Abstraction: The extra returns for stock monetary value in January have long been documented in literatures. Numerous surveies have been carried out with the effort to research the root cause of this January anomalousness. In this paper, three classs of accounts, viz. price-pressure hypothesis, omitted hazard factors and dealing costs, mismeasurement jobs, are presented with both supportive and opposing statements from empirical groundss. It can be concluded from the reappraisal of those surveies that the being of January Effect can non be attributed to a individual factor, and the ground for the presence every bit good as continuity of this anomalousness is still a mystifier worth look intoing.
Keywords: January, anomalousness, hypothesis, account
The January consequence, which refers to the phenomenon that stocks exhibit higher returns in January, has been a het issue in the academic field, since it presents regularities in stock returns, seems to afford profitable chances and contradict to the efficient market hypothesis.
With respect to the root cause of the January anomalousness, three classs of accounts are relatively representative and convincing. The first group of account, dwelling of tax-loss merchandising hypothesis, window-dressing hypothesis and so on, focal points on investor behaviour around twelvemonth terminal and emphasizes merchandising or purchasing force per unit area exerted on stock monetary values. The 2nd one assumes that the unnatural returns in January are compensation for underlying hazard, therefore the consequence does non belie to the efficient market hypothesis. The last type of account by and large suggests that January consequence is an unreal consequence due to data-mining and measurement jobs, or can non be exploited in the presence of dealing costs.
Pulling on empirical groundss, a reasonable decision can be made that neither of these accounts are able to warrant the being every bit good as continuity of January anomalousness wholly, since all of them are tested in assorted surveies with conflicting consequences. Consequently, we conclude in the paper that the ground for the being of January consequence is still a mystifier in the academic universe and farther probes are in demand to pull a clearer image sing this well-known anomalousness existed in the fiscal market.
In recent old ages, a turning organic structure of academic literatures has documented the anomalous regularities in security return at bend of the twelvemonth. These anomalousnesss are by and large referred to as the January consequence. Harmonizing to Seyhun ( 1993 ) , the January consequence ( besides called the turn-of-the-year consequence ) can be defined as the phenomenon that stocks exhibit big and positive unnatural returns during the first few hebdomads in January. The first terminal paper brought the January consequence to the attending of modern finance is Rozeff and Kinney ( 1976 ) . They study the being of seasonal forms in an equal-weighted index of New York Stock Exchange monetary values over the period of 1904-1974 utilizing both non-parametric trials ( Kruskal-Wallis trial ( Conover, 1971 ) ) and parametric analysis ( Bartlett ‘s ( 1937 ) trial ) . The statistically important higher mean of return in January found in the trial confirms the being of January consequence.
Explanations for the January anomalousness have invoked het argument in the academic field. The on-going treatment sing the root cause of January consequence falls into three wide classs. First, price-pressure hypothesis efforts to explicate the turn-of-the-year consequence from the position of investor behaviour, such as tax-loss merchandising hypothesis and window-dressing hypothesis. Second, accounts based on hazard factors suggest that January anomalousness does non belie to the efficient market hypothesis since higher returns in January are simply a compensation for higher hazard. The last possible accounts for January consequence are attributed to data-mining, mismeasurment jobs and bid-ask spread bounciness, which suggest that the anomalousness is empirical illusory or can non be exploited as profitable chances. In this paper, I will present each of these accounts and present empirical grounds for and against these statements, while trying to pull a reasonable decision sing the cause of January consequence.
2. Price-pressure hypothesis
2.1 Tax-loss merchandising hypothesis:
The tax-loss merchandising hypothesis is the most controversial account for the January consequence since Branch ( 1977 ) , who asserts that stocks exhibiting high returns in January have experienced diminution in monetary values during the old twelvemonth. Harmonizing to Thaler ( 1987 ) , the tax-loss merchandising hypothesis indicates that shortly before the twelvemonth terminal, investors will sell stocks with diminution in monetary values during the old twelvemonth in order to recognize capital loss for revenue enhancement intent. Then, in January, monetary values of those stocks will bounce in the absence of selling force per unit area.
Numerous surveies have been conducted to turn out the cogency of this hypothesis. Reinganum ( 1983 ) finds grounds to back up the tax-loss merchandising hypothesis and argues that January returns are higher for previous-year & A ; lsquo ; also-rans ‘ and the extra returns can non be observed for little & A ; lsquo ; victors ‘ in the first five trading yearss in January. Schultz ( 1985 ) and Steven et Al. ( 1991 ) besides provide grounds in support of the hypothesis by analysing the seasonal consequence before and after the debut of personal income revenue enhancements in 1917. Ritter ( 1988 ) explains the monetary value motions at bend of the twelvemonth at the degree of single investors and concludes that the return anomalousness is associated with purchasing and selling wonts of “ little ” investors who tend to sell poorly-performed stocks in December to recognize loss and reinvest in a wide scope of small-cap stocks in January with hard currency extract from end-year fillip or capital additions from the sale of large-cap stocks. He besides offers a “ parking the returns ” hypothesis as to why higher returns are mostly concentrated in little house stocks, particularly those who have experienced monetary value lessening in the anterior twelvemonth.
In add-on, Griffiths and White ( 1993 ) warrant the tax-induced hypothesis by analyzing Canadian and U.S. informations with the intent of know aparting between tax-motivated and other year-end effects. Poterba and Weisbenner ( 2001 ) find the alterations in capital additions revenue enhancement regulations have important influence on the January consequence. Chen and Singal ( 2004 ) examine the hypothesis by commanding the consequence of other factors such as window dressing and bid-ask bounciness. Starks et Al. ( 2006 ) usage municipal bond closed-end financess to analyze the cogency of tax-loss merchandising hypothesis because these financess are chiefly held by tax-sensitive single investors. All of these empirical surveies provide grounds in support of the tax-loss merchandising hypothesis. Besides, Hang and Hirschey ( 2006 ) study that tax-loss merchandising by little investors instead than institutional investors contributes to the anomalous form of stock returns at bend of the twelvemonth, which consistent with the survey of D’Mello et Al. ( 2003 ) .
However, harmonizing to Thaler ( 1987 ) , international grounds suggests that revenue enhancement may partly explicate the unnatural returns in January, but it is non the full account. For case, Japan with no capital additions and loss offset exist can besides detect January consequence ( Kato and Schallheim, 1985 ) . Canada had a January consequence before 1972 when capital additions revenue enhancement is still non implemented ( Berges et al. 1984 ) . Great Britain and Australia have the turn-of-the-year consequence although their revenue enhancement old ages begin on April 1 and July 1, severally ( Thaler, 1987 ) .
Despite the fact that overpowering empirical grounds have been presented in support of the tax-loss merchandising hypothesis, opposite statements still exist. Roll ( 1983 ) regards the tax-loss merchandising hypothesis as a obviously absurd and argues that despite the fact that some investors may merchandise for revenue enhancement intent, others would hold already bought the stocks in expectancy of the higher returns in January, taking to the disappear of January consequence. Reinganum ( 1983 ) shows that unnatural returns can besides be observed in little house stocks without prior-year monetary value worsening. Constantinides ( 1984 ) oppose the hypothesis by bespeaking that detaining the loss realisation until December is non an optimum revenue enhancement trading scheme. Jones et Al. ( 1987 ) study that before the infliction of income revenue enhancements, the January consequence had already existed in U.S. Furthermore, Hang and Hirschey ( 2006 ) have proved continuity of January consequence in their trial after the Passage Tax Reform Act of 1986, when all seasonal inclinations associated with tax-motivated merchandising by institutional investors should hold been eliminated at calendar twelvemonth terminal, bespeaking that tax-loss merchandising hypothesis for institutional investors should be rejected. Therefore, empirical groundss sing the tax-loss merchandising hypothesis provide assorted consequences.
2.2 Window-dressing hypothesis:
An option but non needfully reciprocally sole account for the January consequence is the window-dressing hypothesis proposed by Haugen and Lakonishok ( 1988 ) and Lakonishok et Al. ( 1991 ) . Specifically, the hypothesis indicates that portfolio directors are evaluated on both their investing public presentation and consistence of their investing doctrine. They tend to include hazardous and little house stocks in their portfolios to gain higher returns and sell them before the calendar twelvemonth terminal to avoid uncovering them in year-end retention. In January, they would change by reversal this procedure by reinvesting in those little house stocks, typically including some past & A ; lsquo ; also-rans ‘ .
Empirical groundss refering this hypothesis provide conflicting sentiments. Musto ( 1997 ) concludes that the window-dressing hypothesis contributes to the January consequence to some extent because he finds turn-of-the-year consequence can be observed in money market in which instruments do non bring forth capital losingss through revenue enhancement consequence. Maxwell ( 1998 ) asserts that for noninvestment class bonds window dressing is a major factor behind the January anomalousness. Chopra and Ritter ( 1989 ) and Meier and Schaumburg ( 2004 ) besides find grounds consistent with the window-dressing hypothesis. However, Hang and Hirschey ( 2006 ) claim that it is sensible to presume that window dressing by big institutional investors would be a large-cap house phenomenon. Since most unnatural returns in January are concentrated in small-cap houses, window-dressing hypothesis has limited relevancy with the turn-of-the-year consequence.
However, it is notable that both the tax-loss merchandising hypothesis and window-dressing hypothesis rely on year-end merchandising force per unit area and supply same anticipation sing the extra returns at bend of the twelvemonth. Therefore, it is hard to distinguish the influence of the two hypotheses and determine which 1 has derived the January consequence ( Steven et al. , 1991 ) , Chen and Singal ( 2004 ) and Starks et Al. ( 2006 ) ) . Due to this testing job, several surveies have been designed taking to separate the two hypotheses. For illustration, controlled trials conducted by Sias and Starks ( 1997 ) measure these two hypotheses individually and happen tax-loss merchandising hypothesis provides stronger power in explicating January consequence. Furthermore, base on the thought that institutional investors will window dress their portfolio more than one time in a twelvemonth when necessary, Chen and Singal ( 2004 ) analyze the stock return behaviour around semi-annual shutting when intervention of tax-motivated merchandising is eliminated. Both trials tend to reject the cogency of window-dressing hypothesis. Hence, empirical groundss are inclined to back up revenue enhancement hypothesis instead than window-dressing hypothesis.
2.3 Differential information hypothesis
In conformity with Chen and Singal ( 2004 ) , the extra returns in January can be attributed to the influence of important information releases at bend of the twelvemonth. There are different versions of information hypothesis relation to the unnatural returns in January.
Barry and Brown ( 1984 ) assert that deficiency of information is by and large considered a higher hazard index, ensuing in the disposition of investors to see those houses with less information as bearing higher hazard than those with relatively equal information although the beta hazard of both types of companies are equal. Due to the fact that returns exhibit entire hazard involved in the stocks whereas plus pricing theoretical account simply monetary value systematic hazard entirely, the extra return for less-information houses may be treated as unnatural returns. For proving this hypothesis, they use naming period as a placeholder for obtainability of information and happen differential production of information among houses can partly explicate the January consequence.
Merton ( 1987 ) proposes the investor acknowledgment hypothesis as another manner of construing the information hypothesis. Specifically, since January is the common period for houses to print their information, investors would be given to purchase the stocks when they become more cognizant of the companies through those freshly published information, taking to ascertained January consequence.
Chen and Singal ( 2004 ) test the differential information hypothesis based on stock returns and turnover. To be specific, if the hypothesis holds, non merely in January but besides in other months should higher returns for little house stocks be observed when listed houses submit their accounting information quarterly in conformity with the demand of Securities Act of 1933. Furthermore, less trading volume in December is expected since investors would wait until the publication of new information. However, consequences from both facets nem con suggest that information hypothesis can non be the primary driver of January consequence.
2.4 Turn-of-month liquidness hypothesis
Ogden ( 1990 ) argues that the January consequence in U.S. market can be, at least in portion, due to the standardisation in payment system, which leads to a concentration of hard currency flow at bend of the twelvemonth. Specifically, significant hard currency grosss at bend of the twelvemonth enable investors to put in a broad scope of stocks ensuing in higher demand of stocks and a rush in stock returns in January. He tests the deduction of this hypothesis based on value-weighted and equally-weighted stock index returns and concludes that this hypothesis can be a partial account for the January consequence.
3. Risk-based accounts
The Risk-based accounts suggest that higher hazard is the root cause of unnatural returns at bend of the twelvemonth. For illustration, Chan et Al. ( 1985 ) use a multi-factor pricing equation as an account of the house size consequence and concludes that the size anomalousness can be justified by extra hazard factors involved in small-cap houses.
Base on the same thought, Rogalski and Tinic ( 1986 ) effort to explicate the turn-of-the-year consequence in footings of hazard. Specifically, they argue that old surveies by and large assume that systematic hazards and equilibrium rate required on stocks returns remains changeless over calendar months. However, this premise is neither required by equilibrium plus pricing theoretical accounts in theory nor consistent with stock monetary value motions in pattern. As a consequence, they correct this job by analyzing nonstationarities in entire, systematic and diversifiable hazards of common stocks over calendar months and claim that a marked addition of little house beta is observed at the beginning of the twelvemonth ; hence higher returns in January are risk premium but non anomalousnesss.
However, a contrary decision is drawn by Seyhun ( 1993 ) . He tests the hypothesis that omitted hazard factors contribute to the big January returns, using a stochastic laterality attack whose consequences do non depend on risk-return trade-offs generated by plus pricing theoretical accounts. Findingss in the paper suggest that extra returns in January are excessively high to be equilibrium returns, therefore omitted hazard factors in explicating January returns are less plausible. Furthermore, Keim ( 1983 ) opposes the hazard hypothesis advanced by Brown, Kleidon and Marsh by reasoning that even if those unspecified hazard factors are able to explicate portion of the size consequence, they are deficient to warrant those January behaviours since the return premium is observed in the same month each twelvemonth.
4. Bias, dealing costs and mismeasurment jobs
4.1 Bid-ask spread bounciness
The bid-ask spread bounciness documented by Keim ( 1989 ) suggests that Investors tend to merchandise at the command in late December and at ask in early January, therefore the bid-ask bounciness gives an feeling of big positive January returns, when the fact to be non the instance.
Base on this premise, Blume and Stambaugh ( 1983 ) study that stock returns computed by shuting monetary values are imparted an upward prejudice, chiefly due to bid-ask consequence. Bhardwaj and Brooks ( 1992 ) demonstrate that the bend of the twelvemonth switch from commands to inquire leads to a positive bid-ask prejudice in deliberate returns estimated at about 1 % and this prejudice is big and important for low-price stocks. Hence, the positive bid-ask prejudice in returns has overestimated the January anomalousness. Griffiths and White ( 1993 ) besides conclude that a systematic displacement from seller-initiated minutess at command monetary value to buyer-initiated trades at ask monetary value at bend of the twelvemonth drive the outgrowth of January consequence.
Due to this concern, specific steps have been taken in many empirical surveies in order to command the consequence of bid-ask prejudice when analysing the January consequence. For illustration, Cox and Johnston ( 1998 ) correct the bid-ask bounciness by excepting low-price stocks while Chen and Singal ( 2005 ) use center of the bid-ask monetary value for ciphering returns. Most of these surveies suggest that the bid-ask spread bounciness may upwards bias the extra returns in January but is non the full account.
4.2 Transaction costs
Transaction costs contain committee fee, bid-ask spread and the market impact of the dealing ( Seuhun, 1993 ) . A sensible inquiry would be if dealing costs are responsible for the being of January consequence since such costs prevent arbitrageurs from working anomalous forms and extinguishing monetary value inefficiency.
Empirical trials are inclined to believe that dealing costs eliminate profitable chances from January anomalousness, ensuing in the continuity of unnatural returns. Bhardwaj and Brooks ( 1992 ) examine returns after dealing costs on stocks within different monetary value scopes, taking to find whether typical investors are able to obtain extra returns. The consequence of the survey is time-dependant. For the first ten-year period of 1967-1976, low-price stocks outperform their opposite numbers after sing dealing cost. However, in the undermentioned 10 old ages, a reversal is observed when high-price stocks dominate low-price stocks. Therefore, economically and statistically important extra returns can non be earned by typical investors from merchandising about January. This determination is consistent with the statement presented by Haugen and Jorion ( 1996 ) , viz. turn-of-the-year consequence is non a manifestation of market inefficiency since the unnatural returns are non exploitable for typical investors. Seyhun ( 1993 ) argues that investors with dealing costs lower than 300-basis-points, typically big investors, are able to gain from scheme derived from January consequence while for those who bear dealing costs higher than 500 footing points, a bargain and clasp scheme would be more ideal.
Hence, empirical groundss by and large suggest that dealing costs prevent typical investors gaining from those anomalousnesss and contribute to the continuity of January consequence to some extent.
4.3 Data-mining and mismeasurment jobs
Lokonishok and Smidt ( 1988 ) present three generic considerations sing the dependability of January consequence grounds. The first 1 is called ennui factor, which stresses the danger that academic field attaches undue significance to surveies documenting anomalousnesss due to a choice prejudice ( Merton, 1985 ) . Then, noise in security returns emphasizes troubles in separating anomalousnesss with noise in the market ( Fischer Black, 1986 ) . The last consideration is called informations snooping, which represents the act of utilizing the same information to detect or prove hypothesis in different surveies. All of these jobs may take to colored consequences for stock return forms.
After taking these factors into history, Lokonishok and Smidt ( 1988 ) still find grounds for the continuity of anomalous returns at bend of the twelvemonth. This consequence strongly refutes the hypothesis that seasonal anomalousnesss are merchandise of trying mistake and information excavation ( Lokonishok and Smidt, 1988 ) . Gultekin and Gultekin ( 1983 ) besides proves that January consequence is non simply a statistical artefact. However, Steven et Al. ( 1991 ) turn out in their survey that different methods of calculating returns may take to different consequences of empirical trials. Therefore, measurement jobs may impact the survey consequences of January anomalousness to some extent, but are non primary accounts.
To reason, it can be seen from empirical groundss that accounts refering the being of January consequence are assorted consequences. Price-pressure hypotheses based on investor behaviour can afford partial account to the January anomalousness but surveies barely reach a consensus. Risk-based accounts are reasonable per se since most January effects are little house phenomenon and little houses are by and large considered to bear higher hazard. However empirical trials can non pull a definite decision for this account either. The last class of account by and large suggests that anomalousnesss do non be in the first topographic point. They are unreal consequences due to data-mining and bid-ask prejudice or can non be exploited by arbitragers due to dealing costs. In this sense, it is improbable to show a individual account with respect to the January anomalousness ( Lokonishok and Smidt, 1988 ) . More surveies are expected to decide the mystifier in the hereafter.