Score-driven panel data models of the capital structure of US firms

Abstract

ABSTRACT Capital structures of United States (US) firms may be influenced by changes in interest rate levels, which may be in relation to lower borrowing costs of firms in different periods of conventional and unconventional monetary policies (UMPs). The present article is the first comparative study of the capital structure of Standard & Poor’s 500 (S&P 500) firms for subperiods between 1997Q4 and 2019Q3. Subperiods are defined based on the evolution of the effective federal funds rate (EFFR), which include a period of decreasing EFFR in response to the Dotcom Bubble, and the near-zero EFFR for the period of UMP. The hypothesis of this article is that S&P 500 firms increase the proportion of debt financing in periods of decreasing or low interest rate levels, and the opposite is implemented in periods of increasing interest rates. To provide a robust test of the hypothesis, the new score-driven panel data model is introduced. The hypothesis is supported for debt-to-capital of S&P 500 firms by the empirical results.