11 May 2015

Supply-side solutions waste your money

Better for oil or milk than water...
Some people are suggesting that California's drought be "solved" by building a pipeline to Alaska or shipping water in by rail. These ideas, while feasible from a technical perspective, are a terrible idea economically.

First, let's ignore the environmental cost of pumping water up and down hills or fueling trains.

Now, let's just consider cost, i.e., 100 tanker cars carrying about 34,500 gallons (each) would bring just over 10 acre-feet of water. Under normal conditions, farmers would be willing to pay about $50/af or $500 in total for the water. But there's a drought, so let's assume they would pay $1,000/af, or $10,000.

Now, the cost of shipping water from a wet place to California is about 3 cents/ton/mile by tanker and 0.4 cents/ton/mile for a pipeline.[1] Using a distance of 750 miles from Seattle to Sacramento and a weight of 8 pounds/gallon,[2] we're taking about a cost of $324,000 for the train and only $43,000 for a pipeline delivery.

So, that's a pretty heavy cost for delivery of something "worth" far less.

Besides the other alternative -- dragging bags of water behind barges[3] -- it seems that the cheaper supply-side alternative is desalination, which only costs about $2,000/af. Given California's dire need for water now and the 20 year process of getting a desalination plant built, it seems that bags are the way to go -- from a supply perspective.

On the demand side, of course, you can get water "for free" by raising prices so that people use less outside on lawns, leaving more for indoor use. When it comes to agricultural irrigation, you can "free up" water by facilitating markets where buyers can pay sellers $1,000/af  (delivery included when they share canals, groundwater or rivers).

Bottom Line: Most "solutions" to drought and shortage fail the basics of economics. Hopefully politicians check their real facts before spending YOUR money.

  1. Lots of people are saying "if we can waste $80 billion on a high speed rail, why not build a pipeline to Washington State." That's false logic if you step back and consider that both projects are a terrible idea.
  2. It's around here that we are reminded how great the metric system is, i.e., one car with 131,000 liters (131 m^3, each weighing one ton) means each train that weighs 13,100 131 tons.
  3. The cost from Humboldt to SF or SD ranges from $550 to $2500/af. The cost on a train or trucks is probably lower (but getting a train over the coastal range?)
H/T to RM


  1. My head is still a little muzzy, but I don't think that 131,000 liters of water weighs 13,100 tons.

    131,000 liters = 131,000 kg = 131 metric tons.


  2. Yup totally agree. Although it would be interesting to see whether spending the HSR money on other nonsensical projects would be less bad.

  3. Remember to emphasize the margin, where a one-year irrigation water value of $1000/af would be hard to imagine.

    Desal costs confidently stated at this level, as if they're predictably stable in the face of risky energy prices, always concerns me. Another concern is the potentially underappreciated environmental impacts (and external costs) of desal. And if you're going to salute desal as the go-to marginal supply source, then you should also point to it as bounding the MC of raw water, with evident implications for the efficient pricing of water. Not only might this info guide volumetric water pricing, it also has direct implications for pricing new connections (i.e. for new homes).

    Saying that raising prices is "free" is in error and even conflicts with the intuitions of noneconomists. People don't think this is free because it isn't. There are consumer surplus implications, though the loss in CS is low enough that better pricing still trumps the supplyside options. Just say it is cheaper though not free.

    Climate change ratchets up the importance of better signaling – not just of water use and conservation activity but where people live and what they work upon there.

    I'm trying to imagine the potential stakes in the Southwest. Is it overzealous to imagine harsh declines in CA housing values, because of the region's inability to regularly support (at current population levels) the comforts and landscapes that people enjoy? Or will no one leap from the pot as it is brought to a boil? If it's time for people to move to the water rather than the other way around, then the sooner we get the signals corrected the better.
    ron griffin

    1. Great comment. Feel free to send a guest post on marginal and dynamic supply and demand policies/implications!

    2. Ps. More supply has opportunity costs in the same way that less demand does, so we're really talking about minimizing those, via policy, over the long run.

  4. "Saying that raising prices is "free" is in error and even conflicts with the intuitions of noneconomists. People don't think this is free because it isn't. There are consumer surplus implications, though the loss in CS is low enough that better pricing still trumps the supplyside options. Just say it is cheaper though not free."

    It's better than free... The price of water is below efficient pricing, so raising the price will boost producer surplus more than consumer surplus is reduced. Because water is a public good, this could be rebated directly (or indirectly through public programs, etc) to consumers for a net gain.

    Reducing DWL is not costly on the net just because CS could be reduced.

    1. DB, I'd juggle some of your language here, for better compliance with strict definitions, but your central thought is not objectionable. "Producer surplus" and "DWL" don't really work, because most water is supplied by a nonprofit entity, and the institutional conditions are nonmarket. Any gains that might accrue to US utility/districts are devoted to clients instead. So it really is about consumer surplus and consumer-received rents, as influenced by the overall rate system (including volumetric and nonvol. fees). Water doesn't qualify as a public good in economics (nonrival AND nonexclusive) although some water uses are public goods (e.g. biodiversity support, scenic views). [Lots of sloppiness and inconsistent usage even by economists here.] Terminology aside, yes! better pricing does yield net benefits because of the opportunity costs incorporated and the supply costs dodged. Some of these gains may accrue to future consumers instead of current ones, but they still count.
      ron griffin

  5. These are great examples but I wish when you calculate the cost of various alternatives that you also include the cost per gallon of better leak detection and maintenance. While I don't know the current leakage rate in California, I do know that in most developing countries these rates are typically 40% to 60%. This means that even after we pay the cost of treating and transporting the water nearly half of it is wasted. Singapore, by contrast, has reduced leakage to below 5% while most American cities still have rates of 20% or more. Consequently, in many areas of the world, by far the most cost-effective means of "providing" safe clean water is simply to reduce the waste of water.

    1. DF, 40-60% is very high for American contexts and a 5% claim is usually suspect. But regardless of the % loss, imo it is not wasted water unless the cost of reducing the loss (at the margin) is less than the value of saved water. That is, other resources have value too, and other resources must be expended to reduce water loss.
      ron griffin

    2. Leaked water isn't necessarily "wasted," either. Often it recharges aquifers and thus gets back into the system, and sometimes it has more complex benefits (cf. Imperial Valley, Salton Sea).


Read this first!

Make sure you copy your comment before submitting because sometimes the system will malfunction and you will lose your comment.

Spam will be deleted.

Comments on older posts must be approved (do not submit twice).

If you're having problems posting, email your comment to me