by John Fleck
With that in mind, Carpenter steered the 1922 Colorado River Compact toward an agreement that would hold the line against California while preserving water for later use by other states. The Compact Commission divided the river basin in half, drawing a line at Lee’s Ferry, at the upper end of the Grand Canyon.5 The states above that—Wyoming, Utah, Colorado, and New Mexico—would get half the water. The states below Lee’s Ferry—Arizona, Nevada, and California—would get the other half. The upper states were growing far more slowly, but the scheme was intended to ensure that, when they finally needed it, there would still be some water left.
The agreement contained two provisions that would haunt water managers by the 1990s. The compact specified amounts of water rather than percentages—the Upper Basin would get 7.5 million acre-feet of water per year, and the Lower Basin would get 7.5 million acre-feet. The pact’s authors failed to write clear rules about what would happen if the river didn’t have that much water in it. Second, they built in a mechanism that ensured that unused water in the system could benefit downstream states—particularly California.
For nearly a century, the other states weren’t using their full share, allowing extra water to slosh in California’s direction. And just as Carpenter had feared, California had come to depend on it. By the late 1980s, California, which had an official allocation of 4.4 million acre-feet per year, routinely used more than 5 million, thanks to the other states’ unused allotment. The bonus water went to the Metropolitan Water District of Southern California (“Met”), which served the growing cities on the Los Angeles–San Diego coastal plain. The other states, especially those of the Upper Colorado River Basin, were concerned “that California’s growing dependence on surplus water would one day ripen into a legal entitlement,” said Colorado water lawyer Jim Lochhead.6
Simon Rifkind, retained by the US Supreme Court to try to untangle the Colorado River mess, recognized as early as 1960 that the Metropolitan Water District had come to depend on such a surplus, but said there was no reason to worry because the surplus would be around for a very long time. “I am morally certain that neither in my lifetime, nor in your lifetime, nor the lifetime of your children and great-grandchildren, will there be an inadequate supply of water for the Metropolitan Project,” Rifkind told lawyers during a 1960 hearing. “I am morally certain, as certain as I am of the multiplication table, that not within the span of the ages indicated there will be any diminution either in the present uses of the Metropolitan Aqueduct or its contemplated expansion.”7
By the early 1990s, it was becoming clear Rifkind’s multiplication table had come up short.
In response, the states were pushing the Bureau of Reclamation to do what they believed was the agency’s job: forcing California to begin living within its means. “Appropriate enforcement,” Utah’s Larry Anderson said, “is critical to protecting our allocations under the Law of the River.”8
“A New Era of Limits”
The easy comfort provided by surplus on the river began to slip away in 1990, when Southern California needed its usual dose of extra water, above and beyond its minimum entitlement, to tide it over. But the other basin states had had enough.
1990 was the driest year in nearly two decades in coastal Southern California, an exclamation point on a drought that had lingered, with minor breaks, for most of the 1980s.9 Southern California usually used the Colorado River’s distant watersheds as a hedge against local drought, but in 1990 the Colorado River’s ability to play that role was shrinking. The first half of the 1980s had been the wettest years the Colorado River Basin had seen, leaving the reservoirs full. But beginning in 1988, nature turned off the tap, and the three years ending in 1990 were the driest to date in the Colorado River Basin.10
In response to local drought conditions, California government agencies took steps that would once have been unthinkable. Santa Monica developers had to pay for low-flow toilets in existing homes to conserve enough water to supply any new homes they built. Elsewhere in the basin, water districts refused entirely to hook up new customers, and the Metropolitan Water District, once known for its expansionist tendencies, stopped annexing new territory.11 And California again turned to the Colorado River’s federal managers, looking for an allocation of surplus water.
The basic dilemma will be familiar to anyone with a bank account: how much to spend now and how much to save for later? But the consequences were profound. The reservoirs in the basin serve as a massive water savings account. The rules for how much water to move, where, when, and for whom, were byzantine, but the underlying questions were simple: how much should be sent downstream for LA’s lawns, swimming pools, and toilets now, and how much should be saved as a hedge against prolonged drought? Any answer to that question involved trade-offs, with someone benefiting by being able to use the water now and someone suffering the risk if the reservoirs were empty if and when the dry days arrived.
A “use it now” approach would benefit California, providing the surplus needed to keep the Colorado River Aqueduct to Los Angeles and San Diego running full. But the other states argued that this would increase the basin’s vulnerability to the risks posed by a future run of dry years like those the basin had seen from the 1930s to the 1950s. Such a multiyear drought would leave the reservoirs unable to deliver the minimum legal allotment to all users. If that happened, Arizona and Nevada would be the most vulnerable, because their shares would be the first to be cut in a true shortage. California would benefit from a surplus declaration while the risk would be borne by Arizona, Nevada, and the other upstream states.
The other six Colorado River Basin states were unwilling to accept the resulting risk, and vehemently disagreed with California’s request for a surplus declaration. In response, the Bureau of Reclamation set down an important marker for how the “drain or save” argument was to be resolved. In 1990, the agency’s staff concluded they could find “no clear basis in the existing legal and institutional framework of the Colorado River” to give California surplus water “without the consensus of all seven Basin States.”12 All for one, and one for all. California could only get the extra water if the other states agreed.
It was a critical move by the bureau that would shape the debates that followed, and remains as an underlying principle for Colorado River management today. The bureau was wagging its finger at the states and saying, in essence, “It’s up to you to come up with a way to solve this. Don’t expect the federal government to take sides in your disputes over how to allocate the river’s water.”
Everyone understood the risks of going to court. Water managers hate uncertainty. Litigation is uncertainty writ large, and the Arizona v. California battle over Colorado River water allocation had created a decade of high-risk uncertainty. Where the give-and-take of negotiated solutions can add flexibility and adaptive capacity, litigation is suited to settling narrow questions or, as in West Basin, questions on which the parties all agree and the courts are merely a tool to implement a collaborative solution. As a tool to settle a conflict, litigation tends to constrain future water operations. Lesson learned.
The basic framework of a deal was clear from the beginning, but it took more than a decade to work out the details. Nature helped out, with enough wet years to help refill the reservoirs, reducing the problem’s urgency, and the negotiations dragged on.
Colorado governor Roy Romer sketched out the rough terms in an August 1991 letter to California governor Pete Wilson. Romer recognized that California had a real and immediate problem: hammered by drought, California needed its surplus water in the short term to serve the parched cities of Southern California. But Romer’s letter also recognized that, in the long run, the rest of the basin states needed some assurance that California would eventually learn to live within its means. California had to agree to a concrete and enforceable plan that would bring its Colorado River water use from more than 5 million acre-feet to its allotted 4.4 million acre-feet per year. The deal was to be quantified in
very specific rules governing how the secretary of interior was to declare the “surplus” each year that would give California its extra water. No longer would basin water managers face the uncertainty they faced in the summer of 1990 over how much water would be available to each.
Essentially, Romer offered California what came to be called a “soft landing.” “We want to help Southern California with its drought situation. But it also is in our self-interest to get California to live within its entitlements in the river,” Romer said.13
Bureau of Reclamation officials signaled what the discussion was really all about in a presentation to one of the early meetings entitled “Managing the Colorado River in the Lower Basin in an Era of Limits.” It was a recognition that Rifkind had been wrong, and that the surpluses had run out.14
California Struggles to Comply
The approach embodied a core principle of Law of the River. The Colorado River Compact, the Upper Basin Compact, and statutes approved by Congress set each state’s overall water entitlement. It is then up to each state to figure out how to divide the water among its users. California’s case was complicated by the federal government’s role in administering contracts for the allocation and distribution of that state’s Colorado River water. But as a practical matter, no one could come in from the outside and tell California how to get to 4.4 million acre-feet.
This is the most difficult part of making Colorado River water management work. Any agreement to deal with the basin’s overall water problems inevitably must be implemented at the local level, one irrigation district, city council, and water user at a time. That is because people at the most local level are the ones actually using the water: a farmer deciding which crop to plant this year, a resident turning on the tap. State representatives may negotiate a deal that works at the basin scale, but if they can’t then make it work back home, the whole thing falls apart.
At the basin scale, state leaders had agreed that California would have to live within its 4.4 million acre-feet per year allocation. In order for the other states and the federal government to give California time to get there—the “soft landing”—California had to come up with an acceptable plan to do that, with binding commitments and enforceable penalties if they didn’t. They called the unspecified solution “The 4.4 Plan” or, more elaborately, the “Quantification Settlement Agreement,” and the task of building a workable proposal looked intractable.
California’s difficulties embodied the tension between the farm country that the Colorado River’s developers originally imagined and the urban nation we have become. As the state’s water managers tried to figure out how to live within its means, Californians had to contend with longstanding water allocations that gave most of the river’s water to desert farms. America plumbed the Colorado River to turn desert into farmland, both out of a philosophical desire to embody the Jeffersonian ideal of the yeoman farmer, and a practical need to feed itself. A system designed to spread water across vast acres of farmland was having a hard time making the transition to the post–World War II era of growing cities. Repurposing the physical plumbing was a manageable problem, but the institutional plumbing needed to reallocate the water was another matter. The rules were all wrong, and the politics made it difficult to change them.
In 1931, the Southern California agencies that used the most Colorado River—the Palo Verde Irrigation District, the Imperial Irrigation District, the Coachella Valley Water District, the Metropolitan Water District of Southern California, the City of Los Angeles, the City of San Diego, and the County of San Diego—signed an agreement that gave farmers first dibs on the bulk of the water—87.5 percent—released from Hoover Dam.15 The Metropolitan Water District was entitled to the other 12.5 percent. It was a recognition of the doctrine of prior appropriation: the farm districts had a legitimate “first in time, first in right” claim. It was also consistent with the reality of water in that day and age. Whether measured by volume or cultural importance, water’s primary use was on the farm.
But the deal included a safety valve that, by the 1990s, was coming back to haunt Southern California. Beyond the initial allocation of 4.4 million acre-feet, the so-called Seven-Party Agreement said that if there was a surplus, cities could use it. That surplus had been keeping Metropolitan happy all these years, enabling Los Angeles and the other cities of coastal Southern California to grow. Met routinely took double its basic allotment. But by 1990, as urban demands grew and pressure rose from the rest of the Colorado River–using states, the scheme’s flaws were becoming increasingly apparent. It was not just California water users in general who were vulnerable if the state was cut back to 4.4 million acre-feet. It was one class of water user in particular—the cities. In hindsight, it is easy to say simply that Southern California should not have allowed so much growth, but by 1990, such hindsight did not help much in dealing with the facts on the ground. Changing the allocation would require the cooperation of the big farm districts that controlled the bulk of the water, and their political supporters. And therein lay the problem.
Since at least the 1980s, Californians had been fighting over whether the Imperial Irrigation District was wasting water.16 The battles had ended, unproductively, in court. So the agencies turned instead to voluntary deals under which the cities would pay for agricultural water-efficiency improvements in return for a share of the saved water. On paper, this should have led to equitable arrangements. Economists love this sort of simple “efficiency” argument: let the markets work and the problem will take care of itself. But as we have seen, simply taking farmland out of production eats away an agricultural community, creating significant political opposition. And the law tends to give those farm communities the power to block change if they don’t like the terms of the deal.17
Some low-hanging fruit was obvious. Earlier agreements had lined previously unlined irrigation canals that carried water from the Colorado River through the desert to the farmlands of the Imperial and Coachella valleys. This required no farmland to be taken out of production. But that only saved 2 percent of California’s Colorado River water use. A broader agreement, which would require more farm water to be moved to city use, proved elusive. Within California, things were a mess.
The two biggest farm districts, the Imperial Irrigation District and the Coachella Valley Water District, couldn’t agree on how to divide up their shares of the agricultural water. The two biggest municipal agencies, the Metropolitan Water District of Southern California and the San Diego County Water Authority, couldn’t agree on terms for the use of Met’s canals to move water from desert farms to urban users in the San Diego area. They couldn’t settle on a plan to manage environmental impacts from the deal. And no one could agree on how to save enough water overall to make the deal work.18
By 1997, a frustrated interior secretary Bruce Babbitt issued a warning: if Southern California water agencies didn’t get their act together and come to a mutually acceptable deal, he would be forced to step in and slash water allocations for them.19
Prodded by the threat, California’s water agencies finally worked out a framework for a deal. San Diego would fund a slowly expanding conservation effort in the Imperial Irrigation District, with the saved water moving to the cities. San Diego and Met reached a deal on the use of Met’s canals to move the saved water. The Coachella and Imperial water districts settled their dispute over who was entitled to how much of the desert’s agricultural water supplies.
In the basin, there was a bureaucratic sigh of relief. It seemed that the water war had been averted. In 2001, Babbitt, in the last public event of his eight-year tenure, stood at a San Diego ceremony beneath a banner that said “Peace on the River.” The deal announced that day formalized the “interim surplus guidelines,” rules striking a balance between releasing enough water from the Colorado River’s reservoirs to give California a “soft landing” and holding enough water back in savings to protect the other six basin states from the risk of future shortage. But th
ere was a caveat. California’s deal, while close, was not quite done, so the basin-wide agreement included a firm commitment from California to finalize its 4.4 Plan by December 31, 2002, along with a threat: if California failed to come to terms with its own problems, there would be no soft landing.
Again, the final California negotiations lagged, and the other states worried. Would the federal government have the nerve to wield the enforcement hammer come January 1, 2003? With the clock ticking, California’s negotiators came to what they thought was a deal on the terms of the Quantification Settlement Agreement with less then two months to spare. Bob Hertzberg, one of the senior statesmen of California politics and the man brought in to broker the deal, again trotted out the rhetoric of “lasting peace on the river.”20
But without the approval of the board of the Imperial Irrigation District, the Colorado River’s largest farm-water agency, Hertzberg’s deal would collapse. All eyes in the basin focused on an emotional meeting of the Imperial Irrigation District board of directors on December 9, 2002. The farm district was under enormous pressure, and there were fears that if they did not agree to a deal that included compensation for their water, it would simply be taken away.21
Gustave Aguirre of the United Farm Workers urged the IID board to approve the deal only if there was an added $70 million for economic transition assistance for the chronically poor communities of the Imperial Valley. Antonio Ramos of Calexico dramatically held up a milk jug with little water in it and recalled the other infamous agricultural-to-urban transfer in California that dried up basins in the eastern Sierra Nevada in order to supply water to Los Angeles: “If it goes through we will have no water and the valley will become another Owens Valley—dead.”22
In the end, the board voted three to two against the deal.
Federal officials wasted no time in responding. Assistant Secretary of the Interior Bennett Raley wrote a letter to the California delegation saying that the failure to formally quantify how much water would be shifted from the Imperial Valley to the coastal cities meant California had failed to meet the terms required to earn the soft landing they’d been hoping for. In what is surely the most literary moment in the history of the Law of the River, Raley referenced F. Scott Fitzgerald: “The Department has no interest in a (Quantification Settlement Agreement) that does not represent a long-term Quantification of the parties’ portion of California’s apportionment of Colorado River water, lest in fifteen years we find ourselves as Gatsby did—‘So we beat on, boats against the current, borne back ceaselessly into the past.’”23
