It has been common to assess the equity risk premium (the difference between returns on equities and bonds), allowing for all of the capital growth observed in the past. In fact, almost all of the volatility in equity returns stems from capital growth rather than from dividends.

The history chart below looks at what the risk premia would have been over rolling periods of 15 years if the capital return component had been 50% (“CapGrow_050%”) OR 75% OR 100% of what had been observed.

The components chart looks at the income and capital percentage components of the risk premium (with 100% capital return).  For periods of 15 years starting from 1962, the equity return capital component averaged 56% of the total return (standard deviation 15%). Interestingly, 50% was assumed throughout the 1980s by at least one major UK consultancy.

Finally, for each of the four periods considered, the table shows the likelihood of the equity risk premium being as high as 0% pa, 1% pa, 2% pa and 3$ pa. For the “blend” period, in which I’m most interested, the premium would be as high as (or higher than) 3% pa over 15 years 43% of the time.