What if someone told you that there is an indicator that has scarcely ever been wrong about a recession? As one of the few economic statistics generated by the collective insight of the bond markets, its surprising that the yield curve has shown a recessions months before one had actually occurred -- each and every time (excluding a few false positives)

The market for U.S. treasuries is one of the most liquid in the world, and no surprise either, investing in the U.S. government has been one of the safest investments you can put your money in. But not all treasuries are created equal, some are bought for a 3-month timeframe, while other last quite a bit longer -- extending all the way to 30 years.

What determines how much interest you'll receive on what is essentially a loan to the U.S. Treasury Dept? Well, not much can happen in 3-months, so there is not much risk associated with the 3-month treasuries. On the other hand, quite a bit can happen over 30 years (e.g. laws change, bond ratings drop, gdp falls or rises, etc...). Naturally, you'd want a higher interest rate for locking up your money for longer periods of time -- even if it is the U.S. Government.

The Normal Yield Curve

And that's how the market for treasuries works, at least when its operating normally. The 30 year bond usually trades at around a 2.5% premium over the 3-month bill. It has a fairly consistent slope across all maturities, usually indicating that the economy is running smoothly. If you watch the yield curve from 1986 to 1987, you won't see anything special. That's because the bond markets didn't have much to worry about: they were already 3 years away from the savings & loans crisis and there didn't seem to be any sign of a bubble. GDP growth was fairly consistent, and the Fed had controlled inflation. As a result, the 30-year bond commanded its normal premium over shorter maturities.

This isn't always so, however, as fears of a recession or optimism for a bubble can make things a little funky.

Inverted Yield Curve

So you might think the yield curve will always have a positive slope. Seems perfectly logical, right? But take a look at the series from 2000 to 2001 (during the dot-com bubble) where after a few variations, the interest rate on 3-month bills were greater than the yields on the 30-year bonds.

That means people who were taking much less risk in their 3-month investments were making more profit than those buying 30-year bonds!

What seems like a paradox at first can actually be explained by the collective fear of the bond markets. The time between 2000-01 was the precursor to the dot-com crash, which would inevitably result in a recession. During a recession, the Fed tends to lower its Fed Funds Rate (to promote growth) by increasing the supply of dollars through the purchase of treasuries. With so much government demand for treasuries, the average interest rate on t-bills and bonds will fall precipitously. Therefore, a smart investor would want to lock in high interest rates now before a downturn takes place. The 5-year bond tends to go to the lowest yield since that's the average period the Fed lowers and raises rates back up again.

Is the bond market correct in locking in high interest rates? Absolutely, in our dot-com bubble example from 2002 to 2003, the 3-month bond falls from 6% annualized interest to a paltry .75% interest in just the course of a year.See if you can calculate the gains in buying the supposedly "lower-yielding" 5-year bond, below:

From an economic point of view, the inverted yield curve has been one of the most reliable indicators of an upcoming recession. In our example, the following months after the yield inversion, led to a widespread collapse of the tech companies that fueled the Dow. From 2000 to 2002, the market value of the tech industry fell by $5 trillion, wiping out investor portfolios. Over the years there have only been two false positives of this curve, one in 1966 and another in 1998.

While rare, the inverted yield curve is an incredibly important event