Even Warren Buffett has been bamboozled by oil.

He admitted it in his latest annual report to the shareholders of Berkshire Hathaway -- the holding company he runs. In his own words: "I bought a large amount of ConocoPhillips stock when oil and gas prices were near their peak. I in no way anticipated the dramatic fall in energy prices that occurred in the last half of the year."

Specifically, he made the bulk of his purchases during the six months ending Sept. 30, 2008 -- you know, the same time in which oil prices peaked near $150 a barrel.

Despite a recent run-up, the price of oil is currently half that peak, and ConocoPhillips' stock price has tanked in lockstep with the oil free fall. Buffett clearly bought oil too early. But is it still too early for us to buy oil stocks now?

Now may be the time
Those bullish on oil point to the inevitability of "peak oil," arguing that the time will come when we hit the peak of global oil production. From that point on, we'll be able to pump less and less oil out of the ground. In economic terms, we'll face decreasing supply.

Meanwhile, bulls argue that demand will increase greatly, as China and other emerging markets fuel their economic growth with oil. On average, each person in the U.S. consumes about 25 barrels of oil a year; each person in China consumes just more than two. That's a lot of possible future demand.

And all of us amateur economists know what happens when you restrict supply while simultaneously increasing demand: Prices rise.

But then again ...
Um, weren't these the same arguments made when oil was at $147 a barrel? Yup. At that price, all of these favorable supply and-demand assumptions were baked in, and then some. The subsequent price fall highlights that we'll only make great returns if we buy at low prices.

With oil prices at half of their year-ago highs, oil plays are certainly tempting now. Getting in at steep discounts to the prices Buffett paid is a wonderful thing. However, when we look back in time, we see that current oil prices are about seven times the lows of the late 1990s.

In other words, looking at price movements by themselves just isn't that helpful. We need to estimate oil's intrinsic value.

How do we do that?
Beyond bubbles and busts, oil should sell at its marginal cost of production, plus some profit. Unfortunately, that's not easy to calculate with much precision. Some oil sources are really easy to find and extract (traditional onshore) while others are especially onerous (for example, oil sands and deepwater).

Then there's the Achilles' heel of oil: alternative fuels and the vehicles they power. Just as the solar technology made by JA Solar and Trina Solar becomes more attractive when fossil-fuel prices rise, high oil prices increase demand for hybrids, electrics, and hydrogen-cell cars from Ford (NYSE: F), Honda (NYSE: HMC), Toyota, and the rest. The development of these sorts of substitutes for the fuels can act as a price ceiling for oil -- affecting all of the oil players, from the titans like ExxonMobil to the merely large like Transocean (NYSE: RIG) to the small ones like Dawson Geophysical (Nasdaq: DWSN).

Thus, I view the promise of alternative energy as a long-term capping mechanism on runaway oil prices.

OK, so is oil a buy?
The question boils down once again to supply and demand. If peak oil is a ways off, demand slackens, and alternative-energy options evolve quickly, a high oil price isn't justified. But if our oil supplies become constrained, the world greatly increases its energy lust, and alternative-energy players hit snags, it's off to the races.

Here's an additional data point to keep in mind. After admitting his timing error on ConocoPhillips, Buffett went on to say, "I still believe the odds are good that oil sells far higher in the future than the current $40-$50 price." Before we dismiss his opinion because of his poor judgment on ConocoPhillips, let's remember his investment in PetroChina.

In 2007, he sold shares he'd bought just five years before, for more than a 700% gain! Since then, PetroChina's stock price has plummeted.

Buffett's optimism is certainly encouraging. But regardless of the supply and-demand outlook, I think some exposure to oil companies makes sense as an insurance policy. When the price of oil rises, most companies -- from manufacturers such as Boeing (NYSE: BA) to retailers such as Home Depot (NYSE: HD) to shippers such as FedEx (NYSE: FDX) -- suffer from higher input costs and slackening demand. An investor's best defense lies in owning stock in the oil companies that stand to benefit.

In the near term, our dependence on oil isn't going anywhere, and the general trend of rising marginal costs of production provides a cushion for oil prices. That was certainly true in Buffett's $40-to-$50 oil range. The argument to load up on oil stocks is less compelling in the $60s and $70s, but the argument for a reasonable oil exposure as a hedge against skyrocketing energy prices still holds.

Our Global Gains newsletter service recommends a number of oil plays, including one from China. (No, it's not PetroChina.) If you're looking for oil ideas, or are just interested in learning about some intriguing international investments, I invite you to enjoy a 30-day free trial by clicking here. You can read up on all of the team's recommendations with no obligation to subscribe.

Already subscribe to Global Gains? Log in at the top of this page.

This article was originally published April 3, 2009. It has been updated.

Anand Chokkavelu owns shares of Berkshire Hathaway and Toyota. Berkshire Hathaway and FedEx are Motley Fool Stock Advisor selections. Berkshire Hathaway and The Home Depot are Motley Fool Inside Value picks. Dawson Geophysical is a Motley Fool Hidden Gems selection. The Fool owns shares of Berkshire Hathaway and has a disclosure policy.