A survey of CEOs at Fortune 1,000 firms asked about their firms’ hurdle rates and time horizons. Survey results suggest that most U.S. firms use hurdle rates that are higher than standard cost-of-capital analyses would suggest. The average discount rate applied to constant-dollar cash flows was 12.2 percent, distinctly higher than equity holders’ average rates of return and much higher than the return on debt during the past half-century. At the time of the survey, the fall of 1990, U.S. CEOs believed that their firms had systematically shorter time horizons than their major competitors in Europe and (especially) Asia. U.S. CEOs also thought that government policy is a powerful agent affecting corporate planning horizons. They saw several policy reforms, including a cut in corporate tax rates, a permanent R&D tax credit, a corporate tax deduction for dividend payments, and a credible commitment to a stable tax policy for the next decade, as policies that could lengthen planning horizons.
1. For recent data on manufacturing productivity in the United States, Japan, and other G-7 nations, see:
E.R. Dean and M.K. Sherwood, “Manufacturing Costs, Productivity, and Competitiveness, 1979–1993,” Monthly Labor Review 117 (1994): 3–16.
2. For a summary of the policy debate about the role of time horizons in exacerbating the competitiveness problems of U.S. firms, see:
M. Jacobs, Short-Term America (Boston: Harvard Business School Press, 1991).
For a survey of studies on the cost of capital in U.S. and Japanese firms, see:
J.M. Poterba, “Comparing the Cost of Capital in the United States and Japan: A Survey of Methods,” Federal Reserve Bank of New York Quarterly Bulletin, Winter 1991, pp. 20–32.
For a discussion of various policy measures that might lengthen corporate time horizons, see:
National Research Council Board on Science, Technology, and Economic Policy, Investing for Productivity and Prosperity (Washington, D.C.: National Academy Press, 1994).
3. There are dozens of studies of corporate capital budgeting practices. See, for example:
L. Schall, G. Sundem, and W. Geijsbeek, Jr., “Survey and Analysis of Capital Budgeting Methods,” Journal of Finance 33 (1978): 281–287; and
J. Binder and J.S. Chaput, “A Positive Analysis of Corporate Capital Budgeting Practices” (Chicago: University of Illinois-Chicago, mimeo, 1991).
However, the competitiveness policy debate has largely ignored them.
4. A copy of the survey is available from the authors on request.
5. In April 1991, we also sent the questionnaire by fax to the fifty largest firms in Germany, Japan, and the United Kingdom. We discuss these responses later.
6. We view both the long-term R&D share and the hurdle rate as measures of corporate time horizons. We therefore examined their correlation in our sample and found that the share of long-term R&D is (weakly) negatively correlated with the real hurdle rate.
7. Ibbotson Associates reports a geometric average return of 10.3 percent (nominal) on common stocks of large firms, 12.4 percent on common stocks of small firms, 5.6 percent on long-term corporate bonds, and an inflation rate of 3.1 percent for the 1926–1993 period. See:
Ibbotson Associates, Stocks, Bonds, Bills, and Inflation: 1994 Yearbook (Chicago: Ibbotson Associates, 1994).
8. M.S. Feldstein, J.M. Poterba, and L. Dicks-Mireaux, “The Effective Tax Rate and the Pretax Rate of Return,” Journal of Public Economics 21 (1983): 129–158.
9. Competitors of the firms we analyzed may not be representative of Asian or European firms; they may be more export oriented and potentially longer horizon firms than companies engaged only in domestic sales.
10. Another explanation for this finding is that the CEOs who responded were more attuned to problems of time horizons than a purely random sample. Their firms could typically have longer time horizons than their domestic competitors.
11. We present detailed information on the responses to each question because two questions may have identical mean responses but quite different distributions in the intensity of views. Two questions may have equal average responses (say, 3), but one might have an equal number of 2, 3, and 4 responses, suggesting no very strong views on this policy, while another could have the same mean but half of the respondents indicating 1 and the other half 5. Table 5 provides this information.
12. This is a surprising finding since a “rifle-shot” capital gains tax cut for equities would make them more attractive than other gain-producing assets, reducing the cost of equity more than a capital gains tax cut for all assets.
This project was carried out with the support of the Time Horizons Project of the Harvard Business School and the Council on Competitiveness. We are grateful to George Hatsopoulos, Jay Lorsch, Michael Porter, and two anonymous referees for helpful comments and suggestions, and to Doug Hendrickson, Elizabeth MacIver, and especially Jim Saylor for assistance in carrying out and analyzing the survey. The research was conducted when Summers was a professor of economics at Harvard University. Views expressed here do not represent the position of the U.S. Treasury Department or any other organization.