Objective research and reviews to aid investing decisions
 Looking for information on debt consolidation loans? Want to know more about reverse mortgages? |
Does the stock market react reliably and exploitably to the monthly announcements of the change in the Consumer Price Index (CPI) by the Bureau of Labor Statistics? To check, we examine the typical behavior of stocks during the five trading days before and the five trading days after CPI release dates. Using non-seasonally adjusted total and core CPI data, associated release dates and contemporaneous daily S&P 500 index data for the period 1/94-4/08 (171 announcements), we find that...
Because U.S. stock markets closed for part of the ten-day interval of analysis for 9/01, we exclude the release of CPI data during that month.
The following chart compares change in non-seasonally adjusted total CPI with the opening level of the S&P 500 index on CPI announcement dates across the sample period. Visual inspection suggests that: (1) CPI is more volatile in the 2000s than in the 1990s; and, (2) large positive changes in CPI might be related to dips in the stock market.
For greater precision, we examine daily changes in the S&P 500 index around announcements of CPI changes.
The next chart plots the average return for the S&P 500 index during each of the five trading days before and five trading days after monthly CPI releases, with one standard deviation variability ranges above and below the averages, for all announcement dates in the sample period. Day 1 is the day of (pre-market) announcement. The average daily return for all days during the sample period is 0.04%. The chart shows that, in comparison with standard deviations, all average daily returns around announcement dates are quite close the average for all days in the sample. There is no exploitable anomaly.
Might detailed distributions of the data reveal irregularities for specific (extreme) ranges of CPI changes?
The following scatter plot relates the cumulative return for the S&P 500 index for the five trading days after monthly CPI releases to the change in total CPI. The Pearson correlation for the two series is 0.06, and the R-squared statistic is 0.00. There is no systematic relationship between changes in total CPI and five-day movements of the stock market after announcement. Moreover, the fairly uniform dispersion of the data points offers no hope that any particular ranges of total CPI changes, whether extremely high or extremely low, predict the short-term reaction of stocks.
Might the distribution for changes in core inflation be more useful?
The next scatter plot relates the cumulative return for the S&P 500 index for the five trading days after monthly CPI releases to the change in core CPI. The Pearson correlation for the two series is also 0.06, and the R-squared is again 0.00. There is no systematic relationship between changes in core CPI and five-day movements of the stock market after announcements. And again, the fairly uniform dispersion of the data points offers no hope that any particular ranges of core CPI changes, whether extremely high or extremely low, predict the short-term reaction of stocks.
Given the apparent increase in volatility of CPI during the 2000s compared to the 1990s, have stock returns around CPI releases been stable across the entire sample period?
The final chart compares average daily returns for the S&P 500 index around CPI release dates for two subsamples: the 1990s (72 announcements), and the 2000s (99 announcements). It shows a reversal in average behavior in the three-day interval bracketing release dates. During the 1990s (2000s), this interval is positive (negative), suggesting a stream of good (bad) CPI surprises. Long-term inflation trend may determine the likely response of the stock market to CPI releases.
Use of changes in seasonally adjusted rather than non-seasonally adjusted CPI makes no significant difference in the typical short-term stock market reaction to CPI releases. The media generally highlight seasonally adjusted changes in CPI.
In summary, CPI announcements by themselves have little or no explanatory power for short-term stock returns over the period 1/94-4/08, with a negative average immediate effect in the 2000s canceling a positive average immediate effect in the 1990s.
Note that the above analysis does not take into account how CPI releases compare with the consensus expectations of experts. This baseline for defining "surprise" might expose systematic relationships between CPI changes and short-term stock returns.
See Blog Synthesis: The Economy and the Stock Market for analyses of the relationships between other economic indicators and stocks.
