CHAPTER 2 THEORITICAL FOUNDATION: LITERATURE REVIEW. According to Wild, Subramanyam & Halsey (2007), financial statements are the

Transcription

1 CHAPTER 2 THEORITICAL FOUNDATION: LITERATURE REVIEW 2.1 Financial Statements According to Wild, Subramanyam & Halsey (2007), financial statements are the major source of information used by its users to have an understanding about a company. There are five components of financial statements that are available to the users, which are balance sheet, income statement, owner s equity statement, cash flow statement and notes to financial statements. The preparation of financial statements is required to provide information to its users regarding the financial condition, performance and future prospects of a company (Alfredson, Leo, Picker, Pacter, and Radford, 2005). Alfredson et al., (2005) identify that each financial statement has its own objectives. Balance sheet is prepared to provide information about a company s financial condition, while income statement is used to provide information about a company s financial performances. The owner s equity statement is used to assess the full performance of a company, it includes revenues, expenses, gains or losses that are reported in income statement and also transactions related to equity holders (such as, owner s investments and withdrawal). Moreover, the statement of owner s equity acts as a connection between the balance sheet and income statement. Cash flow statement is prepared to support users with the information about the movement of cash (inflows and outflows) in the company. The last statement which is the notes to 10

2 11 financial statements contains additional information to improve the understanding of the other four statements mentioned above. Basically, the objective of financial statements is to provide useful information to its users. However, the preparation of financial statements that contains useful information requires the preparers to understand the conceptual framework of financial reporting. The framework prepared by accounting standard boards will act as a guidance to identify the restrictions of financial reporting, identify the events that are required to be reported, the methods to recognize and measure those events and the procedures to summarize and report the events (FASB, 2006). 2.2 Conceptual Framework for Financial Reporting The Financial Accounting Standards Board (FASB, 2006) determines the conceptual framework for reporting financial statements, which defines the objective of financial reporting, qualitative characteristics of accounting information, elements of financial statements, and the operating guidelines of financial statements. The conceptual framework prepared by FASB can be divided into three different levels consisting of (Kieso, Weygandt and Warfield, 2004): Level 1: Objectives of financial reporting The why Goals and Purposes of accounting Level 2: The bridge between level 1 and level 3 consists of the qualitative characteristics of accounting information and the elements of financial statements

3 12 Level 3: Operating guidelines (Recognition and Measurement Concepts) - The how implementation The three different levels can be illustrated into a diagram that is prepared by Kieso et al., (2004). Source: Kieso, Weygandt and Warfield (2004), p.30 Figure 2.1 Conceptual Frameworks for Financial Reporting FASB (2006) identifies that the objectives of financial reporting are to provide useful information for investing and financing decisions, for assessing future cash

4 13 flows, and report information regarding the company s resources, claims on the resources and the changes in resources and claims. Besides FASB, the Indonesian Institute of Accountants (IAI) also publishes an accounting standard called Standar Akutansi Keuangan (IAI, 2007) that determines similar financial reporting s objectives. In order to provide useful information, FASB (2006) and IAI (2007) require accounting information to possess qualitative characteristics which are relevance, reliability, comparability and consistency. FASB (2006), IAI (2007), Weygandt, Kieso and Kimmel (2005), Stice, Stice and Skousen (2000) and Alfredson et al., (2005) describe each of the qualitative characteristics of accounting information as follows; accounting information is relevant when it is able to influence the decision of its users (relevant), it is reliable if it is free from material error and bias (reliability), it should be comparable with other entity or comparable accounting information over time within the entity itself (comparability), and financial statements are prepared using the same accounting principles over time (consistency).

5 14 Source: Kieso, Weygandt and Warfield (2004), p.32 Figure 2.2 Hierarchy of Accounting Qualities Figure 2.2 shows the relationship of the accounting qualities defined by Financial Accounting Standard Board. It shows that accounting information needs to be understandable, and useful for decision making. The four qualitative characteristics are further classified by FASB, Statement of Financial Accounting Concepts No. 2 (2008) into two groups, which are primary qualities and secondary qualities. The primary qualities consist of relevance and reliability, while the secondary qualities consist of comparability and consistency. It

6 15 can be seen from the figure that each primary quality can be described by three different components. FASB (2008), Weygandt et al., (2005), Stice et al., (2000) and Alfredson et al., (2005) identify that the components of relevance in accounting information are feedback value, predictive value and timeliness. Feedback value allows information to be useful for the confirmation and correction of past expectations. Predictive value allows accounting information to forecast the outcome of similar future events. Timeliness means that information is always updated and available whenever the users need to make their decisions. Besides that, they also identify the components of reliability, which are verifiability, faithful representation, and neutral. Verifiability requires the information to be prepared under a method that can be verified by other accountants using the same method. Faithful representation requires information to present the facts of an event or transaction. Neutral means that the information is not made to support one particular party over another and it does not have the intention to influence any users to make a particular decision that is in favor of its preparers. Stice et al., (2000) argue that the main purpose of financial reporting is to provide valuable information to its user decision making process. Despite of all the qualitative characteristics of accounting information that have been described above, this paper focuses only on the relevance of accounting information which aims to

7 16 identify whether accounting information provided in financial statements are able to influence its users on their pricing decisions. 2.3 The Value Relevance of Accounting Information Accounting information will be considered as relevant if it is able to influence the decision of its users. Wild et al., (2007) suggest that value relevance of accounting information can be observed by examining the explanatory power of accounting information provided in financial statements, to reveal share prices. Francis and Schipper (1999) point out four possible interpretations of value relevance in their studies and one of the interpretations is similar to the suggestion given by Wild et al., (2007). They state that: A statistical association between accounting information and market value of returns, particularly over a long window, might mean only that the accounting information in question is correlated with information used by investors (Interpretation 4). Under this view, value relevance is measured by the ability of financial statement information to capture or summarize information, regardless of source, that affects share values. Besides of these studies, many other researchers have used the same interpretation to test the value relevance of accounting information. Therefore, this paper will perform an analysis on the relationship of accounting information and share prices, to determine the value relevance of accounting information. It is known that various researches on the value relevance of accounting information had been done by researchers around the world. They had used different accounting

8 17 variables as the proxies of accounting information, such as earnings, residual income, dividends and book value. However, it is identified that earnings and book value have been the most common variables that are used to represent accounting information (Collins et al., (1997), Lev & Zarowin (1999), Francis and Schipper (1999)) The Value Relevance of Earnings In financial statements, earnings are usually known as net income, it is the bottom line of income statement. Earnings allow the users of financial statements to know about a company s performance. According to Wild et al., (2007), earnings are more relevant compared to cash flow due to the reason that it can predict future income and measure the financial condition and performance of a company. Ball & Brown (1968), Easton & Harris (1991) and Ohlson & Shroff (1992) identify that there is a relationship between earnings and share price. These findings help to discover that earnings can be the explanatory power of share prices. Many studies had been done to determine the value relevance of earnings. Bowen, Burgstahler and Daley (1987), Abukari, Jog and McConomy (2000), and Easton, Harris & Ohslon (1992) have the same findings that earnings are one of the variables that explains share price. Pradhono and Christiawan (2004) conduct a research involving manufacturing firms that are listed in Indonesia Stock Exchange (IDX). They discover that earnings have a significant relationship with share prices.

9 18 Many studies conducted in different countries were able to identify that the value relevance of earnings has been increasing over time. Hutagaol and Pranatio (2009) conduct a research using LQ 45 companies that were listed in IDX during 2003 to They found that value relevance of earnings have been increasing over time. Kadri, Ibrahim and Aziz (2009) also identify that there is an increasing trend on the explanatory power of earnings. They used high technology firms as their sample because previous research performed by Dontoh, Radhakrishnan and Ronen (2004), and Liang and Yao (2005) suggest that those firms are the industries that are most affected by the decline in the value relevance of accounting information. Gjerde, Knivsflå* and Sættem (2009) perform a research to test the value relevance of financial reporting in Norway during 1965 to 2004 and they also discover that the value relevance of earnings have been increasing over time. In contrast to the studies performed above, there are also empirical studies that show the decline in the explanatory power of earnings. Davis-Friday and Gordon (2002) conduct a study using Mexican firms during the period 1992 to 1997 and find that the explanatory power of earnings decline due to the presence of negative earnings. Broedel Lopes (2002) signifies that earnings do not provide relevant information for the equity valuation in Brazil. Rahmawati (2005) had performed a research by collecting the financial statements of 41 listed manufacturing firms in IDX. The result of her research determines that earnings had lost its value relevance during 1998 to Research performed by

10 19 Collins et al., (1997), Barth, Beaver and Landsman (2001) and Lev and Zarowin (1999) indicate that there is a significant decrease in the value relevance of earnings over time The Value Relevance of Book Value According to Spremann and Gantenbein (2002), book value of equity has been included as one of the variables that may explain share price, due to the framework created by Feltham-Ohlson (1995). The Feltham Ohlson framework define share price as a function of earnings and book value of equity. Spremann and Gantenbein (2002), Collins, Pincus and Xie (1999) also state that book value of equity represents the net assets used to generate normal earnings that are expected to persist in the future. Several studies performed by Abukari et al., (2000), Broedel Lopes (2002), Hutagaol and Pranatio (2009) and Burgstahler and Dichev (1997) have also specified that book value should be included as one of the explanatory variable of share prices. Some studies discover that the value relevance of book value has been declining over time. Kadri et al., (2009) show that there is a downtrend in the value relevance of book value over time. Lev and Zarowin (1999) also document that the relationship between book value and capital market values has weakened due to the change in business environment and the inadequacy of the change in accounting treatment. In contrast to studies performed by Kadri et al., (2009) and Lev and Zarowin (1999), other studies performed by Suwardi (2005), Francis and Schipper (1999) and Collins

11 20 et al., (1997) discover that the value relevance of book value increases over time. Collins et al., (1997) explain that the increase in the value relevance of book value may be influenced by several factors such as the increase in non-recurring items, firm size, negative earnings and intangible firms. Many study such as Collins et al., (1997), Spremann and Gantenbein (2002), Subramanyam and Venkatachalam (2000) state that book value can be used as a proxy to forecast future earnings when transitory components are material in current earnings. Besides that, it is also useful because it acts as a proxy for the liquidation or abandonment option for firms with poor performance. With the support of these studies, Collins et al., (1999) find that the omission of book value from the simple earnings capitalization model creates a negative (positive) bias in the coefficient of earnings for loss (profit) firms. Therefore, book value is said to be able to contribute in the elimination of the negative relation between price and return found in loss firms. However, research performed in Greece by Papadaki and Siougle (2007), discover that the inclusion of book value in the simple earnings capitalization model does not assist in eliminating the negative coefficient on earnings of loss firms The Value Relevance of Cash Flow In addition to earnings and book value, many studies have also included cash flow to determine share prices. Sumarni and Rahmawati (2007), Bowen et al, (1987), Pradhono and Christiawan (2004) discover that cash flow contains information to explain share prices.

12 21 According to Keown, Martin, Petty and Scott (2005), cash flows are important compared to profits because it is cash that is actually received and reinvested by firms. Cash flows are also the important components used in capital budgeting. It can be used to project the Net Present Value (NPV) of an investment and the NPV will be used to make investment decision. Keown et al., (2005) and Berk, Demarzo and Harford (2008) indicate that Free Cash Flow (FCF) Valuation Model can be used to value a firm. Firm value can be found using the FCF model by dividing the total present value of all future free cash flows to the company s weighted average cost of capital. Through this model (given in Figure 2.3), share value can be determined. Source: Keown et al, (2005), p.440 & 443 Figure 2.3 Free Cash Flow Valuation Model FCF Valuation Model is one of the examples to indicate that there is a relationship between cash flows and share price. Darwin (2008) conducts a research in Indonesia and observes that there is no significant evidence to conclude that cash flows are related to share prices during 1992 to Nevertheless, she observes that there is a significant relationship between cash flow and share price during 1996 to She predicts that cash flow

13 22 statements of Indonesian companies were still ineffective during 1992 to 1995 and caused cash flows to have an insignificant relationship with share prices during that period. Juniarti and Limanjaya (2005) conduct a research involving 50 Indonesia s listed companies and find that cash flow is more relevant than earnings for companies that are in the growth and mature stage. Lev and Zarowin (1999) state that cash flow is more informative than earnings as the probability of manipulating cash flow is lower than earnings. However, they do not find the evidence that the value relevance of cash flow increases over time. Cash flow is also determined to have higher explanatory power in equity valuation when earnings are transitory (Charitou, Clubb and Andreou (2000), Cheng, Liu and Schaefer (1996) and Habib (2008)). On the other hand, Habib (2008) finds that the explanatory power of earnings is higher than cash flow; however the difference appears to be insignificant. Thinggaard and Damkier (2008) also show that investments decisions that are made based on accrual information are more profitable than the investment decisions that are made based on cash flow information. This means that accrual information (earnings) is more relevant than cash flow. FASB (2006) clarifies in Conceptual Framework for Financial Reporting that: The accrual accounting information in financial reports about an entity s resources and claims and changes in resources and claims generally provides a better basis for assessing cash flow prospects than information solely about the entity s current cash receipts and payments. Without accrual accounting,

14 23 important economic resources and claims to resources would be excluded from financial statements. These statements highlight the importance of variables prepared by using accrual method over cash flow variables Prior Research on the Trend of the Value Relevance of Accounting Information The value relevance of accounting information can be tested by examining whether the information published in financial statements is useful to influence user s decisions. Financial statements provide much information that can be used as a proxy for accounting information. However, many prior researches have used earnings, book value or cash flow to represent accounting information. Many studies have confirmed that combining several variables into one model can provide higher explanatory power for share prices. Bae & Kim (1998) suggest that book value (earnings) can explain some factors of share prices that are not explained by earnings (book value). They also find that both book value and earnings have value relevance towards share prices and both are complementary to each other. Bowen et al, (1987), Charitou et al., (2000), Habib (2008) identify that cash flow (earnings) have incremental information content beyond earnings (cash flow) in explaining security returns, which signifies that combining both cash flow and earnings will provide higher explanatory power for security returns.

15 24 Studies have been made to discover the trend in the value relevance of accounting information over time. Graham and King (1998) show mixed results in determining the value relevance of accounting information for different countries over time. They identify that the value relevance of accounting information in Korea, Philippines and Taiwan increases over time, while Indonesia and Thailand experience a decline in the value relevance of its accounting information. Eng, Li and Mak (1999) identify that Australia, France and Germany has an increasing trend in the value relevance of accounting information, whereas United States of America, Japan and Sweden show an opposite trend. Davis-Friday and Gordon (2002) perform a study in Mexico and identify that the value relevance of Mexican accounting information does not show a decline during the economic crisis. With the support of their findings, they state that the accounting system in Mexico can be used to improve the financial reporting in United States. Dontoh et al., (2004) find that the R² of the model used in the research has decline over time which means that the explanatory power of accounting information falls and it was affected by the increasing amount of Non-Information Based trading. Cortijo, Palmon and Yezegel, Core, Guay and Buskirk (2001), Lev and Zarowin (1999) indicate that there is a decline in the usefulness of accounting information due to the change in business environment (from low-tech to high-tech firms). Gu (2007) and Brown, Lo and Lys, (2000) find that there is a decline in value relevance when

16 25 alternative measure such as regression residual dispersion are used to measure value relevance. In the contrary to the research pointed above, Thinggaard and Damkier (2008) note that the result of their research indicates that the value relevance of accounting information in Denmark remained unchanged over time ( ). Moreover, several numbers of studies (Landsman and Maydew (2002), Gjerde et al., (2009) and Kadri et al., (2009)) had found that the value relevance of accounting information does not decline over time. Collins et al., (1997), Francis and Schipper (1999), Ely and Waymire (1998) observe that there is an uptrend in the value relevance of accounting information during the mid of 1950s to the early of 1990s. As stated before, Graham and King (1998) find that the usefulness of Indonesian accounting information falls. In contrast to their study, Hariani and Nashih (2006), Hutagaol and Pranatio (2009), and Darwin (2008) find an uptrend in the value relevance of Indonesian accounting information. The inconsistency of their findings may be due to the difference in the period in which they conduct the research. Graham and King (1998) use 338 firm years in Indonesia to conduct the research and the data is taken from the year 1991 to Financial firms and firms with negative book values are removed from the sample collections in their research. Hariani and Nashih (2006) also conduct their research in Indonesia, but they obtain their data from the year 1993 to Besides that, they exclude new firms and firms with

17 26 negative earnings from their samples. Hutagaol and Pranatio (2009) perform a research to identify the change in the value relevance of accounting information by using samples from LQ 45 firms in Indonesia. Their samples are collected based on the list of LQ 45 at the end of December The data are obtained from the period of 2003 to 2007 and 220 firm years are used as their final samples. Darwin (2008) performs a research in Indonesia by including data from 1992 to 2005 and obtains a final sample of 1151 firm years. She excludes financial firms and firms that report their financial statements in currencies other than Rupiah. The research performed by Darwin (2008) involves using the same number of firms each year; it includes 86 firms that are listed every year during 1992 to As the research is conducted using the data from 1992 to 2005, the author decides to divide the sample into two periods which are the pre-crisis and post-crisis period. This is done to prevent distortions in the research that can be caused by the change in the financial condition and structural change in Indonesia Factors Affecting the Value Relevance of Accounting Information Collins et al., (1997) classify that there are four factors that may affect the value relevance of accounting information. The four factors are the increase in the number of intangible firms, increase in non-recurring items, the presence of negative earnings and difference in firm sizes.

18 27 Cortijo et al., (1997) find that the accounting information provided by hightechnological firms (contain higher intangibles) are less relevant compared to lowtechnological firms. Lev and Zarowin (1999) also conduct a research to test the impacts of intangible firms on the value relevance of earnings information. They find that increasing intangible investment, such as research and development (R&D) can cause a decline in the earnings explanatory power. Core et al., (2001) show that using subsamples that include high-technological firms have caused the explanatory power of accounting information to decline over time. On the contrary, Kadri et al., (2009) and Broedel Lopes (2002) state that intangible firms do not cause a decline in the explanatory power of accounting information. Francis and Schipper (1999) also find that high-technological firms do not show higher decline in value relevance of accounting information compared to low-technological firms. Davis-Friday and Gordon (2002) and Barth et al., (2001) identify that the decline in the explanatory power of earnings are caused by the presence of negative earnings. Research conducted by Hutagaol and Pranatio (2009) fail to discover that there is an impact of negative earnings on equity valuation. They predict that the use of a small number of samples containing negative earnings may be the reason of the unexpected findings. Filip and Raffournier (2010) conclude that the relationship between earnings and market return increase significantly when they exclude negative earnings, this means that the presence of negative earnings decrease the relationship between earnings and market return.

19 28 According to Collins et al., (1997), non-recurring items are also considered as transitory components. Brown and Sivakumar (2003), Elliot and Hanna (1996), Elliot and Shaw (1988) and Ali and Zarowin (1992) conclude that accounting information that contain less transitory components or non-recurring items are more relevant than information with more transitory components. Firm size can be identified by determining the life cycle stage of a company. Collins et al., (1997) point out that startup companies may be considered as smaller firms that are likely to report losses. They expect that book values will be more relevant for smaller firms. Hayn (1995) find that information content of earnings increase with the reduction in the frequency of losses. As smaller firms are predicted to report losses more frequently than larger firms, it can be regarded that the information content of earnings for smaller firms are lower than larger firms. 2.4 Methodology for Value Relevance Analysis Value relevance can be tested by identifying how accounting information can reveal share prices. However, the measurement of how well accounting information can reflect share prices require the use of statistical tools. Different methods have been used to measure value relevance, such as Coefficients of Determination (R-Squared or R²), R-squared after controlling for scale effects and the regression residual dispersion.

20 29 R-Squared method is used frequently by many researchers to measure value relevance (Brown et al., (2000)). Researchers such as Collins et al., (1997), Francis and Schipper (1999), Hariani and Nashih (2006), Hutagaol and Pranatio (2009), Kadri et al., (2009), Suwardi (2005) and Ely and Waymire (1998) employ the R- Squared method to investigate whether accounting information is relevant and whether it has changed over time. According to Brown et al., (2000), R-Squared is a measure of the explanatory power of the independent variables in a linear regression. R-Squared is also known as the coefficient of determination; it explains the degree of dependent variables that can be explained by independent variables (Keller, 2005). In this study, the independent variables will be the variables that have been chosen to represent accounting information, while the dependent variable will be the share price. Although R-Squared method is commonly used by researchers, studies have been done to prove that the R-Squared method is unreliable. Brown et al., (2000), Gu (2007), Chang (1999) and Cortijo et al., suggest that R-squared method are not a reliable statistic due to the exposures to scale factor and heteroscedasticity in regression variables. They suggest some approaches or techniques to control the exposures that are faced by R-Squared method. Brown et al., (2000) are concerned regarding the bias that may be caused by using the Coefficients of Determination (R²) in price levels regression. They believe that R-Squared method is exposed to scale effects because different samples may have

21 30 different scaling factor. They identify that there is a positive relationship between R² and the coefficient of variation (CV) of the scale factor, therefore, even when the value relevance remains unchanged, R² will be higher due to the increase in the CV of the scale factor. In response to the concern mentioned by Brown et al., (2000) regarding the different scaling factors that exist in different samples, this study have dealt with this issue by conducting an extended research that performs a quartile analysis on the research samples. Two approaches are suggested by Brown et al., (2000) to control the scale effects in R², which are an empirical analysis between R² and scale effects, and an analysis from deflated regressions. Empirical analysis between R² and scale effects involves the investigation of whether an increase in CV can affect the increase in R² and how it correlates with time. After the findings are obtained, the trend in the R² of accounting information after controlling for changes in the CV of the scale factor will be investigated. The next approach which is the analysis from deflated regressions involves the deflation of the variables by a scale proxy (a proxy for unobservable scale factor). By ignoring the scale effects, the R² of the model will be upwardly biased. The amount of the bias depends on the CV of the scale factor. This study has also control the scale effects in R² by deflating the research model using a size proxy which is the number of outstanding shares. Gu (2007) has proposed the use of the regression residual dispersion for determining the value relevance of accounting information. He believes that the use of R-squared

22 31 method to explain the value relevance of accounting information is incomparable across samples. Gu (2007) interpret residual dispersion as the pricing errors that are not captured by accounting information. However, even though regression residual dispersion is exposed to nonlinear scale effects that are related to the magnitude of price/ return levels, Gu (2007) suggests some techniques to control the effects. The techniques are to measure scale-adjusted abnormal pricing errors (also called as sizeadjusted abnormal returns), examine time patterns of the pricing errors within the deciles where scales have already been aligned, and find scale-matched samples. This study does not apply the method that has been proposed by Gu (2007). However, the author would like to recommend the use of the regression residual dispersion method in future research. This is to identify whether the method proposed by Gu (2007) will affect the results obtained from the R-Squared method.