How will demography and entitlement spending affect future interest rates and investment returns?

The Congressional Budget Office today released an interesting and readable explanation of the ways in which demographics and government spending can affect future interest rates and returns on investments such as stocks. This is relevant to the Social Security debate as some analysts have argued that future stock returns will be much lower than in the past, thereby undercutting at least part of the rationale for reforms including personal retirement accounts.

Several points seem worth highlighting:

• The interest rate over the long term is a function of the relative sizes of the capital stock and of the labor force. If slower population growth reduces the size of the labor force relative to the capital, each worker will have relatively more tools and machines with which to work, raising worker productivity and wages. However, more capital per worker also implies a lower marginal return to capital, and therefore a lower interest rate.
  
• Using a standard economic model and reasonable assumptions for economic parameters, CBO shows that a one percentage point reduction in the growth of the labor force would produce a reduction in interest rates around 1.5 percentage points. For illustration, from 1960 through 2008 the labor force grew at annual rate of around 1.7%; over the next 75 years the projected rate of labor force growth averaged around 0.5%. All else being equal, this 1.2 percentage point reduction in the rate of labor force growth would reduce the average interest rate by around 1.8 percentage points.
  
• A lower interest rate would tend to reduce returns on stocks, making personal accounts less attractive. The return paid to stocks is a function of the general interest rate paid to all investments and of the premium paid to riskier investments, such as stocks. The discussion in this CBO paper focuses on changes in the overall interest rate, not upon the size of the risk premium. All other things equal, differences in returns between safe in risky investments would remain roughly constant even if average returns rose or fell. (That said, I agree with CBO's general practice of abstracting from the equity premium so to avoid the misinterpretation that stocks produce "free money." A large equity premium doesn't necessarily imply that stocks are a "good deal"; it may just imply that they're very risky or that investors are very risk averse.)
  
• At the same time, lower interest rates would make a long-term Social Security actuarial deficit larger. A reduction in the real interest rate of 1.8 percentage points would increase the 75-year Social Security deficit from 2 percent of payroll to around 2.9 percent of payroll. While a high ratio of capital to labor would also increase wages, this very likely wouldn't be enough to offset the poorer actuarial balance due to lower interest rates. (A 1 percentage point reduction in labor force growth would reduce interest rates by around 1.5 percent and increase the wage rate by around 0.6 percent. However, the wage rate differs from the growth of wages. Thus, we could see a single year in which wages grew 0.6 faster than ordinary or, more likely, a number of years with very slightly faster annual wage growth until the new equilibrium level was reached.)
  
• While demographic changes are likely to reduce interest rates, budget deficits associated with financing Social Security and Medicare would tend to increase interest rates. CBO concludes that in most cases the positive effect on interest rates of rising budget deficits would outweigh the negative effect from demographic change. In other words, we're as likely to see higher interest rates and investment returns in the future as we are to see lower ones.
  

Overall this is a very good piece of work on an interesting and important topic – technically solid but accessible enough to be understandable to the public.