19 Mar 2009

Yes to Prices

(via PW) Professor Stavins is back at the Huff Post:
[We] analyzed the relative merits of price and non-price approaches to water conservation. We reviewed well over a hundred studies, and found strong and consistent empirical evidence that using prices to manage water demand is more cost-effective than implementing non-price conservation programs.
He goes on to "debunk" four misconceptions about water and prices, saying that:
  1. Demand WILL respond to price.
  2. Customers DO respond to price changes.
  3. Increasing block rates do NOT affect demand -- controlling for price levels.
  4. Price increases may NOT reduce quantity demanded because demand may only increase more slowly.
To these I replied with:
  • Point elasticity estimates are NOT going to be useful when price increases exceed 5-10% or so. With 200% increases (as I advocate), you not only get an elasticity response but you also get an "on the radar screen" response -- as we saw with $4gas. What happens next will be attributed to elasticity, but it's NOT a certeris paribus response. (Also note the "20/80" response -- 20 percent of people will conserve because its "right," 80 percent will because it's expensive NOT to...)

  • I am not sure about your frequency comment [billing cycle length doesn't matter]. Are you saying that monthly billing will generate the same demand response as annual billing? Doubtful, if you believe in bounded rationality.

  • I suggest that you pay attention to the "equity" third rail on water. Many people worry that "the poor" will be priced out of water if the price is allowed to rise to clearing levels. The obvious way to address that is through price tiers (I suggest per capita tiers), and cross-subsidies from high to low-consuming customers are both progressive AND "humane."

  • I suggest that most of the command and control responses are derived from the engineer culture that dominates water. They see the "need" for a 20% reduction and ORDER 20% reductions, as if people will respond like valves. Engineers inability (or weak attention to) the human element is also shared by those who favor C&T because they think that a "cap" will really happen. Good luck.
I do not agree with his third point, since that implies that people only look at the marginal price of water, but do not pay attention to the discrete move from one block of consumption to another.

Bottom Line: The best way to reduce the demand for water is through prices, not non-price methods, e.g., command and control, BMPs, regulation, etc.


  1. Point 3 suggests that “Increasing block rates do NOT affect demand—controlling for price levels.” In the article this is taken from Stavins writes “…analysis indicates that increasing-block prices, per se, have no impact on the quantity of water demanded, controlling for price levels.”

    These are interesting comments considering that in an earlier paper he makes the following claim ( there are a # of versions of this paper, I took these directly from "Muffled Price Signals: Household Water Demand Under Increasing-Block Prices." Working Paper, December 20, 2001. With S.M. Cavanagh and W.M. Hanemann.):

    “We find that the sensitivity of residential water demand to price is quite low, and that the effect of price structure may be more influential than the magnitude of marginal price itself.” ( in Abstract)


    “Water demand among households facing uniform marginal prices appears to be signiÞcantly less elastic than among households facing block prices, according to one meta-analysis of selected water demand studies from 1963-1993 [26]. If a house-hold knows that higher levels of use result in higher prices, it will be more sensitive to price. In fact, when we estimate our DCC model on households facing two-tier block prices alone, we obtain a price coefficient of approximately -1.00. For households facing uniform marginal prices alone, we obtain a coefficient of -0.19. This suggests that price sensitivity is higher for households facing block pricing.” (p. 26)

    I am not sure if they have changed these findings or come out with new ones, but these comments seem to suggest, at a minimum, that people behave differently under block rate pricing. Are these findings consistent with Point 3?

    There are a few other papers I could cite suggesting that households, holding prices constant, consume less (perhaps due to uncertainty) when facing block rate pricing, but I thought I would start with this.

  2. A few years ago the City of Lubbock Texas raised water rates to encourage conservation (cutting per meter fees and base rates, raising rates on high volume users). It apparently worked "too well", water consumption went down and water department revenue fell too much. Now the city is raising the per meter fee and base rates (and increasing the rates for high volume users).

  3. Mike,
    For Lubbock, is your comment the same as saying that the opposing forces are yearly income versus long term strategy and that yearly income wins?


Spam will be deleted. Comments on older posts must be approved.
If you're having problems posting, email your comment to me