Why Don't Economists Follow The Leaders?
Paul L. Kasriel
The "leaders" to which I am referring are those in the Conference Board's index of leading economic indicators (LEI). The Conference Board on Thursday announced that the LEI fell 0.2% in April after having dropped a downwardly-revised 0.6% in March. On a year-over-year basis, the April LEI index is down 1.2% -- the largest year-over-year percent decline since the 2.0% drop in April 2001. I tuned in CNBC at 10:01 am EDT on Thursday morning, one minute after the release of the April LEI index report. Perhaps the CNBC commentators mentioned the report as it hit the wires, but there was no discussion of it in the minutes immediately after the release. In the Thursday edition of The Wall Street Journal, some economists were disparaging the LEI index as information that not only was "old," but also subject to revision. Moreover, the LEI index, in the opinion of these economists, is irrelevant in today's services-oriented U.S. economy. It is amusing to hear economists deride the LEI index for containing data subject to revision. Economists attach great importance to a report released the first Friday of the month, a report that is a coincident indicator of economic activity, which is subject to huge revisions. As for containing stale information, mainstream economists pay little attention to the two most highly-weighted components of the LEI index when they first become available, namely the yield spread between the Treasury 10-year bond and fed funds and the real M2 money supply. So, I suspect that it is not because the LEI index contains old news that most economists ignore it. Rather, it is because the LEI index contains some variables that most economists believe are uninformative. Why do they find these variables uninformative? Probably because they took the word of some "authority" figure rather than having the curiosity to find out for themselves what the relationships are.

I think I have figured out why economists downplay the significance of the LEI index. It is to preserve their jobs. After all, if the LEI index does a better job of forecasting the behavior of the economy than do economic forecasters, why do we need economic forecasters? A few months ago, we called the Blue Chip survey folks to find out if the consensus of the best and brightest had ever forecast a recession. Blue Chip's answer was "no." As Chart 1 shows, the LEI index has done an excellent job of forecasting recessions. The year-over-year percent change in the LEI monthly index has gone negative prior to the onset of every NBER-designated recession since 1960.

Chart 1

Now, it is true, that the LEI index has sent some false recession warnings. That is, there have been some occasions when the year-over-year percent change in the LEI index has gone negative and no officially-designated recession ensued. But that does not mean that the weakness in the LEI index was not signaling a developing weakness in the growth of economic activity. Chart 2 shows the relationship between growth in the LEI index and the Coincident Economic Indicators index. (The Coincident Economic Indicators index is made up of nonfarm payrolls, real personal income less transfer payments, real manufacturing and trade sales and the Fed's index of industrial production.) We can see in Chart 2 that although a contraction in the LEI index may not always signal an NBER-designated recession, it does signal weaker growth in the Coincident Economic Indicators index. By the way, the highest correlation, 0.77, between the growth in the Coincident index and the Leading index occurs when the Leading index is advanced by 7 months. A correlation coefficient of 0.77 is not perfect. Perfect would be 1.00. But folks, a correlation of 0.77 is fantastic for government work and its darned good for private sector work, too.

Chart 2

But how relevant is the Coincident index to real GDP in today's post-modern economy? Chart 3 shows the relationship between growth in the Coincident Economic Indicators index and growth in real GDP. The correlation between the two has diminished in recent years. Between 1960:Q1 and 1994:Q4 the correlation was 0.91. It has slipped to 0.79 in the period between 1995:Q1 and 2005:Q1. I don't know about you, I would say that a correlation of 0.79 still is relevant.

Chart 3

Lastly, another reason you should be paying attention to the behavior of the LEI index is that the Fed has been raising its policy rate. As Chart 4 shows, a rising fed funds rate goes hand-in-hand with a declining growth rate in the LEI index. Perhaps economists are surprised to learn that when the Fed starts to restrict the flow of the credit it creates out of thin air, the pace of economic activity slows down. But I bet you are not surprised by this.

Chart 4


May 25, 2005

Paul L. Kasriel, Director of Economic Research (plk1@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.