Now, we have come to appreciate in the 30+ years of being directly or tangentially involved in the art of forecasting that there is no single monetary policy indicator that sends a true "signal" all of the time. But when two indicators that have shown a great deal of reliability over time line up on the same side, we take notice. And both the slowdown in real M2 growth - outright contraction in the second quarter of this year - and the steady narrowing of the yield spread are sending a strong signal that economic growth, which already has moderated on a year-over-year basis, is likely to moderate further.
Mention of these two indicators of the thrust of monetary policy, real M2 and the yield spread, are a segue into a discussion of the Conference Board's index of Leading Economic Indicators (LEI). These two monetary policy indicators happen to be components of the LEI. The LEI is not a constant through time. Periodically, some components are dropped, new ones are added and computational methodologies are changed. And, with the release of the June LEI, there will be some changes in it - computational changes. The principal change will be how the yield spread computationally affects the index. Heretofore, a narrowing in the yield spread has resulted in a negative contribution to the LEI. With the new version of the LEI, a narrowing in the yield spread will not make a negative contribution to the LEI until the spread itself becomes negative - i.e., until the fed funds rate is above the yield on the Treasury 10-year bond. A narrowing in a positive yield spread will now just make a less positive contribution to the LEI. There are some other computational changes, which you can read about at www.conferenceboard.org/economics/bci/RevisionsLEI_2005.pdf
In Chart 2 we have plotted the year-over-year percent change in quarterly averages of the "old" version of the LEI and the "new" version. As you can see, the principal difference between the two is of scale. The growth trends move in tandem. It is just that the old version of the LEI has lower growth rates. Whereas the year-over-year percent change in the old LEI for Q2:2005 (using the April-May average as a proxy for the Q2 average) is minus 1.5%, the change for the new LEI is plus 1.69%. But the qualitative story is the same - after peaking in mid 2004, growth in the LEI - old and new - has been on a sharp downward trend. Is it just a coincidence that this major trend change in LEI growth commenced as the Fed began its "measured" pace of interest rate hikes?
So, big deal, the growth trends in the old and new LEI move in tandem. But, as Maria Bartelomo would screech, "How do I make money with this?" In other words, how good is the new LEI at predicting real GDP growth? Actually, better than the old LEI, which was no slouch itself. Chart 3 shows the relationship between the year-over-year percent change in the new LEI (advanced by one quarter) and the year-over-year percent change in real GDP. As you can see, the fit is almost as tight as O.J.'s glove. The correlation coefficient between the two is 0.795, about one and onehalf points better than what was obtained with the old LEI. Nothing is perfect in the business of forecasting economic growth. But Chart 3 presents a compelling case to us that economic growth is on a decelerating path and will likely to continue along that path until there is a major reversal in the growth trend, not a one-month reversal, of the LEI.
Before leaving the subject of the LEI, we would like to make one more point. The revisions that the Conference Board made to the LEI did not include the removal of real M2 or the yield spread from the components. This implies that despite the low repute with which mainstream forecasters, including the Fed, hold these two indicators of monetary policy, the Conference Board still believes, based on empirical evidence, they are worthy candidates to be included in the LEI. Interestingly enough, the Conference Board has not seen fit to include in the LEI the Fed's and many mainstream forecasters' favorite monetary policy indicator, the real fed funds rate. The real Fed Funds rate must not do a very good job of forecasting economic growth or else it would have been added to the LEI.
What does all of this have to do with the title of this month's commentary? The LEI suggests that the trend toward slower economic growth remains intact. The "new" LEI is not yet suggesting a recession is on the horizon. But if the Fed does not soon call a cease fire in its interest rate increases, the LEI is likely to begin emit recession-warning signals. Why? Because continued increases in the fed funds rate could very well place it above the Treasury 10-year bond yield and could very well slow nominal and real money growth even more.
The Fed has raised the funds rate a total of 100 basis points in the first half of this year. Given the lags in the effects of monetary policy, the full impact of this cumulative 100 basis point increase in the fed funds rate on economic growth has not yet occurred. As can be seen in Chart 4, the year-over-year percent change in real GDP growth in Q1:2005 was 3.7%, which is down considerably from the 5.0% year-over-year growth in Q1:2004. Given the lagged effects of policy-induced interest rate hikes already in the pipeline, would it not be reasonable to expect a continued moderating trend in real GDP growth? As shown in Chart 5, both all-items and core-inflation appear to have peaked and are currently at relatively low levels. Given that inflation tends to be a lagging economic indicator (the CPI for services is a component of the Conference Board's index of Lagging Economic Indicators), given that it already has peaked and given the economic growth could be expected to moderate further, is it not reasonable to expect inflation to be contained going forward? That's what real people betting real money think, as shown in Chart 6.
It is not clear to us why the Fed would want to raise the funds rate on August 9. But given no indication to the contrary from public comments by Fed officials, we surmise that they are committed to at least one more rate hike. And, as a result, we are revising down are Q4:2005 real GDP growth forecast by 20 basis points. But, currently, we can think of only one rational reason why the Fed would continue to raise the funds rate after August 9. That reason is to remove some of the froth in the housing market. This would represent a departure from its stated policy of not moving the funds rate to influence asset prices. Moreover, the Fed and other regulatory agencies appear to be applying regulatory pressures on banks and thrifts in order to get them to rein in their housing-related lending. But for whatever reason, if the Fed insists on raising the funds rate to 4% or above this calendar year, we believe the risks of a 2006 recession move up exponentially. And we believe that what might at first look like a mild recession could turn into a severe one if it burst the housing bubble. Given how contained price increases of goods and services appear to be and given the high probability that the economy is settling into its potential growth rate or a little below, we do not think that the Fed wants to risk overdoing it now. One of the reasons we have been reluctant to issue a 2006 GDP forecast is that what the Fed does after August 9 will have profound effects on 2006 economic growth. If the Fed pauses after August 9, economic growth in 2006 looks to be in the range of 3% to 3-1/2%. If the Fed marches the funds rate up to 4% or 4-1/4% by December 31, 2005, then 2006 looks to be a tough year, economically speaking.
July 20, 2005
Paul L. Kasriel, Director of Economic Research ( plk1@ntrs.com )
Asha G. Bangalore, Economist ( agb3@ntrs.com )
The information herein is based on sources which The Northern Trust Company believes to be reliable, but we cannot warrant its accuracy or completeness. Such information is subject to change and is not intended to influence your investment decisions.