Watering the Big Apple

In April 1985, for the second time in five years, the mayor of New York City declared a water emergency. Lawn watering was limited. Open hydrants and the washing of cars, sidewalks, and streets with a hose were banned. Businesses and institutions were ordered to curtail water use by 15 percent. When these measures failed to halt the decline in reservoir levels, progressively more severe restrictions were imposed on water users during each of the summer months. It was late November before they were removed.

The shortages and restrictions seemed to confirm what the Army Corps of Engineers had been warning for some time: New York's water-supply system is not sufficient to meet normal levels of use during periods of drought. (Tables 1 and 2 give water supply and demand figures for New York City; figure 1 shows reservoir levels during recent drought years compared with a normal year.) The system's safe yield (the quantity of water that would be available under conditions similar to the worst recorded drought) fell short of the Corps' demand estimates by 120 million gallons per day (mgd) in 1970 and 227 mgd in 1980. Assuming that no changes in water prices take place but that water conservation devices are gradually introduced, the gap is expected to peak at about 300 mgd in 1990.

New York City's permanent water-supply system has not grown since early 1964 when the Cannonsville project nearly doubled the quantity of water the city could extract from the Delaware River basin. When it was authorized in the mid-1950s, the Cannonsville addition was expected to provide adequate supplies until at least 1980. Yet despite the additional capacity, New York City experienced shortages during 1964 and 1965. The record drought that afflicted the Northeast from 1961 to 1965 led officials to lower their estimates of the system's safe yield and to search for additional water supplies.

Urban water supplies generally are a local responsibility. But the drought in the 1960s prompted Congress to direct the Army Corps of Engineers to lead the planning for the long-term water needs of the densely populated Northeast. In 1971, after a decade of study, the Corps concluded that supplies were totally inadequate for New York City's growing population and recommended taking prompt action on its proposal to skim high flows from the Hudson River about 80 miles upriver from the city. After treatment, the water would be conveyed through a deep tunnel into the existing distribution system. The proposed Hudson River project would increase the system's annual safe yield by 390 mgd at an estimated cost in excess of $6.5 billion (in 1985 dollars). Even with a repeat of the worst drought on record, the resulting 30 percent increase in capacity would meet projected needs for the indefinite future.

Action stalled

Despite the sense of urgency conveyed by the Corps and the declaration of two drought emergencies (in 1981 and 1985) and one alert (in 1983), no action has been taken on the Hudson River project. Instead of being hailed for the relief it might bring, the proposal has been widely belabored. It has been attacked, for example, for giving inadequate attention to environmental concerns, such as potential effects of the project on fisheries and water quality. But even more fundamental are criticisms that the Corps failed to justify the need for additional water supplies or to address the problems that stem from an antiquated, leaky, and undersized distribution system.

The need for a 30 percent increase in supply capacity has been challenged on several grounds. The demand projections assume water use will increase roughly in step with population, and significant water savings are anticipated by industrial users. But as long as residential water use (which currently consumes 75 percent of the total) is unmetered, most consumers have little or no incentive to conserve. On balance, the Corps assumes that the effect of economic growth and of conservation measures will offset one another, keeping per capita use roughly constant at about 220 gallons per day.

Without meters, the city is unable to determine even how much water is being lost in the distribution system or to use price to influence use. By the Corps Engineers' own estimates, universal metering would curb use by 50 mgd, use of domestic conservation devices (restrictive shower heads, for example) would save 85 mgd, and leakage controls would save another 25 mgd. Combined, these savings would close more than half the projected peak gap between use and safe yield, at estimated cost of about 5 percent of the Hudson River project. These are some of the considerations underlying the proposal set out by New York City's mayor, Edward Koch, in January 1986, to require all residential buildings to install water meters by 1995. The proposal, which must be approved by the state and the city, does not require submetering of individual apartments. Consequently, even with its adoption, many households still will have little incentive to conserve on water use.

The Corps, moreover, dismisses pricing as a desirable tool to influence demand in an area of high population density and many multiple-unit dwellings. It argues that price would have to be increased in multiples of 100 percent to affect use significantly; in its view, equity, or concern for fairness, rules out such large price hikes.

Although municipal water demand is relatively insensitive to price, higher prices encourage conservation, especially where use previously has been unmetered. If consumers are expected to pay the costs associated with new supplies, the multibillion-dollar Hudson River project would imply sizable increases in residential water bills. If universal metering also were introduced to enable bills to reflect consumption, the higher price of water probably would dampen aggregate use significantly.

The Corps treats its projections of future use as requirements to be met under foreseeable conditions and regardless of cost; its planning essentially attaches infinite cost to system shortages. Recent droughts suggest, however, that shortages are not unmitigated disasters. To be sure, shortages have their costs, but they are not unbearably large when compared with the costs that would be incurred in constructing the Hudson River project.

The most serious threat to New York City's water supplies comes from an old, undersized distribution system. Prone to breakdowns, it is plagued by an average of nearly two water-main failures every day. Millions of dollars are spent each year for repairs, and disruptions in service impose additional social and economic costs on residents and businesses.

Two main tunnels link the city to its eighteen reservoirs in the Croton, Catskill, and Delaware watersheds. Fortunately, neither has failed, but if one were to do so before a third major tunnel under construction is completed in about 1995, New York truly would have a water crisis. Of concern since at least the 1950s, the two tunnels have been in continuous service since 1917 and 1936; on peak days they carry flows that are 60 percent above the levels for which they were designed. They never have been repaired or even inspected, and there is no way to do so, until the third tunnel is completed, without leaving as much as half the city without water.

In New York City, as in many parts of the country, good-quality water is becoming increasingly scarce relative to demand. As a result, it is becoming ever more costly, futile, and inefficient to maintain an ethic based on abundant, cheap water. New York retains vestiges of this ethic: in 95 percent of the residential units water is not metered; the average family pays less than $100 a year for supposedly unlimited supplies of high-quality water; per capita consumption is well above the national average. And this in the face of unmistakable signs that water is not abundant and cheap. Two drought emergencies and thousands of breakdowns in the city's water mains over the past five years illustrate its vulnerability to the weather and to an aging distribution system. New York has embarked on a $5 billion project—the third tunnel—to improve distribution, and the Corps is recommending an even more expensive project to increase supplies. At what point will New Yorkers' expectations of unrestricted access to water cost too much to fulfill?

Why not efficiency?

Implications of an efficient city water strategy run counter to the traditional, supply-oriented approach of the Corps of Engineers. For instance, the losses associated with temporary and infrequent restrictions on use surely are less than the costs of expanding supplies to meet all contingencies. Planners, therefore, should expect to need water-use restrictions during extreme weather. Also, it would seem prudent to give at least short-term priority to improving the distribution system rather than seeking large increases in supplies. To guide them in making their decisions, planners should attempt to minimize the sum of the capital costs for expanding the system's supply and distribution capabilities and the expected losses resulting from drought and failure of the distribution network.

Finally, assuming that water costs cannot be shifted to the state or nation as a whole, the city would benefit from a strategy that relied more on prices than on restrictions to bring water use in line with supply, especially during nondrought periods. Higher prices would encourage economically sound conservation investments and practices and reduce the expense of administering the system. Moreover, the increased revenues could help finance the enormous investments required to improve the system.

Unlike, say, the Southwest, the natural endowment of water is not the principal problem for New York City. Nevertheless, water is now a scarce resource in the region, and it will be extremely costly and inefficient to construct and manage a system capable of preserving an outdated water ethic even under extreme weather conditions.