So you’re a macro trader and one of your assets is fixed income. You focus on Fed policy, economic growth, and inflation. We all know that a strong stock market is not likely to be a great help to your long bond positions but you really have no edge when it comes to forecasting the S&P 500. If you knew the stock market was going to fall tomorrow then you could buy bonds the night before (and the converse) and make a fortune.
Does it help to know what yesterday’s stock performance looked like? Certainly but not as much as you’d expect. (See below) We should expect an up day in S&P’s will be followed by a reversal the next day (or at least a day of paltry gains) so we should buy bonds and the converse. Doing so would flip this 2nd plot, upside down. (more on this later).
Here’s a different question: Does today’s relative performance of tech stocks vs. the S&P tell us anything about tomorrow’s bond market behavior:
This is all done after the fact so we don’t need to forecast the Nasdaq’s relative performance. Clearly, bonds like to see the S&P outperform the Nasdaq. Here’s the combined P&L:
I’ve applied this to the US 10 Year contract but seeing how we are measuring long-term risk factors it makes more sense to apply it to US 30’s:
Since the middle of 2015 this has worked relatively well – gaining 55 points or so while the market has gained only 5.
Why is this true?
It would appear that a day when the S&P outperforms the Nasdaq is followed by a day of S&P weakness or at least a day of catchup where the Nasdaq does better but the net changes are small. Bonds seem to behave well under those conditions.
Back to Looking at Stock Market Performance, Outright.
What if we just wait for the stock market to rise (or fall) by a significant amount (say) equal to half a standard deviation.
Now we see how the clear daily mean reversion nature of stocks can be used to gain an edge trading long bonds. You just have to filter out inconsequential changes. Putting the two sets of trades together we earn over 85 points for the 5 year period while the market gains about 18. It would seem that the buy signal is more reliable than the sell signal. Once again the bond market anticipates a sort of dissipation in the equity trend and stands ready to recover from what may have been a bad day after a significant stock market rise (or the converse).
I’m going to leave the final step to you – combining the two signals so we buy bonds if the S&P is up and the S&P has just outperformed the Nasdaq, and we short after the converse. It does seem clear that bond traders can benefit from looking in the rear view mirror at both the stock market and at tech stocks relative performance.