More Misconceptions About Japanese Equities

Continuing last week’s discussion on the role of the minority interest in Japanese equity returns, below we perform the same correlation test on universes of DM Americas and EMEA companies. To recap, we tested the following metrics, drawn from each company’s fiscal year-end reports and stated in USD, and correlated them with the percent price return in local currency over the past five years.

  • Accumulated minority interest as a percentage of total capital and of total assets.
  • The percentage of shares outstanding is closely held (by insiders, institutions, etc.).
  • Minority interest expense is a percentage of the cost of goods sold.
  • Equity in affiliate income as a percentage of sales.

The hypotheses for our test were as follows:

H0 (null hypothesis): There is no statistically significant correlation between a given financial metric and a five-year % price return. This would be verified by a correlation’s P-value above 0.05.

H1 (alternate hypothesis): A given financial metric has a statistically significant correlation with price returns. A correlation P-value below 0.05 would identify a given correlation as likely statistically significant and would be the cause for rejecting the null hypothesis for a given metric.

As a little stats refresher, the p-value of a statistical hypothesis test is the likelihood that the null hypothesis is correct. Standard practice is to reject the null hypothesis if the p-value for any metric is below 0.05.

Looking at the results of these tests, we drew the conclusion that Yen-denominated price returns on mid and large-cap Japanese companies over the past five years were statistically unrelated to the metrics we tested. It is important to note that not all the companies tested had non-zero values for the metrics we tested. The following table breaks down the percentage of companies that had values greater than zero for each item, by area.

Share of Companies with Nonzero Value