This study investigates whether market learning explains the absence of the accrual anomaly in recent years by examining three conditions associated with the presence of the anomaly in prior research: (i) a differential relation between future earnings and cash flows versus accruals, (ii) incorrect weighting of cash flows and accruals by investors when predicting earnings, and (iii) association of earnings forecast errors with returns. All of these conditions are widely documented in the anomaly period. In the no-anomaly period, I continue to find a differential relation of cash flows and accruals with future earnings. However, investors appear to correctly weight accruals and cash flows in their earnings predictions implicit in beginning-of-year security prices, consistent with learning. This study also investigates whether improvements in analyst forecasts contribute to investor learning and the absence of the anomaly. The association between analyst optimism and accruals is weaker in the no-anomaly period, but is still statistically significant. Furthermore, the anomaly ended simultaneously for firms followed by analysts and for non-followed firms, suggesting that improvements in analyst forecasts alone cannot account for improved market efficiency with respect to accruals. The results suggest that the anomaly was similar for firms held by institutional investors and for firms with no institutional holdings before the discovery of the anomaly while the anomaly ended sooner for held firms than for non-held firms after the discovery of the anomaly, consistent with the conjecture that arbitrage by institutional investors reduce the anomaly. Overall, the findings are consistent with market learning and suggest that improvement in investors' interpretation of accruals after the discovery of the anomaly explains the end of the anomaly. This improvement in investor learning is not due to changes in analysts' forecasting behavior, however.

A presente pesquisa foi desenvolvida com o objetivo de avaliar a capacidade dos investidores em interpretar os dados emanados pela Contabilidade; mais especificamente, analisou sua habilidade em compreender informações relativas ao lucro. De forma complementar, buscou analisar a existência de oportunidades de obtenção de ganhos econômicos por intermédio da adoção de estratégias de investimento com base em informações relativas aos accruals. A amostra compreendeu empresas não financeiras para as quais o banco de dados Economática dispunha de informações relativas ao período de 1995 a 2010. Foram descartadas da amostra as empresas com patrimônio líquido negativo, companhias com dados faltantes (missings), bem como observações com comportamento distinto dos demais (outliers). Por conta do baixo impacto dos números contábeis no mercado de capitais brasileiro detectado por Lopes (2005) esperava-se uma baixa presença da anomalia dos accruals no mercado de capitais brasileiro. Isso porque, países em que a importância dos lucros para os preços de mercado é reduzida, a precificação de ações seria menos influenciada pela fixação funcional no lucro final reportado o que, por sua vez, provocaria interferências na anomalia dos accruals (EL MEHDI, 2011). Para avaliar se a persistência dos componentes dos accruals era significativamente inferior aos componentes de fluxos de caixa, estimou-se uma regressão com dados em painel, na qual foi possível comprovar a hipótese de que os ajustes do regime de competência exibem uma menor persistência, com um parâmetro padronizado e estatisticamente significativo na regressão estimada de 0,43, enquanto os componentes de fluxos de caixa apresentaram um parâmetro de 0,53. A falta de significância estatística entre os accruals correntes e retornos anormais futuros das companhias estudadas, bem como, da ausência de retornos anormais significativos de estratégias baseadas em accruals demonstraram que uma baixa qualidade dos lucros correntes - devido a um alto nível de accruals - não resultou em retornos anormais negativos no período posterior. As proxies relativas a adoção das normas IFRS (International Financial Reporting Standards) e ao nível de investimentos - incluídas no modelo de regressão - compreendem parte das contribuições deste tralho, ainda que não se tenha identificado significância estatística para tais variáveis. Isso porque, por intermédio do teste-t, foi explicitada a ocorrência de uma relação entre o nível de acrruals e o crescimento do imobilizado. Tal resultado sugere indícios de que ambas as variáveis captariam o mesmo efeito, qual seja, a atividade investimento por parte das firmas (WEI; XIE, 2007; ZACH, 2007). Os resultados coadunam com as evidências detectadas por Cupertino (2010), ampliando os indícios sobre o comportamento do mercado frente a informações emanadas pela Contabilidade em mercados emergentes, além de explicitar a ausência da denominada anomalia dos accruals no mercado de capitais brasileiro. / This research was developed to evaluate investors\' ability to interpret Accounting data, more specifically, to examine its ability to effectively understand earnings information. As a complement, we have analyzed the existence of economic opportunities to obtain abnormal returns through investment strategies based on accruals. The sample was composed by nonfinancial companies with available information in Economatica database from 1995 to 2010. We\'ve excluded firms with negative equity, missing data, as well as outliers. In countries in which profits importance to market price is lower, pricing of shares would be less influenced by the bottom line functional attachment, which in turn, would decrease the Accruals Anomaly (El MEHDI, 2011). Since Accounting numbers in Brazilian stock market have demonstrated low impact (LOPES, 2005) we previously expect a lower presence of the Accrual Anomaly. To assess whether persistence of accruals was significantly lower than cash flow component, we\'ve estimated a panel data regression, in which it was possible to prove our first hypothesis, that accrual\'s exhibit a lower persistence with a 0.43 estimated parameter, while the cash flows have presented a 0.53 parameter, both significantly different from 0 at the 0.05 level. The lack of statistical significance between current accruals and future abnormal returns among studied companies and the absence of significant abnormal returns in strategies based on accruals have demonstrated that a low quality of current earnings - due to a high level of accruals - did not result in a negative abnormal return, thereafter. Adding proxies to IFRS adoption and investment level can be considered as an additional contribution. Although these variables have shown no statically significance, we\'ve found a relationship, explicit by T-test, between accruals level and inventory growth, providing evidences that both variables would capture the same effect, namely, investments activity by firms (WEIK; XIE, 2007; ZACH, 2007). The results are consistent with Cupertino (2010) research, have increased evidences about market behavior to Accounting information in emerging markets, and explicit the absence of the Accrual Anomaly in Brazilian stock market.

A presente pesquisa foi desenvolvida com o objetivo de avaliar a capacidade dos investidores em interpretar os dados emanados pela Contabilidade; mais especificamente, analisou sua habilidade em compreender informações relativas ao lucro. De forma complementar, buscou analisar a existência de oportunidades de obtenção de ganhos econômicos por intermédio da adoção de estratégias de investimento com base em informações relativas aos accruals. A amostra compreendeu empresas não financeiras para as quais o banco de dados Economática dispunha de informações relativas ao período de 1995 a 2010. Foram descartadas da amostra as empresas com patrimônio líquido negativo, companhias com dados faltantes (missings), bem como observações com comportamento distinto dos demais (outliers). Por conta do baixo impacto dos números contábeis no mercado de capitais brasileiro detectado por Lopes (2005) esperava-se uma baixa presença da anomalia dos accruals no mercado de capitais brasileiro. Isso porque, países em que a importância dos lucros para os preços de mercado é reduzida, a precificação de ações seria menos influenciada pela fixação funcional no lucro final reportado o que, por sua vez, provocaria interferências na anomalia dos accruals (EL MEHDI, 2011). Para avaliar se a persistência dos componentes dos accruals era significativamente inferior aos componentes de fluxos de caixa, estimou-se uma regressão com dados em painel, na qual foi possível comprovar a hipótese de que os ajustes do regime de competência exibem uma menor persistência, com um parâmetro padronizado e estatisticamente significativo na regressão estimada de 0,43, enquanto os componentes de fluxos de caixa apresentaram um parâmetro de 0,53. A falta de significância estatística entre os accruals correntes e retornos anormais futuros das companhias estudadas, bem como, da ausência de retornos anormais significativos de estratégias baseadas em accruals demonstraram que uma baixa qualidade dos lucros correntes - devido a um alto nível de accruals - não resultou em retornos anormais negativos no período posterior. As proxies relativas a adoção das normas IFRS (International Financial Reporting Standards) e ao nível de investimentos - incluídas no modelo de regressão - compreendem parte das contribuições deste tralho, ainda que não se tenha identificado significância estatística para tais variáveis. Isso porque, por intermédio do teste-t, foi explicitada a ocorrência de uma relação entre o nível de acrruals e o crescimento do imobilizado. Tal resultado sugere indícios de que ambas as variáveis captariam o mesmo efeito, qual seja, a atividade investimento por parte das firmas (WEI; XIE, 2007; ZACH, 2007). Os resultados coadunam com as evidências detectadas por Cupertino (2010), ampliando os indícios sobre o comportamento do mercado frente a informações emanadas pela Contabilidade em mercados emergentes, além de explicitar a ausência da denominada anomalia dos accruals no mercado de capitais brasileiro. / This research was developed to evaluate investors\' ability to interpret Accounting data, more specifically, to examine its ability to effectively understand earnings information. As a complement, we have analyzed the existence of economic opportunities to obtain abnormal returns through investment strategies based on accruals. The sample was composed by nonfinancial companies with available information in Economatica database from 1995 to 2010. We\'ve excluded firms with negative equity, missing data, as well as outliers. In countries in which profits importance to market price is lower, pricing of shares would be less influenced by the bottom line functional attachment, which in turn, would decrease the Accruals Anomaly (El MEHDI, 2011). Since Accounting numbers in Brazilian stock market have demonstrated low impact (LOPES, 2005) we previously expect a lower presence of the Accrual Anomaly. To assess whether persistence of accruals was significantly lower than cash flow component, we\'ve estimated a panel data regression, in which it was possible to prove our first hypothesis, that accrual\'s exhibit a lower persistence with a 0.43 estimated parameter, while the cash flows have presented a 0.53 parameter, both significantly different from 0 at the 0.05 level. The lack of statistical significance between current accruals and future abnormal returns among studied companies and the absence of significant abnormal returns in strategies based on accruals have demonstrated that a low quality of current earnings - due to a high level of accruals - did not result in a negative abnormal return, thereafter. Adding proxies to IFRS adoption and investment level can be considered as an additional contribution. Although these variables have shown no statically significance, we\'ve found a relationship, explicit by T-test, between accruals level and inventory growth, providing evidences that both variables would capture the same effect, namely, investments activity by firms (WEIK; XIE, 2007; ZACH, 2007). The results are consistent with Cupertino (2010) research, have increased evidences about market behavior to Accounting information in emerging markets, and explicit the absence of the Accrual Anomaly in Brazilian stock market.

Esta pesquisa buscou investigar a anomalia dos accruals nos retornos de cinco países da América Latina (Argentina, Brasil, Chile, México e Peru), bem como observar se os aspectos institucionais impactaram nos retornos anormais de accruals dos países. Para tanto, foram analisadas séries temporais para cada país, agrupadas em carteiras de accruals e carteiras de setor, para os modelos CAPM, 3-fatores e 4-fatores, incluindo o fator accruals e observados os retornos anormais das séries, avaliando-os por meio da estatística GRS e pelas estatísticas de teste do procedimento de Fama e MacBeth. Em seguida, foram construídos modelos de dados em painel para o período de 2004 a 2017, em que as variáveis dependentes foram os retornos anormais de accruals, resultantes das séries temporais mensais de cada uma das empresas, e as variáveis independentes foram os fatores institucionais característicos dos países e as variáveis a nível de empresa. O estudo esperava que, se confirmados os retornos anormais de accruals nas séries temporais da América Latina, os fatores específicos de cada país contribuiriam para explicar as diferenças de retornos anormais existentes entre eles. Os resultados confirmaram a presença de retornos anormais em todos os países da amostra, em todos os modelos e o comportamento dos accruals explicou parte dos retornos. Além disso, as variáveis institucionais dos países estudados impactaram os retornos anormais de accruals. Por fim, esta pesquisa fornece evidências quanto a generalização da anomalia dos accruals, enriquecendo a literatura sobre países ainda pouco explorados e de importância econômica crescente, além de prover indícios sobre os efeitos dos aspectos institucionais na anomalia dos accruals na América Latina. / This research aimed to investigate accrual anomaly of returns from five countries in Latin America (Argentina, Brazil, Chile, Mexico and Peru), and to observe whether institucional factors impacted the accruals abnormal returns in these countries. For that purpose, time series for each country were built, and grouped in accruals and industry portfolios for Capital Asset Pricing Model (CAPM), 3-factor and 4-factor models, including the accrual factor (ACC). Abnormal returns were detected through the GRS statistics and the Fama and MacBeth procedure. Then, panel data models were used considering the period from 2004 to 2017, in which the dependent variables was the accrual abnormal returns resulted from the individual monthly series for each analyzed company; and the independent variables were the country-specific institutional factors and the company-level variables. Considering previous literature, in this study, it was expected that, if the accrual abnormal returns were confirmed in the time series from Latin America, the specific factors from each country would contribute to explain the differences between the returns. The results confirmed the presence of abnormal returns in every country from the sample and in all models, plus, the behavior from accruals partially explains returns. Furthermore, the institutional variables of countries influence the accruals abnormal returns. Finally, this research provides evidence over the generalization of accruals anomaly, contributing to literature on poorly studied, but economically relevant countries, besides providing signs regarding effects of institutional factors effects on the anomaly in Latin America

Esta pesquisa buscou investigar a anomalia dos accruals nos retornos de cinco países da América Latina (Argentina, Brasil, Chile, México e Peru), bem como observar se os aspectos institucionais impactaram nos retornos anormais de accruals dos países. Para tanto, foram analisadas séries temporais para cada país, agrupadas em carteiras de accruals e carteiras de setor, para os modelos CAPM, 3-fatores e 4-fatores, incluindo o fator accruals e observados os retornos anormais das séries, avaliando-os por meio da estatística GRS e pelas estatísticas de teste do procedimento de Fama e MacBeth. Em seguida, foram construídos modelos de dados em painel para o período de 2004 a 2017, em que as variáveis dependentes foram os retornos anormais de accruals, resultantes das séries temporais mensais de cada uma das empresas, e as variáveis independentes foram os fatores institucionais característicos dos países e as variáveis a nível de empresa. O estudo esperava que, se confirmados os retornos anormais de accruals nas séries temporais da América Latina, os fatores específicos de cada país contribuiriam para explicar as diferenças de retornos anormais existentes entre eles. Os resultados confirmaram a presença de retornos anormais em todos os países da amostra, em todos os modelos e o comportamento dos accruals explicou parte dos retornos. Além disso, as variáveis institucionais dos países estudados impactaram os retornos anormais de accruals. Por fim, esta pesquisa fornece evidências quanto a generalização da anomalia dos accruals, enriquecendo a literatura sobre países ainda pouco explorados e de importância econômica crescente, além de prover indícios sobre os efeitos dos aspectos institucionais na anomalia dos accruals na América Latina. / This research aimed to investigate accrual anomaly of returns from five countries in Latin America (Argentina, Brazil, Chile, Mexico and Peru), and to observe whether institucional factors impacted the accruals abnormal returns in these countries. For that purpose, time series for each country were built, and grouped in accruals and industry portfolios for Capital Asset Pricing Model (CAPM), 3-factor and 4-factor models, including the accrual factor (ACC). Abnormal returns were detected through the GRS statistics and the Fama and MacBeth procedure. Then, panel data models were used considering the period from 2004 to 2017, in which the dependent variables was the accrual abnormal returns resulted from the individual monthly series for each analyzed company; and the independent variables were the country-specific institutional factors and the company-level variables. Considering previous literature, in this study, it was expected that, if the accrual abnormal returns were confirmed in the time series from Latin America, the specific factors from each country would contribute to explain the differences between the returns. The results confirmed the presence of abnormal returns in every country from the sample and in all models, plus, the behavior from accruals partially explains returns. Furthermore, the institutional variables of countries influence the accruals abnormal returns. Finally, this research provides evidence over the generalization of accruals anomaly, contributing to literature on poorly studied, but economically relevant countries, besides providing signs regarding effects of institutional factors effects on the anomaly in Latin America

Problembakgrund: Finansiella rapporter är ett sätt för företagen att kommunicera med investerare och intressenter. Samtidigt som det finns lagar och förordningar som styr redovisning och rapporterns utseende finns det goda möjligheter för företagsledningen att påverka de siffror som presenteras i de publika rapporterna genom periodiseringar. Denna studie tar konceptet kring att investera baserat på graden av periodiseringar till den svenska aktiemarknaden. Problemformulering: Är det möjligt att generera riskjusterad överavkastning på den svenska aktiemarknaden med periodiseringsbaserade investeringsstrategier? Syfte: Huvudsyftet med studien är att undersöka investerares möjlighet till riskjusterad överavkastning på den svenska aktiemarknaden utifrån investeringsstrategier baserade på graden av periodiseringar. Ett delsyfte är att jämföra två delperioder för att undersöka om periodiseringsanomalins förekomst förändrats över den totala undersökningsperioden. Ett annat delsyfte är att undersöka om den svenska aktiemarknaden effektivt värderar offentlig redovisningsinformation. Teori: Studien baseras huvudsakligen på tidigare periodiseringsbaserad forskning utförd på den amerikanska aktiemarknaden där Sloan (1996) anses vara den som introducerade teorin kring periodiseringsanomalier så som den ter sig idag. Metod: En kvantitativ studie med en deduktiv ansats. Den empiriska undersökningen baseras på redovisnings- och aktiedata under tidsperioden 1996-12-31 till 2013-06-30 som kommer från Thomson Reuters Datastream och Worldscope. Empiri & analys: Resultatet av studien tyder på att det finns en negativ korrelation mellan periodiseringar och avkastning samt visar stöd för att periodiseringsanomalin existerar på den svenska aktiemarknaden. Av de periodiseringsmått som testats visade det sig att ΔNOA och ΔWC gav högst överavkastning. Slutsats: Studien indikerar att felprissättningar förekommer och att periodiseringsanomalin existerar på den svenska aktiemarknaden. Den höga exponeringen mot riskfyllda småbolag innebär att vinstmöjligheterna minskar något på grund av höga implementeringskostnader samtidigt som det gör strategierna mindre attraktiva för institutionella investerare. Det är därför upp till privata investerare att utnyttja dessa felprissättningar.

Cette thèse se compose de trois essais sur deux anomalies bien documentées : effet momentum et anomalie des ajustements comptables. Le premier essai examine si l'ampleur de l'anomalie des ajustements comptables est entraînée par la probabilité de détresse financière. Les résultats indiquent que l'anomalie des ajustements comptables est économiquement et statistiquement positive pour les entreprises avec une faible probabilité de détresse financière, mais non significative pour celles avec une forte probabilité de détresse financière. Cela signifie que cette anomalie des ajustements comptables est omniprésente, mais pas limitée aux entreprises avec une faible probabilité de détresse financière. Le deuxième essai étend la question de recherche abordée dans le premier essai au marché boursier émergent du Vietnam. Comme pour les résultats du premier essai, les résultats indiquent que l'anomalie des ajustements comptables est limitée aux stocks avec une faible probabilité de détresse financière. Le dernier essai examine si l'effet momentum se produit sur le marché boursier vietnamien. Les résultats confirment la présence de momentum dans le court terme et révèlent aussi que les rendements gagnants et perdants sont faiblement persistants, mais que la forte corrélation entre ces rendements gagnants et perdants crée des bénéfices momentum significatifs. / This dissertation consists of three essays on two well-documented anomalies: momentum effect and accrual anomaly. The first essay investigates whether the magnitude of accrual anomaly is driven by the financial distress probability. The results indicate that accrual anomaly is economically and statistically positive for firms with low financial distress probability, but insignificant for those with high financial distress probability. This means that that accrual anomaly is not pervasive but limited to firms with low financial distress probability. The second essay extends the research question addressed in the first essay into the emerging stock market of Vietnam. Similar to the findings in the first essay, the results indicate that the accrual anomaly is limited to the stocks with low financial distress probability. The last essay examines whether the momentum effect occurs in the Vietnamese stock market. The results support the occurrence of momentum in the short-run and also reveal that winner and loser returns are low persistent, but the strong correlation between winner and loser returns creates significant momentum profits.

Sloan（1996）研究指出，投資人無法完全地分辨出應計項目與現金流量間盈餘持續性的差別，導致對應計項目資訊反應過度，而對現金流量資訊則反應不足，因此公司擁有相對較高（低）的應計項目使用金額，預期會有負（正）的未來股票異常報酬率，此種存在於應計項目與未來來股票異常報酬率間之負向關係，即本文所稱之「應計項目異常現象」。 投資人方面，本研究依資訊取得優勢，區分為內部關係人、機構投資人（外資、投信、自營商）與自然人；投資人持股行為則分別以持股比例與持股比例變動代表。此外，並將應計項目分別以總應計項目與總應計項目組成要素下之個別營運資金應計項目（應收帳款變動數、存貨變動數與應付帳款變動數）作衡量。首先測試應計項目異常現象是否存在於我國，再利用應計項目異常現象建構之套利投資組合，買進最低應計項目金額的投資組合而賣出最高應計項目金額的投資組合，探討應計項目異常現象與投資人持股行為之關聯性。 實證結果顯示，應計項目異常現象存在於我國，亦存在於個別營運資金應計項目。持股比例方面，外資與內部關係人似乎能利用應計項目異常現象形成之套利投資組合；當總應計項目的金額愈低，持股比例會愈高，但在不同應計項目的衡量方法下會有不同的結果。持股比例變動方面，除內部關係人與自然人稍佳之外，本研究設計之迴歸模型並無對應計項目與投資人持股比例變動間之關聯性有足夠的解釋能力。此外，第二年度的內部關係人持股比例變動雖與總應計項目、存貨變動數有負向的關聯性，惟統計結果並不顯著。 關鍵字：應計項目異常現象、投資人、持股行為、應計項目、機構投資人、內 部關係人、自然人、套利投資組合 / Sloan（1996）results indicate investors failing to distinguish fully between the different properties of the accrual and cash flow components of earnings. This leads to overreaction of the information contained in the accrual components of earnings and underreaction of the cash flow components of earnings.Consequently,firms with relatively high (low) levels of accruals experience negative (positive) future abnormal stock returns. The negative relationship between accounting accruals and subsequent stock returns calls the “Accruals anomaly” in this paper. With repect to the investors, I distinguish them from the advantage of obtaining the information into insiders, institution investors (QFII, mutual funds, security dealers), and individual investors; then use the percentage of the investors’ holding and the percentage of the investors’ holding change to represent the investors’ holding behavior. Besides, I use the total accruals and individual working capital accruals（change in accounts receive, change in inventory, and change in minus accounts payable）to measure accruals. Firstly, I test whether the accruals anomaly exists in our country or not, then exploit the hedge portfolio formed by accruals anomaly,by taking a long position in the stock of firms reporting relatively low levels of accruals and a short position in the stock of firms reporting relatively high levels of accruals generates positive abnormal stock returns to probe into the association between accruals anomaly and investors’ holding behavior. The results suggested that accruals anomaly indeed exists in our country and the individual working capital accruals. With regard to the percentage of the investors holding, QFII and insiders seems to capable of exploiting the hedge portfolio formed by accruals anomaly; when firms with relatively low levels of total accruals experience the percentage of the high investors holding,but there have different results of using dissimilar measurement of accruals. For the percentage of the investors holding change, this paper’s regression model doesn’t have enough capability of explaining the association between accruals and percentage of the investors holding change except insiders and individual investors. Furthermore, although the percentage of the insiders’ holding change in the second year is negatively correlated with total accruals and change in inventory, the empirical results are not significant. Key words: accrual anomaly, investors, holding behavior, accruals, institution investors, insiders, individual investors, hedge portfolio