The Jorda et al. estimate of the world Market Risk Premium

The Market Risk Premium (expected excess return of equities and other risky assets over risk free assets) is an important element in asset pricing models particularly the Capital Asset Pricing Model. Estimating the Market Risk Premium from historical data is very difficult because of high volatility – the sample mean over even many decades of data is subject to too large a sampling error. For example, reliable historical data on risk premiums in India goes back less than three decades, and we worry whether the realized risk premium over this period is representative of what premium will prevail in future. Data going back around a century is available for the United States, but use of this data raises serious issues of survivorship bias, as the US is clearly one of the best performing economies of the last century.

I think the NBER Conference paper by Jorda, Knoll, Schularick, Kuvshinov and Taylor “The Rate of Return on Everything, 1870–2015” is a valuable new estimate of the Market Risk Premium. First they have put together a large sample: 16 advanced economies over almost 150 years (the length of the sample varies from country to country). Second, they compute the Market Risk Premium using not merely equities, but also housing which is the most important risky asset outside of equities. In finance theory, the Market Portfolio in theory includes all risky assets, and including housing moves the empirical estimation closer to theory. Pooling data across all countries, they arrive at the following conclusion:

In most peacetime eras this premium has been stable at about 4% – 5%. But risk premiums stayed curiously and persistently high from the 1950s to the 1970s, despite the return to peacetime. However, there is no visible long-run trend, and mean reversion appears strong. The bursts of the risk premium in the wartime and interwar years were mostly a phenomena of collapsing safe rates rather than dramatic spikes in risky rates. In fact, the risky rate has often been smoother and more stable than safe rates, averaging about 7% – 8% across all eras.

It is interesting to observe that the Capital Asset Pricing Model was created during the period of high risk premiums in the 1960s, and its obituaries started being written in the 1980s and 1990s when the risk premium collapsed to very low levels (Figure 10 in the paper).

Jorda et al. also provide an estimate of another important risk premium using the same long period multi-currency sample: the term structure premium or the liquidity risk premium (bonds versus bills). This risk premium is around 1.5% for the full sample, but somewhat larger during the last quarter century (Figure 3 of the paper).

Posted at 5:01 pm IST on Mon, 4 Sep 2017         permanent link

Comments