The Reality of Trade Credit and its Link to Bank Borrowing and Inventory: (2) Correlation Coefficients and Multiple Regressions

This is the last of the 4 discussion papers that, together with the Introduction and Summary paper (Miwa, 2010c), comprise the report of my recent investigation: "A Study of Financing Behavior of Japanese Firms with Firm-Level Data from Corporate Enterprise Quarterly Statistics -- 1994~2009". In the third paper, I examined the place of trade credit in Japan. Here, I continue that exploration, and turn to trade credit patterns during the "Credit Crunch." I examine the correlation among various financial variables, including the ratio of specific variables to total assets (level variables), and the ratio of the change in specific variables to total assets (difference variables). The results are surprisingly stable over time, and similar between level variables and difference variables. The only stable relation is between payables and receivables. This appears in multiple regression studies both on level variables and on difference variables -- and appears for the size and t-value of regression coefficients and adj. R squares. Moreover, the results are stable over time, and similar between level variables and difference variables. Observers argue that Japan experienced a serious "Credit Crunch" during 1997-1999, but has avoided much of the confusion caused by the recent world-wide "financial crisis." The data I studied suggest the opposite. I find no indication of any serious "credit crunch" ten years ago, but find a drastic decrease both in payables and in receivables in firms of all sizes.