Use Prices to Conserve Water when Supplies are Scarce

During times of drought, many cities try to reduce water consumption by restricting outdoor water use. Recent research suggests that a market-based policy—raising the price of consuming water—would be much less costly and achieve the same benefits.

With a historic drought affecting southern states from Florida to Arizona, millions of urban households are experiencing the typical approach to managing water shortages—outdoor watering restrictions. In much of south Florida, households can water, at most, two days per week. Major cities in Texas are also restricting outdoor water use, to two days per week in Austin and one day per week in San Antonio. In some arid cities, like those in much of the western United States and Australia, a summer without watering restrictions is unusual. Outdoor water use comprises, on average, about 30 percent of total consumption for U.S. households, but the fraction is much higher in arid climates and during the dry season. Restricting outdoor water use during a drought is common, but is it smart public policy? 

Basic economics suggests that this method imposes greater social costs on regulated households than a market-based strategy—that is, raising prices—for the same quantity of water saved. Higher prices harness households’ own incentives to reduce their water bills, according to their own preferences. Using this approach, utilities set an aggregate target for water use reduction and allow households to undertake conservation efforts that are in their own interests and collectively meet the aggregate standard. Watering restrictions, in contrast, require households to undertake similar shares of a conservation burden regardless of cost.

Our recent research quantifies the costs of rationing outdoor water in 11 urban areas in the United States and Canada, testing the hypothesis that raising prices could conserve the same amount of water at lower economic cost. To see whether households really are different in how they would prefer to conserve water, we divide our sample into four subgroups, based on income and lot size. Those with both incomes and lot sizes above the sample medians are categorized as “rich, big lot” households; those with both incomes and lot sizes below the medians are categorized as “poor, small lot” households; and so on for the two groups in between.  If these groups differ significantly in their response to a price change, in terms of their expected change in indoor and outdoor water use, this indicates that their preferences for water consumption in various uses are also different.  And the more different are household preferences, the more costly it is to ration water during a drought, compared to raising the price.

We can measure this difference in responsiveness by comparing price elasticity—the percent decrease in demand expected for a percent increase in price—across groups, for indoor and outdoor use. The price elasticity of indoor demand is zero or close to zero for each of the groups we examine, meaning that indoor water consumption will not change much with price increases. This is not surprising. Some indoor uses, such as drinking and cooking, fulfill basic needs. Others, such as showering and clothes-washing, do respond to prices. But households can reduce outdoor uses, such as lawn-watering, car-washing, and the filling of pools with less of a loss of well-being (and perhaps more easily, depending on households’ choices of water-using fixtures and appliances). In our sample, a 10 percent increase in price reduces outdoor demand by about 6.2 percent, on average. If current regulations target precisely the uses that households, themselves, would cut back in response to price increases, why should cities consider alternative policies?

The key to this question is the variation in the price-responsiveness of outdoor water use that we observe across groups. The rich, big-lot group has the least elastic outdoor demand (-0.42), and the poor, small-lot group has the most elastic outdoor demand (-0.79). The two middle groups appear to be about equally price elastic outdoors, and in between the “bookend” groups. We can use these different degrees of price-responsiveness to compare what happens when these cities restrict outdoor watering to what happens if, instead, they raise prices to achieve the same reduction in citywide water demand.

Consider a common policy, a two-day-per-week watering restriction (like that currently in place in many cities in Florida and Texas), which would reduce summer water consumption in our sample by about 32 percent. When constrained by such a policy, households’ average willingness to pay for an extra unit of water outdoors—if they could use more (their “shadow price”)—is about $5.36 per thousand gallons, almost three times the marginal price they actually pay during the arid season. But shadow prices vary dramatically across households within a city, a strong indicator of the potential gains from trade achievable using a market-based approach. Next, we estimate the prices that would be necessary to conserve the same aggregate quantity of water each utility would have saved by implementing the two-day-per-week policy.

We use all of these estimates—demand elasticity, shadow prices, and market-clearing prices—to simulate the welfare losses for each water utility from a two-day-per-week watering policy over a single summer, relative to the introduction of a market-clearing price. The estimates range from $2.09 per household in Cambridge, Ontario, to $407.66 per household in the service area of the Las Virgenes Municipal Water District in California. In our sample, society would be better off by an average of $96 per household per summer if a two-day-per-week watering restriction were replaced with price increases. In large cities, the economic loss from the current approach could amount to millions of dollars per summer.

The shift from outdoor watering restrictions to a price-based municipal drought policy would create economic gains, but prices would redistribute scarce water so that those with high willingness-to-pay would consume more than they do under outdoor use restrictions, and those with low willingness-to-pay would consume less. Thus, if one purpose of drought policy is to ensure that everyone does their share in shouldering the burden of brown lawns, communities may value this uniformity despite its economic costs.

Perhaps more importantly for political feasibility, households in each group would be worse off under drought pricing if water utilities were allowed to keep the extra revenues (about $152 per household) they would earn from raising prices during a drought. The solution is straightforward, and would likely be mandated by existing law—U.S. water utilities are usually restricted to zero or very small profits. The extra revenues could be rebated to households without dampening the effect of the price increase, so long as the rebate was not tied to current consumption. Returning the same amount to each household would make the overall distributional effects of the policy progressive. Without a rebate, drought pricing would be regressive, conditional on a household’s lot size.

Compared to the other arenas in which prices have been proposed to allocate scarce resources, such as congested roadways and airport runways, municipal water markets may be a relatively easy case. Household water use is metered. Compliance with raised prices would likely be higher than with watering restrictions; in the current context, cheaters can water at night or otherwise avoid observation by utility staff or vengeful neighbors. In contrast, cheating is very difficult under a price-based mechanism (households would have to figure out how to consume piped water off-meter).

This analysis has important parallels in other work in environmental economics. Researchers have demonstrated that flexible policies such as environmental taxes and tradable permit systems can reduce pollution at smaller social cost than prescriptive approaches. The reasoning is similar—market-based approaches exploit the variation in polluting firms’ marginal abatement costs, reducing total compliance costs relative to regulations that require the same behavior by all firms. Our research suggests that households, too, stand to lose when regulators adopt inflexible approaches to allocating scarce resources.

​