In the late 1990s, California came under increasing fire from the six other states that share the waters of the Colorado River for taking more than its 4.4-million acre-foot allocation of the river. The river dispute made the ag card an even more important piece of Met's portfolio. Palo Verde's farmers watched as several agencies negotiated a complex set of deals to get water from the nearby Imperial Irrigation District, and they sensed that it wouldn't be long before they heard from Met once more themselves. "Met kept going to all these meetings and giving these PowerPoint presentations, and it always included 100,000 acre-feet from Palo Verde," says Smith. "They never bothered to tell us that, but we knew they were gonna come after it." Palo Verde was particularly attractive to Met because it had the oldest recognized rights on the entire Colorado River - which, in the first-in-time, first-in-right hierarchy of water law, also makes them the most dependable. 

By that point, Palo Verde's farmers had resigned themselves to the inevitability of Met's presence in their lives. "We know there's 18 million people over there, and 2,700 of us here. They're gonna come get some water," Smith says. "And we want to make a deal that's best for us." 

In late 2000, one of the irrigation district's board members called Smith and instructed him to drive out to the Blythe airport to pick up a couple of representatives from Metropolitan. The two sides began negotiating an agreement that would take water transfers to a higher order of sophistication. As in the earlier deal, farmers in the irrigation district would generate water for Met by fallowing their lands. Unlike the previous arrangement, however, this one would stand in place for 35 years. 

"Both sides spent a lot of time trying to look for bogeymen," Smith says. "If you're going to be in this 35 years, you have to think about a lot of things." 

Simply taking land out of production for that length of time would have been tantamount to drying it up permanently. But by crafting a "rotating" fallowing program, in which any given field would be idled for no more than five years in a row, the new deal makes the impacts temporary and distributes them throughout the district. 

The negotiations were not without their rough spots. Met, for instance, initially insisted on approving the yearly plan that each participating farmer submits to his bank for loans. "That was an absolute deal-killer," Smith says. "We will guarantee that if you tell us to fallow some ground, we will fallow some ground. The rest of that ground, you have nothing to do with it. It's none of your damn business." 

Finally, after nearly three years of negotiations and a set of meticulously written contractual protections, the Palo Verde Irrigation District agreed to be drawn within Met's orbit. Farmers idle between 7 to 29 percent of the land in the district each year. In exchange, they received a sign-up bonus of $3,170 per acre, plus about $630 for each acre they fallow in any given year. (In each subsequent year of the deal, that amount will increase to cover inflation.) 

The fallowing program allows Metropolitan to draw a steady "base load" of at least 25,000 acre-feet from the district each year. During dry years, Met can call for a maxed-out "peak load" of 111,000 acre-feet. For the valley's farmers, the deal allows them to continue farming the majority of their acreage during normal years and compensates them for income lost from any fallowed acreage. 

When Metropolitan's managers talk about the agency's diversified water-supply portfolio, they tend to emphasize the interplay between Met's water-conservation programs and its massive expansion of storage projects. Per capita water consumption in Metropolitan's service area has been decreasing since 1990, while Met has expanded its water storage capacity by a factor of 10 over the same period. The agency can stash whatever water its efficiency efforts save during wet years. 

But wet years are increasingly rare, and farm-water transfers play a critical role in the total mix; during droughts, they're the one reliable way Met can "backfill" shortfalls and make good on its pledge of 100 percent reliability. To ensure access to that water, Quinn - who was rising through Met's ranks on his way to becoming deputy general manager - began working to establish relationships with irrigation districts in other parts of the state. 

"It wasn't like there was a long line of people wanting to talk to us," he concedes. "My nose got broken 86 times with agricultural district doors slamming shut on it, saying, 'What part of "no" don't you understand?' " 

But Quinn learned. Eventually, he began touting the ways that Met could use its considerable financial resources to help farmers hedge their own risk, by providing them a supplemental source of income that was more dependable than variable crop prices. 

Quinn also got better at sniffing out people like Van Tenney, who ran the Glenn-Colusa Irrigation District, more than 400 miles north of Los Angeles. The district supplies water to rice farmers near Sutter Buttes, the eerily beautiful carcass of an extinct volcano that sits smack-dab in the center of the Sacramento Valley. In spring and fall, the sky is alive with ducks and geese winging their way in and out of several wildlife refuges in the area. 

In 2001, Glenn-Colusa banded with other local irrigation districts to sell water to the Westlands Water District, an agricultural powerhouse in the San Joaquin Valley, south of the Delta. "That was a pretty major deal," says Tenney, "and I think Metropolitan was watching." 

Still, in the years after Boronkay left Metropolitan in 1993, Met's top brass did little to help Quinn's cause. In the late '90s, Boronkay's successor, John "Woody" Wodraska - who, not long afterward, would go on to work for Enron's water subsidiary - delivered an incendiary speech to the area's rice farmers. "He basically came up and told the audience, 'If you guys don't cooperate with Metropolitan, we're gonna take your water,' " says Tenney. "It was the most in-your-face thing I've ever seen." 

Wodraska's departure from Met, and a series of conversations with Quinn, finally convinced Tenney that the agency was, as he puts it, "going through a fairly significant attitudinal change." He agreed to talk. 

Quinn, meanwhile, had been thinking about how to move beyond the simple first-order strategy of portfolio diversification as a way to manage risk. He set out to hedge two of Metropolitan's biggest risks - running short of water, and spending heavily on backup supplies that it might not need. 

"The really bad years don't happen that often," Quinn says. "We had enough water most of the time." Metropolitan might only need to backfill its supplies one year in five. "I wanted to find some sort of mechanism that would solve that one-in-five-year problem," Quinn says - a way that Met could ensure itself access to backup water, but only in the worst drought years. 

As a graduate student at UCLA, Quinn was interested in stock-options trading. Electrical utilities had adapted that tool to put together portfolios that supplemented their "firm" supplies - such as power plants that they themselves owned and operated - with options to buy power from other producers on an as-needed basis. Quinn borrowed that idea and tailored it to the rusticated world of Western water. 

In the winter of 2002, Quinn arranged a series of meetings with farmers in Glenn-Colusa and 10 other Sacramento Valley irrigation districts and began negotiating one-year options contracts for part of the farmers' water. Initially, says Tenney, "Probably half my farmers would've strung me up. (Shipping) that water south for money was just viewed as a terrible moral thing to do." But for the farmers, the options deals opened up a wider array of - what better way to put it? - options than just growing a crop with their water. 

Under the terms of the contract, Met paid the irrigation districts $10 an acre-foot for the option to buy sometime before the following March. If Met called the option, it would pay an additional $90 per acre-foot to buy the water. The irrigation districts would then "float" it down the Sacramento River and into the Delta; State Water Project pumps at the Delta's southern edge would transfer the water into the California Aqueduct, which would then carry it to Southern California. 

If Metropolitan didn't need the water, it wouldn't exercise the option. Farmers would pocket the option fee and then use the water on their farms. For Met, the arrangement was attractive because the agency could keep extra water on-call, but pay just a fraction of what it would have cost to buy the water outright and run the risk of not actually needing it. 

It was an elegant idea - on paper. All told, Met signed options for more than 146,000 acre-feet of water. The winter was dry, and that spring, Met called the options. Almost immediately afterward, the heavens opened, and it rained like hell. "April was off the charts," says Quinn. "April and May combined were one of the biggest flows in history." 

Because so much water had fallen on Northern California, the State Water Project pumps were running day and night, sucking water into the California Aqueduct. "The pumps were so busy that they didn't have any room for option water," Quinn says. And, in a sort of dejavu, $8.3 million worth of water floated down the Sacramento River and out to sea. "We watched tens of thousands of acre-feet of water go out to the Golden Gate Bridge," says Quinn, "because we had to call it too early."