THE EFFECT OF EARNING PERSISTENCE TOWARD FINANCIAL PERFORMANCE OF LISTED BANKS IN INDONESIA STOCK EXCHANGE

Getting information about earnings is important for the parties who are interested in investment in a company. The purpose of this study is to examine the earnings persistence effect toward performance of Indonesian bank listed in Bursa Efek Indonesia (BEI), with company’s size as control variable. This study use a sample of Indonesian bank listed in BEI from 2007 to 2010. Bank’s financial performance is measured using seven indicators, namely Cash and Bank to Total Deposits (CBTD), Loan to Total Deposits (LTTD), Equity to Total Asset (ETTA), Operating Profit Margin (OPM), Net Profit Margin (NPM), Return on Equity (ROE), and Return on Investment (ROI) as dependent variables. Earnings persistence as the independent variable is the regression coefficient of Earning per Share at one year before t (EPSt-1) when it regressed toward EPSt. Bank size is used as control variable. The result of regression model indicates that earnings persistence influence the performance of Indonesian bank listed in BEI. Of the seven measures of bank performance, only LTTD and ROE are not influenced by earnings persistence.


INTRODUCTION
The growing business world today indicates the presence of competition among the firms.Such competition led the company's management to provide confidence and show good performance in their activity.In general, the performance of a company is measured by its profitability (Penman in Sunarto 2008).The company that has a good net income (profit) will attract investors to invest in its stock while that with a low profit will lead investors to withdraw their investments.Thus, earnings quality reflects the earnings that continue in the future (sustainable earnings) which is determined by the components of accruals and cash flows.
Financial reports are used by management to determine the conditions the company facing the company at that time.In relation to this case, financial accounting standards (IAI 2010) No.1 states that the objective of financial statements is to provide information about the financial position, performance, and cash flows of the company for the benefit of the majority of the users so that they can make economic decisions.This can also show top management accountability in using resources entrusted to them.
Earnings persistence is a measure of profit based on the views that the more sustainable the earnings, the better their quality.For example, Penman, in Sunarto (2008) classifies it into two groups, namely sustainable earnings and unusual earnings.Earnings are said to be persistent when they are good indicators for the future earnings generated by the company repeatedly in the long term (sustainable), while the unusual earnings are those that are generated temporarily and not repeatedly so that they could not be good indicators of the future earnings (Sunarto 2008).Sloan in Sutopo (2005) used regression of the regression between accounting profit of current period and future periods as a proxy for accounting earnings persistence.In this case, accounting profit is considered to be persistence, when the coefficient of variation is smaller.Sloan also states that the accrual component of current earnings are less likely to recur or less persistence to determine future earnings as they are accrual and deferred basis, as well as based on allocations and assessments that are subjective.Some analysts prefer to use operating cash flows as the determinant of the quality of the earnings because the cash flow is considered more persistent than the accrual component.They believe that the higher the ratio of operating cash flow to net income, the higher the quality of the earnings.
This study focuses only on earnings persistence as the measure of earnings quality.No indication of earnings management means that the quality of earnings is good.Earnings persistence is measured based on the ability of current earnings to predict future earnings.So, the current earnings are used to predict the future earnings by looking at the real bank's financial statements.

THEORETICAL FRAMEWORK AND HYPOTHESIS Definition of Earnings
Profit (earnings) is the element of financial statement that is the most concern by the users because its amount can provide rich information regarding a company's overall performance.In addition, profit is the increase in capital (net assets) from side transactions or irregular transactions of a business entity during a certain period, except those arising from revenue or the owner's investment.
Beside that, profit is used as a measure of a company's performance or as the basis assessment of other measures such as stock per share.It consists of revenues and expenses.By grouping items into revenues and expenses, the company can get the results of different measures of income, namely gross profit, operating profit, profit before tax and net profit.Profit in the financial statements is an indicator that can be used to measure the operating performance of a company.Both lenders and investors use profit to evaluate the performance of management, to predict earnings power and to predict the future earnings.

Earnings Quality
Earnings in the financial statements are used by management to attract potential investors and borrowers so that these are managed in such a way by management to influence the final decision of such parties.This is consistent with the signaling theory showing the tendency of information between management and the outsides.Company's internal parties in general have more information about the real condition of the company's current and future prospects than that of external parties.Therefore, the quality of accounting earnings reported by management will be the main concern of outsiders.Quality accounting earnings is accounting earnings with little or no impaired perception and can reflect the actual financial performance of companies (Wijayanti 2006).
Hayn (1995) explains that impaired perception in accounting earnings caused by transitory events or application of accrual accounting concept.Transitory events are events that happen at certain times and only affect the period of the incident.In this case, high-quality earnings are earnings that have relatively low variability or smooth earnings.In accounting research literature, there many definitions of earnings quality from the view points of togetherness and decision-making.Schipper and Vincent (2003) in Margani (2009) classify quality constructs of earnings, based on the time series properties of earnings and the qualitative characteristics of the conceptual framework.The earnings relationships can be summarized as follows: 1.Based on the time series properties of earnings, the qualitative characteristics of the conceptual framework, the relationship of accrual earnings and implementation decisions.The persistence of earnings as earnings quality is determined based on the perspective of its usefulness in decision making especially in equity valuation, for example, to predict future earnings.2. Earnings quality based on cash earnings accrual relationships that can be measured with a variety of measurements, namely: the ratio of operating cash to earnings, changes in total accruals, estimated abnormal / discretionary accruals (abnormal accruals / policy), and the estimation of accruals -cash relation.By using a measure of operating cash to earnings ratio, earnings quality is indicated by the closeness of income to operating cash flow.3. Earnings quality could be based on qualitative concept of conceptual framework.Quality earnings are earnings that are useful in making decision, that are earnings that have the characteristics of relevance, reliability and comparability or consistency.4. Earnings quality based on decision implementation includes two approaches.

Agency Theory
Agency theory is based on the theory of the entity.A company is seen as an entity separated from its owners and creditors.It means that there is separation between the management and the owner of company.Management to obtain the trust of the owner of the company to manage the company and management will be compensated for it.Agency theory also says that the actions taken are likely to maximize the value of the company if it is in the interests of the manager.Conflict of interest between managers and owners arise when they try to maximize their own interests.Anthony and Govindarajan (in Diastity 2010) explains the agency relationship exists when one party (the principal) hire another person (agent) to perform a service.In doing so, the principal delegates the authority to make decisions to the agent.In agency theory, the principal is shareholder and the agent is management of the company.
Agent is assumed to get satisfaction not only from financial compensation but also from visible agency relationship, while the principal is assumed only interested in financial returns derived from their investment in the company.Basically agency theory is a model used to formulate the problems between management (agent) and owner (principal) (Sunarto 2008).

Concept of Earnings Persistence
Earnings persistence is a measure of income based on the view that a sustainable profit is profit that has a better quality.The definition of earnings persistence according to (Sunarto 2008, Fanani 2010, Djamaludin et al. 2010, and Margani 2009) distinguishes income into to groups, namely sustainable earnings and earnings unusual.The persistent earnings are earnings that have ability as a good indicator of the future earnings generated by the company repeatedly in the long term (sustainable), while the unusual earning is temporary generated earnings and it cannot be produced repeatedly, and therefore it cannot be used as a good indicator of future period earnings.

Theory of Performance
Performance measurement is a process to determine how well the business activities carried out in order to achieve strategic objectives, eliminate waste and present information on timely basis to make continuous improvement.The purpose of performance measurement is to motivate employees to achieve the organization's objectives by complying with the standards of conduct that have been established before, in order to produce the desired actions and outcomes (Mulyadi 2001: 415-416).In the context of bank, the financial condition of a bank can be seen from the financial statements presented by the bank periodically.The financial statements also illustrate the performance of all banks for a period.

Bank Performance
Bank Financial statement is a form of man-agement accountability to the parties concerned with the performance of the banks achieved during a certain period (Ismail 2009: 15).Financial performance is an analysis conducted to see the extent to which a company has implemented financial principles correctly and properly (Irham 2011: 2).

Firm Size
Size of the firm may influence how easy the firm to raise funds from the capital market, bargaining power in financial contracts and the scale of the costs and return.These lead bigger companies to produce higher profits.Machfoedz states that size of the company can be determined based on sales, total assets, labor, etc., which all are highly correlated (Diastity 2010).
Based on this theory, the framework of thought can be described as Figure 1.Earnings quality of banks is good when there is no indication of earnings management made by the bank.The management does not commit any fraud or manipulation on bank's financial statements.The good quality of earnings significantly affects bank performance, both internally and externally, because the banks are able to maintain the stability of the health of the bank itself.
Based on the formulation of the problem, the purpose of the research and the theoretical basis, the hypothesis proposed in this study are as the following: H1: Earnings persistence affects the performance, terms of CBTD (Cash and Bank to Total Deposits), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable.H2: Earnings persistence affects the performance, ETTA (Equity to Total Assets), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable.H3: Earnings persistence affects the performance, LTTD (Loan to Total Deposits), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable H4: Earnings persistence affects the performance, OPM (Operating Profit Margin), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable.H5: Earnings persistence affects the performance, NPM (Net Profit Margin), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable.H6: Earnings persistence affects the performance, ROI (Return On Investment), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable.H7: Earnings persistence affects the performance, ROE (Return On Equity), of bank listed on the Jakarta Stock Exchange (IDX), with bank size as the control variable.

RESEARCH METHOD Research Design
The research uses secondary data of financial statement and quantitative approach (2008).It uses the annual financial statements of bank listed on the Indonesian Stock Exchange (IDX).

Operational Definition and Variable Measurement
There are three groups of variables used in this study, namely: the dependent variable, the independent variables, and the control variables (variables that are controlled or held constant so that the relationship between the independent variables and the dependent variable is not influenced by other factors beyond those examined).

Dependent variables
The dependent variables are bank performance, measured using financial ratios from For example; it is ICMD which includes CBTD (Cash and Bank to Total Deposits), ETTA (Equity to Total Asset), LTTD (Loan to Total Deposits), OPM (Operating Profit Margin), NPM (Net Profit Margin), ROI (Return on Investment) and ROE (Return on Equity).This measurement is according to Handayani (2005) and Irham (2011).

Independent variable
The independent variable is earnings quality measured by the earnings persistence.This measurement is according to Fanani (2010), Djamaludin et al. (2010), andMargani (2009).It is assumed that good quality earnings reflect the actual earnings or earnings that are not managed for a specific interested party.In this study, earnings quality is measured using earnings persistence.Thus, earning persistence is an earnings measurement based on the view that sustainable earnings have good quality.It is measured using: = Constant e = Time series variance of disturbance to income from accounting earnings EPS t = EPS firm i at year t EPS t-i = EPS firm i at one year before t.Based on equation 1, the residual (e) of the effect of previous earnings on current earnings is obtained.The Residual values are then sought for their variants.The higher the residual value variance, the lower the quality of earnings will be.This means that profits are increasingly able to predict it own.

Control variable
The control variable is total assets.

Population, Sample, and Sampling Technique
The population is all the elements that can be used to make some conclusions.The population in this study is a bank listed in the Indonesia Stock Exchange, while the sample in this study is banking companies whose data on financial statements is available from 2007 -2010 and published consecutively in rupiah.The data is collected from www.idx.co.id and Indonesian Capital Market Directory (ICMD).Sampling is conducted using purposive sampling to obtain representative samples based on the predetermined criteria

DATA ANALYSIS AND DISCUSSION Normality Test
This study tests the normality of data using Kolmogorov Smirnov test (Imam 2011).Normality in this case is the normality of residual.Residual values are considered to be normal if the significant probability > 0.05.Normality test aims to test whether the residuals of the regression models normally distributed or not.Regression model is good when the residual values are normally distributed.Summary of test results for normality is presented in Table 1.

Hypothesis Testing
The research hypotheses are tested using regression test (Imam 2011).Regression test results are summarized in Table 2. Based on the results in Table 2 shows that Bank financial ratios cover CBTD (Cash and Bank to Total Deposits), ETTA (Equity to Total Assets), LTTD (Loan to Total Deposits), OPM (Operating Profit Margin), NPM (Net Profit Based on the summary in Table 2, it appears that only CBTD, OPM, NPM, and ROI are influenced by earnings persistence, after controlling for the total assets of the bank.While earnings persistence does not affect ETTA, after controlling for the total asset of bank.In addition, earnings persistence, along with total asset as control variable, does not affect LTTD and ROE.

CONCLUSION, IMPLICATION, SUG-GESTION, AND LIMITATIONS
This study aimed to analyze the effect of earnings persistence on corporate performance of commercial banks in Indonesia in the period from 2007 to 2010.In theory, the higher the quality of the earnings, the better the bank's corporate performance.Therefore, better profit will attract more investors to invest their funds.In this study, quality of earnings is measured using earnings persistence.Accounting profit is considered to be persistent if the coefficient of variation is small.Current accrual components of earnings is less likely to recur in the future or it is less persistent to determine future earnings as they are based on accrual and deferred basis, as well as the subjective allocations and assessments.Some financial analysts prefer operating cash flows as a determinant of the quality of the earnings because the cash flow is considered more persistent than the accrual component.They believe that the higher the ratio of operating cash flow to net income, the higher the quality of the earnings Based on the sampling criteria, the number of commercial banks in this study is 12 commercial banks for 4 years, so the numbers of observations are 48 data.
Based on the description above, it can be concluded that the persistence of earnings significantly influence future performance of banks in terms of CBTD, ETTA, OPM, NPM, and ROI so that hypothesis 1, hypothesis 2, the hypothesis 4, hypothesis, and hypothesis 5, and hypothesis -6 can be accepted.While hypothesis 3 and hypothesis 7 cannot be accepted as there is no significant impact of earnings persistence on bank performance in terms of LTTD and ROE.
The limitations and suggestions on this study are as follows: 1. Originally, the data for prediction covers the period of 2010 -2003, however much of the data is not complete for the year.There are only 12 bank1 with consistent date over 2007 -2010.2. The study focused only on earning persistence as a measure of earnings quality.It is not combined with other measures of earnings quality.Future research should undertake other measures of earnings quality such as persistence, predictability, variability, smoothness, abnormal accruals, and accruals quality.
3. This study used samples of commercial banks in Indonesia.Future researches are expected to do similar research using manufacturing firms so that they can obtain more general picture associated with earnings quality.