Sunday, November 4, 2007

How Long Term Interest Rates Limit the Federal Reserve's Options

You get the impression that, all public protestations aside, the Fed would like to bail out the CDO and mortgage-backed securities markets. Its surprise half-point fed funds rate cut on September 18, and the more recent quarter-point cut on Oct. 31, won't have much impact on overall economic growth for months, perhaps many months, since interest rate changes need time to work their way through the system. (Check to see if the interest rates on your credit cards have dropped--you'll see what we mean about things taking time.) The justification in the economic data for these cuts was skimpy, especially the quarter-point drop on Oct. 31. But the financial markets wailed loudly and threatened to throw a monumental fit, so the Fed indulged them. The best of all worlds for the Fed would be to take monetary action that can be explained as beneficial for the economy as a whole, but which quietly alleviates the pain that so many market participants are feeling.

Yet, in spite of the Fed's interest rate cuts, the major banks reported some mega losses for the third quarter, topped by Merrill's $8.4 billion. This illustrates a limitation on the Fed's ability to deal with the subprime mess.

The Fed's power to affect interest rates is focused on the short term debt market. Fed funds are overnight loans that banks make to each other. Overnight is about as short a term for a loan as you can get.

But CDOs and other mortgage-backed securities are usually medium to long term investments. Most mortgages are paid approximately 2 to 10 years after they are taken out. That's roughly how long people wait to either refinance or sell their homes. Consequently, mortgage-backed securities tend to have longer terms than fed fund loans. Of course, it's possible to create short-term derivatives from mortgages. But at the end of the day, someone has to hold the right to the long term payments. So, there will always be longer term interests in any securitization of mortgages. And given the trillions of dollars of mortgages that have been securitized, the long term interests exist in large quantity. Perhaps one might think that lowering long term interest rates would increase the value of mortgage-backed securities (in the way it would increase the value of a traditional bond). If that could be done, we'd have a bailout, presto quick. But let's look at interest rate movements.

Since June 2004, the Fed has raised the fed funds rate from 1% to 5.25%, and then moved it back down to 4.5%. Short term interest rates, such as banks' prime lending rates and interest rates on money market funds, followed the upswings and downswings in the fed funds rate closely. What about longer term rates?

The most important long term security in the financial markets is the 10-year U.S. Treasury Note. It serves as the benchmark for all other debt of comparable maturity, because it is effectively free of credit risk. Credit risk, as we know so well from the credit crunch, is the murkiest of all risks that debt instruments bear. Eliminate credit risk, and you can get a much clearer picture of the time value of money. The 10-year Treasury Note is issued in abundance, since it is the Treasury Department's principal long term borrowing instrument. (Thirty-year Treasury bonds were not issued between 2001 and 2006.) Investment banks and other market players use the 10-year note as a hedging tool (because of its creditworthiness and easy availability), and institutional investors worldwide seek out the 10-year note as a safe haven.

During the time that the fed funds rate was moving within a range of 4.25 percentage points, the 10-year Treasury Note, starting in June 2004, varied between roughly 3.9% and 5.3%, a range of 1.4 percentage points. Moreover, the 10-year note was often the most inverted part of an inverted yield curve, dropping even as the fed funds rate was rising.

In other words, changes in the fed funds rate don't have much impact on longer term interest rates. In June 2004, the 10-year Treasury Note was yielding at around 4.7% when the fed funds rate was 1%. Today, it yields around 4.3%, even though fed funds rates today are 3.5 percentage points higher. The Fed can cut short term interest rates as much as it wants, and it won't have much impact on mid to long term interest rates.

It's also important to remember is that a drop in longer term interest rates may not increase the value of CDOs and other mortgage-backed securities. A drop in longer term interest rates can increase the frequency of refinancings, as borrowers seek to lower their monthly payments by getting a new and cheaper mortgage, and using it off the old one. Thus, in an environment of falling interest rates, the pool of mortgages supporting an existing mortgage-backed security may pay off faster than investors had hoped. They receive their principal back sooner, and must now re-invest at the lower prevailing rates. This represents a loss of long term income to them. Thus, a drop in longer term interest rates, which would ordinarily increase the value of existing longer term debt, may actually decrease the value of a mortgage-backed security. (However, the opposite isn't likely to be true--an increase in longer term interest rates probably won't do much to increase the revenue stream from the mortgage pool, but would depress the value of that stream.)

So, even if the Fed could lower long term interest rates, such a move might backfire by hurting the value of mortgage-backed securities. This is probably an important reason why three major banks and the Treasury Department have proposed a Super Conduit (or Super SIV, as some commentators call it) to bail out SIV-bedeviled banks. This is one respect in which the Fed can't ride to the rescue. Holders of mortgage-backed securities will likely have to work their way through their problems themselves. The fact that an ad hoc solution like the Super Conduit was even proposed tells you that there are no obvious solutions to the problem (see our blog at

The financial markets have been pouty ever since the Fed signaled that their Halloween candy might have to last for a while. Inflation risks are rising (as we predicted in, and the Fed may be constrained from more interest rate cuts. That may be just as well, since it is doubtful that interest rate cuts could do much to alleviate the subprime mess. But they could fuel inflation. Let's not forget that the foundation of today's prosperity (yes, the economic data continues to show that the U.S. is prosperous, housing downturn notwithstanding) emanates from Fed Chairman Paul Volcker's unwavering stand against inflation at the end of the 1970s and in the early 1980s, which threw the U.S. into a painful recession that saw unemployment rise to more than 10%. Inflation is the biggest asset bubble of all, one that pumps up the price of all assets, and working our way out of an inflationary bubble would be far more painful than dealing with the current mortgage mess.

Legal Update: coin-flipping judge gets the boot. (That's right; a judge flipped a coin to make a decision.)

No comments: