If you haven't got the message that something disturbing is happening in the oil world, stop by my office. On my desk, I have a pile of books a foot high with titles like Out of Gas, The End of Oil, and Twilight in the Desert. The authors range from geologists to journalists to policy wonks, and they all tell the same story.

For years, oil industry executives dismissed fears of an energy crisis, attributing rising gasoline prices to unrest in the Middle East, Wall Street speculation, and temporary interruptions in supply. But recently, as the price of crude has bounced around $100 a barrel, even some establishment figures have been making alarmist noises. The Paris-based International Energy Agency warned of a possible "supply crunch" within five years. Its chief economist, Fatih Birol, said prices could reach such a high level that "the wheels may fall off" the global economy. In the U.S., the National Petroleum Council, a federal advisory group, said that as the economies of China and India continue to expand, global energy consumption will rise by 50 percent over the coming quarter of a century. "There is no quick fix," said Lee Raymond, former chairman of Exxon Mobil, who leads the council.

Perhaps not. But the experts who are predicting the worst, based on geology and geopolitics, are missing the crucial role that economic incentives play in determining the price of crude. The tripling of oil prices since the summer of 2003 has unleashed forces that within the next two or three years will bring oil prices tumbling back down to below $50 a barrel. Looking even further ahead, prices could easily fall to $30 a barrel or even lower. So before you trade in your Cadillac Escalade for a Toyota Prius, think twice: $1.50-a-gallon gas might not be gone forever.

The key to understanding where prices are headed is distinguishing between the short run and the long run. In a time frame of anything shorter than five years, the supply of crude is more or less fixed. Drilling for oil is an arduous and unpredictable process. Even after a new hydrocarbon reservoir is discovered, ramping up output takes years. Current production capacities reflect investment decisions made in the late 1990s or earlier.

Today, OPEC has the ability to produce about 35 million barrels of crude a day; the rest of the world can produce perhaps 50 million barrels a day. As recently as 2003, this seemed like plenty. Since then, though, global demand has grown rapidly, and a series of catastrophes—some natural (hurricanes Rita and Katrina), some man-made (war in Iraq and unrest in Nigeria and Venezuela)—have curtailed production, causing supply to dip below demand. In September, the global demand for crude reached 85.9 million barrels a day, whereas global supply was just 85.1 million barrels a day, according to I.E.A. figures.

When shortages emerge in any market, prices spike. If the imbalance is expected to continue, speculators move in and drive prices even higher. Oil is no exception. In the fall, as crude inventories declined and the rhetorical battle between the U.S. and Iran escalated, trading volume shot up.

With prices close to the inflation-adjusted record, energy companies and governments are investing heavily in facilities that generate crude and crude substitutes. Consumers of fuel oil and gasoline are starting to economize, and over time, these changes in behavior will shift the balance of power in their favor. When that happens, an oil glut will emerge, and the price will plummet.

Already, in Texas and California, hundreds of mothballed, low-producing stripper wells have been brought back into production. In Africa, the Chinese government is making development deals with Sudan, Chad, the Congo Republic, and other impoverished nations with unexploited reserves. In the Canadian province of Alberta, Shell and other energy companies are building massive strip mines to access local tar sands, which can be converted into synthetic oil or refined directly into petroleum at a cost of roughly $30 a barrel. Some experts believe the sands contain more oil than the subdeserts of Saudi Arabia.

Not very long ago, energy companies were slashing their exploration and drilling budgets, refusing to finance any project unless it could generate crude for $15 or $20 a barrel. But since 2003, when the price of crude rose above $30 a barrel, the industry has relaxed its financial assumptions and beefed up capital spending. In the past four years, Exxon Mobil, the world's largest oil company, has invested more than $60 billion in exploration and development. Between now and 2010, the company plans to begin pumping oil or gas from no fewer than 20 new projects.

Besides Canada, the oil majors are also returning to areas that weren't economically viable when oil was cheap, including the Arctic Ocean and the deep waters of the Gulf of Mexico. The industry's efforts aren't confined to searching for new reserves. It is also investing heavily in high-tech imaging machines and steerable drills that raise yields from existing reservoirs, where historically only the most readily available crude, typically 30 to 40 percent of the total, was recovered. (Extracting the rest was considered too costly, so it was left alone.)

When experts claim that oil is running out, what they really mean is that cheap oil is running out. About this, they may be right. Outside of Saudi Arabia, Iraq, and a few other countries, it is no longer possible to recover large quantities of crude for a dollar or two a barrel. But there are plenty of places where oil can be produced for $20 or $30 a barrel, let alone the $100 range where it has been trading recently.

And the list of potential substitutes for crude is long. Natural gas can be converted to a liquid fuel that produces few pollutants. Venezuela has big reserves of tar sands, as does Utah. Neighboring Colorado has oil trapped in shale, which industry engineers are trying to extract by slowly heating the rock under the Green River Basin. Corn, sugar, and potatoes can be distilled into ethanol, a perfectly good transport fuel, as can wood chips, straw, and other biomass. And as demand for ethanol has surged in recent years, farmers throughout the Midwest have taken advantage of generous federal subsidies to convert their fields to corn, the price of which doubled in the past 18 months. (When oil prices fall, such crop switching may prove to be a costly mistake.)

With energy supplies expanding and the demand for oil showing signs of faltering, it won't be very long before economic fundamentals reassert themselves. If oil were a normal commodity, competition would eventually drive the price down to a level close to the current cost of production, which at the margin is probably somewhere between $20 and $30 a barrel.

Of course, the oil market is hardly a textbook case of open competition: The OPEC cartel controls 40 percent of the supply, and geopolitics is an ever-present factor, as is speculation. The recent surge toward $100 a barrel was a dramatic demonstration of how traders can cause prices to become unmoored from costs for a lengthy period. But that also means that once market sentiment turns, the fall in prices could be just as dramatic.

Nobody in the oil market—not Wall Street, not Exxon Mobil, not even OPEC—can sustain prohibitively high prices for very long, a point that Sheik Yamani, the Saudi oil minister during the oil price shocks of the '70s and '80s, recognized. "If we force Western governments to invest heavily in finding alternative sources of energy, they will," he said in 1981, shortly after OPEC production cuts caused the price of crude to hit a record of $39.50 a barrel—roughly $100 a barrel in 2007 dollars. "This will take them no more than seven to 10 years and will result in their reduced dependence on oil as a source of energy to a point which will jeopardize Saudi Arabia's interests."

Most people ignored Yamani's warning, but he was right. Between 1979 and 1983, oil consumption in the non-Communist world fell by 6 billion barrels a day, or more than 10 percent. Motorists bought smaller cars. Homeowners threw out their oil furnaces. Power stations switched to coal, nuclear fuel, and natural gas. And this all happened at a time when new oil fields in Alaska, Mexico, and the North Sea were coming onstream in a big way. The result was an excess supply of crude and a huge drop in prices. In 1986, the cost of a barrel of crude fell to as low as $11.

The oil industry entered a prolonged slump, devastating Texas and other producing areas. For most of the '90s, the cost of a barrel of crude stayed below $20. At the end of 1988 and the start of 1989, it fell below $10, and you could get change out of a dollar for a gallon of gas.

I'm not saying that the oil price will slink all the way back to $10 a barrel. But a reckoning is inevitable. Serious divisions are emerging within OPEC about 2008 production levels. Presidential candidates in the U.S. are calling for tougher fuel-economy standards. Many Western countries, the U.S. and Britain included, have been making plans for a new generation of nuclear power plants. In the oil market, the laws of supply and demand sometimes appear to have been suspended. Ultimately, however, they do work.