Portfolio Choice and the Debt-to-Income Relationship
NBER Working Paper No. 1545 (Also Reprint No. r0628)
The ratio of outstanding debt to gross national product in the United States has shown essentially no time trend over a period measured not in years but in decades. The research reported in this paper indicates that lenders' portfolio behavior exhibits characteristics that could provide aplausible explanation of this phenomenon. Given the long-run stability of the U.S. economy's wealth in relation to income, the question of lenders' behavior explaining the stable aggregate debt-to-income ratio turns on whether investors treat debt and other assets as close or distant substitutes in their portfolios. Analysis of financial assets' respective risk properties indicates that debt and equity are indeed sufficiently distant substitutes for lenders' behavior to confine the debt-to-income ratio within relatively narrow limits. In particular, the substitutability of debt and equity securitiesis sufficiently limited that very large movements in expected return differentials -- movements so large as presumably to elicit offsetting responses from borrowers -- would be required to induce major changes in the debt share of investors' aggregate portfolio, and hence in the economy's aggregate debt-to-income ratio.
Published: Friedman, Benjamin M. "Portfolio Choice and Debt-to-Incoome Relationship." American Economic Review, Vol. 75, No. 2, (May 1985), pp. 338-343.