Don’t be locked into a start year for historical returns

BVWireIssue #195-3
December 19, 2018

cost of capital
cost of capital, private company valuation, risk analysis, business valuation resources, cost of equity

During a recent demo of BVR’s new Cost of Capital Professional, an audience member asked whether he could choose the starting year used to calculate the historical equity risk premium and historical size premium. Yes, the platform is designed to be flexible, so you can choose any starting year you want. You can choose preselected dates (1928, 1963, or 1982), but you can enter a custom year. The CRSP data go back to the 1920s, and Compustat data, used by some analysts, go back to the 1960s. The size premium was first documented in academic research in 1981, and it has diminished or disappeared since that time, hence the 1982 selection.

Another audience member asked: “How do I determine the start date for the historical returns?” That boils down to the analyst’s professional judgement as to what version of the past is best indicative of the future, which is not something that can be determined easily, the presenters said. In essence, the issue is whether something important enough has changed in the markets that changes rates of return. For example, if you look at the past nine years, you logically would have to admit that markets have changed. No laws of nature say that all rates of return for the S&P 500 or size premiums go back to a certain level of equilibrium.

Free trial ending soon. Take a free test drive of the Cost of Capital Professional before the end of the year—the free trial period ends December 31. After that, the annual price will be $295 for the first user and $100 for each additional user. There’s a special offer of 14 months for the price of 12 if you subscribe now. Check out a demo and free webinar that explains the thinking behind the platform. BVR welcomes feedback and comments from all users.

Please let us know if you have any comments about this article or enhancements you would like to see.