Investor sentiment and accruals anomaly:

European evidence

Francisca BEER
California State University of San Bernardino
fbeer@csusb.edu
Tel: 001 909 537 5709

Badreddine HAMDI
University of Burgundy
Badreddine.Hamdi@u-bourgogne.fr
Tel: 0033 380 395 320

Mohamed ZOUAOUI†
University of Burgundy
Mohamed.Zouaoui@u-bourgogne.fr
Tel: 0033 380 395 435

April 10, 2013

Abstract
This paper examines whether investor sentiment affects accruals anomaly across 15
European countries. In line with recent evidence for the U.S., we find that sentiment causes
accruals mispricing across European countries. The effect is pronounced for stocks whose
valuations are highly subjective and difficult to arbitrage. Our results also reveal evidence in
favor of managers' opportunistic disclosure behavior. The accruals reported are higher in high
sentiment periods as compared with low sentiment periods. Finally, we employ a cross-
country perspective and provide evidence that sentiment influences accruals anomaly in
countries with weaker outside shareholder rights, lower legal enforcement, lower equity
market development, dispersed ownership, higher allowance of accrual accounting and in
countries where herd-like behavior and overreaction behavior are strong. These countries that
present significant accruals mispricing in high sentiment periods also display high level of
accruals management, suggesting that accruals management is an important source of accruals
mispricing for individual investors.

Keywords: Investor sentiment; Accruals anomaly; Accruals management; Cross-country

study

†
Corresponding author.

-1-

Electronic copy available at: http://ssrn.com/abstract=2249675

Introduction

Classical finance theory posits that market prices fully reflect all available information.
Thus, in many cases, researchers rely on the maintained assumption that investors use all
available information and do so perfectly to set prices. Behaviorists, challenge this
assumption by addressing the implications of investors’ cognitive biases on the price
formation process. In particular, studies show that investors have limited attention and limited
aptitude to process information. Libby, Bloomfield, and Nelson (2002) and Barber and Odean
(2008) provide evidence that investors and financial professionals concentrate on a few salient
stimuli and, consequently, are more likely to neglect part of the relevant information. These
studies use evidence in social psychology about individual decision-making biases as
explanations for the fact that individuals can only process portions of available information.
Attention necessitates energy and since the amount of information accessible is very large,
investors must be selective (Kahneman, 1973). This inadequate way to operate might explain
why investors value firms based on their earnings performance rather than all the financial
variables available.
Hirshleifer, Hou, Teoh, and Zhang (2004) argue that when investors are subject to limited
attention, they do not make full use of the balance sheet information and purely focus on
accounting profitability neglecting cash profitability. Hirshleifer, Lim and Teoh (2011)
provide evidence that investors with limited attention attend to earnings announcement but do
not impound the information about earnings components (i.e. cash flows from operations and
accruals) into their valuation, irresponsive of the fact that cash flows from operations are
better at forecasting future profitability than accruals (e.g. Todd and Tim, 2000; Hirshleifer,
Hou and Teoh, 2009, 2012). To neglect the difference between these variables might result in
over-optimistic prospects for firms with high accruals and in gloomy one for firms with low
accruals. Consequently, firms with high accruals might end up overvalued relative to those
with low accruals. In the long run, such mispricing will eventually be corrected. This pattern,
known as the accruals anomaly, originally documented by Sloan (1996), shows that investors
overestimate the persistence of the accrual component of earnings when forming earning
expectations and suggests that firms with high accruals underperform firms with low accruals.
In a recent study, Ali and Gurun (2009) use investor sentiment to capture limited
attention. They hypothesize that optimism (pessimism) decreases (increase) the attentiveness
of investors. When sentiment is high, retail investors are trading more enthusiastically and
paying less attention to the impact of earnings components on stock prices than in low

-2-

Electronic copy available at: http://ssrn.com/abstract=2249675

sentiment periods. Furthermore, since high-sentiment periods are characterized by overly
optimistic expectations about stocks future prospects, it is plausible that during these periods,
investors will neglect the losses attributed to mispricing. Conversely, during low sentiment
periods, when retail investors are more suspicious about stocks future prospects, it is probable
that these losses will be considered to be significant. During these periods, retail investors are
more likely to pay attention to the earnings components. This assumption is convincingly
supported by research in social psychology where one can find evidence suggesting that
human behavior is significantly different in times of anxiety and fear than in periods of
prosperity and tranquility (Tiedens and Linton, 2001). Investors’ decision to buy, sell or hold
is thus impacted by the investor’s psychological state of mind when the decision is made. In
particular, the psychology literature shows that emotions affect information processing.
Schwarz (2002), for example, notes that investors in an optimistic emotional state are likely to
proceed as usual while those experiencing a pessimistic emotional state have the tendency to
proceed in a more drastic and details oriented manner.
Our study speaks to the call by Mian and Sankaraguruswamy (2012) for a better
understanding of the relation between investors’ cognitive biases and the market’s reaction to
accounting information. We examine the impact of limited investor attention, measured by
market sentiment, on accruals anomaly. Our analysis is predicated on the assumption that
investors are less vigilant during optimistic periods than during pessimistic one. Optimistic
views about stocks have the tendency to be excessively optimistic in high sentiment periods,
leading to a larger overvaluation of accruals as compared with low sentiment periods.
Furthermore, there is substantial evidence that firms with more subjective valuations and
greater limits to arbitrage are more affected by investors‘ behavioral biases (Baker and
Wurgler, 2006; Lemmon and Portniaguina, 2006 and Qiu and Welch, 2006). Thus, our first
hypothesis predicts that accruals mispricing is greater in high sentiment periods as compared
with low sentiment periods, especially for firms hard to value and difficult to arbitrage.
Several empirical studies find results broadly consistent with the established finding in the
field of psychology that investors use different decision rules and consequently, react
differently to certain information during optimistic periods than in pessimistic periods. Ali
and Gurun (2009) note that accruals mispricing is stronger in high sentiment periods as
compared with low sentiment periods, especially for small stocks which tend to be held by a
greater proportion of individual investors. Livnat and Petrovits (2008) show that in
pessimistic sentiment periods, companies with low accruals produce considerably higher
excess returns than during optimistic sentiment periods. Similarly, Stambaugh, Yu and Yuan

-3-
(2012) examine eleven well documented anomalies in the finance literature (including
accruals anomaly) and find that their profitability is stronger in high-sentiment than low
sentiment states. In a related literature, Mian and Sankaraguruswamy (2012) examine the
association between investor sentiment and stock price reaction to earnings news. They find
that the stock price response to good earnings news is higher during periods of high sentiment
periods, whereas stock price reaction to bad earnings news is higher during periods of low
sentiment. The response is more important for stocks that have more subjective valuations
such as small stocks, young stocks, high volatility stocks, non-dividend-paying stocks, and
stocks with extremely high and low market-to-book ratios. Hribar and McInnis (2012) show
that investor sentiment is associated with bias in analysts’ earnings expectations, mainly for
difficult-to-value firms. They find that part of the forecast errors occur during periods of high
sentiment.
Our study adds to recent accounting research which examines possible explanations to the
accruals anomaly. In this respect, Xie (2001) and DeFond and Park (2001) find that the
accruals anomaly is attributable to the mispricing of discretionary accruals, which are
expected to be less reliable and more subject to managerial discretion. Richardson, Sloan,
Soliman and Tuna (2005) confirm this result and conclude that the magnitude of the accruals
anomaly is significantly greater for the discretionary component of total accruals, which have
been linked to earnings management in a large volume of empirical studies. In an
international setting, Pincus, Rajgopal, and Venkatachalam (2007) examine various
explanations for the accruals anomaly and find that earnings management by means of accrual
manipulation best explains this anomaly. Their findings show also that the anomaly persists as
the barriers to arbitrage due to the absence of close substitutes for mispriced stocks.
We connect the sentiment-related mispricing to managers' opportunistic disclosure
behavior. Recent studies show that managers intentionally disclose information in accordance
to investors’ current sentiment. These studies also presume that managers distinguish between
periods of high or low sentiment and adjust their disclosure strategies accordingly. Bergman
and Roychowdhury (2008) examine how investor sentiment influences managerial discretion
in the financial reporting process. They argue that managers, in an attempt to sustain
optimistic valuations, have incentives to reduce voluntary disclosure when investor sentiment
is bullish and when their expectations of future earnings are optimistic. In contrast, when
sentiment is low and expectations are less optimistic, managers increase voluntary disclosure,
with the objective of adjusting expectations upward. Rajgopal, Shivakumar and Simpson
(2007) investigate the overall impact of market sentiment on earnings management. They find

-4-
that managers cater to overall earnings optimism by increasing abnormal accruals. These
catering incentives are relatively stronger for smaller stocks and those susceptible to more
subjective valuations. Ali and Gurun (2009) find that managers of small firms increase
accruals during periods of high sentiment to exploit investors’ limited attention for
understanding the earnings components. Thus, our second hypothesis forecasts that managers
of firms’ hard to value and difficult to arbitrage report higher accruals during high sentiment
periods than low sentiment periods in an attempt to take advantage of investors’ greater
overvaluation of accruals during these periods of high sentiment1.
Our study differs from previous research in three different ways. First, we extend Ali and
Gurun’s findings to a large set of European countries that allows comparisons with U.S. data.
Pooling data also is known to increase the power of statistical tests providing better estimates
(Ang and Bekaert, 2007). Second, we deepen the cross sectional relationship between investor
sentiment and accruals anomaly. The sentiment-driven accruals mispricing in stocks is
expended by integrating several characteristics other than firm size such as dividend policy,
profitability, and tangibility. Third, we analyze the cross-country variations in the sentiment-
accruals anomaly. A cross-country study can provide evidence on how cultural differences, as
well as institutional differences, affect the sentiment-accruals anomaly relationship.
Our analysis examines 15 European countries and spans the period 1994-2011.
Specifically, we investigate whether consumer confidence – as a proxy for individual investor
sentiment– affects accruals mispricing across the European countries. We find that accruals
mispricing is higher in high sentiment periods than in low sentiment periods. The effect of
sentiment on the accruals mispricing appears to be pronounced for stocks hard to value and
difficult to arbitrage. Further, we note that the accrual component of earnings tends to be
higher in high sentiment periods as compared with low sentiment periods. This result suggests
that managers of these firms report higher accruals in high sentiment periods to take
advantage of investors’ greater overvaluation of accruals during these periods. Finally, we
employ a cross-country perspective and provide evidence that both cultural values and
institutional structure have explanatory power for the accruals anomaly. We also find that the
countries that present significant mispricing accruals in high sentiment periods also display
high level of accruals management. This finding provides support for the hypothesis that

1
For example, the managers’ compensation is often increased when their stocks are overvalued, even though the
overvaluation is only temporary and will be corrected when accruals reverse.

-5-
earnings management using accruals is a key component to the occurrence of the accruals
anomaly.
The remainder of the paper is organized as follows. Section 2 discusses the data, the
sample selection, and describes variables measurement. Section 3 presents the research
design. Section 4 reports our results on the effects of sentiment on accruals anomaly. Section
5 reports results of cross-country analysis and section 6 concludes.

2. Data sources and variable definitions

Our study covers all common stocks listed on the major stock exchanges of 15 European
countries for the period 1994-2011. The countries included in the study are Belgium, Czech
Republic, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Netherlands, Portugal,
Slovenia, Spain, Sweden and the U.K. We extract stock return data and company financial
data from the Datastream/Worldscope database. Ince and Porter (2006) caution researchers
for potential flaws in Datastream pertaining to coverage, classification, and integrity of data;
particularly for low-price stocks and/or small stocks. To clean the return data2, we apply
screens advocated by the authors by removing penny stocks, stocks with extremely low
market capitalization and stocks with unrealistic returns behavior. Penny stocks are defined as
stocks trading for less than one Euro. Stocks with extreme market capitalization are stocks
with a market capitalization lower than 1% of the country market capitalization during any
given month. Unrealistic returns are returns above 300% that are reversed within one month3.
Similarly to previous research examining accruals, we exclude financial firms (SIC codes
6000-6999) and utilities (SIC codes 4900-4999) from our sample. Financial firms and utilities
are known to exhibit investing, capital structure, and other characteristics that differ from
other industries. We also exclude firm-year observations with non-positive total assets and
book value of equity, as these variables are used to standardize other variables and thus
cannot be zero or negative. A complete set of desired data is required for a firm-year
observation to be included in our sample. Our data requirements produce a final sample
comprised of 46,386 firm-year observations for 4,787 European firms.

2
We measure all stock returns in Euro using the stock return index which is adjusted for dividends and stock
splits. When a stock is not traded, Datastream replicates the stock return index of the earlier month to the current
month. Hence, a stock return index equalling zero may be the result of the absence of trading. In the same way as
Ince and Porter (2006), we delete all zero returns from the end of the sample until the first non-zero return to
eliminate this type of error.
3
Similar to Ince and Porter (2006), we set monthly returns to missing if Rt or Rt-1 is greater than 300% and
(1 + Rt)(1 + Rt-1) −1 is less than 50%. Where Rt and Rt-1are the stock returns on month t and t-1, respectively.

-6-
 2.1. Accruals measures

We use two accrual measures: total accruals and current accruals. Along the lines of prior
research (Dechow, Sloan and Sweeney, 1995, and Sloan, 1996), we measure total accruals as
change in non-cash working capital minus depreciation expense. Specifically, we estimate
total accruals for firm i in year t as follows4:

Total Accruals it = ( ∆CA it - ∆CASH it ) − ( ∆CLit − ∆STDit − ∆TPit ) − DEPit (1)

Where ∆CAit represents the annual change in total current assets; ∆CASHit is the change in
cash and cash equivalents; ∆CLit is the change in total current liabilities; ∆STDit is the change
in short-term debts; ∆TPit is the change in income taxes payable; and DEPit is the depreciation
and amortization expense. Total accruals component is scaled by lagged assets for meaningful
cross-sectional and cross-countries analyses. Notice that because accruals are mechanically
related to ending total assets, average total assets cannot be used as a deflator (Ali and Gurun,
2009).
The current accruals for firm i in year t is measured as:

Current Accrualsit = ( ∆CA it - ∆CASH it ) − ( ∆CLit − ∆STDit − ∆TPit ) (2)

The current accruals component is also scaled by lagged assets. We use current accruals as
a proxy for accruals management because managers have greater discretion over current
accruals than over noncurrent accruals (Sloan, 1996). Managers are less likely to manage
earnings by manipulating depreciation because the disclosure requirements make changes in
depreciation highly visible (Beneish, 1998 and Botsari and Meeks, 2008). Earnings
management accomplished by timing capital expenditures also has the potential of been costly
if profitable investment opportunities are reduced or postponed.

2.2. Firm characteristics

Baker and Wurgler (2006, 2007) show that the effect of investor sentiment on stock
returns is more pronounced for certain categories of stock, particularly those that are hard to
4
To compare our results with those published by Ali and Gurun (2009) using U.S. data, we rely on the same
measure of accruals. Notice, however that other measure of accruals computed directly from the cash flows
statements were also used (Collins and Hribar, 2002). Findings, available upon request, are similar to those
reported in this paper.

-7-
value and to arbitrage. One stream of research shows that arbitrage tends to be particularly
risky and costly for young, small, unprofitable, or extreme growth stocks (D’Avolio, 2002;
Wurgler and Zhuravskaya, 2002). Likewise, it is difficult to use traditional models to value
firms characterized by high amounts of intangible, the absence of earnings history and
significant growth opportunities. Finally, stocks not paying dividends are also difficult to
value as their fundamentals are speculative prospects (Pontiff, 1995).
Following Baker and Wurgler (2006), we classify small firms, young firms, unprofitable
firms, intangible firms and firms that do not pay dividends as firms that are difficult to value
and costly to arbitrage. The firms’ characteristics used are: size, age, profitability, tangibility
and dividend policy. Size is the market capitalization measured as price time number of
shares outstanding. Age is the number of months since the firm’s first appearance in our
database. Profitability is captured by the return on assets defined as earnings divided by total
assets. Tangibility is captured by property, plant and equipment over total assets. Dividend
policy is approximated by the ex-date dividends per share multiplied by shares outstanding
divided by book equity.

2.3. Investor sentiment

As of today, there is no uncontroversial and universally accepted sentiment measure.

Several empirical studies have attempted to quantify investor sentiment5. These studies
identified direct and indirect sentiment measures. Direct sentiment measures are derived from
surveys directly asking individuals how they feel about current or future economic and stock
market conditions while indirect ones rely on economic and financial variables susceptible to
capture investors’ state of mind.
For our international analysis, we favour the consumer confidence index for the following
reasons. First, the relationship between the consumer confidence index and international stock
markets is well documented. Prior studies show that consumer confidence index, as a proxy
for investor sentiment, has a significant impact on stock returns across countries even after
controlling for other standard risk factors and expected business conditions (Schmeling, 2009;
Zouaoui, Nouyrigat and Beer, 2011). Second, the consumer confidence index constructed by
the European Commission is available for several European countries for long and regular
periods of time. The index is accurate measure of individual investor sentiment because it is
based on a monthly survey of a large number of households about their current and expected

5
See Baker and Wurgler (2007) for a recent and comprehensive survey of investor sentiment measures.

-8-
financial situation and their beliefs about the economy. Third, more importantly, as these
surveys are harmonized since the mid-1980s, comparisons across countries are achievable.
The raw sentiment indicator encompasses a psychological component related to sentiment
and a rational component related to economic fundamentals. The bullishness or the
bearishness of an investor can reflect rational or irrational future expectations or both. To
mitigate this possibility, we decompose the raw sentiment indicator into two components: a
rational one, reflecting the economic fundamentals and a psychological component, reflecting
investor sentiment. Specifically, we treat the residuals from the equation (3) as our measure of
sentiment unwarranted by fundamentals:

= + , + , (3)

Where Senti is the raw sentiment variable of every country, αi is the constant and βi are the
parameters to be estimated. Fund is the set of fundamental variables representing rational
expectations based on risk factors. Similar to previous studies, we use data on growth of
industrial production, inflation, term spread, and growth in durable, nondurable and services
consumption. The fitted values of Equation 3 capture the rational component and the residual
captures the psychological component. Finally, we construct an annual measure of sentiment,
Sent┴, by averaging the consumer confidence index orthogonalized by fundamental variables
across twelve months in every year.

3. Research design

3.1. Accruals mispricing

As stated previously, our hypothesis asserts that accruals mispricing for stocks hard to

value and difficult to arbitrage is greater in periods when investor sentiment is high than in

periods when investor sentiment is low. This hypothesis is tested using the following model

of Ali and Gurun (2009):

Eretit +1 = β 0t + β1t Dummyit + β 2t TotalAccrualsit + β 3t Pr ioir Re tit

+ β 4t BM it + β 5t Dummy × TotalAccrualsit + µ it +1 (4)

β 5t = γ 0 + γ 1Sentt⊥ + ν it +1 (5)
Eretit+1 is stock return in excess of market return for month t+1. Dummyit is a dummy
variable that takes the value of one when a firm characteristic (i.e. size, age, profitability,

-9-
tangibility and dividend policy) is below the median for the month t, and is zero otherwise. At
the beginning of each month t, all stocks are ranked as difficult to value and hard to arbitrage
relative to easier to value and easier to arbitrage, using the median of each firm characteristic
as the cutoff point. Fama and French (1992) show that a delay of six months is needed to
insure that accounting data are published and available to investors. For instance, when a firm
publishes its accounting data as of December 31 of year t, these data are only available to
investors at the end of June of t+1. According to Fama and French, a gap of at least 6-month
between fiscal year end and the return measurement month is a safe way to operate. To relate
future returns to the most recently reported accounting information, Total accrualsit is total
accruals for the fiscal year that ended at least six months before but less than eighteen months
before the return measurement month. BM is book-to-market ratio measured as the book
value of equity divided by the market value of equity. Book value of equity is for the fiscal
year that ended at least six months before, but less than eighteen months before the return
measurement month. Market value of equity is for the most recent calendar year-end.
PriorRetit is prior returns measured by 6-month cumulative stock returns preceding the future
return measurement period.
Sent┴ is the explanatory variable of Model (5) and represents our sentiment measure
defined previously. Since we investigate the relationship between investor sentiment and
accruals mispricing, our independent sentiment variable must capture investors’ state of mind
at the time they value the firm reported accruals. The dependent variable of Model (5) (β5t)
represents the value of average accruals mispricing of firms difficult to value and hard to
arbitrage relative to those easier to value and easier to arbitrage for a month t. This dependent
variable is estimated using cross-sectional regressions, one for each month in our sample
stocks that includes accruals from firms with different fiscal year-ends. Thus, the sentiment
measure that corresponds to the accruals mispricing may be either for the most recent
calendar year before the return month or for the calendar year before that. As Ali and Gurun
(2009), we retain the average value of the sentiment measures for the two most recent
calendar years.
When investors value a firm, they should distinguish between the higher persistence
associated with cash flows component of earnings and the lower persistence associated with
accrual component of earnings. However, in high sentiment periods naïve individual investors
are likely to fixate more on the total amount of reported earnings without regard the
persistence differential of accruals and cash flows. Because they anchor on total amount of
earnings, investors systematically overreact to accrual earnings. We predict a negative

-10-
coefficient for the variable Sent┴ in Equation (5). For firms hard to value and difficult to
arbitrage, a negative coefficient is coherent with significant overvaluation per unit of accruals
when investors are bullish.

3.2. Accruals management

The managers can use earning discretion to distort financial information in order to
maximize their own utility. For instance to conceal their private control benefits, managers
may be inclined to use earnings management to mislead outsiders about the firm
performance6. Assuming that investors’ limited attention increases during high sentiment
periods, it is conceivable that managers will attempt to exploit the greater mispricing by
reporting higher accruals in these periods. Therefore, we evaluate the following model
inspired by the study of Ashbaugh, LaFond and Mayhew (2003):

TotalAccrualsit = α + β1 Sent t⊥ + β 2 Dummyit + β 3 Sentt⊥ × Dummyit

+ β 4TotalAccrualsit −1 + β 5 MAit + β 6 Financing it
+ β 7 Litigationit + β 8 Leverageit + β 9 MBit + β10 Lossit
+ β11CFOit + ΣIndustryit + ε it (6)

The model (6) has been modified by the addition of the variable Sent┴ designed to
capture investors’ state of mind7. The dependent variable in equation (6) was defined
previously. Dummy is a dummy variable that takes the value of one when a firm characteristic
(i.e. size, age, profitability, tangibility and dividend policy) is below the median for the year t,
and is zero otherwise. Total accrualsit-1is prior year total accruals used to control for the
accruals reversal. MAit is an indicator variable set to 1 if the company has engaged in a
merger/acquisition activity, and zero otherwise. Financingit is one if MAit is not equal to one
and the number of outstanding shares increased by at least 10 percent, or long-term debts
increased by at least 20 percent, or the firm first appears on the Thomson monthly returns
database during the fiscal year, and zero otherwise. Litigationit is an indicator variable,
controlling for ex-ante litigation risk, that equals one if the firm operates in a high-litigation
risk industry, and zero otherwise. High-litigation industries are industries with SIC codes of
6
Private control benefits are various perquisites to which only managers are entitled (e.g. bonus, prestige…,
etc.). Most of these perquisites provide immediate benefit, although some may result in an immediate harm to be
avoided, such as a drop in stock price.
7
We decide not to use discretionary accruals, the dependent variable used in the model of Ashbaugh, LaFond
and Mayhew, to evaluate earnings management. Discretionary accruals are customarily estimated as the
residuals of a yearly cross-sectional regression of an accrual model. The residuals mean been zero, this technique
of evaluation prevents us from examining the relation between investor sentiment and accruals management.

-11-
2833-2836, 3570-3577, 3600-3674, 5200-5961, and 7370-7374. Leverageit equals the total
debt divided by total assets at the beginning of the fiscal year. MBit is market-to-book ratio,
defined as the market value of the firm’s equity scaled by the book value of its common
equity at the end of year. Lossit is an indicator variable set to 1 if the firm reports a net loss for
the fiscal year, 0 otherwise. CFOit is cash flow from operations, divided by beginning of year
total assets. Industryit are indictors based on the Thomson database industry groups.

4. Results

Table 1 depicts some descriptive statistics for the variables total accruals, current accruals
and the control variables used in the study. Table 1 shows that on average total accruals tends
to be negative with a mean of -0.037, while mean and median for current accruals are
positive. This negative sign is the result of the subtraction of depreciation expenses from total
accruals while capital expenditures are incorporated in investing cash flows.

4.1. Results for accruals mispricing

Panel A of Table 2 presents the results of Fama-MacBeth regressions of Model (4) with
country dummies to capture the possible country effect. The reported estimates are the time-
series averages of the monthly estimated coefficients. The p-value is based on the standard
deviation of coefficient estimates. Results in the table are comparable for all characteristics
studied. Similar to prior studies, we find that the coefficients for the variables book-to-market
and prior returns are positive and significant. Findings are thus consistent with the arguments
that the variable book-to-market capture distress risk and/or mispricing effects and the
variable prior returns capture the risk of momemtum and/or mispricing effects. We also find
that the coefficient for the variable total accruals is negative and significant. This finding,
which is also consistent with previous studies, indicates the presence of accruals mispricing in
our international sample (LaFond, 2005 and Pincus, Rajgopal, and Venkatachalam, 2007).
Additionally, we observe that the coefficient for the interactive variable between total accruals
and the dummy variable is negative but insignificant. A non-significant coefficient suggests
that on average there is not much difference in accruals mispricing between small and large
stocks, young and old stocks, profitable and non-profitable stocks, tangible and non-tangible
stocks and dividend and non-dividend paying stocks.
Panel B of Table 2 presents the results of the estimation of Equation (5). In that equation,
the dependent variable represents the time series of coefficients β5 estimated in Equation (4)

-12-
and the independent variable is our measure of sentiment. We observe that all estimated
coefficients are negative and significant. The adjusted R2 for the characteristics studied range
between 0.9% and 2.4%. The highest adjusted R2 is found for the characteristic tangibility.
Generally speaking, these findings indicate that the mispricing per unit of accruals for stocks
hard to value and difficult to arbitrage is related to our sentiment measure, which is consistent
with our prediction. To check for robustness of our results, we replace the time series of
coefficients β5 by the time series of coefficients β2 in Equation (4). The model is as follows:
β 2t = γ 0 + γ 1 Sentt⊥ + ν it +1 (7)
An insignificant relation between accruals mispricing of all the stocks and our measure of
sentiment should further support our hypothesis. The results of the test of model (7) are
reproduced in Panel C of Table 2. We found a negative but insignificant link between accruals
mispricing of all stocks included in our sample and investor sentiment. The results validate
the hypothesis that sentiment does not significantly impact the accruals mispricing of stocks
easier to value and to arbitrage.
Finally, we run additional regressions by replacing the dependent variable total accruals
by current accruals. The results summarized in Table 3 show that the estimated coefficients
and their significance levels are comparable to those presented above. This result supports the
hypothesis following which accruals mispricing for stocks hard to value and difficult to
arbitrage is greater in periods when investor sentiment is high than in periods when it is low.

4.2. Results for accruals management

Table 4 reports the regressions result of Model (6) using pooled data across time and firms
in all countries. The regression equation is estimated with the ordinary least square (OLS)
estimation procedure clustered by country and time using pooled data. All control variables
included in model are significant and display the expected sign. The variables MA and
Financing have both positive coefficients significant at 1%. This result induces us to conclude
that companies engaged in mergers and acquisitions and those ensuring their funding through
a capital increase or debt, tend to manage upwards their results (higher accruals).
The coefficient on the litigation variable is also positive and significant (p-value of 0.000).
This result confirms that firms subject to high litigation risk avoid negative earnings surprises
by overvaluing accruals component of earnings (Matsumoto, 2002). The leverage coefficient
is negatively significant at 1%. In light of this result, it appears that managers of indebted
firms do not tend to engage in upwards earnings management. Leverage restricts the

-13-
opportunistic behavior of managers. In highly leverage firms, free cash flows is used to
service loans, precluding managers from investing in sub-optimal projects constraining them
to manipulate earnings to conceal these projects poor performance.
Furthermore, in all five panels, the value for coefficients for the variables Loss and CFO
are negative and significant respectively at 1% and 5%. This result reveals that the firms
reporting a net loss are less likely to report higher accruals. As far as cash flows are
concerned, firms with high operating cash flows may not necessitate manipulating accruals to
meet earnings targets.
Our results also show after controlling for the variables impacting accruals, there is a
positive and significant link between the interactive variable Sent┴×Dummy and total
accruals. This result holds for all the characteristics studied and supports our hypothesis that
accruals are higher for stocks hard to value and difficult to arbitrage in high sentiment periods
as compared with low sentiment periods. It suggests that managers of these firms report
higher accruals in high sentiment periods to take advantage of investors’ greater overvaluation
per unit of accruals during these periods.
As robustness check, we re-run regressions of Model (6) using current accruals as the
dependent variable. Results from this robustness exercise are presented in Table 5. We find
that the estimated coefficients and their significance levels are comparable to those presented
above.

5. Cross-country analysis

As established above in this study, accruals mispricing and accruals management are
pervasive for stocks hard to value and difficult to arbitrage in high sentiment periods
(optimism) as compared with low sentiment periods (pessimism). This section contributes to
our understanding of the international differences in accruals mispricing and accruals
management phenomena8. Specifically, we examine whether differences in market integrity
and culture between countries influence the accruals mispricing and the managerial behavior
in earnings discretion. To measure cross-country differences in institutional and cultural
values, we turn to the institutional variables of Leuz, Nanda and Wysocki (2003) and the
cultural variables as defined by Hofstede (1980). All market integrity and cultural variables

8
In the literature, studies use aggregate measure of earnings management to capture a large range of
different earnings management activities (e.g. the magnitude of accruals, the tendency of firms to avoid
small losses, the smoothness of earnings relative to cash flows and the correlation of accounting accruals
and operating cash flows). In this study, the term ‘accruals management’ is used to imply earnings
management by means of accrual manipulation.

-14-
are country-level observations. Sub-sections 5-1 focuses on accruals anomaly and sub-section
5-2 concentrates on accruals management.

5.1. Accruals mispricing cross-country

The evidence on the international differences in accruals mispricing phenomena is sparse

and controversial. LaFond (2005) determine how country differences in accounting and
institutional structures explain the accruals anomaly. Using data from 17 countries, the author
reports that international factors are not significant contributing factors to predict the accruals
anomaly across all markets studied. He finds evidence that the accruals anomaly is a global
anomaly, there is not much difference in accruals mispricing between countries with code and
common law, countries with wide uses of accounting standards and countries with different
amounts of shareholder protection. However, this result is challenged by Pincus, Rajgopal and
Venkatachalam (2007). Using country-level data from 20 countries, they find that accruals
anomaly is more likely to occur in countries with common law, weaker outside shareholder
rights, higher equity market development, lower concentration of share ownership and higher
allowance of accrual accounting. Other studies have investigated the accruals anomaly in an
international context but none of these studies have deepen the link between accruals anomaly
and cultural factors (Liodakis, Brar, Gadaut, and Sharma, 2004 and Leippold and Lohre,
2012)9.
• Institutional factors
To examine if our findings are linked to the countries distinctive features, we split them
into two groups depending on some determinants of market integrity. Specifically, we use our
cross-section of countries to determine if there is evidence that the pervasiveness of the
accruals anomaly in high sentiment periods is related the level of development of their
financial institutions and to the level of sophistication of their equity market. Our market
integrity indicators obtained from Leuz, Nanda and Wysocki (2003) include: (i) Outside
investor rights; (ii) Legal enforcement; (iii) Importance of the stock market, and (iv)
Ownership concentration. Our last market integrity indicator is the Hung (2001)’s index
measuring the extent of accrual accounting permitted in a country.
The first and second market integrity variables are well-known proxies for country
shareholders’ protection mechanisms and the ability of the country to enforce them. Outside
investor rights are proxied using the anti-director rights index created by La Porta, Lopez-de-

9
A notable exception is the paper of Papanastasopoulos (2011) that shows that accruals mispricing varies
according to a country individualism level.

-15-
Silanes, Shleifer and Vishny (1998). This index captures how strongly the legal system favors
minority shareholders against dominant shareholders. Legal enforcement is measured as the
mean score across three legal variables (1) the efficiency of the judicial system, (2) the
assessment of rule of law, and (3) the corruption index. Hung (2001) shows that the presence
of shareholder protection attenuates the negative impact of accruals on the value relevance of
earnings. We expect that the occurrence of the accrual anomaly will be negatively related to
the strength of shareholders’ protection since it reduces incentives to manipulate accruals by
decreasing the opportunities to benefit from these manipulations.
The third and fourth market integrity variables are used to represent the characteristics of
equity markets, including their importance as a source of capital and the concentration of
share ownership. The importance of the stock market in the local economy is measured by the
mean rank across three variables: (1) the ratio of the aggregate stock market capitalization
held by minorities to gross national product, (2) the number of listed domestic firms relative
to the population, and (3) the number of IPOs relative to the population. In countries where
investors focus heavily on earnings to value firms, managers might have more incentives to
manipulate accruals. In developed equity markets, earnings are more value relevant than in
less developed markets (Alford, Jones, Leftwich and Zmijewski, 1993; Ali and Hwang,
2000). Consequently, investors in developed markets confer more importance of earnings for
security pricing. This market characteristic is insidious because it has the potential to motivate
managers to manipulate accruals to keep investors happy all the time. We thus expect a
positive relationship between the occurrence of the accruals anomaly and the importance of
the equity markets in a country.
The ownership concentration variable is measured as the median percentage of common
shares owned by the largest three shareholders in the 10 largest privately owned non-financial
firms. High dispersion of ownership is likely to increase the information asymmetry between
managers and stock market participants. Information asymmetry induces investors to rely on
reported and forecasted earnings to value firms. As a greater focus on earnings increases the
probability of accruals mispricing, the occurrence of the accruals anomaly should be
negatively correlated to the degree of concentration of share ownership in a country.
Hung (2001) develops an accrual index based on an analysis of 11 accruals accounting
standards presented by Coopers and Lybrand (1993) for twenty-one countries10. The accruals
index corresponds to the extent to which the accounting system moves away from a cash

10
We obtain data for the index from Hung (2001). Among these countries, data were not available for Czech
Republic, Greece, Portugal and Slovenia.

-16-
method measure of performance. A higher index value indicates higher use of accrual
accounting in a country is permitted. Hung (2001) finds that the degree to which accrual
accounting is permitted in a country negatively affects the value relevance of earnings. Since
a higher use of accrual accounting provides managers with more opportunities to manage
earnings, we expect higher mispricing anomaly in countries with higher allowance of accrual
accounting. To sum-up, we hypothesize that the accruals mispricing in high sentiment periods
is more pronounced than in low sentiment periods in countries with weaker outside
shareholder rights, lower legal enforcement, well-functioning equity markets, dispersed
ownership, and higher allowance of accrual accounting.
As shown in Table 6, the countries are allocated to one of two groups depending on
whether they are above or below the median of a specific characteristic of market integrity.
Regression results of Model (5) for each group of country show there is a negative and
significant relation between sentiment and accruals mispricing in countries characterized by
weaker anti director rights, weaker legal enforcement, lower equity market development,
lower ownership concentration and more extensive accrual accounting. In the countries with
contrasting features, the accruals anomaly is not significantly related to investor sentiment.
Our results are in line with our hypothesis except that related to importance of equity
markets. In accordance with the classic literature on accruals mispricing11, the anomaly is
expected to be more prevalent when earnings are more value relevant in security pricing, an
attribute of developed equity markets. Our result, however, can be reconciled with the idea
that large equity markets should benefit from a better flow of information and therefore be
more efficient. In an efficient market, the effectiveness of the accrual strategy is thus
questionable.
• Cultural factors
Hofstede’s (1980, 2001) cultural dimensions serve as the proxy for national culture.
Hofstede (2001) defines culture as “the collective programming of the mind” separating the
members of one group from another. Culture, a by-product of our environment, can be
captured by five dimensions: power distance, uncertainty avoidance, individualism,
masculinity and long-term orientation. Findings from studies on the importance of cultural
differences in equity markets show that only two dimensions, namely individualism and
uncertainty avoidance, exhibit a clear relationship with mispricing in capital markets
(Schmeling, 2009; Chui, Titman and Wei, 2010; Zouaoui, Nouyrigat and Beer, 2011).

11
See for example Pincus, Rajgopal and Venkatachalam (2007) and Papanastasopoulos (2011).

-17-
 Individualism refers to the extent to which people emphasis their own abilities to
differentiate themselves from others. Collectivism, on the other hand, pertains to people
(stakeholders) motivation to be assimilated and sheltered by their organizations. Further,
because collectivistic countries are countries in which people are integrated into strong
groups, consensus opinions prevailed. Collectivism leads to imitative or herd-like behavior.
Herding can be defined as behavior patterns that are correlated across individuals. Noise
traders’ actions are correlated as they are mimicking each other actions based on their similar
overly optimistic or pessimistic expectations. This tendency to invest with the herd is exactly
what is assumed to drive the relationship between investor sentiment and stock returns in
financial markets.
Uncertainty avoidance has the potential to be the most relevant dimension to the cultural
aspect of equity investment (Lucey and Zhang, 2010). According to Hofstede (2001), the
fundamental issue pertaining to uncertainty avoidance is whether a society tries to control the
future. The uncertainty avoidance index indicates to what extent people believe they are
content regardless of uncertainty. In countries with a higher uncertainty avoidance index,
people prefer predictable outcomes, are reluctant to accept risks and supposed to be more
emotional than in countries with lower uncertainty avoidance index. People in the later
countries can handle more uncertainties, tolerate more risks and are described as
contemplative and thoughtful. As a result, we use the uncertainty avoidance index as proxy
for the tendency of investors to overreact. To sum-up, we hypothesize that the accruals
anomaly in high sentiment periods is more pronounced than in low sentiment periods in
countries with low level of individualism and high uncertainty avoidance.
Table 6 shows that accruals mispricing in high sentiment periods is higher than low
sentiment periods in countries scoring low on individualism and countries scoring high on
uncertainty avoidance, supporting the idea that investors in different cultures have different
biases. This result validates our hypothesis that investor sentiment causes accruals mispricing
in countries where herd-like behavior and overreaction behavior are strong.

5.2. Accruals management cross-country

There is a growing literature that examines systematic differences in earnings

management across countries. Numerous studies establish that cross-national differences in
earnings management can be explained by two complementary forces: institutional factors
and cultural factors.

-18-
 • Institutional factors
Leuz, Nanda and Wysocki (2003) analyze accounting data for over 8,000 firms within 31
countries. They find that cross-national differences in the level of investor protection and the
quality of the legal system explain differences in earnings management. Specifically, they
show that the pervasiveness of earnings management increases in countries characterized by
inadequate outsider rights, weak legal enforcement, small stock markets, high ownership
concentration and low disclosure. In these countries, insiders have more incentives to manage
accounting earnings because low protection extends insiders ‘ability to acquire private control
benefits.
Burgstahler, Hail and Leuz (2006)’s study lead to a similar conclusion. The authors
document differences in earnings management across private and public firms from 13
Europeans countries. They show higher levels of earnings management for private firms than
for public firms. For public companies, a strong system of securities regulation reduces
earnings management more than the country legal system. Their findings re-establish that
institutional factors, such as such as book-tax alignment, outside investor protection, and
capital market structure, impact earnings quality of private and public firms differently.
Further, they show that legal systems and capital market forces reinforce each other to deter
managers from reporting earnings that do not accurately reflect economic performance in both
private and public firms. Although, other papers can undoubtedly be cited, the large majority
of the publications on the topic reach almost the same conclusions (Haw, Hu, Hwang and Wu,
2004; Gopalan and Jayaraman, 2012). Accordingly, we hypothesize that accruals
management in high sentiment periods is more pronounced than in low sentiment periods in
countries with weaker outside shareholder rights, lower legal enforcement, lower equity
market development, concentrated ownership, and higher allowance for accrual accounting.
Similarly to what was done for the accruals anomaly, we now examine whether our results
established in section 4 depending on some determinants of market integrity. To carry this
investigation, we operate as previously stated, i.e. we re-estimate the equation (6) in dividing
our sample of countries into two groups using market integrity. Findings are depicted in Table
7. As expected, we observe that accruals management in periods of high sentiment is higher
than in low sentiment periods in countries with weak and less-enforced outsider rights, small
stock markets and high allowance for accrual accounting. Contrary to what was anticipated,
we find that accruals management is higher in countries with low concentrated ownership.

-19-
Our result, however, can be reconciled with efficient monitoring hypothesis12 which suggests
that small shareholders extend the scope of managerial opportunism. Indeed, small
shareholders do not have strong motivation to scrutinize and influence firm management as
their investment is not substantial enough to justify such actions.
• Cultural factors
Several studies also establish that cultural factors contribute to explain the difference in
earnings management across countries. These studies conclude that certain cultural
dimensions are significantly related to earnings management, even after controlling for
investors’ protection and other legal institutional factors. In particular, individualism and
uncertainty avoidance appear to be the cultural dimensions significantly related to earnings
management. Findings, however, lead to conflicting results regarding the relationship
(positive or negative) between each cultural dimension and earnings management.
Gray’s (1988) model is a theoretical framework used by several studies to explain
differences in accruals management across counties. The model associates four accounting
values of professionalism, uniformity, conservatism, and secrecy to systems with Hofstede’s
(1980) cultural constructs. According to Gray (1988), individualism and uncertainty
avoidance are the most variables significantly related to these accounting values and can act
as proxies for them. In highly individualistic countries, accountants have a lot of autonomy in
measurement (non-uniformity, i.e. flexibility) and substantial freedom to rely on their
personal judgment (professionalism). The only cultural restriction they faced is their
obligation to disclose information (transparency). In these countries, earnings management by
accountants is thus expected to be more frequent than in collectivistic countries. In highly
uncertainty avoidant countries, accountants are compelled to use standardized rules
(uniformity) discouraged from relying on their personal judgment (professionalism) and
required to be cautious with measurement (conservatism). In these countries, conservatism
and uniformity should lead to fewer opportunities for earnings management.
Gray’s (1988) theoretical framework has been empirically tested by Guan, Pourjalali,
Sengupta and Teruya (2005). Using data from five Asian countries, they find that the cultural
dimensions of power distance and individualism are positively associated with earnings
management whereas the cultural dimensions of uncertainty avoidance and long term
orientation are negatively associated with earnings management. In a similar study, Han,

12
This hypothesis contrasts with the expropriation hypothesis which states that in firms with concentrated
ownership conflicts of interest between majority and minority shareholders are likely to emerge (Shleifer and
Vishny, 1997). These conflicts may end up with the expropriation of minority shareholders by the majority
shareholders encouraging managers to engage in earnings management for their private benefits.

-20-
Kang, Salter and Yoo (2010) test the Gray’s (1988) model using a large sample of 32
countries. They report that the institutional structures as well as cultural values of
individualism and uncertainty avoidance contribute to explain earnings management. Findings
also show that the influence of both factors on earnings management is conditional on each
other, i.e. the positive (negative) relationship between individualism (uncertainty avoidance)
and earning management is particularly pronounced in strong investor protection regimes.
The impact of individualism and uncertainty avoidance on earnings management is
debated and controversial. Based on Hofstede’s (1980) research, different alternatives
hypotheses about the relationship between cultural values and earnings management have
been proposed. According to Hofstede (1980), individualism is the rejection of hierarchy and
discrimination. In low individualism countries (high collectivism), people are often expected
to show excessive loyalty to the extended family. This type of behavior, unfortunately, has
been shown to lead to nepotism and corruption (Hofstede, 1980; Husted, 1999; Licht
Goldschmidt and Schwartz, 2005). In these countries, employees want to be treated like a
family member, i.e. they desire to be taking care of and protected by the organization. Hence,
the hypothesis that low levels of individualism should lead to higher levels of earnings
management. Hofstede’s (1980) also emphasizes that in high uncertainty avoidant countries,
there is low tolerance for uncertainty and ambiguity. People in these countries aspire to
certainty and clarity to satisfy their needs to have control over their future. In this context,
earnings management should be used by managers to control the future of their firms, high
levels of uncertainty avoidance should thus lead to more earnings management.
Doupnik (2008) examine the effect of Hofstede’s five cultural dimensions on earnings
management. He finds that earnings management is negatively associated with individualism
and positively associated with uncertainty avoidance. Power distance, masculinity and long-
term orientation are not significant in explaining the international differences in earnings
management. In addition, he finds that the cultural dimensions have a stronger impact on
earnings management than legal institutional factors. Notice however that the effect seems
stronger with earnings smoothing than with earnings discretion. In a similar study, Callen,
Morel and Richardson (2011) investigate the impact of culture as defined by Hofstede and
religion on earnings management. They find that earnings management is negatively related
to individualism and positively related to uncertainty avoidance. In contrast, earnings
management is unrelated to power distance, masculinity, long-term orientation and both
religion affiliation and the degree of religiosity. Notably, their finding also establishes that
after controlling for culture, the result of Leuz, Nanda and Wysocki (2003) that high legal

-21-
enforcement lead to less earnings management is not validated. To sum-up, there is no
consensus in studies examining the relationship between cultural dimensions and earnings
management, so our hypothesis is non-directional and states that: the relationship between
investor sentiment and total accruals is related to the degree of individualism and uncertainty
avoidance.
For cultural factors, Table 7 shows a positive and significant relationship between
investor sentiment and total accruals only in low individualism societies and in societies with
high levels of uncertainty avoidance. The positive relationship between investor sentiment
and accruals management in collectivist societies confirms that managers in low
individualism countries exhibit allegiance to the extended family which, in turn, leads to the
development of powerful networks of kinship and nepotism susceptible to lead to earnings
management. The positive relationship between investor sentiment and accruals management
in societies with high levels of uncertainty avoidance explain people uneasiness when faced
with uncertain circumstance. Earnings management could thus be interpreted as a mechanism
that eases uncertainty.
Our results support that culture is an important environmental factor influencing earnings
management. Notice also that when a country is characterized by high level of accruals
mispricing, it also displays significant amount of earnings management. This is a critical
result as it suggests that the accruals anomaly may be due to the use of accruals to manage
earnings.
Finally, we also have run regressions on the full sample of countries and introduced in the
Model (5) interactive terms between the variables identifying institutional factors, herd-like
behavior and sentiment. Results13 show that the impact of the sentiment variable is
discriminate by the adjunction of the interactive terms. This finding confirms that differences
between the groups of countries are significant for both institutional and cultural factors. We
also note that using current accruals in the different models used in this section instead of total
accruals does not change qualitatively the results.

Conclusion

Prior literature has related limited investor attention to stock market misreactions to
accounting information and to managers' opportunistic disclosure behavior (Stein, 1996;
Hirshleifer and Teoh, 2003; Hirshleifer, Lim and Teoh, 2009). In this study, we examine

13
We do not report the results for brevity but are available on request.

-22-
whether limited investor attention, measured by market-wide investor sentiment, affects
accruals anomaly in a European setting. Limited investor attention is expected to increase
(decrease) in high (low) sentiment periods as a result of individual investors’ overoptimistic
(over-pessimistic) belief about stock future performance. Moreover, limited attention in high
(low) sentiment periods is also expected to reduce (increase) individual investors’ concerns
about the losses they could experience from mispricing.
We analyze a sample of 46,386 firm-year observations from 15 European countries for the
period between 1994 and 2011. In line with recent evidence for the U.S., we show that
investor sentiment causes accruals mispricing across European countries. The effect is
pronounced for stocks whose valuations are highly subjective and difficult to arbitrage,
consistent with the predictions of noise trader’s models. Specifically, we find that mispricing
per unit of accruals for stocks hard to value and difficult to arbitrage is greater in periods
when investor sentiment is high then in periods when investor sentiment is low. This result is
due to naïve individual investors’ fixation, in high sentiment periods, on the total amount of
reported earnings without regard to the persistent differential between accruals and cash
flows. Our results also show that accruals reported for stocks hard to value and difficult to
arbitrage are higher in high sentiment periods as compared with low sentiment periods. This
additional result supports the opportunistic behavior hypothesis stating that managers report
higher accruals in high sentiment periods to take advantage of investors’ greater overvaluation
during these periods.
In subsequent empirical analyses, we hypothesize and test whether the differences in
market integrity and culture across countries explain the magnitude of accruals mispricing and
managers’ decisions to manage earnings. Our results indicate that both institutional factors
and cultural factors have explanatory power for accruals anomaly in the European countries
sampled. In particular, we find that investor sentiment causes accruals mispricing in countries
characterized by weaker outside shareholder rights, lower legal enforcement, lower equity
market development, dispersed ownership, higher allowance of accrual accounting and in
countries where herd-like behavior and overreaction behavior are strong. These countries that
present significant accruals mispricing in high sentiment periods also display high level of
accruals management, suggesting that accruals management is an important source of accruals
mispricing for individual investors.
Our research has implications for accruals anomaly research in at least fourth contexts.
First, findings suggest that the stock market misreactions to accounting accruals information
are related to periods of high investor sentiment. Hence, it is important for future studies to

-23-
consider investor sentiment as an important time-series determinant of the accruals anomaly,
particularly for stocks hard to value and difficult to arbitrage. Second, findings suggest the
generalizability of the accruals anomaly in European countries characterized by different
levels of economic development, legal-tradition and accounting standards. Evidence on the
pervasiveness of the accrual anomaly in these diverse countries contributes to evaluate the
adequacy of the capital market theory. With respect to accounting information, the accruals
anomaly remains a challenge to the capital market theory. Third, findings enhance our
comprehension of why the anomaly occurs. The scrutiny of the magnitude of the relationship
between the occurrence of the anomaly and cross-country differences in institutional
structures and accounting regimes, provide insights into the informational, corporate
governance, and capital market factors most associated with the anomaly. This is potentially
beneficial in setting and regulating financial accounting and reporting standards. For example,
more transparent and timely reporting of accruals might be used to mitigate the potential for
market mispricing of accounting information. Fourth, findings suggest to caution individuals
investors. These investors would be wise to take into account the impact of mood on the
performance of their portfolio. They must keep in mind that periods of high optimism are
accompanied by high level of accruals and followed by low future stock returns.

References

Alford, A., Jones, J., Leftwich, R., & Zmijewski, M. 1993. The Relative Informativeness of
Accounting Disclosures in Different Countries. Journal of Accounting Research,
31(3): 183-223.
Ali, A., & Gurun, U. G. 2009. Investor Sentiment, Accruals Anomaly, and Accruals
Management. Journal of Accounting, Auditing & Finance, 24(3): 415-431.
Ali, A., & Hwang, L.-S. 2000. Country-Specific Factors Related to Financial Reporting and
the Value Relevance of Accounting Data. Journal of Accounting Research, 38(1): 1-
21.
Ang, A., & Bekaert, G. 2007. Stock Return Predictability: Is it There? Review of Financial
Studies, 20(3): 651-707.
Ashbaugh, H., LaFond, R., & Mayhew, B. W. 2003. Do Nonaudit Services Compromise
Auditor Independence? Further Evidence. Accounting Review, 78(3): 611-639.
Baker, M., & Wurgler, J. 2006. Investor Sentiment and the Cross-Section of Stock Returns.
Journal of Finance, 61(4): 1645-1680.
Baker, M., & Wurgler, J. 2007. Investor Sentiment in the Stock Market. Journal of
Economic Perspectives, 21(2): 129-151.
Barber, B. M., & Odean, T. 2008. All That Glitters: The Effect of Attention and News on the
Buying Behavior of Individual and Institutional Investors. Review of Financial
Studies, 21(2): 785-818.
Beneish, M. D. 1998. Discussion of "Are Accruals during Initial Public Offerings
Opportunistic?" Review of Accounting Studies, 3(1/2): 209-221.

-24-
Bergman, N. K., & Roychowdhury, S. 2008. Investor Sentiment and Corporate Disclosure.
Journal of Accounting Research, 46(5): 1057-1972.
Botsari, A., & Meeks, G. 2008. Do Acquirers Manage Earnings Prior to a Share for Share
Bid? Journal of Business Finance & Accounting, 35(5/6): 633-670.
Burghstahler, D. C., Hail, L., & Leuz, C. 2006. The Importance of Reporting Incentives:
Earnings Management in European Private and Public Firms. Accounting Review,
81(5): 983-1016.
Callen, J. L., Morel, M., & Richardson, G. 2011. Do culture and religion mitigate earnings
management? Evidence from a cross-country analysis. International Journal of
Disclosure & Governance, 8(2): 103-121.
Chui, A. C. W., Titman, S., & Wei, K. C. J. 2010. Individualism and Momentum around the
World. Journal of Finance, 65(1): 361-392.
Collins, D. W., & Hribar, P. 2002. Errors in Estimating Accruals: Implications for Empirical
Research. Journal of Accounting Research, 40(1): 105-134.
Coopers, & Lybrand. 1993. International Accounting Summaries: A Guide for Interpretation
and Comparison Wiley; 2 edition
D'Avolio, G. 2002. The market for borrowing stock. Journal of Financial Economics,
66(2/3): 271-306.
Dechow, P. M., Sloan, R. G., & Sweeney, A. P. 1995. Detecting Earnings Management.
Accounting Review, 70(2): 193-225.
Defond, M. L., & Park, C. W. 2001. The Reversal of Abnormal Accruals and the Market
Valuation of Earnings Surprises. Accounting Review, 76(3): 375-405.
Doupnik, T. S. 2008. Influence of Culture on Earnings Management: A Note. Abacus, 44(3):
317-340.
Fama, E. F., & French, K. R. 1992. The Cross-Section of Expected Stock Returns. Journal of
Finance, 47(2): 427-465.
Gassen, J., & Markarian, G. 2009. Investor Fear and Earnings Management:VIX-based
Evidence. Unpublished Working Paper.
Gopalan, R., & Jayaraman, S. 2012. Private Control Benefits and Earnings Management:
Evidence from Insider Controlled Firms. Journal of Accounting Research, 50(1):
117-157.
Gray, S. J. 1988. Towards a Theory of Cultural on the Development of Accounting Influence
Systems Internationally. Abacus, 24(1): 1-15.
Guan, L., Pourjalali, H., Sengupta, P., & Teruya, J. 2005. Effect of Cultural Environment on
Earnings Manipulation: A Five Asia-Pacific Country Analysis. Multinational
Business Review (St. Louis University), 13(2): 23-41.
Han, S., Kang, T., Salter, S., & Yong Keun, Y. 2011. A cross-country study on the effects of
national culture on earnings management. Journal of International Business Studies,
41(1): 123-141.
Haw, I.-M., Hu, B., Hwang, L.-S., & Wu, W. 2004. Ultimate Ownership, Income
Management, and Legal and Extra-Legal Institutions. Journal of Accounting
Research, 42(2): 423-462.
Hirshleifer, D., Hou, K., & Hong Teoh, S. 2012. The Accrual Anomaly: Risk or Mispricing?
Management Science, 58(2): 320-335.
Hirshleifer, D., Hou, K., & Teoh, S. H. 2009. Accruals, Cash Flows, and Aggregate Stock
Returns. Journal of Financial Economics, 91(3): 389-406.
Hirshleifer, D., Hou, K., Teoh, S. H., & Zhang, Y. 2004. Do investors overvalue firms with
bloated balance sheets? Journal of Accounting & Economics, 38(1-3): 297-331.
Hirshleifer, D., Lim, S. S., & Teoh, S. H. 2009. Driven to Distraction: Extraneous Events and
Underreaction to Earnings News. Journal of Finance, 64(5): 2289-2325.

-25-
Hirshleifer, D., Lim, S. S., & Teoh, S. H. 2011. Limited Investor Attention and Stock Market
Misreactions to Accounting Information. Review of Asset Pricing Studies, 1(1): 35-
73.
Hirshleifer, D., & Teoh, S. H. 2003. Limited Attention, Information Disclosure, and Financial
Reporting. Journal of Accounting and Economics, 36(1-3): 337-386.
Hofstede, G. 1980. Cultures Consequences: International Differences in Work Related
Values. Beverly Hill, CA: Sage Publications
Hofstede, G. 2001. Culture’s Consequences: Comparing Values, Behaviors, Institutions, and
Organizations Across Nations. Sage Publication, Beverly Hills.
Houge, T., & Loughran, T. 2000. Cash Flow Is King? Cognitive Errors by Investors. Journal
of Psychology & Financial Markets, 1(3/4): 161-175.
Hribar, P., & McInnis, J. 2012. Investor Sentiment and Analysts' Earnings Forecast Errors.
Management Science, 58(2): 293-307.
Hung, M. 2001. Accounting standards and value relevance of financial statements: An
international analysis. Journal of Accounting & Economics, 30(3): 401-420.
Husted, B. W. 1999. Wealth, Culture, and Corruption. Journal of International Business
Studies, 30(2): 339-359.
Ince, O. S., & Porter, R. B. 2006. Individual equity return data from Thomson DataStream:
handle with care! Journal of Financial Research, 29(4): 463-479.
Kahneman, D. 1973. Attention and effort Englewood Cliffs: Prentice-Hall.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. 1998. Law and finance.
Journal of Political Economy, 106(6): 1113-1150.
LaFond, R. 2005. Is the Accrual Anomaly a Global Anomaly?: Massachusetts Institute of
Technology (MIT), Sloan School of Management, Working papers.
Leippold, M., & Lohre, H. 2011. Data Snooping and the Global Accrual Anomaly. Applied
Financial Economics, 22(7-9): 509-535.
Lemmon, M., & Portniaguina, E. 2006. Consumer Confidence and Asset Prices: Some
Empirical Evidence. Review of Financial Studies, 19(4): 1499-1529.
Leuz, C., Nanda, D., & Wysocki, P. D. 2003. Earnings management and investor protection:
an international comparison. Journal of Financial Economics, 69(3): 505-527.
Libby, R., Bloomfield, R., & Nelson, M. W. 2002. Experimental research in financial
accounting. Accounting, Organizations & Society, 27(8): 775-810.
Licht, A. N., Goldschmidt, C., & Schwartz, S. H. 2005. Culture, Law, and Corporate
Governance. International Review of Law and Economics, 25(2): 229-255.
Liodakis, M., Brar, G., Gadaut, J., & Sharma, J. 2004. Searching for alpha-quantifying
earnings quality. Working paper, Citigroup Smith Barney Industry Report.
Livnat, J., & Petrovits, C. 2009. Investor sentiment, post earnings announcement drift, and
accruals. Unpublished working paper, New York University.
Lucey, B. M., & Zhang, Q. 2010. Does cultural distance matter in international stock market
comovement? Evidence from emerging economies around the world. Emerging
Markets Review, 11(1): 62-78.
Matsumoto, D. A. 2002. Management's Incentives to Avoid Negative Earnings Surprises.
Accounting Review, 77(3): 483-514.
Mian, G. M., & Sankaraguruswamy, S. 2012. Investor Sentiment and Stock Market Response
to Earnings News. Accounting Review, 87(4): 1357-1384.
Papanastasopoulos, G. 2011. Accounting Accruals and Stock Returns: Evidence from
European Equity Markets. Unpublished Working paper.
Pincus, M., Rajgopal, S., & Venkatachal, M. 2007. The Accrual Anomaly: International
Evidence. Accounting Review, 82(1): 169-203.

-26-
Pontiff, J. 1995. Closed-end fund premia and returns Implications for financial market
equilibrium. Journal of Financial Economics, 37(3): 341-370.
Qiu, L., & Welch, I. 2006. Investor Sentiment Measures. Brown University and NBER.
Rajgopal, S., Shivakumar, L., & Simpson, A. 2007. A Catering Theory of Earnings
Management. Unpublished Working Paper.
Richardson, S. A., Sloan, R. G., Soliman, M. T., & Tuna, Ä. r. 2005. Accrual reliability,
earnings persistence and stock prices. Journal of Accounting & Economics, 39(3):
437-485.
Schmeling, M. 2009. Investor sentiment and stock returns: Some international evidence.
Journal of Empirical Finance, 16(3): 394-408.
Schwartz, N. 2002. Situated Cognition and the wisdom of feelings: Cognitive Tuning. In B.
Feldman & P. Salovey (Eds.), The wisdom of feelings (pp. 144 - 166). New York:
Guilford Press.
Shleifer, A., & Vishny, R. W. 1997. The Limits of Arbitrage. Journal of Finance, 52(1): 35.
Sloan, R. G. 1996. Do Stock Prices Fully Reflect Information in Accruals and Cash Flows
About Future Earnings? Accounting Review, 71(3): 289-315.
Stambaugh, R. F., Yu, J., & Yuan, Y. 2012. The Short of It: Investor Sentiment and
Anomalies. Journal of Financial Economics, 104(2): 288-302.
Stein, J. C. 1996. Rational capital budgeting in an irrational world. Journal of Business,
69(4): 429-455.
Tiedens, L. Z., & Linton, S. 2001. Judgment Under Emotional Certainty and Uncertainty: The
Effects of Specific Emotions on Information Processing. Journal of Personality &
Social Psychology, 81(6): 973-988.
Wurgler, J., & Zhuravskaya, E. 2002. Does Arbitrage Flatten Demand Curves for Stocks?
Journal of Business, 75(4): 583-608.
Xie, H. 2001. The Mispricing of Abnormal Accruals. Accounting Review, 76(3): 357-373.
Zouaoui, M., Nouyrigat, G., & Beer, F. 2011. How Does Investor Sentiment Affect Stock
Market Crises? Evidence from Panel Data. Financial Review, 46(4): 723-747.

-27-
 Table 1: Descriptive statistics

The full sample consists of 46,386 firm-year observations for the period 1994 to 2011 across 15 European countries and 4,787 European firms. Financial firms and utilities are
excluded from our sample. To be included in our sample, a complete set of desired data is required for a firm-year observation. Financial and accounting information are
obtained from the Datastream/Worldscope database. Total accruals are calculated as: (∆current assets - ∆cash) - (∆current liabilities- ∆short-term debt - ∆taxes payable) -
depreciation expense, scaled by lagged assets. Current Accruals are measured as: (∆current assets - ∆cash) - (∆current liabilities- ∆short-term debt - ∆taxes payable), scaled
by lagged assets. Prior Returns is measured by 6-month cumulative stock returns preceding the future return measurement period. Market value is measured as year-end stock
price times the number of shares outstanding. Book-to market ratio is calculated as the the book value of common equity divided by market equity. Age is the number of
months since the firm’s first appearance in our database. Profitability is captured by the return on assets defined as earnings divided by total assets. Tangibility is captured by
property, plant and equipment over total assets. Dividend policy is approximated by the ex-date dividends per share multiplied by shares outstanding divided by book equity.

Current Market Value Book-to Market Age Dividend

Total Accruals Prior Returns Profitability Tangibility
Accruals (in millions Euros) Ratio (in months) Policy
Mean -0.037 0.026 0.070 855.97 0.852 68.303 0.005 0.311 0.389
Medium -0.040 0.006 0.046 44.05 0.549 71 0.048 0.165 0.376
Std. Dev. 1.022 0.985 0.346 2388.66 0.488 39.85 0.252 0.553 0.391
Min -0.298 -0.222 -0.324 4.853 0.442 1.292 -4.261 0.226 0.000
Max 0.213 0.297 0.794 1845.98 1.541 133.198 1.231 0.883 0.123

-28-
 Table 2: Investor sentiment and accruals mispricing (Total accruals)

Panel A of this table reports the results of Fama-MacBeth regressions of Model (4) with country dummies. The
statistics tabulated in parentheses correspond to the p-values. The dependent variable, Eret, is stock return in
excess of market return. Dummy is a dummy variable that takes the value of one when firm characteristic (i.e.
size, age, profitability, tangibility and dividend policy) is below the median for the month t, and is zero
otherwise. Total accruals are calculated as: (∆current assets - ∆cash) - (∆current liabilities – ∆short - term debt -
∆taxes payable) - depreciation expense, scaled by lagged assets. Prior Returns is measured by 6-month
cumulative stock returns preceding the future return measurement period. Book-to-Market ratio is measured as
the book value of equity divided by the market value of equity.
Panel B reports the regression estimates of model (5), where the dependent variable, , is the incremental
mispricing per unit of accruals of stocks hard to value and to arbitrage relative to stocks easy to value and to
arbitrage, obtained from Fama-MacBeth regressions of Model (4). Sent┴ is the component of consumer
confidence index that is orthogonal to economic fundamentals.
Panel C reports the regression estimates of Model (7), where the dependent variable, , obtained from Fama-
MacBeth regressions of Model (4). The sample consists of 46,386 firm-year observations for the period 1994 to
2011 across 15 European countries.

Panel A: The dependent variable is stock return in excess of market return for month t+1, Erett+1
Firm characteristics
Size Age Profitability Tangibility Dividend Policy
0.768 0.827 1.023 0.927 0.751
Intercept
(0.001) (0.001) (0.000) (0.000) (0.005)
1.628 1.346 1.123 1.987 1.098
Dummy
(0.021) (0.042) (0.048) (0.011) (0.072)
-2.354 -2.732 -2.423 -2.825 -2.223
Total Accruals
(0.023) (0.012) (0.019) (0.011) (0.029)
0.721 0.736 0.724 0.742 0.688
Prior Ret
(0.000) (0.000) (0.000) (0.000) (0.000)
0.245 0.203 0.235 0.282 0.231
Book-to-Market
(0.021) (0.051) (0.032) (0.019) (0.041)
-0.135 -0.123 -0.101 -0.112 -0.099
Dummy × Total Accruals
(0.234) (0.329) (0.423) (0.345) (0.542)
R2 Adjusted 0.020 0.018 0.011 0.024 0.012
Panel B: The dependent variable is the incremental mispricing per unit of accruals of stocks hard to value and to
arbitrage relative to stocks easy to value,
-0.502 -0.487 -0.498 -0.673 -0.522
Intercept
(0.322) (0.287) (0.325) (0.234) (0.294)
-1.082 -1.112 -1.091 -1.237 -1.076
Sent┴
(0.042) (0.036) (0.041) (0.019) (0.049)
R2 Adjusted 0.011 0.012 0.010 0.014 0.009
Panel C: The dependent variable is the incremental mispricing per unit of accruals of stocks hard to value and to
arbitrage relative to stocks easy to value,
-0.723 -0.812 -0.731 -0.693 -0.894
Intercept
(0.302) (0.212) (0.298) (0.356) (0.184)
-0.541 -0.556 -0.544 -0.612 -0.806
Sent┴
(0.298) (0.358) (0.242) (0.254) (0.136)
R2 Adjusted 0.001 0.001 0.001 0.002 0.003

-29-
 Table 3: Investor sentiment and accruals mispricing (Current accruals)

Panel A of this table reports the results of Fama-MacBeth regressions of Model (4) with country dummies. The
statistics tabulated in parentheses correspond to the p-values. The dependent variable, Eret, is stock return in
excess of market return. Dummy is a dummy variable that takes the value of one when firm characteristic (i.e.
size, age, profitability, tangibility and dividend policy) is below the median for the month t, and is zero
otherwise. Current Accruals are measured as (∆current assets - ∆cash) - (∆current liabilities- ∆short-term debt -
∆taxes payable), scaled by lagged assets. Prior Returns is measured by 6-month cumulative stock returns
preceding the future return measurement period. Book-to-Market ratio is measured as the book value of equity
divided by the market value of equity.
Panel B reports the regression estimates of model (5), where the dependent variable, , is the incremental
mispricing per unit of accruals of stocks hard to value relative to stocks easy to value, obtained from of Fama-
MacBeth regressions of Model (4). Sent┴ is the component of consumer confidence index that is orthogonal to
economic fundamentals.
Panel C reports the regression estimates of Model (7), where the dependent variable, , obtained from Fama-
MacBeth regressions of Model (4). The sample consists of 46,386 firm-year observations for the period 1994 to
2011 across 15 European countries.

Panel A: The dependent variable is stock return in excess of market return for month t+1, Erett+1
Firm characteristics
Size Age Profitability Tangibility Dividend Policy
0.743 0.822 1.041 0.899 0.692
Intercept
(0.001) (0.001) (0.000) (0.000) (0.005)
1.611 1.309 1.119 1.986 1.093
Dummy
(0.022) (0.043) (0.047) (0.011) (0.073)
-2.523 -2.826 -2.519 -2.822 -2.324
Current Accruals
(0.019) (0.009) (0.012) (0.010) (0.021)
0.701 0.683 0.709 0.711 0.699
Prior Ret
(0.000) (0.000) (0.000) (0.000) (0.000)
0.248 0.204 0.232 0.276 0.218
Book-to-Market
(0.020) (0.051) (0.033) (0.021) (0.043)
-0.102 -0.099 -0.073 -0.112 -0.099
Dummy × Current Accruals
(0.436) (0.536) (0.643) (0.345) (0.542)
R2 Adjusted 0.019 0.012 0.007 0.019 0.009
Panel B: The dependent variable is the incremental mispricing per unit of accruals of stocks hard to value and to arbitrage
relative to stocks easy to value,
-0.524 -0.501 -0.534 -0.693 -0.606
Intercept
(0.312) (0.262) (0.313) (0.309) (0.298)
-1.066 -1.073 -1.043 -1.219 -1.098
Sent┴
(0.043) (0.034) (0.056) (0.021) (0.033)
R2 Adjusted 0.010 0.009 0.009 0.012 0.008
Panel C: The dependent variable is the incremental mispricing per unit of accruals of stocks hard to value and to arbitrage
relative to stocks easy to value,
-0.734 -0.833 -0.752 -0.706 -0.900
Intercept
(0.301) (0.208) (0.264) (0.324) (0.189)
-0.612 -0.587 -0.572 -0.724 -0.824
Sent┴
(0.223) (0.332) (0.239) (0.192) (0.133)
R2 Adjusted 0.002 0.001 0.001 0.003 0.003

-30-
 Table 4: Investor sentiment and accruals management (Total accruals)
This table reports regression results of Model (6) using pooled data across time and firms in all countries. The
statistics tabulated in parentheses correspond to the p-values. The dependent variable is Total accruals calculated
as: (∆current assets - ∆cash) - (∆current liabilities – ∆short - term debt - ∆taxes payable) - depreciation expense,
scaled by lagged assets. Sent┴ is the component of consumer confidence index that is orthogonal to economic
fundamentals.
Dummy is a dummy variable that takes the value of one when firm characteristic (i.e. size, age, profitability,
tangibility and dividend policy) is below the median for the year t, and is zero otherwise. LTotalAccruals is prior
year total accruals used to control for the accrual reversal. MA is an indicator variable set to 1 if the company
has engaged in a merger/acquisition activity, and zero otherwise. Financing is one if MA is not equal to one and
the number of outstanding shares increased by at least 10 percent, or long-term debts increased by at least 20
percent, or the firm first appears on the Thomson monthly returns database during the fiscal year, and zero
otherwise. Litigation is an indicator variable, controlling for ex-ante litigation risk, that equals one if the firm
operates in a high-litigation risk industry, and zero otherwise. Leverage equals the total debt divided by total
assets at the beginning of the fiscal year. MB is market-to-book ratio, defined as the market value of the firm’s
equity scaled by the book value of its common equity at the end of year. Loss is an indicator variable set to 1 if
the firm reports a net loss for the fiscal year, 0 otherwise. CFO is cash flow from operations, divided by
beginning of year total assets. Industry are indictors based on the Thomson database industry groups. The sample
consists of 46,386 firm-year observations for the period 1994 to 2011 across 15 European countries.

Firm characteristics
Size Age Profitability Tangibility Dividend Policy
-0.023 -0.042 -0.021 -0.052 -0.062
Intercept
(0.001) (0.001) (0.002) (0.000) (0.000)
-0.005 -0.006 -0.005 -0.008 -0.002
Sent┴
(0.002) (0.001) (0.001) (0.004) (0.009)
0.008 0.009 0.005 0.009 0.002
Dummy
(0.002) (0.001) (0.008) (0.009) (0.012)
0.003 0.002 0.002 0.005 0.003
Dummy × Sent
(0.031) (0.039) (0.047) (0.027) (0.033)
0.033 0.042 0.028 0.039 0.024
LTotalAccruals
(0.000) (0.000) (0.000) (0.000) (0.000)
0.042 0.039 0.040 0.040 0.037
MA
(0.001) (0.001) (0.001) (0.001) (0.002)
0.032 0.029 0.037 0.036 0.039
Financing
(0.002) (0.002) (0.002) (0.002) (0.003)
0.004 0.003 0.004 0.005 0.002
Litigation
(0.000) (0.000) (0.000) (0.000) (0.000)
-0.018 -0.026 -0.032 -0.028 -0.021
Leverage
(0.000) (0.000) (0.000) (0.000) (0.000)
0.001 0.001 0.002 0.001 0.002
MB
(0.000) (0.000) (0.000) (0.000) (0.000)
-0.042 -0.040 -0.032 -0.039 -0.035
Loss
(0.006) (0.007) (0.003) (0.007) (0.009)
-0.022 -0.025 -0.021 -0.026 -0.029
CFO
(0.012) (0.018) (0.018) (0.011) (0.016)
Industry Yes Yes Yes Yes Yes
R2 Adjusted 0.151 0.143 0.126 0.185 0.124

-31-
 Table 5: Investor sentiment and accruals management (Current accruals)
This table reports regression results of Model (6) using pooled data across time and firms in all countries. The
statistics tabulated in parentheses correspond to the p-values. The dependent variable is Current accruals
calculated as: (∆current assets - ∆cash) - (∆current liabilities – ∆short - term debt - ∆taxes payable), scaled by
lagged assets. Sent┴ is the component of consumer confidence index that is orthogonal to economic
fundamentals.
Dummy is a dummy variable that takes the value of one when firm characteristic (i.e. size, age, profitability,
tangibility and dividend policy) is below the median for the year t, and is zero otherwise. LCurrentAccruals is
prior year total accruals used to control for the accrual reversal. MA is an indicator variable set to 1 if the
company has engaged in a merger/acquisition activity, and zero otherwise. Financing is one if MA is not equal to
one and the number of outstanding shares increased by at least 10 percent, or long-term debts increased by at
least 20 percent, or the firm first appears on the Thomson monthly returns database during the fiscal year, and
zero otherwise. Litigation is an indicator variable, controlling for ex-ante litigation risk, that equals one if the
firm operates in a high-litigation risk industry, and zero otherwise. Leverage equals the total debt divided by total
assets at the beginning of the fiscal year. MB is market-to-book ratio, defined as the market value of the firm’s
equity scaled by the book value of its common equity at the end of year. Loss is an indicator variable set to 1 if
the firm reports a net loss for the fiscal year, 0 otherwise. CFO is cash flow from operations, divided by
beginning of year total assets. Industry are indictors based on the Thomson database industry groups. The sample
consists of 46,386 firm-year observations for the period 1994 to 2011 across 15 European countries.

Firm characteristics
Size Age Profitability Tangibility Dividend Policy
-0.015 -0.028 -0.028 -0.044 -0.059
Intercept
(0.004) (0.005) (0.004) (0.003) (0.006)
-0.002 -0.004 -0.006 -0.007 -0.006
Sent┴
(0.008) (0.007) (0.001) (0.004) (0.009)
0.009 0.007 0.008 0.008 0.004
Dummy
(0.002) (0.007) (0.009) (0.007) (0.010)
0.002 0.004 0.004 0.006 0.004
Dummy × Sent
(0.030) (0.033) (0.044) (0.024) (0.031)
0.044 0.039 0.032 0.041 0.029
LCurrentAccruals
(0.000) (0.000) (0.000) (0.000) (0.000)
0.041 0.037 0.039 0.044 0.039
MA
(0.002) (0.001) (0.001) (0.000) (0.001)
0.035 0.031 0.034 0.034 0.033
Financing
(0.002) (0.003) (0.003) (0.002) (0.003)
0.003 0.002 0.004 0.005 0.002
Litigation
(0.000) (0.000) (0.000) (0.000) (0.000)
-0.015 -0.022 -0.029 -0.025 -0.020
Leverage
(0.000) (0.000) (0.000) (0.000) (0.000)
0.001 0.001 0.002 0.001 0.002
MB
(0.000) (0.000) (0.000) (0.000) (0.000)
-0.039 -0.035 -0.029 -0.031 -0.029
Loss
(0.004) (0.005) (0.003) (0.005) (0.006)
-0.024 -0.031 -0.029 -0.026 -0.033
CFO
(0.010) (0.015) (0.014) (0.010) (0.012)
Industry Yes Yes Yes Yes Yes
R2 Adjusted 0.154 0.146 0.129 0.189 0.129

-32-
 Table 6: Cross-country analysis of the sentiment-accruals mispricing
relation

This table presents the results of estimating the model (5) when countries are pooled according to one of the
determinants shown in the first column. For succinctness only the coefficients of the variable Sent┴ are
presented. The countries are allocated to one of two groups depending on whether they are above or below the
median of a specific determinant. Sent┴ is the component of consumer confidence index that is orthogonal to
economic fundamentals. The sample consists of 46,386 firm-year observations for the period 1994 to 2011
across 15 European countries. ***,**,* indicate statistical significance at the 0.01, 0.05 and 0.10 level,
respectively.

Panel A: Size
Countries below median Countries above median
Sent┴ Adj.R2 Sent┴ Adj.R2
Cultural factors
Individualism -1.202** 0.019 -0.924 0.005
Uncertainty avoidance -1.056* 0.009 -1.098** 0.012
Market integrity
Anti-director rights -1.192** 0.016 -0.904 0.004
Legal Enforcement -1.199** 0.019 -0.909 0.004
Importance of Stock Market -1.204** 0.021 -0.919 0.005
Ownership Concentration -1.107** 0.014 -0.892 0.003
Accrual Index -0.906 0.004 -1.194** 0.016

Panel B: Age
Countries below median Countries above median
Sent┴ Adj.R2 Sent┴ Adj.R2
Cultural factors
Individualism -1.218** 0.020 -0.926 0.005
Uncertainty avoidance -1.076* 0.010 -1.116** 0.013
Market integrity
Anti-director rights -1.204** 0.017 -0.936 0.004
Legal Enforcement -1.201** 0.015 -0.941 0.004
Importance of Stock Market -1.199** 0.015 -0.956 0.005
Ownership Concentration -1.109** 0.013 -0.933 0.003
Accrual Index -0.938 0.004 -1.197** 0.014

Panel C: Profitability
Countries below median Countries above median
Sent┴ Adj.R2 Sent┴ Adj.R2
Cultural factors
Individualism -1.212** 0.017 -0.969 0.005
Uncertainty avoidance -1.055* 0.007 -1.161** 0.014
Market integrity
Anti-director rights -1.098** 0.008 -0.945 0.004
Legal Enforcement -1.205** 0.016 -0.859 0.002
Importance of Stock Market -1.182** 0.014 -0.870 0.003
Ownership Concentration -1.199** 0.015 -0.854 0.002
Accrual Index -0.935 0.003 -1.109** 0.012

-33-
Panel D: Tangibility
Countries below median Countries above median
Sent┴ Adj.R2 Sent┴ Adj.R2
Cultural factors
Individualism -1.483** 0.025 -0.942 0.005
Uncertainty avoidance -1.082* 0.011 -1.382** 0.021
Market integrity
Anti-director rights -1.502** 0.026 -0.973 0.007
Legal Enforcement -1.487** 0.022 -0.965 0.006
Importance of Stock Market -1.391** 0.019 - 0.979 0.007
Ownership Concentration -1.462** 0.023 -0.972 0.007
Accrual Index - 1.041* 0.009 -1.252** 0.016

Panel E: Dividend Policy

Countries below median Countries above median
Sent┴ Adj.R2 Sent┴ Adj.R2
Cultural factors
Individualism -1.163** 0.016 -0.949 0.005
Uncertainty avoidance -1.013* 0.007 -1.137** 0.013
Market integrity
Anti-director rights -1.076** 0.008 -0.926 0.004
Legal Enforcement -1.180** 0.015 -0.841 0.002
Importance of Stock Market -1.158** 0.014 -0.852 0.003
Ownership Concentration -1.175** 0.015 -0.836 0.002
Accrual Index -0.916 0.004 -1.086** 0.011

-34-
 Table 7: Cross-country analysis of the sentiment-accruals management
relation

This table presents the results of estimating the model (6) when countries are pooled according to one of the
determinants shown in the first column. For succinctness only the coefficients of the interactive variable
Sent┴×Dummy are presented. The countries are allocated to one of two groups depending on whether they are
above or below the median of a specific determinant. Sent┴ is the component of consumer confidence index that
is orthogonal to economic fundamentals. Dummy is a dummy variable that takes the value of one when firm
characteristic (i.e. size, age, profitability, tangibility and dividend policy) is below the median for the year t, and
is zero otherwise. The sample consists of 46,386 firm-year observations for the period 1994 to 2011 across 15
European countries. ***,**,* indicate statistical significance at the 0.01, 0.05 and 0.10 level, respectively.

Panel A: Size
Countries below median Countries above median
Sent┴×Dummy Adj.R2 Sent┴×Dummy Adj.R2
Cultural factors
Individualism 0.006** 0.160 0.001 0.143
Uncertainty avoidance 0.003** 0.153 0.004** 0.155
Market integrity
Anti-director rights 0.005** 0.156 0.000 0.142
Legal Enforcement 0.005** 0.157 0.001 0.144
Importance of Stock Market 0.004** 0.154 0.000 0.143
Ownership Concentration 0.004** 0.153 0.000 0.142
Accrual Index 0.001 0.145 0.005** 0.154

Panel B: Age
Countries below median Countries above median
Sent┴×Dummy Adj.R2 Sent┴×Dummy Adj.R2
Cultural factors
Individualism 0.004** 0.149 0.000 0.138
Uncertainty avoidance 0.001* 0.137 0.002** 0.141
Market integrity
Anti-director rights 0.003** 0.145 0.000 0.137
Legal Enforcement 0.004** 0.147 0.001 0.139
Importance of Stock Market 0.003** 0.144 0.000 0.138
Ownership Concentration 0.002** 0.145 0.000 0.139
Accrual Index 0.001 0.141 0.003** 0.146

Panel C: Profitability
Countries below median Countries above median
Sent┴×Dummy Adj.R2 Sent┴×Dummy Adj.R2
Cultural factors
Individualism 0.005** 0.134 0.000 0.117
Uncertainty avoidance 0.002** 0.125 0.003** 0.129
Market integrity
Anti-director rights 0.004** 0.130 0.001 0.120
Legal Enforcement 0.004** 0.133 0.001 0.122
Importance of Stock Market 0.002* 0.127 0.000 0.118
Ownership Concentration 0.002* 0.124 0.000 0.118
Accrual Index 0.001 0.119 0.003** 0.131

-35-
Panel D: Tangibility
Countries below median Countries above median
Sent┴×Dummy Adj.R2 Sent┴×Dummy Adj.R2
Cultural factors
Individualism 0.007** 0.191 0.000 0.179
Uncertainty avoidance 0.002* 0.181 0.005** 0.183
Market integrity
Anti-director rights 0.008** 0.193 0.001 0.181
Legal Enforcement 0.006** 0.189 0.001 0.180
Importance of Stock Market 0.005** 0.184 0.000 0.178
Ownership Concentration 0.005** 0.183 0.000 0.180
Accrual Index 0.001 0.181 0.008** 0.192

Panel E: Dividend Policy

Countries below median Countries above median
Sent┴×Dummy Adj.R2 Sent┴×Dummy Adj.R2
Cultural factors
Individualism 0.004** 0.156 0.000 0.140
Uncertainty avoidance 0.002* 0.142 0.004** 0.151
Market integrity
Anti-director rights 0.006** 0.152 0.000 0.139
Legal Enforcement 0.005** 0.153 0.001 0.141
Importance of Stock Market 0.004** 0.150 0.000 0.140
Ownership Concentration 0.004** 0.149 0.000 0.139
Accrual Index 0.001 0.142 0.004** 0.149

-36-