Macroeconomic news and stock prices over the FOMC cycle, Accessible Data

Figure 1. Evolution of sources over time.

Shows the evolution of the number of macro releases (sources) that we include in our two indexes, and their sum, over time. As it is evident in the figure, the set of macro news sources increased steadily from 2000 to 2013, and has remained relatively consistent since.

As is evident in the figure, the set of macro news sources increased steadily from 2000 to 2013, and has remained relatively consistent since.

Source: Bloomberg Finance LP, Bloomberg Terminals (Open, Anywhere, and Disaster Recovery Licenses); and Bloomberg Finance LP, Bloomberg Per Security Data License.

Return to text

Figure 2. News indexes over time.

Shows our news indexes and the Citigroup US Economic Surprise Index (CESI) over time, plotted against the average of daily return of the S&P500. In particular, we show the 8-FOMC moving average of our indexes and the CESI. We aggregate the daily activity news index $$ani$$, price news index $$pni$$, and CESI as the mean over the inter-FOMC period, then scale each variable to a mean of zero and standard deviation of one. We take the 8-period moving average here to reflect the average number of FOMC meetings for each year, smoothing out seasonal fluctuations. The left axis shows the standard deviation of each index, and the right axis shows the corresponding smoothed average daily return of the S&P 500 over the intermeeting period, in percent. While ani shows a clear positive correlation with the S&P500 returns, pni shows a negative correlation. CESI looks like having a positive correlation but not as strong as ani.

This figure plots the 8-FOMC moving average of our two macro news indexes and the CESI for US. We aggregate the daily $$ani$$ and $$pni$$ as the mean surprise over the inter-FOMC period and CESI as the average daily change over the same period, then scale each variable to a mean of zero and standard deviation of one. We take the 8-period moving average here to reflect the average number of FOMC meetings for each year, smoothing out seasonal fluctuations. The left axis shows the standard deviation of each index, and the right axis shows the corresponding smoothed average daily return of the S&P 500 over the intermeeting period, in percent (dashed line).

The shaded bars indicate periods of business recession as defined by the National Bureau of Economic Research: December 2007–June 2009 and February 2020.

Source: Bloomberg Finance LP, Bloomberg Terminals (Open, Anywhere, and Disaster Recovery Licenses); and Bloomberg Finance LP, Bloomberg Per Security Data License.

Return to text

Figure 3. Inter-FOMC S&P 500 returns vs. news indexes.

Shows scatterplots of the Inter-FOMC S&P 500 returns vs. news indexes. These scatterplots show the relationships between the average daily inter-FOMC S&P 500 returns and the average daily standardized activity news index, price news index, and CESI. The thick black line shows the predicted values from a univariate regression on each. The standardized ani has a coefficient of 0.06 with significance at the 1% level and an R2 of 17%; the standardized pni has a coefficient of -0.02, with significance at the 1% level and an R2 of 3%; and the CESI has a coefficient of 0.02, with no statistical significance and an R2 of 2%.

These scatterplots show the relationships between the average daily inter-FOMC S&P 500 returns and the average daily standardized activity news index, price news index, and CESI. The thick black line shows the predicted values from a univariate regression on each. The standardized $$ani$$ has a coefficient of 0.06 with significance at the 1% level and an R2 of 17%; the standardized $$pni$$ has a coefficient of -0.02, with significance at the 1% level and an R2 of 3%; and the CESI has a coefficient of 0.02, with no statistical significance and an R2 of 2%.

Source: Bloomberg Finance LP, Bloomberg Terminals (Open, Anywhere, and Disaster Recovery Licenses); and Bloomberg Finance LP, Bloomberg Per Security Data License.

Return to text

Last Update: October 14, 2020