Tax-Loss Selling and the January Effect: Evidence from Municipal Bond Closed-End Funds

This paper provides direct evidence supporting the tax‐loss selling hypothesis as an explanation of the January effect. Examining turn‐of‐the‐year return and volume patterns for municipal bond closed‐end funds, which are held mostly by tax‐sensitive individual investors, we document a January effect for these funds, but not for their underlying assets. We provide evidence that this effect can be largely explained by tax‐loss selling activities at the previous year‐end. Moreover, we find that funds associated with brokerage firms display more tax‐loss selling behavior, suggesting that tax counseling plays a role.     

Institutional Ownership and Return Reversals Following Short‐Term Return Consistency

Securities with consistently strong positive (negative) returns during the previous two weeks have future returns that are higher (lower) than those that do not. The results hold for various robustness checks, including those involving firm size, share turnover, past return levels, and bid‐ask bounce. The returns to short horizon consistency trading strategies are reliable through time and are both economically and statistically significant. There is also some evidence that longer periods of consistency lead to greater risk‐adjusted profits. Most surprising is that this effect holds only for those firms with high institutional ownership.     

Retail investors exonerated: the case of the January effect

We dispel the belief that the January effect is due to retail investor trading. Previous studies suggest that retail investors, affected by behavioural biases and disproportionally invested in small capitalization stocks, are the source of the January effect. Furthermore, the literature regards retail investor trading and the tax‐loss selling hypothesis as essentially the same explanation. We separate tax implications and market capitalization to show that retail traders are not the cause of the January effect. Our study is an important direct test of whether retail trading causes market anomalies.     

Tax Calendar Effects in the Municipal Bond Market: Tax‐Loss Selling and Cherry Picking by Investors and Market Timing by Fund Managers

Examining municipal bond returns, bond fund flows and buying activities by fund managers over the period 1990–2009, we find evidence of tax calendar‐related rational opportunistic trading patterns by fund investors and fund managers. Specifically, fund shareholders conduct tax‐loss selling in December and re‐invest in January. In April, June, and September, fund investors rationally cherry pick to sell their shares of short‐term bond funds instead of their shares of long‐term bond funds to raise cash to pay estimated taxes. Unlike fund shareholders, fund managers adopt a contrarian strategy of buying in December and selling in January.     

Price Movement Effects on the State of the Electronic Limit‐Order Book

This paper investigates public‐trader order‐placement strategies by examining the relations between the state of the limit‐order book and previous price movements. There is support for an information effect, as traders become more aggressive in buying and more patient in selling after previous positive stock returns. The widening of the bid‐ask spread also causes traders to place less aggressive orders. However, there is no evidence of the options effect on limit‐order trading. This study also reveals that orders at the best quotes react faster and complete the adjustment earlier than orders that are far away from the best quotes.     

Intraday Patterns in the Cross‐section of Stock Returns

Motivated by the literature on investment flows and optimal trading, we examine intraday predictability in the cross‐section of stock returns. We find a striking pattern of return continuation at half‐hour intervals that are exact multiples of a trading day, and this effect lasts for at least 40 trading days. Volume, order imbalance, volatility, and bid‐ask spreads exhibit similar patterns, but do not explain the return patterns. We also show that short‐term return reversal is driven by temporary liquidity imbalances lasting less than an hour and bid‐ask bounce. Timing trades can reduce execution costs by the equivalent of the effective spread.     

DOES VOLATILITY DECREASE AFTER REVERSE STOCK SPLITS?

Previous research documents that volatility decreases after reverse stock splits. I show that measurement effects bias observed volatility, especially for lower priced stocks. Based on observed returns, volatility decreases 25% after reverse splits. Controlling for bid–ask bounce, volatility still decreases for stocks with prices above $5.00. However, for stocks below $2.00, volatility increases slightly. The portion of observed volatility attributable to measurement effects declines as the stock price increases and as the minimum tick size decreases. Finally, there is a significant and positive cross‐sectional relation between changes in the number of trades and changes in volatility after reverse splits.     

Against the tide: the commencement of short selling and margin trading in mainland China

China's recent removal of short‐selling and margin trading bans on selected stocks enables testing of the relative effect of margin trading and short selling. We find the prices of the shortable stocks decrease, on average, relative to peer A‐shares and cross‐listed H‐shares, suggesting that short selling dominates margin trading effects. Contrary to the regulators' intention and recent developed market empirical evidence, liquidity declines and bid‐ask spreads increase in these shortable stocks. Consistent with Ausubel (1990), these results imply that uninformed investors avoid the shortable stocks to reduce the risk of trading with informed investors.     

Effect of the ban on short selling on market prices and volatility

We examine the effects of the short‐selling ban, imposed by Australian regulators in the wake of the global financial crisis, on the trading of financial stocks. Our findings argue against commonly stated reasons for imposing short‐sale bans. We find no evidence that short‐sale restrictions provide support for stock prices or that they reduce volatility. Moreover, stocks subject to the short‐selling ban suffered a severe degradation in market quality. Controlling for the adverse effects of the financial crisis on markets, we show that short‐selling restrictions increase intraday volatility, reduce trading activity and increase bid–ask spreads.     

Explaining Monday Returns

The Monday effect is reexamined using two stock indexes and a sample of 452 individual stocks that trade on the London Stock Exchange. The results based on conventional test methods reveal a negative average return on Monday. Extending the analysis to examine the effects of various possible influences simultaneously, the average Monday return becomes positive and does not differ significantly from the average returns of most other days of the week. Fortnight, ex‐dividend day, account period, (bad) news flow, trading activity, and bid‐ask spread effects are all controlled for. The results broadly support the trading time hypothesis.     

Can Tax-Loss Selling Explain the January Seasonal in Stock Returns?

This paper analyzes the tax-loss selling hypothesis as an explanation of the January seasonal in stock returns and argues that rational tax-loss selling implies little relation between the January seasonal and the long-term loss. Empirical results show that the January seasonal is as strongly related to the long-term loss as it is to the short-term loss. The evidence is inconsistent with a model that explains the January seasonal by optimal tax trading.     

The Choice and Role of Lockups in IPOs: Evidence from Heterogeneous Lockup Agreements

This paper analyses heterogeneous lockup agreements from the London Stock Market. With hand‐collected data, I compare and contrast absolute‐date lockups with the relative‐date lockups and single lockups versus staggered lockups. This paper tests several potential explanations for the choice of lockup contracts: (i) information asymmetry, (ii) signaling, (iii) agency problem, and (iv) certification. I find strong evidence for information asymmetry and certification (VC and prestigious underwriters) and partial support for agency explanation for the choice of lockups. The insider selling activity and lockup expiration returns are also consistent with asymmetric information, certification and agency hypothesis.     

Information Disclosure and Stock Liquidity: Evidence from Borsa Italiana

This article examines the effect of increased corporate information disclosure on stock liquidity. Using the adoption of International Financial Reporting Standards (IFRS) in Italy as a natural experiment we extend previous work examining the effect on one measure of liquidity—bid‐ask spreads—to others, specifically depth and the price impact of transactions (or effective bid‐ask spreads). Consistent with previous research we find that bid‐ask spreads of stocks decline following the introduction of IFRS, which implies that stock liquidity increases for small traders. However, we also provide evidence that depth at the best quotes declines, which challenges the proposition that liquidity increases for large trades following an increase in disclosure. In additional tests, we find that effective bid‐ask spreads of block trades also decline following the introduction of IFRS. Overall, this evidence confirms that stock liquidity for both small and large trades increases following an increase in corporate information disclosure.     

The Puzzling Increase in the Underpricing of Seasoned Equity Offerings

Using a sample of over 3,000 seasoned equity offerings (SEOs) from 1983 to 1998, we test the hypothesis that the U.S. Securities and Exchange Commission's Rule 10b‐21, which disallows the covering of short positions with newly issued SEOs, makes pre‐offer stock prices less informative, which, in turn, causes the new seasoned equity to be priced at a discount. Consistent with the hypothesis, we find that the year the rule went into effect coincides with the year from which we begin observing significant SEO discounts. Further, we find that ex ante uncertainty and SEO discounts are positively related. We also conduct tests specifically related to short selling, and we also consider an exhaustive set of alternative explanations for the discounts. Based on all of the evidence, we conclude that it is the rule that makes issue discounts larger in the 1990s.     

Credit Contagion Channels: Market Microstructure Evidence from Lehman Brothers’ Bankruptcy

Immediately after Lehman Brothers’ bankruptcy, many firms disclosed their financial exposure (or lack thereof) to Lehman. This offers a clean setting to test two credit contagion channels through which a financial firm's bankruptcy can affect other firms—“counterparty risk” and “information transmission” channels. Using market microstructure variables to measure the various dimensions of contagion effects, we provide robust evidence supporting the significance of counterparty risk. Firms with exposure to Lehman suffered more severe negative effects—wider bid‐ask spread, higher price impact, greater information asymmetry, and greater selling pressure—than unexposed firms. We find mixed evidence regarding the information transmission hypothesis.     

Stock return seasonalities and the tax-loss selling hypothesis: Analysis of the arguments and Australian evidence

A ‘tax-loss selling’ hypothesis has frequently been advanced to explain the ‘January effect’ reported in this issue by Keim. This paper concludes that U.S. tax laws do not unambiguously predict such an effect. Since Australia has similar tax laws but a July–June tax year, the hypothesis predicts a small-firm July premium. Australian returns show pronounced December–January and July–August seasonals, and a premium for the smallest-firm decile of about four percent per month across all months. This contrasts with the U.S. data in which the small-firm premium is concentrated in January. We conclude that the relation between the U.S. tax year and the January seasonal may be more correlation than causation.     

DAY‐END EFFECT ON THE PARIS BOURSE

We study the day‐end effect on the Paris Bourse, a computerized order‐driven market with competing dealers. The day‐end return is approximately double the magnitude found in U.S. data and is nearly four times larger for stocks trading with a registered dealer. However, this is largely explained by the time between trades and the bid‐ask spread. Unlike the U.S. data, the effect does not decline as stock price increases, probably because of a variable tick size in the Paris market. Finally, a change to a closing call auction in May 1996 for a subset of stocks did not reduce the day‐end effect.     

Pricing Efficiency in a Thin Market with Competitive Market Makers: Box Spread Strategies in the Hang Seng Index Options Market

Using a box spread arbitrage strategy, we examine the pricing efficiency of the emerging, thinly traded Hang Seng Index options market in Hong Kong, where market makers operate under a competitive open outcry system. In 20 months of tick‐by‐tick bid‐ask and transaction quotes we find very few arbitrage opportunities. Our examination of the reporting time of quotes shows that in effect, all the apparent mispricings are deceptive and could be explained by stale quotes. The absence of real arbitrage opportunities supports the pricing rationality hypothesis in the Hong Kong options market.     

The Post-Reform Bid-Ask Spread Disparity between Nasdaq and the NYSE

We report further evidence of the difference in execution costs between Nasdaq and the NYSE before and after the 1997 market reforms. We find that informed trading costs are consistently higher on Nasdaq both before and after the reforms. In the pre‐reform period the Nasdaq‐NYSE disparity in bid‐ask spreads cannot be completely attributed to the difference in informed trading costs. However, in the post‐reform period the spread difference between these two markets becomes insignificant with the effect of informed trading costs controlled. Our findings are consistent with the contention that the reforms have largely reduced noninformation trading costs and dealers' rents.