Media Mutters

David Brock has a new blog out covering the abuses of the Right Wing Media called Media Matters.

Their very first post is called "Backdating the Recession" and it deals with unscrupulous efforts by the Republicans to convince people that the recession began during the Clinton Administration.

Now, it is true, as Media Matters goes into great detail in explaining, that the recession is considered to have begun in or around March 2001, about 2 months after President Bush was elected. The group that determines this is known as the National Bureau of Economic Research. However, as Donald Luskin points out, the NBER determined the starting date of the recession in November 2001, and several of the data that they may have relied on to make that determination have since been revised for the worse.

Most important, back then third quarter 2000 GDP growth was reported at 1.3% -- but now it's been revised all the way down to a negative 0.5%. NBER had no way of knowing that then.

But even if we accept the March 2001 starting date, does that mean that it's Bush's fault? No, because the forces in play to cause the recession had already started well before his inauguration.

Having worked in finance for most of my life, I always look at the Treasury Yield Curve. This is presented in graph form in every issue of the Wall Street Journal, so it's not as if its importance is unknown. The Yield Curve provides a graphic look at the tradeoff between term and rate.



(Note: the graph above is dynamic and may change over time, so the interest rates discussed may be different from what you see above if you're read this in the archives).

The graph above shows the current treasury yield curve. The numbers on the Y-axis are interest rates, and the numbers along the X-axis represent the term or length of time till maturity of the particular treasury. For example, looking at the black line on the chart, we can see that the yield (effectively the interest rate) for a 3-month treasury is somewhere between 0.75% and 1%, that the rate for a 6-month treasury is between 1% and 1.25% and that the rate for a 2-year treasury is between 2.25% and 2.5%. You can also see that as the maturity gets longer, the interest rate generally rises. This makes sense, as the longer one lends somebody money (even the US Treasury) the more risk there is.

The yields on long term treasuries, for example the 10-year and 30-year instruments, are determined by the market. In theory, the yield, or interest rate, has three components:

The safe rate (i.e., the rate the investor would anticipate given no inflation and no risk);

The inflation rate (because the investor does not want his return eaten up by inflation); and

The risk rate for the particular investment.

The shorter term rates (3-month and 6-month) are effectively set by the Fed.

Most of the time the yield curve looks something like the above. But sometimes, in order to battle inflation, the Fed is forced to raise short term interest rates enough that the yield curve becomes flattened, or even inverted, with the short term interest rates higher than longer term interest rates. This means the Fed is putting strong brakes on the economy.

As happened in the last year of the Clinton Administration. In January of 2000, the treasury yield curve was still normal, but quite a bit higher than today's rates--as of 1/14/00 the rates were 5.41% for 3 months (which I'll call the short term rate) and 6.69% for 10 years (which I'll call long term rates).

However, over the next six months things changed substantially. The 3-month yield moved steadily higher, finally reaching 6% just before the Fourth of July. Meanwhile, the 10-year treasury had dropped to 6%, making the yield curve virtually a flat line with a few small bumps. By August the 3-month was yielding a full 1/4% more than the 10-year. The yield curve was thus inverted, predicting recession ahead.

By September of 2000 the spread between the 3-month and the 10-year had widened to almost 6/10ths of a percent. The 3-month topped out on 11/6-11/7/00 at 6.42%, which was still 1/2% higher than the 10-year.

The brakes obviously worked. As Luskin remarks, real GDP growth crashed from 5% to 2% for the last two quarters of 2000, to 0% in the first quarter of 2000. It's quite obvious that the economy was cooling in the last quarter of 2000 under the effect of the inverted yield curve.

The Fed and the bond market began to realize that they had tightened too much and that the economy was slipping towards recession. So they both eased off on interest rates, but rates fell faster at the longer terms. By the end of December 2000, the 3-month was down to 5.89%, while the 10-year had declined all the way to 5.12%. Still a strong brake on the economy. By 1/19/01, the last trading day before the inauguration of President Bush, the 3-month had dropped to 5.24%, while the 10-year had risen slightly to 5.19%.

The next trading day after Clinton left office, the 3-month treasury yielded 5.23% and the 10-year finally inched above it again, yielding 5.25%. It was the first time that the shorter term rate had yielded less than the long term rate since 7/19/00. By the end of March, the short term rate had dropped to 4.3%, while the longer term rate was 4.93%.

How does the inverted yield curve work at predicting economic recessions? Pretty good actually. Obviously it predicted the last one. In mid-1989 it predicted a recession; the economy hit one in the third quarter of 1990. At no other point in the last two decades has the inverted yield lasted for more than a day or two; each time a recession followed.

IOW, a recession was largely inevitable.