Leading Economic Index and the Stock Market
Posted in Economic Indicators
February 17, 2012
The Conference Board “publishes leading, coincident, and lagging indexes designed to signal peaks and troughs in the business cycle for ten countries around the world,” including the widely cited Leading Economic Index (LEI) for the U.S. Does the LEI predict stock market behavior? Using the as-released monthly change in LEI from archived Conference Board press releases and contemporaneous dividend-adjusted daily levels of SPDR S&P 500 (SPY) for June 2002 through January 2012 (116 monthly observations), we find that:
The Conference Board releases the LEI for a month about three weeks after the end of the month. To analyze its relationship to the stock market, we consider market behavior relative to both the month measured by LEI and the LEI release date. Since the Conference Board regularly recalibrates the level of LEI, we focus exclusively on the monthly change in LEI.
The following scatter plot relates the monthly SPY total return to the monthly change in LEI for the same calendar month over the entire sample period. The average monthly change in LEI is 0.16%, and the average monthly change in SPY is 0.44% over this period. The Pearson correlation between the two series is 0.39, and the R-squared statistic is 0.15, implying that monthly variations in LEI explain 15% of concurrent stock market movements. However, as noted, the LEI reading for a calendar month is not available until well after the month has ended.
To evaluate the exploitable monthly predictive power of LEI for stocks, we focus on LEI release dates.
The next scatter plot relates SPY total return to change in LEI from the close on the day before the LEI release date to the close on the day before the next LEI release date over the entire sample period. The Pearson correlation between the two series is 0.10, and the R-squared statistic is 0.01, indicating little or no useful predictive power for LEI over the month after its release.
As a separate simple ranking test, we calculate the average SPY returns during the month after LEI release for the lower half of changes in LEI (average SPY return 0.54%) and the upper half of changes in LEI (average SPY return 0.70%) over the entire sample period. Results suggest some discriminating power of LEI for next-month SPY returns.
Might LEI changes be looking ahead more than one month?
The next chart summarizes Pearson correlations between monthly changes in LEI (returning to calendar months measured rather than release-to-release months) and monthly SPY returns for lead-lag relationships ranging from stocks lead LEI by six months (-6) to LEI leads stocks by six months (6). The strongest indication is that stock returns lead LEI changes by one month. There is some indication that LEI weakly leads stocks by one to four months.
What about trading in the few days around releases of new LEI data?
The final chart shows the average daily SPY returns from three trading days before (-3) to three trading days after (3) LEI release dates (day 0 is the release date) over the entire sample period for:
- All release dates, with one standard deviation variability ranges.
- Release dates when LEI decreases.
- Release dates when LEI increases.
The average SPY daily return over the entire sample period is 0.03%. Results suggest that the average response to LEI decreases is most expoitable as reversion after a negative announcement day reaction, but variability generally swamps averages.
In summary, evidence from simple tests on a limited sample offers little support for the belief that the Conference Board’s Leading Economic Index for the U.S. exploitably predicts the behavior of the U.S. stock market over the short term or the intermediate term.
Cautions regarding findings:
- The sample period is short in terms of number of economic cycles.
- The Conference Board may occasionally adjust the LEI calculation method, confounding measurement of predictive power.
- The fact that the S&P 500 Index is one of ten components of LEI introduces some positive coincident correlation between the two series. This intersection makes it difficult to distinguish any predictive power of LEI from that for stock market intrinsic momentum (autocorrelation).
You May Also Enjoy...
- ECRI’s Weekly Leading Index and the Stock Market
- Stock Market Reaction to FOMC Meeting Minutes Releases
- Distress Index and Stock Returns
It costs less than a single trading commission. Learn more here.
