1

Empirical Tests for Market Timing Theory of Capital Structure on the Indonesian Stock Exchange
Ignatius Roni Setyawan (ignronis@gmail.com) (Lecturer from Tarumanagara University (UNTAR), Jakarta, Indonesia) Budi Frensidy (frensidy@yahoo.com) (Senior Lecturer from Indonesian University (UI), Jakarta, Indonesia)
Paper submitted to 14th Malaysian Finance Association Conference 2012 1 - 3 June 2012, Penang, Malaysia (http://gsb.usm.my/mfac2012) Abstract This study has aim to examine the validity of Market Timing Theory (MTT) from Baker and Wurgler (2002) in the Indonesian context. The essence of MTT is when the market price overvalued, the firms will take debt financing and otherwise for undervalued condition. MTT is actually the development of Pecking Order Theory (POT) and Static Trade-Off Theory (STT).The motivations of this study are to test the dispute level of pro and cons empirical studies about MTT such as Alti (2003) and Wagner (2007) and to check the consistency result of empirical studies of MTT in Indonesian from Dahlan (2004), Kusumawati and Danny (2006), Susilawati (2008) and Saad (2010). In order to realize the objective, this study will reuse the empirical model OLS from Baker and Wurgler (2002) with adaptation in Indonesian context. The empirical model OLS from Baker and Wurgler (2002) has a specific uniqueness i.e. the negative relation between leverage and market to book ratio. That negative relation is controlled by several factors such as EAT, Total Asset and Fixed Asset. The other specific uniqueness is empirical models of MTT are generally applied for IPO-firms. The result of this study supports the hypothesis of MTT from Baker and Wurgler (2002) in Indonesian Stock Exchange (IDX) with the main finding i.e. market to book ratio has the negative impact to market leverage. While the relevant factor for supporting the hypothesis of MTT is EAT. The implication is when firm achieve the earnings growth due to increasing of EAT; the stock price will be overvalued as an impact from investor positive sentiment. This situation suggests the firm in IDX should conduct the debt financing. Key word : Market Timing Theory, IPO, Market To Book Ratio, Book Leverage, Market Leverage : G3;G31;G32

JEL Classification

Empirical Tests for Market Timing Theory of Capital Structure on the Indonesian Stock Exchange
INTRODUCTION Management does not know when the optimal capital structure especially in the capital market investors. The issue becomes complex when management must decide determinant factor "when" optimal capital structure. So it is no longer such an argument Elliot, et.al. (2004) i.e., how many servings of leverage so that the optimality is reached. Theory - the theory of capital structure such as the traditional pecking order theory (POT) and the Static Trade-Off Theory (STT) has not been satisfactory financial managers in determining the best capital structure policy. Instead they compete with each other in determining the best proxy determinant factor [see Frank and Goyale (2003) and Liu (2005)]. Both quantitative theory-minded, STT is more emphasis on optimal leverage so that the company is safe from financial distress while more emphasis POT optimum priority in issuing capital. Whereas psychological factors in the capital structure decision according to the behavioralist view like Kant (2003) and Miglo (2010) is interesting to consider. This is because the study of Graham and Harvey (2001) has accommodated the psychological approach of capital structure through a survey of CFOs in the USA. The emergence of Market Timing Theory (MTT) from Barker and Wurgler (2002) is expected to provide "answers"; but will not be as easy as imagined. MTT proxy in general is the market to book ratio, i.e., in cases of IPO. Many academics as quoted Huang and Ritter (2005) criticized this proxy because generally the market to book ratio is a proxy of investment decisions; the under-valued or over-valued him a stock. Barker and Wurgler (2002) claimed market timing is "the cumulative outcome of past attempts to time the equity market".1 Two assumptions are such as: 1. Asymmetric information occurs varies in the capital market, then the rational management is reluctant to make adjustments to the target leverage. 2. Management believes can do the "timing" of the equity market. And claims Barker and Wurgler (2002) was successfully derived in empirical models. However, MTT from Barker and Wurgler (2002) raises a lot of pros and cons academically. Pros and cons rather than on the assumption of the 2nd as well as allegations of this study; but instead on the assumption that sooner or absence of a management perform adjustment of the target leverage. Based on a survey study of Huang and Ritter (2005), scholars who pros with MTT include Welch (2004); Kayhan and Titman (2005) and Lemmon, et al. (2005). While the cons include the MTT Leary and Robert (2005), Alti (2003) are skeptical of the definition of market timing Barker and Wurgler (2002) and Hovakimian (2005). Pros and Cons of the Market Timing Theory according to the behavioralist such as Kant (2003) and Miglo (2010) derived from the condition of the company's internal and external factors (capital market situation). Pros to MTT when the capital market situation and conditions established investor sentiment affects the company's internal management acts in making funding decisions. Whereas the opposite conditions prevail cons MTT [see Vasiliou and Daskalakis (2007)].2
1 Although Kant (2003) and Miglo (2010) says market timing is not a new idea. His view is reinforced also by
Graham and Harvey (2001) that there is an indication of behavioral management in the equity issuance. Another Hovakimian, et.al. (2001) when the stock price rises, companies will make equity offerings. Then Tobing (2008) and Saad (2010) mention the problem of adverse selection as the issuance. [details see also Frank and Goyale (2003), p.7]. Dittmar and Thakor (2007) stated the company issuing equity-related problems also with the indications of irrationality among investors and management. Investors and management were no longer much use of intuition the rationality of the Bayes theorem (Neoclassical Paradigm) in decision making. According Vasilou and Daskalis (2007) concept of market efficiency of Fama and perfect market of MM Theorem is threatened.

1.

Leverage company publicly traded on the Stock Exchange before the financial crisis has increased sharply and after the financial crisis tends to fall. A sharp increase before the crisis can be seen from the convenience. On the other hand, welcomed the company's management conglomerate. But then that happens because bad credit bank loans used for own business group (related lending) which ignores the principle of prudential (prudential banking). So after the 1998-2002 monetary crisis many commercial banks are forced to "freeze operation" and "take-over" by regulator such Indonesian Banking Restructuring Agency (IBRA). Medium conglomerate as many bank shareholders to restructure debt and business efficiency. The phenomenon of companies on the Stock Exchange during the financial crisis did not prove easy to implement an optimal capital structure targets. Theory POT, STT and MTT are expected to provide potential solutions for the target leverage (Tobing, 2008). But targeting cannot leverage on the basis of practical judgment alone but must be of empirical studies. On the basis that this study intends to test market timing of capital structure in Indonesia. There are two motivations that first, the reconciliation debate theoretical study of MTT as Alti (2003) and Hogfeltd and Oborenko (2005) [cons to MTT] and the study of Kayhan and Titman (2005) and Wagner (2007) [pros to MTT]. Second, new research MTT was done four times in each of the BEI Kusumawati and Danny (2006) which emphasizes the effects of long-term persistence of capital structure with MTT method and OCS (optimal capital structure to STT) and Dahlan (2004) which focuses on the existing any indication of the policy in Indonesia's capital structure leads to the MTT. In addition there are also from Susilawati (2008) and Saad (2010). The general objective of this study is to prove that MTT could be applicable on the Stock Exchange, while specific objectives3 are: to analyze the influence of market to book ratio on leverage and analyze the influence of other variables (control variables) such as net property, plant and equipment; After Tax and Total Earning Assets of leverage. Urgency general purpose is to look for evidence of indications of MTT in the BEI, i.e. market value to book ratio will negatively affect leverage. The logic at the time the company experienced high growth (one of its proxies is the market to book ratio), then the company will tend to reduce the use of debt (one of his proxies is leverage). This is because at that time in the capital market investors will assess the company is under-valued so that the cost of equity would be less than the cost of debt. And usually this condition will occur when a company (which is experiencing high growth) does an IPO. While the urgency of a special purpose lies in the discovery of the control variables of MTT. There are proxies of the studies Baker and Wurgler (2002) and Huang and Ritter (2005) such as Net Property, Plant and Equipment; After Tax and Total Earning Assets. The role of these variables influence the relationship in the market to book ratio and leverage is interesting to study. Because the variable market to book ratio will not stand alone as a separate variable in the model. From Dahlan (2004), Kusumawati and Danny (2006) and Saad (2010), identified the role of each determinant of leverage as control variables in addition to the market to book ratio, which proved to be a major determinant of leverage to indicate the validity of the MTT in IDX. Some of these control variables such as EBIT, SIZE, Net Working Capital and the lagged-leverage different levels of significance and is one motive for the study of this study.

Some understanding of the variables to proxy MTT will be explained in detail will be discussed at the operational definitions of variables. Special variables used leverage of book value and market value of such studies Huang and Ritter (2005). The difference between market and book leverage in the market value component of total assets.

Thus, this study has several contributions that are redesigning MTT in terms of assumptions, the core, the explanatory variables and model research. Speaking of assumptions, according to this study there are three that targets leverage is great but when it reaches the optimal leverage is much more important. This will entirely depend on the equity issuance. Another assumption is the company will experience a deficit financing, since it is not enough just to rely on internal financing. Finally a proxy other than the cost of capital such as the characteristics of firms and market conditions are also important [Huang and Ritter (2005) have shown]. MTT according to the core of this study is the company should use the equity when the cost of equity capital is cheap. And vice versa when the cost of debt using debt capital is also cheap. But regardless that companies can use a combination of both when the cost of equity capital amounting to the cost of debt capital. This means that this created a perfect optimality of capital structure. Another thing is the funding decisions are also influenced by the current situation of the company's IPO or SEO either. His theory of IPO and SEO will affect the company's capital structure. If the explanatory variables then the ratio M/B, EFWA M/B, the intensity of fixed assets of Baker and Wurgler (2002) can all be applied. Provided with a record variable ratio M/B and EFWA M/B has a negative impact on leverage. Study of Huang and Ritter (2005) succeeded in adding a variable Equity Risk Premium, profitability, firm size, level of sales and net working capital as well as macro variables such as taxes and GDP. The addition of variable extends the findings of Baker and Wurgler (2002). Models of research still refer to OLS Baker and Wurgler (2002), although it could be a panel data regression as study Huang and Ritter (2005). REVIEW OF THEORY Development of Capital Structure Theory As shown in Figure 1, this study introduces the emergence of market timing theory also starts from the conventional MM theory of capital structure in the late 50's. Modigliani and Miller (MM) issued two propositions. The first proposition associated with leverage, arbitrage4 and firm value. While the second related to leverage, risk5 and cost of capital. Berk and De Marzo (2007) stated that both these propositions lead to the assumption that the leverage does not affect firm value, although the main requirement is the perfect market as there are no transaction costs; risks of business every business the same; equitable access to information (symmetric); rationality and homogeneity of expectations among investors. Then we all know that no relevant assumption of debt is a lot of controversy. Because there is evidence that with leverage, EPS will rise, even without leverage was no dilution of ownership of shares. If the study looks, the key to the perfect market assumptions. Then after the revision of the MM theory, there are also alternative theories like the pecking order and static trade-off based on the assumption of imperfect markets such as the existence of asymmetric information and the emergence of financial distress due to the use of debt. With Figure 1, this study suggests the pecking order and the Static Trade-Off has a strong dominance in the 60's s to d 80's. Pecking order starts from the Fortune 500 survey that

2. 2.1.

This is based on the allegations because the interest rate the individual is treated the same as the interest rate the institution; then the individual investor has access to extensive in the capital markets to trade shares of the levered and unlevered firms through homemade leverage. In equilibrium conditions will be created so that the arbitrage process will not be a difference anymore between the levered firm value and the unlevered. So high and low debts will not be different for the firm. Risk factor is the difference arising from capital cost of levered and unlevered firms multiplied by the amount of debt. The larger the debt would make the cost of equity capital increases. Because of tax-shield effect of debt will offset the increase in capital costs alone. That situation would make the amount of debt is irrelevant to the increase in firm value.

generates a sequence of funding. The cost of retained earnings of the least expensive capital in the view of filler survey was relevant, because the management does not need to issue a high opportunity cost for capital access at the owner (principal). While the Static Trade-off begins with the rise of financial experts discuss financial distress as the negative implications of the use of debt. According to STT, shall be used in an optimal debt until the restrictions will not reduce the value of the company. What's interesting how the best proportions still vary for different industrial sectors; giving rise to "optimal leverage puzzle". In the decade of the 80's and 90's many advanced research capital structure refers to the STT and POT. Miglo Studies (2010) noted the two study groups are the pros and cons of the POT. The group's pros to POT as Myers (1984), Baskin (1989), Allen (1993) and Adedeji (1998), whereas those who cons like Shyam-Sunders and Myers (1999) and Frank and Goyale (2003). Manurung (2004) states the pros and cons of research results of the two groups were due to differences between the OLS model that is research and GLS are always competing to be the best estimation model and the need for the industrial sector as a determinant of leverage. GLS will be effective when a large sample studies (involving the industrial sector) or cross-country studies such as Mahajan and Tartaroglu (2007). {Figure 1 here} Since both STT and MTT theories still exist, so this study has the POT and STT scenarios which inspire the emergence of MTT.6 Why can appear MTT? Baker and Wurgler (2002) had stated that the capital structure decisions related to the company's efforts to do the timing of the capital market. Which the reason is POT and STT are not capable of maximizing the value of the firm and MTT that has a character of "persistent" is expected to be a means of goal achievement finance. Persistent keywords into MTT excellence in implementation. In the following section after a detailed discussion of the POT and the STT, the study will discuss it separately. But as Alti (2003) questioned the nature of such persistence MTT; this study suspect there are many research gaps that could be a reference to subsequent studies. Gap research mainly concerned about the reliability of MTT from Baker and Wurgler (2002) as contemporary theories of capital structure and potential problems MTT is dependent on the sample of IPO firms, because it is believed that the IPO will invite investor reaction as in the phenomenon of underpricing. If you look at companies that are not IPO, then according to the opinion of Baker and Wurgler (2002), market to book ratio will not have a significant effect on leverage. 2.2. Static Trade-off and Pecking Order Theories In table 1 of this study and STT POT peeling out using four pillars namely assumptions, core, variables and research models. Selections of these four pillars to more easily discuss a study of the theory by looking at the methodological elements. Seen in table 1 the real difference between STT and POT. POT more emphasis on hierarchical funding, while STT dotted-optimality heavily on funding. Although there are striking differences, but at the core of discussion, both focus on the Cost of Capital. POT focus on the cheap while the STT sticks on the minimum of COC which indicated it is remained as the main target capital structure decisions.

This study argued that MTT is a "slice" of the POT and the STT. This can be evidenced by MTT recognition that the company should set a target leverage, and provides a strong argument when going to use a source of either debt or equity funds.

{Table 1 here} Some of the explanatory variables, this study taken from a study of Pangeran (2004). The main model is a logistic regression with an option for financing equity and debt financing options to 0. As per the study of Pangeran (2004), POT significant explanatory variables are profitability; stock prices and capital market conditions with the direction of positive influence. Then there is no STT explanatory variable is significant, so that claims from Pangeran (2004) POT is more relevant in Indonesia compared to STT. Allegations of this study are related to the period 1991-1996 the data are more bullish. Interestingly, Pangeran (2004) adopted the explanatory variables and STT of POT Bayless and Diltz (1994) (see underlined italic print table 1). That being the case means there is a wedge or a linkage between STT and POT. Deviation of the target leverage can occur because of the size of the stock offerings and stock prices. When the higher the size of the stock offering, the target leverage will be low. Meanwhile, when the stock price higher, then the target leverage will increase, because the company is better to seek additional lenders in the market because investors intend to sell back the shares after the IPO. If the company wants to find a new investor can pass private placement procedures. 2.3. Market Timing Theory (MTT) As per the study Kusumawati and Danny (2006), this study could eventually define operationally MTT easily. This is important because Baker and Wurgler (2002) have made little justification of the MTT, not to mention the groups of researchers are the pros and cons of MTT just too busy with the affairs of the persistence of MTT in econometrics alone. From studies Dahlan (2004) and Kusumawati and Danny (2006), MTT showed more important implications of the choice of debt or equity at various time points compared with the search for the optimal leverage ratio. Saad (2010) mention two points in time that investor sentiment conditions and financial constraints. My study do not use financial constraints factor on the grounds that the sample is not exposed to company-induced effects of the global financial crisis and if taken, be biased, because the context of MTT is good reaction from investors. Then the MTT approach with regard to the release of activity both at the IPO shares, or SEO. Baker and Wurgler (2002), Huang and Ritter (2005) and Saad (2010) gives 4 arguments about effectiveness of MTT (Market Timing Theory) as follows: 1. Companies tend to sell its stake in lieu of debt when market value is high relative to book value and market value of the past is high, and tend to buy back shares when the market value is low. 2. Through the analysis estimates earning prospects and the expected realization of stock prices around the release of stocks, companies tend to sell its stake at the time investors have attitude optimism and high enthusiasm. 3. If the company experienced financial constraints, then with debt funding will be prioritized. For clarity in the bond contract will discipline the managers. If you wish to remain MTT issuing the variation that can be done is SEO or pre-emptive rights (for IDX). 4. MTT should be done when the company has high growth (growth in the Product Life Cycle) because it would invite a lot of market sentiment. Both of these indicate the importance of over-valued or under-valued of a stock, when the company will sell its stake in the stock. And interesting activities will affect the release of shares of capital structure. This issue of the MTT version of the Baker and Wurgler (2002). If

the activity is more prospective release of these shares; then it should have a negative effect between markets to book ratio of equity to leverage.7 Study of Baker and Wurgler (2002) reinforce the findings of the study Fama and French (2002) concerning the negative relationship, and even give recommendations on how the company manages its optimum leverage associated with the market to book equity ratio (M/B). So if the ratio of M/B is low, firms with high leverage may sell its stake. It is the contrast ratio would apply to M/B high. Finally this study to discuss the relationship between the ratio M/B with the leverage tends reciprocal. If it is fundamentally opposed leverage and equity. Seen in the arrangement of postal liabilities; when debt rises and consequently increase the leverage of equity will fall. What makes the debt rises, the company reduces internal funding or increase debt financing. The logic is a good company with a rating of debt, hoping the stock price will rise. But according to the assumption that the opposite occurs MTT high leverage will reduce the price of the stock through a decline in market value of equity. This is because the behavior of investors pessimistic on the company's high leverage despite her debt rating was good. 2.4. Academics Research Pros and Cons against Market Timing Theory Pros and cons of the MTT ranged about the persistence of capital structure could be long term or not. The study results Baker and Wurgler (2002) successfully demonstrated the persistence of the effect of equity issuance net effect is still there. If it was the persistence of the effect is still there then the company does not need to rush to do adjustment of leverage. In Huang and Ritter (2005) revealed two groups, each of which the pros and cons of MTT. Among other groups of pros to Welsch (2004); Kayhan and Titman (2005) and Lemmon et al. (2005). They claim that the sample of firms that had IPOs, the effect of persistence is still so strong even up to 10-20 years. But with nearly the same samples were also Leary and Robert (2005), Alti (2003) and Hovakimian (2005) found the persistence of the effects of missing a few years after the IPO. Their study has alleged problem analysis methods; framework of panel data and a new variable as the trigger factor. Leary and Robert (2005) using GLS which is certainly more robust than OLS from Baker and Wurgler (2002). Meanwhile, Alti (2003) already incorporate elements of hot and cold IPO markets during the panel data framework, although the same model that is OLS. Last Hovakimian (2005) has included new variables such as size, tangibility and profitability in addition to the ratio M/B; ratio of PPE/Assets and the ratio of EBITDA/Asset in the study of Baker and Wurgler (2002). If so the root problems in the definition of the persistence of the "loaded" with the field of econometrics, the study agreed with Huang and Ritter (2005) that need to model appropriate analysis. It seems to be an alternative panel data regression to explain the phenomenon of persistence. Huang and Ritter (2005) successfully demonstrated the persistence of the effect of persistence, although quite weak.

3. 3.1.

RESEARCH METHODOLOGY Research Procedures First, researchers will collect data on the Stock Exchange listed companies with active status. Second, researchers will collect data variables to be tested for each company. Third, researchers will conduct OLS regression to test H1 until H4.

Because this argument is very important for the existence of MTT, the study will peel out in section 3.4 the pros and cons of academic research MTT. Besides there is also criticism of this study to the definition of persistence Kusumawati and Danny (2006) that are too heavy to econometric alone.

3.2.

Data Sources and Sample Type of data to be retrieved by this study is companies that go public in 2008 until 2009. Company data is obtained two sites such as www.idx.co.id and www.finance.yahoo.com and the Indonesian Capital Market Directory (ICMD) 2009 and 2010 and cross-check to the database of OSIRIS from PDBI Indonesian University (UI). 3.3. Sampling Techniques The number of companies that go public in the year 2008 up to 2009 is as much as 52 companies including financial institutions. By using a purposive technique, then collected 28 companies; with details of 14 companies IPO in 2008 and 14 companies IPO in 2009. Purposive sampling criteria are: 1. The Company is not in the financial sector characterized by highly-regulated. 2. The Company is not exposed to the status of delisting during the period 2008-2009, it means not experiencing negative earnings or negative equity due to the global financial crisis of 2008. Thus the sample selected is companies that go public with a high success rate are not affected by the crisis. 3. The company has a complete financial statement information primarily leverage ratio, the number of shares outstanding and stock prices as of 31 December.8 3.4. Operational Definitions and Relations between Variables There are two types of variables that are independent variable and the dependent variable. Dependent variable is the level of debt leverage of the company that affects a company's capital structure. There are two proxy of leverage that is the book leverage and market leverage. Book leverage is measured by the ratio of debt and total assets. While the market leverage is measured by the quotient between total debts minus total equity multiplied by the value of market capitalization and total assets. The independent variable is by definition refers to previous studies. But for the development of relations between independent variables with the dependent variable of this study modification itself; in detail can be described as follows: 1. Market to Book Ratio is the ratio between the value of market capitalization plus total debt to total assets. Market to book ratio is negatively related to leverage expected (H1) on the grounds when the company made the IPO market to book value ratio will be high, this will encourage companies to reduce the debt financing. Saad (2010) stated the high market to book ratio is due to the positive sentiment of investors who believe the prospect of a good company. If H1 accepted, the mean MTT effect on the BEI. 2. Property plant and equipment is net book value of fixed assets which is obtained from the difference in acquisition cost and accumulated depreciation of "books" to walk. Also expected to have a negative influence on leverage (H2) because when the IPO fixed assets will not serve as collateral for debt financing. When the IPO was increasing fixed assets with equity funding of new shareholders. 3. After Tax Earning is the net income tax cut in interest expense and current year. Also expected to have a negative influence on leverage (H3) because when IPO firms will experience an increase in profits. Then the tax-shield effect due to the use of debt will begin to decrease.

Some non-financial firms excluded from the sample. It is based on the consideration that the company's stock price data is too extreme for the calculation of market to book ratio.

4. Total Assets is total assets consist of current assets and fixed assets. Expected to be positively related to leverage (H4). For when the IPO, it should be an increase in equity and the assumption of fixed rate debt. The increase in equity capital due to the addition of new shareholders in turn will increase the size of the company as measured by total assets. 3.5. Model Analysis Benchmark study is a model Dahlan (2004) which still refers to the model of Baker and Wurgler (2002). The reason of this study is the model of Baker and Wurgler (2002) is cited by many research groups such as Susilawati (2008) and Saad (2010). Additionally OLS model of Baker and Wurgler (2002) which is suitable for simpler because of short data period especially 2 years or less. Analysis model appears as follows:

BL t = 0 + 1(M/B)t-1 + 2 PPE t-1 + 3 EAT t-1 + 4 TA t-1+ ML t = 0 + 1(M/B)t-1 + 2 PPE t-1 + 3 EAT t-1 + 4 TA t-1+

(1) (2)

Where = BL = Book value leverage is expressed as the difference 9

ML = Market value of Leverage which is also expressed in the difference M/B PPE EAT TA = Market to Book Ratio = Net Property, Plant and Equipment = Earnings After Tax = Total Asset

As is known in models 1 and 2, then H1 until to H4 in order to be accepted then the value of each coefficient 1<0; 2<0; 3<0 and 4>0. Also statistically each coefficient has a value of t-count significant at minimum level (p-value) 10%. To be used as a predictive model for capital structure decisions in the future, then the models 1 and 2 also have to pass the test classic assumptions.

In the preliminary study results of testing with the absolute value is not very satisfying. Then this study decided with the difference of such value. Since taking the difference to BL and ML as the dependent variable, then all free variables model (1) and (2) is expressed in the lag (t-1).

10

4. 4.1

ANALYSIS RESULT Descriptive Statistics Based on table 2, then almost all the important variables in the model have unique characteristics. BL and ML have different characteristics. Market value leverage is generally higher than book value leverage. This is in harmony with the explanation Kusumawati and Danny (2006) on two factors determining the deduction from total equity and market capitalization value-adding factors. Causes of negative values in the book and market leverage due to decreased levels of debt in the sample which implies acceptance of the hypothesis MTT. Most of the MTT with equity financing. So Saad (2010) refer to as the Equity Market Timing MTT (EMT). {Table 2 here} Independent variables such as M/B, PPE, EAT and the TA has the characteristics of the data that is standard deviation which larger than mean. This is because the problem of sample data to the extreme. Extreme data samples are potentially affect the results of testing the hypothesis but the impact will be small if the number of samples above 30 as a condition to condition i.i.d (independent and identically distributed). To further test the hypothesis is valid prior to analysis, it is necessary to check the correlation between independent variables in Table 3. {Table 3 here} Based on these tables, see the variable delta market leverage has a negative close relationship with the market to book ratio is significant at the level of 1% as an early indication of the enforceability of MTT. Also visible is variable and EATt PPEt-1-1 also has a negative relationship with leverage market delta, although not significantly. This indicates the initial support of the H2 and H3, which means equity financing in accordance MTT applies when negative earnings growth and companies are not urgent to invest in fixed assets. But these findings are not supported by the fact that the total assets negatively correlated with delta market leverage, although it is not the case for the delta book leverage.

Result of Hypothesis Testing If we follow the table 4, the apparent acceptance of H1-H4 will tend to be oriented on the market leverage. In models 1 and 2 the coefficient of market to book even positive, it is rejecting the hypothesis MTT. Although to a variable of fixed assets and total assets instead provide results that support H2 and H4. While in models 3 and 4 then received H1-H4. These results support Dahlan's study (2004); Kusumawati and Danny (2006) and Susilawati (2008) and of course Barker and Wurgler (2002). So for the fourth time MTT proved valid in IDX. But you need to know is the market to book ratio (t-1) would be more fit with market leverage as compared to book leverage. This is the reason the main components of the calculation of market leverage is the market value of total assets. Saad (2010) states the terms of enforceability from EMT (Equity Market Timing) are a factor of market sentiment to be inherent in the market value of total assets. That means investors in capital markets more able to control the optimal market leverage than optimal book leverage. When the price of the stock is overpriced (over-valued) then the market value of total assets increased sharply which makes investors reluctant to buy shares of the company. If the project is very urgent funding needs, then the debt-financing alternatives will be the best choice or the manager if the company has retained earnings of the fund may also use internal equity. Only alternative

4.2.

11

is not consistent with MTT, MTT relevant today because the stock is under-valued. So the determination of over-valued and under-valued will show the validity of the optimal leverage will be an EMT or not. {Table 4 here} In econometrics, model 3 clearly "not feasible" because of multicollinearity between TA(t-1) and PPE (t-1) with VIF limits more than 10. So this study tested again by dropping the model 4 the PPE(t-1) and TA(t-1). The result is same ie H1 and H3 are still acceptable and multicollinearity have disappeared. Actually, to overcome the multicollinearity is a proxy of total assets in model 3 can be replaced with the sale. With so only models 3 and 4 that can be used as a basis for proving MTT, i.e. when market leverage is used as a proxy capital structure. 4.3. Result of Hypothesis Testing There are two models of the study Kusumawati and Danny (2006) and Dahlan's study (2004). Kusumawati and Danny (2006) managed to observe MTT in Indonesia with a data sample of 400 observations from 1991 to 2001 for non-financial companies. GMM models presented in the form as follows (a black mark for the significant variables): BL = 0.0197 (M/B)t-1 0.0473 (EFWA M/B) t-1 + 0.0048 (PPE/A)t-1 0.1274(EBITDA/A)t-1 + 0.0631 ln (A) t-1 0.0019 (S/A) t-1 - 0.4537 (NWCA) t-1 + 0.0698 DUM k ML= 0.0306 (M/B)t-1 0.2946(EFWAM/B)t-1 + 0.108(PPE/A)t-1 0.0836 (EBITDA/A)t-1 + 0.0844 ln (A) t-1 0.0061 (S/A) t-1 - 0.2291 (NWCA) t-1 + 0.0265 DUM k With this GMM model, Kusumawati and Danny (2006) succeeded in proving the persistence of the effect of MTT even if only for short-term (1991-1995) and (1997-2001). While studies Dahlan (2004) successfully introduced the effects of MTT on the Stock Exchange for non-financial firms (1990-2000). Bleak as Kusumawati and Danny (2006), then Dahlan (2004) also use dummy variables crisis and GLS models. Only important difference is Dahlan (2004) emphasized the variable market leverage is not in the magnitude (level) but the difference. The results of the study equation Dahlan (2004) appears as follows (black mark for the significant variables): LEVt = - 0.533(M/B)t-10.098 PPE t-1 0.418 EBITt-1 + 9.503 SIZEt-1 0.294 LEVt-1 LEVt = -0.51(M/B)t-10.11 PPE t-10.418 EBITt-1 + 10.414 SIZEt-10.283 LEVt-1 1.192 DCris t-1 Based on these two studies both Dahlan (2004) and Kusumawati and Danny (2006, then the MTT applicable to BEI. But there are challenges that arise in subsequent research is to find the effect of interaction between the crisis dummy independent variable in the MTT test and try a simpler model if it is the number a little sample. Because crisis dummy is not proven as a determinant of leverage on Dahlan (2004) and Kusumawati and Danny (2006) it is necessary an alternative model for robustness check. The importance of this model to test the effect of persistence of MTT as Baker and Wurgler (2002) using samples IPO companies.

12

Alti (2003) highlights the companies that go public will be exposed to the phenomenon of long-term market underperformance that is a decrease in performance (stock price) because more investors are making a sale of shares. And as a result the company cannot conduct an EMT again. The model for this robustness check GLS model is adopted from model 3. Results of the Data Model for Random Effect Model 3 Robustness Check Based on the simulation model with the GLS random effects in Table 5 looks all the independent variables including the M/B(t-1) as evidence of significant MTT. That way the results of MTT testing hypotheses with GLS are better than OLS testing on the table 4. This indicates that testing MTT Indonesia for more relevant data by GLS, because it is so varied data leverage, market to book ratio, EAT, PPE and total assets between samples non IPO company that makes the difference between individual between samples is less can be captured by the OLS. Then some further research MTT post Baker and Wurgler (2002) recommend the GLS model of fixed effects, random effects, GMM (General Methods of Moment) to SUR (Seemingly Unrelated Regression). GLS models are capable of lowering the level of autocorrelation which was still frequently encountered in the OLS model, making a low adjusted-R2 as shown in the table 4. With the GLS random effects model in Table 5, in addition to all the components of MTT has worked well so with H1 until H4 is received then the individual effects when tested proved to 78.57% of samples met the conditions for MTT. This finding is one of the GLS models of excellence that is capable of sorting out individual samples which are aligned with MTT and which ones are not. And the test results of this study are still valid because it matches with the MTT more than 50%. Suitability of the MTT is decreased market leverage. This is the argument when market leverage is associated with a market to book ratio, the market leverage should be low in order to meet the conditions of MTT or EMT. {Table 5 here} CLOSING NOTES Conclusion Based on these studies, it appears that the OLS-model hypothesis MTT was successfully received. This means that this gives a wide space for other investigators who want to try the sample period and different industries. The study also looked at the issue of familiarity panel data regression is also a consideration. This model of this study admits better but with a long sample period. Then associated with the acceptance of this hypothesis MTT Similarly, this study observed that IPO firms are viewed as a sample of MTT hypothesis. There is one unique character of the company's 2008-2009 IPO stock price data that is likely to be reduced after the IPO. The phenomenon that occurs is the new investors will make profit-taking; with the assumption that debt levels will rise and stock prices is one component of market to book then clearly the relationship between market to book and leverage will be negative. The unique character of the market turns again to book also influenced by the EAT which gives a negative effect on leverage. The phenomenon that occurs is the medium-term post-IPO underperformance. The bottom line profits will tend to decline even in the first year after the IPO tend to rise. The cause of decline in earnings is management action to pay the debt interest bill or fund specific projects. It can be found in the case of pre-emptive rights. 5. 5.1. 4.4.

13

Thus the general purpose of the research to prove the validity of the MTT Baker and Wurgler (2002) was achieved for the case of IPOs (2008-2009) on the Stock Exchange. The results of this study support the findings of Dahlan (2004), Kusumawati and Danny (2006), Susilawati (2008) and Saad (2010). When the analysis is a special purpose, all determinants of market leverage both market to book, PPE, EAT and TA have a significant effect both when tested with OLS and GLS with high R2 values (above 70%). This signifies the MTT model validation Baker and Wurgler (2002) means that all four independent variables as determinants of market leverage can be used in a variety of sample conditions on the condition that the company is doing an IPO. If the company does not do an IPO, it is necessary to add a relevant determinant variables such as financial constraints as measured by the Kaplan-Zingales Index [see study Saad (2010)]. When in the conditions of financial constraint, then the negative market sentiment, so it is better to delay EMT until the company into a positive sentiment. 5.2. Implication 5.2.1. Theoretical Implications One of the biggest problems in testing the MTT is dependence on the sample of firms that conduct an IPO of shares. This study had a discussion with peers and get the argument that the IPO shares is a unique phenomenon of the company with the cheapest fee according to the perspective of investors because it is usually an IPO during the growth phase of the PLC. This would support the view EMT, due to the number of funds that will achieve the best level of stock prices are set by the underwriter and management. And validity of this company EMT occurs when a company is successful IPO process (eg Krakatau Steel on November 10, 2010) and long-term post-IPO performance of companies will remain good (long-run performance out). If conditions are met then the assumption of the persistence of the MTT will remain in effect so that worries Alti (2003) regarding the inability of long-term MTT test can be neutralized because there are many other studies failed MTT. If the company will conduct a second-IPO stock or bond IPO even then whether EMT or MTT still relevant? This could be explained that the EMT is dependent on the negative relationship between market leverage and market to book ratio. If the context of debt, then the only way to stay behind to do that is a positive relationship between market leverage and market to book ratio. It's just a slightly modified MTT essence of the company will add leverage when market to book ratio is still low. The addition of this leverage will not change the context of EMT where optimal market leverage is still relevant, in this case the management company is still following the static trade-off theory. When a context switch to second-IPO shares of the context of EMT will still be oriented at a negative relation between market leverage and market to book ratio, just do not as strong as the power level of IPO shares. This is because the market to book ratio is no longer a minimum, then the company cannot lower market leverage lower or increase the equity ratio is even greater because it will penetrate the optimum level that will make the second-IPO will be less successful. From this argument then the dependence on IPO shares to the context of EMT can be overcome by transfer of bonds or a second sample of IPO-IPO stock market as long as leverage remains optimal can be maintained at any time by management. 5.2.2. Managerial Implications When it was learned that the EMT or the MTT can be done not only before the IPO shares, then the main task of management is to make the process of adjustment at any time to set the optimal market leverage. This adjustment process is very important because it

14

determines the level of market to book ratio that would still have the negative relationship with market leverage. One application of the theory of post-MTT can be used for dynamic capital structure that actually has the initial idea to seek the optimum level of debt arising from the tax reduction benefits and potential costs of bankruptcy by looking at various parameters such as the condition of the company free cash flow, ownership structure and policy strategies long-term investment. Dynamic adjustment process in capital structure will certainly strengthen the essence of EMT because it will provide information to management moment when should set back the level of a negative relationship between market leverage and market to book ratio, especially after the post IPO. If this is an important prerequisite for an IPO, the company can use the initial essence of this theory that the optimum level of book value leverage derived from the initial idea of static trade-off theory (STT). 5.3. Limitation There are five limitations of this study are: First; study did not wear a long period of data such as MTT studies in the USA are on average over 20 years. This is because the study was to focus the uniqueness of data 2008-2009. Both with respect to its length are not the study period, so this study could not have used panel data regression model (GLS). For this study, this study tries OLS method based argumentation "parsimony". Third, the study did not use the data because of the financial sector leverage its behavior is different from ordinary companies and highly regulated government rules. Fourth, the company that the sample should not be exposed to the effects of the global financial crisis that is a company that has no export sales division or have a payable on the imported raw material components. Fifth, several independent variables determinant of leverage to prove the hypothesis of MTT was derived from previous studies Baker and Wurgler (2002), Dahlan (2004), Danny and Kusumawati (2006), Susilawati (2008) namely market to book ratio and financial ratios. Variable financial constraints Saad (2010) has not been discussed because it would bias the analysis of samples using an air-IPO companies. Moderate IPO itself occurs in the phase of growth. 5.4. Suggestion There are two things namely the first, the sample period should be extended to test the effects of the persistence of the MTT. Because this is so intensively "attacked" by Alti (2003). Persistency effect is one of the characteristics that have emerged from the MTT. Only problem is that this effect can only be maintained in the short term, while the capital structure decisions (MTT) is a long-term decisions. Presumably the use of GMM Kusumawati and Danny (2006) may be a solution with a time unit record data is the minimum quarterly. Second, testing in several industrial sectors and groups of dummy interaction effects of the global financial crisis in the USA in every industrial sector. The importance of seeing the effects of this crisis is to analyze the truth of the market sentiment from global investors in addressing the effectiveness of the MTT of a company. They should not see any financial information to estimate the market value to book but also look at non-financial information such as the CGPI (Corporate Governance Perception Index) and using CSRI (Corporate Social Responsibility Index).

15

REFERENCES Adedeji, A. (1998), Does the Pecking Order Hypothesis Explain the Dividend Payout Ratios of Firms in the UK? Journal of Business Finance & Accounting Vol. 25 No.9-10 pp. 1127-1155. Allen, D.E. (1993), The Pecking Order Hypothesis: Australian Evidence, Applied Financial Economics, Vol. 3 No.2, pp. 101-112. Alti, A. (2003), How Persistent Is the Impact of Market Timing on Capital Structure, Working Paper from University of Texas Austin, pp. 1-35. Baker, M. and R. Wurgler (2002), Market Timing and Capital Structure, Journal of Finance 57, pp. 1-32. Baskin, J. (1989), An Empirical Investigation of the Pecking Order Hypothesis, Financial Management, Vol. 18 No.1, pp. 26-35. Bayless, M.E. and J.D. Diltz (1994), Securities Offerings and Capital Structure Theory, Journal of Business Finance & Accounting Vol. 21 No.1 pp. 77-91. Berk, J. and P. De Marzo (2007), Corporate Finance, Pearson International Edition, Chapter 14 dan 15. Chirinko, R. and A. Singha (2000), Testing Static Trade off Against Pecking Order Models of Capital Structure: A Critical Comment, Journal of Financial Economics Vol. 58, pp. 417 425. Dahlan, I.O. (2004), Market Timing dan Struktur Modal: Studi pada Perusahaan Non Keuangan Tercatat di BEI, Tesis S2 PSIM UI. Donaldson, G. (1961), Corporate Debt Capacity, Division of Research, Graduate School of Business, Harvard University, Chapter 1. Elliot, W.B., J.K. Kant and R.S. Warr (2004), Further Evidence on the Financing Deficit: The Impact of Market Timing, Working Paper from Oklahoma State University, pp. 1-32. Fama, E. F., and K. R. French (2002), Testing Tradeoff and Pecking Order Predictions About Dividends and Debt. Review of Financial Studies Vol. 15, pp. 1-33. Frank, M.Z. and V.K. Goyale (2003), Capital Structure Decisions, Working Paper from www.ssrn.com,pp. 1-56. Graham, J.R. and C.R. Harvey (2001), The Theory and Practice of Corporate Finance: Evidence from the Field, Journal of Financial Economics 60,pp. 187-243. Hogfeldt, P. and A. Oborenko (2005), Does Market Timing or Enhanced Pecking Order Determine Capital Structure? Working Paper from Stockholm School of Economics, pp. 1-48. Hovakimian, A. (2005), Are Observed Capital Structure Determined by Equity Market Timing? Working Paper from Baruch College, pp. 1-45. Huang, R. and J.R. Ritter (2005), Testing the Market Timing of Capital Structure. Working Paper from University of Florida, pp. 1-44. Kant, J.K. (2003), Valuation Errors at the Time of Security Issuance dan the Market Timing Theory of Capital Structure, Doctoral Dissertation from Oklahoma State University, pp. 1-123. Kayhan, A. and S. Titman (2005), Firms Histories and Their Capital Structure. NBER Working Paper, pp. 1- 51. Kusumawati, D. dan F. Danny (2006), Persistensi Struktur Modal Pada Perusahaan Publik Non Keuangan yang Tercatat di BEI: Pendekatan Market Timing dan Teori Struktur Modal Optimal, Jurnal Ekonomi STEI 15 (32), hal. 1-24. Leary, M.T. and M.R. Robert (2005), Do Firms Rebalance Their Capital Structure?, Journal of Finance, Vol. 60 No. 6, pp. 2575-2619.

16

Lemmon, M.L., M.R. Robert and J.F. Zender (2005), Back to the Beginning: Persistence and the Cross-Section of Corporate Capital Structure, Working paper from University of Colorado, pp. 1-40. Liu. L.X. (2005), Do Firms Have Target Leverage Ratios? Evidence from Historical Market to Book and Past Returns, Working Paper from Hongkong University of Science dan Technology, pp. 1-48. Mahajan, A. and S. Tartaroglu (2007), Equity Market Timing and Capital Structure: International Evidence, Working Paper from Texas A dan M University, pp. 1-32. Manurung, A.H. (2004), Teori Struktur Modal: Sebuah Survei, Manajemen dan Usahawan Indonesia Vol. 33. No.4, hal.20-25. Miglo, A. (2010), The Pecking Order, Trade-Off, Signaling and Market Timing Theories of Capital Structure: A Review, Working Paper from University of Bridgeport, pp. 1-26. Myers, S. C. (1984) The Capital Structure Puzzle. Journal of Finance, 39, 575-592. Pangeran, P. (2004), Pemilihan Antara Penawaran Sekuritas Ekuitas dan Utang: Suatu Pengujian Empiris terhadap Pecking Order Theory dan Balance Theory, Manajemen dan Usahawan Indonesia Vol.33. No.4, hal. 27-36. Saad, M.D.P (2010), Pengaruh Sentimen Investor dan Kendala Keuangan Terhadap Equity Market Timing, Disertasi PPIM FE-UI, bab I. Shyam-Sunder, L., and S. Myers (1999), Testing Static Tradeoff Against Pecking Order Models of Capital Structure. Journal of Financial Economics, 51, 219-244. Susilawati, C.E. (2008), Implikasi Market Timing Pada Struktur Modal Perusahaan, Makalah Bahan Presentasi 3rd The Doctoral Journey of Management at Crowne Plaza Hotel Jakarta, hal. 1-15. Tobing, L.R. (2008), Studi Mengenai Perbedaan Struktur Modal Perusahaan Multinasional Dengan Perusahaan Domestik yang Go-Public di Pasar Modal Indonesia: Perspektif Teori Keagenan dan Teori Kontijensi Dalam Mengoptimalkan Struktur Modal Perusahaan, Disertasi Program S3 Ilmu Ekonomi UNDIP, hal. 1-26. Vasiliou, D. and N. Daskalakis (2007), Behavioral Capital Structure: Is the Neoclassical Paradigm Threatened? Evidence from the Field, Working Paper from Hellenic Open University, pp. 1-31. Wagner, H.F. (2007), Public Equity Issues and the Scope of Market Timing, Working Paper from London Business School, pp. 1-59. Welsch, I. (2004), Capital Structure and Stock Return, Journal of Political Economy Vol. 112, pp. 106-131.

17
PeckingOrder Theory Modigliani-Miller Theorem

Capital Structure Theory Pattern (Main Stream)

Market Timing Theory

StaticTradeOff Theory

1. MTT = POT & STT ? 2. MTT = IPO sample?

Gap Theory

Figure 1. Map of Capital Structure Theories Result Analysis from many literature (2010) Table 1. Comparative between POT and STT
Dimension Assumption Core Explanatory Variable Research Model POT Preference of internal funding for fear of asymmetric information Strict dividend policy to "sticky" is usually a low dividend payout ratio Funding by looking at the sequence of the Cost of Capital Cost of Capital (COC) is cheap generally more dominated by internal financing, Profitability & stock price Stock Offering Size Baskin (1989) Bayless and Diltz (1994) Allen (1993) STT Target leverage Internal and external funding for the maximization of corporate value Funding to find the optimal leverage ratio The use of excessive debt will pose a risk of bankruptcy due to financial distress condition Business Risk Deviation from target leverage Stiglitz (1969) Bayless and Diltz (1994) Frank and Goyale (2003)

Source: Result Study (2010) from Manurung (2004) and Pangeran (2004)

18

Table 2. Descriptive Statistics [56 (28X2) observes IPO 2008-2009] Variable Mean Std.Dev 0.1501992 2.880701 3.0960674 0.1580649 0.013 0.2201024 Min -0.5699379 -16.23754 .02289079 0.000103 -0.033934 0.020481 Max 0.4485325 4.833801 17.69324 0.723647 0.041606 1.021668

-0.0178231 Book Leverage (BL) -0.8324106 Market Leverage (ML) 2.183696 Market to Book (M/B) t-1 0.0982001 PPE t-1 0.0026806 EAT t-1 0.2131082 Total Asset (TA) t-1 Source: Result Study (2010) with STATA 9.0
Table3.CorrelationamongIndependentVariables
BL BL Pearson Correlation Sig. (2-tailed) N ML Pearson Correlation Sig. (2-tailed) N M/B Pearson Correlation Sig. (2-tailed) N PPE t-1 Pearson Correlation Sig. (2-tailed) N EAT t-1 Pearson Correlation Sig. (2-tailed) N TA t-1 Pearson Correlation Sig. (2-tailed) N 1 . 56 -.192 .156 56 .101 .460 56 .144 .291 56 .312(*) .019 56 .233 .084 ML

M/B .101 .460 56 -.852(**) .000 56 1 . 56 .206 .128 56 .147 .280 56 .141 .301 56

PPE t-1 .144 .291 56 -.228 .091 56 .206 .128 56 1 . 56 .360(**) .006 56 .953(**) .000 56

EAT t-1 .312(*) .019 56 -.243 .071 56 .147 .280 56 .360(**) .006 56 1 . 56 .464(**) .000 56

TA t-1 .233 .084 56 -.152 .262 56 .141 .301 56 .953(**) .000 56 .464(**) .000 56 1 . 56

-.192 .156 56 1 . 56 -.852(**) .000 56 -.228 .091 56 -.243 .071 56 -.152 .262 56

56 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

19

Table 4. Result of Hypothesis Testing [Modification from Dahlan (2004)]

Independent Variables

0.0055903 0.0028761 (0.85) (0.45) 1.12 1.02 -0.7493058 PPE (t-1) (-1.7)* 12.92 1.782002 3.398325 EAT (t-1) (1) (2.23)** 1.44 1.02 0.6115446 Total Asset (t-1) (1.86)* 14.01 -0.0915507 -0.0332131 Intercept (-2.32)** (-1.39) 2.31* 2.78* F-Hitung 2 0.0868 0.0609 Adj-R 2.36 2.188 D-W Source: Result Study (2010) with STATA 9.0 Market to Book (t-1)

Model 1 (dependent variables: Book Leverage t )

Model 2 (dependent variables: Book Leverage t )

Model 3 (dependent variables: Market Leverage t )

-0.7463752 (-10.92)*** 1.12 -7.802858 (-1.71)* 12.92 -40.31233 (-2.18)** 1.44 5.904876 (1.74)* 14.01 0.4133692 (1.01) 39.07*** 0.7346 1.805

Model 4 (dependent variables: Market Leverage t )

-0.7771906 (-11.76)*** 1.02 -26.98061 (-1.72)* 1.02 0.9370608 (3.8)*** 75.12*** 0.7294 1.866

20

Table 5. Result of Hypothesis Testing using Random Effect

Dependent Variable: ML Method: GLS (Variance Components) Date: 01to03to11 Time: 02:08 Sample: 1 2 Included observations: 2 Number of cross-sections used: 28 Total panel (balanced) observations: 56 Variable C M/B t-1? PPE t-1? EAT t-1? TA t-1? Random Effects _AD--C _AK--C _AL--C _AN--C _AR--C _AS--C _BE--C _DA--C _DH--C _DY--C _FO--C _GO--C _IN--C _IO--C _JA--C _KA--C _KI--C _KR--C _LA--C _LP--C _ME--C _PL--C _RI--C _SU--C _SY--C _TP--C _TU--C _WA--C GLS Transformed Regression R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat Unweighted Statistics including Random Effects R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat 0.853762 0.842292 1.144560 2.429771 Mean dependent var S.D. dependent var Sum squared resid -0.826750 2.882119 66.81090 0.815696 0.801241 1.284917 1.927935 Mean dependent var S.D. dependent var Sum squared resid -0.826750 2.882119 84.20157 Coefficient 0.388169 -0.795363 -9.757771 -41.12151 7.468921 -0.386628 0.121852 0.149593 1.788932 -0.017650 -0.463395 -0.007910 -0.091657 -0.513837 -0.002466 -0.297117 0.023260 0.120700 -0.285217 -0.106931 -0.147442 -0.107705 -0.111906 -0.109086 1.061136 0.294825 -0.076533 -0.304147 -0.102310 0.546990 -0.458469 -0.188501 -0.328380 Std. Error 0.442444 0.069572 4.704509 19.12292 3.557510 t-Statistic 0.877331 -11.43219 -2.074132 -2.150378 2.099480 Prob. 0.3844 0.0000 0.0431 0.0363 0.0407

Hypothesis Testing H1 [M/B t-1] ok (MTT) H2[ PPE t-1] ok H3[EAT t-1] ok H4[ TA t-1] ok

AD [ ML] ok AK [ ML] not ok AL [ ML] not oke AN [ ML] not oke AR [ ML] ok AS [ ML] ok BE [ ML] ok DA [ ML] ok DH [ ML] ok DY [ ML] ok FO [ ML] ok GO [ ML] not ok IN [ ML] not ok IO [ ML] ok JA [ ML] ok KA [ ML] ok KI [ ML] ok KR [ ML] ok LA [ ML] ok LP [ ML] not ok ME [ ML] not ok PL [ ML] ok RI [ ML] ok SU [ ML] ok SY [ ML] not ok TP [ ML] ok TU [ ML] ok WA [ ML] ok

Individual's evidentiary effect in MTT if the coefficient is negative intercept of each company, which means a decrease rather than increase ML. MTT proved valid as many as 22 companies from a total of 28 or approximately 78.57%.

ModelValidation: a. HighlyR2 b. TolerantDW(2).