A December Effect with Tax-Gain Selling?

We present evidence on the December effect. When investors do not sell winner stocks in December but postpone their sale to January so that capital gains will not be realized in the current fiscal year, the “winners” appreciate in December. The December effect is relatively easy to arbitrage. We also present evidence regarding the persistence of the January effect and note that the January effect continues because it is difficult to exploit profitably. The January effect is well known and largely explained by tax-loss selling. Tax-gain selling, however, is another activity driven by the tax code and has not received attention. If investors realize capital losses to offset capital gains, a natural strategy for investors is to postpone realization of capital gains so that they can postpone paying taxes on capital gains. Thus, rational investors will sell winner stocks in January instead of December. The selling pressure on “winners” should be small in December, causing the price of winners to rise.Indeed, we found a return of about 2.2 percent in the last five days of December (excluding the last trading day of the year) in the 1988–2000 period for winners, compared with a loss of 0.9 percent in the first five days of the subsequent January. (In our study, winners were those with the smallest price decline from a previously defined high price; losers were stocks with the largest price decline from the high price.) The higher five-day December return was accompanied by significantly lower volume turnover, whereas the five-day January return was accompanied by higher volume. The return and turnover patterns are consistent with tax-gain selling. The January and December effects, in turn, are consistent with the accepted hypothesis explaining turn-of-the-year trading.We analyze the results of simple trading strategies to exploit the December and January effects and discuss the persistence of the effects. The December effect is relatively easy to arbitrage because the winners are usually large-market-capitalization stocks that are liquid and have low transaction costs. A natural proxy for the large winners is the S&P 500 Index. Therefore, a simple, testable trading strategy is to buy S&P futures contracts or SPDRs (Standard & Poor's Depositary Receipts), which serve as an exchange-traded S&P 500 fund, at the close on the seventh trading day before year-end and to unwind the position at the close on the second trading day before year-end. Over the 1988–2001 period, trading SPDRs in this manner would have generated an average return of 1.88 percent. After transaction costs, the net return would have been at least 1.5 percent. Similarly, trading S&P futures would have generated a return of 2.1 percent before transaction costs.With such high realizable returns, a natural question is: Why hasn't the December effect been arbitraged away? The most likely explanation is that the December effect is not well known; it has not been previously documented. The effect is most likely to disappear as soon as arbitrageurs enter the market.The January effect is difficult to exploit profitably. Although losers can command as much as a 10 percent return in the first five days of January, this effect cannot be captured because, unlike strategies for capturing the December effect, low-transaction-cost instruments are not available for trading small-cap stocks. The January effect is concentrated among stocks in the bottom 30 percent of all stocks by market cap. For example, in 2000, the losers had a mean market cap of $108 million, a low median market cap of $26 million, and a median price of only $1.14. Such stocks have high transaction costs, are illiquid, and do not have listed options. Alternative financial instruments are inappropriate because these stocks do not carry a significant weight in any of the tradable indexes (or even in any of the so-called microcap mutual funds). The January effect persists because it cannot be arbitraged.Investors could take advantage of the January effect, however, by altering their trading behavior to hold onto their small-cap losers until the middle of the following January. Such a strategy has a high probability of capturing an extra 10 percent return.