The ever-higher prices investors appear willing to pay for the perceived safety of many traditionally defensive assets has been a recurring theme for some time now, here on The Value Perspective – most recently in articles such as Driving returns and Calm before the storm. Rarely, however, have we seen the issue illustrated as starkly as in the following chart.
Large-Capitalisation Stocks, Highest and Lowest Deciles of Return Correlation with Ten-Year Treasury Bonds Relative Trailing- P/E Ratios 1952 to Late-June 2016
Source: Empirical Research Partners Analysis.
Included in some analysis by our friends at Empirical Research Partners, the chart highlights the relative price-earnings (PE) ratios of large US stocks whose returns have shown the greatest and least correlation with 10-year US government bonds since 1952.
The idea, of course, is that those stocks with the highest correlation to treasuries are the kind of stable, low-volatility, ‘safe-haven’ names whose shares have literally been behaving the most like bonds in recent times – while those with the lowest correlation are thought by the market to be the exact opposite.
Each of the chart’s two lines – for the groups of companies with the highest and lowest correlations with treasuries – has a dot showing where the current valuation stands today in the scheme of the last 64 years. As you can see, the group of stocks whose returns are most correlated with treasuries – what have come to be known as ‘bond proxies’ – is some way towards the higher end of valuations over this period.
In contrast, the stocks whose returns are least correlated with treasuries – are about as cheap as they have ever been in the six-plus decades under review by the Empirical analysts.
Clearly the treasury market is being used here as a proxy for investor perceptions of safety – or the lack thereof in the equity market. Yet, at least to our way of thinking, here on The Value Perspective, this only serves to underline how many investors have, for some years now, viewed large stocks with historically low volatility as bond proxies – the risks of which approach we have discussed in articles such as Oh-oh seven.
To reprise just a couple of the issues here, there will come a time – really there will – when, for whatever reason, these bond proxies are no longer what investors want. It might be because economic growth improves and cheap, economically sensitive stocks suddenly seem exciting rather than risky or it might be that interest rates rise and the eye-wateringly high multiples – PEs of 25x or more – that investors are currently paying for ‘safety’ can no longer be justified.
Whatever the reason, as the above graph shows, these bond-like businesses are close to being the most expensive they have ever been – with history suggesting future returns that will therefore be close to as low as they have ever been. When they fall out of favour, those who are invested in them have little to no margin of safety.
Whereas for the businesses that are least like that – the sort of ones we own in our portfolios? We are not saying their low – admittedly, relative – valuations mean future returns will be as high as they have ever been. Yet history also suggests those willing to buy in the areas on the other curve above, where valuations are as low as they have been in more than 60 years, are giving themselves a great chance of seeing good future returns on a medium-term view.