Chapter No

Chapter No.1
INTRODUCTION
1.1 Background of the Study
Earnings management is a burning issue in the accounting world. To obtain the specific objectives for earning management, different strategies are proposed by accounting managers (Scott, 2003). Healy and Wahlen (1999) predicted that earning management in financial reports to either mislead the shareholders about the performance of firms’ economically or to affect the predetermined results that depend on stated accounting records. Earnings management is very helpful to increase profits and revenues artificially by applying accounting strategies (accrual earnings management). Accrual earnings management is done when books are maintained to report general public and manipulating the process of financial reporting to attain some private gain. Earnings management may incorporate the exploiting opportunities to roll out the accounting calculations that alter the earnings figure expressed in the financial statements. Accounting decisions can, in turn; influence earnings as they can influence the timing of transactions and the measurements used in financial reporting.
Earnings management incorporates the manipulation of firms’ earnings against a pre-determined objective. This objective can be driven by an inclination for more consistent earnings, in which circumstance management is supposed to be accomplished income smoothing. Opportunistic income smoothing can, in turn, demonstrate lower risk and increase an organization’s market importance.
Contemporary research on accruals management has inspected and observed a range of inspirations that lead to earnings manipulation (Healy and Wahlen, 1999). It has also been exposed that firms’ strategically operate earnings for a host of purposes that assortment from endeavoring to affect firms’ bonuses (Healy, 1985; Guidry et al., 1999; Gaver et al., 1995) to expanding earnings prior to seasoned equity offerings (SEOs) and initial public offerings (Rangan, 1998; Shivakumar, 2000). Hayn (1995) and Burgstahler and Dichev (1997) examined the earnings management dependable with income smoothing while Holthausen et al. (1995) scrutinize that directors may use accruals to change earnings over the time. Firms’ use earnings management to increase the firm’s stock price (Collins and Hribar, 2000), and to take lesser cost financing (Dechow et al., 1996). Beneish and Vargus (2002) provide evidence that high accruals are accompanying with insider sales of stocks whereas Philippon (2006) examined high accruals in firms’ where chief executive officer’s (CEOs) reward is secured to the worth of stock and option fortunes. Conversely, Dichev et al. (2013) show that directors have encouragements to include in earning management to escape specified forfeiture, earnings drops, and earnings vanished predictors’ projections influence. Chen et al. (2012) provide evidence that the idiosyncratic return volatility is completely related with the managerial discretion in relations of accruals.

Uncertainty has a vital influence on economic outcomes. An emerging literature examines how macroeconomic uncertainty a?ects the firms’ earning management (Baker and Bloom, 2013), business cycles (Bloom et al., 2011; Basu and Bundick, 2012; Bidder and Smith, 2012; Christiano et al., 2014; Bianchi et al., 2014), economic growth (Bachmann and Bayer, 2014), and equity prices and risk premia (Pastor and Veronesi, 2012). Accompanying with these studies, various papers examine how both aggregate and ?rm-level uncertainty affect parts of managerial earnings decision making, for example, cash flow, stock prices, research and development (R;D) expenditure (Bloom, 2009; Julio and Yook, 2012; Stein and Stone, 2014; Arif et al., 2016; Gulen and Ion, 2016).

Graham (et al., 2005) examines how ?rm-level uncertainty affects an important dimension of managerial earnings decision making: the reporting and administration of accounting earnings. Top executives give an impressive thoughtfulness regarding earnings and their management and an extensive group of literature in accounting and ?nance has examined in different circumstances in which earnings are figured out in order to meet management objectives around different corporate events.

Arif et al. (2016) and Gulen and Ion (2016) investigate how managers entrepreneurially manage their earnings around market participants’ uncertainty about ?rm value, therefore contributing to and connecting the literature on earnings management and the e?ect of economic uncertainty on ?rms’ earning decision making.

Utilizing ?rm-level variation in uncertainty about ?rms’ future projections, Arif et al. (2016) and Gulen and Ion (2016) records empirically that ?rms confronting comparatively higher amounts of uncertainty report more positive discretionary accruals (DA). Empirical evidence suggests that ?rms’ value option-implied volatility and the cross-sectional Jones model (Bartov et al., 2000) increased with return on assets (Kothari et al., 2005), yet, is robust to different measurements of uncertainty and earnings management. These empirical ?ndings are consistent with ?rms opportunistically managing earnings upward during uncertain circumstances. Consistent with the earnings-management hypothesis, they additionally show that the high levels of uncertainty discretionary accrual connection are most pronounced at ?rms whose CEOs have more incentives to manage earnings (Arif et al., 2016; Gulen and Ion, 2016).

Consistent with opportunism, (Ferri et al., 2017) document that this pattern is especially articulated when managers face more incentives or appreciate more ability to manage earnings. Conversely, with settings where variety in the quality of ?nancial reports influence the earnings coe?cients (Fischer and Verrecchia, 2000; Ferri et al., 2017). They focus on a mechanism through which variation in earnings coe?cients a?ects the reporting incentives. The effect of earnings’ value on reporting incentives measured by Abarbanell and Lehavy (2003), Strobl (2013), Jackson et al. (2016), and Fang et al. (2017) and their ?ndings suggest that uncertainty related distinction in earnings response coe?cient (ERCs) influences firms’ value to present motivating forces to achieve their income. These outcomes have new ramifications for the role of uncertainty may play in fluctuating managerial earnings decision making and their reporting.

Theoretical studies have documented numerous channels through which uncertainty may affect the accruals management of a firm. Yet, different studies forecast different signs of the association between firms’ volatility and earnings management. These modifications are mainly attributed to the essential assumption of firms’ volatility with respect to earnings management. The relationship between idiosyncratic volatility and accruals management is expected to be positive if the “synchronicity” or low idiosyncratic risk is related with high transparency or more opacity of firm-specific information. Roll (1988) pointed out that the relationship between idiosyncratic volatility and accruals management is positively strong with the firms. Several studies resulting to that find a documenting evidence that low idiosyncratic volatility is connected with poor information environment of the firm (Bartram et al., 2009; Wei and Zhang, 2006; Ali et al., 2003; Kelly, 2007). Conversely, various studies discuss that low idiosyncratic risk measure could capture a firm’s informational transparency (Bakke and Whited, 2010; and Hutton et al., 2009). This debate could be made that cross-sectional deviation in idiosyncratic risk across firms and its materialistic increase over the time is generally beyond the relentless control of firm’s management. However, if the increased idiosyncratic volatility is related to the poor information environment and to the point that the shareholders demand a better risk premium for holding shares, stocks, and bonds with more idiosyncratic volatility, then such firms face the greater cost of capital. Moreover, if this idiosyncratic risk accompanies variability in earnings, so it gives the poor information for such firms. In such conditions, management may have the incentive to accomplish earnings as an instrument to diminish the negative value of increased idiosyncratic volatility on firm’s stock price and financing costs.

In recent times, the significance of various studies have found the idiosyncratic volatility at firm level have been increased world-wide. For example, Campbell et al. (2001) investigate a significant increase in firm-level volatility comparative to market volatility. The evidence in Campbell and Shiller (1988) and Vuolteenaho (2002) leads to idiosyncratic risk being stimulated by cost shocks that influence the firm’s basic cash flows. Different papers record theatrical increases in idiosyncratic risk resulting deregulation of merchandise markets and increase in worldwide competition, growing the forces of innovative destruction in the economy (Gaspar and Massa, 2006; Irvine and Pontiff, 2009).
Stock returns in the United States (U.S) have become more volatile since 1960 due to improving the financial reporting quality. Rajgopal and Venkatachalam (2011) verify that the declining earnings quality is related to greater idiosyncratic return volatility. They use accrual-based earnings management and accrual quality as the proxy to evaluate the financial reporting quality. However, accrual-based earnings management is not the only procedure to control the earnings. Chang et al. (2015) prove that idiosyncratic volatility is positively accompanying with both accrual-based earnings management and real earnings management without contemplating exogenous shocks. Likewise, idiosyncratic risk is negatively (positively) connected with accrual-based earnings management in the global financial crisis (GFC) period and the opposite result goes out to be accurate when earnings management is correlated with the real earnings management. Consequently, these managers tend to shift from real earnings management performances to accrual-based earnings management performances.

In prior research, Durnev et al. (2000) attempted to relate ?rms’ ?nancing decisions to macroeconomic conditions. However, empirical evidence suggests that firms’ own uncertainty has stronger effects on firms’ earnings as compared to the aggregate uncertainty. Analogously, Badertscher et al. (2008) depicted that idiosyncratic uncertainty has an important role in determining firms’ earning, in particular, when firms’ manage their earnings more effectively. Generally speaking, ?rms’ accrual management decisions to volatile deviations in macroeconomic conditions and conjectures that ?rms’ accrual management is pro-cyclically allied to the macro economy states. In particular, Hackbarth et al. (2006) recommend that the borrowing results of ?rms are pro-cyclical. They also discuss that both the stride and the magnitude of capital structure fluctuations are signi?cantly familiarized by macroeconomic conditions. Although researchers have effectively expressed theoretical interlinkages between ?rms’ accrual management decisions and the uncertainty related to macroeconomic conditions, we have inadequate empirical evidence on this issue. There are a few previous studies that empirically scrutinize the response of ?rms’ earning management decisions to macroeconomic conditions (Hatzinikolaou et al., 2002; Baum et al., 2009; Caglayan and Rashid, 2013). These studies generally document that macroeconomic condition have a signi?cant and adverse impact on ?rm leverage. Rashid (2014) prove that idiosyncratic and macroeconomic uncertainty has a vital role to play in ?rms’ security and repurchase decisions are of abundant signi?cance to ?rm managers and representatives. If several costs and bene?ts connected with ?rms’ ?nancing decisions (debt borrowing, retentions, and equity ?nancing) are influenced by risk. This indicates that ?rms are probable to change their ?nancing decisions when they take the risk, which effects variations in ?rms’ earning management, in turn, affecting the importance of the ?rms.

Some other empirical studies have also documented a significant association between uncertainty and earnings management. Brainard, Shoven, and Weiss (1980) scrutinize that risk procedures are made through the capital asset pricing model (CAPM) have negative effects on the accrual-based management decisions of firms. Beaudry, Caglayan, and Schiantarelli (2001) develop a firm-level dataset for the UK to discover the effect of macroeconomic risk on the firms’ leverage. By creating combined uncertainty from macroeconomy using restrictive variances, they observe a significant and negative correlation between macroeconomic (aggregate) uncertainty and firm-specific (idiosyncratic) risk. Baum, Caglayan, and Talavera (2008) study the relationship between risks and firms’ capital expenditures using a panel of the US firms. They report that firms’ own risk and a CAPM based risk level have substantial negative effects on firms’ investment. Yet, the uncertainty measure constructed on the entire market performance has a positive impact on the firms’ investment and earning management decisions.

In exploring the effects of uncertainty on earnings, several researchers have observed the effects of idiosyncratic (firm-specific) uncertainty on the value enlargement of earnings. Conversely, empirical evidence suggests that the marginal effects of both idiosyncratic and aggregate uncertainty are relatively different. For example, Caballero and Pindyck (1996) provide evidence that aggregate uncertainty, which is connected with the complete macroeconomic environment, has a stronger effect on firms’ symmetry investment than the uncertainty linked with firms’ own business activities has. Similarly, Brainard, Shoven, and Weiss (1980) find that risk procedures caused through capital asset pricing model (CAPM) have negative effects on the investment decisions of firms.

As mentioned above, economic theory suggests that both macroeconomic and firm-specific uncertainties play an important role in accrual management of a firm. However, researchers have mainly focused on only firm-specific uncertainty while the role of uncertainty in firms’ earning management decision was also considered (e.g., (Dechow et al., 1996; Messod, 2001; Goel and Thakor, 2003). Likewise, Rashid (2017) concluded that the greatest exposure to both idiosyncratic and macroeconomic uncertainties tends to smaller investment spending of manufacturing firms. In this study, we first propose a theoretical model for firms’ accrual management decisions based on each individual of the firm by incorporating the role of both idiosyncratic (firm-specific) and macroeconomic uncertainty. We then estimate the proposed model for non-financial firms listed at Pakistan Stock Exchange, over the period 2000-2016. We predicted that a higher exposure to both idiosyncratic and macroeconomic uncertainties leads to higher earnings in Pakistani manufacturing firms. Also, we hypothesize that both types of uncertainties have differential effects on firms’ accrual management. In particular, we predict that the uncertainty related to macroeconomic conditions has a strong negative impact on firms’ accrual management in the Pakistani context.

1.2 Research Gaps
One of the most important gaps in the contemporary empirical literature is that researchers have widely examined the accrual management and uncertainty relationship for firm-level idiosyncratic volatility in industrialized countries but a little consideration has been paid to emerging countries. Consequently, empirical evidence on how firm-level earnings management responds to uncertainty in developing countries is rather finite. To empirically examine how idiosyncratic and macroeconomic uncertainty influence corporate firms’ earning management, firms working in Pakistan seem highly significant and fascinating case. In these days, Pakistan is suffering much from numerous economic and political issues. For example, trade deficit, energy crisis, anti-corruption movements, the war on terror, and the high level of debt burden affected great uncertainty about macroeconomic conditions. In prior literature, the impact of idiosyncratic risk has been empirically investigated on equity, investment, cash flows volatility, accrual anomaly, idiosyncratic volatility, stock return and incentives of firms. However, the effect of idiosyncratic risk has not been measured on accrual management on whole. Similarly, the combined effect of idiosyncratic and macroeconomic uncertainty has not been measured on accrual management of any country or in Pakistan.
1.3 Research Objectives
The study has the following objectives
To analyze the impact of idiosyncratic risk on accrual management.

To investigate the impact of macroeconomic risk on accrual management.

To examine the impact of the interaction between the firms’ specific risk and macroeconomic risk on accrual management.

1.4 Research Questions
In this study, we provide answers to the following questions
How much is an idiosyncratic risk important for accrual management?
What are the motives behind doing accrual management?
Do risks associated with macroeconomic conditions play any role in accrual management?
Does the one type of risk enhance the impact of the other type of risk on accrual management?
Do both types of risk affect differently accrual management?
1.5 Hypotheses Development
Idiosyncratic volatility has a part either in evidence transparency or opacity. Agreeing to Roll (1988), “synchronicity” or low idiosyncratic (firm-specific) volatility could be related to additional transparency or opacity of firm-specific information. Firm-specific price engagements taking private information is exploited more competently into stock prices and upcoming earnings by up-to-date risk arbitrageurs (Morck et al., 2000; Durnev et al., 2003; Ferreira and Laux, 2007). Morck et al. (2000) find price synchronicity in developing economies is better than that in industrialized countries and characteristic this to the strong stakeholder property privileges in industrialized economies. Durnev et al. (2000) show that greater firm-specific ambiguity parallels to informed swapping which they settle leads to stock prices pursuing essentials more closely. If firm-level volatility is excessively high, then the part of the stock price as an “indicator” of the accurate value of the firm is weakened and therefore managers will rely on accruals management to boost the informativeness of earnings. Agreed with the above indication, aim that companies with high idiosyncratic volatility are encouraged to increase the informativeness of their earnings through accruals management to decrease the spillover of volatility into stock returns and we construct the following hypothesis.

H1: Firms facing more idiosyncratic uncertainty are likely to engage more in earnings management.

Jin and Myers (2006) and Hutton et al. (2009) suggested the information opacity increases as synchronicity increases. It is debated that when there is a deficiency of firm-specific information transparency, shareholders will trust more on publicly existing information that contributes to the greater association between stock price and the market. Jin and Myers (2006) hypothesize that evidence opacity swings firm-specific risk from shareholders to managers. One can claim that if the high volatility of idiosyncratic risk is accompanying with more information, then managers have slighter need to cope earnings to help depositors in price detection. Based on the above discussion, we construct the following hypothesis, as an alternative to H1
H2: Firms facing higher idiosyncratic volatility are likely to engage less earnings management.
The answer to the question about risks that has been associated with macroeconomic conditions are used to accomplish accrual management are likely to help standard setters recognize in which macroeconomic conditions are effective in simplifying management communication with shareholders, which lead to resourceful behavior (Hutton et al., 2009; Jin and Myers, 2006). In addition, confirmation on the magnitude and occurrence of earnings management and on resource allocation possessions should help macroeconomic conditions to consider whether shareholders are betrayed by earnings management or the possessions are extensive enough to permit transforming prevailing macroeconomic conditions. Based on the above argument, we construct the following hypothesis.

H3: When macroeconomic conditions are more uncertain then the firm will do more earnings management.

The effects of earnings management on two different uncertainties including idiosyncratic (firm-specific) risk and macroeconomic uncertainty. The global financial crisis (GFC) resulted in the potential risks in the financial frame. Vichitsarawong, Eng, and Meek (2010) reveal that managers were less preservationist and not opportune to address the challenges amid the Asian financial crisis. This study expects that global financial crisis builds the weight on managers which propels them to manipulate earnings. Higher earnings management brings about higher idiosyncratic risk. Chen et al. (2012) demonstrate that the uncertainty of idiosyncratic (firm-specific) and macroeconomic volatility is emphatically connected with the managerial decision in terms of accruals. From the above discussion, this study hypothesizes that the idiosyncratic (firm-specific) and macroeconomic uncertainty is decidedly related either with earnings management practice. Accordingly, we build up our fourth hypothesis as follows:
H4: When idiosyncratic volatility and macroeconomic conditions are more uncertain then the firm will do more earnings management.

1.6 Significance of the Study
There are prior empirical researches as Healy (1985), DeAngelo (1986) and Jones (1991) which have depended on discretionary accruals to recognize the earnings manipulation. In spite of the fact that there are a few types of research concerning earnings management and uncertainty DeAngelo (1988) concern the interest for a better show which can clarify or expose earnings management and firms’ uncertainty. Besides, there are no earlier studies that have actualized these models in an object to identify the impact of earnings management on idiosyncratic and macroeconomic uncertainty at a firm level. Along these lines, examine the quality of the financial report. Consequently, it is imperative to test if these models can be connected seeing the impact of uncertainty at a firm level. Moreover, if globally endeavors are occupied with earnings management.
Idiosyncratic risk it may negatively affect the firms earning management. Economic theory advocates that both firm-specific and macroeconomic uncertainties play an important role in value optimization of a firm. Yet, researchers have predominantly focused on merely firm-specific uncertainty while taking into account the role of uncertainty in firms’ earning management verdicts. This study attempts to provide useful information on the relationship between idiosyncratic risk and earning management. Keeping in view these guidelines firms, investors, financial agencies, stakeholders, managers, and shareholders may manage the idiosyncratic risk more effectively which resultantly can enhance their earrings because, in these years, Pakistan is suffering much from numerous economic and political snags. For example, anticorruption actions, trade deficit, energy crises, the war on terror, and a high level of debt burden triggered extreme uncertainty around macroeconomic conditions. These economic and political problems have also critically affected firms’ business activities, which, in turn, improved the risk accompanying with their maneuvers. This research also helps the investors in knowing better knowledge of the impact of idiosyncratic risk and macroeconomic uncertainty on accrual management. In this study, we include both firm specific-risk and macroeconomic risk. With the help of this study, we can find the impact of both types of risk on accrual management for small, medium and large firms. This research work will be beneficial for students and researchers as it would widen their scope from the information contained in this research work.

Chapter No.2
THEORETICAL FOUNDATION
2.1 Introduction
The present chapter of the study includes an illustration on the most well-known theories of corporate finance. It introduces a chronological overview of the theoretical development that took place in the earnings management field over the time. Earnings, represented by the primary concern of the income statement, are a summary component in financial statements. Financial statements are firms’ essential method for imparting firm value and performance to stakeholders and other essential parties. They give a road through which managers disperse some privately held information. The arrangement of financial statements is guided and controlled by accounting standards, guidelines, rules, or policies, called the generally accepted accounting principles (GAAP). Every country has its own accounting benchmarks.
Additionally, this chapter includes the hypotheses of the study, which are formulated after analyzing the various theoretical and empirical studies. A handful theory is presented on the earnings management behavior with the Agency Theory of earnings put forward by Jensen and Meckling (1976). In the course of time, a bulk of innovation and influential theories were developed such as the Keynesian Approach introduced by Keynes (1936), Big Bath” by Arthur Levitt (1998), Healy and Wahlen (1999) analysis the source of earnings management with a point of view from standard setters in light of empirical evidence.
Similarly, a large number of theories were put forward that combined the earnings behavior in corporate finance and the growth of economic development. Henceforth, these theories will be connected so as to comprehend the earnings management models.

2.2 The Agency Theory and Earnings Management
Agency theory provides a background for accepting how the association of incentives and information asymmetry effect managers’ decisions. Earning management is reflected as a kind of agency cost as directors look after their own interests by illuminating the financial reporting (Prior et al., 2008). Agency theory presents a structure in which some researchers recommend that earnings management might be useful as it raises the information worth of earnings by carrying secluded information to the shareholders and the public. Researchers studied the agency theory as an instrument to differentiate among the contriving and encouraging practices of earnings management. The empirical evidence supports that companies wherever earnings management take place to a higher (fewer) extent suffer fewer (higher) agency costs.

Jensen and Meckling (1976) put forward the agency theory which described the companies’ executives as agents and shareholders as principals. In their examination, a shareholder is inconsistent with executives. In other words, decision-making is designated to executives. But, the issue is that the agents do not decide in favor of principals. One of the fundamental hypotheses of agency theory is that principals and agents are in the irreconcilable situation. In their opinion, management incentives are for private interests which are in opposition to the interests of maximizing shareholders’ capital. As indicated by this perspective, remuneration incentives, contract incentives (comprising agreements with third parties, undisclosed agreements or commission contracts), political incentives, tax incentives, initial offering to the general public and change of executives (particularly the CEO), they are amongst the main earnings management incentives. Moreover, the relationship of earning management and firm value is increases significantly. Above mentioned results exposed that earning management is on average, not destructive
2.3 The Researchers’ Points of View in Connection to Earnings Management
In linking with the earnings management, the researchers have two perspectives; some researchers have faith that with firms’ growing in size, the executives do earnings negative management to decrease the affectability. Another view believes that with companies growing in size, supervision and social impacts of financial reporting results increase; and consequently, the manager’s propensity increases to manage earnings. Despite the over two perspectives, in light of this suspicion that in reality all individuals looking to expand their wealth and prosperity and the executives are likewise not excluded from this lead and the direction of these incentives is not the same in the various executives. In this way, we can expect that earnings management does not rely on the size of the companies. With regard to the behavioral aspects of management, with a similar reasoning, it could be said that earnings management depends upon the firms’ size and their business (Etemadi et al, 2012).
2.4 The Earnings Management and the Quality of Financial Statements
Levitt (1998) believes that the disintegration of the quality of financial statements is because of the high inclination of executives to accomplish the predefined earnings or making reserve funds for false accounting. These statements suggested a connection between the impact of earnings management and the risk and uncertainty of macroeconomic and stakeholders on the earnings quality. The confirmation similarly likewise demonstrates the expansion in economic growth demonstrates the expansion of shareholders doubt, particularly among companies that were presented to earnings management. The financial statements of companies which have been presented to earnings management, rather than the actual presentation of the company, it mirrors the interests of executives. Moreover, the sturdy market response to the general news on earnings management represents the investors’ absence of mindfulness on executives’ incentives (Anvaz, 2011). Regarding the matters presented, followed by the researches did in different countries and with a specific end goal to clear up the part of earnings management incentives and its effect on financial performance, and the relationship between earnings management incentives and earnings reaction coefficients in companies listed in Tehran Stock Exchange, has been considered and the earnings management incentives in earnings reaction coefficient measurements are used to investigate it.

2.5 Basic Concepts of Earnings Management
Healy and Wahlen (1999) investigate the cause of earnings management with a viewpoint from standard setters in the light of empirical evidence. Regardless of the theoretical interest of the academics, they explore management judgment in the financial report. Further, they describe earning management as; “earnings management occurs when managers utilize judgment in financial reporting and in structuring transactions to adjust financials reports to either deceive some stakeholders about the underlying economic performance of the firm or to influence contractual results that depend upon reported accounting numbers” (Healy ; Wahlen 1999). The result indicates that it needs more studies about earnings management. Though, Healy and Wahlen convey different methods to investigate the phenomena the problem is still present.

2.6 Big Bath and Earnings Management
Big Bath” put forward by Levitt (1998) argues the companies their rebuilding charges when companies are experiencing an auxiliary change. The reason is that analysts tend to center their appraisals of future cash flows and does not respond to firm’s restructuring charges as negative at the stock cost since the cost is considered as an expendable thing. Companies can, therefore, change the overstating costs to revenue. As indicated by Watts & Zimmerman (1990) and Evans & Sridhar (1996), earnings management is the key exercise of management inclination over bookkeeping records with or without constraints. For Levitt, earnings management is to exploit a favorable position of the adaptability in accounting in order to keep pace with business innovations. In particular, earnings management is a preparation of creative accounting. In a word, earnings management is neither a real nor an unlawful practice insofar as management preference over accounting numbers or accounting adaptability is worked out.

To evaluate the presence of earnings management for empirical researches, three noteworthy approaches have been utilized in the theoretical foundation: accruals (i.e. the contrast between reported earnings and cash flows from operations), earnings distribution, and return on assets ratio. All the three represent some of the possible results of earnings management. Healy & Wahlen (1999) believe that unpredicted accruals (i.e. the residual item after total accruals are regressed on variables that are indicators for normal accruals and gross fixed assets) are the confirmation of earnings management, in light of the fact that unpredicted accruals are the unexplained part of total accruals. Then again, Messod (2001) utilized particular accruals (e.g. the arrangement for awful obligation; accruals in particular sectors, for example, the claim misfortune hold in the insurance industry) to measure earnings management. Still, the accruals approach is risky for at least three reasons. In the first place, although discretionary accruals may be influenced by managerial choices, the connection between earnings management and unpredicted accruals can be close to an assumption because of information asymmetry; to be specific, the two are not really of the circumstances and end results relationship. Second, unforeseen accruals are an uproarious variable. Third, the accrual approach isn’t comprehensive or inclusive, since accruals are just a single sort of the items that can be controlled, different items incorporate, for instance, product costs; and unforeseen or specific accruals represent, it might, the presence of earnings manipulation only.

2.7 Earnings Management and Idiosyncratic Risk
The prior research investigates that enhancing data exposure and reporting quality can alleviate information asymmetries and lessen the volatility of stock prices (Rajgopal and Venkatachalam, 2011). Kothari (2000) demonstrates that astounding financial information can relieve information asymmetries between managers and outside investors.
However, Dichev et al. (2013) demonstrate that managers have impetuses to take part in earning management to abstain from announcing misfortune, earnings declines, and earnings missing analysts’ forecasts impact. Cohen et al. (2008) show that that the inclination for firms to exchange off real versus accrual-based earnings management activities. Chen et al. (2012) demonstrate that the idiosyncratic return volatility is emphatically connected with the managerial decision in terms of accruals.

2.8 Uncertainty-Earnings Relationship through Financial Constraints
Financial constraints are another channel that explains the uncertainty-earnings relationship. Minton and Schrand (1999) have reported that when there is higher firm-specific volatility and firms confront inner income deficiencies, they essentially put off the discretionary accruals and sometimes forgo the earnings. Further, firms can utilize the outside funding to meet the income setbacks. However, Myers and Majluf (1984) look at that outer capital is more exorbitant than inside capital. Higher idiosyncratic volatility is not the main reason that influences the earning decisions of firms. Income shortage is additionally decidedly connected with a firm’s cost of getting to outer capital. Therefore, following net present value (NPV) decision rule firms do lower earnings by external capital. The financial constraints through getting to outer capital and firms’ volatility are viewed as another explanation for the negative impacts of uncertainty on accruals.
2.9 Uncertainty-Earnings Relationship through Risk Aversion
The negative connection between uncertainty and earnings can likewise be defended through the risk-averse attitude of firms. Some studies, for example, Zeira (1990) and Nakamura (1999), theoretically demonstrate that the risk-averse behavior of firms is one of the significant reasons of the negative connection amongst uncertainty and corporate firms’ earning decisions. In particular, Nakamura (1999) builds up a model, which not just shows the connection between uncertainty and earnings, yet additionally the condition for negativity. It is expressed that if the level of relative risk aversion is higher than the relationship of uncertainty-earnings will be negative.
Researchers have extensively examined the uncertainty-earnings relationship using both aggregate and firm-level data. Some studies have empirically explored the effect of exchange rate volatility on firm-level. For example, Goldberg (1993) and Campa and Goldberg (1995) utilize the exchange rate as a risk measure to explore the effect of exchange rate uncertainty on earnings management. They utilize a US industrial dataset for their empirical findings. They investigate that the exchange rate volatility has a weak negative effect on earnings management.
Ghosal and Loungani (2000) generate the intrinsic uncertainty proxy by computing the standard deviation of firms’ profit for industries. They find that the relationship between uncertainty and earnings management is negative and statistically significant. Solimano (1993) examined the implication of discretionary accrual models by utilizing the uncertainty and found a significant negative impact of uncertainty measures on corporate firms’ earning management.

Chapter No.3
LITERATURE REVIEW
Several studies have been undertaken to investigate the impact of idiosyncratic risk on accrual management of developed countries. In the past decade, many researchers and financial analyst have strived to examine the relationship between idiosyncratic (firm-specific) and macroeconomic uncertainty/risk and accrual management. Theoretical discussion of accrual management has always been of great importance. Stein and Stone (2014) have an active role in corporate finance. They concluded that managers may use accruals in place of earnings over the time. Researchers explained the significant role of accrual management in economic policies and procedures of all economic businesses around the globe (Strobl, 2013; Jackson et al. (2016). They have conducted various studies to establish the effect of idiosyncratic risk on accrual management and the outcomes of all those studies are in different direction. This chapter of the thesis has deliberated about prior research papers and their empirical decisions that are connected to our part examination.
Debnath (2017) has studied a research on evaluating the impact of firms’ growth and performance on earnings management in India. He studied the nature and magnitude of earnings management practices and the effect of firm’s performance and growth on earnings management through the discretionary accruals and took 756 firm-year observations from the corporate sector of non-financial over the period of 2007 to 2015. He used panel data to analyze the data. Firms’ growth has positively momentous effect with discretionary accruals whereas the firm’s performance has adversely associated with discretionary accruals. He used earning management as a dependent variable while the firm’s growth and performance were used as independent variables.

Datta, Iskandar-Datta, and Singh (2017) investigated a research on the impact of idiosyncratic risk on accrual management in the United States. They took 44,599 firm-year observations as a sample over the period from 1987 to 2009. They studied that the significance of firm-specific volatility as a factor of earnings operation, for this study, they used robust econometrics methods to analyze the hypotheses. They found that the relationship between idiosyncratic risk and accruals management was a positively correlated and also observed a significant positive relationship between residuals volatility and discretionary accruals whether accruals were income inflationary or deflationary. Accruals management used as an independent variable whereas idiosyncratic volatility was used as a dependent variable while leverages, market-to-book ratio, asset growth, market capitalization, return on assets, sales and cash flow volatility have been used as control variables.
Farhadvand and Jalilian (2017) conducted a research on investigating risk effect and profit management on bank risk in Iran. They intended to check the risk effect and profit management on banks listed in the Tehran stock exchange. For this purpose, they used bank credit risk as a dependent variable and bank risk as an independent variable while financial leverage ratio, return on assets (ROA), cash flows from debt, bank size, growth, and systematic risk have been used as control variables. 18 corporations were selected as a sample and collected secondary data over the period of 2010 to 2014. Linear regression, correlation, and panel regression techniques were used to analyze the sample data and hypothesis. Their findings of the study exposed that bank risk and profit management have a significant positive effect on bank credit risk and there was a positive association between corporation risk and profit management. They also observed that bank idiosyncratic risk and profit management have a negative effect on credit risk.

Chang, Hsin, and Shiah-Hou (2016) scrutinized the performance of financial hedging and earnings management under diverse corporate information quality in Taiwan. They inspected that financial hedging has a positive valuation premium for firms with better quality of data condition and earning management has a positive valuation premium for firms with poor data quality however negative effect for firms with better information quality for the calculation of study. They used financial hedging as a dependent and earning management as an independent variable while three variables i.e. analyst coverage, corporate governance, and idiosyncratic risk have been used to proxy for the quality of corporate information environment. S&P 500 non-financial firms were selected from Taiwan stock exchange and covering the years from 2001 to 2010. Findings of the study exposed that robust test was applied for financial hedging and they proved that there was an important role of information quality in evaluating the illusion of financial hedging and accrual management.

Ebrahimi, Nasab, and Karim (2016) investigated a research on the effects of uncertainty, accruals, and quality of risk on risk premium of stocks in Iran. They investigated that the accruals and quality of risk had a positive significant effect on uncertainty due to the demand of stock returns. Also, uncertainty had an insignificant effect on accruals and, therefore, on stock returns. For this, they took 70 companies as a sample from Tehran stock exchange over the period of 2003 to 2014. In this study, they used risk premium as a dependent variable while accruals and risk quality, firm size, investment quality and book equity to market equity were used as independent variables and they used panel regression techniques to scrutinize the data.

Sellami (2016) studied the analysis of mandatory IFRS adoption on the interaction between real and accrual-based earnings management in Tunisia. They were using a sample of 124 firms selected from the French-listed firms during the period from 1999 to 2011. Empirical results indicate that French firms’ manage accruals and used real activities to smooth earnings. They focused on real activities manipulation as well as accrual-based earnings management in order to describe the full opportunism of managers when managing their earnings. They used panel regression techniques to examine the facts and figures of the data. Abnormal cash flow from operations and discretionary accruals were used as dependent variables whereas real earnings management used as an independent variable while leverages, cash flow from operations, asset turnover, firm size, and growth have been used control variables.

M and Zh (2016) conducted a research on the effect of the global financial crisis on earning management in Iran. Empirical results indicate that there was a positive significant relationship between the size of the company and the management of profit in the global financial crisis. They also scrutinized that there was a positive relationship between the profitability index of company and earnings management and similarly a significant relationship between sale growth and profit management in global financial crises. They took data from the non-financial firms’ listed on the Tehran Stock Exchange as a sample over the period from 2008 to 2012. They used panel regression techniques in order to evaluate the overall adequacy of the model. For the purpose of measuring the variables, they used earnings management as a dependent variable whereas company size, sales growth, and profitability indicators were used as independent variables while financial leverage and efficiency of assets have been used as control variables.

Chang, Wang, Chiu, and Haung (2015) analyzed the earnings management and idiosyncratic risk evidence from post-Sarbanes-Oxley Act (Global financial crisis) in the United States by using stock returns annual data from the database and the data were collected from 3,940 firm’s observation for the period 2000 to 2010. They used a panel data regression model to examine the association between idiosyncratic risk and earning management. The findings indicated that idiosyncratic risk has a positive impact on earning management whereas idiosyncratic risk has an adverse impact on accrual based earnings management in the Sarbanes-Oxley Act and vice versa. They analyzed that there has a transaction amongst accrual-based and real earnings management. Moreover, they investigated the idiosyncratic risk that affects the earning management procedure. As per investigation earning management has a significantly positive effect on idiosyncratic risk. They used idiosyncratic risk as an independent and earning management as a dependent variable whereas earning management contains two types of earning such as accrual earnings and real earning management.

Chen and Wang (2015) have piloted a research on idiosyncratic risk and the cross-section of expected stock return. They figured out the positive relationship between idiosyncratic risk and cross-section of stock returns when investors were less motivated to diversify and a negative relationship exists when investors have a strong incentive to diversify. They examined that investors’ incentive to diversify varies over the time. For the calculation of the study, they employed threshold regression to expose the basic relationship between idiosyncratic risk and the cross-section of predictable stock returns. The effects were shed light on the prevailing collected work and help to reunite the contradictory results found in the literature. They collected data from the stocks traded on NYSE and NASDAQ as a sample from the period of 1998 to 2012. Idiosyncratic risk used as an independent and stock returns, market returns and risk factors were used as dependent variables while market value and book value of the equity has been used as control variables.

Sellami (2015) has conducted a research on incentives and constraints of real earnings management in Tunisia. He studied frameworks the various forms of real earning management, incentives, and constraints of real earning management. He found that international financial reporting system (IFRS) played a dynamic role in decreasing the earnings management by controlling opportunistic management preferences in conducting real activities. He also examined that real earnings management has a straight influence on cash-flows of the company which was highly leveraged to generate cash and pays down their loans. Real earnings management was used by managers to gratify their particular interests. In order to illustrate the explanatory power of variables, real earnings management was used as a dependent variable whereas capital market, opportunistic motivations, and benefits of real earnings management were used as independent variables. He took data from the Tunisia stock exchange covering the period from 2002 to 2011 and used panel regression methods.

Hwang, Chiou, Hsueh, and Hsieh (2014) studied how earnings management affects innovation strategies of firms in Taiwan. They used a sample of S&P 500 firms from Lexis Nexis academic database covering the period from 1997 to 2009. However, they used earning management as a dependent variable while other essential factors were included in the regression test such as financial leverage, firms’ profitability, a number of outstanding shares, growth, financial constraints and firm size as controlled variables. They found that a company with greater real earning management and the higher level of patent purchase has the lowest firm value. The findings showed that there was a strong bonding between innovation strategy and earning management and such undertakings have small impacts on firm’s cash flow volatility and have a positive significant impact on firms’ value and they used panel regression techniques to analyze the data.

Herskovic, Kelly, Lusting, Nieuwerburgh (2014) investigated the common factor in idiosyncratic volatility quantitative asset pricing implications in the United State. They used idiosyncratic volatility factor as an independent variable and firms cash flow growth as a dependent variable by using a sample of an annual panel of United State public firms and evaluated the data by using the cross-sectional regression techniques of a given volatility panel covering the period of 1926 to 2010. They found that customs import value (CIV) was a valued municipal variable because an upturn in idiosyncratic firm volatility increases the average household’s marginal utility. Also, the ideal theoretical framework depicted the high degree of correlation in idiosyncratic volatility, beta return and many other asset pricing moments.

Patel and Cooper (2014) examined the influence of firm-specific (idiosyncratic) risk the incentives of chief executive officer (CEO) in the United States. They used a large panel for US firms. For the purpose of estimation, researchers used CEO incentives as a dependent variable which was measured by cash, equity, total compensation and pay-performance sensitivity while firm risk as an independent variable which was unique to a particular firm, idiosyncratic risk and was independent of the mutual association of the market whereas other risk measures such as systematic risk and bankruptcy risk, firm performance, firm policies, firm characteristics, executive characteristics, and external monitoring have been used as a controlling variable. Researchers collected data from a sample of five executive firms over the period of 1992 to 2009 and used panel regression techniques to analyze the data. They suggested that firms with higher idiosyncratic risk yield their CEOs and other executives with the higher incentives.

Kitagawa and Okuda (2014) studied a research on management forecasts, idiosyncratic risk and the information environment in Japan. Empirical evidence suggested that management forecast error was positively associated with firm-specific uncertainty, suggesting that high-quality public information decreases the idiosyncratic (firm-specific) uncertainty. Moreover, they examined that management forecast error was less positively linked with idiosyncratic uncertainty in relatively good information environments. They provided evidence that management projections miscalculation was less confidently and linked to idiosyncratic risk in relatively good information environment condensed that risk and firms through one or more specialist enthusiastically following and giving views on them. The dependent variable was the complex measures of management forecast and included sales growth and leverages as independent variables. They developed a generalized method of moments (GMM) used to treat the model with a vibrant panel data technique consists of 8,527 firms year observation covering the period of 2000 to 2008.

Fan and Yu (2013) have conducted a research on accrual and idiosyncratic risk; international evidence. They examined that abnormal returns have positively related to idiosyncratic risk in international equity markets. Empirical results provided strong evidence to support limits of arbitrage theory and the existence of accrual across countries. They also concluded that idiosyncratic risk has less impact on accrual abnormal returns for industrialized countries than emerging countries. A research conducted on accrual anomaly and idiosyncratic risk in the United States. The results depicted that the accrual anomaly has been expanded and implemented in several types of research in both the financial economics and accounting. For the purpose of the calculation, they used abnormal returns in the econometric model as the dependent variable while the idiosyncratic risk was used as independent variables. They took data from Thomson financial DataStream database for the period of 20 years from 1989 to 2009 for firms in 43 countries and used the panel regression model for their research
Angelis, Grullon, and Michenaud (2013) examined the natural experiment on the exploration of downside risk and the design of CEO incentives in the United States. For the calculation of data, 4,036 firm-year observations from the Russell 3000 index listed on NYSE, NASDAQ, and AMEX were taken as a sample over the period of 2001 to 2007. CEO incentives were used as a dependent variable whereas risk used as an independent variable for scrutiny purpose. They observed that whether risk exaggerated the policy of CEO incentives or not. They used a randomized natural testing that exogenously improved downside equity risk over the decreasing of short-retailing limitations on a random sample of US stocks and used panel regression methods.

Sarwar and Muradoglu (2013) have studied on macroeconomic risks, idiosyncratic risk, and momentum profits in the United Kingdom, and all the stocks listed in the NASDAQ, NYSE, and AMEX were selected as a sample period from 1926 to 2005. They investigated major earnings remain the same both at the portfolio and individual stock level by using Fama-French factors, did not eradicate the momentum return and the premium reduces when macroeconomics variables were used. The effects were more noticeable when lagged variables used through the market rise. Momentum return used as dependent variable and idiosyncratic risk was used as an independent variable and cross-sectional and time series regression techniques were used to analyze the momentum returns.

Tariverdi, Keighobadi, and Tavasol (2013) carried out a research on the effective motivations on earning management in Iran. They analyzed the incentives factors such as economic, social and cultural that effects on earnings management. They took firms secondary data from the Tehran Stock exchange and covering the period between 2004 and 2010. They observed that the ownership structure, company growth, and firm size three incentives factors have a significant relationship with earnings management whereas the financial leverage has an insignificant relationship with earnings management. Financial leverage, ownership structure, supply stock, bonuses of directors, company growth and firm size used as dependent variables whereas earnings management (discretionary accruals) were used as a dependent variable for measuring discretionary accrual and they used kasznik model. They used multiple regression statistical models in their study.
Dadbeh and Mogharebi (2013) examined a research on the effect of information asymmetry on earning management in Tehran. They took 47 listed companies as a sample and collected secondary data from Tehran stock exchange covering the years from 2002 to 2008 and panel regression techniques were used to analyze the data. They examined the effect of information asymmetry on earnings management with a bid-ask spread. They figured out the information asymmetry with other models and depicted the effect of information asymmetry on firm value, corporate diversification, and corporate governance. Likewise, they observed that asymmetry information has positive effects on earnings management. They used earnings management as an independent variable while information asymmetry was used as a dependent variable.
Alam and Shah (2013) investigated a research on corporate governance and its impact on firm risk in Pakistan. They examined the relationship between corporate governance and a firm’s risk. They took 106 Pakistani companies as a sample covering the period from 2005 to 2010. They examined that the managerial ownership and family ownership were negatively related to firm risk. Similarly, they found a positive relationship between managerial ownership and chairman/CEO duality with firm risk. They also provided guidelines; firms should aim at non-family directors on the board and should not allow managerial ownership to be their major shareholders since both negatively affect the firms’ risk-taking abilities and thereby hampering its performance. They used firm risk as independent variable whereas family and bank ownership were used as dependent variables while firm size, leverage and growth have been used as control variables. They used generalized method of moments (GMM) estimation techniques introduced by Holtz-Eakin, Newey and Rosen (1988) and Arellano and Bond (1991) to evaluate the data.

Mahmoodabadi, Ziyari, and Dehghan (2013) carried out a research on examining the relationship amongst accrual anomaly and external financing anomaly in Iran. They took 87 firms from the Tehran stock exchange as a sample during the period of 7 years from 2006 to 2012. They examined that the managerial market relationship between external financing and the accrual anomaly was affected by long-term accruals and working capital accruals consistently. They were taken cash flows as an independent variable while accruals earning was used as a dependent variable and panel regression techniques were used to examine the data. The results from cross-sectional regressions showed that the capacity of working capital and long-term accruals in the calculation of stock returns connected with external financing activities and in the existence of long-term accruals the analytical power of the procedures of net external financing for future yields was shortened significantly.

Geng, Wang, Song, Liu, Chi, Yu, Zhang, He, Zheng, Zhu, Zhou, and Li (2013) investigated the effect of listed companies risk on the market pricing of accruals quality in China. They observed that the effect of accruals quality on cost of capital increases with the increase of basic risks. Though, the relationship between fundamental risk and accruals quality were closely related to the cost of capital. Also when this interaction exists, the main influence of accruals quality will diminish. They also observed that there was no internal connecting link between accruals quality and cost of capital which was calculated by the present worth of income of low- fundamental-risk enterprises. They took 6840 firms year observation from the Shanghai stock exchange covering the period of 1999 to 2009 and used panel regression techniques. They used risk as a dependent variable while incomes of stock and accruals quality were used as independent variables.

Rani, Hussain, and Chand (2013) conducted a research on managerial incentives for earnings management among listed firms in Suva, Fiji Islands. 14 companies were selected as a sample over the period of 1980 to 2008. They examined that the earnings management was a universal phenomenon and common incentives for earnings management were management compensation incentive, borrowing cost incentive, an incentive to meet target expectations and increased (decreased) monitoring cost whereas management compensation was the most prevalent incentives for earnings management. The smallest collective motivations were equity assistances due to extremely inactive share market and management takeover as this situation was uncommon among registered corporations to date. They used panel regression techniques to evaluate the data.

Hall and Agrawal (2013) studied a research on the financial statement analysis and earning management in the United States. They examined that the company was expected to have influenced or accomplished the financial statement numbers, and those numbers in the financial statements were most probable to have been achieved, and the extent of the management and they also observed that there was an increased probability of earnings management in any framework in which earnings management has been providing an advantage to the managers or the firm. Discretionary accruals were used as a dependent variable while items of balance sheet and cash flows from operations and income from operations were used as independent variables during the period from 2000 to 2009 and used panel regression techniques.

Zamri, Rahman, and Isa (2013) conducted a research on the impact of leverage on real earnings management in Malaysia. They took 3,745 firm-year observations from the Bursa Malaysia as a sample for the period of 2006 to 2011. They found that an insignificant relationship between leverage and real earnings management (REM). Similarly, they examined that the leveraged companies have lower levels of real earnings management activities which in turn, could influence the quality of accounting earnings. They used real earnings management as a dependent variable whereas leverage used as an independent variable while net interest expense (INTEXP), Return on Assets (ROA) and firm size (SIZE) have been used as control variables. They also observed that the leverage was one of the directing and monitoring arrangement which limits the REM. They used panel regression data.

Sayari, Omri, Finet, and Harrathi (2013) carried out a research on the impact of earnings management on stock returns in Tunisia. They took 33 firms which were listed on the Stock Exchange of Tunis over the period of 1999 to 2008. They used panel regression data for analysis purpose. They used earnings management as an independent variable while stock returns were used as the dependent variable. They examined that the large firms’ preference was reducing their earnings because governments focus closely on their profits to finance the state budget by leveraging greater taxes. They also investigated that the earnings management were allowed companies for increasing abnormal positive stocks returns for large Tunisian firms’ and reducing abnormal negative stocks returns for small Tunisian firms’.

He, Li, Wei and Yu (2012) conducted a research on uncertainty, risk, and incentives in the United States. They found that profitability uncertainty and moral hazard were interrelated with each other, whereas Shareholders who handled more ambiguity desire leading to better learning and therefore offer advanced managerial incentives to encourage higher exertion from the executives and adverse risk-incentive transaction. They observed an affirmative relation amongst profitability uncertainty and managerial incentives of the firm. They used panel data set techniques for the estimation of regression analysis for the period of 1992 to 2008 and the sample included 2,441 firms. The dependent variable was used pay-performance sensitivity (PPS) and the independent variable was used profitability uncertainty whereas firm size has been used as a controlling variable to seizure the side effect of the firm.

Albuquerque, Papadakis, and Wysocki (2011) have shown a research on the impact of risk and monitoring of CEO equity incentives in the United States. They used 23,042 CEO firm-year observations nominated as a sample from ExecuComp and covering the period from 1992 to 2006. Equity incentives were used as dependent variable, risk as an independent variable while firm size, growth, market leverage, performance and research, and development have been used as control variables. Researchers investigated that there was a positive significant relationship between a firm’s operational risk, the monitoring of the manager’s actions and the level of CEO equity incentives. They used robust regression test and the result shows that influence of the level of customer attention on Chief Financial Officer (CFO) equity incentives because CFOs have less control over operational assessments that affect the firm’s risk acquaintance.

McAnally, Neel, and Rees (2010) have carried out a research on CEO incentives and downside risk in the United States. They detected from their research that CEO incentives were negatively related to downside risk but positively related to upside risk because CEO incentives reflected both the asymmetry and the conditional nature of firm risk. Consistent with an assumption that CEOs perceived potential losses as riskier than potential gains. However, they examined that CEO equity incentives were reflected risk asymmetrically and conditionally. For the calculation of outcomes, they used a sample of about 2,600 firms from the period of 1992 to 2008 and applied panel regression methods. The dependent variable of the study was CEO incentives, risk as an independent while number of firms, growth and leverages have been used as control variables.
Markarian and Albornoz (2010) have piloted a research on income smoothing and idiosyncratic volatility in the United States. They examined that there was a negative relationship between smoothing and idiosyncratic volatility, which they interpreted as evidence that income smoothing practices were implemented in order to reduce the stock price idiosyncratic volatility. Likewise, they also examined that increases in income smoothing were related to decreases in idiosyncratic volatility. They observed that this relation was due to managerial prime, not intrinsic smoothness. They used idiosyncratic volatility as independent variable which was measured by residual from market ideal panel regression techniques whereas income smoothing used as a dependent variable and measured by income volatility with respect to cash flow and change in accruals by change in cash flow while market value of equity and return on asset have been used as control variables. 88,577 observations of the cross-section data were selected as a sample for analysis over the period 1989 to 2006.

Rajgopal and Venkatachalam (2010) carried out a study on financial reporting quality and idiosyncratic return volatility in United State and collected data from a sample of 95,270 firms’ year observations from 1962 to 2001 and figured out the data by using cross-sectional and time series regression methods. For the purpose of a study, the dependent variable of the firms’ observation was measured by stock returns volatility. They used two proxies to seizure earning quality by Dechow-Dichev (DD) and squared abnormal accruals for measuring the earning quality and firm’s independent variable was idiosyncratic return volatility used to measure the financial reporting quality of the firm and time has been used as controlling variable. They analyzed that earning quality was accompanying with greater idiosyncratic returns volatility constructed in 40 years covering the period from 1962 to 2001 by the above-mentioned proxies and positive association after controlling the several effects of control variables and the impact of these variables on the firms’ year observation was negative earnings, merger movement, and financial misery.

Bakke and Whited (2010) conducted a research on an analysis of corporate investment decisions on which firms follow the market in the United States. They constructed an econometric methodology that disentangles stock-price activities that were relevant for investment from those that were not. They joined this decomposition with substitutions for private information and mispricing to develop unbiased tests for the e?ects of mispricing and evidence on a stock. Also, they found that stock-market mispricing did not a?ect investment especially that of large firms and firms subject to mispricing. They used GMM techniques to estimate their parameters. They collected a sample that contains between 2,684 and 3,891 observations annually during the period from 1991 to 2004.

Forester, Sapp and Shi (2009) examined on the analysis of the impact of management earnings forecast on firm risk and firm value in Ontario. They found that there has a momentous adverse relationship between the earning management and firm risk measures containing idiosyncratic risk, stock returns volatility, beta and bid-ask spreads advising the issuance of earning management diminished the information asymmetry and thus decreasing the risk. They also found that there was a positive momentous relationship between earning management prediction and firm value as caught by Tobin’s Q when the issuance of earning management was more specific and also investigated a positive relationship when actual earnings encounter the management’s prediction. For the calculation of the study stock return volatility was used as an independent variable and firm-specific risk as well beta was used as a dependent variable. Approximately 3000 US domestic firms year observations were selected as a sample covering the period from 1994 to 2000 and collected secondary data and panel regression procedures were used to scrutinize the data.

Gray, Koh, and Tong (2009) investigated a research on accruals quality, information risk, and cost of capital in Australia. They examined that there was a positive relationship between accruals quality and the firms’ cost of capital. They also found that the costs of debt and cost of equity for Australian firms were highly influenced by accruals quality arising from economic essentials. Furthermore, they investigated that there was a number of important institutional and regulatory differences were hypothesized to affect the relation between accruals quality and cost of capital. 509 firms for the cost of debt and 346 firms for the cost of equity analysis were used and covering the years from 1992 to 2005. They used accruals quality as a dependent variable which was used as a proxy for information risk while the cost of debt and cost of equity were used as independent variables. They used panel regression techniques.

Badertscher, Collins, and Lys (2009) carried out a research on earnings management and the accruals with respect to future cash flow in the United States. They examined that there were two management incentives for managing earnings opportunistic and informational and each has different implications for the resultant numbers capability to forecast the future firm operating cash flows. They found that firms’ with managing earnings for opportunistic explanations were less predictive of future cash flows whereas firms’ with managing earnings for informational reasons were more predictive to the future cash flow. However, they also provided evidence that supports Jensen’s (2005) estimation that overvaluation leads to value-destroying opportunistic earnings management. Conversely, they observed that income-increasing and income-decreasing IP firms’ exhibit significantly different return patterns in the year that earnings were managed in directions. Also, that was consistent with management using earnings management to signal the future prospects of the firm. They used earnings management as dependent variable while average account receivable, accounts payable and other accrual components used as dummy variables for analysis purpose. They took the sample of 845 firms from the General Accounting Office from January 1, 1997, to June 30, 2002. They used a panel regression data for analysis.

Irvine and Pontiff (2008) have conducted a research on idiosyncratic return volatility, cash flows, and product market competition in the United States. They used idiosyncratic volatility to assess the measures of fundamental cash flow volatility. They found that the movement in idiosyncratic cash ?ow volatility reflects the movement in idiosyncratic stock-return volatility, dependable with the market ef?ciency. Also, they examined that the increasing movement in idiosyncratic (firm-specific) volatility was connected with an increasingly competitive environment in which ?rms’ have fewer market influence. Once the success of one ?rm in an industry comes at the expense of another ?rm in that industry, then the competition underwrites to negative covariance in ?rm performance and markets re?ected an environment with less consumer trustworthiness to a speci?c ?rm. They took 577,300 firm-specific observations from the CRSP/Compustat merged database as a sample over the period of 1964 to 2003. They used idiosyncratic volatility and fundamentals of cash flow volatility as independent variable whereas market competition used as a dependent variable. They used a panel regression data for analysis.

Michelacci and Schivardi (2008) conducted a research on why does idiosyncratic business risk matter in Canada. They investigated that idiosyncratic risk depressed business activity and delays growth, with the e?ects being stronger in economies with minor risk diversi?cation chances and also found that those countries with low levels of risk diversi?cation opportunities executed relatively poorer in segments that were categorized by high idiosyncratic volatility and that volatility were endogenous with respect to risk diversi?cation opportunities. They observed that ?rms measured by less diversi?ed owners show lower mean and dispersion of yield growth. They used idiosyncratic risk as a dependent variable whereas performance and diversification opportunities were used as independent variables while growth has been used as a control variable. Panel regression techniques were used to measure the data. They took secondary data from the 20 largest organizations from the stock market as a sample that runs the sample period from 1973 to 2003.

Cornett, Marcus, and Tehranian (2008) conducted a research on the effect of earnings management on corporate governance and pay performance in the United States. They investigated that the institutional ownership of shares, institutional investor representation on the board of directors, and the existence of independent outside the directors on the board all shrink the use of discretionary accruals. Besides this, firstly they show the impact of significant of earnings management that increased the importance of governance variables. Secondly, they also depicted the influence of incentive-based compensation on a company act. They were using S&P 100 firms as a sample for a period of 1994 to 2003. Discretionary accruals used as a dependent variable while performance measures were used as an independent variable. They used panel regression techniques to analyze the data.

Skaife, Gassen, and LaFond (2006) conducted a research on stock price synchronicity that affects the firm-specific information in the United States. They found that the zero-return metric was a superior measure of firm-specific (Idiosyncratic risk) information held into share prices than the synchronicity measure globally. As the zero-return metric was more useful in taking the differences and deviation in stock price synchronicity across firms in transnational markets was not due to changes in firm-specific information. They found a significant positive relationship between the zero-return measure and the magnitude of returns which was reflected in returns. They took all the 2895 firm-year observations for Australia, France, Germany, Japan, the U.K., and the U.S from world scope as a sample size during the period of 1990-2002. They used the stock price synchronicity as independent variable whereas standard deviation of sales and the standard deviation of return on assets used as dependent while regulated industry regulations, industry returns, firm size and firms’ share have been used as control variables and firms (Public and private) information flow as proxy variables and they used large panel regression techniques to examine the data.

Goyal and Clara (2003) examined the research on idiosyncratic risk matters in Los Angeles in the United States. He examined the relationship between stock returns and risk measures and found a significantly positive relation between idiosyncratic risk and average stock returns on the market by using the CAPM model. They also found that conditioning on average stock risk can bring significant economic aids to the shareholders. They took Schwert’s daily data from CSRP as a sample for their study and covering the period of 1962 to 1999. They used average stock prices as the dependent variable and firm-specific (idiosyncratic) risk as an independent variable and idiosyncratic risk was measured by residuals and residuals were calculated by monthly cross-sectional and time-series regression for each stock returns on Fama-French Three-Factors Model. They examined that idiosyncratic risk is an important element of the total risk and time variation in average risk while size and market book value has been used as control variables.
Ali, Hwang, and Trombley (2003) examined a research on mispricing versus risk explanations on residual-income-based valuation predicts future stock returns in the United States. They examined that the price-to-book ratio effect was partially determined around the future earnings announcements, reliable with the mispricing description and the effect of price-to-book ratio was due to market mispricing or mislaid risk factors and also observed that price-to-book ratio has substantially interrelated with certain risk proxies. However, after controlling for these risk factors, price-to-book ratio continues to show a momentous positive relationship with upcoming returns proposing that these risk factors were not accountable for the price-to-book ratio effect. Generally, the results have shown that the mispricing explanation for the price-to-book ratio effect of risk proxies. They used price-to-book ratio as the dependent variable while beta, volatility, debt-equity ratio, firm size and risk of financial distress as independent variables in their study and took the sample from the NYSE and AMEX stock returns during the period of 1976 to 1997 and used Fama-French Three-Factors model to estimate the future residual income.

Beneish and Vargus (2002) carried out a research on insider trading, earnings quality, and accrual mispricing in the United State. They observed that the accrual mispricing was due to the mispricing of income-increasing accruals and income-increasing accruals was significantly lower when the accompanied by abnormal insider selling and higher when buying the abnormal insider and hedging returns was trading strategies based on the direction of accruals and exceed when insider trading significantly based on only accruals and lower of income-increasing when insider selling appears to be at least partly attributable to the opportunistic earnings management and these accruals were also relevant to the policymakers charged with regulating insider trading. They included 1998 tapes of all the firms on the Compustat industrial, research and full coverage as the sample for the period of 1985 to 1997. Total accruals and operating cash flows were used as independent variables whereas income-increasing accrual used as the dependent variable while growth has been used as control variable. They used panel regression techniques to analyze the data.

Vuolteenaho (2002) studied the firm-level stock returns in the United States. He examined that the firm-level stock returns were mostly determined by cash-flow, for a representative stock, the variance of cash-flow was more than double that of expected-return and shocks to expected returns. They also observed that cash flows were positively related to a representative small stock whereas expected-return series were highly interrelated across firms, while cash-flow could generally be differentiated away in collective portfolio. Cash-flow evidence was basically firm-specific. Similarly, they found that expected-return information was mostly determined by systematic, market-wide mechanisms. He used panel regression techniques to analyze the data and took data from NYSE, AMEX, and NASDAQ stocks over the period of 1954 to 1996. He used returns on equity as a dependent variable while leverages and book to market value were used as independent variables.

Xie, Davidson, and DaDalt (2001) conducted a research on earning management and corporate governance in the United States. They examined that board and audit committee members with corporate or financial backgrounds were negatively related to the level of earnings management. They examined that board and audit committee action and their associates’ financial superiority might be essential factors in making the tendency of managers to involve in earnings management. They also found that earnings management was less expected to occur or occur less in corporations whose boards comprise both autonomous external executives and directors with company experience. Therefore, they also recommended that board and committee activity influences members’ ability to serve as more effective monitors of corporate financial reporting. They selected 110 companies from S&P 500 index from every year during the period of 1992 to 1996 and used panel regression procedures in their study to examine the facts and figures of the data. Earnings management was used as an independent variable whereas current accruals were used as a dependent variable while the firm’s growth and sales have been used as control variables.

Heninger (2001) examined the relationship between auditor litigation and abnormal accruals, in purpose to find casual effect between earnings management and auditor litigation. Heninger starting point was to redefine total accruals to abnormal accruals as of earlier studies had disclosed the conflict in their result (Lys & Watts 1994). Heninger also refers to Dechow et al (1995) conclusion, that the modified Jones model was the best approach to identify earnings management. Though, the result was consistent with earlier studies. This means that the coefficient for variable cash sales was positive with income increasing. Moreover, the results of multiple regressions indicated that the coefficient for a variable property, plant, and equipment has a negative effect on total accruals which implies that income decreasing has a negative effect on total accruals. Heninger extended the Jones model by adding additional variables that measure the relationship between audit, client, firm size, and financial position of the firm. He took data from the Chinese companies as a sample for the period of 1984 to 1998. He used panel regression methods to analyze the data.
Campbell, Lettau, Malkiel, and Xu (2001) conducted a research on an empirical exploration of idiosyncratic risk on individual stocks become more volatile in the United States. They examined that increased in the volatility of firm-level relative to the market volatility; correlation among individual stocks and the advisory power of the market model for a representative stock have been deteriorated whereas the numbers of stock required to achieve a given level of variation have increased. They took data from the NYSE, AMEX and NASDAQ stocks covering the period from 1962 to 1997. They used Granger causality test to identify the relationship between variables. To measure the data, they used stock returns as a dependent variable while market, industry and firm level volatility were used as independent variables.

Collins and Hribar (2000) conducted a research on the effect of one or two earnings-based and accrual-based market anomalies in the United States. They collected primary, supplementary, tertiary or research file data from NYSE and AMEX as a sample for the period of 1988 to 1997. They observed that the market’s over-under estimation of the importunity of accruals cash flows indications to mispricing that could be demoralized to create a substantial encouraging abnormal returns over the two-quarters after the portfolio realization date and the accruals cash flow mispricing performed mainly independent of the post-earnings statement implication occurrence extensively. They suggested that post-earnings statement implication was overstated or lessened based on the level of accruals surrounded within the earnings disclosure. For this, they used panel regression data to analyze the research files.

Healy and Wahlen (1999) carried out a research on the earnings management literature and its implications for standard setting in the United States. They investigated that earnings management arose for a variety of causes, contained inspiration the stock market observations, upturn management’s reward, and shrink the probability of violating lending contracts and to escape the regulatory interventions. They took 1294 firms year observation during the period of 1980 to 1990 and to analyze the data, they used panel regression methods.

Becker, Defond, Jiambalvo, and Subramanyam (1998) studied a research on the effect of audit quality on earnings management in the United States. They examined that earnings management was captured by discretionary accruals that were estimated by using a cross-sectional technique and audit quality was treated as a dichotomous variable. They also found that users of non-Big six accountants report were discretionary accruals that increased revenue comparatively more than the discretionary accruals reported by the users of Big six accountants and inferior audit quality was accompanying with more accounting flexibility. 10,379 Big six and 2,179 non-Big six firms were selected as a sample from the period of 1989 to 1992. Discretionary accruals were used as a dependent variable whereas leverages, property plant & equipment, change in revenue, operating cash flows, and total assets were used as independent variables and used panel regression methods to assess the data.

Dechow et al. (1995) examined the phenomena earnings management by evaluating alternative accrual based models. The valuation compared specification and power of commonly used test statistics regarding measures of discretionary accruals in purpose to find significant result which may contribute the importance of controlling for financial performance. They concluded that the modified Jones model was efficient in detecting earnings management. The findings in this study provide major implications for research on earnings management. Empirical evidence suggested that earnings management was significantly positively influenced by executive compensation motivation, whereas this influence was constrained mostly by ownership of the firm. Thus, regardless of the model used to detect earnings management, the explanatory power of these tests was relatively low for earnings management. 1024 firms’ were selected as a sample from the period of 1980 to 1990. They used panel regression methods to estimate the data.

When we review the literature on this issue for Pakistan, we find a handful papers that have explored (only) the impact of idiosyncratic (firm-specific) uncertainty on earning management. These studies have documented a positive effect of idiosyncratic uncertainty on earning management. It should be noted that our study significantly differs from these studies as our focus is on firm-level earnings rather than the aggregate one. Secondly, unlike these studies, we consider the role of both idiosyncratic (firm-specific) and macroeconomic uncertainty on firms’ accrual management. Last, but not the least, we propose a theoretical model for the accrual management relationship based on both types of uncertainty.

3.1 Literature Review Summary of Idiosyncratic and Macroeconomic Uncertainty
Authors
Time
Period Covered Country
Model
Specification
Key Findings
Farhadvand et al. (2017) 2010-2014 Iran Panel regression The positive effect on bank credit due to corporation risk and profit management; negative effect on bank credit risk due to bank risk and profit management.

Debnath (2017) 2007-2015 India Panel regression Firms’ growth has positive significant effect on discretionary accruals; firms’ performance has a negative effect with discretionary accruals.

Datta et al. (2017) 1987-2009 United States Robust Econometrics Techniques A positive relationship between idiosyncratic risk and accruals management whether accruals were income inflationary or deflationary.

Chang et al. (2016) 2001-2010 Taiwan Robust test Financial hedging has a positive valuation premium for firms with better quality and earning management has a positive valuation premium for firms with poor information quality.

Ebrahimi et al. (2016) 2003-2014 Iran Panel regression Accruals and quality of risk had a positive significant effect on risk premium and investment anomaly had an insignificant effect on risk premium.

Sellami (2016) 1999-2011 Tunisia Panel regression Real activities manipulation and accrual-based earnings management to describe the full opportunism of managers when managing their earnings.

M et al. (2016) 2008-2012 Iran Panel regression A positive Significant association between the size of the company and the management of profit in the global financial crisis.

Chang et al. (2015) 2000-2010 United States Panel regression Idiosyncratic risk has a positive impact with earning management whereas idiosyncratic risk has a negative impact on accrual-based earnings management.

Chen et al. (2015) 1998-2012 Taiwan Panel regression A positive relationship between idiosyncratic risk and cross-section of stock returns when investors were less motivated to diversify.

Sellami (2015) 2002-2011 Tunisia Panel regression Real earnings management has a direct influence on cash-flows of the company which was highly leveraged to generate cash and pays the loan.

Hwang et al. (2014) 1997-2009 Taiwan Panel regression Strong bonding between innovation strategy and earning management and have a positive significant impact on firms’ value.

Herskovic et al. (2014) 1926-2010 United States Panel regression Idiosyncratic firm volatility increases the average household’s marginal utility.

Patel et al. (2014) 1992-2009 United States Panel regression Firms with higher idiosyncratic risk yield their CEOs and other executives with the higher incentives.

Kitagawa et al. (2014) 2000-2008 Japan GMM Management projections were less confidently and connected to idiosyncratic risk in relatively good information environment reduced the risk.

Fan et al. (2013) 1989-2009 United States Panel regression Abnormal returns have positively associated with idiosyncratic risk in international equity markets.

Angelis et al. (2013) 2001-2007 United States Panel regression Randomized natural testing that exogenously improved downside equity risk over the decreasing of short-retailing limitations on a random sample of US stocks.

Sarwar et al. (2013) 1926-2005 United Kingdom Panel regression Earnings remain same both at the portfolio and individual stock level.

Tariverdi et al. (2013) 2004-2010 Iran Multiple regression statistical models Ownership structure, company growth, and Firm size three incentives factors have a significant relationship with earnings management while the financial leverage has an insignificant relationship with earnings management.

Dadbeh et al. (2013) 2002-2008 Iran Panel regression Asymmetry information has positive effects on earnings management.

Alam et al. (2013) 2005-2010 Pakistan GMM Bank and family control have an adverse effect on the firm’s risk whereas ownership structure has a significant effect on the firm’s risk.

Mahmoodabadi et al. (2013) 2006-2012 Iran Panel regression A managerial market relationship between external financing and the accrual anomaly was affected by long-term accruals and working capital accruals consistently.

Geng et al. (2013) 1999-2009 China Panel regression Influence of accruals quality on capital cost was rising with the increase of fundamental risks.

Rani et al. (2013) 1980-2008 Suva, Fiji Islands Panel regression Earnings management was a universal phenomenon and increased (decreased) the monitoring cost.

Hall et al. (2013) 2000-2009 United States Panel regression The probability of earnings management was increased and earnings management has been providing an advantage to the managers or the firm.

Zamri et al. (2013) 2006-2011 Malaysia Panel regression A negative relationship between leverage and real earnings management (REM) which in turn, could affect the quality of accounting earnings.

Sayari et al. (2013) 1999-2008 Tunisia Panel regression The large firms’ preference was reducing their earnings because governments focus closely on their profits to finance the state budget by leveraging greater taxes.

He et al. (2012) 1992-2008 United States Panel regression A positive relationship between uncertainty and managerial incentives of the firm.

Albuquerque et al. (2011) 1992-2006 United States Robust Test A positive significant relationship between a firm’s operational risk, the monitoring of manager’s actions and the level of CEO equity incentives.

McAnally et al. (2010) 1992-2008 United States Panel regression CEO incentives are adversely related to downside risk but positively related to upside risk.

Markarian et al. (2010) 1989-2006 United States Panel regression A negative relationship exists amongst smoothing return and idiosyncratic volatility.

Rajgopal et al. (2010) 1962-2001 United States Panel regression Earning quality was accompanying with higher idiosyncratic returns volatility.

Bakke et al. (2010) 1991-2004 United States GMM Stock-market mispricing did not a?ect the investment in big firms.

Forester et al. (2009) 1994-2000 Ontario Panel regression A negative relationship between the earning management and firm risk measures and a positive relationship between earning management and firms’ value.

Gray et al. (2009) 1992-2005 Australia Panel regression Costs of debt and equity for Australian firms were highly motivated by accruals quality arising from economic essentials.

Badertscher et al. (2009) 1997-2009 United States Panel regression Firms with managing earnings for opportunistic explanations were less predictive whereas firms with managing earnings for informational reasons were more predictive to the future cash flow.

Irvine et al. (2008) 1964-2003 United States Panel regression Movement in idiosyncratic volatility was positively linked with a progressively competitive environment in which ?rms have less market influence and negative covariance in ?rm performance.

Michelacci et al. (2008) 1973-2003 Canada Panel regression Countries with low levels of risk diversi?cation opportunities performed relatively poorer in segments that were categorized by high idiosyncratic volatility.

Cornett et al. (2008) 1994-2003 United States Panel regression Impact of earnings management significantly increased the dignified importance of governance variables and declined the impact of incentive-based compensation.

Skaife et al. (2006) 1990-2009 United States Panel regression A significant positive relationship between the zero-return measure and the magnitude of returns which was reflected in returns.

Goyal et al. (2003) 1962-1999 United States Panel regression Significantly positive relationship between idiosyncratic risk and average stock returns on the market by using the CAPM model.

Ali et al. (2003) 1976-1997 United States Fama-French Three-Factors model A positive relationship with returns proposing that risk factors were not accountable for the value-to-price ratio effect.

Beneish et al. (2002) 1985-1997 United States Panel regression Income-increasing accruals were significantly lower when the accompanied by abnormal insider selling and higher when buying the abnormal insider.

Vuolteenaho (2002) 1954-1996 United States Panel regression Firm-level stock returns were mainly determined by cash-flow, cash flows were absolutely linked to a representative stock whereas expected-returns were tremendously interrelated across the firms.

Xie et al. (2001) 1992-1996 United States Panel regression Earnings management was less expected to occur or usually less occurs in corporations whose boards include both autonomous external executives and directors with company experience.

Heninger (2001) 1984-1998 China Panel regression Auditor litigation has a negative effect on total accruals which implies that income decreasing has a negative effect on total accruals.

Campbell et al. (2001) 1962-1997 United States Granger causality test Increased in the volatility of firm-level relative to the market volatility; correlation among individual stocks and the advisory power of the market model for a representative stock have been weakened and the number of stocks has increased.

Collins et al. (2000) 1988-1997 United States Panel regression Post-earnings statement implication was overstated or declined, based on the level of accruals surrounded within the earnings disclosure.

Healy et al. (1999) 1980-1990 United States Panel regression Increased in earnings management was due to the stock market observations, management’s reward, and shrink the probability of violating lending contracts and to escape the regulatory interventions.

Becker et al. (1998) 1989-1992 United States Panel regression Earnings management was captured by discretionary accruals and audit quality was accompanying with more accounting flexibility and increased revenue comparatively more than the discretionary accruals.

Dechow et al. (1995) 1980-1990 China Panel regression The earnings management was significantly positively influenced by executive compensation motivation. Explanatory power of accrual based models was relatively low for earning management.

Chapter No.4
RESEARCH METHODOLOGY AND DATA
4.1 Introduction
The third chapter of the thesis intends to introduce the econometric model, measuring accrual management, measuring both types of uncertainty i.e. idiosyncratic risk and macroeconomic risk, data, and sample.

4.2 Empirical Framework
The study includes seven variables in order to explore the effect of idiosyncratic (firm-specific) and macroeconomic uncertainty/risk on accrual management as shown below:
2990850227330Explanatory Variables
Explanatory Variables
76200227330Dependent Variable
Dependent Variable

Figure SEQ Figure * ARABIC 1: Empirical Framework
The above figure 1 shows the conceptual framework of the study. Firm-specific being independent variable are measured by sales volatility, Cash flow volatility, and return on asset volatility and macroeconomic uncertainty are the explanatory variables and measured by incorporating consumer price index, market interest rate, industrial production index, and exchange rate whereas Firm absolute discretionary accruals (earning management) being dependent variable for the study.

4.3 Econometrics Model
The major goal of the current study is to identify the factors, firm-specific and macroeconomic specific, that influence the accrual management. This chapter, particularly, discusses the empirical techniques and approaches that are implemented to estimate and test the hypotheses developed in the current study. The baseline model of the study identifies the important factors of idiosyncratic (firm-specific) and macroeconomic uncertainty/risk on accrual management, which is presented in Equation (1). The baseline model of the study is split into two models that identify the total accruals and discretionary accruals, which are presented in Equation (2), and (3), respectively.
Similarly, the baseline model of the study is extended by incorporating firm-specific and macroeconomic specific factors in Equation (1). We follow the following model based on the absolute discretionary accruals for our empirical ?ndings using the well-known model of Datta et al, (2017).

ADAit=?+ IVit?1+MVt?+VCVit?+fi+ft+ ?it (1)
where ? denotes year dummies to control for business progression effects, IVit is a vector of ?rm-speci?c treatment variables, which contains sales volatility, return on asset volatility, cash flow volatility and standard deviation of daily returns. MVt is a vector of macroeconomic uncertainty which contains industrial price index (IPI), consumer price index (CPI) and the market interest rate. VCVit denotes a vector of firm-level control variables i.e. asset growth, leverage, firm size and age, fi and ft capture firm fixed effects to control for deviations across the firms’ and ?it is the white-noise error terms in the regressions are assembled at the firm level. The dependent variable in the regression is the absolute discretionary accruals (ADA) flattened by the preceding year’s total assets.
4.4 Measuring Accrual Management
To estimate accruals management we have to differentiate between two types of accruals first one is non-discretionary accruals and the second one is discretionary accruals. Non-discretionary accruals that are indispensable accounting modifications and discretionary accruals are made at the prudence of managers to manipulate earnings. First of all, we estimate non-discretionary accruals and evaluate this accrual from total accruals to calculate the discretionary element and total accruals are defined as the difference between the net income and cash flow from operations. We use the Kothari et al. (2005) model to estimate discretionary accruals and discretionary accruals are measured by using the cash flow. Especially, we calculate non-discretionary accruals from cross-sectional regressions of total accruals on changes in sales minus change in receivables, property, plant, and equipment (PPE), and lagged return on assets (ROA) for each Fama-French model. We use lagged return on asset (ROA) as an additional regressor to normalize the impact of performance on a firm’s accruals. This approach is based on the assumption that firms in a corporation at a definite point in time are homogeneous with respect to their fundamental operations and strategy.
TAitAit-1=?11Ait-1+?2?REVitAit-1+?ARitAit-1+?3PPEitAit-1+?4ROAit-1+?it (2)
where i represent firms, t represents time. TAit is the total accrual which is calculated by net income (NI) minus cash flow from operations. ?REVit is the changes in sales, ?ARit is the change in receivables, and PPEit is the total property, plant, and equipment. All these variables are scaled by the lagged value of assets. ROAit?1 in equation (2) is calculated by using Net Incomeit-1/Ait-1. We use the estimated coefficients i.e.?1,?2,?3, and ?4 to estimate discretionary accrual as follows:
DAit??it=TAitAit-1-?11Ait-1+?2?REVitAit-1+?ARitAit-1+?3PPEitAit-1+?4ROAit-1(3)
Large values of discretionary accruals (DA) are usually taken to specify earnings management. As discretionary accruals could be optimistic (when firms’ expand earnings) or destructive (when firms’ in good ages managers cover earnings for future use), both optimistic and destructive values capture earnings management. We use GMM to estimate the impact of both types of uncertainties on accrual management.

4.5 Measuring Uncertainty
In this study, we use the following types of risk to measure the uncertainty of firms.

4.5.1 Idiosyncratic Risk
Researchers have applied different methodologies to generate a proxy for the firms’ idiosyncratic risk. For instance, Huizinga (1993) uses the conditional variance attained from a GARCH form speci?cation on income and materials cost and the Carhart (1997) uses idiosyncratic volatility by using the residuals from the four-factor model which includes Fama and French’s three factors plus momentum. To calculate the day-to-day residuals for the Fama-French model, they merely regress daily return on the market index, firm size, and book-to-market ratio factors for every month and every stock. Ghosal and Loungani (2000) analyze the ?rm level risk by the standard deviation of the ?rm’s volatile pro?t. Bo and Lensink (2005) use stock price volatility as well as the volatility of the number of personnel to examine ?rm level uncertainty. Baum, Stephan, and Talavera (2009) obtain idiosyncratic risk by measuring the standard deviation of the stock returns of the firms’. We analyze idiosyncratic volatility with accruals obtained on the basis of the financial year; the annual volatility at this point is based on the volatility of sales, Return-on-Asset, cash flows and standard deviation of daily returns. We used the generalized method of moments (GMM) to estimate the idiosyncratic risk.

4.5.2. Macroeconomic Risk
Researchers have also applied various methodologies to fabricate measures of macroeconomic risk. One common method is to use ARCH/GARCH class models in developing a measure of macroeconomic risk. For example, Driver, Temple, and Urga (2005), and Baum, Stephan, and Talavera (2009) are among others who use this methodology. Another possibility is to utilize the influencing standard deviation of a variable as in Ghosal and Loungani (2000) and Korajczyk and Levy (2003) and Graham and Harvey (2001). Though, the standard deviation-based measures undergo from extensive ongoing correlation complications in the invented chain. Baum, Caglayan, and Talavera (2010) calculated macroeconomic uncertainty by using the conditional variance found by approximating the generalized ARCH model the index of foremost macroeconomic components. Driver, Temple, and Urga (2005) using GARCH model generated a proxy for macroeconomic uncertainty from the conditional variance of industrial output. For our study, we implement consumer price index (CPI), industrial production index (IPI) and the market interest rate methods in our study to measure the macroeconomic risk by ARCH/GARCH Model.

4.5.3. Generating Macroeconomic Uncertainty
Similar to the instance of evaluating idiosyncratic uncertainty, researchers have used various methods to calculate the measures of macroeconomic uncertainty. Aizenman and Marion (1999) used conditional variances acquire from government outlays as a part of GDP, nominal money growth, and the real exchange rate. Ghosal and Loungani (2000) used the moving standard deviation of energy prices. Kaufmann, Mehrez, and Schmukler (2005) and Graham and Harvey (2001) utilized survey-based procedures based on the dispersion of forecasts, which are gathered from the firm or bank managers, as a measure of macroeconomic uncertainty. Driver, Temple, and Urga (2005) using GARCH specification assembled a proxy for macroeconomic uncertainty from the conditional variance of industrial output. Baum, Caglayan, and Talavera (2010) calculated macroeconomic uncertainty by using the conditional variance attained by calculating the generalized ARCH model the key of the important macroeconomic factors.

To construct macroeconomic uncertainty, we estimate an ARCH/GARCH model. We obtain conditional variances by calculating the model using quarterly data on consumer price index, industrial production index, market interest rate, and exchange rate covering the period 2000 to 2016. The mean equation of the GARCH model with ARMA (1 1) is presented as follows.
Yt=?+?Yt-1+ ??t-1+?t(4)
where Y is the underlying macroeconomic variables, namely as consumer price index, industrial production index, market interest rate, and exchange rate, ? and ? are parameters to be estimated, Yt-1 is the lagged value of the uncertainty variable and ? is the error term. The conditional variance (? 2t) of the uncertainty variables is demonstrated as follows.

?t2=?1+?2?t2-1+?3?t2-1(5)
where ?2 is the square of the lag value of error term. The conditional variance series is used as a proxy for macroeconomic uncertainty. Once we get the conditional variances for each series, we then annualize the monthly conditional variances by taking the average over monthly to match the occurrence of our uncertainty measure with that of the panel data.

4.6 Estimation Method
While explaining the role of idiosyncratic and macroeconomic uncertainty on accrual management in Pakistani Non-Financial firm-level panel data covering the period 2000 to 2016.  The autoregressive conditionally heteroskedastic (ARCH) model is a statistical model for time-series firm-year observation data that defines the alteration of the existing error term or modification as a determination of the real dimensions of the preceding time periods error terms. To make the time variance of firm-specific risk, we estimate the ARCH/GARCH model on firms’ volatility of sales for each individual firm incorporated in the sample over the scrutinized period. ARCH model is used to explain an altering, feasibly volatile variance while an ARCH model could probably be used to define a slowly increasing variance over the time. The ARCH model is suitable as the error variance in a time-series pursue an autoregressive model if an autoregressive moving average model (ARMA) model is anticipated for the error variance, the model is a generalized autoregressive conditionally heteroskedastic (GARCH) model. Firm-specific uncertainty is also estimated based on the square of the residuals of firms’ sales. Three measures of macroeconomic uncertainty are calculated by using the conditional variance attained by estimating the ARCH/GARCH model for consumer price index (CPI), industrial production index (IPI), market interest rate (INT), and exchange rate (REX). Various alternative methods to calculate both types of uncertainties are used to ensure the robustness tests of uncertainty effects. To ease the problem of endogeneity, the robust two-step system generalized method of moments (GMM) estimator is used to evaluate the econometric model originally designed by Arellano and Bover (1995), and then additionally proposed by Blundell and Bond (1998). The GMM is the keystone of semi-parametric valuation frameworks and it is a common technique for measuring the econometric model. This method allows researchers to incorporate lagged values of the variables as tools to lessen the problem of endogeneity and the GMM operates all the linear moment settings stated by the model.

4.6.1: Measurement of Variables
Table 4.1 represents the construction and measurement for various variables are incorporated in the regression model of our study for exploring the influence of idiosyncratic (firm-specific) and macroeconomic uncertainty/risk on accrual management listed at Pakistan Stock Exchange (PSX).

Table 4.1: Measurement of Variables
Variables Measurements
Dependent Variables
Absolute Discretionary Accruals (ADA) The absolute discretionary accruals deflated by previous year’s total assets (TA). Total Accruals which are calculated by net income minus cash flow from operation. We used GMM method to estimate the ADA
2847975781050Explanatory Variables
Firm-specific Uncertainty Cash flow Volatility Previous year period scaled by one-year lagged assets and obtained from the Autoregressive (AR) model
Sales Volatility Return on Asset (ROA) Volatility Macroeconomic Uncertainty 18027652540000Consumer Price Index (CPI) Macroeconomic uncertainty is based on conditional variances by measuring an ARCH/GARCH model
Market Interest Rate (INT) Industrial Production Index (IPI) Exchange Rate (REX) 4.7 Data and Sample
The ?rm-level data for Pakistan are haggard from Pakistan Stock Exchange during the period 2000 to 2016. The actual sample covers a total of 400 non-financial ?rms’ listed at Pakistan Stock Exchange (PSX). In order to eliminate the effects of heterogeneity across ?rms, we scale all the ?rm-speci?c and macroeconomic variables are determined by the availability of the key variable, cash flow from operations, volatility of sales, volatility of return on asset, the standard deviation of daily returns, industrial production index, consumer price index, market interest rate and exchange rate, to estimate accruals. We define each firm’s year observation based on an ARCH/GARCH model. We include 400 non-financial firms’ in our sample because of the differential nature of their financial statements. This study runs the Generalized Method of Moments (GMM) to examine the effects of earning management on two different uncertainties containing idiosyncratic and macroeconomic uncertainty. We first conduct all empirical analysis on the whole 400 sample period then we evaluate the results in each firms’ year observation period. The choice of the sample period is based on the reflection of adequate observations so that meaningful factor assessments can be obtained. Data are taken from several matters of Balance Sheet Analysis issued by State Bank of Pakistan (SBP). We permit entry and exit of firms’ to alleviate the problem of sample collection biases. The data for macroeconomic risk variables are taken from International Financial Statistics (IFS) database.

Chapter No. 5
EMPIRICAL FINDINGS
5.1 Summary Statistics and Correlations
The summary statistics of firm-specific variables used in the estimation is presented in Table 5.1. It summarizes a given set of data through several measures i.e. mean, standard deviation (Std.Dev), Minimum and Maximum values of the firm-specific and Macroeconomic uncertainty variables of a data set. The mean value of cash flow volatility shows that firms invest about 14.9% of their total asset, on average with 184.4% standard deviation. The standard deviation indicates that the cash flow volatility of firms is highly disseminated across the firm-year observations. The exchange rate volatility has the lowest standard deviation with a value of 0.000 among the firm-specific variables.

Table 5.1: Summary statistics of idiosyncratic and macroeconomic uncertainty
Variables
Obs
Mean
Std. Dev
Min
Max
Absolute discretionary Accruals 2732 0.004 1.067 8.87e-07 55.785
Lagged Absolute discretionary Accruals 2408 0.049 1.136 8.87e-07 55.785
Cash Flow Volatility 6177 0.149 1.844 0 102.149
Sales Volatility 6137 0.394 0.396 0 6.235
Return on Asset Volatility 6188 15.927 182.327 0 9721.409
Consumer Price index Volatility 6437 0.028 0.018 0.009 0.147
Market Interest rate Volatility 6437 0.186 0.071 0.082 0.415
Industrial price index Volatility 6437 0.001 0.003 0.000 0.022
Exchange Rate Volatility 6436 0.000 0.000 0.000 0.001
Size
5734 14.436 1.745 2.564 20.194
Market Leverage 4839 0.283 0.380 0 1
Asset growth
5213 0.095 0.250 -1.841 6.934
Age 5752 31.856 17.802 0 155
Notes: Table 5.1 presents summary statistics of the uncertainty measures used in the study. The sample of the study included 400 non-financial firms listed at Pakistan Stock Exchange (PSX) and the sample period is from 2000 to 2016. Firm-Specific uncertainty is measured by firm uncertainty such as volatility of cash flow, sales volatility, and return on asset volatility. However, Macroeconomic uncertainty is based on the conditional variances of Industrial Production Index, Consumer Price Index, market interest rate, and market exchange rate that were estimated by using ARCH/GARCH model. Likewise, size, leverage, Asset growth, and Age are used as control variables. The symbols used for firm-specific and macroeconomic uncertainty are as shown in table 5.1 above.
We observe the average value of sales volatility is 0.149 with 0.396 standard deviations and a maximum value of 6.235. Similarly, the mean value and standard deviation of return on assets volatility presented in the table is 15.927 and 182.327, respectively. The standard deviation of return on assets volatility indicates a higher volatility.

The mean (SD) value of the macroeconomic uncertainty measures based on consumer price index, industrial production index, market Interest rate, and exchange rate (REX) are 0.028 (0.018), 0.186 (0.071), 0.001 (0.003), and 0.000 (0.000) with a standard deviation of 0.000, lowest among other variables and this standard deviation indicates that the ratio is less volatile than the other firm-specific variables in the estimation. The mean value and standard deviation of firm size, market leverage, asset growth and age reported in the table are 14.436 (1.745), 0.283 (0.380), 0.095 (0.250) and 31.856 (17.802) respectively.
The mean value of absolute discretionary accruals is 0.00469707 which shows that the mean of absolute discretionary accruals is 0.00469707 with the largest observation of 55.78576 and minimum observation of 8.87e-07 and the standard deviation is 1.067341. The mean value of lagged absolute discretionary accruals is 0.0498169 with observation 2408 and the largest value is 55.78576 and minimum observation of 8.87e-07 and the standard deviation is 1.13685.
Various statistical methods, such as moving standard deviations, autoregressive (AR) speci?cations, ARCH/GARCH displaying, stochastic volatility models, and overview based strategies, have been utilized as a part of literature to generate measures of risk. For instance, Driver et al. (2005) and Baum et al. (2009) use the conditional variance got from the estimation of GARCH speci?cations for manufacturing output and the list of driving macroeconomic indicators, respectively, to measure macroeconomic uncertainty. Similarly, Aizenman and Marion (1999) also use the conditional variance obtained by measuring a GARCH model for government spending, the nominal money supply, and the real exchange rate to develop a proxy for macroeconomic uncertainty. Conversely, Caglayan and Rashid (2013) and Rashid (2013) utilize the conditional variance by measuring an ARCH model for quarterly real GDP to create a proxy for macroeconomic uncertainty. Ghosal and Loungani (2000) utilize the moving standard deviation of the federal funds rate (FFR) and vitality costs as a proxy for macroeconomic uncertainty/risk. Kaufmann et al. (2005) and Graham and Harvey (2001) use overview construct techniques situated in light of the scattering of the figure to proxy macroeconomic uncertainty/risk.

However, we examine an ARCH/GARCH model for a regularly balanced month to month real GDP to create a proxy for macroeconomic risk. The monthly conditional variance acquired from the ARCH/GARCH estimation is annualized by taking a month to month average and utilized as a proxy for macroeconomic risk in our study of empirical analysis. Macroeconomic uncertainty is based on the conditional variances of Industrial Production Index, Consumer Price Index, market interest rate, and market exchange rate by measuring the ARCH/GARCH model over the examination period and macroeconomic volatility can be estimated based on historical data. The ARCH/GARCH model is estimated for the examination period 2000-2016 to get robust, balanced, and smooth estimates. The basic ARCH/GARCH results are given in tables B.1, B.2, B.3, and B.4 below in Appendix B, which shows that the coefficient values are positive and summation of RESID and ARCH/GARCH is slightly less than unity. The estimated coef?cient of the ARCH/GARCH term (0.965, 0.812, 0.849, and 0.353 respectively) is less than one and appears statistically signi?cant. The diagnostic test statistics show that there are no staying ARCH/GARCH effects left in the uniform residuals.

Table 5.2: Summary Statistics of Idiosyncratic and Macroeconomic Uncertainty
Variables Obs Mean Std.Dev Min Max
Absolute Discretionary Accruals 2732 0.046 1.067 8.87e-07 55.785
Firm-Specific Uncertainty
6137 3.769 7.910 0 141.39
Macroeconomic Uncertainty 6436 0.054 0.017 0.031 0.116
Notes: Table 5.2 presents summary statistics of the uncertainty measures used in the study. The sample of the study included 400 non-financial firms listed at Pakistan Stock Exchange and the sample period is from 2000 to 2016. Firm-Specific uncertainty is measured by firm uncertainty index which is composite of cash flow volatility, sales volatility and return on assets volatility. However, Macroeconomic uncertainty is based on the macroeconomic uncertainty index which is the combination of conditional variances of Industrial Production Index , Consumer Price Index, market interest rate, and market exchange rate that were estimated by using ARCH/GARCH model.

5.2 Correlation Matrix
To investigate whether the uncertainty proxies measure similar movements in the business earnings activity and macroeconomic environment, we estimate their correlations. The estimated correlations are presented in Table 5.3. The table indicates that the correlation coefficients are very low and they are also statistically insignificant at any acceptable level of significance. Hence, we conclude that each of our measures captures a different element of the risk that a firm may face in its earnings management decisions.

Table 5.3: Correlation between Dependent and Explanatory Variables
Explanatory Variables
Correlation coefficients P-Value
Lagged Absolute Discretionary Accruals 0.0029 0.8888
Cash flow Volatility -0.0098 0.6078
Sales Volatility -0.0126 0.5099
Return-on-Asset Volatility -0.0080 0.6768
Consumer Price Index Volatility 0.0389 0.0419
Market interest rate Volatility -0.0065 0.7346
Industrial Price index Volatility -0.0052 0.7868
Exchange Rate Volatility -0.0062 0.7455
Size 0.0173 0.3666
Market Leverage 0.0300 0.1239
Asset Growth 0.5137 0.0000
Age -0.0171 0.3717
Note: Table 5.3 shows the correlation between the dependent variable i.e. Absolute Discretionary Accruals (ADA) and the Explanatory Variables are shown in the table. Absolute Discretionary Accruals is defined as the total accruals are management’s decisions and measurements about cash flows for making accounting earnings well imitate a firms’ underlying economic presentation. Total accruals are the sum of discretionary and non-discretionary accruals. Cash flow is income before tax plus depreciation and amortization. Sales of a firm, and return on asset indicate the one-year lagged ratio of income to assets and are calculated by dividing net income by average total assets. However, Macroeconomic uncertainty is based on the conditional variances of Industrial Production Index, Consumer Price Index, market interest rate, and market exchange rate by measuring an ARCH/GARCH model over the examination period. The firm size which is the natural logarithm of total assets, leverage is calculated as the firm’s overall debt divided by its total assets. Ag indicates the asset growth which is calculated as the change in total assets scaled by one-year lagged assets. Age represents the firms’ establishment year.

In Table 5.3, we represent a simple correlation with p-value between dependent and the other explanatory variables used in the practical analysis. The correlation coefficient of asset growth is significantly higher than the coefficient of any other variable in the table. The correlation coefficient suggests that all the three firm-specific uncertainty have a negative association with firms’ earning management and appear statistically highly insignificant.
In connection with correlation with the macroeconomic uncertainty, Table 5.4 provides evidence for the presence of a significant and negative relationship between all types of macroeconomic risks except consumer price index and appears statistically highly insignificant. Likewise, we observe that macroeconomic uncertainty measure based on a conditional variance of Consumer price index has a positive correlation with the firms’ earning management and appear statistically highly significant at the margin. The firm size and market leverage are insignificant and have a positive correlation with firms’ earning management because firms with high leverage have been connected with proximity to the violation of debt covenants, such firms’ may utilize discretionary accruals to manage their earnings upward (Defond and Jiambalvo, 1994) . The term asset growth has a positive correlation and is significant. Whereas, the last term representing age has a negative correlation and appears insignificant. These correlation estimates provide preliminary evidence on the negative relationship between firms ‘earning management and both firm-specific uncertainty as well as macroeconomic measures.

Table 5.4: Correlation between Uncertainty Measures
Variables
Firm-Specific Uncertainty Macroeconomic Uncertainty
Firm-Specific Uncertainty
1.0000 Macroeconomic Uncertainty 0.0002
(0.9888) 1.0000
Note: Table 5.4 shows the correlation matrix between two different risk measures such as Firm-Specific (Idiosyncratic risk) and Macroeconomic uncertainty. Firm-Specific Uncertainty is measured by firm uncertainty index which is composite of cash flow volatility, sales volatility, and return-on-assets volatility. However, Macroeconomic uncertainty is measured by macroeconomic uncertainty index which is composite of conditional variances of industrial production index, consumer irice index, and market interest rate and market exchange rate that were measured by using ARCH/GARCH Model.

The estimated correlations are presented in Table 5.4. The table shows that the correlation coefficients are very low but they are statistically significant at any acceptable level of significance. Therefore, we conclude that each of our measures seizures a different component of the risk that a firm may face in its earning management and financial decisions.

5.3 The Baseline Model
We estimate a baseline model to compare our results with the studies that do not take into account uncertainties effects while examining the determinants of the firms’ earning management. We present the estimated results of the baseline model in Panel A and provide diagnostic tests in Panel B of Table 5.5.

Table 5.5: Explaining the impact of idiosyncratic and macroeconomic risk on earnings management
Model 1
Without Uncertainty Model 2
With Firm Specific Model 3
With Macroeconomic Model 4
With Uncertainty
Panel A: Dependent Variable: Absolute Discretionary accruals
Variables Coefficient
P value Coefficient
P value Coefficient
P value
Coefficient
P value
Lagged Absolute discretionary Accruals -0.001***
(0.000) 0.004***
(0.001) 0.006***
(0.001) 0.006***
(0.002)
Cash Flow Volatility 0.049***
(0.004) 0.046***
(0.006)
Sales Volatility -0.007***
(0.002) -0.010***
(0.002)
Return on Asset Volatility -0.000***
(0.000) -0.000***
(0.000)
Consumer Price index Volatility 0.021***
(0.001) 0.018***
(0.001)
Market Interest rate Volatility -0.024***
(0.0006) -0.024***
(0.000)
Industrial price index Volatility 7.970***
(1.632) 5.821***
(0.928)
Exchange Rate Volatility -10.940***
(0.264) -10.640***
(0.323)
Size
-0.002***
(0.000) -0.003***
(0.000) -0.005***
(0.0006) -0.004***
(0.000)
Market Leverage 0.001***
(0.000) -.0003***
(0.000) -0.000***
(0.000) -0.000***
(0.000)
Asset growth
0.049***
(0.000) 0.095***
(0.000) 0.102***
(0.0002) 0.104***
(0.000)
Age
-0.000***
(0.000) 0.0001***
(0.000) 0.000***
(0.000) 0.000***
(0.000)
_cons
0.071***
(0.000) 0.066***
(0.001) 0.105***
(0.002) 0.087***
(0.002)
Panel B: Diagnostic Test
Hansen Statistics 328.45832 193.11201 157.78200 143.90 197
P-value 1.000 0.643 0.988 0.998
AR (2) -0.26
0.91 1.33 1.35
P-value 0.797 0.365 0.185 0.178
Firm-year obs. 2231 2231 2231 2231
No. of Firms 362 362 362 362
No. of instruments 210 210 210 210
F-Statistics 3.17e+12
1.18e+10 7.79e+10 3.39e+08
P-value
0.000 0.000 0.000 0.000
Note: *** specifies the level of statistical significance at the 5%, and 1%, respectively. Values in the parenthesis are p-values, J-statistic the Hansen (1982) test of over classifying limitations. Values in the brackets are the degree of freedom. The serial-correlation in residuals is estimated by using AR(2), the Arellano and Bond (1991) test. In all four models, the one period lagged values of the first deference of the right-hand side variables aside from uncertainty estimates are used as tools for the equations in variables level for both firm-specific and macroeconomic uncertainty.

Table 5.5 shows that the value of F-Statistics of model 1 which indicates that there is no autocorrelation. The results indicate that the overall model is significant because the p-value is 0.000 which is less than 0.05. The AR(2) value is negative i.e.-26%. We estimate four different models where each model has at least one different type of uncertainty measure except Model 1. The estimated coefficients for uncertainty variables are statistically significant and have negative signs across all four models.
However, in Model 2, we find that there is a negative relationship between firm-specific uncertainty (volatility of sales and return on assets volatility) and the accrual management of the firms, significant showing that firm-specific uncertainty have an impact on the degree of accruals management undertaken by the firms whereas volatility of cash flow has a positive relationship with accrual management. That is, firms when facing higher firm-specific uncertainty decreases their earning management. Firms do so because when they become unclear about their future earnings than they are likely to cut down their earnings. Thus, from the above results of model 2, it is concluded that the model is statistically fit for the study, and empirical findings supports the hypothesis of the study i.e. H2. All control variables are of the estimated sign and significant in all the four models. However, earnings management mechanism that is used more aggressively by the firms’ with high idiosyncratic (firm-specific) volatility (John et al. 1998).

When we turn to the effects of Model 3 by excluding firm-specific uncertainty and including macroeconomic uncertainty measure based on CPI, IPI, INT, and REX, we observe that macroeconomic uncertainty has also a positive (negative) and significant impact on the accrual management undertaken by the firms. However, in Model 3, we find that there is a positive relationship between CPI and IPI with the absolute discretionary accrual whereas the negative relationship between INT and REX with the absolute discretionary accrual, both positive and negative discretionary accruals of a strong positive association between firm-specific, macroeconomic volatility, and earnings management. This result suggests that firms increase their earnings when macroeconomic conditions become uncertain. Hence, from the above results of model 3, it is concluded that the model is statistically fit for the study and supports the hypothesis of the study that is H3. Whereas, the negative effect of macroeconomic uncertainty on firms’ earning is consistent with the findings of Caballero and Pindyck (1996), Bergstresser, and Philippon (2006). One potential reason for this negative effect is the unprofitable financial markets and delicate economic circumstance that is a collective phenomenon of underdeveloped countries like Pakistan.

When we compare the impact of firm-specific uncertainty on firms’ accrual management with the macroeconomic uncertainty, in Model 4, we examine that both types of uncertainty have differential effects on the earnings of the firms. However, the coefficient of macroeconomic uncertainty looks positive (negative) and statistically significant. Likewise, the results concerning other firm-specific variables are also similar, both in terms of sign and statistical significance, to those in Model 2 and 3. This Model suggests that our results on the effect of firm-specific uncertainty on the accrual management of firms are robust to several proxies of firm-specific uncertainty. Especially, we find that the magnitude of the coefficient of the macroeconomic uncertainty is greater than that of the firm-specific uncertainty in all four models. This suggests that firms are more delicate to macroeconomic uncertainty than idiosyncratic uncertainty. Therefore, our results are consistent with the study of Graham et al. (2005) that managers favor smooth earnings because volatile earnings lead to greater valuation risk and estimated models support the hypothesis of the study i.e. H4 that firms’ earnings are more volatile to the firm-specific and macroeconomic uncertainty, and hence to higher risk premia. Though, our results are contrasted with the findings of Hutton et al.’s (2009).
5.4 The Augmented Firms’ Absolute Discretionary Accrual Model
To empirically analyze how firms react to idiosyncratic (firm-specific) uncertainty and variations in macroeconomic conditions, the proposed accrual management model represented in equation 1 is estimated for Pakistani non-financial firms. We use different measures for each type of risk in our empirical analysis, for firm-specific we use sales volatility, cash flow volatility and return-on-assets volatility whereas, for macroeconomic uncertainty, we use consumer price index, industrial production index, interest rate, and exchange rate.
Table 5.6: The effect of idiosyncratic risk volatility on the absolute level of discretionary accruals
Model 1
With CF Model 2
With CF and Sale Model 3
With CF, sales and ROA
Panel A: Dependent Variable: Absolute Discretionary accruals
Variables Coefficient
Coefficient
Coefficient
Lagged Absolute discretionary Accruals 0.006***
(0.000) 0.006***
(0.001) 0 .004***
(0.001)
Cash Flow Volatility 0.032***
(0.000) 0.030***
(0.001) 0 .049***
(0.004)
Sales Volatility -0.000***
(0.000) -0.007***
(0.001)
Return on Asset Volatility -0.000***
(0.000)
Size
-0.004***
(0.000) -0.004***
(0.000) -0.003***
(0.000)
Market Leverage
-0.000***
(0.000) -0.0004***
(0.000) -0.0003***
(0.000)
Asset Growth
0.094***
(0.000) 0.095***
(0.000) 0.095***
(0.000)
Age
0.0001***
(0.000) 0.0001 ***
(0.000) 0.0001***
(0.000)
_cons
0.070***
(0.000) 0.075***
(0.001) 0.066***
(0.001)
Panel B: Diagnostic Test
Hansen Statistics 178.93203 171.91202 193.11 201
P-value 0.887
0.939 0.643
AR (2)
0.89 0.90 0.91
P-value 0.374 0.369 0.365
Firm-year obs.

2231 2231 2231
No. of Firms 362 362 362
No. of instruments 210 210 210
F-Statistics 6.06e+10 1.03e+09 1.18e+10
P-value
0.000 0.000 0.000
Note: *** specifies the level of statistical significance at the 5%, and 1%, respectively. Values in the parenthesis are p-values, J-statistic the Hansen (1982) test of over classifying limitations. Values in the brackets are the degree of freedom. The serial-correlation in residuals is estimated by using AR(2), the Arellano and Bond (1991) test. In all three models, the one period lagged values of the first deference of the right-hand side variables together from firm-specific uncertainty estimates are used as tools for the equations in variables level for uncertainty measures in all three Models.

As one can see that the absolute discretionary accruals are highly significant with a positive sign across all three models in Table 5.6. This implies that there is a persistent effect in firms earning management. It is also verified that the firms’ cash flow volatility is statistically significant with a positive sign and indicates a positive relationship with the absolute discretionary accruals of the firms whereas firms’ sales volatility and return on asset volatility have a negative sign in model 2 and 3 respectively.
Table 5.7: The effect of Macroeconomic uncertainty volatility on the absolute level of discretionary accruals
Model 1
With CPI Model 2
With CPI and INT Model 3
With CPI, INT and IPI Model 4
With CPI,INT, IPI and REX
Panel A: Dependent Variable: Absolute Discretionary accruals
Variables Coefficient Coefficient Coefficient
Coefficient
Lagged Absolute Discretionary Accruals
0.005***
(0.000) 0.005***
(0.001) 0.006***
(0.002) 0.006***
(0.001)
Consumer Price index Volatility 0.087***
(0.000) 0.043***
(0.002) 0.033***
(0.003) 0.021***
(0.001)
Market Interest rate Volatility -0.035***
(0.000) -0.045***
(1.357) -0.024***
(1.632)
Industrial price index Volatility 10.224***
(0.000) 7.97***
(1.632)
Exchange Rate Volatility -10.940***
(0.264)
Size
-0.005***
(0.000) -0.005***
(0.000) -0.005***
(0.000) -0.005***
(0.000)
Market Leverage
0.001***
(0.000) 0.0001***
(0.000) -0.0003***
(0.000) -0.0006***
(0.000)
Asset Growth
0.096***
(0.000) 0.098***
(0.000) 0.098***
(0.000) 0.102***
(0.000)
Age
0.0001***
(0.000) 0.000***
(0.000) 0.0001***
(0.000) 0.000***
(0.000)
_cons
0.091***
(0.000) 0.100***
(0.000) 0.101***
(0.001) 0.105***
(0.002)
Panel B: Diagnostic Test
Hansen Statistics 169.74 203 159.72202 158.60201 157.78200
P-value 0.957 0.987 0.988 0.988
AR (2) 0.86 1.03 1.06 1.33
P-value 0.390 0.305 0.290 0.185
Firm-year obs. 2231 2231 2231 2231
No. of Firms 362 362 362 362
No. of instruments 210 210 210 210
F-Statistics
9.74e+09 1.92e+07 2.30e+08 7.79e+10
P-value
0.000 0.000 0.000 0.000
Note: *** specifies the level of statistical significance at the 5%, and 1%, respectively. Values in the parenthesis are p-values, J-statistic the Hansen (1982) test of over classifying limitations. Values in the brackets are the degree of freedom. The serial-correlation in residuals is estimated by using AR(2), the Arellano and Bond (1991) test. In all four models, the one period lagged values of the first deference of the right-hand side variables together from macroeconomic uncertainty estimates are used as tools for the equations in variables levels for uncertainty measures in all four Models.

As one can see that the absolute discretionary accruals are highly significant with a positive sign across all four models in Table 5.7. This implies that there is a persistent effect in firms earning management. It is also verified that the consumer price index is statistically significant with a positive sign and indicates a positive relationship with the absolute discretionary accruals of the firms whereas interest rate and exchange rate have a negative sign in model 2, 3 and 4 respectively. Likewise, industrial production index is statistically significant with a positive sign and indicates a positive relationship with the absolute discretionary accruals of the firms.

5.5 Robustness Checks
While scrutinizing the effects of uncertainty on firms’ earning management we include both firm-specific (idiosyncratic risk) and macroeconomic uncertainty in the same model as we conjecture that these uncertainties are jointly operational. Though, one may contend that the firm-specific uncertainty may drive the effect of macroeconomic uncertainty on the earning management of firms or vice versa. Thus, to confirm that our results on the influence of one type of uncertainty on firms’ earning management are driven by another type of uncertainty, we measure only the effect of a single type of uncertainty on firms’ earning. For this purpose, we estimate three models, each model incorporate one type of risk measurement and then combine both types of risk in Model 3. Specifically, Model 1 comprises the effect of firm-specific uncertainty only, whereas, Model 2 estimates the impact of macroeconomic uncertainty and in Model 3, we incorporate the impact of both types of uncertainties. The results of the robust two-step system GMM estimator are described in Table 5.8.
Table 5.8: The effect of idiosyncratic and macroeconomic risk index on earnings management
Model 1
With Index Fu Model 2
With Index Mu Model 3
With Index Fu and Mu
Variables Coefficient
Coefficient
Coefficient
Lagged Absolute Discretionary Accruals
0.006***
(0.000) 0.006***
(0.000) 0.006***
(0.000)
Firm-Specific Uncertainty Index
-0.000***
(0.000) -0.000***
(0.000)
Macroeconomic Uncertainty Index
-0.156***
(0.000) -0.156***
(0.000)
Size
-0.004***
(0.000) -0.004***
(0.000) -0.004***
(0.000)
Market Leverage
-0.000***
(0.000) -0.001***
(0.000) -0.001***
(0.000)
Asset Growth
0.093***
(0.000) 0.095***
(0.000) 0.095***
(0.000)
Age
0.000***
(0.000) 0.000***
(0.000) 0.000***
(0.000)
_cons
0.082***
(0.000) 0.094***
(0.000) 0.094***
(0.000)
Panel B: Diagnostic Test
Hansen Statistics 174.77203
167.87203 173.47202
P-value
0.925 0.966 0.928
AR (2)
0.86 1.08 1.08
P-value
0.388 0.281 0.281
Firm-year obs.

2231 2231 2231
No. of Firms
362 362 362
No. of instruments 210 210 210
F-Statistics
9.42e+08 1.86e+10 2.19e+10
P-value 0.000 0.000 0.000
Note: *** specifies the level of statistical significance at the 5%, and 1%, respectively. Values in the parenthesis are p-values, J-statistic the Hansen (1982) test of over classifying limitations. Values in the brackets are the degree of freedom. The serial-correlation in residuals is estimated by using AR(2), the Arellano and Bond (1991) test. In all three models, the one period lagged values of the first deference of the right-hand side variables together from macroeconomic uncertainty estimates are used as tools for the equations in variables levels for uncertainty measures in all four Models.

In Table 5.8, Model 1, we estimate the impact of firm-level uncertainty proxies by the composite index constructed based on the unpredictable parts of firms’ volatility of cash flows, volatility of sales and return on assets volatility and has a negative relationship with firms’ accrual management. Similarly, in Model 2, the measure of macroeconomic uncertainty included in the model is also derived from the conditional variance obtained by estimating the ARCH/GARCH models for consumer price index, industrial production index, interest rate and exchange rate and has a positive relationship with firms’ absolute discretionary accruals. The estimated coefficients demonstrate that both of uncertainty indices are positively (negatively) and statistically significantly related to the absolute discretionary accruals of the firms. Further, one can also observe from the table that consistent with our earlier findings, the composite index of firm-specific uncertainty is more negatively related to firms’ accrual management. This suggests that as compared to macroeconomic uncertainty, uncertainty about firm-specific more negatively impacts firm-level earnings. The results reveal that both types of uncertainty negatively and statistically significantly affect the earning management of firms, although we include them separately. This confirms that the one type of risk measure does not drive the effect of the other type of risk measure.

5.6 Testing of Hypotheses
From the results discussed above, we have accepted or rejected the following hypotheses:
Particulars Accept/ Reject
H1: Firms facing more idiosyncratic uncertainty are likely to engage more in earnings management. Rejected
H2: Firms facing higher idiosyncratic volatility are likely to engage less earnings management. Accepted
H3: When macroeconomic conditions are more uncertain then the firm will do more earnings management. Accepted
H4: When idiosyncratic volatility and macroeconomic conditions are more uncertain then the firm will do more earnings management. Accepted
Based on the above literature review, ?rms’ report more discretionary accruals when managers are confident about their future prospects (Datta et al, 2016). Hence, the above-discussed results confirm that our results are robust to the different proxies of both firm-specific and macroeconomic uncertainty and Indicate that the effects of both types of uncertainty remain significant on firms’ earning management. Furthermore, these findings demonstrate the results in Kelly (2007) and support the above-mentioned hypothesis that firms with high idiosyncratic (firm-specific) risk are likely to engage fewer earnings. Likewise, empirical findings suggest that firms’ increases their earnings when macroeconomic conditions become more uncertain (Bergstresser, and Philippon 2006). Generally, we ?nd that this phenomenon is more articulated at ?rms whose managers have more incentives to manage earnings and to boost the growth of the economy. Furthermore, we find that the idiosyncratic risk and macroeconomic risk proxies and variables are associated with earnings management.
The study adds an important new dimension to the earnings management and variables using prior literature within each construct studied under the corporate finance and management fields that are discretionary accruals, corporate strategy, investments, and firms’ performance. Hence, the objectives of the study have been achieved because the impact of one type of uncertainty on firms’ earning management is driven by another type of uncertainty. Similarly, the results revealed that both types of uncertainty have significant and negative effects on firms’ earning management.

Chapter No.6
CONCLUSION
6.1 Introduction
This study is an attempt to identify various idiosyncratic (firm-specific) and macroeconomic level determinants of earning management for the sample firms. We developed the baseline econometric model of the study based on various firm-specific characteristics which incorporate sales volatility, cash flow volatility, and return on asset volatility, to capture the differential effects of the firm-specific characteristics.

Next, the baseline model of the study is extended by incorporating various macroeconomic level factors. Likewise, the extended model of the study incorporates the consumer price index, market interest rate, industrial production index, and the exchange rate.

The study adds an important new dimension to the earnings management literature by founding a relationship between idiosyncratic risk (firm-specific) and macroeconomic uncertainty and the degree to which firms manage their earnings. We developed the baseline econometric model of the study based on various firm-specific and macroeconomic conditional variance characteristics and the model was based on a comprehensive sample of 400 non-financial firms during the period covering 2000 to 2016.

6.2 Key Findings
The results demonstrated momentous and negatively statistical coefficients of uncertainties with their one-period lagged values which suggest the presence of significant earnings management to the volatility of cash flows, sales, return on assets, consumer price index, industrial price index, market interest rate, and exchange rate for the sample firms over the period from 2000 to 2016. We document that both firm-specific and macroeconomic uncertainty has a significant and negative impact on earnings management. Simply put, higher idiosyncratic risk induces managers to more viciously manage earnings. Precisely, our results suggest that firms are prone to earn more when they face variations in their sales, cash flows, and return-on-assets. This negative effect on earnings management of firm-specific uncertainty makes sense as risky firms generally do more effort to manage earnings. Regarding the role of macroeconomic uncertainty, our results suggest that macroeconomic volatility has also a significant and negative impact on firms’ earning management but firms increase their earnings when macroeconomic conditions become uncertain. These results confirm that our results are robust to the different proxies of both firm-specific and macroeconomic uncertainty and Indicate that the effects of both types of uncertainty remain significant on firms earning management. Furthermore, we find that the idiosyncratic risk and macroeconomic risk proxies and variables are associated with earnings management.
Comparing the magnitude of the estimated coefficients, we see that as compared to firm-specific uncertainty, macroeconomic uncertainty have a more negative influence on the earnings management of firms operating in Pakistan. This piece of evidence suggests that firm managers take into consideration the state of macroeconomic conditions at a great point when making earning decisions. In the country, such as Pakistan, where institutions are weak, financial and economic policies lack consistently, financial markets suffer from brushings, and there is lack of earning-favoring environment, it is obvious that variations in macroeconomic indicators make firms reluctant in earning their businesses.
We examine that ?rms report more discretionary accruals when ?nancial markets are confident about their future prospects. Consistent with markets being more likely to characteristic performance to fortunes during uncertain periods, in this way, we create motivation for the firms’ to report higher earnings and to maintain reserves during higher-uncertainty periods. Generally, we ?nd that this phenomenon is more articulated at ?rms whose managers have more incentives to manage earnings and to boost the growth of the economy, these results have new ramifications for the role of idiosyncratic (firm-specific) and macroeconomic uncertainty may play in modifying the managerial earnings decision-making.

The research findings suggest that firm managers should significantly consider the uncertainty discretionary accruals relationship, particularly while formulating earning policies. Firms should design such sales strategies that help stabilize firms’ sales and cash flow streams, and thus, they prevent their accruals not to be negatively affected. These strategies may contain sales variation at domestic markets, internationalization, product quality, competitiveness, and more R&D expenditures to innovate their products. The analysis also suggests that macroeconomic uncertainty has a comparatively stronger negative influence than firm-specific risk on firms’ earning management.
6.3 Policy Recommendations
A large strand of empirical literature is present on the accruals management and both types of uncertainty that have been taken place across the different economies. The current study covers the research line in the literature of firms’ earnings management and in particular provides analysis for the impact of both firm-specific and macroeconomic conditional uncertainty factors, on the absolute discretionary accruals for the firms in the context of Pakistan.

Arif et al. (2016) state that understanding of the accrual management is crucial for the managers. For instance, the finding shows that above the appropriate levels, both firm-specific and macroeconomic uncertainty has a negative influence on the accrual management performance of the firms. Thus, the managers are compelled to consider the negative influence of firm-specific and macroeconomic uncertainty prior to any decisions regarding the firm level adjustments. Furthermore, the investors can also consider the uncertainty of a firm to make appropriate investment decisions.

In this way, our research is an attempt to contribute to the field of corporate finance. Eventually, our study will attract the attention of corporate executives, top management, firms manager, academia, researchers, and investors to understand how the role of idiosyncratic (firm-specific) and macroeconomic uncertainty affects the firms’ accrual management and their decisions of earnings. The outcome that firms make their earnings in times of both the firm-specific and macroeconomic uncertainty indicates that increase in firms’ accrual management during the periods of firm-specific and macroeconomic turmoil may boost the process of more volatile earning recovery. Ultimately, the policymakers should design such policies that encourage firms to invest more in economic crisis periods, which, in turn, would boost the growth of the economy and help to overcome the problem of a downturn/recession.

6.4 Limitation and Future Direction
Our study, however, has a couple of restrictions. It is important that examining earnings is a controversial subject (Courteau et al. 2011). Our commitment identifies with substitution factors utilized while empirically approving theoretical assumptions. Our conclusions, however, ought to be deciphered with caution. Firm-specific and macroeconomic uncertainty used to characterize the earnings management may to numerous different impacts. Some variables are examined through a panel regression technique. In addition, the used methodologies slip taking into account the interactions between independent variables. In any case, these associations are overlooked by multivariate systems which instinctively process weight of the different independent variables. At this level, we point out that Dechow et al’s (2012) new approach of identifying dynamic accruals-based earnings management was truly censured as a few authors believe that despite everything it battles with the shortcomings of discretionary accruals’ traditional estimation techniques (Gerakos, 2012).

This study can also be completed by taking different measures of firms’ earnings management other than overproduction. Sample period of this study is taken as 2000 to 2016; further analysis can be done by taking data of more number of years. In this study upcoming performance is only tested after one year of the firms’ engaged in accruals/earning management activities, the study can be extended to examine the earnings management’s on the second year and so on. This study has measured the impact of both types of uncertainties (i.e. Firm-specific and macroeconomic uncertainty) on accruals management but future research can also be done to see the earnings management impact on large and small firms separately. It can also be examined which type of the accruals/earnings management has the simplest impact on future performance.
This study would be beneficial for investors and analysts to determine how firm’s earnings are enhanced. It would be helpful in an understanding of investors how managers play with their wealth and they would be able to consider the accurate earnings of the firm. It would open a window for investors to identify the intrinsic value of the firm. This study is also valuable for managers to judge the after effects of earnings management. Our thesis has laid empirical findings on the “Accrual management in Pakistani non-financial firms: Explaining the role of idiosyncratic risk and macroeconomic uncertainty” of 400 non-financial firms’ listed at Pakistan Stock Exchange. It also provides a pathway for the further researches. The study proposes that further research may include more variables and stretch the study period to achieve more accurate and reliable results.

REFERENCES
Abarbanell, Je?ery, and Reuven Lehavy. (2003). Biased forecasts or biased earnings? The role of reported earnings in explaining apparent bias and over/underreaction in analysts’ earnings forecasts, Journal of Accounting and Economics 36, 105–146.
Aizenman, J. and Marion, N. (1999), “Volatility and investment: interpreting evidence from developing countries”, Economica, Vol. 66 No. 262, pp. 157-179.
Alam, A., & Shah, S. Z. (2013). Corporate Governance and its Impact on Firm Risk. International Journal of Management, Economics and Social Sciences, 2(2), 76 –98.
Albuquerque , A., Papadakis , G., & Wysocki , P. (2011, May). The impact of risk and monitoring on CEO equity incentives, 3(2), 1-50.

Alia, A., Hwang, L. -S., & Trombley, M. A. (2003). Residual Income Based Valuation Predicts Future Stock Returns: Evidence on Mispricing vs. Risk Explanations. The Accounting Review, 78(2), 377-396.

Angelis, D. D., Grullon, G., & Michenaud* , S. (2013, August 28). Downside Risk and the Design of CEO Incentives: Evidence from a Natural Experiment. EVCA Working Paper, 6647
Arif, Salman, Nathan Marshall, and Teri Lombardi Yohn, 2016, Understanding the relation between accruals and volatility: A real options-based investment approach, Journal of Accounting and Economics 62, 65–86.

Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58(2), 277-297.

Ashbaugh-Skaife , H., Gassen, J., & LaFond, R. (2006). Does Stock Price Synchronicity Represent Firm-Specific Information? The International Evidence. working paper, 6167, Massachusetts Institute of Technology, Cambridge, MA.
Bachmann, Ru¨diger, and Christian Bayer, 2014, Investment dispersion and the business cycle, American Economic Review 104, 1392–1416.

Badertscher , B. A., Collins , D. W., & Lys , T. Z. (2009, May). Earnings Management and the Predictive Ability of Accruals with Respect to Future Cash Flows. Working Paper Series 4149, European Central Bank.
Baker, Scott R., and Nicholas Bloom, 2013, Does uncertainty reduce growth? Using disasters as natural experiments, Working paper, National Bureau of Economic Research.

Bakke , T.-E., & Whited , T. (2010, May). Which Firms Follow the Market? An Analysis of Corporate Investment Decisions. The Review of Financial Studies, 23(5), 1941-1980.
Bartov, Eli, Ferdinand A. Gul, and Judy S. L. Tsui, 2000, Discretionary-accruals models and audit quali?cations, Journal of Accounting and Economics 30, 421–452.

Basu, Susanto, and Brent Bundick, 2012, Uncertainty shocks in a model of e?ective demand, Working paper, National Bureau of Economic Research.

Baum, C. F., Caglayan, M., & Talavera, O. (2008). Uncertainty determinants of firm investment. Economics Letters, 98(3), 282-287.

Baum, C. F., Caglayan, M., & Talavera, O. (2010). On the sensitivity of firms’ investment to cash flow and uncertainty. Oxford Economic Papers, 62(2), 286-306.

Baum, C. F., Stephan, A., & Talavera, O. (2009). The effects of uncertainty on the leverage of nonfinancial firms. Economic Inquiry, 47(2), 216-225.

Beaudry, P., Caglayan, M., & Schiantarelli, F. (2001). Monetary instability, the predictability of prices, and the allocation of investment: An empirical investigation using UK panel data. American Economic Review, 91(3), 648-662.

Becker, C. L., Defond, M. L., Jiambalvo, J., & Subramanyam, K. R. (1998). The Effect of Audit Quality on Earnings Management. Working Paper, 2109, European Central Bank.

Beneish, & Vargus. (2002). insider trading, earnings quality, and accrual mispricing. The Accounting Review, 77(3), 755-791.
Bianchi, Francesco, Cosmin Ilut, and Martin Schneider, 2014, Uncertainty shocks, asset supply and pricing over the business cycle, Working paper, National Bureau of Economic Research.

Bidder, Rhys M., and Matthew E. Smith, 2012, Robust animal spirits, Journal of Monetary Economics 59, 738–750.

Bloom, Nicholas, 2009, The impact of uncertainty shocks, Econometrica 77, 623–685.
Bloom, Nicholas, Max Floetotto, Nir Jaimovich, Itay Saporta-Eksten, and Stephen Terry, 2011, Really uncertain business cycles, Working paper, National Bureau of Economic Research.

Caballero, R. J., & Pindyck, R. S. (1996). Uncertainty, Investment, and Industry Evolution. International Economic Review, 37(3), 641-662.
Caglayan, M., & Rashid, A. (2013). The response of firms’ leverage to risk: evidence from UK public versus non-public manufacturing firms. Economic Inquiry, 52(1), 341–363.

Campbell, J. J., Lettau, M., Malkiel, B. G., & Yexiao, X. (2001, February). Have Individual Stocks Become more Volatile: An Empirical Exploration of Idiosyncratic Risk. The Journal of Finance, 6(1),220-250.
Campbell, J. and Shiller, R. (1988), “The dividend-price ratio and expectations of future dividends and discount factors”, Review of Financial Studies, Vol. 1 No. 3, pp. 195-228.

Chang , S. H., Wang , T. S., Chiu , A. A., & Huang , S. Y. (2015). Earnings management and idiosyncratic risk-evidence from the postSarbanes-Oxley Act period. Investment Management and Financial Innovations, 12(2), 117-126.

Chang, F. Y., Hsin, C. W., & Hou, S. R. (2016). Performance of Financial Hedging and Earnings Management under Diverse Corporate Information Quality. The Accounting Review, 57(4), 1-52.

Chen, C. R., & Wang, S. Y. (2015, July). Idiosyncratic Risk and the Cross-Section of Expected Stock Return: A Threshold Regression Approach. Proceedings of the Second European Academic Research Conference on Global Business, Economics, Finance and Banking (EAR15Swiss Conference), 53(3), 1-16.

Chiou , J. R., Hsueh, M. H., & Hsieh , C. C. (2014, May). How Does Earnings Management Affect Innovation Strategies of Firms? Journal of Accounting and Economics, 39(2), 1-41.
Christiano, Lawrence J., Roberto Motto, and Massimo Rostagno, 2014, Risk shocks, American Economic Review 104, 27–65.

Collins, D., & Hribar , P. (2000). Earnings-based and accrual-based market anomalies: one effect or two. Journal of Accounting and Economics, 29(1), 101-124.

Cornett, M. M., Marcus, A. J., & Tehranian, H. (2008). Corporate Governance and pay-for-performance: The impact of earnings management. journal of Finanacial Economics, 87(4), 357-373.
DeAngelo, Linda Elizabeth, 1988, Managerial competition, information costs, and corporate governance: The use of accounting performance measures in proxy contests, Journal of Accounting and Economics 10, 3–36.

Debnath , P. (2017, February). Assaying the Impact of Firm’s Growth and Performance on Earnings Management: An Empirical Observation of Indian Economy. International Journal of Research in Business Studies and Management, 4(2), 30-40 .

Dechow, P. M., Sloan, R. G. & Sweeney, A. P. (1995). Detecting Earnings Management. The Accounting Review, Vol. 70, No. 2, pp. 193-225.
Dadbeh, F., & Mogharebi , N. (2013). A study on effect of information asymmetry on earning management: Evidence from Tehran Stock Exchange. Management Science Letters, 3(1), 2161–2166.
Demsetz, Harold, & Kenneth lehn. (1985). The Structure of Corporate Ownership: Causes and Consequences. journal of Political Economy, 33(2), 1155-1177.

Dennis, P., Strickland?, D., & Hall, K. (2009, November). The Determinants of Idiosyncratic Volatility. FEDS Working Paper No. 2003-20.
Dichev, Ilia D., John R. Graham, Campbell R. Harvey, and Shiva Rajgopal, 2013, Earnings quality: Evidence from the ?eld, Journal of Accounting and Economics 56, 1–33.

Driver, C., Temple, P. and Urga, G. (2005). “Pro?tability, Capacity, and Uncertainty: A Model of UK Manufacturing Investment”, Oxford Economic Papers, Vol. 57 No. 1, pp. 120-141.
Datta , S., Datta , M. I., & Singh, V. (2017). The impact of idiosyncratic risk on accrual management. International Journal of Managerial Finance, 13(1), 70-90. doi:10.1108/IJMF
Ebrahimi , S. K., Nasab , A. B., & Karim , M. (2016). “Evaluating The Effect of Accruals Quality, Investments Anomaly and Quality of Risk on Risk Premium (Return) of Stock of Listed companies in Tehran Stock Exchange”. Problems and Perspectives in Management , 14(3),115-140.

Fama, E., & French, K. (1993). Common risk fatosr in the returns on stocks and bonds. Journal of Financial Economics, 33(3), 3-56.

Fan, S., & Yu, L. (2013). Accrual anomaly and idiosyncratic risk: international evidence. The International Journal of Business and Finance Research, 7(4), 63-75.
Fang, Vivian W., Allen Huang, and Wenyu Wang, 2017, Imperfect accounting and reporting bias, Journal of Accounting Research .

Farhadvand, S. M., & Jalilian, O. (2017). Investigating Risk Effect and Profit Management on Bank Credit Risk. International Journal of Economics and Financial Issues, 7(3), 548-554. Retrieved from http: www.econjournals.com
Ferri, Fabrizio, Ronghuo Zheng, and Yuan Zou, 2017, Uncertainty in managers’ reporting objectives and investors’ response to earnings reports: Evidence from the 2006 executive compensation disclosures, Working Paper 17-14, Columbia Business School.
Fischer, Paul E, and Robert E Verrecchia, 2000, Reporting bias, The Accounting Review 75, 229–245.

Foerster, S. R., Sapp, S. G., & Shi, Y. (2009, May 3). The Impact of Management Earnings Forecasts on Firm Risk and Firm Value. Journal of Business, 58(2), 1-34.

Fu, F. (2009). idiosyncratic risk and the cross-section of expected stock returns. journal of Financial Economics, 81(1),24-37.

Geng, Z., Wang, Z., Song, T., Liu, T., Chi, W., Yu, Y., . . . Li , T. (2013). The Effect of Fundamental Risk of Listed Companies on the Market Pricing of Accruals Quality. iBusiness, 5(2), 6-12.
Ghosal, V., & Loungani, P. (2000). The differential impact of uncertainty on investment in small and large businesses. Review of Economics and Statistics, 82(2), 338-343.

Goyal, A., & Clara, P. S. (2003, June). Idiosyncratic Risk Matters! The journal of Finance, 3(1), 975-1007.

Graham, John R., Campbell R. Harvey, and Shiva Rajgopal, 2005, The economic implications of corporate ?nancial reporting, Journal of Accounting and Economics 40, 3–73.

Graham, J. R., & Harvey, C. R. (2001). The theory and practice of corporate finance: Evidence from the field. Journal of Financial Economics, 60(2), 187-243.

Gray, P., Koh , P. S., & Tong, Y. (2009). Accruals Quality, Information Risk and Cost of Capital: Evidence from Australia. Journal of Business Finance & Accounting, 36(1 & 2), 51-72.
Gulen, Huseyin, and Mihai Ion, 2016, Policy uncertainty and corporate investment, Review of Financial Studies 29, 523–564.
Hackbarth, D., Miao, J. and Morellec, E. (2006), “Capital structure, credit risk, and macroeconomic conditions”, Journal of Financial Economics, Vol. 82 No. 3, pp. 519-550.

Hall, S. C., Agrawal, V., & Agrawal, P. (2013). Earnings Management and the Financial Statement Analyst. Accounting and Finance Research, 2(2),462-475.
Hatzinikolaou, D., Katsimbris, G.M. and Noulas, A.G. (2002), “In?ation uncertainty and capital structure: evidence from a pooled sample of the Dow-Jones industrial ?rms”, International Review of Economics & Finance, Vol. 11 No. 1, pp. 45-55.

He, Z., Li, S., Wei, B., & Yu, J. (2012, January). Uncertainty,Risk,andIncentives:TheoryandEvidence. International Economic Review, 37(4), 641-662.

Healy, P., & Wahlen, J. M. (1999). “A review of the earnings management literature and its implications for standard setting”. Accounting Horizons, 13(4), 365-383.

Heninger, W. G. (2001). The Association between Auditor Litigation and Abnormal Accruals, The Accounting Review, Vol. 76, No1, pp. 111-126.

Herskovic , B., Kelly , B., & Lustig, H. (2014, December 2). The Common Factor in Idiosyncratic Volatility: Quantitative Asset Pricing Implications?. International Economic Review, 37(3), 1-64.

Holthausen, R., Larcker, D. and Sloan, R. (1995), “Annual bonus schemes and the manipulation of earnings”, Journal of Accounting and Economics, Vol. 19 No. 1, pp. 29-74.

Huizinga, J. (1993). Inflation uncertainty, relative price uncertainty, and investment in US manufacturing. Journal of Money, Credit and Banking, 27(2), 521-549.
Hutton, A.P., Marcus, A.J. and Tehranian, H. (2009), “Opaque financial reports, R2, and crash risk”, Journal of Financial Economics, Vol. 94 No. 1, pp. 67-86.

Irvine , P. J., & Pontiff , J. (2009). Idiosyncratic Return Volatility, Cash Flows, and Product Market Competition. The Review of Financial Studies, 22(3), 1149-1177.
Jackson, Andrew B., Brian R. Rountree, and Shiva Sivaramakrishnan, 2016, Earnings comovements and earnings manipulation, Working paper, The University of New South Wales.

Jones, J., 1991, Earnings management during import relief investigations. Journal of Accounting Research 29: 193-228.

Jin, L. and Myers, S. (2006), “R2 around the world: new theory and new tests”, Journal of Financial Economics, Vol. 79 No. 2, pp. 257-292.
Julio, Brandon, and Youngsuk Yook, 2012, Political uncertainty and corporate investment cycles, Journal of Finance 67, 45–83.

Kelly, P. (2007), “Information efficiency and firm-specific return variation”, working paper, University of South Florida, Tampa, FL.

Kitagawa, N., & Okuda, S. Y. (2014, December). Management forecasts, Idiosyncratic Risk, and the Information Environment. Journal of Political Economy, 88(3), 1-40.
Kothari, S., Leone, A. and Wasley, C. (2005), “Performance matched discretionary accruals”, Journal of Accounting and Economics, Vol. 39 No. 1, pp. 161-197.

Levitt, A. (1998). Remarks by Chairman Arthur Levitt, the “Numbers Game”, Securities and Exchange Commission, NYU, Center for Law and Business, New York, N.Y. pp.1-14.

M, R., & ZH, N. (2016). The Effect of the Global Financial Crisis on Earning Management in Tehran Stock Exchange (Evidence From of the Products of Petroleum and Chemical Industry). Journal of Business & Financial Affairs, 5(4),475-498.

Mahmoodabadi, H., Ziyari , S., & Dehghan , F. (2013, june). Investigating the relationship between accrual anomaly and external financing anomaly in Tehran Stock Exchange (TSE). International Journal of Economics, Business and Finance, 1(5), 108 – 119.

Markarian*, G., & Albornoz , B. -d. (2010, November). Income Smoothing and Idiosyncratic Volatility. Journal of Financial Economics, 60(2), 187-243.

McAnally , M. L., Neel , M., & Rees , L. (2010, December). CEO Incentives and Downside Risk, Working Paper Series,6695,215-238,v National Bureau of Economic Research, Inc.

Michelacci , C., & Schivardi, F. (2008). Does Idiosyncratic Business Risk Matter? Working Paper,995, University of Cagliari, Canada.
Minton, B. A., & Schrand, C. (1999). The impact of cash flow volatility on discretionary investment and the costs of debt and equity financing. Journal of Financial Economics, 54(3), 423-460.
Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13(2), 187221.

Nakamura, T. (1999). Risk-aversion and the uncertainty–investment relationship: A note. Journal of Economic Behavior & Organization, 38(3), 357-363.
Pastor, Lubos, and Pietro Veronesi, 2012, Uncertainty about government policy and stock prices, Journal of Finance 67, 1219–1264.

Patel, & Cooper . (2014, September). The Effects of Idiosyncratic Risk on CEO Incentives.Review of Financial Studies, 29(3), 523-564.
Rajgopal, S. and Venkatachalam, M. (2011), “Financial reporting quality and idiosyncratic return volatility”, Journal of Accounting and Economics, Vol. 51, pp. 1-20.

Rani, P., Hussain, F. F., & Chand, P. V. (2013). Managerial incentives for earnings management among listed firms: evidence from fiji. Global journal of business research, 7(1), 21-31.
Roll, R. (1988), “R-Squared”, Journal of Finance, Vol. 43, pp. 541-566.

Sarwar, S. M., & Muradoglu, G. (2013). Macroeconomic Risks, Idiosyncratic Risks and momentum profits. Bosra Istanbul Review, 13(4), 99-114.

Sayari, S., Omri, A., Finet , A., & Harrathi , H. (2013, May). The impact of earnings management on stock returns: The case of Tunisian firms. Journal of Accounting and Auditing, 1(1), 7-21.

Sellami , M. (2015). Incentives and Constraints of Real Earnings Management: The Literature Review. International Journal of Finance and Accounting, 4(4), 206-213 .

Sellami, M. (2016). The interaction between real and accrual-based earnings management: analysis based on the mandatory IFRS adoption. International Journal of Accounting and Economics Studies, 4(1), 24-31.
Shivakumar, Lakshmanan M., 2000, Do ?rms mislead investors by overstating earnings before seasoned equity o?erings, Journal of Accounting and Economics 29, 339–371.

Skaife , H. A., Gassen , J., & LaFond , R. (2006). Does Stock Price Synchronicity Represent Firm-Specific Information? The International Evidence. Journal of Financial Economics, 60(2), 137-163.
Stein, Luke C. D., and Elizabeth C. Stone, 2014, The e?ect of uncertainty on investment, hiring, and R&D: Causal evidence from equity options, Working paper, Arizona State University.
Strobl, Gu¨nter, 2013, Earnings manipulation and the cost of capital, Journal of Accounting Research 51, 449–473.

Tariverdi, Y., Keighobadi, A. R., & Tavasol, R. (2013). The Effective Motivations OnEarning Management. International Research Journal of Applied and Basic Sciences, 5(6), 730-741.

Vuolteenaho, T. (2002). What Drives Firm-Level Stock Returns? The Journal of Finance, 7(1),575-598.

Wei, S. and Zhang, C. (2006), “Why did individual stocks become more volatile?”, Journal of Business, Vol. 79 No. 1, pp. 259-292.

Watts, R.L., and J.L. Zimmerman. ‘Positive accounting theory: a ten year perspective’. The Accounting Review 65: 131–156. 1990.

Xie , B., Davidson , W., & DaDalt , P. (2001, July). Earnings Management and Corporate Governance: The Role of the Board and the Audit Committee. Journal of Corporate Finance, 2(1),295-325. doi:10.1016/S0929
Zamri, N., Rahman, R. A., & Mohd Isa, N. S. (2013). The Impact of Leverage on Real Earnings Management. Procedia Economics and Finance, 7(6), 86 – 95.
Zeira, J. (1990). Cost uncertainty and the rate of investment. Journal of Economic Dynamics and Control, 14(1), 53-63.

APPENDIX
Appendix A: Construction of Variables
Variables Definition
Idiosyncratic risk Idiosyncratic risk is the logarithm of the volatility of the residuals managed from Carhart (1997) four-factor models. We simply revert the daily stock additional return on the daily market portfolio’s additional returns, firm size, book-to-market ratio, and momentum elements for each month. Idiosyncratic risk is measured on the basis of the square of the residuals of the firms’ sales obtained from the autoregressive (AR) model for each firms’ year observation incorporated in the sample.
Market-to-book ratio Market-to-book ratio is a ratio used to estimate the worth of a firm by equating the book value of a company to its market value. Book value is measured by viewing at the firm’s historical cost, or accounting value. Market value is determined in the stock market through its market capitalization. Market-to-Book is the logarithm of the one-year lagged ratio of market capitalization to the book value of the firm.

Return on assets (ROA) Return on assets (ROA) is the one-year lagged ratio of income to assets and is calculated by dividing net income by average total assets.
Sales, ROA, and cash flow volatility Sales volatility, ROA volatility, and cash flow volatility are the standard deviation of actual trades, cash flows, and return on assets individually, measured over the prior year period scaled by one-year lagged assets.

Asset growth Asset growth is calculated as the change in total assets scaled by one-year lagged assets.
Variables Definition
Leverage Leverage is calculated as the firm’s overall debt divided by its total assets.

Standard deviation Standard deviation is a measure of the spread of scores within a set of data and daily returns are calculated as the logarithm of the one-year lagged standard deviation of stock returns of the firms’.

Macroeconomic uncertainty Macroeconomic uncertainty is based on the conditional variances of Industrial Production Index (IPI), Consumer Price Index (CPI) and market interest rate (INT) and market exchange rate (REX) by measuring an ARCH/GARCH model over the examination period.

Earning Management Earnings, sometimes entitled the bottom line or net income, are the only most essential element in the financial statements. These financial statements indicate the point to which a company has involved in value-added activities. Earnings management is used to provide financial reports that present an overly positive view of a firm’s business activities and financial position. 
Property, Plant And Equipment
Property, Plant, and equipment (PPE) is a firm asset that is dynamic to business operations but cannot be easily liquidated, and the overall value of PPE can range from very low to very high compared to total assets. PPE is tangible items that are projected to be used in more than one period and that are used in manufacturing, for rental, or for administrative management.

Total Accruals The total accruals are management’s decisions and measurements about cash flows for making accounting earnings well imitate a firm’s underlying economic presentation. Total accruals are the sum of discretionary and non-discretionary accruals. 
Variables Definition
Discretionary Accruals The accruals module managers choose inside the adaptability of accounting principles in modifying a firm’s cash flows is the discretionary accruals. 
Non- Discretionary Accruals The part of the accrual that is executed by the accounting regulator in adjusting a firm’s cash flows is the non-discretionary accruals.

Appendix B: ARCH/GARCH) Tables
Table B.1 Consumer Price Index (CPI) for Fixed Effect
Dependent Variable: D(LNCPI) Method: ML – ARCH Date: 12/05/17 Time: 17:08 Sample (adjusted): 2000M03 2016M12 Included observations: 202 after adjustments Convergence achieved after 31 iterations MA Backcast: 2000M02 Presample variance: backcast (parameter = 0.7)
GARCH = C(4) + C(5)*RESID(-1)^2 + C(6)*GARCH(-1)
Variable Coefficient Std. Error z-Statistic Prob.  
C -0.002270 0.008142 -0.278816 0.7804
AR(1) 0.964500 0.034268 28.14583 0.0000
MA(1) -0.984942 0.029266 -33.65488 0.0000
Variance Equation C 0.001298 0.000833 1.558431 0.1191
RESID(-1)^2 0.164732 0.045132 3.650024 0.0003
GARCH(-1) 0.801201 0.040253 19.90409 0.0000
R-squared 0.009398     Mean dependent var 0.001038
Adjusted R-squared -0.000558     S.D. dependent var 0.174236
S.E. of regression 0.174285     Akaike info criterion -0.803862
Sum squared resid 6.044650     Schwarz criterion -0.705597
Log likelihood 87.19007     Hannan-Quinn criter. -0.764104
Durbin-Watson stat 2.101651 Inverted AR Roots       .96 Inverted MA Roots       .98 Table B.2 Market Interest Rate (INT) for Fixed Effect
Dependent Variable: D(INT) Method: ML – ARCH Date: 12/05/17 Time: 17:10 Sample (adjusted): 2000M03 2016M12 Included observations: 202 after adjustments Convergence achieved after 49 iterations MA Backcast: 2000M02 Presample variance: backcast (parameter = 0.7)
GARCH = C(4) + C(5)*RESID(-1)^2 + C(6)*GARCH(-1)
Variable Coefficient Std. Error z-Statistic Prob.  
C -0.015384 0.066179 -0.232467 0.8162
AR(1) 0.873057 0.137256 6.360783 0.0000
MA(1) -0.760268 0.184860 -4.112670 0.0000
Variance Equation C 0.031682 0.014560 2.175962 0.0296
RESID(-1)^2 0.055902 0.028333 1.973002 0.0485
GARCH(-1) 0.757483 0.104826 7.226126 0.0000
R-squared 0.025209     Mean dependent var -0.025990
Adjusted R-squared 0.015412     S.D. dependent var 0.413852
S.E. of regression 0.410651     Akaike info criterion 1.074245
Sum squared resid 33.55821     Schwarz criterion 1.172510
Log likelihood -102.4987     Hannan-Quinn criter. 1.114003
Durbin-Watson stat 2.243205 Inverted AR Roots       .87 Inverted MA Roots       .76
Table B.3 Industrial Production Index (IPI) for Fixed Effect
Dependent Variable: D(LNIPI) Method: ML – ARCH Date: 12/05/17 Time: 17:09 Sample (adjusted): 2000M03 2016M12 Included observations: 202 after adjustments Convergence achieved after 127 iterations MA Backcast: 2000M02 Presample variance: backcast (parameter = 0.7)
GARCH = C(4) + C(5)*RESID(-1)^2 + C(6)*GARCH(-1)
Variable Coefficient Std. Error z-Statistic Prob.  
C 0.006502 0.003382 1.922389 0.0546
AR(1) 0.912648 0.123584 7.384822 0.0000
MA(1) -0.826385 0.165374 -4.997051 0.0000
Variance Equation C 3.82E-05 2.13E-06 17.95516 0.0000
RESID(-1)^2 0.057937 0.013532 4.281644 0.0000
GARCH(-1) 0.792036 0.012797 61.89180 0.0000
R-squared -0.076137     Mean dependent var 0.006714
Adjusted R-squared -0.086953     S.D. dependent var 0.044264
S.E. of regression 0.046149     Akaike info criterion -4.906067
Sum squared resid 0.423808     Schwarz criterion -4.807802
Log likelihood 501.5128     Hannan-Quinn criter. -4.866309
Durbin-Watson stat 2.922322 Inverted AR Roots       .91 Inverted MA Roots       .83
Table B.4 Exchange Rate (REX) for Fixed Effect
Dependent Variable: D(LNREX) Method: ML – ARCH Date: 12/05/17 Time: 17:12 Sample (adjusted): 2000M04 2016M12 Included observations: 201 after adjustments Convergence achieved after 25 iterations Presample variance: backcast (parameter = 0.7)
GARCH = C(4) + C(5)*RESID(-1)^2 Variable Coefficient Std. Error z-Statistic Prob.  
C 0.000777 0.001011 0.768861 0.4420
AR(1) 0.150120 0.091533 1.640065 0.1010
AR(2) -0.032358 0.075445 -0.428898 0.6680
Variance Equation C 0.000140 1.87E-05 7.490672 0.0000
RESID(-1)^2 0.353895 0.121010 2.924511 0.0034
R-squared 0.005243     Mean dependent var 0.000863
Adjusted R-squared -0.004805     S.D. dependent var 0.014887
S.E. of regression 0.014923     Akaike info criterion -5.641212
Sum squared resid 0.044091     Schwarz criterion -5.559040
Log likelihood 571.9418     Hannan-Quinn criter. -5.607961
Durbin-Watson stat 2.104588 Inverted AR Roots  .08+.16i      .08-.16i