Everyone is warning about bubbles. These warnings have all been unconvincing to me because they imagine that the next bubble will look like the last one.

Until today. Gillian Tett in the November 23 Financial Times has come up with the first description of a bubble that I’ve seen that is both convincing to me in its mechanism and in its unexpectedness. The bubble she describes is still inflating and isn’t about to break tomorrow, but it is worth taking very seriously if you’re building a portfolio with a time frame of five years or longer.

Warnings are a dime a dozen these days. In the last two weeks, for example, the Chinese and German governments have both lectured the United States and particular the U.S. Federal Reserve about the danger that low U.S. interest rates and a weak U.S. dollar are fueling a new global bubble. (It’s probably just a coincidence that the two countries doing the lecturing as two of the world’s  biggest exporters are among the countries hurt the most by a weak dollar. How China, which has pegged its currency to the dollar, is hurt by a weak dollar is a long and complicated story that I’ll have to save for the future.)

The weakness in all these warnings about a new bubble is that when they get around to describing the next bubble it sounds an awful lot like the last bubble. We’ll see an unsustainable increase in real estate prices, for instance, or commodity stocks will rocket and then burn, or banks will use cheap money to rebuild risky derivative portfolios.

But if past bubbles should have taught investors anything it is that the next disaster never duplicates the last disaster. That’s because we put rules—official and ad hoc—in place after each disaster to prevent a replay. It takes a long time for memories to fade so that we unwind those safeguards. For example it took 66 years before Congress dismantled the 1933 Glass-Steagall Act, designed to keep commercial banks out of investment banking with the 1999 Gramm-Leach-Bliley Act. And then it took almost another decade before banks imploded in crisis that would have immediately recognizable to Senator Carter Glass and Representative Henry Steagall in 1933.

That’s too slow a pace if you’re looking for another bubble and bust to follow this one on something like the schedule on which this crisis followed the bursting of the technology stock bubble in 2000.

This is where Tett’s piece comes in.

First, she posits a completely unanticipated location for the bubble: In the markets for what’s called sovereign debt. Sovereign debt is made up of the bonds and such that countries sell to finance their budget deficits. There’s a lot of this stuff rolling off the printing presses from Berlin to Washington to Tokyo right now to fund stimulus packages, bailouts of the financial system, and unfunded long term liabilities for an aging world.

What’s surprising about this market is that the huge supply of this paper hasn’t resulted in a massive discount to its price. On the contrary, yields are miniscule, which means bond prices are high. For example, a three-month U.S. Treasury bill showed a yield of just 0.005% on November 20. A five-year Treasury bill yields just 2.18%.

A yield like that says either that investors are expecting current deflation, running at an annual rate of 0.2% in October, to continue for five years or more. That seems unlikely since if you take out volatile food and energy prices the U.S. economy isn’t experiencing deflation at all but instead a 1.7% annual rate of inflation. To believe in deflation, you’ve got to believe in really weak economic growth that is enough to push down the price of oil and food for the next five years.

Or, government bonds are over-priced. With a real yield of just 0.48% on a five-year Treasury (if you use the core inflation number) that’s a possibility that should be taken seriously.

There are lots of reasons that sovereign debt could be over-priced right now.  The financial markets and the global economy are scary enough so that investors are willing to take a very low yield in return for high safety, for example.

And some of the biggest buyers in this it’s-safe-so-who-cares-about-the-low-yield market are the world’s big banks, Tett points out. In response to a financial crisis that was set off and then exacerbated by bank’s over-investing in risky and illiquid derivatives such as collateralized debt obligations (CDOs, remember them) regulators around the globe have pushed banks toward increasing their holdings of sovereign debt. It’s risk free, the regulators have told banks, and the regulatory system is set up to reward safety by decreasing the amount of capital a bank needs to hold in reserve as the safety of its capital portfolio holdings increases.

And second, Tett posits a mechanism for the bubble to keep inflating and then bursting.

No one wants to say, Hey, these sovereign bonds really aren’t risk free. The banks certainly don’t want to say it, because recognition of the risk in these bonds would drive down their prices. The central banks and national treasuries don’t want to say it because they need to sell a lot more paper in the years ahead. The governments of the developed economies need to sell about $12 trillion in bonds in 2010, Tett calculates.

Instead of selling sovereign debt to reduce their holdings, banks and other big holders of these bonds, have recently turned to the derivatives market, particularly to an instrument called a credit default swap that is supposed to insure against a government defaulting on its debt. Not surprisingly Italy, which one of the worst long-term deficit problems leads the developed world in the amount of credit default swaps outstanding. Recently there were $216 billion in these derivative insurance policies outstanding against Italian sovereign debt, according to figures pulled together by the Financial Times.

That makes the $10 billion in these derivatives outstanding on U.S. debt, the $15 billion on Japanese debt, and the $24 billion on United Kingdom debt look like pocket change. But before you get too complacent, that pocket change amounts to a 150% increase, a 114% increase, and a 100% in the amount of credit default swaps outstanding on U.S., Japanese, and United Kingdom debt, respectively.

So what could cause a bust?

Two things would have to happen. First, some event, a default or near default on sovereign debt by a major developed country, would have to send the prices of sovereign debt plunging. That would trigger huge losses in bank portfolios.

Second, the credit default swaps market, the market that is supposed to insure against these losses, would have to behave in the same kind of unexpected fashion as the derivatives market did in 2007 and 2008. Big losses at the banks holding sovereign paper would by themselves be extremely painful. But those losses wouldn’t be enough to trigger a true bust large enough to shake the financial system again unless a derivatives market designed to insure against loss actually worked to magnify losses and to create a liquidity freeze so that banks that needed to raise cash and/or capital because of a big drop in the price of sovereign debt couldn’t raise the money they needed.

Let’s not forget that one of the major causes of the crisis in the financial system—the reliance of banks and other big financial institutions the short-term commercial paper market—hasn’t changed significantly for many institutions. If banks can access that market to raise new cash or to roll over existing commercial paper borrowings, it would indeed produce another crisis.

For that to happen bank and other financial institutions have to keep building their portfolios of sovereign debt and to continue to rely on the derivatives market for insurance on those portfolios. To build the kind of leverage structure that’s a precondition to a bubble and a bust, I think pattern of activity will have to run through 2010 and 2011.

It’s not a problem for today but it sure is a worry for tomorrow.