Every few years for at least the last 2 decades we’ve read that “60 percent stock/40 percent bond portfolio” investing is dead. Some shock, so the storyline goes, will force down both fixed income and equity asset prices.
Pop quiz: guess how many times both the S&P 500 and 10-year Treasuries have posted total return losses in the same calendar year. The answer, back to 1928, is just 4 times:
- 1931: -43.8 percent for the S&P 500, -2.6 percent for the 10-year
- 1941: -12.8 pct for the S&P, -2.0 pct for the 10-year
- 1969: -8.2 pct for the S&P, -2.0 pct for the 10-year
- 2018: -4.2 pct for the S&P, -0.02 pct for the 10-year
Even if you expand the definition of “bonds” to include US BBB corporates, you only add 4 years to the list:
- 1937: -35.3 percent for the S&P 500, -4.4 pct for BBB bonds
- 1966: -10.0 pct for the S&P, -3.5 pct for BBB bonds
- 1974: -25.9 pct for the S&P, -4.4 pct for BBB bonds
- 2008: -36.6 pct for the S&P, -5.1 pct for BBB bonds
As comforting as that historical record is, the world is somewhat different now. First, rates are lower so duration risk is higher. We can’t as easily make up principal loss with coupons. Second, we have the dual specters of Fed bond purchase tapering and relatively high current inflation. Combine those 2 factors in a sloppy market environment and both stocks and bonds could come out the worse for wear.
Now, neither low rates nor tapering are exactly new phenomena; we had both in the 2010s. 10-year Treasury yields were sub-2.0 percent in 2012 and going into the May 2013 “Taper Tantrum”. Now, 10-years are a touch lower than then (1.33 pct today) but the Fed bond purchase tapering chatter is with us again.
That means we can learn something useful by looking at US equity correlations relative to long-dated Treasuries as well as investment grade/high yield corporates from 2012 – present. To simplify this analysis we used ETF prices (IVV, TLT, LQD and HYG, respectively) and calculated trailing 100-day price return correlations between US stocks (IVV, the S&P 500) and the 3 fixed income options.
Here is what we found:
Long dated Treasuries rarely show positive correlations to US stocks.
- The average correlation back to 2012 is -0.37 (r-squared of just 14 percent).
- As the chart below shows (where the line touches the x axis about a third of the way across the time series), the trailing 100-day correlation after the May 2013 “Taper Tantrum” was still just -0.04 (r-squared of 0).
- For the most recent 100 days, the correlation between long-dated Treasuries and the S&P 500 is -0.15 (r-squared of 2 percent).
Investment grade US corporate bonds average essentially zero correlation to the S&P 500, at -0.005 (r-squared of zero).
- As with Treasuries, you can make out the 2013 Taper Tantrum about a third of the way across the time series shown below.
- IG correlations to the S&P 500 hit +0.34 (r-squared of 12 percent) in October 2013, the highest levels until the Pandemic Crisis hit.
- At present, IG correlations to the S&P 500 are 0.14 (r-squared of 2 percent).
US high yield corporate bonds are much more correlated to US stocks, as you’d expect.
- Their average 100-day price correlation to the S&P is 0.70 (r-squared of 49 percent).
- The 2013 Taper Tantrum was not kind to this asset class; spreads widened and underlying yields rose. That actually reduced their price correlation to stocks (which did quite well in the 100 days after the “Tantrum” started) to 0.57 (r-squared of 32 percent) but not in a “good way” – they lost money in this period.
- HY correlations to the S&P 500 are 0.75 (r-squared of 56 pct) at present, slightly higher than the long run average.
Takeaway: Treasuries and investment grade corporates have structurally low correlations to US stocks, and even the 2013 “Taper Tantrum” only boosted them temporarily. Both are, therefore, still valuable diversifiers for equity investors. High yield corporates are another matter entirely; they essentially trade like equities. We still prefer HY over IG, but if you are concerned about a 2021 – 2022 “Taper Tantrum” then history says IG or even Treasuries will provide better diversification.
Historical stock and bond returns: https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/histretSP.html