Determinants of wealth maximization and optimal cost of capital for non-financial and non-utility firms

The problem. Historically, researchers could only agree that the optimal capital structure was the one that maximized the firm's wealth. Assuming that a common set of factors exists that can be combined to describe both the maximization of the firm's value (i.e., shareholder wealth) and minimization of the firm's capital cost (i.e., determination of optimal capital structure), the major research problem is identifying the proper combination of financial factors and calculating the determinants of those factors. This becomes even more critical as governmental agencies begin to investigate the creative financing schemes affecting today's markets.; Method. This paper assumed that the exhaustive research of the last 44 years has identified all of the factors that determine the minimum cost of capital (i.e., optimal capital structure) and maximization of the firm's wealth. The heterogeneous sample to test the factors and compute the determinants was drawn from S&P 500 for the years 1995 to 2000. The sample was delimited to financial, insurance, and utility companies, and analyzed cross-sectionally and longitudinally.; Results. Using stepwise multiple regressions of cross-sectional and longitudinal data obtained from the large heterogeneous sample of S&P 500 companies and the determinants of a standard set of factors related to cost of capital and firm wealth, the data were combined into 2 integrated models, the integrated firm wealth model (R2 = 77.2%) and the integrated cost of capital model (R2 = 39.6%). Each model demonstrated different levels of accountability in the variances of their respective measures, and each incorporated different (not inversely related) determinants. This tends to refute the accepted theory that “the optimal capital structure is the one that maximizes the firm's wealth” and indicates that each measure is independent. The longitudinal data did not substantially alter the cross-sectional models.; Future research might break the data down into more homogeneous groups by using Global Industry Classification System (GICS) codes to avoid the broader results of the heterogeneous model.