This study examined whether asset revaluation accounting earnings, when provided supplementarily to historical cost accounting earnings, possess incremental information content, combined information content, and interaction information content, by investigating their relationship with stock returns. Stock returns were measured over two different test periods. The first period, during which the information content of accounting earnings was presumed to be reflected in stock prices after financial statement disclosure, ran from April of each year to March of the following year, covering 82 months over seven instances during the study period. The second period, during which the information content of accounting earnings was assumed to be reflected in stock prices prior to financial statement disclosure, ran from January to December of each year, covering 71 months over six instances during the study period. Six experimental portfolios and one control portfolio were constructed, and after controlling for other variables by setting the market beta of all firms within each portfolio to one, it was statistically tested whether significant differences existed among the monthly average returns of the seven portfolios. The test results showed that during both periods, no evidence was found that asset revaluation accounting earnings possessed incremental information content. No evidence of combined information content or interaction information content was obtained either. These results imply that asset revaluation accounting earnings are not relevant to investors' decision-making in Korea's capital market. In other words, in the case of Korea's capital market, it can be inferred that there is a greater tendency to rely on other information sources rather than asset revaluation accounting earnings information when estimating a firm's future cash flows.