Social discounting and the equity premium

There has been a 40-year divide in economics on the relevance to public funding of the equity premium (in particular, today, the consumption capital asset pricing model, or CAPM).

The costs and benefits of public spending are often correlated with income, but conventional neoclassical analysis suggests that the cost of this systematic risk in publicly funded activities is usually trivial. On the other hand, it is often asserted that equity market premiums, which are very much higher than would be estimated from neoclassical analysis (the equity premium ‘puzzle’), should apply also to public funding.

This paper, which aims largely to help government administrations, assembles a picture of the evolving research on and understanding of the premium. Public funding by taxation does impose social costs and lacks the incentives of private financing. There is, however, no evidence to support assertions that the equity risk premium anomaly is relevant to the cost of public funding. In any case, the cost of systematic risk in the benefits of public funding does not fall as an annual percentage rate to financiers, but as an absolute cost to public service beneficiaries.

These considerations reinforce other arguments for discounting over time at a social time preference rate in climate change analysis.

Key points for decision-makers

• The equity premium, the extent to which returns to equity exceed the risk free rate, is a central concept in financial economics.
• It is often asserted that the rates of return to financing revealed by competitive financial markets are a social cost that applies equally to public funding of similar activities. However, the differences between financing by investors seeking a financial return and funding by taxation are fundamental.
• Private financing has crucial, beneficial incentive impacts that are lacking with public funding. However, there is good evidence that the freedom of equity markets that contributes greatly to these benefits also brings significant costs and that these costs are not relevant to public funding.
• Any cost of systematic risk in the benefits of public funding does, anyway, not fall as an annual rate of return, but as a reduction in the value of the benefits to beneficiaries.
• The benefits of publicly funded projects, in developed as well as less developed economies, are sometimes strongly covariant with income, but the cost of virtually any plausible covariance appears to be trivial. However, professional consensus on this matter may be a long way off.
• This applies to all public spending analysis, but is of course especially important for analysis over the very long term.