This study tested the pecking order theory of capital structure using the Frank and Goyal (2003) model for 299 firms during the analysis period from 1995 to 2001. The results showed that the financing deficit variable had a significant positive (+) effect on leverage changes at the 1% level, confirming that it is one major factor explaining leverage changes. However, the magnitude of its effect on leverage was not large enough to eliminate the effects of traditional variables on leverage changes, so the pecking order theory could not be said to be supported. Sub-sample analyses based on chaebol affiliation, chaebol ranking, and pre- and post-IMF bailout periods showed no substantial differences from the results for the full sample.