move to the cities because they need a new job? I absolutely have no idea here, anyone care to speculate?
There’s a pretty simple alternative to alfalfa. It’s alfalfa, just grown in places where there’s more water. Allow the price of alfalfa to reflect the inputs and other sources will meet the demand.
We’re focusing way too much on California as a standalone. Sure, California beef prices might rise, but that’s just going to incentivize ranchers everywhere else to grow their herds.
And I’d bet that all of the alfalfa and beef farmers who would be forced out could fit in a single SF highrise.
The market isn’t magic, it just looks that way because none of us have the perfect information needed to see all of the moving parts.
Alfalfa actually grows best in places where its irrigated. Its deep taproots allow it to grow well with long intervals between watering, which most weeds/forbs can’t handle. Irrigation allows you to space the waterings properly so there isn’t much competition for the alfalfa. On the other hand, alfalfa is particularly sensitive to too much water. Finally, no water or rain when you are drying the hay is vastly important.
Most crops do better on irrigation simply because they get more sunlight (if you are irrigating, presumably its a dry area). But with alfalfa the advantages are compounded.
This isn’t an argument for or against, simply adding some relevant background here.
Clarification: Beef cattle in California eat little if any alfalfa. They’re grazed on non-arable hills, and supplemented with what would otherwise be agricultural waste: used brewing grains, rice bran, almond hulls, orange peels, etc. (The exception is feedlot forage, which I believe is mostly corn.)
Alfalfa is for the dairy industry. They can’t rely on grazing: they need to pack many animals into a small space around the milking equipment, and they need a consistent protein-rich diet. Those cows are high-performance and high-maintenance. They need different feed mixes according to their reproductive cycle, and if you screw this up they will, I’m not kidding, disarrange their internal organs and you’ll be covered in blood up to your shoulders and the vet will tsk-tsk at you while reassembling your cow. So yeah, alfalfa is a big part of this.
There ARE alternatives (white beans and other legumes, maybe even soy) but it’s much the same problem as with gasoline: switching the existing equipment over to it is a gigantic pain in the ass. You want to convince every dairy operator in Merced County to change to your new feed that replaces half of the alfalfa with peanuts or something, good luck. They are change-averse for a reason.
Also worth noting that alfalfa is a nitrogen fixer, so if your replacement crop doesn’t do that you’ll either need to rotate something in that does (dramatically decreasing per-acreage efficiency) or deplete your soils.
I was about to point this out and noticed you said it much better than I could have.
If all else fails, there are alternatives to meat.
I think what you mean is there are alternative meats. Like vegans. I hear they’re delicious.
Hammers are not magic, but they’re damn good for driving nails. Allocating scarce resources to produce commodities efficiently is the thing in this world markets are best at. Most of the problems with markets come when they’re applied outside this scope.
But how do you ensure that the water you ‘saved’ by paying alfalfa farmers doesn’t just get used for the next-cheapest-crop, and you’re back at the original problem?
I guess what you effectively want to do is buy water off alfalfa farmers; not sure how that’d work (but I’m not sure how water distribution works in general). Then there’s no extra water for other crops to pick up easily.
The next cheapest crop uses less water.
Just to clarify this, stop distorting the markets means stop giving away water for free. Paying people to not grow alfalfa is just distorting it more, and will likely make the problem worse. After all, you can’t get paid to not grow alfalfa if you don’t grow any in the first place.
In the glorious pre-market distortion era of 800BC, the water would have been free anyways because no one would have owned it (when I draw on the water table under my house, the water under your house leaves you of its own free power and moves in under me, I didn’t steal anything.) So charging money for water would involve a different intervention, but that wouldn’t be the same thing as no intervention.
You’re confusing the thing and the name for the thing; the name for the thing is water rights, but the thing is self-defense, and if your water extraction is a threat to my existence I self-defend against it.
@ Irrelevant: That logic applies just as well to either party, if *both* parties need the water to survive. Just because the water is under your land doesn’t make it self-defense to defend the water; that’s *property* defense, and property is explicitly based on government intervention to say people own things.
The difference between property rights and the current situation in California is that property rights- in theory- a continuous and general set of rules the government sets up, whereas “government intervention” usually refers bureaucrats and special interest groups haggling over individual issues like which crops to grow, hence distorting the market.
Property requires intervention. But there’s a standard way to do it.
Yep. Which is precisely why there’s a strong incentive to work something out that doesn’t require some higher body intervening.
Property rights are a natural emanation of self-defense, and property rights created government, not vice-versa.
If you feed the cows whatever’s cheap, at a certain level of food quality (ground up cow bones and rocks, for example) the cow is going to die. A serious cost ranchers are already facing is between the day to day cost of feeding the cow, and the long term costs associated with attaining the cows health. So if you place an incentive that drives up the cost of day to day feeding, then the compromise line is going to move in the direction of reduced long term animal herd health.
One of the big stumbling blocks to a perfectly efficient market is that some needs are inelastic.
If the demand is inelastic, the supply will change. If the supply is also inelastic, then you’re doomed no matter what you do.
I’ve noticed a lot of people in this thread who seem invested in the idea that California’s doomed no matter what, or at least doomed if it doesn’t follow their favorite policy prescriptions.
While it sounds like Scott is describing something like cap’n’trade for water, a normal market for water would solve the problem more cleanly. As water becomes more expensive, the alfalfa growers either find ways to use it more efficiently, or they eventually get displaced by people growing the stuff in states that aren’t short on water.
Figuring out efficient ways to allocate resources is precisely what free markets are good at.
The problem is that when the market goes free, ONLY the people who can use water most efficiently can afford to do so. What happens if residential uses aren’t cost-efficient enough?
Sorta like how the free market in food means we’re all starving to death because only physicists and oil tycoons can afford to eat?
, but perhaps it means other people are starving to death to increase profits for rich westerners.
This article (there are a bunch of others around, it’s not just one person’s conspiracy theory) makes the case that deregulation of commodity futures trading in 1999 exposed prices of commodities (including food) to a lot of influence from financial speculators, and that this has in fact changed food prices drastically in ways that have nothing to do with the actual amount of actual food actual people actually need or want to eat, and that this has had severe adverse effects on a huge number of poor people.
That article’s behind a get-one-free-then-pay-wall, so here’s another, with a somewhat more rabble-rousing tone. There are rebuttals around too; I don’t want to give the impression that this is an uncontroversial position.
Now, my understanding is that we’re consuming all the food we can produce worldwide, which suggests (1) that maybe lower prices wouldn’t help that much or would even hurt by reducing supply, and (2) that maybe high food prices are simply a matter of supply and demand. But I think the story above is consistent with consuming all the food we can produce (it’s being consumed by different, better-off, people) and with lower prices being better (supply would be reduced, but supply to people close to starvation at prices they can afford wouldn’t).
Here’s a not-so-partisan discussion in a pretty reputable publication; seems to come down in favour of the hypothesis.
Here’s a scholarly article from a reputable research institution that constructs a model that fits actual food prices well, and suggests that the rising trend in food prices is mostly because of corn-to-ethanol conversion in the US, and the huge volatility in food prices is mostly because of financial speculation.
So, Irrelevant: apparently the free market in food means that (not we, but) poor people in poor countries are starving to death, not because only physicists and oil tycoons can afford to eat but because investment banks like to turn a profit and because US politicians need votes from farmers in Iowa.
[EDITED to fix a minor typo.]
I’m not sure what happened to the very start of my comment — perhaps I inadvertently mangled it while editing to fix a typo elsewhere. Anyway, I’ve only just noticed the error and it’s too late to edit now; please imagine it begins “No”.
While there are certainly people starving in other countries, I was under the impression that obesity and diabetes are becoming a public health problem even in third world countries (e.g. India), which seems like evidence that the food supply is growing faster than the population is.
The US government, in its wisdom, bans speculation on onion futures.
Onion price swings make oil prices look tame.
We clearly aren’t consuming all the food we can produce globally. There’s lots of undeveloped land, we grow tons of luxury things that don’t maximize calories per acre, and urban agriculture towers are held back mainly by economic rather than technological concerns.
How many poor people are starving so rich Westerners can eat non-GMO organic heirloom kale?
This is a (semi)serious question.
http://www.fao.org/news/story/en/item/243839/icode/
Those are absolute numbers, which have been decreasing even though world population has increased by over a billion people in the period.
Physicists?
I took a +200% raise by leaving physics.
“Physicists and oil tycoons” was supposed to be a category sketch, not literal. I trust you realize that, but I bring it up because I know there’s an actual name for that sort of idiomatic usage (better/more formal than “category sketch”) and I can’t recall it. Anyone know?
That’s… suspiciously periodic. Since it’s tracking monthly prices, I strongly suspect we’re looking at seasonality; it seems a little unfair to call that a price swing in the same sense that oil has price swings.
Irrelevant: Is the word you’re looking for “metonymy”? It doesn’t fit what you did precisely but it’s fairly close.
Nah, metonymy is way more general.
Why would you expect speculation to increase price swings? A speculator makes money by buying something when it is cheap, selling when it is expensive, which means driving up the price when it is low and driving down the price when it is high–precisely the opposite of what some commenters seem to be assuming.
The belief that speculators cause shortages has to be one of the most lethal errors ever.
@Nornagest: economists have also looked at onion price volatility before and after the ban and found the post-ban period had higher volatility.
David: Omniscient speculators in a perfectly rational market would stabilize prices by the mechanism you describe. On the other hand, in the tragically flawed real world, there is another dynamic that also occurs: a commodity is expensive relative to equilibrium but prices continue to increase because market participants see an upward slope and expect it to continue, producing a boom. After a while the wiser speculators notice how far the prices are from equilibrium and start selling; as the prices go down, everyone else piles on, and you get a crash.
The paper on the other end of the last link in my comment above has a model of speculator behaviour along these lines — a mixture of “follow the trend” and “return towards equilibrium” — and its results seem to match observed commodity prices well. (I am not expert enough to judge how well, or how much evidence that gives us that their model is along the right lines.)
g: You might consider that speculators who behave as you describe on average lose money, and what that implies about changes over time in the composition of the pool of speculators.
The standard economic analysis doesn’t require omniscience, just that the people who are willing and able to bet large amounts on their predictions and continue doing so, are better at predicting than most other people.
That model, of course, assumes speculators are never replaced. If we suppose intelligent speculators have greater longevity than weak speculators (ie they are only replaced after retirement or greatly unexpected shifts), then the market should shift towards a proper speculation-regulated model over time…until this cadre of speculators retires. At which point you start “fresh” again and your progress is reset, since the population of potential speculators has not changed significantly. Meanwhile, you have a sizable population of weak speculators coming in and distorting things until they (the speculators) fail and fall out of the market. Well, them and whatever producers actually making things got blown out by the bubble the weak speculators caused. In any case, the weak speculators are then replaced by a group of new speculators, many weak and some smart. The process then repeats.
It may make sense to subside water for the personal use of the people who really can’t afford it.
Let’s say you could pay water with non-transferable food stamps.
Subsiding farmers who grow crops for commercial use, not even for human consumption, is textbook harmful market distortion.
People who imagine that pricing water means the poor going thirsty have a wildly inaccurate picture of water consumption. Almost none of it goes for drinking.
Suppose the price of water increases enough so that nobody grows alfalfa in California. That means that the 5.3 million acre feet currently used for the purpose have increased in price enough to eliminate the profit currently made on revenue of $860 million. Call that an increase of $100 an acre foot, to estimate high–I doubt profit is anything close to that high a fraction of revenue. An acre foot is about 326,000 gallons. So you are talking about a price increase of about one cent per 30 gallons.
The problem with the bureaucratic system (which is what California now has), is that ONLY the people who can lobby the government most efficiently can get water. What happens if residential uses aren’t lobby-efficient enough?
Oh, I know this one: widespread inconveniencing and costs (water saving toilets, showerheads with low-flow devices that must be manually removed,…), high fines for unapproved uses, and still a water shortage.
*
There are many more restaurants with cheap food than Michelin star restaurants.
I’m not a dogmatic free-markek plusplusgood liberatarian, but I don’t think your objections are particularly applicable to this situation.
1)I have a fairly high belief that water in California will never become so expensive that residents of one of the richer states in the US will suddenly be unable to afford it.
2)Water is an illiquid commodity, ironically enough. Transaction costs are high enough that you can’t really expect water to divert away from California to Israel or Saudi Arabia.
3) The worst case scenario, where water becomes so expensive(like, to pull a number out of my ass, 20 times more expensive) that poorer people can’t afford residential uses of water, is a case where the gov’t could step in and offer “water stamps.” Case transfers are generally better ways to alleviate key resource shortages than direct subsidies, e.g. as fairly exhaustively shown in the work of Amartya Sen
http://www.amazon.com/Poverty-Famines-Essay-Entitlement-Deprivation/dp/0198284632
This seems like a weird problem to worry about because we currently have the situation that government interference has pushed down the price that wealthy farmers pay at the expense of ordinary people who are being harassed with all sort of water regulation.
“the price that wealthy people have to pay.”
Agriculture is a competitive market. Increasing costs by charging for water means that food prices go up. The farmers will sell less and be worse off, but it isn’t as if the extra money is coming directly out of their pockets.
David, it depends on the elasticities, but given the size of the California market (on the producer side) relative to the world market, most of the incidence will fall on the landowners. Some may fall on the employees, but the sign of that is actually indeterminate because we don’t know what substitutions will happen – the farmers could swap out of water-intensive crops into more (or less) labor-intensive crops.
Residential uses cost around nothing. And getting enough water is obviously more important than getting food that takes a lot of water to grow. That’s not the most cost-effective place to cut it.
The advantage of only allowing the most cost-effective use of water is that anything that ends up not getting water is something less important than the stuff that did. The only way residential usage gets cut is if there’s something even more important so it’s a good thing we didn’t save the water for residential uses.
How exactly are we defining “cap and trade”? If you mean give a certain amount to everyone based on how much water they normally use or how many campaign contributions they make or something like that, then I’m against it. I think the term also includes stuff where the government just sells the water to everyone though, so the only distinction from “normal” markets is that instead of setting the price the government sets the amount. I’m in favor of setting the price for CO2 emissions on the basis that the marginal cost won’t change much on the scale of one state, but for water it seems pretty clear they need to set the amount.
And when you displace your alfala crops to other states, you lose that revenue stream, so who pays for the efficient infrastructure of water? People in the cities? Slap the extra charges from the lost revenue onto their utility bills?
I’m certainly agreeing that a more efficient system of using the water is vital, but “let’s drive out our industry” is not a viable solution.
Perhaps the other 99+ percent of California GDP? I wasn’t able to find state-specific stats, but according to the USDA, hay (including alfalfa) totals a 22 billion dollar industry US-wide. Compare that to the state’s GDP at 2.2 trillion, and even if California’s producing all of it (it isn’t), it would be no more than 1% of the state’s economy.
I’m not saying this is a perfect solution. I’m pretty sure it isn’t. But I’m pretty sure “who pays for water infrastructure if the alfalfa farmers don’t” is not a valid objection here.
Nornagest, when I was scrabbling around for statistics, the largest part of the Californian economy is classified “government”.
So industries which generate revenue, including agriculture, even if they’re only a fraction of the 2.2 trillion GDP are contributing to the revenue stream. And if you’re stopping sales of $86 million, and either paying subsidies to not grow the crop or letting the farmers leave or sign on the dole, then you are generating extra costs to be paid out from a reduced tax take.
This means that on top of having to pay for the water as usual, the state government has extra expenses. The farmers aren’t paying (however low a price) for water any more because they’re not using it and they are claiming extra money. The cost of upkeep for the pipes, sewers, treatment plants, pumping stations, wages, etc. etc. etc. of providing water still remains the same as before the alfalfa farmers all stopped using it.
Where is the money coming from to pay the bills? Even if the domestic consumer pays only cents extra per gallon, they’re still going to have to pay more. I’d like to see some pricing on this (I have a particular interest because our current government is bringing in charging for water/wastewater provision, and making a pig’s ear of the whole matter).
The point is, the alfalfa economy is one or two orders of magnitude too small for that objection to bear any weight. Yes, you’re reducing revenue (by how much is complicated, given existing agricultural subsidies, but it’s probably within spitting distance of what you’d naively estimate from industry revenue), and you’re not substantially decreasing the costs of maintaining the irrigation infrastructure. But the difference is tiny and tax revenues in California are comparatively vast. There are any number of things you could do to make up the shortfall, the simplest being charging California taxpayers $2.05 (or something) for every $2.00 used to buy out the alfalfa industry.
Someone has to pay. Someone always has to pay. But in this case, no one has to pay very much.
who pays for the efficient infrastructure of water?
The taxpayers of the U.S. in general, as they already do. Much of California’s water-delivery infrastructure is owned by the U.S. Army Corps of Engineers, and the U.S. Bureau of Reclamation. Most of the rest is owned by urban water districts, the Los Angeles Metropolitan Water District chief among them; the water districts get their revenue from selling water to urban customers and occasional tax subsidies.
They can probably get alfalfa from a state with more water at a slightly higher price.
Are there states with significant untilled / underutilized land AND a lot of underutilized water?
The Midwest has a lot of water (often too much) but most of the attractive land is already being farmed (of course if the price of crops goes up, more land becomes attractive…)
Most of the Western states are pretty dry away from the major rivers – and some of the major rivers (particularly the Colorado) are being diverted to California already. I guess you could turn off the spigot to California and grow in Arizona, but you’d be using the same water.
Caveat: I know very little about farming, but a little about economics and logistics.
I believe that there’s so much farmland in the midwest that they use circular irrigation patterns (http://en.wikipedia.org/wiki/Center_pivot_irrigation) in order to save capital costs, so if they really needed more farmland, they could get it. I think the reason that the Midwest grows mostly soy, corn and wheat, and not vegetables is that corn and wheat are much less labor intensive than vegetables. And the Midwest is less accessible to cheap labor (Mexico/Central America). It’s also further from the coasts, which makes export more difficult.
That said, a quick google shows that Alfalfa should be as easy to harvest as wheat or corn and is already being grown in large quantities in Montana and Idaho, so yeah.
Re: export they’ve got tons and tons and tons of riverine transport options through the Missouri and Mississippi rivers, down to the port of New Orleans. Riverine transport is super-cheap compared to trucking and even railroads.
Center-pivot irrigation happens in the West, not the Midwest. In central Iowa, there generally isn’t irrigation for crops. (Source – my father-in-law.) Further west, they might divert local creeks in ditches, but that’s nothing like the scale we irrigate in California.
The “Midwest” goes from Ohio to Iowa, while center-pivot irrigation starts half-way across Nebraska. (You can check this on Google Maps.)
I figure you feed them something that takes less water, which would be more expensive, but someone just paid them an extra $2.
Also, you’d have to ensure they don’t replace the alfalfa with some other crop.
In addition to feeding them other things, we could produce fewer cows, or import alfalfa from other states.
There’s a whole bunch of solutions to that problem, including but not limited to:
1) Ship alalfa from areas with more water.
2) Feed the cows a less water-intensive crop.
3) Shift the cow-raising activities to areas with more water.
4) Have people eat less beef and dairy.
5) Keep growing alfalfa, but reduce production of some other crop.
6) Have residents take showers every other day, and reduce consumption of foods that waste water, like pasta.
The regulatory solution to that is to hire some fairly smart people to make the decision, then have all those stakeholders hire lobbiests to present their evidence to those smart people/protest in the streets to explain why they need more water than another group/etc. Then the smart people decide who should get less water, and they try to enforce that rule somehow.
The market solution is to let people pay a little more for water, and people who don’t want as much water as they are currently using will find another way to make things work.
I generally default to the market, but the argument against it is “I personally think that people should eat more yogurt and take shorter showers, so I would prefer a rule that takes water away from shower takers and gives it to yogurt raisers, and your precious ‘market’ might result in more showers and more expensive yogurt.” The problem is that it’s hard to know which of those yogurt versus shower rules are correct, and that sooner or later, you get buried in rules that have consequences you’ve never thought of.
Someone may have said this already but if you’re going to subsidise alfalfa farmers you could also just subsidise the dairy producers to buy the alfalfa they need from elsewhere. Even if it costs twice as much as getting it from California farms you’d still only be paying 2-3 billion to solve the water crises (or i.e. 80 dollars per Californian).
If you no longer grow alfalfa, the cows will have to be fed something else. I don’t know what, but it will be more expensive than alfalfa, which means the price of beef will go up. That might be fine, but it’s a cost you’ll have to take into account too. (Or is it? Someone with better economics intuition can correct me. David Friedman, help?)
So stop feedings cows alfafa AND stop giving out ethanol subsidies.
Whatever the cows eat will probably be more expensive in terms of dollars paid by the ranchers, but may not be as expensive in total cost – since whatever they eat instead may not have as much subsidized water included in it.
The accounting should take into account the higher prices of cow-derived goods, of course, but the distortions from the subsidies make this a complicated calculation.
You don’t have to pay the revenue of the alfalfa farmers only the profits, even then you only have to pay some of them because you’re substituting a risk-bearing return for a risk-free return. The consumers of alfalfa will substitute at a marginally higher price input but that is only a small fraction of the cost of producing beef so beef gets marginally more expensive. The overall cost to consumers will be much less than $2 each.
That’s all in a free market where farmers would have the right to choose whether to buy their water and make a crop, sell the water or anything in between. The water market in California is hardly free, there are layers of water rights and water districts that prevent the free exchange of water.
You don’t have to pay them ANYTHING – just stop insulating them from the true cost of their inputs. They’ll be forced to either raise prices, find a less water-intensive way to farm, find another career, or some combination of the above.
And when you go down to the supermarket and find the price of a head of lettuce has gone up dramatically (not by a few cents)? Vegetables? Fruit?
“True cost of inputs” will come from the consumer. Farmers have to raise prices, they’ll be looking for higher prices from the processors, who in turn will be charging higher prices to the retailers, who in turn will charge higher prices on the shelves and cabinets.
There’s an argument to be made that Americans are paying (by global standards) artificially low prices for their food, as for their petrol. “True input costs” might come as a shock to more than alfalfa farmers.
As for “finding another career”, this is already happening. American farms are being consolidated into large agri-business industrial producers or small hobbyist farms, middle-sized farms are declining. It’s the huge agri-industry farms that are the ones hoovering up the subsidies and the water.
A price increase isn’t necessarily a “cost” in a social sense. Because the price of water, and thus alfafa, and thus beef is lower than the cost society pays to provide it then people will eat too much beef. Suppose:
– The cost society pays to produce a pound of beef is 7$
– The amount I’m willing to pay for it is 6$
– The list price, thanks to artificially cheap water, is 5$
So I buy the beef and come out 1$ ahead, while everyone else as a whole comes out 2$ behind. Because the price was too low the world is a dollar poorer. Roughly speaking.
If California runs out of water altogether then there will be no more water to grow the alfalfa, so the price of beef will go up.
It is of course desirable to take all the costs into account, but when it comes to environmental resources in limited supply (even if we can’t measure that limit precisely) eg the maximum sustainable fishing catch, or the total amount of water, then a lot of times not taking an action just means delaying the inevitable.
In a previous post, I calculated a very rough figure of a price increase of about .03 cents per gallon to drive out the alfalfa farmers. That’s almost certainly much too high, since if you actually increase the price the result will be to make people economize on water usage in whatever are the least expensive ways, which is unlikely to be by abolishing one crop—call it .01 cent/gallon for a simple back of the envelope guess, I suspect still high.
To convert that to an increase in the price of cattle you would need to know how much water they consume, both directly and via their feed. A little googling suggests that direct consumption per animal is probably on the order of a few thousand gallons, so that would increase the cost of producing cattle in California by something under a dollar an animal. I don’t have numbers for the indirect water consumption.
If you had all the figures, that would give you an upper bound on the increase in beef prices, since as the cost of water in California rises producers will substitute to less California water intensive ways of producing beef.
Maybe I missed something, but what does the vertical dimension in your chart map to?
Confused about that as well.
Nothing. Including it was probably a bad idea – I was going for something like a visualization of “how much is left after x is taken away”
Yeah, I was surprised that you made a mistake in your infographic that is usually done on purpose by data-distorters, since you always seem to be on the lookout for distortions in statistics.
People estimate the size of regions on charts based on their areas, not linear measures of their height or width. For example, if you represent numbers with different-sized circles, people think circles with twice the diameter represent numbers four times as large, because they have four times the area.
Because of this principle, your chart initially overemphasizes the water usage in categories on the left, because they are especially tall. For example, water usage for Agriculture (34 MAF) looks about 4 times as large as for Wild Rivers (23 MAF), instead of only 1.5 times as large. Readers will eventually notice that the areas aren’t meant to correspond to the numbers, but the lack of obvious meaning of bar height will still confuse them.
One clearer representation would be bars on a bar chart with equal widths and varying heights, where sub-categories are closer to each other. A more interesting representation would be a treemap, with thinner borders between sub-categories than between main categories.
As a side note, I would also recommend defining “MAF” somewhere on the chart, rather than defining it in the middle of the second paragraph after the chart.
Something like “(MAF = million acre-feet)” right below the “Total (80 MAF)” at the top, the first usage of “MAF”. That’s only necessary because the units are so obscure – “kWh” would be less likely to need explanation.
Except that this chart does use areas.
No, it doesn’t. If areas were used, wild rivers (23 MAF) would look larger than urban (9 MAF).
oops.
So width is the amount of water and height is the anti-cumulative water remaining?
IT MADE SENSE IN MY HEAD.
(I fixed the graph)
I take it that if a three dimensional object is represented, the size is interpreted as being the volume? Although if the depiction only weakly gives the impression of being 3D, it would probably lean more towards areas.
I’m not sure defining MAF is all that important, since it’s the relative size that matters. And it’s not too obscure in the water context; I was able to figure it out.
If you’re going to adjust your chart based on reader comments, you should also make a note of the revision history, so as to reduce confusion created by comments referring to the chart as it existed at the time of the comment.
Agriculture in California is crucial for ensuring that enough votes get to politicians with (D) after their name because it provides jobs for the first migrant in a line of chain migrations.
Agriculture can never be cut because of this.
Or, if you wanted to put in other terms, because agriculture is a hugely profitable industry in California and there are many politicians who represent districts where it’s one of the primary drivers of economic activity.
You really don’t need to go to talking about immigration in order to explain why California politicians are unwilling to piss off farmers. It is, I think, pretty easy to explain simply as politicians acting in the interests of their constituents and / or donors.
California agriculture is only 2% of the state economy.
It is a $42.6 billion dollar industry – smaller than Google’s revenue of $66 billion, or even Apple’s $52 billion operating income.
And much smaller than the USA’s GDP of $17T.
I don’t think that’s really inconsistent with what I said. I mean, that’s a lot of money, and it’s a lot of jobs, and it’s a lot of votes in agricultural districts. I’m not sure how useful the comparison to a pair of multinational corporations is, in the context of California politics.
California has been losing native population for decades and yet it grew by about 10 million people in 25 years anyway due to immigration. It went from consistently supporting one national party in presidential elections to supporting the other.
That is something that doesn’t happen by accident and it sure as hell doesn’t happen because some industry that’s 2% of the state economy and dwarfed by the income of two corporations in the same state wanted to save a few bucks on water and labor.
Solving the water crisis interferes with that project because as unimportant as the 2% of the state economy it still employs the first wave of immigration that allows the rest.
Basically you could solve the water crisis if the Democratic party was allowed to collect votes in Mexico rather than having to import Mexicans to collect the votes. Then you could cut back on agriculture’s water use without worrying that the voters will move back to Mexico.
@Steve Johnson:
When was California a conservative state? It may have had nominal Republican government, but it’s not been a conservative state.
And immigration from where? Mexico? Or India? Or elsewhere in the US? My prior for CA is that the immigration picture there is not simply agricultural.
California went Republican in presidential elections from 1952 to 1988, with the exception of the 1964 election. However, Reagan and Nixon were both California natives, and presidential elections have historically shown a strong native-son effect.
Before that, it went Democratic from 1932 to 1948; and before that, results from 1852 to 1928 are mixed but show a definite Republican lean. I’m not sure how well pre-WWII (or even pre-Sixties) results map onto current politics, though.
@Nornagest:
That really has very little bearing on whether CA was a conservative state. Pre-1988 is before the great sorting.Heck, Nixon essentially put Medicaid in place. Republican didn’t mean conservative back then.
Steve Johnson: you are wrong. The water-intensive crops are less labour-intensive; the labour-intensive crops (which need the cheap migrant labour) use less water.
So you can “solve the water crisis” or “send the illegals back home”, but doing one will not do the other. In fact, if the water-guzzlers like alfalfa are stopped, it may well be that the land (instead of lying fallow) will be converted to other crops – such as the less water-consuming, more labour-consuming, requiring cheap labour which means migrant labour, varieties.
HeelBearCub – California is the home of 1976’s Proposition 13, where the people voted to cut their property taxes by two-thirds in one swell foop. It’s the state that elected Ronald Reagan governor twice, after he got tired of making ads promoting capitalism for GE.
In 1994, California voters passed Proposition 187, which would have denied most government services to illegal aliens, and Proposition 184, which toughened up the “three-strikes” law passed by the legislature (under public pressure) earlier that year.
Even as late as 2008, California voted to put its ban on same-sex marriage into the state Constitution. In the same election, voters defeated two propositions which would have softened the “three-strikes” law.
These are not signs that California is not a conservative state.
@Anthony: Pointing at the result of direct referendums is hardly evidence that allows you compare California to other states and say whether CA is more or less conservative than other states, as most other states do not have that system in place.
But interestingly, the fact that the direct referenda system exists is actually a point of evidence in favor of the idea that California has long had a progressive agenda, as they were passed into law in 1911 during the Progressive era as a means to break the the control of party bosses over policies and elections.
Agricultural districts are probably more conservative.
Isn’t it so in California too that urban areas are mostly Democratic, and rural areas are more Republican? And don’t they also have obnoxious supermajority requirements that makes it very easy to block legislation but very hard to pass it?
Where in the US is that not the case?
Well in that case, it’s not necessary to assume a Democratic conspiracy to replace the population, is it?
Labour-intensive market gardening depends on cheap labour, which is immigrants. Crops like wheat, soy and alfalfa which can be machine-harvested are much less labour-intensive and so don’t rely on cheap labour.
Do you like tomatoes?
Another thing, is the variety of rice we grow some special strain or something? I understand that typically rice needs to practically down to grow well, and also that CA is one of the top producers of it, and that drought years are more common than not.
Surely one of these things I think I know has to be wrong?
Subsidies at work again. We need to support American rice farmers and not outsource jobs to other countries, even if the environment for rice in CA is horrible, goes the logic.
I think that alfalfa requires 5 feet of water, while rice requires 10 feet.
Well then, let’s do both! 😆
Does it matter how deep the water is? 10 feet deep water doesn’t evaporate any faster than 5 feet.
I’m not sure he meant all at once.
I think the value is “10 acre-feet per acre per year”.
It matters how much water is sucked up and bound by the plant too. I assume rice does that more, since it needs more.
I have been told that rice doesn’t actually need to grow wet; we do that because rice can grow in water and most other plants can’t, so weeds aren’t an issue.
I cannot personally attest that this is true. :/
I suppose before I say how inefficient that is I need to know how much water goes into making weed killer, etc.
I doubt it. Anyhow, rice in California is grown under five inches of standing water.
The One Straw Revolution— a book about growing rice without starting it underwater.
Rice fields are typically flooded, and in most places, by several feet of water. Thanks to the development of laser-leveling technology for farm equipment, it’s possible to grow rice by flooding its fields to about 4 to 6 inches (10-15 cm), and achieve similar results. Other commenters mention something about drowning weeds; I suspect that application of 4 to 6 inches of water is sufficient for that, but not necessarily enough to keep the plant healthy from seed to harvest.
One useful resource on the general topic is David Zetland’s “Living with Water Scarcity” which is available as a free .pdf at http://livingwithwaterscarcity.com
When doing these calculations, I think you need to account for the other side of the equation – namely those that are currently purchasing that alfalfa. If the beef industry truly needs that much alfalfa (I have no idea if there are less water intensive replacement crops, or other alternatives) you’re just going to create a bidding war between people trying to buy “water offsets” and people actually trying to buy alfalfa. I don’t know where that bidding war would end, but it would certainly complicate the equation.
Well, the government can mandate that alfalfa farmers take their subsidies, so they’ll probably win.
No, it won’t be a bidding war between alfalfa and water offsets, it be a bidding war between water and alfalfa. If I pay $2 to use X more water, then X water will be removed from the market, and either alfalfa farmers or someone will be bid out of the market. Someone, somewhere, will be out X water, but will be compensated for the loss with $2. And we can be confident that someone will willingly make that trade, because otherwise the free market price of water would be more than $2 for X.
Well, that’s why we have markets. Markets deal with limited resource availability all the time. They’re basically a massive computational device. If someone had invented them consciously we’d admire them as a genius.
California gets about 80 million acre-feet of water a year. California can’t use more than that in the long run. So things are going to shake out sooner or later for the alfalfa industry, or, if not them, some other industry or service (Hollywood?). Anyone with an ounce of environmental responsibility hopes for sooner.
I was in Brisbane (Australia) for the drought and water shortage we had there about a decade ago. Through extensive public campaigns and water restrictions — we were strongly encouraged to limit showers to 4 minutes, and at the peak of the drought we couldn’t use hoses or sprinklers outside, only buckets — average residential water use fell to 140 litres per day. And it was annoying, and green front lawns became brown, and generally it’s not something that you feel a first-world city ought to have to do, but it wasn’t really that bad a hardship.
Converting your figure of 6.6 mega-acre-feet (acre-feet is the worst Imperial unit I’ve ever come across, BTW) to metric, and dividing by 40 million people and 365 days, currently Californians use an average of 560 litres per day. So I’d guess that there’s scope to solve much of the water crisis through reduced residential usage.
(Yes, it’d probably be preferable to have better-functioning water trading, and it seems inequitable to make urban residents make most of the adjustment when they use a small minority of the state’s water. But I think the Brisbane experience shows that markets aren’t always necessary to solve scarcity problems.)
I’m not sure all the lawn water should be considered residential (I’m assuming you added the 3.8 MAF for lawns to the 2.8 MAF misc. residential to get the 6.6 MAF figure).
College campuses, businesses, and other impressive-looking institutions seem to have much bigger lawns than most people would care to water for their private residence. Of course, this is based on my experience from not-California. Maybe in California only private homes keep lawns?
You could be right, I don’t know — I’m coming from my Australian perspective where almost every house in the suburbs has a front and back lawn. (Gradually changing as we move to higher-density housing.)
Aha, buried in Scott’s “Urban Water Use Efficiency” source (pg 8), we get a figure of 2.7 MAF “residential internal” and 3.0 MAF “residential external”. So most of it is residential after all.
(Edit: most of the water used for lawns, in case the context is unclear)
My university, along with a lot of public lawns and roadway landscaping, uses recycled water (recycled to plant-watering but not people-watering standards) and has some fairly nice landscaping.
Most Californians would be happy to *water* corporate-campus sized lawns, if they had that much land and the water was cheap enough. Not nearly so many would want to *maintain* such huge lawns. In general, suburban single-family lot sizes in California are smaller than in the East, but need more watering because even in wet years, it doesn’t rain from May through October.
That.
Mind you, CA lawns never made any sense to me, since they need non-zero maintenance, but aren’t big enough to actually be usable.
Either give me my half-acre where I pay the kid down the street $15 to mow it every week, and throw down some mulch one weekend a year, but in the meantime, have what is basically my own de facto private park, or give me my cool condo in an interesting place where I don’t have to do any yardwork at all.
Anything that cannot be sustained will not be sustained, so markets are never “necessary” to solve scarcity problems. They are simply faster and more pleasant.
That depends on what you consider a “solution”. Is “a large portion of the population dies from starvation, and the food supply is sufficient to keep the survivors alive” a solution to famine?
yes, starvation is a solution to famine, although it is a very bad solution.
I think most people would consider preventing starvation to be a terminal goal, and addressing famine to be an instrumental goal, so it’s a bit odd to refer to abandoning the terminal goal to accomplish the instrumental goal as a “solution”.
It’s not about what you want, it’s about what the great formless thing wants.
I think most people would consider preventing starvation to be a terminal goal, and addressing famine to be an instrumental goal, so it’s a bit odd to refer to abandoning the terminal goal to accomplish the instrumental goal as a “solution”.
History shows that you’re wrong. Plenty of Communists don’t believe that preventing starvation is a terminal goal.
I believe the medical phrase is “all bleeding stops eventually.”
I think he’s saying that you could alter the state intervention to fix the shortages. For example, by restricting agricultural water usage. It will take longer and be more unpleasant.
If you mean me, I’m emphasizing that there is no option in the matter. The worry isn’t whether we can solve things without trade, the worry is that the method of solving things without trade is to bang our non-negotiable requirements against each other and see which breaks first.
“acre-feet is the worst Imperial unit I’ve ever come across, BTW”
How do you feel about kW Hr/year?
Ha! The funny thing is that it feels perfectly natural when the sticker on my fridge tells me how many kWh it uses per year.
My problem with acre feet and such units is that of conversion factors.
1 acre feet = 325 851.429 gallons.
A more sensible unit would be something that just added or subtracted a number of zeroes. So a MG or mega gallon would be much easier to use.
In the metric system, the basic unit is a litre. For large volumes you would use m3 which is just 1000 litres.
For water and land use, you would use mm per hectare which is 10 m3. (or possibly cm, dm or m/hectare which would add 0, 00 or 000 respectively)
keeping track of zeroes is easier than keeping track of arbitrary conversion factors.
Now, kWh/year is a bit of an ugly because both hour and year are horrible units that does not follow the numbers of zeroes scheme, but I can’t say it’s actually worse than acre feet and changing the definition of a year to something sensible will be an even bigger challenge than getting Americans to use the metric system.
You mean like using directed nuclear blasts to slow the planet’s rotation so that it spins on its axis 100 times for every orbit of the sun? I want to be uncharitable and say that still sounds easier than getting everyone to adopt the metric system 🙂
It should be 101 times, if you want 100 solar days per year.
Curses. I fail at mentally rotating shapes in the most paradigmatic way possible.
Metric is great in base ten. If we weren’t base 10, metric would be awful.
And given that our brains aren’t intrinsically base anything, I don’t know that metric is absolutely better in an objective sense. There is a post I can’t find arguing that Farenheit is objectively better than Celsius.
Past the number 3, there is no hard-wired understanding of numerical concepts in the brain. We are hard-wired to be able to discern ratios of large numbers, but not the numbers themselves.
A constant base is better than different bases depending on what level you’re at, e.g. 16 ounces per pound, 2000 pounds per ton. And what does a hypothetical world in which we weren’t base have to do with it? In that universe, they would have picked a different base for their version anyway.
And the choice between celsius and farenheit was arbitrary. It wasn’t really determined by the previous choices in the metric system.
Metric and Conventional are optimized for different things. Conventional optimizes for providing a selection of units convenient to particular usage domains, while Metric optimizes for ease of conversion between usage domains.
The utility of the former is demonstrated by the concept creeping in around the edges in Metric/SI usage. In various domains:
– parsecs are used instead of petameters
– AU are used instead of gigameters
– calories are used instead of joules
– electron-volts are used instead of attojoules
– Atomic mass units instead of yoctograms
@RCF: I was merely pointing out that the preference for metric is based on a learned familiarity with base 10. For those who learned a familiarity with non-metric measures, the ability is not hampered at all by the use of teaspoons, tablespoons, 1/4 cups, cups, pints, quarts and gallons in the recipes, despite their lack of base-10-ness.
The recipes tend to make good food in the right amounts. I don’t really need to scale those measurement by a factor of 1000 (and recipes don’t work that way anyway).
You originally said that if weren’t base ten, then metric would be awful. But then you say that we deal just fine with other bases. Fluid units are mostly base 2, but there are three teaspoons in a tablespoon. So if you like base 2, fluid measures would be improved by having the unit below a tablespoon being one fourth of it.
If I were designing a system of measures, I might be tempted to use base 12: it factors cleanly into halves, thirds, quarters, or sixths, whereas base 10 can only do halves or fifths.
@RCF:
If you were familiar with base-8 and not base-10, trying to use the base-10 metric system would be hair-pullingly awful. Using the “octic” system would be pleasant though.
That was my original point, not that a base-8 based measurement system would be somehow intrinsically bad.
— ISOMORPHISMeS, “Why Fahrenheit is Better than Celsius”
The thing that post doesn’t consider is that there’s no choice about Kelvin/Celsius if you want to do sciency things, so a fair few people are going to learn Kelvin/Celsius just for that.
Doesn’t matter if Fahrenheit uses its digits more efficiently if the entire measurement system is extra information.
You could only teach Kelvin/Celsius to scientists/engineers/other people who need to do science/maths things with temperature, but I’m not sure you want scientists/engineers to be non-native speakers of their measurement units, and now there’s an extra translation step in-between science-stuff and normal-people-stuff.
(Also, celsius can represent every outside temperature you’re ever going to run into on Earth in two digits and a potential negative. Fahrenheit needs three digits. Celsius is more efficient in that sense, which is the exact opposite of what the blog post is talking about – efficiency is in the eye of the beholder and/or use case)
Don’t think it’s entirely fair to equate Kelvin with Celsius. The unit size might be the same, but the unit size is the least interesting and least useful thing about a measure of temperature.
If you know one, you almost certainly know the other (or can learn it trivially – if you’ve only been taught kelvin (and what anchors do they give you for the scale?), then you only have to remember “subtract -273” after all).
A lot of science on temperatures can be done equally well with celsius as with kelvin (how much energy it takes to heat something up by some amount, for example).
Let me put it this way: You must know Kelvin to do science involving temperatures. You’d also like a scale for non-science uses, but preferably it should be moderately simple to translate between them so that lay people understand science stuff easier and scientists don’t have to carry two completely different measurements around in their head. Celsius fits that description (unless you prefer Rankine to Kelvin, I guess, but that way insanity lies).
But acre-feet are perfectly sensible: there are 43,560 square feet in an acre, and thus 75,271,680 cubic inches in an acre-foot; and one (U.S.) gallon is exactly 231 cubic inches, so it’s possible to do exact conversions, giving you 325,851 3/7 gallons per acre-foot.
Actually, the main benefit of a unit like acre-feet is that if you have a sense of one acre, it’s very easy to visualize one acre-foot. (One acre is between 3 and 10 suburban home lots in the U.S., or about 3/4 of a U.S. football field, or a little more than half a regulation FIFA pitch.)
At my workplace, I routinely use units of W/in^2 (“Watts per square inch”) and W/in^2.K^4 (“Watts per inch squared, Kelvin to the fourth”).
That’s more about unusual dimensions. The actual units you’re using aren’t terrible, although W/m^2.K^4 would be more sensible.
A watt is kg m^2/s^3, so one could argue that the proper unit would be kg Hz^3 per K^4.
Acre-feet are an incredibly useful domain-specific unit for this purpose. They don’t convert easily to gallons or liters, but you don’t need to do that. You need to answer questions like “How much water does California need for rice farming?” Answer: “the number of acres of rice, times ten feet.”
“How much water does Australia need for rice farming?”
Answer: “The number of square metres of rice, times three metres” works just as well, but isn’t weirdly rectangular and generally all imperial.
(Answer: More water than Australia has, that’s for damn sure. Seriously, guys. Indonesia is right there. Just trade two sheep for a rice and a wood.)
Australian land is measured in square meters; American land is not.
Just ran the numbers. One million acre-feet is almost exactly 0.30 cubic miles. (Or 1¼ km^3)
Its pretty functional if you are farming. The amount of water you want is based on your acreage and how deep you want to flood your field, which will depend on the crop, the time of year, etc.
Like most imperial units, it was selected for ease of use in a particular setting, not for high interoperability.
Much of the alfalfa grown in California is for export, not local consumption. China is a big buyer. The same is true for much of the other ag outputs, though the destination for those exports also includes other US localities.
IOW, it’s not an issue that is usefully analyzed as just a local issue.
It is also worth considering that there are few places on the planet where water yields so much food, fuel and fiber. Due to a benign climate several crops a year come from each plot of land. There’s little bare soil or fallow time so long as there is water. The assumption that each bit of land is fungible leads to fantasy analyses.
It may be worth stepping back and asking if this is truly a distribution/allocation issue at all? A few desalination plants could ease all of the urban needs but in California they build bullet trains which cost 68 times as much as a desalination plant. A city perched on the sea shore really has no excuse for not being self sufficient in water. The same tech will also ease their waste water issues.
Well, if you want to get technical about it, I’m not sure we’re actually building any bullet trains.
We’re certainly spending money on bullet trains but I’m not sure there’s any actual building going on.
If you want to be really technical, I think we’re building trains, but not much in the way of track.
If desalination plants are a cost-effective way to get water to California agriculture, then how about we charge market prices for the damn water and see what the revenue figures would be?
Then maybe the farmers should pay for the water they’re using!
Someone has to use less water, or more water has to get into the system.
Charging market prices for water means the price must go up until enough people can’t afford the water they’re currently using that demand for water roughly matches supply. That means either some really productive farmland is no longer being used, or some residents have drastically cut back on water use.
The argument being made above is presumably that both of those options suck and building a desal plant is better. I don’t actually know much about how functional desalinisation plants are, but hey it’s a possibility.
(Incidental: Pssh, California has it easy. The Australian version of this problem crosses several state borders, and we’ve got the same outcome of agricultural uses, mostly people growing stupidly water-intensive crops, are the largest component. Farmers are a rather large voting bloc.)
Or maybe the really marginal farmland is laid fallow. Or maybe the farmers invest in irrigation systems that use water more efficiently. Or maybe they will switch to crops that are slightly less profitable but require much less water.
With water massively subsidized, do you really think farmers are using it as efficiently as they possibly can?
Bonus points: we don’t even have to figure out how! That’s their job, and with market prices they’d have the incentive to actually do it.
Let’s not pile on the farmers exclusively; drive the alfalfa farmers of California out of business and a company like Monsanto will scream like a fire alarm, because that’s a huge proportion of its market for seed and pesticides you’ve just forced to go belly-up.
You really think their lobbyists aren’t out there working for subsidies and supporting large farmers’ organisations?
Or maybe people who like dairy will pay a little more for it, and some people will consume a little less. It all works out.
(Though the dairy market is itself highly regulated, which imposes its own distortions).
If it uses water worth more than what it’s producing then the farmland isn’t actually productive.
-That means either some really productive farmland is no longer being used
Does it really count as productive farmland if there are periodically multi year droughts?
“That means either some really productive farmland is no longer being used, or some residents have drastically cut back on water use.”
Probably not. It probably means that farmers switch either crop or irrigation technology to consume less water per acre. Leaving productive land vacant is unlikely to be the optimal solution.
People, especially non-economists, tend to badly underestimate how flexible real economies are because they implicitly assume that how things are being done at the moment is the only way they can be done. In that sense, almost everyone is a conservative.
Bastiat’s unseen: laying fallow in the imagination of humanity since, like, history.
Well, with the water ‘debt’ from draining the aquifers, they’re going to hit that barrier anyway. One might say the farmland’s productivity is artificially high.
That means either some really productive farmland is no longer being used, or some residents have drastically cut back on water use.
Third option – agricultural users figure out how to pay for using less water for the same (or pretty close) productivity. Underpriced water leads to huge waste, but using less water requires spending some money; either on equipment, or on people supervising the actual water use in the field. If the price of water goes up, some of those costs become worthwhile.
Desalination of one acre feet has a cost of roughly 1000 USD according to wikipedia, so cost of water would be roughly 10x the value of the alfalfa.
Probably not cost effective.
It’s worth pointing out that Saudi Arabia grew wheat with desalination-produced water for a while, and gave it up a decade ago as uneconomic.
Generally, energy-intensive things that Saudi Arabia gives up as uneconomic probably don’t need to be reconsidered elsewhere.
I always liked desalination plants as an idea, so I tried to run a few numbers using the Carlsbad plant under construction in San Diego as a guideline. The numbers look, at a glance, fairly similar for the Huntington Beach plant also being built in California.
Lots of assumptions go into these numbers, and I have no special/inside insight into desalination costs. Just setting up a few ballparks.
The Carlsbad plant is listed as 50 million gallons per day facility. Or, in other words, ~153 acre-feet per. day.
How many plants needed will depend on how fast California wants to make up the shortfall, and how much extra water capacity California wants to add to the system. Arbitrarily, let’s say that California wants this to be a near-term solution. So, let’s say they want to make up the shortfall in six years. The California plants have ~3 year construction time, so they would half three years with which to run the new plants.
Let’s assume California can get as many desalination plants it wants working at ~153 acre-feet per. day. I’m not sure how much prime desalination real estate exists in California, but it’s got 840 miles of coastline. Hopefully that’s enough. Assuming the 6 million acre-ft shortfall does not grow, that would mean ~36 plants operating for three years.
The cost of Carlsbad desalination plant seems to be right along with the implied ~$1 billion. So, the cost of making these plants would ~$36 billion.
Then California would have to power and operate the damned things for three years. Unfortunately, I’m having difficulty finding any good numbers on commercial energy prices and non-energy operating costs.
So, let’s look at a different ballpark where California “incentivizes” the private market to build the damned things for them instead.
Let’s assume they would all charge California at roughly the cheapest price Carlsbad has agreed to, $1849 per. acre-ft in “2012 dollars”. That is ~$11 billion to make up the shortfall. Add another half a billion if we assume they’re using the CPI to adjust to “2015 dollars”, so $11-12 billion. Only one sixth the project cost to build the high speed railway.
Granted, if only those two plants were built then it’d take ~54 years or so to make up the shortfall.
Sources:
http://poseidonwater.com/
http://www.sdcwa.org/seawater-desalination
http://en.wikipedia.org/wiki/Carlsbad_desalination_plant
I didn’t check your numbers, but to me that read as if you propose to cover a yearly shortfall over 6 years?
If you want to cover the yearly shortfall, surely you need to do it every year?
Did I miss something crucial?
It wasn’t clear to me whether 6 million acre-ft was the yearly shortfall or the current cumulative shortfall. I assumed it was the latter, and that California would have some leeway in how long it takes to make up the shortfall.
Most of the data is in precipitation rates, but I found California reservoir levels listed on the US Department of Agriculture website. Properly we’d want to do some sort of running average and/or do a fit to the data to estimate the yearly shortfall. But I don’t want to spend the time copying all the appropriate numbers right now. So let’s simply take January 2014 and January 2015 reservoir levels as the guide (where reservoir levels being equal across years is taken as no shortfall). In that case, the yearly shortfall is ~587,300 acre-ft.
Carlsbad produces ~153 acre-ft per. day, or ~55,845 acre-ft. a year if it can run non-stop. So, if California wanted to make up that alone (letting the 6 million acre-ft deficit stand), then California would need eleven Carlsbad-like desalination plants. That would give California a little bit of overhead.
So, ~$11 billion to build the necessary desalination plants to make up the yearly shortfall (assuming California has enough viable locations for eleven Carlsbad-like desalination plants). About as much as it would cost to purchase the 6,000,000 acre-ft water deficit from Carlsbad-like desalination plants.
Edit: Whoops, forgot the number for if my assumption is wrong about the six million acre-feet. If the yearly shortfall is 6,000,000 acre-ft, then it would take ~108 Carlsbad-like desalination plants to make up the yearly shortfall.
Source:
http://www.wcc.nrcs.usda.gov/basin.html
Kept running into the unhelpful 11 trillion gallons (34 million acre-feet) figure, but I finally found confirmation that it’s 6 million acre-ft per. year shortage. “Sustainably provide” is the mystery criterion provided, which makes it sound like 1-2 million acre-feet shortfall if you’re willing to let the Sacramento-San Joaquin estuary die.
So, that’s ~108 Carlsbad-like desalination plants necessary to make up the difference on their own.
Sources:
http://www.nasa.gov/press/2014/december/nasa-analysis-11-trillion-gallons-to-replenish-california-drought-losses
http://pacinst.org/publication/ca-water-supply-solutions/#issuebriefs
Here’s a suggestion to make your chart more readable: attach the labels to the sides of the bars rather than to their bottoms. I think the reason it appears confusing is because readers are used to bar charts comparing absolute figures, rather than cumulative sums. You could also put a “}” bracket next to one or all of the labels to make it clear that you are labelling the gap between adjacent bars, not the absolute height.
Also, it is more typical for “0” to be located at the bottom of the graph, rather than the top. If you flipped the graph vertically, I think it would aid in comprehension as well.
EDIT: Taking a closer look at the graph, it seems like there’s some sort of error. If the vertical axis is supposed to be cumulative amounts, why does alfalfa correspond to a jump on its right-hand side, and “all others” correspond to a jump on its left-hand side?
I would suggest just reducing it to a 1d graph If you just present the horizontal axis on its own, it’s easy to visually compare the cumulative amounts without introducing a second axis with different scale (and some sort of plotting error?).
As other people point out, that diagram needs quite some work. In general we completely suck at estimating areas or volumes. Lengths is what we’re good at. It’s not at all obvious that wild rivers is the second largest factor for instance.
The order of things make no sense – the major groups don’t come in order of size, and the subgroups within each group have different order. And especially bad, the lengths of each bar is _not_ proportional to the magnitude – Alfalfa is the most water-using crop, but has the shortest bar.
I’d suggest using stacked bar diagram instead, with the bars from largest to smallest group, then the subgroup stacked with the largest at the bottom and smallest at the top. With wide bars you have room to put the labels right in each box if you want to.
Here’s a quick revision as a stacked bar for comparison: http://i.imgur.com/ZzGa4Dt.png
Edit: the original chart confused me more than I thought; lawn and plumbing values should be reversed. This is likely to have other errors as well; caveat emptor.
I too am confused about whether we care about the area or the width of each line. And what does the height represent?
This is a good straightforward post. I think you’re basically right that economists having it about right… the only miss here, is what ‘The Economist’ has to say on the matter.
http://www.economist.com/news/united-states/21647994-why-golden-state-so-bad-managing-water-price-wrong
The key piece the above article adds to the picture relates to a potential role for desalination plants. California, does after all sit right on the ocean. Here is The Economist’s take on that.
“Other places have dealt with drought better than California. Israel, for example, has built large desalination plants that helped the country, which is 60% desert, cope with a seven-year drought between 2004 and 2010 and the driest winter on record in 2013-14. In California, desalination is harder because electricity is costly, thanks to a renewable-energy programme. And green rules make building anything slow. A company called Poseidon will this year complete a $1 billion desalination facility to increase San Diego’s water supply by 7%, but only after six years of permitting and litigation. Many other desalination projects around the state have stalled or simply been abandoned.”
Keep in mind that Israel has ~8MM people and California has ~ 38MM people so there is a scale difference but its not even an order of magnitude.
(One post script: for those concerned about pollution associated with heating ocean water I see no reason why the desalination couldn’t be paired with solar technology for energy generation – California has near ideal amounts of sun for solar, solars cost has plummeted in recent years, and finally intermittent generation and energy storage– the typical problem with renewables really isn’t a problem for creating ‘fresh’ water from salt water, like it is say, for running air conditioners, or automobiles. If we were talking about using solar 10 years ago for energy generation it would be prohibitively expensive but due to exponential productivity growth (ala Moore’s law) solar costs have come down dramatically. Indeed my knock on The Economist as of late has been that I get the feeling it’s left hand and right hand don’t realize they’re hooked up to the same magazine. See graphic at http://www.economist.com/news/special-report/21639016-biggest-innovation-energy-go-without-invisible-fuel for a quick insight as to costs associated with solar per kWh over the last ~ 5 years. Again these stats can be misleading because intermittent generation and storage issues are the Achilles’ heel for renewables (in the same way that not knowing how to swim can lead you to drop in a river that is on average 1 foot deep), but I don’t think they apply in the case of desalination.)
I suppose that if there’s large overhead costs that apply even when the plant isn’t operational, then the intermittent nature of solar could be relevant.
Fair point. In terms of either fixed costs during downtime (which I am skeptical of) but also huge capital costs up front. In the cited case they spent $1bn over 7 years (and let’s be honest having spent $1bn over the past 7 years is a lot more costly than if someone had just spent $1bn last year – as compounding imputed interest on $1bn over that time is quite costly) for a single plant, you’re right that downtime for say half the day, could make the economics a lot worse.
If costs during down time are low but you are only running the plant for 12 hrs a day instead of 24 hrs a day, you are in effect doubling the cost of the plant. Very valid point.
I suspect that there may be economies of scope with other electricity storage activities – like say thermal storage. I would not be surprised if molten salt (perhaps the very salt from ocean water) makes sense here.
(And if all else fails, you could burn natural gas at night to keep the plants running.)
The problem is that you really need a lot different plants to be quickly constructed by a lot of different smart and talented people, tinkering along the way for ways of having adequate renewable power at night. One or two of these groups is likely to have some serious improvements along the way.
California could be a great ecosystem for such a thing. Instead what you get is years of delays, litigation, and mass cancellations of desalination projects.
The six fish are steelhead trout. I used to live in rural New York, and in the fall you could reach into any shallow stream that emptied into a Great Lake and just pick one up (NB: this is illegal because of fair chase laws). So only this particular population of the fish is endangered.
I think the problem (both politically and ethically) with a market-based solution to the water crisis is the underlying realitiy of economic inequality. Markets give you information about relative demand by using the prices people are willing to pay to determine the distribution of resources. That’s great in principle, but in practice there are many people who don’t have enough money to adequately signal their demand for resources. If you’re talking about luxuries that’s something you can ignore with minimal consequences, but when you’re talking about something that’s an absolute necessity, that’s a major problem.
Now, I’m sure that you could devise a system of subsidies that would be better than what currently exists. But doing away with subsidies entirely is probably at least politically unfeasible for that reason, and possibly ethically unacceptable as well. (I’m unsure of how the math works out, but regardless of whether there is an actual problem, people will perceive one.)
Hey look, it’s a poster from the parallel universe where California’s water policies are designed to aid the citizenry at the expense of the richest special interest groups, instead of vice versa.
In a free market, being poor sucks. In a politicized market, being poor sucks substantially more.
That depends on the market, unless we’re considering any market where state intervention is a net positive “not politicized.” (I don’t believe there are no markets where state intervention is a net positive, although some commenters here may believe that.)
Well, first off, I’ll assume you agree with me on why all the interventions that make life worse for the poor make life worse for the poor, so I’ll just explain why the interventions that make life better for the poor make life worse for the poor.
One of the most disastrous types of market correction a person can suffer through is learning they live in (often, were born in) the wrong place and need to move. This form of disaster is especially bad for the poor, as they often lack the resources to make it over the hump to a more favorable market and are stuck in their now-terrible spot. Any price control or subsidy sufficient to
buy votesnoticeably aid present quality-of-life in a region holds the economic “potential energy” to create a disaster of that form. Under a long enough time horizon that whatever reason the market wasn’t doing things the way the politics wanted catches up with you, that disaster happens and selectively screws the poor.New York controls rents in certain spots? Prices go up everywhere else, the poor are screwed now. California doesn’t raise its water costs for a quarter-century? California runs out of water, the poor are screwed now. Egypt massively subsidizes grain? The agricultural sector collapses, the poor are screwed and also can’t have democracy now.
Which is not to claim that there are no net-positive interventions possible (in fact, it suggests one: subsidizing relocation costs), just that the political marketplace actively selects for disastrous ones.
Not to be pedantic, but how much money they are able and willing to spend on resources is their demand.
Now you can say that’s not fair and I agree with you, it’s not fair. But that unfairness is why the market works in the first place: we all bid for resources and those who put them to the most profitable use have more money in the next iteration of bidding. Removing the connection between past performance and current bidding power is going to play hell with the actual operation of the system we’re expecting to solve the problem with.
(I am not an economist, and am certain that there are several errors in my explanation)
Steelmanning it, I suppose it would be “Some people don’t have enough money to signal how much utility they would receive from it.” Although that would still be a problematic statement.
Isn’t it normal that we don’t have enough money to signal how much we value a lot of things?
Eg, how much money would you spend trying to save your child’s life? All of it, plus whatever you could borrow? That’s what people spend on getting food in famines. What if you have more than one child but you’d still spend that much on saving the life of any single one? Then you can’t possibly signal how much you value your kids’ lives, even if you’re a multi-millionaire.
And yet, most of the time, you don’t need to. Because a year’s worth of food and water per kid is mostly within people’s incomes, and $50 of antibiotics, or $150 for a vaccination, is a lot less than “everything I can possibly raise”. Market signals work fine here, even though no one is signalling how much utility they actually get from starving to death, or not dying from an infected paper-cut.
Now, not being able to signal your full demand is a problem for things like cancer treatments.
But, from the sound of Scott’s figures, buying off the alfalfa farmers is in the order of “buying food for the week” or “$50 of antibiotics”. The cost of doing so is so much lower than the utility of doing so, that who cares that theoretically people can’t signal their full utility?
I’m sure the argument has been made in other ways, but there is an issue of disconnection between, say, the development of life saving drugs and market prices of only the successful drugs.
So once you get past the drugs that are applicable to very broad populations, and low hanging fruit in drug development, the market for the individual successful drug/procedure is trying to bear the cost not just of its own development, but the failed attempts at developing drugs for other diseases.
Theunregulated market does not have very good solutions for this.
Hmm, I don’t quite follow the argument. If that drug development was exclusively government-funded there would still be the problem that society would need to bear the cost of those failed attempts, as well as the successful ones.
And the issue of developing the drugs is only a small part of actual delivery of them. Those $50 antibiotics, or $100,000 chemotherapy drugs, won’t do any good if the pharmacy is out of it because the delivery driver’s van broke down. There’s a lot of work involved in producing and distributing a drug en masse, and probably that’s work that’s not as innately intellectually satisfying as coming up with a new drug in the first place.
I’m stating this in the context where you are claiming it’s not an issue that “we don’t have enough money to signal how much we care about the life of our child” but that it’s no big deal because antibiotics only cost $50.
As a society we may be perfectly willing to spend an additional $50 a year to attempt to develop drugs for a whole host of rare diseases, knowing that the success rate of the development is low. But in the free market, only the successful drugs are saleable, meaning that the overall cost is born by only a few patients.
This means that it is hard to optomize around the desired outcome where everyone insures themselves against the rare case.
HeelBearCub: I specifically said that signalling your full demand was a problem for things like cancer treatments. To be more explicit, I agree that paying for high-cost life-saving treatments, be they for cancer or something rarer, is a difficult problem.
But I think you’re getting into a whole conversation about healthcare funding that I’m not going to do because it doesn’t strike me as relevant to California’s water supply issues. If I’ve misunderstood your comment, which is entirely possible, and you’re not wanting to talk about healthcare funding, but some more general issue with not being able to fully signal one’s utility gain, I wish you’d explain more fully.
@Tracy W:
I agree that this doesn’t seem to apply to CAs water supply issue, but you brought it it in … so I guess it seemed to to ask to be challenged.
I think if you look at places where water supply to citizenry is not managed by the government (slums in most of the world, for example) you can see fairly stark examples of not being able to fully signal. Children die of water-born diseases due to poor “WatSan” issues at staggering rates.
HeelBearCub: South Australian water is provided by a government corporation which owns all the relevant assets, and contracts out to private entities for some parts of the operation. I’ve yet to catch a water-borne illness from the taps here.
(It’s not free, though, maybe that’s what you meant?)
@James:
That is the case in almost all of the US as well, unless you are on a private well (rural) your water is provided by a municipal water company which is usually also charged with sanitation.
But I was speaking about places like the slums of Calcutta, where the water market is completely unregulated and the government essentially takes no responsibility for either end of the equation (water or sanitation).
HeelBearCub: if you look at the third world, you’ll see countries where markets struggle to operate. For example, in India, water supply is a state responsibility under the Indian Constitution. If the lack of water supply in slums in India is down to signalling problems, it’s because the poor are not managing to signal their demand for water politically.
And I have never ever heard of anyone before describing an Indian economic activity as “completely unregulated”. (I’ve done occasional projects for work on India, and worked with some people who were doing specific work on supplying toilets and drinking water in India). Do you have some source for this remarkable claim?
I am going off primarily off of what I remember from
The Big Necessity by Rose George. I need to see if I can find my copy of the book, although I think it might be in (some box) in the garage.
Since you have practical experience in this area, I’d love to hear if you have read the book. It concentrates on the general issue of “WatSan” (with an emphasis on the sanitation side) and how it affects and is affected by poverty around the world.
@HeelBearCub: I’m not an expert in sanitation and water systems. I’m not an expert in India either, there’s way too much to learn in a life time. But, you don’t need to be an expert in American politics to be skeptical about claims that, say, the US government is discriminating against Christians. The Indian politicians who led the fight for Indian independence and ruled for the first few decades, like Nehru, were mostly on the socialist spectrum.
@TracyW:
I have no idea what India’s background as socialist country has to do with the conversation at hand.
The question at hand is whether it is possible to have too little money to adequately signal your need for clean water. I am maintaining that it is most definitely possible and that there is a significant portion of the population of the globe that lives under these conditions.
I’d actually go further and say that not only can’t they signal the actual value they do place on clean water, that the actual value they place on water is not high enough to be rational. The benefits of sanitation are not obvious to the naked eye, especially when you have grown used to an unsanitary environment.
HBC, you say you speak of places “where the water market is completely unregulated and the government essentially takes no responsibility for either end of the equation (water or sanitation)”.
This is quite surprising. I expect that no government anywhere leaves the water market completely unregulated, which is especially true when they have “a background as a socialist country”. I will be surprised if it’s even legal for a private entrepreneur to up and start offering water services in India without a government mandate.
(Sanitation in India may not be the best example for cultural reasons. Indians apparently reject using toilets even when the government installs them)
I’ve never been to India, but there can be a big difference between de-jure and de-facto regulation in cases like this.
Sure. Maybe that means the government (and officially designated private partners) doesn’t follow its own rules or reach its own standards, but it doesn’t usually mean that independent private initiative will be allowed.
—
Wikipedia has a large page on water and sanitation in India. The last thing that comes to mind is “unregulated”.
@HeelBearCub, yes I should have said “in rich countries”. Although, the reason that people in rich countries are generally well able to afford things like food and drinking water and $50 antibiotics is because of markets.
No, that is not the question at hand. The question at hand is whether it is possible to have too little money to adequately signal the need for clean water that HeelBearCub thinks he would have were he a different person and using a non-economic definition of need. And the answer to that question is yes.
Or more specifically, “yes, and…?”
@Cauê: I already made the point that those who grow up in unsanitary conditions will place an irrationally low-value on sanitation. This is one way a market for water and sanitation can fail (or fail to reach a sustainable condition).
As to the statements about regulation, I was specifically referring to the slums, where ” water supply to citizenry is not managed by the government”. Slums typically exists essentially extra-legally. The Indian government does not manage water or sanitation in the slums.
Typically markets work on the concept of ownership, but ownership of water is a very tricky business. If you own the water in section of the Ganges, can you remove all of it? What if someone upstream of you does that, is your ownership meaningless? Can you dump whatever you want in your section?
We are very far afield from whether the poor in CA could adequately signal their need for water (and sanitation), all I was trying to challenge was the idea that it couldn’t possibly be true that poor people can’t signal enough need and point out that it could be true.
@Irrelevant: My prior is that a huge proportion of historical and current deaths were/are due to poor water sanitation. You don’t think that falls under the definition of economic need?
@HeelBearCub:
Another remarkable statement. Do you have any support for this one? Because, yet again, this strikes me as a claim about as unbelievable as “US Government is discriminating against Christians”.
From this article:
So, according to this source, the Indian government is managing water in the slums, by refusing applications for new connections if the applications don’t have the right paperwork. (The linked article also says that slum households have slightly better access to tap drinking water than non-slum urban households, it’s just more likely to be outside the home in slums).
@Tracy W: That point and my point are the same. The slums are vast, water gets in waste goes out, but not in any way that is sanctioned or managed by the government. The slums continue to exist, the “price” of water and sanitation is very high, and the need for clean water and sanitation is not met.
If you were a relatively well off, but eccentric person, who lived in the slums for eccentric reasons, you would pay enough to get clean water. The poor in those slums can’t.
If you think my point was that the situation exists because a completely free market was tried and then failed, I think you drew a conclusion about my line of reasoning that was not my intention.
The only question I was trying to address is can water prices be so high that the poor can’t pay for what they need.
HBC, my main point is that the government doesn’t allow the market to act in this situation, so it’s a poor example.
@Caue: But water is being delivered. It appears that it is typically hand carried the last mile from tankers. It’s just expensive (time and money), which could be thought of as an artificial modeling of natural scarcity causing higher costs.
Regardless of the source of the high-price, the price still exists, and it causes real loss of life. This isn’t a question of whether this market is working in an optimally efficient manner, but whether the poor can be priced out of clean water. I’m merely saying that this is possible. I’m not even making a claim that this would (or would not) be the case in CA.
> Although that would still be a problematic statement.
Not in the same way. If we classify humans as having terminal value, then economic value is purely instrumental and of a different category.
In addition, rich people aren’t actually utility monsters, so there actually is a decreasing marginal value to them having more money, compared to preventing starvation.
Yeah, I was attempting to indicate that I was talking about demand in a more general sense than the more strict economic one, but apparently I failed.
Your point about the connection between past performance and future performance is well taken, but that’s only true up to a point, and unfortunately we live in a society where both wealth and poverty are often hereditary. If we had, say, a guaranteed basic income, I think “let the market decide” would be both more politically attractive and ethically sound.
Also, because some commenters seem to have missed it, note that I am not advocating for the current system of subsidies; I’m specifically opposing a purely free-market alternative to it. I’d prefer a system that that subsidized the consumption of individuals up to a sane limit and largely left businesses to fend for themselves.
It could be a very simple system where the water is split into two pools: one that is distributed equally among residents, and another that is bid on in a market. The per-resident pool could scale linearly with population, and the for-sale pool would fill the gap between the per-resident pool and the total available. I mean heck, there are a ton of ways that it could be distributed, but the basic idea is that if you want to use way more water than the average resident, you should have to pay a market price for it.
Note that that’s really the same thing as UBI+market, but disguised so that people won’t recognize it as such.
Mmm. Yes, but … it may be preferable for a host of sanitary reasons to discourage certain trade-offs. I want to say this is a herd-benefit issue, but I’m not sure if that is actually the case.
But we do want everyone to wash their hands after using the toilet, and that is a shared benefit, not just accruing to the individual in question, so I think the case can be made. There is a reason the sewers and aqueducts of Rome are considered key benefits of the Roman Empire.
Given American intuitions about things like UBIs, it might be more politically feasible to do it that way, while at the same time lowering the overall political distortion of the water market.
Yes, it would make sense to have a “basic income” of water guaranteed for each person–enough to have plenty to drink/cook/bathe with–and then you have to pay for additional water beyond that.
Where I live, you pay one rate for water up to a certain point (I think about 30 cubic meters per 2-month period, with the actual number varying depending on the number of people in your household), and a higher rate for water usage above that point.
Counterpoint: Food.
Food is essential for everyone. Food prices are determined by the free market. Food is really cheap.
Poor people also get food stamps. Maybe we should get to use them to buy water with? They already work for buying bottled water, but that’s pretty inefficient for showering etc.
“They already work for buying bottled water”
Are you sure? Republicans might have added that to their absurd restrictions on using food stamps to buy “luxury foods” (like bulk dry beans).
Food production is also subsidised almost everywhere in the world. For “bread and circuses” reasons maybe, or for strategic reasons (not being dependent on other countries for food in case of war or disaster etc.) Anyway, the consequence is that a lot of food is dumped on the market that wouldn’t be profitable to produce without subsidies, and which quite possibly people won’t eat. So of course it’s cheap.
If we could somehow wish away the possibility of food riots – everyone being perfectly respectful of property rights, and content with the lot the market would give them, even if it was starvation – who knows what would happen? Maybe all of California’s fields would become dedicated to producing a half-util’s worth more of happiness for Carlos Slim.
Subsidies cost money. Subsidies cost more money than the money paid out in subsidies, as you have the cost of collecting the taxes, and the cost of transactions that don’t happen because of the taxes (dead weight losses), and the costs of checking that the subsidies are going to farmers and not just any random bloke who can fill in a piece of paper.
Consequently, if you do away with food subsidies, people will have more money to buy food (or anything else they want), more than the loss of subsidies to the farmers.
If you’re worried about distributional issues, upping payments to the poor at the same time as you get rid of the production subsidies is more cost-effective, as we already have the infrastructure in place to make payments to said poor.
Well I agree that getting rid of food subsidies and just increasing transfer payments to the poor would be more efficient, and preferable in all sorts of ways.
But if I were a politician who really wanted to preserve existing power relationships – not giving too much more power to the poor, not taking power away from incumbents such as agribusinesses, not ceding control over critical resources to rivals (foreign countries) and lower the risk of catastrophic challenges to the social order (food riots and revolutions), then food subsidies make more sense than transfers to the poor and free trade.
New Zealand did away with subsidies to farmers in the 1980s and we didn’t have food riots or revolutions over the issue.
As for preserving existing power relationships, I think they’re going to be smashed up anyway when California can’t make up its water deficit. Environmental hard limits don’t go away just because someone doesn’t like the social fallout from dealing with them.
Sanity check: do you actually believe there is anyone in California who would be priced out of household water use under market prices?
Yes. There are a non-negligible number of very poor people in California. Now, am I completely certain that those people would be worse off in a subsidy-free water market? No. But I do think that they would be better off in a market with sensible subsidies than in a totally free market or in the status quo. Better still would be a society that guaranteed a basic standard of living to its citizens, but that’s beyond the scope of water markets.
Sanity check: do you actually believe there is anyone in California who would be priced out of household water use under market prices?
Some USians are being priced out of household water (and sewage service) under the current prices. ( Iirc, Detroit and Baltimore have better nearby sources of water than California does.)
http://www.needhelppayingbills.com/html/detroit_water_bill_assistance.html
City water bills are high enough for people to notice; but prices per unit residential water are fantastically, sillily higher than the prices of the agricultural or environmental water that we’re mainly concerned with here. To some extent this makes sense (city water distribution is a lot more expensive per gallon than agricultural irrigation is, and almost infinitely more expensive than a river you don’t touch except for flood control), but you’re paying for city infrastructure, not the water itself, even if it’s itemized on a per-cubic-foot basis. And it should be relatively insensitive to prices in a state-level water market. Unless California manages to fuck it up, which I wouldn’t put past it.
Detroit, too, is a bit of a special case: being a city with a shrinking population and a water infrastructure designed for its peaks, its per-capita fixed costs would be unusually high. I don’t know exactly how it’s dealing with that, but I wouldn’t be surprised to find it reflected in the residential bottom line. (This doesn’t apply to Baltimore, though.)
The Detroit Water and Sewerage Department has a lot of issues, but not a shrinking customer base. The department serves the entire metro area, with four times as many customers outside the boundaries of Detroit than inside.
In fact, the nature of the customer base is much of the problem. The department is controlled by the City of Detroit (4 of the 7 commissioners are City of Detroit appointees, the other three have to be confirmed by the Mayor of Detroit anyway), while its monopoly customer base consists mostly of people outside the city. So there’s very little pressure to actually be responsive to customers, while there’s lots of opportunity for sinecures and contracts for the politically connected.
And if Detroit had anything remotely resembling market water pricing, it might be remotely relevant to the question of whether people would be priced out of water under market pricing.
However, it doesn’t, so it isn’t.
There are rural water markets in Australia. These markets are popular with users, so I suspect some of the ethical objections would disappear with experience.
One way of addressing common equity concerns would be to give people property rights over the water they are currently entitled to. They can continue to use this water and would be unaffected by the changes (except for externalities), or they can trade. I think it is typically better to address equity objectives directly by redistributing income or endowments rather than distorting prices.
The climate in Australia’s Murray Darling Basin is similar to California’s. An interesting result is that water markets, and increasing water scarcity, have not lead to the expected destruction of industries such as cotton and rice (which use a substantial volume of water to generate a dollar of output). This is because, unlike horticulture, these are opportunistic cropping activities that can be mothballed when water prices are high. The implication is that it is difficult to know where additional water for, say urban users, will come from under a market system. The advantage of markets is that we don’t need to know.
To the best of my knowledge there are still a lot of dumb water shortage problems going around – there’s really never enough water actually flowing through the Murray, so the mouth is backfilling with saltwater and silting up and Adelaide’s got some moderate water shortage threats as a result, groundwater is being extracted faster than it recharges, and the government is continually having to try and buy back water from farmers.
Environmental damage does not necessarily imply poor management from an efficiency perspective. Some damage (to agriculture and/or the environment) is a sad but inevitable consequence of scarcity.
It will be extremely costly to maintain the lower lakes in reasonable ecological condition, especially if climate change progresses as expected. Is it worth it? I’m honestly not sure, but it’s possible that the water would have greater benefits in other uses, such as watering environmental sites upstream.
In fact, “environmental damage” seems to have a valence of actually meaning damage that is somehow “inappropriate”. If there used to be a lake, but it’s been drained, that’s perceived as “environmental damage”, but if there used to be a large coal deposit, but it’s been mined, that’s not perceived as “environmental damage” to the same extent. Every time we mine or log or plow or hunt or build, we affect the environment, but only some activities are referred as “damage”.
@RCF:
Mmmmm, that really depends on how the coal was mined. If it is dug out by way of a mine, then there is less environmental damage (because there wasn’t any biological activity occurring inside the seam and everything else could theoretically remain relatively unscathed).
Of course in practice, various toxic by-product substances tend to pile up and cause watershed damage (as runoff is poorly controlled), or they don’t dig down for the coal, they just remove the whole mountaintop and deposit the rubble into the valley, which pretty much destroys the local Eco-system.
So, no, coal mining is definitely considered to cause ecological damage.
And broadly speaking, the environmental movement wants there to be a great deal of wild-space which is not touched by the works of humans, so I’m not sure where you are sourcing the idea that there isn’t understanding that any of the activities of humans can damage the environment.
Roger Douglas had the solution to that 40 years ago. Give the poor money and let them decide how they want to spend it.
Or, if that really sticks in your craw, give everyone an X allowance of water a day, and have the market account for the rest. You could even divvy up Scott Alexander’s 2.8 million acre-feet of water evenly amongst the population, that would leave the vast majority of the rest in play.
You mean the $2 t o$5 Alex calculated? Or more like the farmers?
“but when you’re talking about something that’s an absolute necessity, that’s a major problem.”
Drinking water is a necessity. As per my earlier posts, the price increase needed to deal with the California shortage would probably be under .01 cents/gallon. Do you think that would result in people dying of thirst?
You might consider that the main purpose of federal agriculture policy ever since the New Deal has been to hold up the price of farm products in order to buy farm votes, hence to increase the cost of food—by a lot more than the equivalent of raising the cost of water by $.0001/gallon.
If you don’t believe my figures, try a price of a penny a gallon, which also wouldn’t result in people dying of thirst. That comes to a bit over $3000/acre foot, or $3 billion/million acre feet, or $14 billion for the amount of water used to grow alfalfa. Compare that to a total alfalfa revenue of $860 million.
Treating water as a necessity on the assumption it is being drunk is a very natural mistake, but in the modern context it’s complete nonsense.
Either the US is weirder than I thought, or the reason for propping up farming with subsidies is not to buy votes, especially since there are so few farm votes left.
Farming is fairly unique among markets in that you normally want a large over-production. Crop yields will fluctuate from the normal year by +/- 50% or so depending on weather. 50% extra food is not a problem as such, but if you let the market optimise so that you get exactly the right amount of food, you will get the odd year where half your population starves. Subsidies is the simplest way we have found to keep up the buffer production.
That, and also paying farmers to use planting rotations that don’t exhaust their land, or else the short-sighted who do can in the short run outcompete the far-sighted who don’t, and by the time the short-sighted’s land gets exhausted the far-sighted have went out of business.
There are several problems with this. Firstly, crop yields of a particular crop may fluctuate from a normal year by +/- 50% or so depending on weather, but crop yields of all crops across the entire world don’t fluctuate by that much.
Australia had a drought for most of the 2000s, but didn’t go hungry as Australia could import food from places that weren’t having droughts.
Even leaving aside geographical variation in weather outcomes, if we have multiple crops then when wheat is scarce, we can eat more rice.
Secondly, a number of crops, like wheat and rice can be stored for multiple years. Thus the Biblical story of Joseph and the dream about the seven fat years and the seven thin years. In a market, specialists arise in storing foods and examining information about the probability of coming shortages.
Thirdly, food is a normal good. As incomes rise, people purchase more food, in part to eat more, in part because we need to be less careful about wastage along the way, and so we can do things like feed grains to animals to eat their meat. This adds cushioning.
Famines disappeared from English (famously not Irish) and Dutch history in the 17th and 18th centuries, long before agricultural subsides became a thing (although there were rudimentary public systems for feeding the poor, which probably contributed to the disappearance of famines, see Amartya Sen for more).
The reason famines disappeared from England was precisely because it didn’t disappear from Ireland. If you are rich and powerful you can starve the poor Irish to compensate for your own lack of buffer.
If you have over production, you don’t need to starve the Irish even when you have a bad year or 7, hence subsidies.
The reason New Zealand can get by happily without subsidies is because it has massive over production in the form of exports so that the internal fluctuations are zero even when export varies rather a lot.
Edit:
and the +/- 50% was on a continent scale, the individual farmer faces rather larger fluctuations. My worst years were 1992 and 2013 where crop yields were 11 and 12.5% of average respectively.
@Richard: an interesting claim, but I am skeptical about it. A quick search for historical grain prices in England found this article, which shows higher prices in 1839 for oat, wheat and barley than in 1845, which is when the Irish potato famine started, and then high prices again in 1854, when the Irish potato famine had ended (see table 2, page 9). The same source has grain sales (table 4, page 11) and just by eye-balling it I don’t see any connection to the Irish potato famine, while if your story was true then I’d expect sales to go up over 1845-1851 as English farmers had to buy more grain as a result of their own shortages (or possibly down if the British government was buying up grain and redistributing it to the English poor).
I don’t understand your continent point. Australia is a continent. We are agreed that Australia imported food during its drought, why would this change for any other continent?
Yes, imports/exports is how New Zealand can get by without subsidies on agriculture, but this isn’t some trick only us Kiwis can pull off. Any country can deal with internal agricultural fluctuations by selling something on the world market so as to buy imports, eg coal in the case of Australia.
@Tracy
I admit I’m on shaky ground here as it’s about 4 decades since I learned this in school, but I think the mechanism was this:
1: The Irish were only allowed to grow on parts of the land.
2: As the grain harvests got worse, the land set aside to the Irish grew smaller
3: Since potatoes gives higher yields, Irish diet ended up being just about only potatoes
4: When the potatoes caught some disease, the Irish starved.
In short, the English outsourced their famine, which explains the price trends you mentioned rather well. I’m sure Deiseach can correct me if any of the above is incorrect 🙂
As for the export option: What I’m trying to say is that you want a steady over production. You can do that with either subsidies, exports or a combination of the two. You can not scale exports so that everyone exports a lot of food because that would crash prices, so someone must subsidise in normal years.
And for the continent bit: The individual farmer has larger fluctuations than a continent which has a larger fluctuation than the world. This is not important for the basic argument.
I vaguely remember that there was a large price hike in soy beans about 5(?) years ago. I remember at the time that the analysis was that the only reason that didn’t cause a famine in china was that the US had a regular production much higher than the market would normally allow.
@Richard: While your old high school story may have been accurate (about the only thing I know about Irish land tenure during British rule is that everyone seems to regard it as terrible, including Burke), it doesn’t support your hypothesis that under a free market, markets would optimise to provide the “exact” amount of food.
Okay, evil British farmers steal Irish land to grow grain for the British market. But, we are agreed that in the 19th century they weren’t being subsidised to produce. So, let’s assume that British farmers could provide 20 million bushels of grain a year on average, and British demand for grain in the 19th century was growing at 1 million bushels a decade (entirely made up numbers). So every decade the British would steal enough Irish land to grow another million bushels. But the weather keeps on varying each year, and I think we both agree that British and Irish weather tends to be highly correlated. So, when there’s an excess of grain growing in Britain, there’s an excess in Ireland. And when there’s a shortage in Britain, there’s a shortage in Ireland.
So, under your theory, how would stealing land in Ireland make up for the occasional deficit in British production caused by bad harvests? If stealing Irish land allowed Britain to produce so much grain that it could amply cover the British market in a bad year, then, if I understand the working of your theory, in a good year the price of grain would collapse. Which would create a financial problem for the grain farmers in Britain and Ireland as their revenues would collapse. Thus, according to your theory, farmers would cut back production to avoid going bankrupt in good years, causing famines in bad years.
Now, how would using Irish land as well as British land avoid this? British farmers in Ireland would have all these Irish poor around, who presumably would be willing to eat grain as a change from potatoes. But Ireland was a notoriously poor agricultural country in the 19th century (thanks to British policies), the Irish poor had very little money to buy grain. During the potato famine, Irish grain was still being exported to Britain because the Irish poor could not afford to buy grain, and millions of Irish starved to death. Even laying aside moral concerns, there’s still a big problem with being a big imperialist nation that keeps foreign people poor to enrich your own, and that’s that you’re shrinking the size of your own producers’ markets. Empires aren’t just evil, but stupid.
I am glad we agree that the continent thing is a distraction.
As for what you are trying to say, I already knew that. What I’m doing is disagreeing with you. I don’t want an over-production of food. I don’t want everyone to be exporting food. I want some people to be exporting food some years, when they have a surplus, and different people to be exporting food other years when they have a surplus. Plus all the other methods of dealing with good and bad harvests, such as storing food, or switching to alternative crops or reducing waste. To be very clear: I agree with you that it’s silly to have everyone exporting foods. I just go further than you and think it’s also silly to have everyone over-producing because of subsidies.
“Crop yields will fluctuate from the normal year by +/- 50% or so depending on weather. 50% extra food is not a problem as such, but if you let the market optimise so that you get exactly the right amount of food, you will get the odd year where half your population starves.”
::sigh::
Optimize for what? It makes no sense to say “optimize” without there being some metric that you’re trying to optimize. There is no such thing as “the right amount of food”, there is only “the amount of food that optimize some metric”. If there’s a famine, food prices will skyrocket, and the farmers that were overproducing in a normal year will make a killing. So there’s a market incentive for farmers to overproduce. Now, you may think that the market valuation of overproduction is “too low”, but if that’s a value judgment, and it should be presented as a value judgment, rather than presented as an objective fact.
50% too much food is a problem. It means extra resources have gone into producing food. Everything is commensurable. There is some level of starvation at which it is preferable to let those people starve than spend more money preventing those deaths. The decision is then what point that is.
Either the US is weirder than I thought, or the reason for propping up farming with subsidies is not to buy votes, especially since there are so few farm votes left.
Weirder than you thought. It’s an artifact of the federal system – while there aren’t very many farmers left in the U.S., along with closely-associated industries, significant in several states, each of which has two Senators. Also, the number of people very directly dependent on farm incomes is much larger than the number of farmers; lots of people drive trucks to move crops and farm supplies, lots of people work in food-processing, and in a few states, a surprising number work temporarily in agriculture.
The fact that farmers are a fairly small group makes them more influential politically, not less–a more concentrated interest.
Gary Becker used to claim that countries divided into two groups, as defined by farm policy. Where farmers are a small minority, as in the U.S. and Europe, farm policy is designed to hold prices up to buy their political support. Where they are a large part of population, as in some third world nations, farm policy is designed to hold prices down, so as to buy the support of the urban mob at the expense of the dispersed peasantry.
Your explanation makes very little sense, for at least three reasons:
1. Output varies year to year, but by nowhere nearly that much.
2. Many crops can be stored if output varies enough to make it worth doing.
3. Farm products are bought and sold on an international market, and bad weather is a relatively local phenomenon.
The only problem I have with the “free market will fix this” idea, is the way such ideas generally seem to play out. If the “free market” solution is to raise the cost of water for alfalfa growers, cool. However, if the idea is to raise the price of water on everybody, then you’ll just be screwing over poor people who cannot easily absorb even small increases in price. We’re seeing a lot of this now in the demands on different neighborhoods to cut water usage in different ways:
http://www.nytimes.com/2015/04/27/us/drought-widens-economic-divide-for-californians.html
First you stop the regressive subsidization of water for alfalfa and almond production. Then you stop the regressive subsidization of water for L.A. Only if that still hasn’t solved the problem do you start raising prices across the board. I don’t know why this is difficult to understand.
You need to work on that math. California grows about 7 million tons of alfa a year. That means growing a ton of alfa takes a little less than an acre foot of water. A ton of alfa sells for about 300 bucks. an acre foot of water is around 300,000 gallons, which is more than most people will use in a decade. the idea of people being too poor to afford tap water is patently absurd.
Well, since clearly people are already not able to afford current water bills, let’s dispense with the absurd and go to the math.
By my math, using the 2011 average daily water usage of a California household of 360 gallons, the average single-family uses about 131,400 gallons per year. Using your alfa ratio, that would mean their base yearly bill should $131.40, or about $11 per month.
Obviously, that’s not the price residents pay, or else that housekeeper wouldn’t have a $250 water bill over a two-month period with significant conservation efforts in effect. I suppose this means that the alfa growers get their water basically for free, but I wonder what the price point would need to be for them to grow significantly less.
Do you have a source on that housekeeper? Because in San Francisco, $250 buys you about 9,000 gallons of water when you include sewage costs, and it seems unlikely that someone attempting to conserve water would spend that much. I mean, that’s equivalent to running the shower 35 minutes every day with a normal-flow head.
I have paid water bills in California, and they were much, much cheaper than $125 per month. Of course, I acknowledge that the rates I was charged were much higher than are charged to farmers. That’s not too surprising though, most of the costs of operating a municipal water system are fixed costs. What you are really paying for is the ability for clean, unlimited amounts of water to come out of your tap at the turn of a knob 24/7/365. Since the water company doesn’t really pay for that water, their cost structure is the cost of maintaining the pipes and a small marginal cost for pumping and such.
When I was responsible for a water bill, I was paying somewhere around $200-300 per year, which isn’t an order of magnitude more than your calculations suggested I should be paying.
Currently in California, different people are charged different amounts for water. In general, urban users pay much higher prices than farmers. Alfalfa can only be cost effectively grown because the price of water is kept artificially low. If you charged everyone market rates, it’s the farmers of water invasive crops that would get priced out of the market, not consumers.
Depending on the water district, most of the price of the water bill is to have the service. If you rented a house, and didn’t use *any* water, you’d still pay a noticeable bill. As other commenters have mentioned, that money pays for maintaining the system that delivers water from snowmelt hundreds of miles away. If I’m remembering correctly, in Oakland, that’s on the order of $50/month, though I think that also includes sewer service.
There’s a perfect example in your linked article of how environmentalist sentiments are acting as a shield for much more mercenary interests, by the way:
Two possible keyhole solutions:
Set up tiered rates for residential customers: charge a lower, subsidized rate for the first XX gallons per household per month, to make sure a reasonable allocation for drinking, cooking, toilet use, and washing is cheaply available, and then charge market rates for “excess” water use. Charge agricultural customers the same market rates, adjusted for any differences in the cost of getting the water to farms compared to urban/suburban residences.
Charge market rates to all water users, and increase payouts for state-administered antipoverty transfer programs (TANF, etc) by enough to compensate for the cost increase on the first XX gallons/month.
Already the case.
It is the case in my city, but my understanding is that it varies by jurisdiction.
Yes, we must forever lock ourselves in to massively inefficient policies because changing them might possibly have some miniscule negative impact on a poor person, somewhere.
Because making sure the burden doesn’t continue to fall on the poor, as the linked story clearly indicates it currently does, might get dirt on the pretty principle of it all.
Did you even see the chart Scott made above? Or read anything he wrote about how ag use is several times that of residential use? If the price of water goes up, the burden overwhelmingly falls on the owners of farmland. (Who are, in case you didn’t know, disproportionately wealthy.)
I suspect if I proposed raising taxes by one-hundredth of a percentage point on the incomes of the poor, and twenty percentage points on the incomes of the rich, you would also complain about how awful this was.
Yes, I did read the post and all of the comments. Did you even read the article I linked that showed very clearly how even small increases in price are not borne well by poor communities? If it works out to be a few pennies a month for them, fine, but if it’s more like $5+ per month, that’s a big deal.
The pricing for residential use is in a completely different universe from the pricing for agricultural use. Some of that difference is accounted for by the infrastructure requirements of city water delivery, but I suspect most is not. Thus, the burden for this shouldn’t even begin to fall on residential users in any amount until California gets much closer to price-parity between the two.
Members of the population you’re describing regularly face income or expense fluctuations far in excess of $5/month. The idea that there is this huge population of people at the cusp of utter disaster, staved off only by their low water bills, is absurd. (It’s also not actually supported anywhere in your article.)
Have you ever lived or worked in poor communities? The way they generally absorb those fluctuating costs is by either not paying certain bills or cutting back on necessities like food an medicine. Perhaps you have a different definition of “disaster” that does not include the inability to fulfill one’s basic needs.
And how did you get the idea that a $250 water bill for two months was low for a poor family? They don’t have low water bills; they have excessively high ones that they are already struggling to pay and shouldn’t be pushed higher. You can stop with the straw men any time now…
Zephae, you should be more skeptical of what you read in the papers.
Zephae, everyone in California has too low of water bills, because they didn’t raise prices for decades. Those costs are now suddenly playing catchup, and to a dramatically higher price point than they would have had the cost been allowed to float up naturally. If you feel the state is therefore morally culpable for false advertizing and should pay the cost to resettle people to a lower-COL region or whatever, I’m listening, but “the community won’t bear it well” is not an argument about what the price should do.
I’ll put this mathematically. Generally, the cost (in utils) of a “small” increase or decrease in income (or expenses) should be linear in that cost. Diminishing marginal returns to income means that for “large” changes a linear approximation is inaccurate. The question is, does a $5 increase or decrease in income for this population represent “small” or “large”? For the decrease to be “large”, it would need to have a vastly disproportionate effect – you would need the affected people to be presently situated at a very finely positioned inflection point of the curve.
I claim that the proportion of people situated at this point (if it exists) is small, because:
1. It is a priori unlikely, because even among the poor there is wide variation in incomes, so most people should be far away from this point (if it exists).
2. Even within an individual poor household, the monthly variation in income or expenses is frequently larger than $5. This means that at any given month an individual household is unlikely to be within $5 of that point.
Therefore, the cost of losing $5 of income is about the same as the value of gaining $5 of income, to a member of this population.
Therefore, if you think enacting this policy would be bad distributionally, you must be in favor of the inverse policy: one that gives $5/month to every poor household, and also several times that total amount to all of the rich farm owners in California.
The meaning intended was “lower than otherwise”, not “objectively low”.
@Alex Godofsky
As a counterargument, poor people can offset small random variations of income or expenses over time or over different poor people though solidarity within their community and/or averaging over time: e.g. if you can’t pay the bills this month because of an unexpected expense, your uncle may help you, and then you will help him when he needs it.
A price rise on a commodity, however, would make everybody permanently poorer by some amount in a way that can’t be offset by time-averaging or solidarity.
$5/month, every month for everyone in your community may not look like a big deal, but it may be for certain classes of people.
Anyway, I agree that there aren’t probably many people in this condition in a first-world country, therefore the state could just give them $5/month in cash, food stamps or whatever and be done with it.
“the burden overwhelmingly falls on the owners of farmland. ”
Some of it falls on the owners of farmland. Some, I suspect a larger proportion, falls on consumers of agricultural products.
David: see my reply to you in the other thread. To elaborate more, if most of the affected arable land has low value for non-agricultural uses (which is likely since most of it is necessarily rural) then almost all of the incidence is going to fall on the landowners. Ever field laid fallow represents a loss of ~100% of the value of their property. And if landowners are able to spend money on e.g. capital improvements that increase water efficiency, or switch to crops that use less water, they will be absorbing even more of the cost.
In real life, other folks have raised this point to me. My guess is similar to Scott’s, in that we should move toward a free market in water, or at least smooth out some of the bigger inefficiencies. But if raising the price of water means raising the price of food, that might unduly impact the poor. We should be sensitive to legitimate distributional concerns.
And that’s why I like this alfalfa talking point!
Food is traded in a global market. Just how much of an effect do you think this is going to have on total global food production?
If California is using more water than it can sustainabily, eventually it is going to run out. And that running out will impact the poor too. I’m having trouble coming up with a scenario by which the poor are less affected than anyone else by the water running out in California.
It’s like Canada and its codfishery, politicians didn’t want to do anything very effective to stop overfishing because of the social impact. When the cod fishery collapsed, the resulting social fall-out was far larger.
Any Pigovian Tax can be made non-regressive by remitting poor people money from the revenue of that tax.
Raising the price of water does not need to raise any revenue to be effective. Making the marginal price of water reflect it’s value propagates accurate information, allowing people to make more intelligent and informed decisions about their usage.
If you are concerned that this will impact the poor too much, then just *give them some cash* (or water credits or tax credits or whatever) to offset.
The really cool thing about water is that every person has a pretty similar baseline (cooking, drinking, cleaning) and then rich people use way the fuck more water than everyone else. It would be relatively simple, all things considered, to establish a reasonable ‘basic water income’ so to speak.
But can you pass an adequate remittance in the real world?
The state pays X% of the first Y gallons per month used by a certain residence. For exemple.
That kind of attitude (“The right solution will never work, let’s do this instead”) is how we got into this mess in the first place
Doing this with every Pigovian tax is a bureaucratic mess. Why not just one Basic Income like thingy?
If one wishes to improve the lies of poor people, subsidizing water is a really poor way of doing so.
Plus importing millions of them from a middle income (in world terms) country to a drought prone state where they apparently are at risk of starvation if they pay the cost of water might not be the best idea either.
On the other hand, they vote (D) reliably once they move there so we’ll just have to ignore the problem and hope it goes away. Maybe we can yell at people about taking too long in the shower instead.
“where they apparently are at risk of starvation if they pay the cost of water might not be the best idea either.”
I would think Scott had provided enough data so that everyone commenting on this could see what scale of nonsense that kind of talk is. Raising the cost of water by something like a hundredth of a cent a gallon is unlikely to result in anyone starving.
My point was that no one actually believes the “increased water prices will lead to starvation of the poor” story.
If they did then they’d advocate for preventing mass helot class resettlement in California.
It certainly wasn’t a sincere argument against importing tens of millions of helots because the poor dears would be at risk of starvation due to high water prices.
But, obviously enough, if the government simply stops subsidizing alfalfa and gives this money to the poor already everybody is far better off and every residential user can have as much money as they need. This is fairly clear from Alex’s math.
Just reduce their other taxes. You’re not taxing them because you need the money.
” However, if the idea is to raise the price of water on everybody, then you’ll just be screwing over poor people who cannot easily absorb even small increases in price.”
To begin with, we are talking about a level of price rise that would be almost invisible to individual users, even poor ones—something like $.0001/gallon (see more details in my earlier posts).
Beyond that, raising the price to agricultural producers means that crops are more expensive, which means that food is more expensive, which again affects poor people as well as rich.
So…I’ve got some beef with the apparent fact that the top two agricultural water hogs are basically just producing meat.
It’s definitely not the biggest or most direct culprit here, but it is an issue, and in more places than just California. What I’m saying is, if we’re going to offer incentives for things like low-flush toilets, why not also for vegetarianism?
While true in a global sense, you have to remember that California is growing the crops for the entire meat-eating world, but is only consuming a state’s worth of meat. In a marginal sense, making half the state go vegetarian is significantly less helpful that cuting half of the state’s “meat” production.
We don’t need to directly incentivize vegetarianism. Just stop subsidizing water used by alfalfa growers. The price of alfalfa will rise, the price of meat will rise, and peope will eat less meat.
Apparently alfalfa is primarily for dairy cows, so you’d have to go full vegan.
Also from marginal revolution: Coase theorem (I.e. No one’s paying alfalfa farmers to do nothing because market frictions)
http://grist.org/food/california-has-a-real-water-market-but-its-not-exactly-liquid/?utm_campaign=daily_feed&utm_medium=email&utm_source=newsletter
Good find.
I think there’s a business opportunity here. Carbon offsets have already proven there’s a market for selling indulgences to the environmentally-concerned, so there would probably be people willing to buy water offsets from a middle-man who would find farmers with senior water rights who are willing to take a check to leave parts of their field fallow for water conservation.
One issue with this analysis is that it’s treating ‘water in California’ as an undifferentiated mass.
Water isn’t electricity, and even electricity has transmission issues that are obscure to anyone who isn’t trying to load-balance.
From the consumer perspective, it looks like it’s trivial to turn off a faucet here and turn on one there, and so divert the water. YmcY’s link covers many of the practical problems — it’s not helpful to pay the alfalfa farmers to do nothing if that water can’t be made to reach Los Angeles. And, right now, it can’t.
The infrastructure for transfers on that scale doesn’t exist across any sizable geographic range. The accounting for it doesn’t exist either, often enough. Substantial agricultural uses are often done on an honor system based on guesstimates. cite
Look at YmcY’s link for an example of how the existing infrastructure handles transit of even a small purchase: each transaction involves a multi-month sale process with multiple regulatory veto points… and what does a sale actually look like? “We had $13 million worth of [water] isolated above the Delta, and no ability to get it. Well, two-thirds of that water went out into the ocean. We did end up getting a third, seven years later.”
Could a more efficient, liquid, market like Australia’s be put in place? Sure… but that will take years.
Building the physical infrastructure is still more difficult than reforming the regulatory infrastructure. Different incarnations of the peripheral canal plan (or tunnels, or etc.) necessary to move water south of the Delta on that scale have been floating around since 1919; the latest stall started in the 1970s and shows no sign of stopping.
This isn’t to say that the idea of looking at the uses and analysing it as our host does above is fruitless… just that ‘California’ isn’t a useful unit of analysis. The cited Water Use in California gives some data on the Sacramento River Hydrologic Region; the underlying data used by that blogger (along with much else!) is available here. Which is more numbers and more staring, yes, and it’s not as if there isn’t room to dispute the facts asserted, but still… a good starting place.
For example: the Grist article runs with a hypothetical of someone in the San Joaquin valley trying to sell to LA: look at the San Joaquin River Hydrologic Region report, poke around until you find figure SJR-15, and you’ll get a nice summary of what percentage of water sold south of the Delta was actually delivered, broken out by year and use.
Even if the transport capacity existed today, there are other issues in play: e.g., using reservoirs for transport means not having capacity to absorb unexpected flooding.
In short, the current shortfall does not seem — as a matter of engineering — to be susceptible to the solution of buying out all alfalfa farming. (Let alone politically susceptible!)
(In place of going straight to the CWP, you might prefer some focused bloggers: consider this for LA, or this.)
That said, it’s not that buying out ag water is a bad approach… but it is one that would require nontrivial execution.
FTR I live in Australia, and there’s a few points worth making:
* Physical infrastructure has mostly already been in place for years to handle physical transfers across on two of our most important hydrological basins, the Hawkesbury-Nepean (Sydney) and the Murray-Darling (Adelaide, Canberra, 50%+ of national irrigated agriculture). Not sure about Melbourne, Brisbane, or Perth, but I don’t think there’s any major issues like water getting stuck either side of a delta.
* There are tiered products, which sounds like a big part of what addresses some of the problems in California. To simplify it down to the core idea, rights holders own a license to withhold or extract, effectively, X% of a river’s total flow. I can buy a bunch of actual physical megalitres off you, that comes out of your allocation for this year. I can also buy your allocation outright (say if you’re an alfalfa farmer looking to abandon irrigation altogether). But this doesn’t guarantee me a fixed amount of water; if rainfall in a year is only 10% of average I’m only going to get about 10% of my nominal litres. This part is critical given how volatile rainfall is, I think maybe even more so than California.
* Water rights are still politically contentious here, despite the more functional economics. The Federal government has been buying entitlements from farmers in the Murray-Darling in recent years to return to environmental flows. But even though it was all voluntary market transactions, there was still very vocal opposition – essentially the second order effects others have spoken about, where communities heavily depended on irrigated agriculture worried their local economies would disappear altogether passed a certain point.
* The major cities all now have desalination plants as drought insurance. They’re mostly on different hydrology basins to significant agriculture (the Murray-Darling being the exception); and other more cost effective of getting guaranteeing enough drinking water, like recycling or building more dams, haven’t been used as much as they might for essentially political reasons. And they all still use tiered water restrictions, not dynamic unit pricing, to reduce household consumption when needed (I believe industry has been incentivized largely with much more efficient mechanisms).
I could be wrong, but isn’t it the case that California is exporting meat, and possibly even raw alfalfa, to other states (and possibly countries) ? If California ends up growing less alfalfa, how big of a dent will this put in the budget ? Or, in other words, can California even afford to grow less alfalfa at this point ?
The point is that alfalfa grosses California (or, more precisely, alfalfa farmers) $860 million/year. So not growing alfalfa would cost, at most, $860 million/year.
Bugmaster, I think you’re a bit confused.
How much California exports is relevant to the *balance of trade* (or trade surplus/deficit) not to the state’s budget (budget surplus/deficit).
California’s state government is required to balance its annual budget. That is, it has to take in as much money in taxes as it spends.
There is no requirement to ensure that the state doesn’t run a trade deficit with other states or countries. (I’m not even sure that we keep figures on interstate trade.) In general, economists aren’t overly worried about trade deficits, and they aren’t under the direct control of the government regardless. (Individuals choose where they want to sell their products [within the existing trade rules negotiated between federal governments], not the government.)
I was under the impression that California collects taxes on all that trade, which means that tax revenue is proportional to trade income. But perhaps I was wrong.
To the contrary, US states are generally prohibited from levying tariffs. Article 1 Section 10, the Import-Export clause.
A tariff is just one type of tax. States are allowed to levy an income tax, and if farmers are selling alfalfa, that will generate income.
Yes, but for some reason the OP’s context was specifically interstate/international trade.
@Alex Godofsky:
Yes, I was under the impression that California exports most of its alfalfa and/or meat, so I ignored in-state sales.
What is the difference between in-state and out-of-state sales to your point bugmaster?
Mostly great (the Nestlé thing in particular is crazy), but was the calculations for the downside of having less/no alfalfa left out? This seems like a big omission… Yes, the sellers get less income so paying them will fix that, but the buyers will get less alfalfa too, and the solution to that will also require water; the next best option may even be less efficient water-wise.
It may indeed be the case that diverting half the continent’s water so California can grow more food is economically rational. But if that is true, it will still be rational when you remove the water subsidies.
California has some of the world’s most productive farmland. Of the top ten U.S. counties for agricultural production, nine are in California.
If alfalfa growing were banned in California (without changing any other policies), the alfalfa land would be used for growing something else. No water would be saved.
But any other crop (except rice) uses less water per acre. Even if they switched to almonds, that would save 20% of their water use (4 feet vs 5 feet).
Ah! Missed that detail. My apologies.
I was assuming we’d be paying the alfalfaists to let their land lie fallow.
I prefer the term “alfalfador”.
http://en.wikipedia.org/wiki/Owens_Valley
Consider what happened when Los Angeles bought out the water rights to the Owens River.
“As a result of these acquisitions, the lake subsequently dried up completely, leaving the present alkali flat which plagues the southern valley with alkali dust storms.”
What does letting land lie fallow mean in California? A whole ecosystem dies. That’s part of the concern for curtailing ag water in the Imperial Valley. If the Salton Sea dries up there will be poisonous dust storms that may have far reaching effects. It is the tail water from ag irrigation that currently maintains that water body.
Once we have an extra 5.3 million acre-feet of water, couldn’t we use some of that to help the Salton Sea and other places?
Much of the alfalfa is grown in the Imperial valley and the tail water supplies the Salton Sea.
It’s a uniquely appropriate place to grow alfalfa. They get 30% to 40% more production than other places due to the long and ideal growing conditions. Just add water. In farmer talk they get 10 cuttings a year whereas 6 is the norm. In places with real winter they may get 3 cuttings, though they have lots of water.
Alfalfa is a perennial legume, meaning that it needs less fertilizer, especially nitrogen, and is not plowed or tilled for each crop like annual cereals or beans. It’s good for the land.
The water comes from the Colorado river, which at one time flowed through the area. The Colorado jumps its bed periodically and follows a different route to the sea, similar to the situation with the Mississippi and the Atchafalaya. A diversion in 1905 blew out and ran for a year, refilling the Salton Sea which had gone dry when the Colorado jumped to its current bed.
To benefit the coastal cities and the central valley it is water from other rivers that are in play, especially the San Joaquin.
I know that you’re making an offhand suggestion but it’s really far from that simple.
Let’s take some particular plan as a starting point: alfalfists get paid the revenue from the prior year’s crop and are required to not grow alfalfa.
So you set up the program but there is very little ability to ensure that the amount that any particular farmer reports as his prior year income from alfalfa sales is accurate. Ok, not an insurmountable problem – state pays out a few tens of millions for fraudulent claims – worse things go on every day – especially in areas where government is giving away money. They want to give away money to buy loyalty so if some get taken dishonestly, no big deal. Program gets up and running cutting checks to farmers based on some self-reported number.
Of course, there’s the inevitable adjustment – all those non-alfalfa growing farmers now step into the alfalfa market. You just cut the supply massively so prices skyrocketed, right? Were you planning on paying every land owner in California who could grow alfalfa to not do so? If not, well, congrats on writing checks to a bunch of farmers for not growing alfalfa. Too bad the alfalfa harvest didn’t decline in a significant way. Did you make it illegal to plant alfalfa to prevent this? Why not just start with that step?
Ok, you banned alfalfa state-wide. Rice cultivation steps up to take it’s place. Looks like some rice growing permits will have to be printed up.
Luckily there’s a technology that’s been discovered and refined by humans over thousands of years for solving these sorts of thorny allocation problems – of course this technology is thoroughly hated by progressives.
“Of the top ten U.S. counties for agricultural production”
Meaningless predicate alert!
“There is currently a water shortfall of about 6 million acre-feet per year, which is being sustained by exploiting non-renewable groundwater and other sources. This is the equivalent of our slightly-richer-than-average family having to borrow $6,000 from the bank each year to get by.”
In order to interpret the significance of this correctly need information like: how much non-renewable groundwater is there; if we deplete the groundwater beyond a certain point, will there be an environmental disaster of some sort; etc, etc. In terms of our household on 80k$/yr piling up a debt of 6000$, this is like not knowing the interest rate being charged, or whether the lender’s goons are coming to break our shins.
If the groundwater is non-renewable, it’s not like borrowing money, it’s like using capital to pay your debts or sellling off the family silver. You had a resource, you used it up, it’s not there anymore, it’s gone.
Technically that groundwater *is* renewable: a small amount of the surface rainfall each year reaches the underground aquifers, which is how those aquifers got there in the first place. However, the rate of recharge of those aquifers is orders of magnitude smaller than the rate of extraction at present. My back-of-envelope calculation is that if we stopped taking any water from aquifers, that they would refill in ~decades, which is a long time politically, but not very long geologically.
To all of the people who are just terrified that if the economists have their way, all of California’s poor will die of dehydration:
Do you know what charging higher prices for water does, in addition to convincing farmers to use it more efficiently? It raises revenue. Revenue that you can then spend on your social programs or your welfare or whatever. Revenue in a state that is famously bad at raising taxes and experiences periodic budget crises that threaten those social programs.
So take the money that’s just sitting on the table already!
How can California be bad at raising taxes when they have some of the highest taxes in the country?
I ask as a person paying an excessive amount of California taxes
See wiki on Proposition 13. Among other things, most tax increases require a two-third majority in the state legislature, and there are also a bunch of poorly-constructed restrictions on property taxes.
So they’re both really bad at it, and really enthusiastic about it?
I pay about 18% of my income as taxes to California. This does not strike me as “cannot raise taxes”.
That’s a fair description of California public policy in general.
That’s a fair description of democratic public policy in general.
California, especially urban California, is both unusually bad and unusually enthusiastic. A lot of the things Moldbug likes to harp on suddenly made sense to me when I learned that he’s from San Francisco, a city where they approach being correct.
That.
I think half the reason why the techies end up being so libertarian is that on the one hand you have an entire private sector that… well, Netflix replaced Blockbuster with 2000 people while rewriting the rules on how we get content, and a public sector whose recent efforts are:
* Spending $6.4 Billion and 14 years on one bridge that was literally about to fall into the Bay.
* Spending 20 years designing and building a 2-mile long subway that they won’t extend to the rest of the city because “Too many people would ride it” (To wit, it was never designed for 4-car trains, so they’d need to rebuild every station from scratch).
Very bad and very enthusiastic would describe much of California very well.
To be fair, the new Bay Bridge is very pretty. Not sure if it’s $6.4 billion of pretty, but it’s very pretty.
And to be extra super fair, a lot of California techies also work for aerospace firms, a sector (SpaceX and its ilk aside) not known for its initiative or its ability to deliver on time and under budget.
Because direct democracy: they keep voting to implement new programs but not to have any more taxes.
So basically they’re good at raising taxes, but not nearly as good as they are at spending the money.
Except that social programs generally kick in to help people once they are in a crisis situation and the goal is to prevent those situations from arising in the first place. You would need a program that uses that revenue to pay each resident, the way that Alaska does with its oil money, and just straight-up giving money to people up-front, preferably on a monthly basis like Social Security, is a nearly impossible policy to sell politically unless, like Alaska, you get big surpluses, which, as you noted, California doesn’t.
OK, I see. So your position is that no further social programs are necessary or valuable, so any extra money the government comes into should be spent on tax reductions that mostly flow to big businesses, since that’ll help prevent people from falling into crisis situations!
Imagine, just for one moment, that we lived in the world where water is priced at market rates. Would you really propose a policy change that would give a tiny bit of money to poor people and give a huge amount of money to rich business owners? Would you defend it on the grounds that it would help the poor?
It really looks to me that you are massively misreading the comment. Zephae is making a descriptive, not normative, statement.
That’s a lot of extra reading into a comment that basically just said the reality is the money wouldn’t go to the poor.
Well, then, instead of giving people money, what if we give them water vouchers, which can be exchanged for a certain amount of water, or traded for goods or services? That would be totally different from giving them money.
Haha, yeah. Little tricks like that make it more sellable, but if my experience with farmer’s market vouchers is any guide, they generally won’t get used at anywhere near the rate they should.
What would happen to the poor if the water does run out in California? I’m having trouble coming up with a scenario where they do *better* than the middle-class rich, or farmers.
The point about the water consumption of alfalfa is a good one for illustrative purposes, but (as you likely realize) paying farmers to stop growing it would be terrible in practice.
Unintended consequences: (1) farmers race to turn both marginal and productive land into alfalfa plots just to receive subsidies, perhaps destroying wildlife habitats and causing perfectly good cropland to lie dormant (2) alfalfa prices increase dramatically, causing farmers in other parts of the west to grow more alfalfa, meaning California’s water problems simply spread to other states.
If you use your imagination, I’m sure you could come up with some other much more fun (and potentially more damaging) scenarios 🙂
Alfalfa is not a highly profitable crop, it is mainly used as cheap feed for livestock and generally grown for lack of good alternatives. (By contrast, those almond farms are worth $2.5 billion.) A spike in alfalfa prices isn’t going to cause people who could be growing something else to suddenly switch to alfalfa. (Which also isn’t guaranteed to be using more water per acre than the other crop.)
Likewise, I don’t think there’s a tremendous amount of land that could be used for crops but just isn’t, and most other crops being grown pay more than the alfalfa subsidies would.
Rabbit. It is used to feed rabbits. Who even eats rabbits? One of the least popular kinds of meat.
This is a very interesting and highly relevant book: http://endofabundance.com/
The author’s blog: http://www.aguanomics.com/
One interesting thing that he pointed out about residential usage is that the cost and the revenue structures are highly asymmetric. The majority of the cost of providing urban water is infrastructure; massive upfront cost, very little per-unit cost. The majority of the price of providing urban water is fee-for-service: one unit of water costs $x.
This causes two interesting things. First, it implicitly subsidizes sprawl, because a given water district is usually legally mandated to charge the same price to every customer (because it would be unfair otherwise). As a result, dense mature urban areas, who have long since paid off the fixed cost of their water infra, are paying the same per-unit rate as people in the ‘burbs for whom new (and more expensive, because lower density) infra needs to be laid. This is an implicit subsidy from the water consumption of dense/mature areas, to the fixed costs for sparse/suburban areas.
Secondly: Most of the cost is fixed. Most of the revenue is variable. The amount of money that a water district needs to bring in every month is almost entirely debt service; if they don’t service their debts bad things happen. The amount of money that a water district brings in every month is variable; as a price per unit, it depends on how much water people use.
Repeat: Water utilities costs are fixed, but their revenue increases in proportion to how much water people use
Conservation efforts get people to use less water. This lowers water utilities’ revenue. Sometimes to the point that they can’t afford to service their debts. That is to say, the pricing structure and set of subsidies is so convoluted that conservation efforts are bankrupting water utilities.
The author of those two links is David Zetland, the same person that Xenophon mentioned, with a third link.
Just give the utilities a bail-out. Problem solved!
Similar complaints are coming from electric utilities.
Perhaps, in a context where conservation and usage reduction is important, they should be non-profits or *gasp* government-run.
…. They already are government run and non-profit. If you read his book you’ll see that his proposed solution is explicitly non-profit.
What stops the water utilities from setting prices that mostly depend on capacity rather than actual water usage?
It’s illegal in most jurisdictions. It’s seen as both unfair and unreasonable for a water utility to say “You owe $50/mo for the privilege of connecting, and your incremental water price is $0.001/gal”.
People get really, *really* angry when you demand upfront payment for things they don’t think they’re using. People don’t understand that the existence of the infrastructure is a significant benefit that they use by virtue of being connected at all.
I can’t see why it should be seen as unfair and unreasonable. In fact, where I live (a Western European country), water is actually priced this way, and the fees for a typical residential user are within one order of magnitude of the numbers in your example.
But landline Internet is typical priced the same way: You pay for the connection capacity, not for the actual data traffic. People don’t seem to have any problem understanding that most of the cost is infrastructure.
A relevant point to this discussion: Most renters in California (SF Bay anyway) do not pay the water; their landlords do. Raising the price of water *will not materially affect poor renters*.
A followup to this: When the mandatory rationing kicked in, I asked my landlord if my water usage was acceptable or if I had to cut it down.
He tells me he has *no way* of knowing what our water usage is. He has one meter for this building (3 units).
How the hell are people supposed to ration water appropriately when they have *no way* of knowing how much they use
It’s true that most renters don’t pay the water bill, but poor renters often do. It’s usually an added charge to their rent that they pay to their landlord, who then pays the bill. A lot of times, it’s a situation like the one you’re in, but the bill just gets split evenly between the 3 units, so price still does affect them. However, you’re right that renters, poor or not, are still very unlikely to know much of anything about their use because they themselves never see a bill for their particular unit.
I’m now trying to think of a polite way of expressing how great an ignorance of economic principles your comment exemplifies. Regardless of who actually sends the check to the water company, the renter is the one paying for the water, just as they’re the ones paying the mortgage interest and the property taxes.
Sure, but the point was that the level of indirection in the price increase prevents it from delivering a useful signal to the renter, so it’s unlikely to change his behavior. At best it might encourage him to move to an apartment complex with individual meters, but that’s a big hammer to use on a small nail.
“Raising the price of water *will not materially affect poor renters*.”
If the cost of water goes up enough, it’ll show up in their rent, or as new fees tacked onto the lease. It might not reflect the renter’s actual usage, but it’ll show up. In your example, the water bill for the building would probably be divided by 3.
I don’t have to know the absolute amount of water I am using in order to use less than I currently am.
Renters pay the water bill. They may not have it itemized, but if you raise the price of water, the price for their rent will increase. It’s sort of like how how much the employer and employee pay for social security is based on the relative elasticity of supply and demand for labor, and has nothing to do with which one the government asks for money from.
Illustrative article:
http://www.dailymail.co.uk/news/article-3075906/What-drought-Aerial-photos-expose-Kim-Kardashian-Jennifer-Lopez-Barbra-Streisand-greedy-celebrities-wasting-water-lush-lawns-green-midst-California-s-worst-drought-history.html (wow, that’s a long URL)
It has such gems as “Experts estimate there will be less than a year’s worth of drinking water in California reservoir’s at the end of 2015”. Never mind that drinking water is a tiny percentage of water use.
“But for the residents of Los Angeles’ wealthy enclaves, a $100 fine for wasting water is chump change and a fee they are apparently glad to pay in order to maintain their almost fluorescent green lawns.”
How dare rich people use their money to purchase more goods and services than the rest of us! And note that’s 50 times the $2 figure that Scott gives.
I wish more people understood fines as not prohibiting something, but simply raising the cost.
“Take what you want, and pay for it.”
I am really jealous. I have been working in California water for years, and though this is nothing new, I have never expressed it this clearly. Hats off.
The basic problem with buying off the growers, which is clearly the right answer, is that a whole economy depends on the growers. How do you compensate the agricultural workers, farm equipment suppliers, etc, who now don’t have anything to do for a living. Thats the tough political and equity problem. Also, the pricing will be a bit more complex because water is physical, conservation of mass and all that. So a lot of the water used for Alfalfa flows out to groundwater and surface flows. You probably don’t get a new full 6 million AF of water that you bought out.
Yes, the dependence of a whole range of associated services and industries on agriculture means that it’s not just “Farmer Jones stops growing alfalfa and lets his fields fallow, and we pay Farmer Jones not to grow alfalfa so he doesn’t lose his income”.
This is why countries spend vast amounts on subsidising agriculture. For example, the EU is a net importer of North and South American soyabean crops, and why the pressure is on to permit more and more GMO crops to be imported (since the American crops are GMO crops, and at the moment animal feedstuffs but not shipments intended for human consumption can be GMO or GMO not yet certified in Europe).
Stopping alfalfa cultivation may cost more in loss of export revenue, and combined with having to pay subsidies to farmers who would otherwise go bankrupt, it may be much more expensive than continuing to (wastefully?) irrigate the fields.
About the same way you compensate anyone else who is out of money because they lost their job?
What do you think will happen when California runs out of its non-renewable water resources and has to cut back water consumption?
This played out in Canada with the cod fisheries. Environmental hard limits don’t go away because the social problems of dealing with them are huge.
The thing with doing things to the environment is that it is always a bit more complicated than you think. For instance, growing alfalfa is a rather good way to combat erosion. I guess other plants could do that as well, but just to stop growing anything might be ill advised.
Otherwise, this reminds me vaguely of something called “Reagan Farms” which actually worked rather well even with all the negative publicity I think?
IIRC, alfalfa is a pretty important crop for putting nitrogen back into the soil. While you can find other ways to do this (fertilizer) I think it may be worth mentioning that its economic value is more than just its sale-price.
There might be some drought-hardy variety of clover that could replace it; that would also fix nitrogen, while requiring less water.
Farming use aside, there is (in principle) a positive point to having government interfere with water prices, then encourage people to take measures to cut use: If you add government pressure to cut use to your motivations list, people who could cut costs in any of a dozen ways cut them by reducing water use, while otherwise they might go for something that requires less motivation, which helps build up an infrastructure that makes it easier to cut water use (low-flow toilets becoming standard, etc.) Not sure this is the best way to handle the issue even if personal use was the biggest problem rather than agricultural use though.
Other question: what about desalination as an alternative solution? Given that most of California’s population lives close to the cost and gets water pumped in from far away, I’m not sure there’s even much of a price differential between desalinating saltwater and pumping in freshwater from a greater distance (which may easily shrink to zero or negative values if you include the shrinking costs of desalination and the fact that IT DOESN’T USE UP THE WATER SUPPLY).
Desalination of seawater is very expensive.
A fission-powered steam distilling plant could produce an absurd amount of water.
Um, using seawater does use up the water supply; the ocean is not a magic place where you can take out millions of gallons for nothing.
Though it would be a solution to rising sea levels due to melting polar icecaps – removing the excess water for use in desalination plants to irrigate California and other dry areas – win-win situation?
It’s hard to significantly change the amount of water stored on continents. It is a thing, and it does have an effect, but it’s at least one, maybe two orders of magnitude smaller than the effect of melting icecaps.
(Fun fact: The floods in Queensland, Australia, in 2010/2011 got so much water on land that they dropped global sea level ~7 mm. By comparison, seas are rising ~3 mm/year)
It really kinda is.
Uh, it may as well be a magic inexhaustible water supply:
Surface area of the oceans = 361,132,000 sq kilometers or 361,132,000,000,000 sq meters
If we take one millimeter off the top, we get 361,132,000,000 cubic meters, or about 293 million acre feet, so if the 6 million acre feet shortage were drawn from the ocean, it would lower the ocean by .02 millimeters per year, assuming somehow none of the water found it’s way back to the ocean, which in practice basically all of it will.
The term “water supply” refers to the supply of water. As in, stuff that’s water and nothing else. Seawater is water plus a bunch of other stuff, and thus is not part of the water supply.
But the government pressure causes people to save in extremely uneconomical ways. Spending hundreds of dollars on a low-flow toilet which will save water worth a few cents (if it is even that much) per year is a giant waste of resources. Raising the cost of water, on the other hand, will cause people to save water in more or less the most economical ways.
I agree that the way government’s handling it doesn’t seem particularly effective, especially considering that most of the waste is in agricultural use rather than home use. In principal though, I’m arguing that a free market on water for home use (while possibly more efficient than the sort of government regulation we have now), isn’t necessarily an ideal solution. For a start, there’s the question of incremental utility – another liter of water costs the same amount for anyone, but someone who’s using it to drink gets way more utility than someone who’s using a far greater amount of water to take an extra ten-minute shower. In a theoretical perfectly rational free market this might be solvable (assuming you have a universal basic income so that everyone can afford basic drinking (and hygiene – there’s also a public health aspect here, which poor people may decide to cut back on at a cost to society)), but this is one of those places where real-world complications can be an issue.
The main practical issue here is conversion – forcing everyone to personally consider how much water they’re spending and personally decide how to cut back takes a lot of mental resources from people who’re already pretty busy, and less cost-effective government-mandated cutback measures which people don’t have to think as much about may allow saving mental resources by being simpler to follow (there’s a tradeoff here which may or may not be efficient depending on your numbers for efficiency differential vs benefit from mental resources).
TLDR: Water’s a very important resource, and while basic free-market algorithms may be fairly effective, they’re not necessarily the most effective (at least for personal use), and it’s possible that the overhead costs of government intervention may be worth the benefits from a better distribution algorithm.
I thought we were discussing a free market?
Isn’t the idea that buying the same thing costs the same for everyone in the definition of a free market?
No, you’re thinking of market equilibrium or the efficient market hypothesis. Free markets matter most in precisely the short-term situations where those principles don’t apply and you instead have price competition: one person offering more value than another for the same resource or action.
A market where there are droughts and water scarcities and yet each ounce of water costs the same to any person in it would have to be a highly regulated one, in a free market the price of water is going to vary constantly based on how valuable the water is to each buyer and how hard it is to get them that water at the time they want it.
Yes, utility will be different for different people and uses, but the market is already the best known way of handling this issue. The person who’s just buying water for drinking will outbid the person who buys water for a swimming pool every time, precisely because the first person values the water much more highly.
What alternative mechanism do you propose? Should there be some government agency that hands out water at different prices based on what it’s used for? Seems like a lot of unnecessary overhead. And more importantly, how will that agency know what you’re using the water for and how urgently you want it?
I don’t get your argument about mental resources. We already have to make these considerations for everything else, whether it’s food, fuel, shoes, or cigarettes. Somehow people are dealing with that just fine. Also, don’t people in California already pay for water?
As for water-saving devices being easier on mental resources, that really isn’t an argument for government interference. People who value not having to think about how much water they’re using can just buy those devices themselves, while everyone who’s not bothered by that doesn’t need to make uneconomical investments.
About the first one, a simple idea is to make water for home use incrementally more expensive (Say you don’t pay for an amount that should suffice for your basic needs, but pay extra for waste). In theory ideal rational players would easily resolve this by free market, but in practice you need some kind of measure to get people to change their water use if you’re having a sudden drought and are risking running out of water.
The point about mental resources is that I (and most other people) are really really bad at taking the step from “man my water bill’s high, wish I could do something about it” and actually researching what an effective solution would be and using it, and a government body that does all the research, figures out what a widespread solution that ensures reasonable water saving without hurting people too much, and forces people to stick to it would be a lot more efficient (assuming you can trust your particular government to do this well). It’s the same principle as the difference between opt-in and opt-out insurance.
Your suggestion of not charging anything for the first X units of water per household would lead to misallocations of resources. Average costs for the water provider actually decrease with greater water usage (since a substantial part of the cost is fixed costs for pipes and such), so having a price structure for consumers that’s the opposite of that will give the wrong incentives. Also, households which, given normal prices, would consume less than X are now incentivised to waste water.
If you are really so concerned with poor people not being able to afford water, then give poor people money, but don’t mess with the pricing of commodities like water. More importantly, such a concern is misplaced. Poor people currently pay regular market prices for other necessities, such as food or fuel, which are far more expensive than water. Water should be the least of your worries.
Speaking of incentives to waste water, I plan to start watering our lawn once this drought and associated water restrictions are removed. Why? Because our water district, like many others, is requiring everyone to use a fixed percentage of their previous usage or face fines or higher rates.
When I was a kid, the last time Jerry Brown was governor and there was a drought, the water district we were in had fixed allocations based on the number of people in the household, and charged a (punitively, sort of) higher rate on excess water use. That system does *not* create an incentive to use more water in non-drought years.
Unless you’re using the water to make rocket fuel, it will go back into the oceans. And if you are, it won’t be enough to stop the sea levels from rising.
I think even rocket fuel reenters the Earth’s biosphere. Most of it is burned while the rocket is still in the atmosphere, right? About the only way to take water out of the oceans is either inland lakes, glaciers, or creating chemically altering into other forms, such as hydrocarbons.
Hmm. This is a singularly useless thought experiment, but allow me to geek out for a minute… you could refill aquifers, or create new ones if you had the right geology to work with. You could inject it into a subducting plate boundary, or directly into the mantle if you had a deep enough hole; some of that would come back up as water-rich magma, but some wouldn’t. You could crack it into hydrogen and oxygen, store the oxygen somewhere, and the hydrogen gas would escape the atmosphere eventually. You could launch it into the upper atmosphere, where the temperature’s high enough to keep it in the gas phase.
There’s probably stuff I haven’t thought of. But that all sounds super expensive.
There’s the old Combine trick of placing a teleporter directly in the ocean and draining it out to some other planet(s). The Visitors also stole our water, but that was just by shipping it out on giant spaceships.
It’s a mystery to me why people think it’s so hard to flush the toilet less or take shorter showers, but think it’s a great idea to just stop growing food. Flush toilets are a convenience; plenty of people live without them. Food isn’t a convenience.
Don’t troll.
Food grown in unsuitable conditions( mainly for export) that requires massive subsidies is.
Food in general isn’t a convenience. Particular kinds of food are. We could feed our cows something other than alfalfa, or we could feed ourselves something other than meat. We cannot literally just stop growing food as Scot Alexander suggests, but it’s not hard to save water on the food front.
Everything is commensurable. A large enough amount of a convenience is more important than a small enough amount of a necessity.
Sorry, I don’t buy that the entire water usage of a CA family can grow only $10 worth of alfalfa.
That’s nice, but show your calculations then?
Are you sure you’re comparing apples to apples? To buy out the farmers wouldn’t necessarily involve paying them the entire value of the crop; it’s profits they care about, not turnover.
Two things worth nothing:
1) People are arguing that paying farmers not to grow alfalfa might have negative effects on the feed supply for cattle. It’s worth pointing out that a large percentage of the alfalfa is exported (I don’t have the exact percentage, but $586 million worth from the US last year). Hence delightful BBC headlines like “California drought: Why farmers are ‘exporting water’ to China”
2) Geography is an issue here. The United States has enormous amounts of freshwater on the whole (those are seriously *Great* Lakes in the northeast) but moving it around is pretty tough. Alfalfa farms become irrelevant if they happen to be in an area that is rich in water but we have no way of getting that water where it’s needed. A thorough analysis would need to take this into account.
Is it just me or does anyone else, while looking at the chart up there, think “wtf… lawns?” If the water shortage is bad enough that people are encouraged to use water-saving toilets and take fewer showers, why don’t they just stop watering lawns? I mean, from my general observations, if you stop paying attention to the lawns, all that happens is that wild weeds grow there instead of lawn grass and you save a shitload of water, some gas (not using the lawnmower) and a lot of human labor.
From my completely ignorant and uninformed view, it seems to me that California is a desert (or very dry) climate in many regions, but people want things like “long, relaxing showers” (and even in wet Ireland, the notion of “long, relaxing showers” is still a very new thing) and green lawns. The idea of having sprinklers running all day to keep the grass green and growing (instead of natural state of scorched brown and dried out vegetation) boggles me, because over here it’s as you say – the climate is wet enough that you can’t stop the damn grass growing in the summertime!
Swimming pools, though, are the one that really make me go “What the heck?” I mean, those are pure status symbols.
Exactly. All of these things are status symbols. How enjoyable is a lawn? I mean, they’re all right, but I don’t have one and I’ve survived.
The big thing is how enjoyable it is to have the only yellow untidy dying lawn on your street. That sounds negative many many enjoyable.
Swimming pools are pure status symbols? I mean, I guess I was jealous of my friend’s swimming pool as a kid, but mostly I just wanted to swim in it. It’s not like they’re not functional.
Kids don’t really have an adult understanding of status. That’s not to say they don’t get status — they totally do — just that they don’t tend to be jealous of adult status symbols and vice versa. When you’re ten, your friend’s Jordans are probably much more impressive than your friend’s father’s wingtips, even though the latter cost three times as much and the former wouldn’t pass any dress code on Earth that cares about shoes beyond “wear some”.
With a few exceptions. I remember being very impressed with the Porsche belonging to a family friend when I was a kid, but young boys are often impressed by things that go fast and are painted red.
A single swimming pool is more functional than every lawn on the planet.
/my totally unbiased and completely 100% accurate opinion.
“The idea of having sprinklers running all day to keep the grass green and growing”
Just for a reality check … . I have a lawn (spotted with fruit trees). Any given sprinkler currently runs for something like ten to twenty minutes a week. If the sprinklers were running all day I wouldn’t have a lawn, I would have a swamp.
As anyone who has ever asked their children to water the lawn knows well.
If the water shortage is bad enough that people are encouraged to use water-saving toilets and take fewer showers, why don’t they just stop watering lawns?
Or, yanno, drain the shower water onto the lawn.
My mom uses her bath water for the lawn and such so she can justify bathing. And now she’s consistently watering all the potted plants she didn’t before.
Great post. This is the type of stuff I tune into this blog for. Debunking of tribe specific rhetoric through the injection of lethal doses of perspective!
The local congress critter is actually on the resources committee and knows his stuff.
California put off building any more storage capacity since the late 70’s and yet our population grew and grew.
He cited one example where we could easily have had more stored water: The Shasta dam was designed to be 800 ft. tall, but they stopped at 600 ft. (IIRC)
Merely building it up to the original specs would have done a lot.
The Auburn dam is another project that should have been done.
This is not rocket science.
So is California an agricultural economy?
By which I mean, never mind all the talk about Hollywood and Silicon Valley, does the economy of the state run on the vineyards, orchards, and industrial-scale commerical market gardening and farming to provide all-year-round fruits and vegetables for the nation? It may be that you will have to go back to seasonal crops and reduce irrigation demands that way. Is the country going to accept “No, you can’t buy lettuce twelve months of the year”?
As for “pay alfalfa farmers not to grow crops”, this is the question of farm subsidies for set-aside and good luck with working that one out in an American context.
You might have better luck with using this as PR for vegetarianism/veganism: “Look how much scarce water is consumed growing fodder for livestock rearing! If people stopped eating burgers and drinking milk, see how much we could save!”
Of course, then you’ve have all the farms switching to growing cereal crops for humans to eat and consuming just as much or more water, but that’s life! 🙂
No, California is definitely not an agricultural economy. Agriculture and mining account for 2 % of California’s GDP:
http://en.wikipedia.org/wiki/Economy_of_California#/media/File:Gross_Domestic_Product_of_California_2008_%28millions_of_current_dollars%29.svg
That graphic represents the flow of money in and around the state (rents, real estate, etc) which don’t bring money into the state. Actually read the text and you will see that California is the largest ag state in the union by far and one of the largest in the world. California agriculture is not as trivial as the Number you quote would suggest. You are referencing the graphic incorrectly. The industry alone generates over $100 billion per year, much of it through export (bringing money into the state, which is a very good thing). Use common sense, too, food is not eternally abundant and when you constrain a finely tuned, massive market like agriculture, you and cause pretty awful side effects in other areas that you may not foresee. I, for one, would leave food supply out of the discussions on how to deal with this problem. It is too important for survival and involves science that most do not even begin to understand. Look at what many people are saying on this thread…”find a better plant than alfalpha” and such. Just because you can think it from your comfortable suburban home does not mean that you have suddenly solved a crisis. A crisis caused by population explosion in the state, not the pre-existing agriculture industry that feeds the nation.
” The industry alone generates over $100 billion per year, much of it through export (bringing money into the state, which is a very good thing)”
Why is producing food, exporting it, and using the money to buy and import something else and consuming that, a better thing than producing food and consuming it in state?
The idea that exports are good, imports bad, embodied in the term “favorable balance of trade,” is eighteenth century economics.
Doesn’t a lot of water “used”, either for agriculture, industry or domestic purposes, make its way back into the rivers? Budgeting solely on rainfall vs water use doesn’t seem right.
In terms of buying out the alfalfa, it is plausible that removing all CA alfalfa production would significantly increase the market price (I can’t find the numbers, but the US is 41% of world production, and CA seems to be a major contributor. Even a few percent drop in global production could have a large price impact), so pricing it at $860M might be off (but probably not by a factor of 2).
Are you Milo Minderbinder?
See, shit like this is why I’m starting to end all my speeches with the line consilium communis rusticarum delendum est.
Yes, markets are clever enough to fix problems caused by huge-stupid-government-programs-that-help-like-5-dudes. But you know what else is clever enough to fix problems caused by huge-stupid-government-programs-that-help-like-5-dudes? Not having any huge-stupid-government-programs-that-help-like-5-dudes.
I don’t want to make fun of this here, but … acre-feet. Really, Americans? Acre-feet?! An acre is not even square. Like, it doesn’t make a difference because we’re only talking about ratios here … but … acre-feet?
43,560 cubic ft.
I also just found out that converting cubic ft to gallons is not an integer only operation. One cubic ft is about (about! I … I don’t … it’s probably an irrational number, isn’t it?) 7.4805 gallons.
*crying European noises*
I cannot convey in words how much I agree with this.
This was my initial reaction as well.
It’s worse up close.
What gets me the most that all fractions of inches are reduced in standard tools, so the order typically goes 3/64, 1/16, 5/64, 3/32, 7/64, 1/8
When that’s really just 3, 4, 5, 6, 7, 8.
FFFFFFFFFFFF….
It’s a weird-looking unit because it’s optimized for doing agriculture in the US. Agricultural land in the US is measured in acres; water needs for crops are measured in inches of water (originally of rainfall) per unit area. So if you’re a farmer and you need to calculate the water you’ll need to buy from an irrigation authority to make up the shortfall after N inches of rain, you get a value in the tens of inches times some number of acres.
It doesn’t cleanly convert to cubic feet, but it doesn’t need to: irrigation flow calculations work just as well if they’re done natively in whatever volume unit. It does get messy once you start interfacing with urban water systems, which do measure in cubic feet, but as we’ve seen that’s a relatively minor use case.
Acre-feet are a very convenient measure for what it’s mainly used for: measuring rainfall, irrigation, and reservoir stocks.
Irrigation rates are generally specified in water depth per time interval (e.g. a common rule of thumb for home gardens and lawns is 1 inch of water per week). To get volume in acre-feet, convert the depth to feet and multiply by the area in acres.
Rainfall rates are similarly measured in depth over time. Same calculation to get volume in acre-feet.
Reservoir stocks are even easier. Multiply the surface area of the reservoir in acres by the average depth in feet.
The conversion factors are ugly, but they’re not used often, and they’re a lot less ugly if you round them to two significant figures (44,000 ft^3/acre-foot and 7.5 gallons per ft^3) which is good enough for most practical purposes.
In general when calculating with US Conventional units, it’s easiest to pick one unit and do your calculations in decimal quantities of that unit (e.g. “8.1 acre-feet” or “2.6e6 gallons”). That’s really what metric does, too, except the unit you pick is standardized and there are order-of-magnitude prefixes in case you don’t feel like using scientific notation.
I feel the same way about acre-feet as I do about kW-hrs: the physicist in me thinks they’re ugly and awful, but I get why they’re useful.
“An acre is not even square.”
I’m having trouble figuring out what that is supposed to mean.
I’m guessing it means it’s not the square of some linear unit or a convenient multiple thereof?
An acre based on the size of a typical manorial farm plot (a chunk of land that would be ploughed as a unit) in Medieval England, which measured 1 furlong (“furrow-length”) by 4 rods (1/10 of a furlong). The 10:1 length:width ratio was based on the farming technology of the time, as plowing with a team of oxen was much more efficient if you plowed long furrows and minimized how often you needed to turn the plow around. So an acre is 1/10 of a square furlong, or 1/640 of a square mile (a furlong is 1/8 of a mile).
One problem is that a major argument for low skill immigration is that someone needs to pick the crops on America’s subsidised plantations. It is already common knowledge that those plantations weren’t and aren’t profitable even in states where the free market price of water is low, but it has changed nothing.
I’m laughing at the thoughts of banning alfalfa production and assuming they’d switch to some other pasture. These guys picked alfalfa not only because it is a legume – high protein, gets nitrogen from the air, long-lasting, high production – but because IT IS A GREAT DROUGHT-RESISTANT FODDER CROP.
Alfalfa is one of the best drought fodder crops there is. Clovers have shallower roots and need more regular water. When everything else has died the alfalfa keeps on growing for a while longer. There’s not a lot else to switch to.
I’m curious where CA alfalfa goes to. I’m guessing local intensive dairies and local beef feedlots. You could truck this stuff in but transport costs are significant for a product that is only a hundred bucks a ton. Milk and beef prices go up a bit and some local production switches to something more profitable. Seems sane, and beef and milk producers in other parts of the country would be happy.
Eventually there will be another El Nino year like 2005 to replenish the reservoirs. In the meantime, the number of cows in California can be adjusted to the price of alfalfa, which can be adjusted to the price of water: just sell surplus cows to the Midwest or slaughter them early for beef.
Almond tree orchards are harder to adjust to the ups and downs of rainfall because the trees take a long time to grow and will die if not watered.
In general, economists over-stylize the California water situation by assuming there is a fungible market for water, which would be nice. But in reality the water market is less efficient because there are a lot of property claims on sources of water and not as many ways to transport all the water.
On the other hand, I don’t really know the magnitude of this inefficiency. What percentage of water in California is re-directable via aqueduct to the highest paying customers?
2005 wasn’t an El Nino year. There was a Nino event in 2006/2007 – that might be what you’re thinking of? Also one in 2009/2010ish, and there appears to be one brewing essentially now.
Source: Australian Bureau of Meteorology
just sell surplus cows to the Midwest
Looks like the Midwest does better for tillage than drystock or dairying:
If this short-on-numbers-long-on-scary-pictures article from the BBC is any indication, it’s going to China.
TIL: The article also seems to indicate that the Imperial Valley (in southeast California) is awash in Colorado River water, which could be transported to metro LA easily enough but not to the Central Valley.
This is a classic freeloader problem. Wells enable farms to drill around water restrictions.
The obvious answers for the long showers and the alfalfa and other crops is to just levy a sales tax on agricultural products, recreational boats and tap water.
Instead of paying people to not grow alfalfa or almonds or take long showers just levy taxes until the farming corporations switch alfalfa to the lush Mississippi basin in the midwest where floods of excess water are a plague.
The tax is perfect because California can keep the agricultural products that are financially viable with 74 MAF and use the money to keep investing in water efficiency. If alfalfa is still profitable, then grow it.
the farming corporations switch alfalfa to the lush Mississippi basin in the midwest where floods of excess water are a plague
Which is exactly why you can’t grow alfalfa there as successfully as in California; Ireland too is a damp climate (ask Scott!) and to quote the Teagasc website:
Lucerne is the main sown forage legume in the world. It is used primarily to provide feedstuff for housed livestock, being conserved as hay, silage, dehydrated pellets and cubes. Its use in Ireland diminished through the 1960’s, due mainly to poor persistence.
Lucerne is another name for alfalfa and it is prone to fungal diseases. You get fungal diseases in wet conditions. Alfalfa doesn’t like wet or cold conditions, and flourishes best in dry climes such as California, where it needs to be irrigated but once harvested dries very successfully in the hot, dry conditions.
As an aside, do we have any farmers on here? Anyone who’s ever drawn a load of silage?
Also, Ireland doesn’t get enough rain for alfalfa.
I’d guess Ireland is too wet for alfalfa, not too dry.
Deiseach was right to say that it’s too wet. But it is also too dry.
Alfalfa hates soggy conditions. It likes it dry-ish. It is perfect for CA and with irrigation you can get many more cuts in a year than you otherwise would.
Is… is that title an actual You Can’t Do That On Television reference?
If so, I stand in awe.
Is alfalfa used to replenish fields with nitrogen? It could be that growing alfalfa is important to the broader agricultural economy.
Or it could be that if water for alfalfa was made more expensive, farmers would find another way to revitalize their fields, such as with chemical fertilizers.
Alfalfa does replenish the fields with nitrogen. It’s the queen of forages in its own right, but you need to run it in a rotation since it starts losing strength after a few years. I think a lot of CA farmers run a few years of alfalfa (3-5) followed by a few years of cereals where they don’t need nitrogen fertilizer since they’re using all the N from the alfalfa.
Pingback: California, Water You Doing? | Official site of DJ Michael Heath
A lot of California is off the grid of the water supply network.
For example, the spectacular oceanfront land northwest of Santa Barbara is largely restricted to sparse dryland cattle ranching precisely because Santa Barbarans didn’t want to tie it into the California Water System, which would have allowed development as weekend homes and golf courses. There’s a 30 mile long strip of gently sloping coastal land about a mile wide out beyond Santa Barbara. If it were in Turkey or Spain, it would be wall to wall condos for the benefit of the vast population of Los Angeles. But Santa Barbara is pretty much where the modern environmental movement started with the oil spill in January 1969, so it’s just occupied by an occasional cow munching on the dry grass.
Pingback: 1p – California, Water You Doing? | Profit Goals
Pingback: 1p – California, Water You Doing? | Exploding Ads
Is this just another way of saying that cutting down on meat consumption, and especially red meat consumption, saves more water than cutting down on showers/laundry/etc.? Is there an available way to quantify this?
No. Because it’s not beef that’s at stake, it’s milk. See above.
Can you point me to the comment you’re referring to? I’ve run numbers and in units of water per kilogram product or even per gram protein, dairy products are much more water efficient than red meat. If losing 10% of alfafa (which is an interesting thought experiment) meant losing 10% of cattle, losing beef cattle rather than dairy cattle makes a lot more sense to me.
Edited: found the relevant comment.
Excellent post!
Allow me to put on my anarchist tinfoil hat for a second and suggest that the issue goes even farther than collectivists demanding water be a “public” resource.
Jerry Brown et al aren’t stupid (wait, wait, stay with me on this). They have access to the same numbers you do. They KNOW that the problem isn’t little Susie wanting to take a long shower, or even Mr. Bill wanting to water his lawn. But these are people who benefit directly economically, as well as have a vested philosophical interest in the expansion of state power.
“Emergencies” like this allow them to invoke the notion of “shared sacrifice” which is always a statist’s best friend. Sure, taking shorter showers won’t solve this problem, but solving the problem isn’t the goal. Legitimizing government intervention into our lives is the goal. They are trying to condition us to the fact that they can order us to do whatever they want, so long as they say it’s an “emergency.” An OPTIONAL “avoid all this nonsense for $2” program doesn’t achieve this – as the optional component makes it an individual sacrifice, no longer shared. If the catastrophe isn’t actually severe enough, (and a $2 per head price would suggest it clearly isn’t in this case), then the only way to get your “shared sacrifice” that you need to legitimize your power is to force everyone to sacrifice whether it is really actually necessary or not.
I would propose to you that this is a lot of the impetus behind various “solutions” to climate change as well, but that’s an entirely separate discussion.
Good point. All governments secretly love “emergencies.”
I think you have ends and means slightly reversed here.
Rather than viewing sacrifices as an arbitrary task to condition compliance, progessives (including my own relatives) typically think of them as “what you ought to be doing anyway” with emergencies as a way of justifying them to a greedy and shortsighted populace.
In their eyes, big green lawns bottled water flush toilets and steaks are symbols of white middle class decadence: unwarranted privileges we should be giving up even if water was plentiful. It’s literally embarrassing to them that people won’t willingly submit to these measures and view paternalistic government as the only response to this childish behavior.
I’m clear on the thinking behind the term “decadence”. It seems to convey some sort of derision for pleasure, that pleasure is not a legitimate terminal value; only pleasure at achieving some “higher” goal should be pursued. The idea that lawns are bad, not because of some other effect, but merely because they are an unseemly form of pleasure, is foreign to me.
Agreed. I like lush green lawns; to me they’re a Good Thing about the world that I want to continue to exist. If they’re too costly (environmentally or economically), then of course I’m willing to sacrifice them. But to sacrifice them for nothing reeks of Puritanism to me.
A lot of lawn-haters just don’t get pleasure from lawns, and don’t know that other people do. Lawns aren’t as pretty as most other ornamental plantings you could put in the same place; they don’t make flowers or food; and they’re itchy to sit on, annoying to maintain, and easily damaged by foot traffic.
With a little typical mind fallacy, it’s easy to assume all people who have lawns don’t really want to, but do it because everyone else is doing it and they don’t want to be looked down on, or possibly because their HOA will fine them if they don’t. I do think these pressures account for a lot of lawns in existence, and that society would benefit from switching to a less lawn-favoring set of social norms.
I think it depends on whether you’re talking about the politicians or the “true believers.”
I think most politicians (left or right) don’t truly believe in the values that they get elected on. They’re power whores who hitch their wagon to whatever side they think gives them the best chance.
For the average progressive on the street, I totally agree with you. But I don’t think this applies to Jerry Brown and his cronies.
There are a lot of people who’d like an excuse to give up taking care of their front lawns. I’m reminded of how the Energy Crisis of 1973 banned Christmas / Hanukkah lights. When the ban was lifted in later years, the percentage of people in the neighborhood who put Christmas / Hanukkah lights fell from maybe 80% to 40%.
Right now it’s still a tradition that you have a green front lawn, but a lot of people have important TV watching to do inside, so they wouldn’t mind if the government ordered them to let it die to save the environment or whatever.
I’m a little confused how you got those $2 and $5 figures.
California’s population is about 39 million people, and $860/39 is about $22, not $5.
That part is confusing. Scott is saying that instead of installing low-flush toilets and taking shorter showers, the residents should pay $2 each to buy out 10% of the alfalfa crop. This wouldn’t solve the whole deficit, the way eliminating alfalfa entirely would, but it would accomplish just as much as the residents are accomplishing with much more effort.
Point is, if those stats I quoted up the thread are right, you don’t want to be paying the alfalfa growers not to grow alfalfa; California grows a fuckload of the agricultural produce (including livestock and dairy) consumed by the U.S.A., on top of its exports.
So cutting water use by agriculture, by encouraging them not to use water in growing crops because they’re not growing crops, is going to have unexpected consequences when it comes to filing the produce and dairy shelves in your local supermarket.
For example – apparently there has been an increase in consumption of dairy products in the U.S.
Improved economy = more people eating out and eating out more often = increase in cheese and butter consumption = increase in demand for dairy = need for fodder for dairy cattle = increase in demand and price of alfalfa (one of the main fodders).
You’re not being asked to pay the price of the water to grow the alfalfa, you’re being asked to pay the compensation for not growing the alfalfa – how much is the farmer losing by not selling his crop? And that, I imagine, would work out to more than $2 a head (the cost of the water).
The best thing is to encourage efficient irrigation and less wastage, but that’s probably going to cost $$$$$$$ in infrastructure and a lot of hairpulling by various bodies about what the best way to do this is.
“how much is the farmer losing by not selling his crop? And that, I imagine, would work out to more than $2 a head (the cost of the water).”
$2 per head is *not* the cost of the water. It’s the cost of the alfalfa. To be precise it’s the cost of just under 10% of the alfalfa crop (which requires 0.5 MAF of water) distributed among California’s population. The entire alfalfa crop (which requires 5.3 MAF of water) would cost $22 / person as per dlr’s calculation right above.
Raise the price till the necessary reduction in use has been achieved, then the most efficient use of water will be found. Anything else, like all regulation, will be gamed. At some point progressives will realize they are not the smartest people in the room and their plans will fail to achieve their hoped for outcome – always.
I’m left of centre. I am aware that markets are very good at what they do. These are some of the reasons I don’t want to just sit down and trust the market for literally everything:
– Externalities. This is essentially the ur-reason here. Markets are excellent at allocating scarce resources in very efficient ways based on the information captured by the market, but that doesn’t necessarily include all relevant information here. Environmental flows, for example, have to be priced in somehow. Essentially everything problematic about markets is a specialised externality problem.
– I value things other than efficiency. For example, it’s entirely plausible to me that it’s economically inefficient to sell effective healthcare to poor people, to sell effective education to poor people, or to sell public transport in poor suburbs. I would prefer a universe where people didn’t die of stuff that was pretty cheap to treat, where essentially everyone has the opportunity to have an education.
– Markets are terrible at long-term consequences. Overfishing is not easily solved by markets, for example. In fact, the very structure of markets forces participants not to pay attention to long-term consequences. If you reduce fish take in order to not overfish in the long term, you’ll be outcompeted by someone less responsible.
There are other reasons that are more minor or just specialised versions of ones I’ve just listed. I genuinely think that a single coordinating body can do a better job in situations where those problems are significant than a market. I think that belief is borne out by observing more social-democracy countries compared to countries that have chosen more market-based approaches to similar problems (for example, tertiary admissions in Australia are very much government-controlled – ‘student loans’ in the sense of ‘loans for money you will use to directly pay for tuition’ are almost entirely government-provided, with very favourable-for-the-student conditions. Our universities are still pretty damn good. I think this works a lot better than commercial student-loan mechanisms for paying tuition).
P.S. Montreal protocol was regulation and achieved its hoped-for outcome of drastic reductions in CFC emissions.
The point under contention there is existence, not efficiency. Effectively educating the poor is the cold fusion of public policy: if it can be done at all, it’s cheap at twice the price.
Are you sure that’s a global view? I don’t think it’s quite the same over here. I referred to HECS above, for example, which is basically a government scholarship for the top n% of high school graduates by high school grades.
Yeah, we’ve got similar programs, and how to fund them is a color-of-the-bike-shed issue. The substantive problems are pass-rates and scalability.
We know a dozen ways to pay for college, and they may be more or less relatively efficient, but we know zero ways to circumvent ἀγεωμέτρητος μηδεὶς εἰσίτω.
No, markets are terrible at shared property. Market agents buy and sell 100-year bonds. The canonical solution here is to auction off the fishing stock, and then the owner of the fishing stock decides how much to draw down on it every year, who can fish, and so on. Even in an anarchic situation where they have to pay for their own enforcement of that property right, that can still be profitable.
Markets account for values other than efficiency. If people preferred expensive, hand-woven clothing to cheap, machine-made clothing, markets would provide inefficient, hand-woven clothing.
2nd. Markets are better at long-term consequences than the alternative, which is government. Democracy, especially, will sacrifice anything for the next election.
Just pointing to the existence of externalities is not a sufficient argument to justify government intervention in markets which are affected by externalities. After all, government itself suffers from massive externalities: most of the benefits and costs of voting acrue to other people, and most of the benefits and costs of decisions made by politicians and bureaucrats acrue to other people.
So while the existence of externalities means that markets will achieve less than optimal resource allocations, externalities distort government decision making to a much greater extent.
> Markets are terrible at long-term consequences. Overfishing is not easily solved by markets, for example.
I disagree. Overfishing is not easily solved because of the tragedy of the commons. If you sold the entire coastline to one company and let them choose how to fish, then either they’d focus on long-term profits and fish responsibly, or focus on near-term profits and sell the coast to someone that would fish responsibly.
Markets do value long-term consequences less than short term, but it’s doing it in the best way given that you’re valuing future money less. If you pay one dollar today to make sure you’ll have two dollars in a hundred years, you wasted your money, since you could have invested it elsewhere and had many dollars in a hundred years.
In support of this, cattle farmers and sheep farmers have yet to drive cows and sheep to extinction. There seems to be no reason to think that farmers are more foresighted than fishers, the difference is property rights to cows and sheep are more easily enforced.
Do farmers expect a new herd of wild cattle/sheep to wander into the slaughterhouse each year?
No, because if they did they would be hunters not farmers. Different techniques.
One factor is that it’s easier to keep track of how many cows are available for you to purchase or breed. It’s hard to collect data on how healthy the fish stocks are.
@Sylocat: Perhaps that’s the difference. But that’s an entirely different argument to James Picone’s claim that markets generally are terrible at long-term consequences.
Which is my point, since we’re talking about overfishing, which to my mind has more in common with hunting than farming. Farmers are, however, using more water than is sustainable, much like fishers overfish.
And I find it somewhat humorous that the “market solution” to this problem is to decrease competition (one entity owns rights to the entire coast, rather than several owning parts), which seems to perfectly agree that market competition forces short-term focus and that less market competition allows for better long-term planning. Of course, you still need enough market competition to avoid the ruinous consequences of a monopoly, and generally keep things lean and efficient.
I just don’t see how a free market is going to spontaneously achieve that equilibrium under its own power. It seems to me like some outside entity needs to apply pressure whenever it swings too far in one direction. It’s a pity that the government is so atrociously bad at doing so.
@InferentialDistance
Okay, so if I understand you all right, your claim is that the comparison to cattle and sheep farming is invalid, because the techniques used are different. Is my understanding right?
And, assuming I have understood your point correctly, if I find you some examples of fishing and other hunting things that take place on private land (or water), where if the owner limits their extractions they don’t have to fear some other fisher/hunter catching the young they left, will you then agree that markets can be quite good at long-term consequences?
Economics has moved on from the classical models of the early 20th century, with infinite competition. I hope one day that the critics of economics will also move on, although this is yet another case where I distinguish between the words “hope” and “expect”. It is not the case that more competition is infinitely good, transaction costs matter, it is not the case that markets can be neatly divided into small segments. Any given species of fish eaten for food is in competition against other seafood species, and other types of food, like meat or vegetables.
If I can give you an example of this happening, will you then agree that markets can indeed in some situations solve things with long-term consequences, and thus that James was wrong to make a sweeping statement like “markets are terrible at long-term consequences”?
[Note: I am not going to claim that markets are perfect, the best summary I’ve heard on this is that “Markets fail. Use markets.” What I am disagreeing with is James’s extreme statement that “markets are terrible.”]
This is the issue. Are you claiming that in aggregate the market pays a lot of attention to long-term consequences? Or that in specific situations the market pays attention to long-term consequences? Because it appears to me both that the market in general is short-sighted, but in certain situations plans for the long term.
@InferentialDistance, my claim is something like that when people can reasonably expect to personally benefit from acting in their long-term interests they will do so (I may not have gotten the wording of this concept quite right), and beyond that, that markets relying on systems with that set up, help people to act more in the long-term than would happen in their absence. So if cattle farmers found themselves facing a situation where if they kept some calves alive to raise the next generation someone would come in and steal those calves to kill, they’d act like fishers in the Canadian cod fishery.
What evidence could convince you that my claim is right, or at least has some good evidence going for it?
Data showing that markets without intervention/regulation do better at long-term planning than markets with. Not a particular market, in a particular measure, but in aggregate.
The issue is that you’re caching a huge amount of information in “reasonably expect to personally benefit”. With what frequency can an individual agent “reasonably expect to personally benefit” from long-term planning in an unregulated market? Across myriad different economies and scenarios?
I don’t disagree with your implication, I disagree about the probability distribution of the antecedent.
@InferentialDistance, thanks for this reply. It’s a good start, now I need some details.
Firstly, the original claim was that markets are terrible at long-term consequences. Now you are talking about “long-term planning”. But it’s possible to be excellent at long-term planning and so bad at implementation that you do far worse at the consequences than someone who is much more spontaneous. Did you intend to change the claim? (If so, I’m not minded to dispute the new claim, I consider it quite plausible that the average government department does much more long-term planning than the average market, it’s only the consequences they fall apart on.)
Secondly, what data counts to you as a convincing measure of long-term consequences (or indeed, long-term planning)?
Thirdly, what does “without intervention/regulation” mean to you? For example, do you regard property rights as an intervention in the same category as The State Central Planning Agency mandating that every farmer should raise 4 more cows next year? How do you think about tax rates as a form of intervention? Do you have a metric in mind for intervention/regulation? If not, can I suggest the Heritage Foundation’s Index of Economic Freedom, or some of its sub-indices? (Note, I’m happy to discuss an alternative measure if you know of one.)
Fourthly, why do you want me to only compare markets with intervention/regulation to markets without it? Isn’t it at least hypothetically possible that markets with intervention/regulation (depending on how you define that) could do better at long-term consequences than markets without it, but both sorts of markets could still do better at long-term consequences than other forms of economic allocation.
I’m a little confused by InferentialDistance’s claims, as long-term investment is far from an exotic thing. That and the general success of capitalist societies in increasing wealth over the last few centuries look like counterarguments to the claim that long-term thinking is an unusual exception (if that’s the claim).
So if cattle farmers found themselves facing a situation where if they kept some calves alive to raise the next generation someone would come in and steal those calves to kill, they’d act like fishers in the Canadian cod fishery.
Alternatively, a culture might develop a strong taboo against killing cattle, involving vegetarianism, some taboo against killing other creatures, etc — ie a ‘fence around the law’.
I think part of the disconnect is the relationship between externalities/shared commons and long time periods. Externalities and shared commons are disproportionately long term. Market competition applies pressure to make long term sacrifices for short term gains. Because any agent who gets enough of a short term advantage can kill off the competition before the long term consequences can come to bite them in the ass. This doesn’t eliminate long term planning, but it does reduce it.
The market solution to a lot of these problems is to adapt to the consequences once they’ve happened, rather than avoid them. The market is good at adapting, but that’s different than being good at long term consequences themselves. The market doesn’t deal with he threat of drought by preemptively decreasing water usage, it drains aquifers and then switches to water efficient options once the water is gone (more specifically, once the cost of extracting usable water increases above the R&D and infrastructure costs of more water efficient technologies).
@Tracy W
I’m using planning and consequences synonymously. When one considers long-term consequences, one is “planning”. And by long-term consequences, I mean benefits and costs of that particular action that are time distant (10+ years) from the point of action.
That is a difficult question. The short answer is that there probably isn’t, because I’m lazy and opinionated. The minimum is a cost benefit analysis that includes the long term consequences of shared commons and externalities, and some indication of what percentage of long term consequences are shared commons and externalities, relative to short term consequences.
Property rights enforced by an armed authority (i.e. police) do not count as regulation/intervention. Taxes are a form of intervention, since they effect transaction costs. I have no particular metric in mind.
Because, colloquially, “markets are bad at X” is shorthand for “we need more intervention/regulation to achieve better X”. Which, for X = “long term consequences”, I broadly agree with. If you’d like the caveat that “most long term consequences I care about” are also shared commons or externalities, well, I also agree with that.
@Cauê
I did not say long term investment isn’t a thing, nor did I say capitalism is bad at generating wealth. Short term investments can have long term consequences. Long term investments can have short term consequences.
“Markets are bad at long term consequences” is not “the future is bad”, it’s “the market is trapped at local maxima”.
@InferentialDistance,
Ugh. Now you’ve introduced a third claim. We’ve gone from “markets are terrible at long-term consequences” to a request to show that “markets are better at long-term planning”, to a statement about whether markets are better at considering long-term consequences. But these are all different things.
Let’s consider winning the lottery as an example.
1. I consider the consequences of winning the lottery. I consider the pros of buying a flash car and a nice house and a swimming pool, and so forth, against the costs of family members hitting me up for loans and so forth.
2. I plan to win the lottery. Eg I plot some cunning way of hacking the lottery so as to increase my odds.
3. I implement my plan to win the lottery, successfully.
I don’t see how being good at 1 is synonymous with being good at 2, let alone how 2 is synonymous with 3. With 1, I’m just daydreaming. With 2, firstly, I may never put the plan into action, secondly, isn’t it entirely possible that even if I did try, I’d stuff it up, get caught and sent to jail and have terrible long-term consequences?
Anyway, I’m not going to defend the idea that markets are better at considering long-term consequences, any more than I’m going to defend the idea that markets are better at planning long-term consequences. The only position I’m going to defend is that markets are better at the actual long-term consequences.
Yay! An actual number! So, just to get confirmation, anything I come up with that involves a horizon of over 10+ years you will agree is long-term? You absolutely promise not to change your mind and suddenly claim that an example that spans 11 years doesn’t count as long-term?
So what metrics do you want to use for the costs and what metrics for the benefits?
Hmm, this one is going to be tough, because I have no idea how you can apply percentages to causality. And I’ve never seen anyone do this, I’ve studied a fair bit of history and I’ve never seen anyone say something like “Okay, WWII was 31% caused by the Treaty of Versailles and 22% caused by the political system of the Japanese…” So, please explain to me how to do this calculation.
Also, I don’t see how this calculation is relevant to any of your three claims. Please explain.
So, are you happy then to use the Heritage Foundation indices? You absolutely promise to accept the Heritage Foundation indices (not necessarily the overall Index) that I use as evidence of market intervention/regulation, and that you will not change your mind afterwards?
Sorry but this statement is wrong. There are plenty of people who think that various activities should be nationalised.
So, just to summarise:
1. You agree that 10+ years counts as long-term.
2. You agree to use the Heritage Index of Economic Freedom or its sub-indices as a measure of intervention/regulation.
3. I’m going to present evidence about markets compared to non-market options as you have given no valid reason as to why anyone should only consider markets.
You still need to say:
1. What claim you actually want me to defend. (Note, I’m not going to defend any input claims. I find it quite plausible that, for example, governments are better at both considering long-term consequences and planning for long-term consequences than markets, my only disagreement is whether non-market systems are better at those actual long-term consequences).
2. What metrics you want to use for costs and what metrics you want to use for benefits.
3. How you calculate the percentage of consequences attributable to any particular cause, as a general technique.
4. Why you think this percentage is relevant to the question of whether markets are better at long-term consequences.
ID, do you mean that when long-term investments happen in markets it’s because people are going for their short-term consequences?
—
Your arguments and examples seem to support not the thesis that markets are bad at long-term thinking, but that they are bad at commons and externalities, which is less interesting.
Just chiming in to note that InferentialDistance is expressing what I actually meant pretty damn well – if a market can trade off short-term benefits for long-term harms, in the presence of competition, almost everybody will do it (i.e. it’s Molochian). “Markets are terrible at long-term consequences” wasn’t intended to be incredibly precise.
It is often the case that the long-term harm being traded off is an externality/commons thing, but I’d expect a number of non-externality equivalents. Hell, sometimes this is a feature – it’s the thing that (most of the time) stops companies forming pricing cartels. Every company in the cartel is better off in the hypothetical universe where they all maintain the cartel, to in the universe where competition drives prices to market-bearing, but some company will probably break the cartel hoping to hit the jackpot (although that’s a shorter-term long-term consequence than InferentialDistance was meaning).
What I think is an appropriate response depends on the specifics of the problem. Sometimes it’s just “accept that this will happen”. Can’t think of an example for that offhand. Cartel-breaking behaviour above I guess, but that’s a good thing, we don’t really want to anything about it. I’m finding it hard to generate non-commons examples that aren’t just “Company did something really stupid”, like the SCO lawsuits (in the short term: share price went up. In the longer term: company died). That might be a warning flag that I’m being beguiled by commons problems often being long-term, I guess.
Sometimes I think regulation to try and prevent specific instances of the problem is a good idea. Example: fish stocks, cap-and-trade or Pigovian taxes, various banking regulations (not an expert, but I do get the picture that what we’ve got now is genuinely better than the good old pre-regulation days of runs on the bank).
I’m finding it hard to generate examples where this particular issues is why I think they should be government-owned. There are economic sectors I think should have a government provider in them, in addition to private providers (healthcare, education), but that’s for other reasons (the ‘the market’s utility function doesn’t match mine’ thing, where I value lives more than the economic value they generate, and education more than the economic value it generates). I do think public telecomms networks are a Good Idea, and this sort of thing is partially why (our previous public telecomms network got privatised, and since then hasn’t bothered to maintain the network. I don’t expect a good national network to arise through patchworky competition. I think a nationally-owned-and-maintained network that commercial providers sell services over solves both problems with acceptable losses).
@James Picone
What? You also think that considering the long-term consequences of something is synonymous with actually achieving those consequences? So, for example, daydreaming about winning the lottery is synonymous with actually winning it?
Thus explaining why cows went extinct. Everyone killed all their calves immediately for short-term profit, rather than bearing all the costs of food and water and maintaining fences in order to raise the next generation of cows.
No, because that has approximately nothing to do with what’s actually being suggested.
(Although I do think actually considering long-term consequences is a good way to get better-than-average outcomes)
No positional advantage here, which is why it doesn’t happen.
Any halfway-functioning system can get long-term consequences right in easy cases. Markets screw up particular classes of hard cases. It’s entirely possible I’m getting dazzled here by commons problems often having the short-term payout / long-term harm form, though, so maybe this is just a reflection of that. And/or individual companies making terrible decisions.
You may be working with an overly restrictive definition of “economic value” there.
@James Picone:
I suggest you re-read Inferential Distances’s comments. ID introduced the word “synonymous”.
But, if you are the one farmer who slaughters all their cows immediately for short-term profit, rather than expending resources on raising the next generation of cows, then you will in the short-term have more money than all the longer-term thinking farmers. So you’d have a short-term positional advantage.
I’m guessing by “positional advantage”, you’re coming back to the point of the common-access tragedy: if anyone could come in and take the calves the farmer is leaving for the next generation, then cows would go extinct. But, if the farmer is reasonably confident that they’ll get to keep the next generation of their cows, then they’ll keep some calves back.
Sure. Markets, to apply the Churchillian quote on democracy, are the worst form of economic system, except for all the others that have been tried from time to time.
This is not supported by the cod collapse in the early 90s.
Scientists had been warning for nearly two decades that the north Atlantic cod stocks were being dangerously overfished and were destined to collapse, and all the fisheries had just stuck their fingers in their ears and sang really loud. And when the fish stocks were finally depleted and no one was catching any cod, people refused to even admit what had happened in front of their faces, casting about for something else to blame besides themselves (they settled for scapegoating seals, thus giving them an excuse to revive the whitecoat hunt).
Would a single company have done better than the smaller fisheries? Maybe, but given the ridiculous incentives for short-term profits at the extent of long-term sustainability, I doubt it.
By this logic, cows have gone extinct. After all, you can always earn far more profit in the short-term by slaughtering your cows than by spending tons of money on food and fences and veterinary care and stud fees to raise another generation of cows to sell for food.
The difference between cattle farming and fishing is not the difference between short-term profit opportunities, it’s that if an individual fisher leaves more fish for next time they can’t stop another fisher from catching those fish instead. That’s the different signal.
A single company undoubtedly would have done better, because it would have owned the fish.
What you are describing is the tragedy of the commons – and what you are ignoring is the solution to it, which is to not have any “commons” in the first place and to make everything privately owned.
To expand on Tracy’s point, compare cattle versus the American Bison. Biologically they aren’t that different of an animal. Bison could produce meat and milk for human consumption easily. So why are cows abundant and bison rare? Because bison were, for a very long time, impossible to “own.” They were common property, which provides an *additional* incentive (above and beyond the short-term economic value) to quickly consume the resource, because if you don’t, someone else will.
If I own the entire coastline, I don’t have to hurry and catch as many fish as possible out of some fear that if I don’t catch them, someone else will. I can take my time and make the best economic decision (which, nearly always, favors a certain amount of sustainability and protection of a resource’s long-term value).
The solution is to make the commons government-owned and have the government license it out. Partially because it often involves ludicrously centralised control of huge resources that, realistically, are never going to all be owned by one company (is the libertarian Montreal protocol implemented by having one company own the entire atmosphere?), partially because there are approximately zero benefits to this being privately owned.
Aren’t monopolies supposed to never happen and/or be unstable? Isn’t private ownership of an entire fish stock / entire coastline basically a monopoly?
@James Picone: the whole concept of a monopoly is one of those things that turns out to be awfully complex when you get into it. There’s a lot of things to eat that aren’t fish, let alone a particular species of fish. If one person controls the entire cod catch of the world, but everyone is happy to switch to eating herring instead, does that person have a monopoly or not? There’s only one supplier of Lady Gaga live performances in the world, is Lady Gaga a monopolist?
And, there’s a lot of coastline in the world. Owning one stretch of coastline is not the same as owning all the world’s coastlines.
Markets are usually better at incorporating long-term consequences than the busybodies constantly questioning them. Markets assign positive, even very substantial values to assets that might not produce profits until a decade from now, or have a high chance of never producing any profits. Then people come in and claim that oh, it must be a bubble, look at these high P/E ratios, blahblahblah.
By contrast, elected representatives perennially fail at properly valuing long-term consequences.
Which is to say, the same thing the state did, except the market would have decided that “once” was circa 1985, instead of 2005.
You are correct that markets are imperfect for at least your first reason–market failure, situations where rational action by individuals does not produce rational action by groups.
But it takes a candidate to beat a candidate. Market failure ultimately comes from situations where individuals are not paying the cost of their actions or are not receiving the benefit of their actions, as in the classic case of air pollution. Such situations occasionally exist on the free market and lead to suboptimal outcomes.
But such situations are the exception on the private market, the rule on the political market. Collecting the information needed to vote intelligently is producing a pure public good with a public, in the national case, of three hundred million people. No congressman, judge, or bureaucrat bears any significant fraction of the costs, or receives any significant fraction of the benefit, from the decisions he makes. So while market failure occasionally occurs on the private market, it routinely occurs on the public market.
Before you conclude that authority ought to be shifted from the former to the latter, you need some argument to show that the latter will actually make more nearly correct decisions.
Your other point has to do not with market failure in the economic sense but with the market producing a different distribution of income than you would prefer. But, while government can redistribute from the rich to the poor, there is neither theoretical nor empirical reason to expect that to be the consistent pattern. In practice, governments redistribute in all directions, buying the votes of concentrated interest groups, rich or poor, with the money of dispersed interest groups, rich or poor. And the redistribution is costly, since everyone involved has an incentive to modify his behavior so as to pay less and receive more.
For more details of the first point, see:
http://www.daviddfriedman.com/Machinery_3d_Edition/Market%20Failure.htm
No congressman, judge, or bureaucrat bears any significant fraction of the costs, or receives any significant fraction of the benefit, from the decisions he makes.
This is only true in the idealized political market; in the real world corruption gives enormous, personally concentrated benefits for the decision maker if they vote/decree a certain way. And the choice that has the reward is nearly certain to be anti-optimal.
(You are of course aware of this, I am spelling it out for the others)
And yet, governments routinely do things that are generally pretty great, that I value, that most of the citizenry of my country value, and that I very much doubt would have occurred in a libertarian context. For example, I can’t sing the praises of our system for government funding of tertiary education enough. Wiki article. Healthcare here isn’t Swedish levels of amazing, in particular dental stuff is almost entirely private, but our Medicare system means that people do not, as a rule, get bankrupted by medical treatment in Australia, and people who are in difficult-to-insure classes won’t be out of pocket.
The only reasons there’s a national telecomms network in Australia is because respective governments built one – it used to be publicly owned, as the company Telstra, until the Howard government privatised it.
Prior to privatisation, it was a highly-regarded institution. I’m not joking – they were considered efficient and effective providers of telephone services with a mandate to provide telephone coverage everywhere in Australia.
Post-privatisation, they don’t quite reach the bottom of telecomms company ratings, but only because more fly-by-night operators are there to be even worse.
(To an extent this is because the Howard government privatised them without splitting apart the telecomms business end and the network – if they’d been turned into two separate companies, things would probably have ended better, rather than them occupying a monopoly space).
There’s currently a government infrastructure project to provide faster internet across Australia – the NBN. I doubt that would have ever happened under private enterprise (note that cable deployments had been limited to small patches in some capital cities prior to this, and were extremely expensive, and there weren’t many regulations covering them), and I’m expecting some serious benefits to the country if everywhere has reliable fast internet, instead of the current situation where it all comes over rotting Telstra copper (once it was privatised it had no incentive to maintain the network, for some reason).
That’s all local. More globally, would lead have ever been removed from petrol under a purely free-market system? Would the Montreal protocol have been signed? I’m not sure. These aren’t minor things. This is government doing exactly what we have it to do – solve big dumb coordination/externality problems with serious consequences.
I have a sneaking suspicion that libertarian impulses are so strong in the US because you just plain have a fundamentally bad system of government.
“I value things other than efficiency.”
No, you don’t.
Efficiency is how well one’s goals are achieved. One’s goals are, by definition, what one cares about.
“For example, it’s entirely plausible to me that it’s economically inefficient to sell effective healthcare to poor people”
How does such a statement make sense? If you want to sell effective healthcare to poor people, then clearly your utility function has a large weight for the sale of effective healthcare to poor people, and thus selling effective healthcare to poor people would be an efficient way of increasing your utility function.
Your statement is not wrong, it’s less than wrong. It fails to assert a coherent claim.
Your statement is not wrong, it’s less than wrong. It fails to assert a coherent claim.
Someone saying that they don’t care about efficiency is in the end saying that they do not care how many resources actually reach the supposedly intended goal. At best this is evidence of ignorance, at worst it is evidence of manipulative villainy.
If I lose money doing that, it’s not economically efficient in the sense I meant.
In a hypothetical free-market utopia, where it’s also the case that it’s actually impossible to provide effective healthcare to poor people without losing money (this is likely the case if you set the boundary for ‘poor’ low enough), then the only way people that poor will get effective medical treatment is charity. I don’t think a system reliant on charity is likely to be very comprehensive.
There are coordination problems that mean that even if everyone in Free Market Utopia prefers this set of people to have medical care, they won’t necessarily set up the charity or fund it sufficiently, even if they’re actually the idealised economically rational agent.
Government has done a pretty good job of solving that problem in a number of countries worldwide, some do better than others.
Alternate explanation: The market is, as far as I’m concerned, an unfriendly AI. Not the actively malevolent case, just the one where its utility function doesn’t match CEV of the country.
…And your point is?
Notice that public transport was also one of the things where I think running unprofitable routes can be overall more satisfying to society’s utility function than not running an unprofitable route. Pick a better example. 😛
I don’t understand your point, though. From where I’m coming from, I value people as more than their economic value – so I support policies that provide more healthcare to the poor than their economic value can justify. Education creeps in by a similar mechanism, public transport is mostly because there are significant positive externalities to public transport. I think if you instituted Free-Market Utopia somewhere on Earth, the numbers would work out such that there are people in society poor enough that they can’t afford what we would consider moderately basic healthcare – say, removal of wisdom teeth. I would prefer people be able to have their wisdom teeth removed even if they won’t be able to pay it back.
Pingback: The California Water Problem | Transterrestrial Musings
There was a time when “drought” meant a direct and serious threat to human survival (die of thirst). Then it became a term representing a broader and lesser threat to human sustainability (failure of agriculture, famine). Now it is term of threatened political leverage and memetic indoctrination (follow my rules or suffer my wrath). Ultimately, this is about the speed of potential harm. Imminent in the first case, gradual in the second case, and chronic in the last case.
Pingback: What is stupid? | Don't Worry About the Vase
Great article. California’s alphalpha producers presumably have some effect on the world alphalpha price, so you’d have to pay them more as you moved up the demand curve.
My personal creeb is that — as a single person with ecologial leanings living in a house in a suburban residential neighborhood near los Angeles — I realized about fifteen years ago that global climate change would probably reduce the amount of water coming into California and result in a serious shortage. So I reduced my personal water use by about 50%. Five years ago I reduced it again (though not by such a large %age) to about 25 gallons per day, which is, IIUC, about 1/3 the average. Then I was forced to reduce it by another 25%. What I’m doing, actually, is reducing landscape watering by about 90% — I really don’t like the idea of flushing the toilet only once a week, rather than twice, or taking a bath only every two weeks, rather than weekly. (Hey, I’m an Old Geezer — nibbling away at my 87th year — and grumpily claim a Need for a weekly Hot Bath…the greywater from which actually takes care of flushing the toilet and watering several plants.
I must say, I am tickled that this post plus comments reads like an issue of The Farmer’s Journal 🙂
Pingback: California Water Numbers, David Henderson | EconLog | Library of Economics and Liberty
California is not interested in simple and sensible solutions to problems, as demonstrated by its complete lack of interest in fixing the water market.
For a solution to be palatable to Californians, it must involve:
– a way to make better off white people feel guilty
– a plentiful helping of feel-good meddling in other peoples’ affairs
– several years, if not decades, worth of expensive studies and reports
– one or more new government agencies
If the new policy actually addresses the original problem, that is merely a fairly unlikely side effect.
Apologies if I’m mistaken, but it is really possible that nobody has linked to this yet? At $700/acre-foot, that’s about five gallons for a penny. The great advantage of prior appropriation doctrine over riparian water rights (as we have back East) is that farmers actually do have a transferable right to water itself. That seems far more straightforward than subsidizing farms not to grow alfalfa, and maybe it’s possible to grow alfalfa in a manner that is more water-efficient.
(Edited for spelling errors)
People have touched on the technical difficulties of selling water upthread: it’s not nearly as simple as that link makes it look.
Fracking doesn’t use much water, but it does contaminate aquifers via improper or illegal disposal of waste water. I don’t have numbers – maybe the amount of water lost to pollution is trivial as well, but a bit more nuance would be appropriate.
Your $860 million figure for Alfalfa is presumably revenue, not profit. Alfalfa farmers have costs. So compensating them for all of the water they now use should cost a lot less than $860 million.
I think he was going for an upper bound.
Can you do one of these for colony collapse disorder? I gave up trying to untangle all of the accusations in a Reddit AMA last night.
My understanding is that there’s always been a nontrivial attrition rate for kept beehives over the winter. The search term is “winter colony losses”.
For the past 10 years or so, the attrition rate has been quite a bit higher than in has been in the past (IIRC, 20-30% instead of 10-15%). This is well within the range where it’s technically feasible to replace the colonies (the total number of active colonies in the US has actually increased slightly over the past 10 years), but it’s worrisome because the cause of the increase is poorly understood (there are several plausible hypotheses, but evidence has yet to converge on a satisfying conclusion) so there’s worry that the cause might be something that could continue to get worse.
The cost of establish a replacement colony is also apparently high enough that the increase risk of winter colony losses represents a hardship to the traditional business model of commercial beekeepers: this can be expected to drive up the cost of honey and the cost of bee-pollinated crops whose farmers need to pay beekeepers to maintain the colonies that pollinate their fields. The latter is particularly an issue since the attrition rate seems to be different depending on what crops the bees pollinate, with almond crops facing the highest attrition rate, up to 70%.
One question about the graph:
I’m wondering where golf courses fall into it. From what I’ve read I get the impression that a lot of the haranguing I’ve heard over them in California is people looking for a boogeyman, but I’ve also heard that they take more water than all residential use combined. (Though, that might just be in the San Diego area, since the radio here is where I’ve heard that figure, as opposed to the state as a whole.)
As such, I’m not really sure where they fall in that graph. I’m assuming they fall under “industrial and commercial,” but that doesn’t jive with that “more than residential use” figure I’ve been hearing. Or do they fall under lawns, because really, a golf course is pretty much just a big lawn? (Though, most likely my guess that this figure only holds true for the San Diego area is probably the truth. Either that, or it’s not true at all. I’ve really only heard it in passing.)
Either way, I’d be curious exactly how much water is used by golf courses. Like I said, enough people are complaining about them that it’s probably worth addressing.
Unfortunately this seems unGoogleable. I’d be astonished if golf courses use more water than residences, though, unless “residences” and/or “golf courses” are being defined in some extremely sketchy ways: they’re inefficient water users, of course, like all lawns are, but in terms of acreage they’re nowhere near big enough to make more than a small dent.
That being said, if we’re looking for symbolic rather than substantive cuts, they’d be a good target. But I suspect most of the column inches devoted to them have little basis in fact and a lot of basis in sensationalism.
Golf courses take up a not trivial amount of land, as you can see out an airplane window.
Say there are 16,000 golf courses in America and the average one is 1/4th of a square mile (160 acres), so that’s 4,000 square miles. If 1/10th of all golf courses in America are in California, that’s 400 square miles, or a 20 by 20 mile square. By way of comparison. Orange County is 948 square miles, a lot of which undeveloped mountains. So California’s golf courses probably take up roughly as much land as the developed part of Orange County that is home to 3 million people (and several dozen golf courses).
The Palms Springs low desert area has 124 golf courses, or maybe 30 square miles of golf courses. Some of those are not watered edge to edge, but many are. Palm Springs has a huge underground aquifer left over from the glaciers on Mt. San Jacinto during the Ice Age, so most golf courses have various grandfathered in rights to cheap water from wells, often on the course.
There are probably a lot of ways to save water on Palm Springs golf courses, although I don’t know if the freed up water can be put into aqueducts for other uses or not.
In some places in California, golf courses are being paid to remove turf and replace it with attractive sandy waste bunkers. I believe in Arizona you are only allowed to water 90 acres of turf, so you typically have to drive your tee shot over the desert to reach the fairway, which is fun.
Another possibility is to stop keeping Palms Springs golf courses open and green all summer when it’s 110 degrees. I’m sure at least half the water is spent in July, August, and September when the courses have to discount severely to get masochistic golfers. Summer shutdowns might involve replanting the courses with grasses that can survive on little water for a few months of extreme heat. I don’t know if such grasses exist, but they probably do.
Third, the elite golf world is enamored of a trend toward “fast and firm” golf courses that aren’t as heavily watered. The ball rolls further and in unexpected fashions, which is fun in the old St. Andrews style, but the turf isn’t Augusta National green. Last June’s U.S. Open at Pinehurst was played with dry, brownish fairways, but I don’t think the average golfer got what the USGA was doing aesthetically. Donald Trump tweeted that Pinehurst looked terrible, and Trump represents mass market tastes in golf courses.
From a 2014 article on Palm Springs golf courses:
Statewide, golf courses are a much smaller consumer of water. A recent industry report prepared for the California Alliance for Golf found that golf courses account for less than 1 percent of total fresh water used in the state.
In the Coachella Valley, in contrast, golf courses consume much more — an estimated 17 percent of the total water use, according to the Coachella Valley Water District. While agricultural irrigation uses more water overall, many farms rely on Colorado River water rather than wells, and therefore use less groundwater.
An estimated 20 percent of the groundwater pumped each year in the valley is used for farms, and 55 percent goes to cities, residential customers and other businesses. The remaining 25 percent flows to golf courses.
The water district says that on average, each course uses about 1 million gallons a day.
http://www.desertsun.com/story/news/environment/2014/03/19/golf-major-drain-coachella-valley-water-supply/6594019/
I wouldn’t be surprised if Palm Springs/Coachella golf courses account for maybe half the water usage by California golf courses due to the extreme dry heat there. So if you could cut Palm Springs golf course usage in half, which seems conceivable, that would save 1/4th of the water used by golf courses throughout the state, or maybe 0.25% of all the state’s water, without inconveniencing most of the state’s golfers.
(Warning: extremely stylized numeric estimates).
The point is that Palm Springs golf courses are extreme water hogs, golf could contribute some sensible reforms in Palm Springs that would save a small but non-negligible amount of California’s total water usage (well under 1%, but probably over 0.1%), without ruining golf in the milder parts of the state where most people live.
Yeah, the idea that golf courses take up more water than all residences combined seemed a little outlandish to me, but like I said, it was something I’ve only heard in passing on the radio, and they were almost certainly talking about San Diego specifically, which makes sense, since there aren’t too many large lawns out here.
That said, I’d imagine they still take up enough water to be visible on the graph, if not to make much of a dent in the theoretical family’s budget.
The air quality benefit you get from living downwind from an area of lush green lawn and shrubs is of the same kind, whether the lawn area belongs to a golf course or to several private residences. Adjusting for likely wind conditions, there’s probably an optimum size and shape for the lawn area to build up an atmosphere of good air, then lose enough at the edges for the neighbors to get a noticeable benefit. Too small a lawn, and the good air gets dissipated too widely to be noticed; too large, and the air in its middle gets better and better but the output at the edges does not get larger.
The problem is that when you pay people not to farm X crop, there are magically more people every year who want to get paid for not farming X crop.
In Afghanistan, they used to pay farmers not to plant poppies. What happened is the number of farmers growing poppies exploded so that they could take advantage of getting paid to destroy their crops.
Also, you shouldn’t just look at how much the industry makes when doing the analysis. The $860m is taxed, so the real cost isn’t just paying off the farmers but also the opportunity cost of collecting their taxes. The state will have to come up with the money to pay off the farmers, and the new farmers added every year, without counting on the taxes generated from this industry. The state will also lose tax revenue from other parts of the industry, like paying employees and employees getting taxed or similar effects in supporting industries.
The solution is to produce more water. Because money is needed to generate more water, the state should not destroy industries that generate revenue that could be used to help this happen. How to do this efficiently, I don’t know. But I do know that agriculture is what makes civilization possible. Without agriculture we will be reduced to hunter/gatherer lifestyle which does not allow for specialization of skills that comes from people not having to grow all their own food. My god people, agriculture is what allowed humans to stop living like animals.
Greater consideration is required if the only solutions you can think of require the destruction of the pillars of civilization.
Spending thousands of dollars to produce more water to grow crops that are only worth hundreds of dollars is a terrible idea.
But it’s not a terrible idea to let people use water for golf courses and lawns and swimming pools?
If we really are at the point of “spending thousands of dollars on water for crops that are only worth hundreds of dollars means banning such crops” then it is worth getting urban users to implement reduction in water usage, despite what Scott says; if growing alfalfa for dairy cattle is unsustainable and unjustifiable, how much more unjustifiable is growing lawn grass that will be mowed and be used for nothing, not even feeding cattle? Or golf courses – how much do you raise on green fees, and is it worth the price of the water?
Not subsidizing something is a lot different from banning it.
Golf courses probably *do* make a profit on their use of water; they’re mostly private businesses.
The Contra Costa Salt Water District (ok, the word “salt” isn’t actually in their name, just in their product), during the last Governor Brown drought, rationed water based on the number of people in the household, and the per-unit price above the limit was about 5 times the per-unit price below. The rations were reasonably generous if you didn’t maintain plants (which takes a lot of water in months where the high temperature was over 104F/40C *every single day*), but required lots of conservation in the house and very efficient irrigation if you wanted to keep anything going outdoors. My parents managed to do that: I was of an age where bathing only weekly was not an imposition to me, and Bermuda grass will turn brown but not die given 6 months with *no* water.
That plan is a much better idea than what most water districts are implementing, which is semi-fixed fines if you exceed 80% of your use in 2013, which penalizes the people who started conserving water in 2012 or earlier.
This might be a dumb question – but golf courses can water the course with gray water, right?
I don’t doubt that government policy is set up in such a way where they don’t need to, but presumably if there was a functioning market here, gray water would be significantly cheaper than potable water, and golf courses would no longer be competing with the urban residential households for the supply of a certain kind of water, yes?
“Gray” water requires a distribution system that generally doesn’t exist. There are places which sell the treated water from their sewage treatment plants, and many require “reclaimed” water to be used for earthwork construction (you would be surprised how much water is needed to move dirt) and dust control. You *can* water lawns, fairways, and greens with it, but except in places where there’s pipeline distribution, golf courses aren’t going to be that interested because of the cost of trucking water.
Right.
But can we assume the fact that using reclaimed water isn’t economical is HIGHLY dependent on the fact that water is rationed by the government and thus does not have a true market price?
If golf courses currently have access to potable water at a rate significantly below what the free market price would be, that provides them an incentive to not bother researching the feasibility of re-using “dirty” water.
A lot of Palm Springs golf courses use water from sewage treatment plants, although this has to be supplemented with freshwater in the summer when there aren’t enough visitors flushing their toilets.
Most of the older private golf courses in Palm Springs have wells into the giant aquifer.
Is it legal in Palm Springs to pump up water and sell it to Los Angeles instead of using it to water your golf course? There are a lot of laws and regulations restricting sending ground water out of the watershed. That sounds strange but there’s an “I drink your milkshake” problem — everybody with a well in Palms Springs is tapping into an aquifer that may or may not be interconnected, so it’s okay to water as much land as you own, but not to hoover up vast amounts of water out of the aquifer and pump it over the mountains in return for money, even though that makes sense.
This puts me in mind of what Douglas Adams wrote about rhinoceros poachers:
Is there anything to stop an alfalfa grower from getting both subsidies? ie. accepting the countersubsidy, packing up, and then accepting the original subsidy to grow more alfalfa (somewhere else, under a different name)?
I may be overlooking some obvious legal or political thing that would stop this from happening.
It’s a lot harder to hide two hundred acres of alfalfa than it is to hide a dead rhinoceros, and unused farmland isn’t common or cheap.
Might be possible under unusual circumstances, but those aren’t what we’re worried about.
Another article from Marginal Revolution on the subject, in which you learn that the production of almond in California cost more in water than they are worth:
http://marginalrevolution.com/marginalrevolution/2015/03/the-misallocation-of-water.html
An article by a French economist on the subject, with plenty of reference on english article.
http://blog.francetvinfo.fr/classe-eco/2015/04/02/secheresse-en-californie-economique-pas-climatique.html
Pingback: California, Water You Doing? « Economics Info
My vague understanding of the Sacramento River Delta thing is not that EVERYTHING is ruined, but that without that flow, the aquifer in the region becomes (too) brackish and agriculture becomes impossible, and there’s a lot of agriculture in the Sacramento River Delta. Also a bunch of things that are conventionally thought of as environmental damage happen like large numbers of animals finding that their current habitat is no longer inhabitable for them. But even if we don’t care about habit loss, etc, at all, letting the Sacramento River Delta salinify would be roughly comparable to the proposed “stopping all growing of alfalfa” in terms of pure loss to agriculture production.
Pingback: Cali Water Debunking | Random Nuclear Strikes
Pingback: Lightning Round – 2015/05/13 | Free Northerner
Just a few observations:
1. It isn’t only six fish, it is six species of fish. We don’t really know the exact number of fish, so that could throw off your price-per-fish computation.
2. The Wall Street Journal is owned by Rupert Murdoch, so of course they are inclined to rail against environmentalists, and they’ll probably make up any statistic they want to make them look bad (just like FOX “News”). The WSJ would also probably want water to be a free-market commodity, because then industry would own it all and they’d be happy to sell it back to you, if you could afford it.
3. If an attempt was made to get the people to pay agriculture not to grow alfalfa, the dairy and beef industry lobbyists would go on the offensive and would probably win, though the real winners in that fight would probably be lawyers. Besides, who wants to have to deal with a lot of hungry, mad cows?
The link refers to a single population of steelhead trout. Not six species.
Sorry, my mistake. However, I suspect there are more than six fish in that population of steelhead.
The precise claim is apparently that they expected 29 fish to out-migrate down the river but most of those have already done so in PRIOR water releases, leaving 6 yet to travel.
Quote:
Although now that I do the math on that, 76% of 29 is 22.04. If 22 fish have migrated out of 29 expected, that tells me they are wasting 30,000 acre feet to help SEVEN fish. Not six.
(And I’m sure you’re all now saying “phew! in THAT case it’s all right then!”)
Actually, Rupert Murdoch just wants all of your water so he can use it to perform dark baptisms on his vile altar to the Koch brothers.
Wow!
what a great lot of research, compilation and analysis!
In Oregon, we are blessed with a lot of water and low population density (except for the retirees from Californai). consequently, most of us don’t have to worry much about water, except those near the Northern California border; there we have to balance things like crops and migratory fish.
Your informations gives me a much better appreciation for the juggling that goes on to make sure that ample water is spread to cover a variety of priorities.
And yes, I find your statistics “entertaining”. also informitive, and just a little bit frightening.
JTG
>Salon goes even further – their article is called Nestle’s Despicable Water Crisis Profiteering: How It’s Making A Killing While California Is Dying Of Thirst
This should definitely be an unvis.it link too. That headline is ridiculous.
According to environmentalists, this Sacramento River Delta water is non-negotiable, because if we stopped sending fresh water there the entire Sacramento River delta would turn salty and it would lead to some kind of catastrophe that would threaten our ability to get fresh water into the system at all.
This isn’t entirely true, but there are direct effects on people from insufficient fresh water getting to the Delta. Some water districts draw their water from the upper reaches of the Delta, and reduced flow into the Delta will make their water supply noticeably saltier. Back in 1976-77, I lived in one of those districts, and the local newspaper ran on the front page, in the weather box, the salinity level of the drinking water. People on low-sodium diets were advised to not drink water when the chloride level was >500ppm, and people on no-sodium diets at >200ppm. Most of the summer of 1977, the chloride level was above 200ppm, and it did occasionally break 500.
While I don’t disagree with your findings, there is a reason alfalfa is grown in CA. I have a farm background, so bear with me. Alfalfa requires a lot of water, but at very specific times in its development. So if you try to grow it in a climate that gets sporadic rainfall you will get inferior hay if it rains at the wrong time. The consequence of this will be that dairy cows (the largest consumers of alfalfa) will need to be fed more cereal grains to make up the difference in protein. This will in turn show up in the price of milk. Since we don’t have viable locations to grow much of what is grown in CA elsewhere (not just the hay, many other crops cannot grow to the same extent elsewhere) importing water for agriculture and passing thru the cost seems logical to me. In fact we will be importing “water” one way or another- either as liquid or finished foodstuffs. In the East this year they had horrific flooding that interfered with their food production due to too much snowfall. There is a solution, if politicians would just seek it out.
That sounds perfectly reasonable. if the factors are so much in favor of that then it should still be economic if the farmers have to pay market prices for the water and pass the cost through to those later in the chain. That would cause alfalfa to be grown where it is most economic to farm it given local weather conditions and water prices.
Very nicely done! I really like the info-graphic. I wish more folks were obsessed with understanding water use statistics, as a basis for some creative thinking, rather than trying to hide behind creatively contrived versions of statistics to support uncreative thinking.
Here’s an angle I haven’t thought about before: I would distinguish between flexible and inflexible uses of water. The amount of rain varies considerably from year to year: I can recall Los Angeles getting as little as 4 inches and as much as 36 inches of rain in a year in this century. Sometimes wet years pile up in a row and reservoirs have to let water out to the ocean unused. Other times dry years pile up in a row, as at present.
An example of an inflexible use of water would be an almond tree orchard, which can’t go without water during a dry stretch of years without killing all the trees, which tooks years to grow. I know very little about agriculture, but I suspect alfalfa might be nearer the flexible pole: it grows super fast. If water got very expensive for a few years, a lot of farmers might choose to sell their water to other users and forego raising alfalfa, then return to alfalfa growing when the rains come back.
Another aspect of flexibility is being able to survive the annual dry season without water. For example, Palm Springs golf courses pour immense amounts of water on their turf in summer partly to get greens fees from early morning golfers, partly so the grass can survive and look great for the big money winter season. (Palm Springs golf courses typically are open 11 months of the year, shutting down in the October shoulder season to be overseeded with winter grass, then reopening in November for the snow birds.) It might be worth studying if there were grasses that could survive dry summers without much water by going brown, but without huge bare patches emerging that would keep the courses from being unplayable in the winter.
Bermuda Grass, but while it’s an excellent lawn for kids, probably not so much for golfers.
Serious disclaimer here in that I’m literally recalling some mutterings from my father who’s the assistant superintendant at a golf course in Michigan from ~ a decade ago.
So even if my remembering is correct, the conditions out East may not apply to out West.
Basically, there’s 2 types of grass used at golf courses.
There’s the type they use up north (Bluegrass?) which is designed to be able to deal with snow and make it through the winter, but not months and months of heat, and there’s the type they use down South, which can’t handle a frost. And then every time there’s a cold winter or hot summer, you read about all the courses along the Mason-Dixon line that guessed wrong
I believe that the type they use down south is Bermuda Grass. Because it can deal with heat and sun.
If anyone has better knowledge than that, I’d love to hear it.
The blame game on this is taking everyone in. WAKE UP! Farmers have been a big part of California from the beginning. In fact, it is a slowly shrinking industry in the state for various reasons. What has increased water demand over the last forty years? The increase in the state population, especially in the more urban/coastal cities (LA, SF, SD). So, the leadership of the state has KNOWN that water (and other infrastructure) demands are steadily rising and have done nothing in at least four decades to account for it so that normal flucuations in rainfall can be mitigated. Now they are pointing to the evil farmers and corporations whose water rights predate the immigration into the state. They are not to blame for the failure in leadership. It is the state representatives who have known about this problem and ignored it that is to blame and they want you to look at anyone else but them as the cause. This problem has been building for a generation, and some hard times are coming before it is fixed. If you kill agriculture, that state will be drinking its own life blood for sustinence. One party controls the state and most of the over 400 elected offices never change hands due to gerrymandering. The state is a hot mess and all the people are still unaware of the true problem. It is a failure to provide the most basic infrastructure. It is failed policy. It is poor governance. This is the tip of the iceberg for California and as the population continues to grow, the sheer lack of infrastrucure will bring it to the breaking point. That is why I left. I still love that state. It is my home, I was born and raised there. My family still lives there. I hope they get it turned around soon!
You appear to have been mindkilled by politics.
Take another look at the graph. The entirety of *every single urban water user* in the state is 9/80ths of the water while farming consumes almost 4 times that.
If you literally lined up half the entire states urban population and shot them you still wouldn’t cut water use by as much as if you simply stopped farming alfalfa.
The alfalfa industry is worth less than a billion. The GDP of the state is 1.9 trillion.
Murphy, how many of the urban water users are contributing large amounts to that GDP of 1.9 trillion?
Going by this chart, the largest slice of the pie is Education, Health and Other Services – 18% and Government – 12%.
On the other hand, Information – 6% and Finance and Insurance – 6% (if they’re the big tech companies in Silicon Valley and the kinds of inward investment capital they attract) may be the ones creating the most revenue.
So if you sack all the schoolteachers and civil servants and tell them clear off with the alfalfa farmers to states with more water, you’ll probably cut your demand for water immensely, leaving more than enough for the remaining wealth creators in the computer/IT/financial sectors!
OBLIGATORY DISCLAIMER: THIS IS NOT A SERIOUS RECOMMENDATION (just in case anyone thinks I really do mean “Sack all the middle-class and make them move away to other states that get rain and stop using the water”).
Also, the taxable revenue from agribusiness, unlike that of (for instance) Apple can’t be parked in overseas subsidaries in legal tax-avoidance schemes. California probably gets a fairer share of tax from the alfalfa growers!
And seeing as how California is a leading agricultural exporter, any changes to agriculture there will have a nationwide effect:
Again, I’m not saying “Don’t make the agricultural industy pay its fair share for water” or “Don’t make them implement more efficient irrigation pratices” or anything like that, because the current situaiton is unsustainable.
But on the other hand, urban dwellers may have to face up to the fact that they can’t have green lawns; they get the sunny climes but they can’t emulate the lush green vegetation of states that are cooler and water-rich. You can’t have it all.
It’s not as simple as “well, just tell the farmers to stop growing such-and-such a crop”, unhappily!
OBLIGATORY DISCLAIMER: THIS IS NOT A SERIOUS RECOMMENDATION (just in case anyone thinks I really do mean “Sack all the middle-class and make them move away to other states that get rain and stop using the water”).
California land-use policies and their effect on housing prices certainly seem to be working in that direction. The problem is that the (drier, hotter) Central Valley hasn’t become nearly so expensive, so the middle class move there, then use more water. So we need to impose the radically restrictionist building policies of San Francisco and Berkeley on Stockton and Vacaville (both of which are commuter suburbs of San Francisco. Look it up on a map sometime) to keep the middle class moving all the way to Texas. And the uppers to Portland.
I’m honestly not sure what point you’re trying to make here.
Pingback: California Water Crisis | Wheat and Chaff
The solution is clear:
Pump seawater from the pacific ocean to the nearest deserts. Problem solved. Surely it would only cost $64 billion or so. A dried up state does not need high speed rail. Look up seawater greenhouse or saltwater greenhouse. You get food, energy and clean water.
Our eventual goal should be to replace all California’s agricultural water with seawater. Those who think this cannot be done have not done their homework on saltwater greenhouses and seawater greenhouses.
“Innovative Ways to Deal with California’s Drought”
If I’m reading this right, it’s a desalination/humidifier system that runs on solar heating? Clever idea, but I don’t think it’d scale well — putting any substantial proportion of California’s cropland under glass at all would be fantastically expensive, and desalination takes a lot of maintenance. I’m also not sure about the part where you use cold, deep ocean water as a heat sink — moving saltwater inland is one thing, but doing so while maintaining its temperature is quite another.
Unless I’m missing something, using groundwater is more like living off an inheritence than taking out a loan. Eventually, it will run out, but it won’t have to be paid back.
It’s more like drawing down a rainy-day fund (lack of rainy-day fund?). It will replenish slowly over time if you leave it alone.
Depends on the groundwater. Some aquifers are replenished from the surface, some aren’t: the search term is “fossil water”.
Excellent blog with analysis of the finger-pointing, but your fair market conclusion does not factor in growth. Making the residential switch from low-efficiency tech to high-efficiency tech is going to cut the current rate of usage and stop long-term residential usage growth. The state has effectively cut an exponentially growing water sector’s use and growth factor. If we were to save 2 billion AF/year today, the difference in growth trend will be much more impressive 5-10 years from now.
As always, neither a truly free nor absolutely controlled market is the way to go. The state is making a go at finding a good middle ground. I think Californians will like the results in 5 years even if the reasons behind the actions don’t make it up their asses (where their heads are).
Good analysis. But there would have to be other costs involved. If there’s no alfalfa, there’s an immediate upward push on the price of beef and whatever else eats alfalfa at least.
Nice analysis.
You might toss in bermuda and sudan grass into the alfalfa mix, along with a third of the corn and other grains per the UC Davis analysis and the Pacific Institute’s county and irrigation district breakouts by crop and irrigation.
A goodly portion of alfalfa and the above are exported to China and Japan along with the nut crop. That raises questions of how to motivate continuing urban sacrifice for whom a benefit is hard to see.
Remember when Wisconsin was the Dairy State.
Maybe it’s time to relocate the 2100 pasture-less CAFO dairies of southern California back somewhere where it rains. Gallons of water per ounce of beef are astronomic. We have Meatless Mondays already, Milkless Mondays could be the next step.
There are some interesting climate calculations here too with respect to global warming, in terms of cow methane belches and spared thermoelectric cooling of cooling power plants.
BuRec sends 24% of the Mead dam power to pump CAP water over the mountains, that’s a lot of juice that could go to a better cause. Double ditto Edmonton Pumping Station in the Tehachapis.
People in Nevada and Arizona might want to know what the level of Lakes Mead and Powell would be today just from cutting down on the flood irrigated alfalfa in the Imperial Valley. It is not too difficult to calculate that using the drip in feet per million acre feet going out the spillway, using http://lakepowell.water-data.com/
In Australia there’s water shortages in a number of areas with similar disputes going on for the last few decades at least. Sometimes the issues involved are centred around rivers that cross state boundaries, but it’s not always the case.
One similar misrepresentation in Australia is discussions that exclude the type of water. Agriculture does use massive amounts of water here too, but for city drinking water its mostly a non-factor. Plan water is derived, generally speaking, from totally different sources to most of our agriculture which is in remote regions. Such water is often from inland aquifers/bores, whereas city water is generally damns of rivers, coastal aquifers, with the odd desal plant. There is quite a big shortage of plan water in several places, but for plan water lawns are definitely the biggest single sponge, not industry or agriculture. So all those campaigns telling us to cut down our garden watering are actually more sensible than they seem when you just take a look at the raw numbers. Both environmentalists and industry are generally pitted against farmers for river usage, rarely city dwellers.
I personally favour temporary water rights should be auctioned off by government to the highest bidder (with the environment dept allocated a certain amount to bid for water that it sees as having maximal value), so you get the benefits of market-based allocation without the downside of water monopolies and other manipulations. Households might also need to accept a floating (lol) water price at some point.
Fracking doesn’t *use* much water. But it involves fracturing gas-bearing rock strata so the gas can be forced out and collected at well points. Groundwater that previously pooled in porous rock strata above that can now move through the coal/oil/gas-bearing rock, becoming contaminated. That’s part of the problem. Also, because the groundwater has moved downwards, that changes the water table, mucking up the hydrology of water wells, springs and creeks.
If I offered to get an oily stain off your swimming pool floor by pulverizing the concrete down to the soil beneath, would you worry about how long I planned to hose down the rubble to wash off the oil?
Letting the Sacramento river delta salt up would not just be an environmental issue, but would directly harm the goal of distributing water because a lot of fresh water is transported through the delta, down one stream and then up another.
I learned that from the nice article that ymcy posted about how the actual existing market works.
Pingback: 1p – California, Water You Doing? | blog.offeryour.com
Pingback: 1 – California, Water You Doing? | Exploding Ads
The bit about buying out the alfalfa growers isn’t quite that simple. If alfalfa production dropped that much, it would cause a lot of desperately starving cows, and the price of alfalfa would spike, and other farmers would start growing it as a result. You could probably get away with paying off, say, half the alfalfa growers to not grow without causing immediate horrible distortions, but nobody really knows. It’s extremely difficult to manipulate markets piecemeal, because you’re deprived of the information gathering role which free markets play and forced to guess at all kinds of things. Simply making farmers pay for water would allow for truly efficient resource allocation to take place. You need to start charging for groundwater as well to do it properly though.
A commenter said “We have a nation to feed, ergo this situation where farmers desperately produce more and more crops while driving the price ever downwards is somehow a bad thing.”
IT’S WHAT CAUSED THE GREAT DEPRESSION, LEARN SOME ECONOMIC HISTORY!!!!!!!!
The Great Depression was caused by cutting the money supply, at least as far as anything can be settled in history. That’s one of the things Milton Friedman won his Nobel Prize for, and as far as I know, no one has disproved the arguments.
The obvious cause of shortages of water, energy and trees is the rampant human breeding. In times of shortage, rationing should be limited to those who have contributed to the shortage by breeding; the non-breeders have already done their part in conservation.
California should pay me for getting a vasectomy and moving to Texas.
Some more data on alfalfa:
http://ucanr.edu/blogs/blogcore/postdetail.cfm?postnum=17721
Alfalfa does not really use more water than other crops. At full canopy (when the leaves cover the soil surface), alfalfa’s water use is not much different than any other crop (think spinach, lettuce, tomato, wheat, almonds or corn) per unit time. The Evapotranspiration (ET) requirement (the amount of water a crop really needs to grow) is remarkably similar across crops at full canopy (see FAO tabulated values for the water requirements of crops).
Alfalfa’s water use profile in California is primarily due to its high acreage and nearly year-round growth pattern in many regions. If spinach were continually grown on 850,000 to 1 million acres all year long, the water use would be about the same as alfalfa, perhaps more.
Further, it’s not so much how much water is used, but how much crop is produced per unit water that is important – also known as water-use efficiency. In that category, alfalfa shines.
Contrary to popular belief, alfalfa has several unique valuable properties and advantages which would enable cropping systems greater resiliency under drought conditions.
It is right to ignore acreage. What is important water use per acre-year.
Weight of crop per unit water is an idiotic measure. What matters is dollars of crop per unit water. Alfalfa is cheap and that’s why California should reduce it.
Yes, the fact that alfalfa does well in a drought is a reason for California to grow it. It is flexible, which means that it can be treated in different ways. But this is only useful if the flexibility is exploited, if it is treated differently in wet years and dry years. It should be irrigated a lot in wet years and hardly at all in dry years. In dry years, the desperate almond growers should pay a lot for water so that their trees survive, while the alfalfa buyers just skip those years.