A reader suggested that we evaluate the performance of Between the Hedges, a “portfolio manager’s commentary on investing and trading in the U.S. financial markets.” One prominent and systematic feature of the commentary in that blog is the daily net portfolio position, expressed as percentage long. This position changes frequently, and the portfolio manager presumably manages it to exploit expected short-term trends in the broad stock market. If the expectation has value, the net portfolio position should relate positively to near-term broad market behavior. Using the Between the Hedges daily net portfolio position for 2/2/04-2/26/08 (1,024 trading days) and contemporaneous daily data for the S&P 500 index, we find that:
In extracting the daily net portfolio position from Between the Hedges, we note that:
- There are frequently intraday changes in net portfolio position. We choose only the final value for each trading day.
- There are a handful of trading days for which there are no final net portfolio positions, so the sample size is 1,015 rather than 1,024.
- Net portfolio positions range from -50% to +150%. Most positions are in increments of 25%.
- There are only 18 negative (net short) positions, and most of these are from 2004.
The following scatter plot relates next-day (close to close) returns for the S&P 500 index to prior-day closing Between the Hedges net portfolio positions over the entire sample period. The Pearson correlation between the two series is -0.05, and the R-squared statistic is 0.003. Even with the large sample, these statistics indicate that there is no dependable relationship between next-day S&P 500 index returns and prior-day closing net portfolio positions. In other words, the net portfolio positions do not effectively anticipate (or benefit from) very short-term broad market trends.
By contrast, the Pearson correlation between daily closing Between the Hedges net portfolio positions and S&P 500 index returns for the same day is 0.39. Evidence suggests that the net portfolio positions are short-term reactive but not predictive.
For a different kind of test, we calculate average next-day S&P 500 index return for specific values of the net portfolio position.
The next chart summarizes the average next-day S&P 500 index return for specific values and ranges of the Between the Hedges net portfolio position, with one standard deviation variability ranges for each. For example, the average next-day (close to close) S&P 500 index return is -.06% when the prior closing net portfolio position is market neutral (0%). The average daily return for all days in the sample is +0.02%.
The results here confirm that there is no systematic relationship between net portfolio positions and next-day broad stock market behavior. A longer net portfolio position does not systematically relate to a higher next-day average stock market return.
In summary, evidence from simple tests does not support a belief that the Between the Hedges net portfolio position anticipates the short-term trend of the broad U.S. stock market.
As noted, the net portfolio position at Between the Hedges often changes intraday. It is possible that using intraday net portfolio positions and stock index values might affect this conclusion.
This review does not evaluate the stock, sector or style picking performance of the Between the Hedges portfolio manager.
