Capital Structure Determinants Based Upon Total Debt, Short- and Long-Term Debt, Industry and Time

Abstract

We investigate important debt level determinants and whether or not these determinants are affected by industry and time. Continuing from our previous paper, we include short- and long-term debt levels and test whether or not there are differences of key debt level determinants between long- and short-term debt levels. We find that unlike Western firms, short-term debt level takes a bigger portion of total debt level than long-term debt level in Korean firms. Although this is an unusual phenomenon reflecting financial theory, this is the way Korean firms manage their debts. This proves that Korean firms rely more upon short-term debt. We find that not only total debt level, but also short- and long-term debt levels also have target debt levels; and this fact implies that adjusting debt level is firms’ top priority as previous debt levels are related to firm‘s bankruptcy probability and efficient debt management policy. This of course supports the trade-off theory. We also find that previous debt levels, total assets and firms’ earnings are the most important capital structure determinants for firms; and these factors are followed by new investment, dividend and tangible assets. Their associations with DLs (total debt levels, and short- and long-term debt levels) are generally consistent. Namely, DLs are decided by the similar key factors. In addition, cash level change, interest rate and time variables show different directions with DLs between short- and long-term debt levels. They also show different associations with DLs in line with industries and time. Namely, some variables play different roles for firms’ short-term and long-term debt level decision in accordance with industries and time. This paper also suggests that annual stock price growth and R&D expenses have less influence on DLs. We find some evidence that industry and time factors influence debt level shifts, but we admit that we do not have crystal clear structural evidence to prove these influences. Finally, this paper mainly supports the trade-off theory. However, as given evidence of the negative relationship between DLs and earnings, we could also partially support the pecking order theory. However, we do not find evidence that support the market timing theory.