“Markets don’t ask why you own a position or how much work you’ve done when they decide if you’re going to make money on it.” That’s something I first heard at the old SAC Capital many years ago and I have never forgotten it. Its folksier version is “when it comes to investing, you can be right for the wrong reasons, but try to avoid the opposite”. Either way, the message is the same: markets price information, but perhaps not the things you think matter most.
Just to make things even harder, the combination of macro/micro information that moves asset prices constantly changes. For example, there are times when Fed policy shifts can lift or submerge all the boats in the harbor. Other times they seem to float more freely, responding individually to small gusts or the wake of a passing vessel nearby.
To assess where we are in that continuum, we look at asset price correlations every month. For example, the chart above shows the average S&P 500 sector correlation to the index over the last 11 years. What it shows:
- From late 2009 to October 2016 correlations averaged 0.83, for an r-squared of 69%. Those readers with pre-Crisis experience will recall that the long run US sector r-squared is 50%. The period from 2009 – 2016 saw higher levels because Fed policy (rates and quantitative easing) were the most important issue to equity investors.
- After the 2016 US elections, correlations dropped to average of 0.56 (r-squared of 31%) through 2017 and 0.67 (r-squared 45%) in 2018. This decline seems to have come from hopes for deregulation/tax code changes as well as (possibly) the Fed’s attempts to normalize interest rates. No matter the cause, daily sector price action clearly broke free from the tractor beam of Federal Reserve policy.
- In 2019, correlations have averaged 0.74 (r-squared of 55%). They were, of course, higher at the beginning of the year (0.88 in January). But once the Fed shifted course and the US-China trade war eased up, they came down to more-average levels.
As for where we are today, the latest sector correlation data looks quite “normal”, so let’s dig a little deeper into the data:
- The average S&P sector correlation was 0.72 over the last 30 days, yielding an r-squared of 52%. That’s slightly better than the last 6-months (0.74) and spot-on long-run averages.
- Rate sensitive sectors like Utilities (0.32 correlation), Real Estate (0.28) and Consumer Staples (0.64) are tracking day-to-day S&P movements the least.
- Financials are the most correlated group this month (0.92), followed by Technology (0.90). That’s one more sign that Financials are the US equity market’s key leadership group just now, not just Tech.
Also worth noting:
- Since the MSCI EAFE Index is heavily weighted to Financials (18%, the largest sector allocation), no surprise that this asset class is trading very much like the S&P 500 with a 0.93 correlation.
- Emerging market equity correlations are, however, at 0.76 to the S&P 500. That is the lowest in the last 3 months and likely driven by hopes for a US-China trade resolution (China is 30% of the MSCI EM Index, the largest of any country).
- Long-term Treasury correlations to the S&P 500 have dropped dramatically in the last 3 months, at -0.27 now versus -0.67 then. That means Treasuries are not hedging daily equity volatility as much as they did when recession fears were top of mind.
As for what all this says about near term market direction, we take a generally bullish message from the data. US sector correlations are within long-run normal bands, which serves to dampen index/portfolio price volatility. Those groups with high correlations – Financials primarily – signal market confidence in economic growth.
That may all seem to be a dramatic about-face from where we were a few months ago, but markets are clearly not rewarding even the most well-researched macro bear case just now. Those may have their day, as we explained in “Markets”. But the old trader’s saying with which we started this section still applies.