March 3, 1999
Professor Martin Feldstein
Department of Economics
Cambridge, MA 02138
Dear Professor Feldstein:
Thank you for responding to our earlier letters. We appreciate your willingness to take part in the debate over whether historic rates of stock returns are consistent with the growth projections in the Social Security Trustees Report.
Before getting to substantive issues we would like to raise a procedural point where we presumably do agree. We assume that you would join us in urging Congress to require the Social Security Administration to make the same sort of projections for the components of stock returns (capital gains and dividend yields) as it does for all other relevant economic and demographic variables (see pp 57-61 of the 1998 Trustees Report). It can hardly make sense to have detailed projections of wage growth, labor force growth, life expectancy and other variables relevant to the solvency of the program, and then rely on assumptions for stock returns that have never been subject to the same degree of scrutiny.
This procedural matter could also have the advantage of resolving our disagreement. If the Social Security Administration can readily produce plausible projections of capital gains and dividend yields giving 6.75-7.0 percent real returns that are consistent with the other projections in the Trustees Report, then our argument will have been proved wrong. Of course, if we are correct, then it will not be possible for the Social Security Administration to produce such projections.
Turning to the substantive issues, it may be possible to resolve our disagreement by clarifying our position. We have not been arguing, as you assume in your letter, about the returns to physical capital. This debate is entirely about returns to financial capital. (In your example, you assume that these are the same). We have not asserted that it is impossible to have high stock returns in a slow growing economy. Rather we are saying that given current market valuations, it will not be possible to have the 6.75-7.0 percent returns generally assumed.
This point about current valuations is key. The run-up in the stock market has pushed the price to earnings ratios to unprecedented peaks of more than 30 to 1 by most measures. This run-up in stock prices has pushed the dividend yield under 2.0 percent compared with an average of more than 3.5 percent over the prior seventy-five years. (Table B-95 in the 1999 Economic Report of the President gives data on price to earnings ratios and dividend yields for the S&P 500.) In our calculations, we assume that the price to earnings ratios will not rise indefinitely (the alternative assumption quickly leads to price to earnings ratios rising to over 100 to 1). This assumption implies that the capital gains on stock prices will be the same as the growth in profits, which by the Trustees assumptions is equal to the projected rate of growth of the economy. The latter is less than 1.5 percent annually over the next seventy-five years, compared to 3.0 percent over the last seventy-five years. The 2.0 percent dividend yield plus the 1.5 percent projected capital gains give us our 3.5 percent projected rate of return on stocks.
While it is possible that this could move slightly higher if more money were paid out as dividends, even if all profits were paid out as dividends the dividend yield would only rise to about 3.0 percent. Furthermore, since the Trustees growth projections imply some amount of capital deepening, and the labor force does continue to grow throughout this period (albeit very slowly), the necessary investment will probably prevent the dividend yield from rising above 2.5 percent, putting a cap on stock returns of 4.0 percent.
Placing this in the framework of the static agricultural economy you set out in your letter, the basic problem is that the price of financial capital is substantially greater than the price of physical capital. As you suggest, in this economy it is certainly possible to earn a 7.0 percent return on physical capital. However, if we assume that the value of outstanding stock in this economy is 1.6 times as large as the value of physical capital (approximately the current ratio), then the return on stock would be just 4.4 percent (7.0%/1.6 = 4.4 %). This return would all be in dividends, since there would no net investment in an economy that is not growing, and therefore all profits are paid out as dividends. Of course, capital gains would also be zero in this economy.
To make the example more closely approximate the current economic situation, assume that instead of remaining static, this economy begins to shrink at the rate of 1.5 percent a year. Half of this shrinkage is due to a declining labor force (0.75 percent annually) and half is due to a falloff in the rate of multi-factor productivity growth of 0.75 percentage points (soil erosion). In this scenario, holders of stock would experience a real capital loss of 1.5 percentage points annually, assuming a constant price to earnings ratio. This capital loss would be partially offset by a slightly higher dividend yield. Firms could pay out an additional amount of dividends equal to 0.75 percent of the capital stock and still keep the capital/labor ratio constant, since this would be the annual decline in the size of the labor force. (The decline in multi-factor productivity growth allows for no similar offsetting increase in dividends.) The new dividend yield would be approximately 5.0 percent (4.4% + (0.75%/1.6)= 5.0%). This dividend yield, added to the capital loss of 1.5 percent, gives a total return of approximately 3.5 percent, 0.9 percentage points less than in the case where the economy was stagnant.
In this scenario, the return on physical capital would have declined, but not by as much as the return on financial capital. The reason is that the impact of the decline in the size of the labor force is fully offset by the decline in the size of the capital stock. This is accomplished though depreciation of the capital stock. The only way to bring about a similar adjustment in the case of financial capital would be to have a one-time decline in share prices. While this is a plausible adjustment process in the stock market, it implies significant negative returns for the period in which stock prices are falling.
One last point: it is also worth noting in the context of this debate that it has been wrongly assumed that the after-tax return to physical capital averages 7.0 percent. We are enclosing a paper that shows that the after-tax return on physical capital has averaged under 6.0 percent in the years since 1952. The methodology used for calculating returns in this paper was exactly the same as the one used in a 1977 Brookings Papers article that you co-authored with Larry Summers.
Thanks again for engaging us in debate on this issue. We look forward to your response.
Senior Research Fellow
 The net worth of debt adjusted capital stock of non-financial corporations was $6,797.5 billion at the end of 1997, the most recent data available (Federal Reserve Board, Flow of Funds Accounts table B.102 line 31). The Federal Reserve Board has not updated its data on financial corporations for several years, but if we assume that the net worth of financial corporations is 20 percent of the net worth of non-financial corporations, this would add $1,359.5 billion to get a total net worth for corporate capital of $8,157 billion. At the end of 1997, the value of corporate equities was $11,483.8 billion (Federal Reserve Board, Flow of Funds Accounts table L.213 line 19). Since the end of 1997, the S&P 500 has risen by approximately 27 percent; if all equities rose at the same rate, it implies a current market value of corporate equities of $14,584.4 billion. If the value of the physical capital stock rose by 10 percent over this period, it would currently be $8972.7 billion, which gives a ratio of the value of equities to physical capital of 1.63.