An empirical test of the relationship between gains/losses on disposal of discontinued operations arising from spin-offs and stock prices based on the extended functional fixation hypothesis

Abstract

Under the Korean equivalents of International Financial Reporting Standards (K-IFRS), which was adopted in 2011, companies are allowed to recognize gains/losses on disposal of discontinued operations (GDDO) when they spin off part of their operations. GDDO reflects the gap between the fair value and the book value of the spun-off operations. However, the fair value is not based on the fair value of the assets classified as held for sale but based on the valuation of the operation by using the Free Cash Flow to the Firm (FCFF) method. Since the FCFF method estimates the value of the operation by discounting all future free cash flows from the operation, it is not accurate but varies remarkably with a small change in the discount factor. Furthermore, only spun-off operations are evaluated, but sustained operations are not. Lastly, shareholders of the parent company receive equivalent shares in the new company in order to compensate for the loss of equity in the original stocks. This implies that there should be no gains/losses from the division of the company unless there is a huge quantifiable impact from spinning off. For these reasons, it is improper to see GDDO as a real gain; rather, it is proper to see it as a paper gain. According to the extended functional fixation view, unsophisticated investors improperly implement a wrong analysis of such complicated accounting information into their investment decision making. Herein I attempt to test whether unsophisticated investors are affected by GDDO with an assumption that there are some unsophisticated marginal investors who see GDDO as a real gain and they impound this information in price. My hypothesis is tested by examining the stock price reaction to quarterly earnings announcements of firms which have undergone spin-offs since K-IFRS was adopted. On the financial performance announcement date, the firm's stock price is theoretically affected by the difference between the real earnings and the expected earnings (earnings surprise: ES). In a case in which the firm announced earnings including GDDO, only the real earnings, excluding GDDO, less the expected earnings should be reflected in the stock price movement-if all investors are sophisticated enough not to regard GDDO as real earnings. However, I test whether GDDO also affects the stock price movement in this situation separately from the impact from the earnings surprise, which implies that there are unsophisticated investors and the extended functional fixation hypothesis holds in the GDDO case. Eleven Korean companies that are traded on the Korean stock market and listed on the Korea Composite Stock Price Index (KOSPI) were subjected to the test. These companies meet the following constraints: (1) They have undergone spin-offs, and (2) They recognized GDDO on their income statement but do not directly add or deduct the amount from equity items on the balance sheet. I run a regression test on the excess stock return which is calculated by using historical two-day return, historical beta, the impact from earning surprise (ES), and the impact from GDDO. The result shows higher R² and lower p-value for the case when both the ES and GDDO are chosen as x-variables than the case when the earnings surprise is solely chosen as an x-variable. The p-values for x-variables in both cases were smaller than 0.01. The test result implies that the extended functional fixation hypothesis might hold in this case. I conclude that including GDDO as an item on the income statement can affect some investors' decision making and demand for the stock.