Tuesday, February 3, 2009

The crisis is passing (7)

Mark Perry has a good piece today on the New York Fed's model for predicting recessions. Happily, the model says the recession is very likely to end this year.
The Fed's data show that the recession probability peaked during the October 2007 to April 2008 period at around 35-40%, and has been declining since then to less than 10% for December 2008 and January 2009. Looking forward through 2009, the Fed's model shows a recession probability of only about 1% on average through the next 12 months, and below 1% by the end of the year (.82% by January 2010). The Treasury spread has been above 2% for the last 11 months, a pattern consistent with the economic recoveries after the 1990-1991 and 2001 recessions.
The chart above is my contribution to the discussion. It shows the slope of the yield curve, as does the Fed's model, but adds the real Federal funds rate as well. As should be obvious, recessions are almost always preceded by a) a significant rise in the real funds rate (i.e., an overt tightening of monetary policy), and b) a significant flattening of the yield curve (which typically confirms that money is tight). Both of those have reversed in the past year, a pattern that has signaled the end of each recession. My prediction at the end of last year still holds: "the economy is going to recover sooner than the market expects, with the bottom in activity coming before mid-2009; the recovery will be sub-par however, due to the drag of increased fiscal spending and slowly rising inflation."

I'm ashamed to say that I failed to pay attention to this chart in late 2007 (I've been using it for many years). I didn't accept its signal then that a recession was coming, because I didn't think the real Fed funds rate had risen enough. The rule of thumb I was using was that the real funds rate needed to be at least 4% in order to signal a recession. I'll have to revisit that rule going forward.


joemckendrick said...

Scott: Is it fair to assume that when the recession "officially" ends (as defined by the NBER, when we begin to ascend from the trough), that most of us likely won't even be aware of it? For example, for all we know, six months from now, NBER may call this month (February) as the end of the recession. Or are there signs economists such as yourself will recognize? -Joe McKendrick

Scott Grannis said...

Good question. The man on the street and the mainstream media likely won't figure out the recession has ended until well after the fact. Financial markets will figure it out first. Traditional indicators to look for would be a peaking in unemployment claims, a bottoming in payroll employment, a bottoming in industrial production, and a peaking in the unemployment rate.

Check this post for a link to a good paper on recession indicators by Ed Leamer.


Chad said...

If you want to see just how slow the media are to recognize the end of a recession, go to www.time.com and check out the "Why We're So Gloomy" article in the January 13, 1992 issue.

For those who don't remember, the 1990-91 recession ended fully 10 months before that issue came out. Yet what was the blaring headline on the cover? "THE RECESSION: HOW BAD IS IT?"

The sub-head was even better: "And what gives on Wall Street?" The editors couldn't figure out how the Dow, NASDAQ, and S&P 500 could be up roughly 35%, 90%, and 40% from their intra-day lows with all the supposed doom and gloom in the economy.

Like Warren Buffet said in his 10/17/08 op-ed piece in the New York Times, "So if you wait for the robins, spring will be over."

Chad said...

...and by intra-day lows I mean the markets' recession intra-day lows that came in October of 1990.

theshoe said...

For those interested in judging the robustness of the term structure of rates with respect to economic forecasting I encourgage a reading of "Financial Market Perceptions of Recession Risk" available at the Federal Reserve's Finance and Economic Discussion Series. It will reaffirm one's use of the yield curve.

Mark A. Sadowski said...

You should be ashamed for not following the yield curve. If you had you would have noticed that it was inverted from mid 2006 through mid 2007. It's one reason why I started forecasting a recession in the summer of 2006.

The problem with using the Federal Reserve model now is that it is based on empirical data collected since 1980. This is the only time in Federal Reserve history that the target federal funds rate has essentially fallen to zero. The model was never designed estimate probabilities accurately in a deflationary liquidity trap and so must be interpreted with skepticism.I personally am still predicting the recovery to begin in the fourth quarter of 2009.

Scott Grannis said...

Mark: I am ashamed that I dismissed the yield curve signal in 2007. But now you're telling me you feel comfortable doing the same?

Mark A. Sadowski said...

The Federal Reserve can’t reduce interest rates any lower, because they’re already zero. It can, however, raise rates. The long-term rate has to be above the short-term rate, because given the current conditions it’s like an option price in that short rates may move up, but they can’t go down.

In fact, if we look at Japan we find that the yield curve was positively sloped during the entire Lost Decade. For example, in 1999-2000, with ZIRP in effect, long rates averaged just under 2 percent, not too far from current rates in the U.S.

In other words, the current yield curve isn't very encouraging at all. It’s only telling us what I already believe, that we are in a deflationary liquidity trap.

Scott Grannis said...

Mark: just because rates are at zero doesn't mean the Fed can't ease further. Of course they can. They can buy things all day long and expand the money supply. If they do that enough, they will push inflation up and that will steepen the yield curve. But perhaps you're saying that the yield curve doesn't tell the whole story and if so, then I agree with you. I think you always have to take everything in the context of all the other things going on. The real funds rate is an important thing to watch. When the real funds rate and the yield curve tell the same story (i.e., now both are saying the Fed is extremely easy), then I think you have to assume that financial conditions are at the very least getting less strained, less tight, and in the context of the financial crisis that is very important.

Mark A. Sadowski said...

I more or less agree with your assessment but I'm still skeptical about how much the Fed can accomplish with quantitative easing when we have a financial sector populated with zombie institutions. So far I don't have much confidence in how TARP has been handled. I'm curious what Geithner is going to announce on Monday.