Subscribe Now Subscribe Today
Research Article
 

Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals in South Korea



Kang Young Lee, R. Mesia and T. Griffin
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail
ABSTRACT

This study aims to investigate that not-for-profit hospitals in South Korea also conduct opportunistic earnings management as for-profit organizations do. Medical corporations except clinics owned by individual doctors are classifies as not-for-profit organizations in South Korea (Inheritance Tax and Gift Tax Act of Korea, 2012). On the considering launching of for-profit hospitals in South Korea, a lack of research exists to understand the types and motivations of earnings management in not-for-profit hospital settings. With the evidence of earnings management in for-profit organizations, this research involves examining whether leaders of not-for-profit hospitals in South Korea manage earnings and how they manage earnings. The study included the logistic regression models of performance-matched discretionary accruals to test whether not-for-hospitals manage reported earnings with a sample of 375 hospitals (1,500 hospital-year observations) from 2007 to 2010. The findings show that not-for-profit hospitals manage earnings to make a cookie jar reserve by using reserve fund and manage earnings negatively to increase donation revenue. Furthermore, this study found that public (large) hospitals engage in less earnings management than small sized hospitals to avoid reputation (political) costs.

Services
Related Articles in ASCI
Similar Articles in this Journal
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

Kang Young Lee, R. Mesia and T. Griffin, 2014. Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals in South Korea. Research Journal of Business Management, 8: 1-27.

DOI: 10.3923/rjbm.2014.1.27

URL: https://scialert.net/abstract/?doi=rjbm.2014.1.27
 
Received: July 19, 2013; Accepted: January 24, 2014; Published: March 08, 2014



INTRODUCTION

The reported earnings of hospitals play important roles in financing, performance evaluation, donation, negotiation talks and tax exemption benefits (Leone and Van Horn, 2005). Earnings are negatively associated with the possibilities of manager terminations (Brickley and Van Horn, 2002). Frank et al. (1990) noted that donators bear the earnings of hospitals in mind in the donation decision-making process. According to Leone and Van Horn (2005), Chief Executive Officers (CEO’s) manage earnings for various incentives such as bonuses, debts and stock prices. Burgstahler and Dichev (1997) classified the kinds of earnings manipulation in for-profit organizations as smoothing earnings for the steady growth of earnings to prevent losses and to avert decreases in earnings. Chen and Tsai (2010) reported that managers manipulated their earnings to avoid a decrease in equity prices, to mitigate tax burdens and to meet the expectations of stakeholders in the capital market.

According to the SKDI and SKHIDI (2009), launching of for-profit hospitals will improve the quality of medical services, help hospitals be innovative and boost global competitiveness by financing stabilized capital. The proponents added that Korean hospitals cannot compete with globally leading hospitals without the hospitals having a for-profit motive (Dong-A Ilbo, 2010). According to McCue and Nayar (2009), for-profit rural referral center hospitals in the United States served less complicated medical cases and had lower costs per bed. For-profit hospitals reported operating margins over 19%, whereas not-for-profit hospitals reported operating margins of only 8.1%.

Some civic groups and medical experts oppose the introduction of for-profit hospitals in South Korea. The South Korean Federation of Medical Groups for Health Rights insisted that adopting for-profit hospitals would motivate the leaders of not-for-profit hospitals to increase medical costs that come from unnecessary medical tests and high salaries for their skilled medical professionals (Bae, 2010).

In a for-profit environment, hospital leaders are compelled to escalate the costs of medical treatments to make earnings for investors, which burden people with heavy medical expenses, by focusing on uncompensated medical services (Woo, 2008). With U.S. hospital samples, Chen et al. (2009) demonstrated not-for-profit hospitals with good financial records are likely to provide more unprofitable medical services than are not-for-profit hospitals with poor records. Chen et al. (2009), however, found that financial performance did not affect unprofitable services in for-profit hospitals. Fifteen of the 25 most profitable hospitals in the United States were for-profit hospitals Some researchers have indicated the large margin is the result of efficiency and high quality in the hospital, whereas others expect the profit might result from overcharging in the local monopoly.

The problem is that researchers have not properly studied the incentives and patterns of earnings management for not-for-profit hospitals in South Korea and regulation setters have limitations in decision making on establishing for-profit hospitals. The leaders of most not-for-profit hospitals want to operate as commercial businesses alleging that it is the way to increase the quality of medical services. The opponents on launching for-profit hospitals asserted the for-profit system in hospitals would lead to increased medical fees that involve incremental charges for medical insurance subscribers.

One of the most popular and discussable topics in accounting research is earnings management. To understand the ways and patterns of earnings manipulation, empirical studies have been conducted with various approaches. CEOs manage earnings by increasing or decreasing earnings with a variety of accounting principles. From the 1970s to the early 1980s, the focus of most researches on earnings management was on the discretionary exercise through the specific choice of Generally Accepted Accounting Principles (GAAP). Since the mid-1980s, researchers have focused on the discretionary and nondiscretionary accruals measurement (Sun and Rath, 2010). Among the two types of accruals, discretionary accruals allow CEOs to move accounting revenue between accounting periods or to defer expense recognition. CEOs manipulate their earnings by discretionary accruals (Dechow et al., 1995; Healy, 1985; Jones, 1991). Imhoff (1977) used income smoothing Sun and Rath (2010) and Burgstahler and Dichev (1997) applied benchmark beating to measure earnings management.

The measurement of managers’ discretion is a key factor in testing earnings management. Three research methods are generally employed to measure earnings management. The first method is to verify discretionary accruals on the evidence of the relationship between total accruals and explanatory components. To measure earnings manipulation, Healy (1985) employed total accounting accruals but DeAngelo (1986) used the difference of total accounting accruals. Jones (1991) presented a regression model, called the Jones model, which clarifies a linear association between the total accounting accruals and the change in Property, Plant and Equipment (PPE) and sales. The current study involved testing whether a relationship exists between the total accounting accruals and the difference in PPE and sales in not-for-profit hospitals by employing the Jones model. McNichols and Wilson (1988) and Wilson (1996) noted the Jones model has validity problems in the nondiscretionary accrual proxies. The second method involved measuring a specific accrual when the accrual is of some size and needs considerable decisions (McNichols, 2000). Beaver and McNichols (1998), McNichols and Wilson (1988), Moyer (1990), Nelson (2000), Petroni (1992) and Petroni et al. (2000) used such an approach. As the discretion of managers is to be influenced in a specific accrual, the specific accrual approach is useful for understanding the behavior of specific accruals. An essential component of the research design approach using a specific accrual is to characterize its discretionary and nondiscretionary behaviors.

With the pervasive evidence of earnings management in for-profit organizations, the study involves examining whether leaders of not-for-profit hospitals manipulate earnings and how hospital leaders manage earnings in South Korea. An expectation for the study is that the methods and patterns of earnings manipulation in not-for-profit hospitals will differ from for-profit organizations and will give conclusive proof to standards setters and regulators for launching for-profit hospitals in South Korea. Besides regulators, academics and practitioners, the results of the study might provide patients with important information on how to understand the hospitals’ financial status and choosing medical organizations as customers. The results of the study also add to leadership knowledge and literature. Safeek (2008) noted chief medical officers should exercise leadership in financing, planning and marketing by establishing customer relationships, margins and outcomes as parts of leadership in the health care industry. The behavior of leadership in the health care industry wields influence over the hospital safety culture and performance (Yang et al., 2009). Regulators, including government, organizational leaders, users of financial reporting data and academic professionals, will play a part in understanding earnings manipulation in the not-for-profit hospital industry.

LITERAURE REVIEW

Prior research has covered the issues of earnings management from various perspectives. The focus of the earlier studies was primarily on the motives, techniques, restrictions and research designs of earnings management. Prior review papers such as those by Fields et al. (2001), Healy and Wahlen (1999) and McNichols (2000) established the structure and research design on earnings management. Healy and Wahlen (1999) reviewed earnings management literature and presented implications for accounting criteria and regulation setters. McNichols (2000) discussed earnings manipulation research designs.

Dechow and Skinner (2000) concluded that understanding CEO’s incentives is vital to know the desire of their involvement in earnings management, which indicates CEOs have strong incentives to achieve targets to encourage an increase in stock price. Fields et al. (2001) discussed accounting choices in regard to earnings management. Some researchers, including Dechow and Skinner (2000), Healy and Wahlen (1999) and Fields et al. (2001) offered some motives regarding why CEOs manage earnings to beat the earnings targets, including bonus plan, debt cost and stock price (Graham et al., 2005). The literature review includes a comprehensive overview about earnings management and about the issues of earnings management discussed in prior research.

Motivations: Following a review of many studies on earnings management, Kothari (2001) noted that investors and the market react to reported earnings. Chief financial officers regard reported earnings numbers such as the earnings per share as a key metric instead of cash flows (Graham et al., 2005). Due to the high cost of information processing, Graham et al. (2005) noted that investors prefer a simple indicator such as earnings per share to evaluate the performance of a company to avoid risks from investing in stocks.

Capital market motivations: One of the most common incentives that researchers discussed on earnings management was strongly associated with the capital market. Because investment is a process of risky decision making for investors, they tend to use the benchmarks of stock analysts’ expectations as a tool that forecasts the likelihood of the successful performance of firms. Graham et al. (2005) conducted a survey with a sample of the 312 Chief Financial Officers (CFOs) of listed firms. They found that 86.3 and 82.2% of the CFOs strongly or generally agreed that beating earnings thresholds helps to gain credibility with the capital market and supports share prices, respectively. Beyer (2009) documented that CEO’s manage earnings to reduce earnings surprise, that is, the expectation error of the manager on the date of an announcement for controlling the reported earnings level. Chen and Tsai (2010) noted that one of the motivations for managers to manipulate earnings in the reported financial status is to maintain confidence by banks. Chen and Tsai noted this type of earnings management comes from altruistic motivation.

Das et al. (2009) noted the style of changes in quarterly earnings plays a role in the behavioral indicator of possible earnings manipulation. Das et al. (2009) also concluded the companies that showed a sign of earnings changes during quarters reversed their direction in the fourth quarter, which indicated the earnings manipulation timing is a critical factor for detecting the earnings manipulation behavior. Failing to meet the targets of the stock market provides investors with a negative influence in stock returns (Barth et al., 1999; Kasznik and McNichols, 2002) so that CEOs seek the positive implications rather than the negative through earnings manipulation. If the targets of analysts are lower than the premanaged earnings, CEOs might decrease the earnings, also called cookie jar accounting, for bad years or might not engage in earnings management for the stock returns.

As not-for-profit organizations do not have stock market and residual claimants, if the earnings of the current accounting period are positive, there are no incentives for CEOs of not-for-profit hospitals to avoid earnings decreases (Leone and Van Horn, 2005). Hoerger (1991) and Leone and Van Horn (2005) noted that managers of not-for-profit hospitals manipulate earnings with the increases or decreases in discretionary expenditures at the end of the fiscal period to meet or beat the levels of earnings they desire.

Although, a number of researchers studied earnings management, the majority focused on for-profit publicly traded companies in the United States. Even in a for-profit environment, Cormier et al. (2000) documented that differences exist among countries. It is, therefore, significant to study if critical but differential factors are not revealed in earnings management in the South Korean not-for-profit hospital industry.

CEO compensation or contracting motivations: To reduce agency costs between stockholders and CEOs, firms often provide CEOs with incentives through stock options. CEOs’ wealth is sensitive to stock price (Habib and Hansen, 2008; Weber, 2006). Bartov and Mohanram (2004), Bushman and Smith (2001), Core et al. (2003), Jensen and Meckling (1976) and Kwon and Yin (2006) documented the relationship between CEO compensation with stocks and earnings manipulation, explaining that stock compensation for CEOs can be an effective incentive for better reported performance.

McVay et al. (2006) examined whether a positive relationship exists between stock sales proxied by managerial incentive and meeting the analyst forecasts benchmark. CEOs manage the accruals of working capital to meet the targets of earnings (McVay et al., 2006). Managers managed earnings with income-increasing discretionary accruals when the compensation for CEOs was more bounded by stock options (Bergstresser and Philippon, 2006). Laux and Laux (2009) noted that more stock incentives for CEOs were not necessary if the board supervised the managers appropriately. Fama and Jensen (1983) insisted that a directorate should approve, monitor and choose the significant decisions of the agents and reward them because agents do not share the wealth created by their decisions.

Other researchers documented that a nonlinear s-shaped relation exists between meeting earnings benchmarks and abnormal returns (Freeman and Tse, 1992; Kinney et al., 2002). Ayers et al. (2006) examined the positive relationship between discretionary accruals and winning earnings benchmarks at other spots in earnings distribution and in earnings change distribution. Ayers et al. (2006) referred to these points as “pseudo targets”.

Cost-of-debt motivations: Earnings benchmarks include just-above-zero earnings, the earnings of the previous accounting period and the expectations of analysts. Jiang (2008) examined the relationship between beating those earnings benchmarks and the debt costs with the proxies of credit ratings and initially yielded bond spread. Jiang noted that the firms that won the earnings targets enjoyed higher credit ratings in the previous accounting period and a smaller initial bond yield. It is not clear why institutional investors who dominate the bond market and can process compound information simply rely on the experience.

Regulatory motivations: As South Korea has adopted the IFRS for listed companies since January 1, 2011, leaders of corporations are expected to improve accounting transparency, accountability, comparability and reliability of financial information in the global market. IFRS accommodates a fair value accounting in firm valuation. According to Healy and Wahlen (1999), fair value accounting is aimed at reducing earnings management, providing stakeholders with proper financial information and improving the quality of decision making by regulators. The IFRS globally unified accounting standard might reduce the cost of financial reporting. Any investments from investors in for-profit hospitals in South Korea require a full understating of earnings management and accurate financial information on hospitals. According to Healy and Wahlen (1999), researchers have examined two types of regulation motivations when looking at earnings management in the perspective of regulatory incentives: industry regulations and antitrust regulations.

Earnings management methods for beating the benchmarks: Healy and Wahlen (1999) examined the distribution of reported earnings and provided evidence that CEOs participated in earnings management when they expected to miss the earnings benchmark. Habib and Hansen (2008) noted that Healy and Wahlen (1999) did not provide an immediate influence on regulators. This section shows how firm managers manage their reported earnings around the benchmarks of earnings, categorized by earnings management through specific accruals, earnings management through specific activities and earnings management through meeting or beating the targets of analysts.

Earnings management through specific accruals: A difference exists in the levels of discretionary accruals between companies’ petty profits and companies’ small losses (Dechow et al., 2003). Dechow et al. (2003) used earnings levels as a benchmark and measured the difference of discretionary accruals employing the frequency distribution of earnings. The difference in discretionary accounting accruals did not exist between companies with petty profits and companies with diminutive losses (Dechow et al., 2003). Firm managers have some incentives when they manage earnings. The companies with earnings just over the target did not show significantly different discretionary accruals compared to the companies with earnings just below the benchmark (Dechow et al., 2003). Hansen (2010) noted that the alternative benchmarks around the earnings level benchmark as well as earnings changes can influence the discretionary accruals levels. The discretionary accruals of the companies that had earnings just over the benchmark are considerably higher than the discretionary accruals for companies with earnings just below the benchmark (Hansen, 2010).

Deferred corporate taxes can detect earnings management using a modified Jones model (Phillips et al., 2003). According to Phillips et al. (2003), the expenses of deferred income tax, total accruals and abnormal accruals are useful in classifying firms as earnings management doers around the earnings changes benchmark. Ayers et al. (2006) found a positive relationship exists between discretionary accruals and beating earnings benchmarks around pseudo-benchmarks that are not at the focal point of earnings thresholds in earnings levels and in earnings changes. Cash flow can be a discretionary accrual proxy, as used by Ayers et al. (2006).

Earnings management using real activities: According to Roychowdhury (2006), some companies manipulate their earnings through real activities instead of financial activities. Real activities include lower prices, inventories and discretionary expenses to achieve the earnings benchmarks. Roychowdhury (2006) found that firms that are just above the benchmarks of earnings provided their clients with price discounts to increase short-term sales, reduced cost of goods sold using an overproduction method and lowered discretionary expenses.

Earnings management through classification shifting: McVay (2006) found that CEOs shifted key expenses such as cost of goods sold to specific accounts within the income statement. The term used for the intentional misclassification of the expense accounts within the income statement is shifting of expenses (McVay, 2006). Through shifting expenses within the income statement, managers cannot change the last calculated earnings but they can exaggerate certain earnings (McVay, 2006). CEOs use classification shifting as a tool of earnings manipulation to achieve or beat the analyst targets benchmark (McVay, 2006). Athanasakou et al. (2009) noted that leaders of larger companies in the United Kingdom shift essential expenses to other particular accounts to beat the analyst targets.

Earnings management through target guidance: To achieve and beat the analysts’ forecast benchmark, Burgstahler and Eames (2006) found that CEOs manage earnings upward and forecast downward. Brown and Higgins (2005) demonstrated that CEOs use forecast guidance by guiding the targets of analysts downward, as examined by Matsumoto (2002), in the nations that protect investors strongly more than in the nations that protect investors weakly to achieve positive earnings surprises. Matsumoto explained forecast guidance as guiding the earnings expectation of analysts downward to make the most of companies’ opportunities to meet or beat the target when accounting earnings are reported. Feng and McVay (2010) demonstrated that analysts valued the guidance of managers more heavily than expected in the perspectives of usefulness and credibility of management guidance. Brown and Higgins (2005) noted that U.S. CEOs tend to use forecast guidance more than CEOs in other nations. Athanasakou et al. (2009) found that managers of firms in the United Kingdom use earnings forecast guidance instead of discretionary accruals to meet analysts’ forecasts.

Research designs employed in prior studies: According to McNichols (2002), prior studies on earnings management mostly used the following research models for detecting discretionary accruals based on aggregate accruals, specific accruals and the distribution of earnings after management. Jones (1991) introduced the first model on aggregate accruals, which is extensively employed in earnings management studies. Hettihewa and Wright (2010) noted that the Jones model provides more descriptive evidence on actual events by weakening the supposition that nondiscretionary accruals are constant over time. The second model on specific accruals was adopted for a better understanding of abnormal and normal elements. The design was widely employed in specific industries such as banking and insurance (McNichols, 2000). Burgstahler and Dichev (1997) and Degeorge et al. (1999) developed a frequency distribution model to identify discretionary accruals founded on the earnings distribution after management. As the model explains the pattern of earnings around a definite threshold, the researcher also introduced studies that employed this frequency distribution approach.

Aggregate accruals models: Researchers in a number of studies identified abnormal accruals through the relationship between the total accruals and other independent variables. Aggregate accrual models include the Healy model, DeAngelo model, Jones model, modified Jones model, instrumental variables and generalized method of moments model, industry model and method of the statement of cash flow. The Jones model and the modified Jones model are widely accepted models to measure discretionary accruals (Dechow et al., 1995; Kothari et al., 2005; Leone and Van Horn, 2005).

Healy model: To compute normal accounting accruals, Healy (1985) employed the total accruals mean that deflated by previous total assets (τ-1) as:

Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea

where, NDAτ: Nondiscretionary accruals in year t, n: No. of years in the period and subscript, t: years in the period. The total accruals minus nondiscretionary accruals are discretionary accruals. The discretionary accruals are abnormal.

DeAngelo model: To measure normal accruals, DeAngelo (1986) employed the last period’s total accruals (t-1) that deflated by t-2 total assets as:

NDAit = Tait-1/Ait-2

where, NDAit: Nondiscretionary accruals in year t, TAit-1: Total accruals in year t-1 and Ait-2: Lagged total assets in year t-2. The difference between total accruals in year t scaled by lagged total assets in year t-1 and the computed nondiscretionary accruals in year t is the discretionary accruals.

Jones model: Jones (1991) proposed a regression model of total accruals on the change of revenues and PPE and this model uses the error term as the proxy of discretionary accruals. The Jones model can be given as:

TAit/Ait-1 = αi(1-Ait-1)+β1i(ΔREVit/Ait-1)+β2i(PPEit/Ait-1)+εit

Where:
TAit = Total accruals in year t for firm i
ΔREVit = Revenues in year t less revenues in year t-1 for firm i
PEEit = Gross property, plant and equipment in year t for firm i
Ait-1 = Total assets in year t-1 for firm i
εit = Error term in year t for firm i
I = 1,..., N firm index
t = 1,..., Ti, year index for the years included in the estimation period firm i

Modified jones model: Dechow et al. (1995) modified the Jones model and called the modification the modified Jones model. The modification involved removing the conjectured inclination that the original Jones (1991) model had from the modified Jones model to measure discretionary accruals with residual (Dechow et al., 1995). Hettihewa and Wright (2010) noted the modified Jones model estimates discretionary accruals with residuals when discretion is exerted over sales to remove the bias of the Jones model. Because discretionary accruals are computed as nondiscretionary accruals when CEOs manage part of sales, Dechow et al. (1995) insisted that a modification is necessary and developed the modified Jones model to calculate proper nondiscretionary accruals as:

NDAt = α1(1/At-1)+α2(ΔRECt)+α3(PEEt)

where, ΔRECt: Net receivables in year t less net receivables in year t-1 scaled by total assets at t-1. Total accruals are computed as:

TAit = α01(1/Ait-1)+α2(ΔREVit-ΔARit)+α3PPEitit

Where:
Tait = Total accruals in year t for firm i
Ait-1 = Total assets in year t-1 for firm i
ΔREVit = Revenues in year t less revenues in year t-1 for firm i
ΔARit = Change in account receivable from year t-1 for firm i
PPEit = Gross property, plant and equipment in year t for firm i
Δit = Error term in year t for firm i

Modified Jones model using book-to-price value ratio and the cash flows of operating activities: Larcker and Richardson (2004) proposed the modified Jones model using book-to-price value ratio and the cash flows of operating activities to measure total accruals. The growth of a firm in a proxy of the book-to-price value ratio and the current performance of operating activities in a proxy of operating cash flows should be controlled in the regression to measure the total accruals properly (Larcker and Richardson, 2004).

Modified Jones model with return on assets: The measures of performance-matched accounting adjustments based on a firm’s Return on Assets (ROA) can improve the inference reliability in studies of earnings manipulation (Kothari et al., 2005). Kothari et al. (2005) proposed a modified Jones model with ROA as the control variable of firm performance in the regression.

Industry model: Instead of using specific variables to detect the nondiscretionary accruals of firms, the industry model was developed under the supposition that companies in the same industry have the variables to measure nondiscretionary accruals in common (Bartov et al., 2000):

NDAit = α12Mediamj(TAt/Ait-1)

where, Mediamj: Total accruals median in year t deflated by total assets for all non-sampling companies in two-digit Standard Industrial Classification (SIC) codes in industry j.

Method of the statement of cash flow: Hribar and Collins (2002) noted that balance sheet models have biases in the detection of discretionary accruals when firms have actual operating changes like acquisitions and discontinued operations during the same accounting period.

The instrumental variables and generalized method of moments model: Kang and Sivaramakrishnan (1995) employed the instrumental variables and generalized method of moments model to detect earnings management. Specific accruals and total accruals models that estimate earnings manipulation are to do with simultaneity, variable error and variable, omitting issues that are likely to lessen the statistical power of a test and to draw erroneous inferences from the data (Kang and Sivaramakrishnan, 1995). Kang and Sivaramakrishnan (1995) measured the causal effect of accruals on earnings manipulation by using the incidence of Type I and II error and used the error term from the balance-based equation of current assets less liabilities as the proxy of discretionary accruals.

Specific accruals models: Researchers use these models to formulate specific accruals, often in a specified industry. The most important function of this research design is simulating the pattern of specific accruals to detect discretionary and nondiscretionary accruals (McNichols, 2000).

Allowance of loan loss: Beaver and Engel (1996) examined how bond investors priced abnormal and normal components of an important accrual with a proxy of the allowance of loan loss in the banking industry. Capital markets priced a nondiscretionary component negatively and a discretionary component positively (Beaver and Engel, 1996).

Uncollectible debt: McNichols and Wilson (1988) employed the residual from a model of bad debt provision as a proxy of discretionary accruals. Firms with exceptionally high or low levels of income manipulate earnings via income-decreasing accruals (McNichols and Wilson, 1988).

Cash flows from operations: Roychowdhury (2006) proposed the cross-sectional regression model to detect the nondiscretionary accruals of cash flows from operations as:

CFOt/At-1 = α01(1/At-1)+β1(St/At-1)+β2(ΔSt/At-1)+εt

where At is the total assets at the end of period t, St is the sales during period t and Δst = St-St-1.

Discretionary expense: Roychowdhury (2006) developed a regression model to measure the discretionary expenses as:

DISEXPt/At-1 = α01(At-1)+Β(St/At-1)+εt

where, DISEXPt is discretionary expenses in period t.

Loss reserve: Petroni (1992) documented that the CEOs of financially weak insurance companies decrease the loss reserve estimates more than other CEOs and that the managers under a legal obligation understated the estimates of loss reserves to a greater degree. Through the opportunistic behaviors of the managers in financial reporting, Petroni insisted that the CEOs enjoy more benefits with fewer costs. Beaver and McNichols (1998) focused on the discretionary accruals of reserve against loss in the property and casualty insurance industry, hypothesizing uncollectible accounts are likely to be a material or significant amount. The development of reserves against loss released 1 year after the reporting date of a balance sheet has strong explanatory power in the valuation of firms (Beaver and McNichols, 1998). Beaver and McNichols (1998) insisted that investors should recognize the managed effects on the loss reserves and modify the values of firms.

R and D: Berger (1993) proposed a model to measure the nondiscretionary accruals with R and D expenditures for the purpose of tax credit and Gunny (2010) proposed a model with R and D expenditures including the market value.

Working capital: Dechow and Dichev (2002) presented a time-series regression model to measure the quality of firm-level accruals with the change in working capital. McNichols (2002) proposed a cross-sectional regression model including the change in revenues from year t-1 to year t and PPE to the equation of Dechow et al. (2000). The model with the change in revenues and PPE strengthens the detection of earnings management (McNichols, 2002).

Beneish probit model: Beneish (1999) presented a probit model to identify earnings management of firms that achieved steep performance in financial reports using the variables of financial statements. Beneish (1999) found negative discretionary earnings from the firms who broke GAAP for 2 years after infringement. Although, Beneish’s probit regression is not a model for detecting abnormal accruals, the approach is used for studies on earnings management.

Frequency distribution models: To understand the earnings behavior to achieve or beat a threshold, the model of earnings distribution after management is employed. The model is to test the earning patterns around a threshold such as zero with the statistical distribution of earnings, which was introduced by Burgstahler and Dichev (1997) as well as Degeorge et al. (1999). Ayers et al. (2006) presented the profit model to test its relationship between discretionary accruals and meeting or beating actual thresholds. Ayers et al. (2006) measured earnings management to investigate the relation between abnormal accruals and meeting or beating the pseudo-forecasts of earnings.

MATERIALS AND METHODS

Hypothesis development: The purpose of the current quantitative study employing a regression model of performance-matched discretionary accruals was to test earnings management in the not-for-profit hospital industry by measuring discretionary accruals. Even though the Jones model and the modified Jones model are generally accepted as able to detect discretional accruals effectively (Dechow et al., 1995; Leone and Van Horn, 2005). Kothari et al. (2005) demonstrated that the methods of performance-matched discretionary accruals enhance the inference reliability from the study of earnings management. The study involved examining the statistical properties, patterns and motives of earnings management in not-for-profit hospitals in South Korea with the research designs verified using a body of prior studies on discretionary accruals. Reserve funds, donations, government subsidies, ownership types, hospital sizes and other control variables used to estimate earnings management were included in the study.

Cookie jar reserve by reserve fund: McKee (2005) mentioned a “cookie jar reserve” techniques in earnings management is one of the most successful and widely employed methods. According to Corporate Tax Act of Korea (2013) not-for-profit domestic corporations can appropriate their reserve funds to perform their purposed essential businesses and to donate to designated organizations prescribed by the law. The relevant reserve fund is accumulated as reserves in settling the profits that accrue during the corresponding business year, which shall be deemed to be included in deductible expenses. This benefit may have not-for-profit hospitals make or maintain tax-exempt status via a reserve fund when their pre-managed earnings exceed the benchmark.

The allowance of reserve funds for essential businesses as deductible expenses might provide the managers of not-for-profit hospitals with substantive motives to manage earnings for meeting or beating the benchmarks. For the essential businesses in the future, Corporate Tax Act of Korea (2013) allows hospitals to reserve 100% of interest and dividend revenues and 50% of the for-profit revenues less donation expenditures. The hypothesis was that the managers of not-for-profit hospitals have a motivation to manipulate earnings positively for creating a cookie jar reserve:

H1: The reserve fund of a hospital is positively associated with its discretionary accruals

Donor’s departure prevention: For-profit organizations have the motivation to increase their firm values in the stock market with stock prices by meeting or beating the targets of analysts. Not-for-profit hospitals are not legally required to pay dividends from the net income they made to their shareholders due to the lack of claim for dividend. Leaders of not-for-profit organizations, however, are under increasing pressure to manage their organizations in a business manner (Chetkovich and Frumkin, 2003; Dart, 2004). Thus donors may be more likely to make donations to the high earning organizations as evidence of the realization of charitable purposes. Frank et al. (1990) documented that donors reckon with the profits of hospitals for charity donations.

On the other hand, donors may tend to make more donations to the financially strapped hospitals because donators do not believe the organizations that have high returns are in financial straits to receive external financial support (Leone and Van Horn, 2005). As hospitals in South Korea are classified as not-for-profit organizations by law, a hospital may have incentives to maintain or increase donation revenues instead of focusing on debt costs and CEO reputation. The hypothesis was that the CEOs of hospitals in South Korea have the motivation to manage earnings for making or maintaining donation revenues:

H2: The donation revenue of a hospital is negatively association with its discretionary accruals

Government subsidies: In addition to central or local governments, not-for-profit hospitals are incorporated by religious, educational, social welfare, charitable and other foundations. For the improvement of public health, Department of Health and Human Services (2013) allows governments to subsidize part or all of the expenses for medical facilities, operations and R and D for hospitals. The CEOs of not-for-profit hospitals might feel motivated to manipulate their earnings negatively to receive government subsidies:

H3: The government subsides of a hospital is negatively associated with its discretionary accruals

Ownership type: Even though the CEO of a privately owned hospital promotes public interest in its operation, the CEO may be also interested in its bonus by reporting outstanding financial performance. Kramer and Santerre (2010) found the managers of not-for-profit hospitals were more motivated by financial compensation at the expense of medical treatments. Meanwhile, public hospitals, generally, may be more affected by the interested parties than private ones in pursuit of public interests. The leaders of public hospitals might feel motivated to manage their earnings negatively to receive good reputation in pursuit of public interests and avoid interested parities’ criticism in excessive profit seeking:

H4: The public hospital is negatively associated with its discretionary accruals

Hospital size: For-profit large sized firms have more rooms to manage earnings and more pressures to meet the targeted earnings (Biger and Hoang, 2008). But the reverse may be the case in not-for-profit hospitals because most of their essential businesses are normally focused on the development of public interest and more rigorously monitored by stakeholders. The hypothesis was that the bigger hospitals are likely to report less earnings for the avoidance of stakeholders’ unwelcome attention:

H5: A hospital size is negatively associated with its discretionary accruals

Research design: The study includes discussions on the importance of earnings manipulation, the specific accounts performed by managers of not-for-profit hospitals for earnings management and the descriptive characteristics of not-for-profit hospitals in South Korea. The study also includes discussions on the existence of different motives and patterns of earnings management between for-profit organizations and not-for-profit hospitals. The main independence variables include reserve fund, donation revenue, government subsidies, ownership types and hospital sizes.

Regression model: To examine the hypotheses, a multiple regression using non-operating expenditure and revenue accounts with dummy variables was conducted. To improve the inference reliability, a firm’s performance-matched discretionary accrual method used by Kothari et al. (2005) was employed as:

Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea

where, DACCit: Performance-matched discretionary accruals deflated by assets, which Kothari et al. (2005) measured in year t-1 for hospital i; Reserve_Fundit: 1 if the account of the reserve fund on the balance sheet is more than zero, 0 otherwise in year t for hospital i; Donation_Revenueit: 1 if the account of donation revenue on the income statement is more than zero, 0 otherwise in year t for hospital i; Government_Subsidiesit: 1 if the account of government revenue on the income statement is more than zero, 0 otherwise in year t for hospital i; Owner_Typeit: 1 if the hospital is owned by central or local governments, 0 otherwise in year t for hospital i; Sizeit: log of total assets on balance sheet in year t for hospital i; TACit: Total accruals at the beginning of the year t for hospital i; Leverageit: Current and long-term debt deflated by total assets in year t for hospital i; Locationit: 1 if the hospital is in a city with a population over 1 million, 0 otherwise in year t for hospital i; Yearit: 1 if observation is in year j of the data, 0 otherwise; ∈it: the error term.

Kothari et al. (2005) documented that the models of traditional discretionary accrual measures are incorrectly specified, so the study involved employing the approach of the performance-matched discretionary accruals. The log of assets and the changes of sales were included in the regression. The location of not-for-profit hospitals in South Korea was also taken into consideration to control for the number of prospective patients. South Korea has seven metropolitan cities that have more than 1 million people.

Independent variables
Reserve fund:
Corporate Tax Act of Korea (2013) allows the domestic hospitals to reserve 100% of interest and dividend revenues as well as 50% of for-profit revenues less donation spending. The purpose of this regulation is to motivate the hospitals to carry out their essential business activities and to donate to designated organizations. From the hospitals’ managerial point of view, the reserve fund is one of the suspected accounts not-for-hospital managers can use in earnings management. As one of the common fields where cookie jar reserves are made is in allowances (McKee, 2005) and the reserve fund accumulated by hospitals is considered to be a deductible expense, this provision in the Act might motivate the CEOs of not-for-profit hospitals to manage earnings positively to make a cookie jar reserve or to use this benefit as a good means of reducing taxation burdens. Lee and Swenson (2011) documented that managers manipulate earnings to reduce taxation burdens via discretionary expenditures that decrease earnings.

Donation revenue: Leaders of not-for-profit hospitals are not permitted to allot their accumulated income to their shareholders because they do not have residual claims, unlike for-profit organizations. Instead of beating the expectations of analysts, managers of not-for-profit hospitals are expected to have incentives to maintain or increase donation revenues with around zero earnings. Ramirez and Saraoglu (2011) documented that program spending (program services/total revenues) of organizations was the most dominant factor for donors to consider. Chetkovich and Frumkin (2003) and Dart (2004), however, noted leaders of not-for-profit organizations are under pressure to operate their organizations in a business manner. In the dichotomous evidence, which benchmarks guide not-for-profit hospital CEOs to manage earnings for maintaining or increasing their donation revenues is to be examined.

Government subsidies: According to Department of Health and Human Services (2013), not-for-profit hospitals, as well as medical care institutions, can receive central or local government subsidies for medical facility purchases, operating expenses and R and D expenditures for the promotion of public health. Not-for-profit hospitals are exempt from paying corporate taxes. Government aid and tax subsidies provided in the law might also motivate not-for-profit hospital managers to manipulate their earnings.

Ownership types: The essential objective of not-for-profit hospitals in South Korea is not to make profits but to advance the public good. Mordelet (2009) explained that public hospitals are constrained by government regulations from applying pertinent operation. Public hospitals are more interested to report less earnings to show more engagement in the promotion of the public interest.

Hospital sizes: Firm size affects corporate culture, organizational structure and internal control system (Biger and Hoang, 2008). Prior studies in for-profit sector explored that there are contrary results in the relationship between firm size and earnings management. Burgstahler and Dichev (1997) examined that large firms are more likely to manage their earnings than small ones do. Kim et al. (2003), however, documented that small firms involved more in earnings management than large-sized firms. But, large hospitals may try to show their promotion not for their own interests but for the public good to the public as they are classified as not-for-profit organizations by law in South Korea. Thus it is expected that large hospitals are more likely to manage their earnings negatively.

Sample selection: The population of the study included all not-for-profit hospitals in South Korea. According to the Health Insurance Review and Assessment Service (2011), there were 2,859 hospitals in South Korea on December 31, 2010. Because Korean regulations do not allow hospitals to provide medical services for profit, except for clinics owned by individual doctors, all hospitals in South Korea are regarded as not-for-profit organizations.

According to the EONFAK (2012), not-for-profit organizations consist of educational, scholarship, social welfare, medical care, cultural and other corporate bodies. Among the 2,859 not-for-profit hospitals, the 375 hospitals that exceeded 1 billion KRW of total asset value and consecutively reported financial statements to the Electronic Disclosure System of the National Tax Service (NTS, 2011) of Korea from 2007 to 2010 were selected. One billion KRW is equivalent to approximately US$865,000, when applying the annual average basic exchange rate in 2010, which was 1,156.46 KRW to US$1 (BOK, 2011). Those not-for-profit hospitals that exceeded 1 billion KRW of total assets have compulsorily disclosed their financial statements to the NTS by law since April 2009 for publicly monitoring their operating activities. The current study included the data on publicly available accounting information on 375 not-for-profit hospitals from the NTS. Data included the 1,500 year observations in 375 not-for-profit hospitals in South Korea from 2007 to 2010.

RESULTS

The purpose of the quantitative study employing logistic regression model of normal accruals was to examine earnings management in the not-for-profit hospital industry in South Korea.

Descriptive statistics
Descriptive statistics on financial positions, performance and financial ratios:
Table 1 shows the results of the descriptive statistics for all 375 hospitals in the sample. The mean (median) of the total assets was KRW 23,625 (KRW 9,523) million. The average exchange rate of KRW against US dollars was 1,156.46 for $1 in 2010 (BOK, 2011). The mean (median) of liabilities was KRW 14,722 (KRW 6,147). The mean of current assets was KRW 8,505 million. The measured values of the current assets in millions in the first (lowest), second, third and fourth (highest) quartiles were KRW 895, KRW 2,176, KRW 5,820 and KRW 683,653, respectively. The mean (median) of sales reported in millions was KRW 22,511 (KRW 8,131). Net income had a mean and median of KRW 157 and KRW 116. The mean (median) of ROA and Return on Equity (ROE) was 0.012 (0.013) and 0.216 (0.038), respectively. The mean of leverage (total liabilities/total assets) was 66.7% for the hospitals in the database.

The 25th percentile of total assets was 4,486, the 50th percentile was 9,523 and the 75th percentile was 20,665 in KRW million. For the liabilities, the 25th percentile was 2,604, the 50th percentile was 6,147 and the 75th percentile was 14,044 in KRW million. The volumes of total assets, current assets, PPE, liabilities, current liabilities and sales gradually increased for the period.

Table 2 shows the financial positions, management performance and financial ratios of the sample hospitals. The means of financial positions and performance in KRW million by fiscal year increased gradually but the means of financial ratios presented such as ROA, ROE and leverage did not have a fixed direction from 2007 to 2010.

Table 3 shows the descriptive statistics by ownership type. The mean of total assets for tax-paid public hospitals was over four times larger than for privately owned hospitals, which were KRW 76,449 million and KRW 17,318 million, respectively. The mean of PPE for government-owned hospitals was quintuple the mean of PPE for privately owned not-for-profit hospitals. The mean of sales, KRW 58,686 million, was for government-owned hospitals that were 3.46 times larger than the mean of sales for privately owned hospitals.

Table 1: Descriptive statistics for all hospitals in KRW million for 2007-2010
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
PPE: Property, plant and equipment, ROA: Net income/total assets, ROE: Net income/total net assets, LEV: Total liabilities/total assets, 1/4 Q: Cuts off lowest 25% of data, 3/4 Q: Cuts off highest 25% of data

Table 2: Financial position, management performance and financial ratio by fiscal year
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
PPE: Property, plant and equipment, ROA: Net income/total assets, ROE: Net income/total net assets, LEV: Total liabilities/total assets

Table 3: Descriptive statistics by ownership type
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
PPE: Property, plant and equipment, ROA: Net income/total assets, ROE: Net income/total net assets, LEV: Total liabilities/total assets

The mean of ROA (net income/total assets) and ROE (net income/owner’s equity) in governmental hospitals was negative, which meant net loss occurred. The means of leverage (total liabilities/total assets) were 57.9% for public hospitals and 66.8% for private hospitals.

Table 4 shows descriptive statistics by hospital location. South Korea has seven metropolitan cities that have populations of more than 1 million. The means of total assets and PPE in millions for metropolitan-based hospitals were KRW 34,259 and KRW 14,624, which were 1.89 and 1.47 times larger than the ones for urban hospitals. The mean of sales for the metropolitan hospitals was KRW 33,252, which was 2.18 times the mean of the district hospitals. Figure 1 shows the comparison in financial positions and performances between metropolitan hospitals and urban hospitals.

Table 5 reports the descriptive statistics on variables in the regression model. The mean of DACC is-0.107, which is negative, while the mean of TAC is 0.048, which is positive.

Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
Fig. 1: The comparison in financial positions and performance by location

Table 4: Descriptive statistics by hospital location
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
ROA: Net income/total assets, ROE: Net income/total net assets, LEV: Total liabilities/total assets

Table 5: Descriptive statistics on variables in the regression model
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
DACC: Performance-matched discretionary accruals deflated by assets Kothari et al. (2005), TAC: Total accruals, 1/4 Q: Cuts off lowest 25% of data, 3/4 Q: Cuts off highest 25% of data

Correlation analysis among major variables
Pearson’s correlation analysis on variables in the regression:
Table 6 shows the results of Pearson’s correlation among variables in the regression model. The coefficient between DACC (performance-matched discretionary accruals) and Reserve_Fund was positive and significant (p<0.1). This result showed that managers of not-for-profit hospitals in South Korea use reserve funds to create a cookie jar reserve, supporting hypothesis H1.

The coefficient between DACC and Donation_Revenue was negative and significant at a 5% level, indicating that hospitals are likely to decrease earnings to increase donation revenue or to prevent donation departure. This result supported the donor’s departure prevention hypothesis H2. The coefficient on Government_Subsidies was negative and significant (p <0.1), indicating that hospitals decrease earnings for central or local governmental subsidies. This result supported hypothesis H3.

A negative relationship exists between DACC and Owner_type at a 1% level, providing evidence that leaders of government-owned hospitals are more likely than leaders of privately owned hospitals to decrease earnings, supporting hypothesis H4. As hospital location and ownership type are significant variables in the computation of medical insurance fees in South Korea, additional analysis was conducted based on the variables in the model by Location and Owner_type. The coefficient on Size was negative and significant (p<0.01), indicating that larger hospitals are to be restrained on earnings management and providing support for the hypothesis H5.

The coefficient between DACC and TAC that was a control variable was positive and significant (p<0.01), indicating TAC controls other variables efficiently. Positive and significant (p<0.05) coefficient on Leverage indicated that managers of not-for-profit hospitals have an incentive to manage earnings to avoid debt covenant violation. The coefficients between DACC and Location were positive and significant (p<0.05), indicating that hospitals in metropolitan cities with more than 1 million people are likely to increase earnings.

Table 6: Peason’s correlation among major variables in the regression
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
The p-values are in parentheses. DACC: Performance-matched discretionary accruals, estimated by Kothari et al. (2005), Reserve_Fund: 1 if the account of the reserve fund was more than 0, 0 otherwise, Donation_Revenue: 1 if the account of donation revenue was more than 0, 0 otherwise, Government_Subsidies: 1 if the account of government revenue was more than 0, 0 otherwise, Owner_Type: 1 if the hospital was owned by central or local governments, 0 otherwise, SIZE: Natural logarithm of total assets. TAC: Total accruals at thebeginning of the year; Leverage: Total liabilities divided total asset, Location: 1 if the hospital was located in seven metropolitan cities, 0 otherwise, ***p<0.01, two-tailed. **p<0.05, two-tailed. *p<0.10, two-tailed

Table 7: Metropolitan hospitals versus district hospitals
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
DACC: Performance-matched discretionary accruals, estimated by Kothari et al. (2005), TAC: Total accruals at the beginning of the year, See other variable definitions in Table 6, ***p<0.01, two-tailed. **p<0.05, two-tailed. *p<0.10, two-tailed

Due to the possibility of multicollinearity in the model, VIF values were also measured. All examined VIF values were less than 10, which were 1.0765 (Reserve_Fund), 1.3975 (Donation_Revenue), 1.4181 (Government_Subsidies), 1.1580 (TAC), 1.1831 (Leverage), 1.1836 (Location), 1.4855 (Owner_Type) and 1.4439 (Size). These VIF measures were also acceptable in the multiple regression models.

Univariate analysis
Student t test by location and ownership type in the model:
Table 7 contains the means and t values on major variables in the regression equation by location to test significant differences between hospitals in metropolitan areas and hospitals in urban areas. The coefficient on DACC was positive and significant at a 5% level, providing evidence that hospitals in metropolitan area manage larger earnings than district hospitals do. Further, the coefficients on Reserve_Fund and Donation_Revenue are both positive and significant at 1 and 10%, respectively. It indicates that hospitals in metropolitan areas recognize larger reserve funds as an expense and receive larger donation revenues than hospitals in urban areas do. The coefficient on Government_Subsidies was negative and significant at a 1% level, demonstrating that hospitals in urban areas receive larger governmental subsidies than hospitals in metropolitan cities do. The result indicates the hospitals in urban areas, where people need medical service but the hospitals have no profitable medical businesses opportunities, receive larger governmental subsidies. The coefficient on Size was positive and significant at a 1% level, indicating that metropolitan hospitals manipulate more earnings than district hospitals do.

Table 8 shows the means and t values on major variables in the equation by ownership type to test significant differences between two groups. The coefficient on DACC was negative and significant at a 1% level. This result indicated that private hospitals manage larger earnings than governmental hospitals that are under government supervision do. The coefficient on Reserve_Fund was negative and significant at a 5% level, demonstrating that private hospitals recognize a larger reserve fund than governmental hospitals do. The correlation coefficients on Donation_Revenue and Government_Subsidies were both positive and significant at 1 and 10%, respectively. These results indicated that government hospitals earn higher donation revenue and receive larger government subsidies than private hospitals do, which is consistent with the belief of donors and governments that government hospitals make greater contributions to the public interests.

Table 8: Public hospitals versus private hospitals
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
DACC: Performance-matched discretionary accruals, estimated by Kothari et al. (2005), Reserve_Fund: 1 if the account of the reserve fund was more than 0, 0 otherwise, Donation_Revenue: 1 if the account of donation revenue was more than 0, 0 otherwise, Government_Subsidies: 1 if the account of government revenue was more than 0, 0 otherwise, SIZE: Natural logarithm of total assets, TAC: Total accruals at the beginning of the year, Leverage: Total liabilities divided total asset, Location: 1 if the hospital was located in seven metropolitan cities, 0 otherwise, ***p<0.01, two-tailed. **p<0.05, two-tailed. *p<0.10, two-tailed

Table 9: Logistic regression of net income on independent variables
Image for - Motivational Factors Affecting Earnings Management of Not-for-profit Hospitals 
  in South Korea
The t value is in parentheses. Reserve_Fund: 1 if the account of the reserve fund was more than 0, 0 otherwise, Donation_Revenue: 1 if the account of donation revenue was more than 0, 0 otherwise, Government_Subsidies: 1 if the account of government revenue was more than 0, 0 otherwise, Owner_Type: 1 if the hospital was owned by central or local governments, 0 otherwise, SIZE: Natural logarithm of total assets, TAC: Total accruals at the beginning of the year, Leverage: Total liabilities divided total asset, Location: 1 if the hospital was located in seven metropolitan cities, 0 otherwise ΣYear: year dummy variable, F-value: F-test statistic, Adj. R2(%): Adjusted R squared. ***p<0.01, two-tailed. **p<0.05, two-tailed. *p<0.10, two-tailed

Multivariate regression analysis
Tests of hypotheses H1 to H5:
Table 9 shows results of estimating the model mentioned above. The first column in the table contains all independent variables in the equation and the second, third and fourth models eliminated Reserve_Fund, Donation_Revenue and Government_Subsidies, respectively, in the regression analysis to find certain evidence. F values are significant at a 1% level and VIF is below about 1.4 on the four models, indicating that the presented multivariate regression models have no problem with model fitness and multicollinearity.

Test of hypothesis H1: The coefficient on Reserve_Fund is positive and significant at the 5 and 10% level. It indicates that managers of not-for-profit hospitals in South Korea manage earnings to make a cookie jar reserve by using reserve fund.

Test of hypothesis H2: The coefficients on Donation_Revenue were negative and significant at a 1 and a 5% level. Leone and Van Horn (2005) noted that the opportunity of increasing donation revenue would be lost as profits increase. These results provided evidence that managers of not-for-profit hospitals in South Korea manage earnings negatively to increase donation revenue or to prevent donation departure, which supports donor’s departure prevention hypothesis H2.

Test of hypothesis H3: The coefficients on Government_Subsidies were negative and highly significant at a 1% level, providing evidence that not-for-profit hospitals decrease earnings for governmental subsidy increases. Thus, hypothesis H3 by government subsidies was supported.

Test of hypothesis H4: The coefficients on Owner_Type were negative and significant at a 5 and 10% levels, providing evidence that public hospitals are motivated to manipulate their earnings negatively for better reputation in the promotion of public interests. Thus, the hypothesis H4 was also supported.

Test of hypothesis H5: The coefficients on Size were negative and strongly significant at 1 and 5% levels. This result proves that earnings management is negatively associated with hospital size as extant literature examined in the for-profit sector, which supports hypothesis H5.

Descriptive statistics, correlation analysis, student t tests and multivariate regression analysis were conducted to find the evidence of earnings management in not-for-profit hospitals in South Korea. These statistical analyses were performed with the sample of 375 hospitals (1,500 hospital year observations) for the period of 2007 to 2010 in South Korea.

DISCUSSION AND CONCLUSION

This quantitative study documented that not-for-profit hospitals in South Korea also conduct opportunistic earnings management as for-profit organizations do. Not-for-profit hospitals in South Korea report more earnings to create a cookie jar reserve by using reserve fund for future management opportunities. The hospitals also manage their earnings negatively to increase donation revenue and to receive more government subsidies. Furthermore, this study found that public (large) hospitals engage in less earnings management than small sized hospitals to avoid reputation (political) costs. The main analytical aim in the study was to test whether leaders of not-for-profit hospitals in South Korea manage earnings with certain directions.

There are five key findings from this research. First, the relationship between discretionary accruals and reserve fund was positive and significant. This result documented that CEOs of not-for-profit hospitals manage earnings positively to create a cookie jar reserve for the future use. Leaders of not-for-profit hospitals can have managerial opportunities to record less expenses in the future by recognizing more expenditure in the current accounting period. Second, the relationship between discretionary accruals and donation revenue was negative and significant. The opportunity of donation revenue is lost when reported earnings increase (Leone and Van Horn, 2005). This result demonstrated that CEOs of not-for-profit hospitals in South Korea manage earnings negatively to increase donation revenue or to prevent donation departure. Third, the relationship between discretionary accruals and governmental subsidies was negative and highly significant. Department of Health and Human Services (2012) allows not-for-profit hospitals that have no economic opportunities in the operation of public health services to receive a subsidy or tax support from the local or central governments. The result of the test indicated that managers of not-for-profit hospitals in South Korea report fewer earnings to increase governmental subsidy. Fourth, the relationship between discretionary accruals and ownership types was negative and significant. Public hospitals are more motivated than privately owned hospitals to protect better reputations in the performance of the public interest by managing their earrings negatively. Kim et al. (2003) mentioned large organizations consider the reputation costs when involving in earnings management. Fifth, the coefficients on size were negative and highly significant. This provides evidence that large not-for-profit hospitals in South Korea are likely to manage their earnings negatively than small hospitals. Large for-profit firms, however, are more actively engaged in earnings management than small firms to meet the expectations of capital market (Kim et al., 2003).

Consistent with prior research, the results of the current study indicated that not-for-profit hospitals in South Korea manage earnings with different incentives. The current study also revealed that leaders of not-for-profit hospitals use non-operating expenditures and revenue increasing accounts to achieve earnings objectives. The results have three significant implications. The managers of not-for-profit hospitals have the incentives to create a cookie jar reserve, to increase donation revenue and to receive government subsidies by engaging in earnings management activities. The first implication was that stakeholders will pay for the value-fluctuating activities of not-for-profit hospitals. The stakeholders include patients, donors, taxpayers, governments and community members. Governments try to meet government objectives with the sources of tax revenue (Eldenburg et al., 2007). The value-decreasing activities of not-for-profit hospitals relate to tax exemption on hospital income, tax deduction for hospital donors and government subsidy spending. Thus, the second implication was that regulatory or policy setters will be interested in the value-decreasing activities of not-for-profit hospitals, especially on the launching of for-profit hospitals in South Korea. Leaders of not-for-profit hospitals prepare and report accounting information for the benefit of people outside. Financial reporting practices mitigate information asymmetry between managers inside and stakeholders outside. Jiraporn and Gleason (2007) provided evidence that earnings management is an agency cost because information users make poor decisions based on the reported accounting numbers. The shareholders of the hospitals have a claim for dividend when launching of for-profit hospitals in South Korea in future. So there will be great likelihood for leaders of the hospitals to follow the same patterns of for-profit organizations in managerial behaviors. Thus, the third implication was that managed earnings will lead the stakeholders of hospitals to bad decisions.

REFERENCES
1:  Athanasakou, V.E., N.C. Strong and M. Walker, 2009. Earnings management or forecast guidance to meet analyst expectations? Account. Bus. Res., 39: 3-35.
CrossRef  |  

2:  Ayers, B.C., J. Jiang and P.E. Yeung, 2006. Discretionary accruals and earnings management: An analysis of pseudo earnings targets. Account. Rev., 81: 617-652.
CrossRef  |  

3:  Bae, J.S., 2010. Health minister-nominee triggers concerns over for-profit hospitals. The Korea Times, August 9, 2010.

4:  BOK, 2011. Exchange rate from Korean won to US dollars. Bank of Korea, Korea.

5:  Barth, M.E., J.A. Elliott and M.W. Finn, 1999. Market rewards associated with patterns of increasing earnings. J. Account. Res., 37: 387-413.
Direct Link  |  

6:  Bartov, E., F.A. Gul and J.S.L. Tsui, 2000. Discretionary-accruals models and audit qualifications. J. Account. Econ., 30: 421-452.
CrossRef  |  Direct Link  |  

7:  Bartov, E. and P. Mohanram, 2004. Private information, earnings manipulations and executive stock-option exercises. Account. Rev., 79: 889-920.
CrossRef  |  

8:  Beaver, W.H. and E.E. Engel, 1996. Discretionary behavior with respect to allowances for loan losses and the behavior of security prices. J. Accounting Econ., 22: 177-206.
CrossRef  |  

9:  Beaver, W.H. and M.F. McNichols, 1998. The characteristics and valuation of loss reserves of property casualty insurers. Rev. Accounting Stud., 3: 73-95.
CrossRef  |  

10:  Beneish, M.D., 1999. The detection of earnings manipulation. Financial Anal. J., 55: 24-36.
Direct Link  |  

11:  Berger, P.G., 1993. Explicit and implicit tax effects of the R and D tax credit. J. Account. Res., 31: 131-171.
Direct Link  |  

12:  Bergstresser, D. and T. Philippon, 2006. CEO incentives and earnings management. J. Finan. Econ., 80: 511-529.
CrossRef  |  

13:  Beyer, A., 2009. Capital market prices, management forecasts and earnings management. Account. Rev., 84: 1713-1747.
CrossRef  |  

14:  Biger, N. and Q. Hoang, 2008. Managerial ownership, firm size and earnings management. Int. J. Finance, 20: 4629-4654.

15:  Brickley, J.A. and R.L. Van Horn, 2002. Managerial incentives in nonprofit organizations: Evidence from hospitals. J. Law Econ., 45: 227-249.
Direct Link  |  

16:  Brown, L.D. and H.N. Higgins, 2005. Managers' forecast guidance of analysts: International evidence. J. Accounting Public Policy, 24: 280-299.
CrossRef  |  

17:  Burgstahler, D. and I. Dichev, 1997. Earnings management to avoid earnings decreases and losses. J. Account. Econ., 24: 99-126.
CrossRef  |  Direct Link  |  

18:  Burgstahler, D.C. and M.J. Eames, 2006. Management of earnings and analysts' forecasts to achieve zero and small positive earnings surprises. J. Bus. Finance Accounting, 33: 633-652.
CrossRef  |  

19:  Bushman, R.M. and A.J. Smith, 2001. Financial accounting information and corporate governance. J. Account. Econ., 32: 237-333.
CrossRef  |  

20:  Chen, H.F., G.J. Bazzoli and H.M. Hsieh, 2009. Hospital financial conditions and the provision of unprofitable services. Atlantic Econ. J., 37: 259-277.
CrossRef  |  

21:  Chen, M.C. and Y.C. Tsai, 2010. Earnings management types and motivation: A study in Taiwan. Soc. Behav. Personality Int. J., 38: 955-962.
CrossRef  |  

22:  Chetkovich, C. and P. Frumkin, 2003. Balancing margin and mission nonprofit competition in charitable versus fee-based programs. Administrative Soc., 35: 564-596.
CrossRef  |  

23:  Core, J.E., W.R. Guay and D.F. Larcker, 2003. Executive equity compensation and incentives: A survey. Econ. Policy Rev., 9: 27-50.
Direct Link  |  

24:  Cormier, D., M. Magnan and B. Morard, 2000. The contractual and value relevance of reported earnings in a dividend-focused environment. Eur. Account. Rev., 9: 387-417.
CrossRef  |  

25:  Corporate Tax Act of Korea, 2013. A unique business purpose of the inclusion of loss reserves: Aticle 29. Ministry of Strategy and Finance, Korea.

26:  Dart, R., 2004. Being business-like in a nonprofit organization: A grounded and inductive typology. Nonprofit Voluntary Sector Q., 33: 290-310.
CrossRef  |  

27:  Das, S., P.K. Shroff and H. Zhang, 2009. Quarterly earnings patterns and earnings management. Contemp. Account. Res., 26: 797-831.
CrossRef  |  

28:  DeAngelo, L.E., 1986. Accounting numbers as market valuation substitutes: A study of management buyouts of public stockholders. Account.Rev., 61: 400-420.
Direct Link  |  

29:  Dechow, P.M. and I.D. Dichev, 2002. The quality of accruals and earnings: The role of accrual estimation errors. Account. Rev., 77: 35-59.
CrossRef  |  Direct Link  |  

30:  Dechow, P.M., S.A. Richardson and A.I. Tuna, 2003. Why are earnings kinky? An examination of the earnings management explanation. Rev. Account. Stud., 8: 355-384.
Direct Link  |  

31:  Dechow, P.M. and D.J. Skinner, 2000. Earnings management: Reconciling the views of accounting academics, practitioners and regulators. Account. Horizons, 14: 235-250.
CrossRef  |  Direct Link  |  

32:  Dechow, P.M., R.G. Sloan and A.P. Sweeney, 1995. Detecting earnings management. Accounting Rev., 70: 193-225.
Direct Link  |  

33:  Degeorge, F., J. Patel and R. Zeckhauser, 1999. Earnings management to exceed thresholds. J. Bus., 72: 1-33.
CrossRef  |  Direct Link  |  

34:  Dong-A Ilbo, 2010. Allow for-profit hospitals. The DONG-A ILBO, November 6, 2010. http://english.donga.com/srv/service.php3?biid=2010110639828.

35:  Eldenburg, L., K. Gunny, K. Hee and N. Soderstrom, 2007. Earnings management through real activities manipulation: Evidence from nonprofit hospital. August 24, 2007. http://www3.nd.edu/~carecob/Workshops/07-08%20Workshops/SoderstromPaper.pdf.

36:  EONFAK, 2012. Establishment and operation of a public service corporation act on: Trial 03.23.2013, [Presidential Decree No. 24415, 03.23.2013, tabeop revision]. Establishment and Operation of Nonprofit Foundations Act of Korea, Department of Justice, Korea.

37:  Fama, E.F. and M.C. Jensen, 1983. Separation of ownership and control. J. Law Econ., 26: 301-325.
Direct Link  |  

38:  Feng, M. and S. McVay, 2010. Analysts' incentives to overweight management guidance when revising their short-term earnings forecasts. Accounting Rev., 85: 1617-1646.
Direct Link  |  

39:  Fields, T.D., T.Z. Lys and L. Vincent, 2001. Empirical research on accounting choice. J. Accounting Econ., 31: 255-307.
CrossRef  |  

40:  Frank, R.G., D.S. Salkever and J. Mitchell, 1990. Market forces and the public good: Competition among hospitals and provision of indigent care. Adv. Health Econ. Health Serv. Res., 11: 159-183.
PubMed  |  Direct Link  |  

41:  Freeman, R.N. and S.Y. Tse, 1992. A nonlinear model of security price responses to unexpected earnings. J. Accounting Res., 30: 185-209.
Direct Link  |  

42:  Graham, J.R., C.R. Harvey and S. Rajgopal, 2005. The economic implications of corporate financial reporting. J. Account. Econ., 40: 3-73.
CrossRef  |  Direct Link  |  

43:  Gunny, K.A., 2010. The relation between earnings management using real activities manipulation and future performance: Evidence from meeting earnings benchmarks. Contemporary Account. Res., 27: 855-888.
CrossRef  |  

44:  Habib, A. and J.C. Hansen, 2008. Target shooting: Review of earnings management around earnings benchmarks. J. Account. Literat., 27: 25-70.
Direct Link  |  

45:  Hansen, J.C., 2010. The effect of alternative goals on earnings management studies: An earnings benchmark examination. J. Account. Public Policy, 29: 459-480.
Direct Link  |  

46:  Healy, P.M., 1985. The effect of bonus schemes on accounting decisions. J. Accounting Econ., 7: 85-107.
Direct Link  |  

47:  Healy, P.M. and J.M. Wahlen, 1999. A review of the earnings management literature and its implications for standard setting. Account. Horizons, 13: 365-383.
CrossRef  |  Direct Link  |  

48:  Hettihewa, S. and C. Wright, 2010. A review of dominant and emerging issues in corporate earnings management. Southern Bus. Rev., 35: 15-36.
Direct Link  |  

49:  Hoerger, T.J., 1991. Profit variability in for-profit and not-for-profit hospitals. J. Health Econ., 10: 259-289.
CrossRef  |  

50:  Hribar, P. and D.W. Collins, 2002. Errors in estimating accruals: Implications for empirical research. J. Account. Res., 40: 105-134.
CrossRef  |  

51:  Imhoff, E.A., 1977. Income smoothing: A case for doubt. The Account. J., 1977: 85-100.

52:  Jensen, M.C. and W.H. Meckling, 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. J. Financial Econ., 3: 305-360.
CrossRef  |  Direct Link  |  

53:  Jiang, J., 2008. Beating earnings benchmarks and the cost of debt. Accounting Rev., 83: 377-416.
CrossRef  |  

54:  Jiraporn, P. and K.C. Gleason, 2007. Delaware incorporation and earnings management: An empirical analysis. J. Applied Finan., 17: 40-51.

55:  Jones, J.J., 1991. Earnings management during import relief investigations. J. Account. Res., 29: 193-228.
Direct Link  |  

56:  Kang, S.H. and K. Sivaramakrishnan, 1995. Issues in testing earnings management and an instrumental variable approach. J. Account. Res., 33: 353-367.
CrossRef  |  Direct Link  |  

57:  Kasznik, R. and M.F. McNichols, 2002. Does meeting earnings expectations matter? Evidence from analyst forecast revisions and share prices. J. Account. Res., 40: 727-759.
CrossRef  |  

58:  Kim, Y., C. Liu and S.G. Rhee, 2003. The effect of firm size on earnings management. University of Hawaii, January, 2003. http://www2.hawaii.edu/~fima/PDF/Finance_Seminar/EarningsMgmt.pdf.

59:  Kinney, W., D. Burgstahler and R. Martin, 2002. Earnings surprise "materiality" as measured by stock returns. J. Account. Res., 40: 1297-1329.
CrossRef  |  

60:  Kothari, S.P., 2001. Capital markets research in accounting. J. Account. Econ., 31: 105-231.
CrossRef  |  Direct Link  |  

61:  Kothari, S.P., A.J. Leone and C.E. Wasley, 2005. Performance matched discretionary accrual measures. J. Account. Econ., 39: 163-197.
CrossRef  |  Direct Link  |  

62:  Kramer, J. and R.E. Santerre, 2010. Not-for-profit hospital CEO performance and pay: Some evidence from Connecticut. Inquiry, 47: 242-251.
CrossRef  |  

63:  Kwon, S.S. and Q.J. Yin, 2006. Executive compensation, investment opportunities and earnings management: High-tech firms versus low-tech firms. J. Account. Audit. Finance, 21: 119-148.

64:  Larcker, D.F. and S.A. Richardson, 2004. Fees paid to audit firms, accrual choices and corporate governance. J. Accounting Res., 42: 625-658.
CrossRef  |  

65:  Laux, C. and V. Laux, 2009. Board committees, CEO compensation and earnings management. Account. Rev., 84: 869-891.
CrossRef  |  

66:  Lee, N. and C. Swenson, 2011. Earnings management through discretionary expenditures in the US, Canada and Asia. Int. Bus. Res., 4: 257-266.
Direct Link  |  

67:  Leone, A.J. and R.L. Van Horn, 2005. How do nonprofit hospitals manage earnings? J. Health Econ., 24: 815-837.
CrossRef  |  

68:  Matsumoto, D.A., 2002. Management's incentives to avoid negative earnings surprises. Account. Rev., 77: 483-514.
Direct Link  |  

69:  McCue, M.J. and P. Nayar, 2009. A financial ratio analysis of for-profit and non-profit rural referral centers. J. Rural Health, 25: 314-319.
CrossRef  |  

70:  McKee, T.E., 2005. Earnings Management: An Executive Perspective. Thomson, Mason, USA., ISBN-13: 9780324223255, Pages: 205.

71:  McNichols, M.F., 2000. Research design issued in earnings management studies. J. Account. Public Policy, 19: 313-345.
CrossRef  |  

72:  McNichols, M.F., 2002. Discussion of the quality of accruals and earnings: The role of accrual estimation errors. Account. Rev., 77: 61-69.
CrossRef  |  

73:  McNichols, M. and G.P. Wilson, 1988. Evidence of earnings management from the provision for bad debts. J. Account. Res., 26: 1-31.
Direct Link  |  

74:  McVay, S.E., 2006. Earnings management using classification shifting: An examination of core earnings and special items. Account. Rev., 81: 501-531.
CrossRef  |  

75:  McVay, S., V. Nagar and V.W. Tang, 2006. Trading incentives to meet the analyst forecast. Rev. Account. Stud., 11: 575-598.
CrossRef  |  

76:  Department of Health and Human Services, 2013. Medical services act of Korea, 2013. Chapter 1, Article 3 and Chapter 8, Article 83. http://www.law.go.kr/lsSc.do?menuId=0&p1=&subMenu=1&nwYn=1&query=%EC%9D%98%EB%A3%8C%EB%B2%95&x=0&y=0#liBgcolor6

77:  Mordelet, P., 2009. The impact of globalisation on hospital management: Corporate governance rules in both public and private nonprofit hospitals. J. Manage. Market. Healthcare, 2: 7-14.

78:  Moyer, S.E., 1990. Capital adequacy ratio regulations and accounting choices in commercial banks. J. Account. Econ., 13: 123-154.
CrossRef  |  

79:  NTS, 2011. Electronic disclosure system. National Tax Service, Republic of Korea. http://npoinfo.hometax.go.kr/distribution/wepa8801.jsp.

80:  Nelson, K.K., 2000. Rate regulation, competition and loss reserve discounting by property-casualty insurers. Account. Rev., 75: 115-138.
Direct Link  |  

81:  Petroni, K.R., 1992. Optimistic reporting in the property-casualty insurance industry. J. Account. Econ., 15: 485-508.
CrossRef  |  

82:  Petroni, K.R., S.G. Ryan and J.M. Wahlen, 2000. Discretionary and non-discretionary revisions of loss reserves by property-casualty insurers: Differential implications for future profitability, risk and market value. Rev. Account. Stud., 5: 95-125.
CrossRef  |  

83:  Phillips, J., M. Pincus and S.O. Rego, 2003. Earnings management: New evidence based on deferred tax expense. Account. Rev., 78: 491-521.
Direct Link  |  

84:  Ramirez, A. and H. Saraoglu, 2011. An analytic approach to selecting a nonprofit for donation. J. Finan. Plann., 24: 59-67.
Direct Link  |  

85:  Roychowdhury, S., 2006. Earnings management through real activities manipulation. J. Account. Econ., 42: 335-370.
CrossRef  |  Direct Link  |  

86:  Safeek, Y.M., 2008. The expanding roles of the chief medical officer. Phys. Exec., 34: 18-22.
PubMed  |  Direct Link  |  

87:  SKDI and SKHIDI, 2009. A study on the introduction of for-profit hospitals in Korea. South Korea Development Institute and South Korea Health Industry Development Institute, Korea.

88:  Department of Health and Human Services, 2012. Special act on designation and management of free economic zones 2012. Chapter, 5, Article 23. http://www.law.go.kr/lsSc.do?menuId=0&p1=&subMenu=1&nwYn=1&query=%EC%9D%98%EB%A3%8C%EB%B2%95&x=0&y=0#liBgcolor6.

89:  Sun, L. and S. Rath, 2010. Earnings management research: A review of contemporary research method. Global Rev. Account. Finance, 1: 121-135.
Direct Link  |  

90:  Weber, M., 2006. Sensitivity of executive wealth to stock price, corporate governance and earnings management. Rev. Account. Finance, 5: 321-354.
CrossRef  |  

91:  Wilson, G.P., 1996. Discussion write-offs: Manipulation or impairment? J. Account. Res., 34: 171-177.
Direct Link  |  

92:  Woo, S., 2008. For-profit hospitals bring pain to the people. Wolgan Mal, 263: 74-81.

93:  Yang, C.C., Y.S. Wang, S.T. Chang, S.E. Guo and M.F. Huang, 2009. A study on the leadership behavior, safety culture and safety performance of the healthcare industry. World Acad. Sci. Eng. Technol., 53: 1148-1155.
Direct Link  |  

94:  Health Insurance Review and Assessment Service, 2011. Press release February 17, 2011. Health Insurance Review and Assessment Service, Seoul, Korea. http://www.hira.or.kr/common/dummyjsp?pgmid=HIRAA060200000000.

©  2021 Science Alert. All Rights Reserved