In part 1 of this series we proposed an alternative approach for measuring the correlation between bonds and equities, showing how the co-movement of these two asset classes has recently been increasing and of late bonds have failed to ‘protect’ during equity sell-offs.
In part 2 we extended our approach to the sectors of an equity benchmark. We showed how being long the benchmark is the determining factor that drives the time-varying correlation with bonds and is not determined by the underlying sectors once the market is removed (beta-hedged).
Here, in part 3, we extend the use of our procedure further to investigate the co-movement relationship between bonds and equity risk factors. We use data from the Kenneth French website where various risk factors are provided.1 The website provides data conveniently grouping stocks into ‘Decile’ portfolios, meaning the variable upon which the factor is based is split into 10 such stock portfolios representing the levels of the factor variable itself – ranging from highest to lowest. Having investigated all factors, we found ‘Market Beta’ to be most interesting!
So, with Decile 1 being a portfolio of stocks with the lowest market betas; Decile 10 being a portfolio of stocks with the highest market betas; and all Deciles in between being a continuum of increasing and sorted betas, we repeat the exercise of looking at the correlation of each with the US 10-year total return index2 after having residualised the Decile portfolios relative to the market.3
We found that not only is there a statistically significant relationship between the beta of the portfolio (the sorting variable) and its residual correlation with fixed income, but that this relationship was perfectly rank-ordered and remarkably consistent through time. See figures 1 and 2.
So, where does this remarkably stable effect come from? Our belief is that this is a risk appetite effect. The stocks in the low market beta Decile buckets are those that are either low volatility or low correlation to the market. In either case, these are generally characterized as being sure-fire payout stocks with stable business models and high near-term cashflow. They tend to be the dividend payers. In our last blog, looking at sectors and how they correlate with bonds after projecting out the market, one observes that the sectors that are most correlated with bonds are those that fit into this category – notably Utilities and Consumer, Staples. The stocks at the other end of the spectrum, meanwhile, are contrarily high volatility, speculative, far dated payout growth type stocks that are a much riskier proposition. These stocks are also therefore, and contrarily, anti-correlated with the sure-fire payout 10-year treasuries. What is interesting is that, at least if our logic is correct, market beta captures this effect better than fundamental metrics used in the other factors – variables such as cashflow or earnings.
The plot thickens still further for long only investors. Residualising stocks to the market before calculating bond-equity correlations cleans up the picture significantly. However, it seems an even cleaner picture emerges when looking at stocks ranked in terms of their Market Beta rather than considerations of which sector they belong to!
1 For the purpose of this post, and to extend the look-back period as far as possible, we use monthly data for each of the equity risk factors. Interested readers can visit the website of Kenneth French for the data and explanatory notes. We tried the following list: Market Beta, Size, Variance, Book-to-Market, Operating Profitability, Investment, Accruals, and Net Share Issues. Results were mixed and the only factor with a statistically significant pattern presented here was Market Beta.
2 For a longer historical look-back period, and to overlap with the data from Kenneth French, here we use data from Global Financial Data (GFD) for the US 10-year total return.
3 We use the ‘Market’ as per Kenneth French for residualising the monthly returns, i.e., ‘Mkt-Rf’ but adding back the risk-free return (Rf) since the Decile portfolio returns themselves are not excess returns.
Disclaimer
ANY DESCRIPTION OR INFORMATION INVOLVING MODELS, INVESTMENT PROCESSES OR ALLOCATIONS IS PROVIDED FOR ILLUSTRATIVE PURPOSES ONLY AND DOES NOT CONSTITUTE INVESTMENT ADVICE. ANY STATEMENTS REGARDING CORRELATIONS OR MODES OR OTHER SIMILAR BEHAVIORS CONSTITUTE ONLY SUBJECTIVE VIEWS, ARE BASED UPON REASONABLE EXPECTATIONS OR BELIEFS, AND SHOULD NOT BE RELIED ON. ALL STATEMENTS HEREIN ARE SUBJECT TO CHANGE DUE TO A VARIETY OF FACTORS INCLUDING FLUCTUATING MARKET CONDITIONS AND INVOLVE INHERENT RISKS AND UNCERTAINTIES BOTH GENERIC AND SPECIFIC, MANY OF WHICH CANNOT BE PREDICTED OR QUANTIFIED AND ARE BEYOND CFM’S CONTROL. FUTURE EVIDENCE AND ACTUAL RESULTS OR PERFORMANCE COULD DIFFER MATERIALLY FROM THE INFORMATION SET FORTH IN, CONTEMPLATED BY OR UNDERLYING THE STATEMENTS HEREIN.
