THE EFFECT OF CAPITAL STRUCTURE AND FINANCIAL STRUCTURE ON FIRM PERFORMANCE (An Empirical Study of The Financial Crisis 2008 and 2009 in Indonesia)

This research aims to identify the impact of capital structure on Indonesian firms’ performance, particularly on the magnitude of impact at the period prior to crisis, crisis, and the period following the crisis that happened in 2008. The Global Financial Crisis grants a chance to scrutinize the impact of crisis between capital structure and firm performance. Proxies used for capital structure are total debt to total assets, short-term debt to total assets, and long-term debt to total assets ratio. Moreover, firm performance is measured by accounting performance (Return on Asset and Return on Equity) and market performance (Price to Equity Ratio and Tobin’s Q). Samples used include all firms listed in Indonesia Stock Exchange (IDX) from the period 2004 up to 2017, excluding financial sector firms. This research posits that capital structure generally impacts firm performance negatively. The Global Financial Crisis (GFC) that happened in 2008 serves a greater negative impact of capital structure to firm performance than it is before and after crisis. This research is intended for use by firms as a perusal in managing its capital structure, for creditors in managing its lending, and for investors in investing, prominently in times of financial crisis.

Kata kunci: Indonesia; krisis; performa; struktur modal INTRODUCTIONS Generally, every company has a similar objective, which is to get the maximum profit and increment the value of the company (Chowdhury and Chowdhury 2010). Profitability and value of a firm is an indicator that shows how well a firm performance is. To accomplish the objective, the firm needs funding from internal and external sources. Internal funding comes from retained net profit after taxes, while external funding may come from the issuance of new shares, proposing for debt to financial institutions or by issuing bonds (Al-Taani 2013). The combination of internal and external funding is known as capital structure.
Each funding option selected certainly has its advantages to be earned and its costs that must be borne by the company. This research will focus on the further explanation of the effect of the utilization of debt on firm performance. Companies gain an advantage from using debt, it is the increase in free cash flow that comes from reduced tax payments for the interest cost is allowed to be deducted in calculating taxable income (Modigliani and Miller 1963). However, referring to the trade-off theory by Modigliani and Miller (1963), funding by debt with high interest will also decrease firm's income thus lower firm profitability. This condition occurred when the benefit of the debt can no longer cover the cost, thus lower the company performance. Subsequently, debt can also act as a control for management in decision making (Jiahui 2015). Management acts as a decisionmaker so it is usual for them to have more in-depth information about the company in comparison to the owner of the company. The separation of functions between management and company owners results in different interests to be achieved and the management tends to act on the behalf of themselves in order to maximize their own wealth. One way to prevent this from happening is to utilize debt (Jensen and Meckling 1976).
The decision of using debt means comes with interest costs that must be borne by the company. Excessive interest costs will pose a default risk, which will adversely affect the company's profitability (Abor 2005). The utilization of debt becomes optimal when the benefits derived from using debt are proportional to the costs to be borne (Modigliani and Miller 1963). Another factor in the form of management activities in managing investment also plays an important role so that the debt can be worthwhile to improve firm performance.
Research conducted by Sheikh and Wang (2013) proves that the utilization of debt has an effect on diminishing firm performance due to the higher interest costs, thus management have to be more cautious in selecting investments. Unlike the results conducted by Zeitun and Tian (2007), the utilization of short-term debt will pose lower interest costs compared to long-term debt. Companies can use short-term debt to increase sales growth. The higher sales growth, the firm performance will likewise increase.
In crisis conditions, the management of capital structure must be given more attention because weak economic conditions can affect management decisions in acquiring funding for the company (Chang et al. 2014). The Global Financial Crisis (GFC) in 2008 began with a financial crisis that occurred in America in 2007. The Asian economy was one of those affected by the crisis through the trade sector in the fourth quarter of 2008. The condition is shown by the significant decline of export growth in Indonesia that was above 10% along with the first until the third quarter of 2008, then plummeted to just 1.99% in the fourth quarter of 2008. Rising imported products' prices do hit Indonesia even more that it poses a serious risk of inflation. In order to control inflation, Bank Indonesia took monetary policy into a measure, which was raising the interest rate (Bank Indonesia 2009).
The monetary policy action taken by Bank Indonesia in 2008 was implemented in multiple stages. The first stage of the interest rate policy was set at 8% which results in no change. As the risk of inflationary pressure increases, Bank Indonesia decided to raise the interest rate to 9.25% (Bank Indonesia 2009). The increase in the interest rate, in turn, raised savings and prime lending rates. The increase in the prime lending rate affects managements' decision in managing the capital structure. Overall, the effect of global financial crisis in Indonesia were showed by the rise of inflation rate, lower economic growth rate, weaker exchange currency value, and loss suffered by Indonesian firms that invested in US firms (Darajati and Hartomo 2015). Khodavandloo et al. (2017) research posit that as a company's level of leverage increases, it decreases firm performance. Besides, the research also does compare the magnitude of negative impact in the period prior to the crisis, crisis, and post-crisis that happened in 2008 on Malaysian listed firms. It is found that capital structure negatively and more severely impacted firm performance in the crisis than it is in pre-crisis and post-crisis. It is due to the sudden increase in the interest rate from a stable level of 2.8% since 2004 and then increased to 3.5% which occurred at the beginning of the crisis in early 2007 in Malaysia. The rise in the interest rate, in turn, contributed to the rise in the prime lending rate. Rising prime lending rate negatively affect firms' profitability, so do its performance. Khodavandloo et al. (2017) finding supports Chang et al. (2014) finding on its research which was conducted on Vietnam listed firms. Deteriorating capital market conditions resulted in companies relying more on the use of debt. Relying on using debt during a crisis that coincides with hyperinflation conditions results in higher interest costs to be borne and lowering firms' profitability that in turn deteriorates its performance.
Overall, the results of previous researchs are still inconsistent. Moreover, the research regarding impact of capital structure on firm performance is still limited in Indonesia, especially for research on the crisis period. Therefore, this research will further investigate the impact of capital structure on firms' performance, particularly its impact on three different periods (precrisis, crisis, and post-crisis).
This research aimed to examine the negative effect of capital structure on firm performance and to examine the differences in the negative effect of capital structure on firm performance in the period before, the crisis period, and the period after the Global Financial Crisis 2008.

LITERATURE REVIEW AND HYPOTHESIS FORMULATION
Generally, there are three ways to obtain funding for a firm. Those are funding from appropriated retained earnings, debt, and issuing new shares Myers and Majluf (1984). The funding for the company is also known as the capital structure. Modigliani and Miller (1958) then stated in their tradeoff theory that firm can achieve the optimal level of leverage with the existence of debt. Moreover, when debt occurs, the creditor will also be overseeing the firm along with agent and principal, thus help lower the agency problems between agent and principal (Jensen Meckling 1976). Shortly, the existence of debt will generate the cost of debt or the interest. This interest can be optimized to lower the taxable income, thus reduce the tax payment. However, when the cost of debt is too high, it will lower the firm performance.

Trade-Off Theory
Trade-off theory is the development of irrelevant theory which was previously found by Modigliani and Miller (1958). It is assumed on the trade-off theory that taxation is present in the real business conditions. With taxation taken into account, there should be tax benefits obtained from the utilization of debt. That benefit is the addition of cash flow derived from the reduction of tax payments. Tax payments reduction is derived from interest expense coming from the use of debt that is included as a deductible expense in the calculation of fiscal profit.
Balance utilization of debt and equity funding will lead to higher company value. One component of the calculation of company value is the company's ability to generate profits, which is earnings before interest costs and taxes (EBIT) that indirectly describe the company's cash flow. Profit before interest and tax will ultimately affect the company's performance. The financial crisis can potentially bankrupt a firm because in general, the cost of debt will rise along with the benefits of the debt it obtains. On the other hand, an increase in the cost of debt will raise the firm's debt expense. Overall, the trade-off theory states that it is possible for a firm to achieve an optimal level of leverage between costs and interest payments that arise due to debt.

Pecking Order Theory
Pecking Order theory was initiated by Myers and Majluf in 1984. The factor that leads to the emergence of this theory is caused by the presence of rights of each party to claim returns on what each party has invested in the company itself. The decision of a company to obtain external financing will most likely cause information asymmetry problems between shareholders and management to arise. The presence of a conflict of interest between those two parties results in three financing options. Internal financing is the most secure and least risky source of financing compared to external financing. Internal financing here refers to appropriated retained earnings.
Alternatively, firms can obtain external financing by acquiring debt. However, if a firm decides to go for external financing, it should not only expect to cover operational expenses but also to meet its obligation to the third party. The obligation will include the debt and the interest. There is also a third option as stated by Myers and Majluf (1984) which is equity financing by issuing new shares. Howbeit, this option will not only pose a negative signal to shareholders for their current ownership will potentially be diluted, but also envisage that the firm might be in distress. When shareholders get a negative signal, it will lower the company value.

Agency Theory
The separation of management and ownership function in a firm causes conflict of interest between each party. Agency theory was initiated by Jensen and Meckling in 1976. Shareholders as the owner of a firm have their own rights to point the agent to make decisions in terms of investment decisions. Management as the managing party will surely get a more comprehensive view of the firm's condition so that they will most likely make investment decisions on behalf of their own-which is to invest in high risk and high return investments (Bodhoo 2009). Build upon the agency theory, this conflict of interest causes firms to incur a cost in order to alleviate the conflict, which is better known as agency costs that consist of monitoring costs, bonding costs, and residual loss (Jensen and Meckling 1976). Moreover, agency problems can be reduced with the presence of debt. When debt occurred, creditor will also have a role in overseeing the firm along with agent and principal. Hence with the addition of debt in the capital structure, it will be shown by high leverage thus lower the agency problem between agent and principal (Jensen and Meckling 1976). This will in turn increase firms' efficiency because with the presence of debt, agents will be driven by incentives to make the firm profitable. Conversely, the more debt that a firm has, the more conservative agents are in terms of making investment decisions that they will prefer the less risky with lower return options so that the firm is not faced with a lot of risks (Khan 2012). This implies that the more the debt the company has, it will cause the firm's profitability to decrease.

Financial Crisis in Indonesia
The Global Financial Crisis (GFC) during 2008-2009 that started in the United States of America managed to affect Indonesia's economy. Generally, it generated several issues. Those were rising inflation rate, low economic growth, weakened currency exchange, and loss for Indonesian companies that invested in the American firms (Darajati and Hartomo 2015). However, another research stated that implication of GFC in 2008-2009 did not worsen the economy as bad as in other emerging countries (Sugema 2012). This is due the level of Indonesia's export market share that was not more than half of its Gross Domestic Product (GDP). It also did not affect the stock market much since only 0.5% of its citizens who involved in the stock market. However, the big impact shown on the high level of poverty and unemployment especially in rural area since majority of Indonesia's export prime commodities were produced in rural area.
Moreover, Ramli et al. (2018) found some firms that mostly affected by the GFC were forced to use the external funds to helped raise their financial performance. The external funds was most likely banks loans since issuing bonds is harder in the emerging countries. However, Imadudin et al. (2014) insinuated that the non-optimal use of debt funding in Indonesian firm postcrisis does not have any impact on firm performance. This occurred because the rise of debt funding utilization was not followed by any increase in its profitability thus has no impact on firm performance.

Effect of Capital Structure on Firm Performance
Generally, in running a business, a company will certainly need funding that can be obtained either internally or externally. The combination of those funding sources in the form of debt and equity is called capital structure. There are various studies that have been conducted related to the implications of capital structure on the firm performance that generates different results. A company with a good performance means operational activities are running optimally as seen from indicators of productivity, growth, and profitability (Tan and Hamid 2016). In a study conducted by Dawar (2014), stateowned banks play a big role in lending money to undeveloped markets in India so that control over lending and the utilization of debt is not used effectively. As a result of credit disbursement not chosen according to the criteria for creditworthiness, funding through debt cannot actually improve firm performance. Le and Phan (2017) found tax discretion in certain countries, thereby limiting management decisions in managing capital structure. The utilization of debt in Vătavu (2015) is not used to develop a company but only necessary when experiencing financial difficulties. Companies will prioritize using internal funding because it is in line with the Pecking Order theorywhich external funding is riskier which can cause poor firm performance (Twairesh, 2014). That explanation is not in accordance with the statement of Modigliani and Miller (1958) which states that companies can benefit from the utilization of debt by balancing the benefits and costs of the utilization of debt. Abor (2005) and Zeitun and Tian (2007) find that the use of shortterm debt in the capital structure has a positive influence on firm performance. This is because short-term debt has relatively low-interest costs so that it can be used to increase company growth. In addition, Ofek (1993) found that companies with high levels of debt tend to make the company disciplined and respond quickly if there is a decrease in performance. Thus companies with high debt levels tend to maintain the going concern of the company. The formation of Hypothesis 1 is based on the research of Le and Phan (2017) because Indonesia also has tax regulations similar to those applied in Vietnam. The limitation to the reduction of deductible expense that is regulated based on the ratio of debt to capital causes the benefits derived from the utilization debt is limited. The usage of debt that exceeds the amount of benefit causes the use of debt will negatively affect the firm's performance. Therefore, the hypotheses formed are as follows: H1: There is a negative influence between capital structures on firm performance

Effect of Capital Structure on Firm Performance in the Pre-Crisis Period of Global Financial Crisis 2008
Quite the same with the explanation on the development of the previous hypothesis, this hypothesis will focus on the influence of capital structure on firm performance in the pre-crisis research period. Before the crisis-affected global economic conditions, interest rates were relatively low and the company had no difficulty in fulfilling obligations related to debt repayment. In the research of Khodavandloo et al. (2017), companies lean on debt funding for operational activities. Although interest rates in Malaysia in the pre-crisis period were stable, the company used a large portion of the debt. When compared with the trade-off theory, the relatively large usage of debt in the period before the crisis will cause companies to gain smaller benefits because the use of large amounts of debt raises the number of interest costs that must be borne by the company. The amount of interest costs that must be borne by the company causes the firm's performance to be negatively affected. Similar results were also found in studies conducted by Chang et al. (2014). In the study of Chang et al. (2014), the existence of capital market limitations and underinvestment problems resulted in companies in Vietnam relying on debt and had to deal with hyperinflation conditions resulting in costs incurred on debt interest exceeding the limit of benefits the company could receive. While research from Hossain and Nguyen (2016) examines the effect of capital structure on firm performance with a focus only on the oil and gas sector in Canada showing external funding in the form of debt plays an important role for capital-intensive industries for investment in machinery and technology. The use of debt is increasingly used especially if the excavation source is found because the company needs funding to complete the excavation project. The use of large debt in the pre-crisis period caused the firm's performance to decline. The formation of this hypothesis is based on research by Khodavandloo et al. (2017) which has similarities in terms of interest rate movements before the crisis. In addition, in Indonesia at the time before the crisis did not occur hyperinflation as in the study of Chang et al. (2014) and affected by the movement of crude oil prices as in the research of Hossain and Nguyen (2016) because this research examines many sectors. Therefore, the hypotheses formed are as follows: H2a: There is a negative effect between capital structure on firm performance in the period of pre-crisis. Modigliani and Miller (1963) state that to achieve an optimal capital structure, there must be a balance between the usage of debt and equity. One of the advantages of using debt is the interest expense that can be used as a deductible expense, thus giving the company more cash flow due to reduced tax payments (debt-tax shield). The trade-off theory proposed by Modigliani and Miller (1963) also states that there is a maximum point between the benefits obtained from the debt-tax shield and the bankruptcy costs that can arise due to large interest costs. In addition, Ofek (1993) found that companies with high debt, when experiencing difficulties tend to immediately take operational actions such as restructuring assets and laying off employees or financial actions such as cutting dividends and debt restructuring. This helps companies avoid losses in the long run. Thus a high level of debt can help maintain the company's going concern.

Effect of Capital Structure on Firm Performance on Firm Performance in the Crisis Period of Global Financial Crisis 2008
Khodavandloo et al. (2017) in his research on publicly listed companies in Malaysia shows that just before the crisis, there was a significant increase in the benchmark interest rate. The increase in interest rates causes loan interest rates to be higher, thus demanding companies to pay higher interest costs. In addition, research by Khodavadloo et al. (2017) is also supported by research conducted by Chang et al. (2014) in a public company in Vietnam. In times of crisis, the capital market in Vietnam experienced a slump that made it impossible for companies to get capital from the capital market. Companies rely on debt from banks as a source of funding. Meanwhile, Vietnam also experienced hyperinflation during the crisis so that interest rates fluctuate. The fluctuating interest rates cause interest costs to be borne by the company to be greater so that it becomes a burden for the company. An increase in the company's interest and principal expenses in times of crisis will have a greater negative effect on performance compared to pre-crisis.
Research conducted by Hossain and Nguyen (2016) on companies in Canada engaged in the oil and gas sector shows that the negative effect of capital structure on performance is smaller during the crisis than pre-crisis. The contributing factor is Canada's strong financial condition in times of crisis, even one of the best. This is influenced by the stable price of crude oil during the crisis resulting in smaller negative effects during the crisis compared to before the crisis. Therefore, research conducted by Khodavandloo et al. (2017) has similarities with conditions in Indonesia, namely, there is an increase in interest rates during crises. Thus it was expected that the use of debt in capital structure will lower firm performance due to the rise in cost of debt, then this study has the following hypothesis: H2b: The negative effect between capital structure on firm performance in times of crisis is greater than before the crisis. Khodavandloo et al. (2017) in his research found that after a crisis, the interest rates became more stable than during the crisis and Malaysian firms became more conservative in obtaining funding using debt. This conservative attitude is shown by the debt ratio which tends to decrease in the post-crisis compared to during the crisis. The addition of the company's free cash flow causes agents to be opportunistic to take investment projects with high returns. The opportunist attitude of the agent has led Malaysia companies to back to use debt which serves as a monitoring fee so that the principal can control the decisions taken by the agent. The higher the debt used means the burden is high so the agent is conservative so the investment taken is an investment with a low level of risk and ultimately cannot maximize the rate of return on investment (Jensen and Meckling 1976).

Effect of Capital Structure on Firm Performance on Firm Performance in the Post Crisis Period of Global Financial Crisis 2008
In research conducted by Chang et al. (2014) found something similar. Chang et al. (2014) show different things about market performance. It was found that capital structure has a positive effect on market performance after the crisis. This is due to the problem of underinvestment that previously occurred now has become more controlled post the financial crisis. Post financial crisis, agents have become more opportunistic in taking investment, so investments taken are investments that have a large rate of return. The existence of a high rate of return affects the increasing performance of the firm.
In contrast to the results of research conducted by Khodavandloo et al. (2017) and Chang et al. (2014), Hossain and Nguyen (2016) found that in the post-crisis, a greater negative effect of capital structure was seen on firm performance compared to the crisis period. The major influencing factor in this research is the price of crude oil which experienced a slump post-crisis period so that the company becomes burdened with fulfilling the legality. The condition of the decline in oil prices in the Hossain and Nguyen (2016) study is different from this study so that the hypothesis was formed based on the research of Khodavandloo et al. (2017). Similar to the research of Khodavandloo et al. (2017), at the time after the crisis the interest rates in Indonesia also became more stable than during the crisis. The stable interest rate means more stabilized condition of firm cost of debt, thus the capital structure represented by debt is expected have deduction in its negative effect to the firm performance. Therefore, the hypothesis proposed is as follows: H2c: The negative effect between capital structure on firm performance in the post-crisis is smaller than during the crisis.
This research tested effect of capital structure on the firm performance during pre-crisis, crisis, and post-crisis period. The capital structure represented by three variables, those are total debt to total assets (TDTA), short-term debt to total assets (STDTA), and long-term debt to total assets (LTDTA). Meanwhile, the firm performance will be represented by four variables. Those are return of asset ratio (ROA), return of equity (ROE) ratio, price to earnings ratio (PER), and Tobin's Q ratio (TOBINS).
As an additional, there will be two control variables, which are firm size (SIZE) and sales growth (SGROW). SIZE and SGROW are chosen because in knowing sales growth and company size, it is sufficient enough to represent the calculation of the four dependent variables utilized in this research. Besides, this research is also using four independent variables with 13 years of coverage as well. Therefore, the overall control variables here are intended to support the main hypotheses only, not as the new concerned main focus of this study. In addition, several similar previous research conducted in Indonesia also add those as the control variables.

Research Model
Referring to the framework above, below are the models and the variables operationalization used in this study: Model 1.
PERFi,t = α0 + β1LTDTAi,t + β2SIZEi,t + β3SGROWi,t + εi,t  Chadha and Sharma (2015) The dependent variable, PERF, is measured by four proxies. Those are ROA, ROE, PER, and TOBINS. Below are details of the variables used in the models explained in the table of variables operationalization:

RESULT AND DISCUSSION Population and Sample
Population data used in this study are all companies listed on the Indonesia Stock Exchange (IDX) from 2004 to 2017 with the exception of companies engaged in the financial sector. Samples were collected by purposive sampling method so that the data taken as samples were taken based on specified criteria. All data used in the study were obtained from Bloomberg data. The separation of the three crisis period groups is based on the 2008 Indonesian Economic Report Book issued by Bank Indonesia. The following is a list of research sample acquisition ( Table 2).
The study uses panel data, namely research observations involving various companies and various time periods. The data used in the study are unbalanced so that the number of observations will be different for each regression result.

Statistical Results and Discussions
Based on the results of descriptive statistical, it can be seen in Table 3 that the overall usage of TDTA debt in publicly listed companies in Indonesia was highest in the crisis period. When examined further the increase in both types of debt, the descriptive results show an increase in STDTA proxy or short-term debt experienced a significant increase during the crisis period. Descriptive result of LTDTA proxies show companies tend to reduce long-term debt funding because when the crisis period interest rates increase according to the policy steps taken by Bank Indonesia during 2008.
Subsequently, ROA and ROE proxies produce ratios that increase compared to the period pre-crisis due to sales generally still shows an increase during the crisis, can be seen sales growth in Table 3. The impact of the Global Economic Crisis did not affect the entire industry but decreased the performance of companies, especially the trading industry. The increase in the ratio during the crisis period was also shown by one of the market performances, namely PER. Like the previous explanation, not all industries are affected by the global economic crisis so it can be concluded that there are still many investors entrusting their investment even though the crisis conditions might affect the firm's performance. In contrast to market performance measured by Tobin's Q proxy which shows the highest firm performance during the period after the crisis. This is because Indonesia is in a 0,51 0,80 0,00)*** 0,44 (0,1)* 0,64 (0,00)*** SGROW 0,00 0,00 -0,00 -0,00 0,00 0,00 -1,40 0,00 (0,00)*** (0,00)*** (0,00)*** 0,50 (0,00)*** (0,00)*** 0,17 (0,01)***  period of economic recovery so as can be seen in Table 3 the high average post-crisis sales growth is much higher than the crisis period.
Lastly, the descriptive statistical results of the control variables indicate that the size of the company during the study period did not have a significant difference or tended to be close to the value. Whilst the proxy for overall sales growth for the period showed that the sample companies were researched except the financial sector always experienced sales growth.
Subsequently, referring to Table 4 in the section of the pre-crisis period, it can be seen that the independent variables were able to explain dependent variables which can be shown by the result of F tests being below 0.05 along the pre-crisis period. Regression results show that Hypothesis 2a is proven when firm performance is measured by ROE. This is depicted by all capital structure variables showing negative significant relationship to firm performance which is measured by ROE during the precrisis period. The negative impact of capital structure on firm performance can be explained by the fact that firms are still not able to manage their capital structure to the point where benefits and costs are in equilibrium, as it is stated in the trade-off theory. The fact that the Indonesian Tax Regulation sets a certain cap for firms to benefit from the debt-tax shield can be one of the causes of the negative impact of debt utilization. Hypothesis 2a cannot be proven when performance is measured by using ROA and Tobin's Q for the regression result shows that the t-test is above 0.05. ROA and Tobin's Q both used total asset which is proven in the descriptive statistics that total asset variance is the largest in the pre-crisis period so that the effect of capital structure on ROA and Tobin's Q is not linear. Regression results between control variables and dependent variables in the pre-crisis period posit that firm size and sales growth positively affect firm performance. It is said that the greater the firm size and the greater the sales growth better firm performance.
The model regression test results during the crisis period resulted that the utilization of debt in the capital structure which is higher will have a significant negative effect on firm performance as measured using accounting performance. Depicted in Table 4 in the section of the crisis period, the results of the F and t-test of the overall capital structure variable show a number below 0.05 which means that the capital structure variable has a significant effect on firm performance through the ROA and ROE proxy. The results of other model regression tests that produce STDTA  proxies have a negative effect on firm performance using Tobin's Q proxy. Therefore, it can be interpreted that increasing the usage of debt in the capital structure during a crisis period will reduce firm performance or Hypothesis 2b is accepted. Moreover, this result is also supported by the regression depicted in Table 5. For this regression test, new variables pretdta, prestdta, and preltdta are generated, measured by precrisis dummy (precrisis = 1; crisis = 0) multiplied by value of tdta, stdta, and ltdta, respectively. The terms of pretdta, prestdta, and preltdta are created to test the null hypothesis H0: βprecrisis = βcrisis. Most of the results of pretdta, prestdta, and preltdta in Table 5 showed significant results and confirming the test comparison between the regression test of pre-crisis and crisis period in Table 4 that H2b is accepted. Regression test results from the firm size control variable showed a significant positive effect on firm performance with ROA and Tobin's Q proxy. Sales growth control variables in the crisis period showed a significant negative effect on company performance with ROA and Tobin's Q proxy. The results of the other control variable regression resulted in five of  Moreover, in accordance with the results of the regression model testing in the crisis period, capital structure variables have a significant negative effect on firm performance during the post-crisis period as measured through the ROA proxy. For regression model testing using ROE proxy, only STDTA and LTDTA proxies have a significant negative effect on firm performance. It was found in this study that Hypothesis 2c was accepted because it could be proven by comparing the results of the regression test of the highly significant negative influence in the crisis period. Hypothesis 2c test results are also supported by the results of regression models through market performance that is Tobin's Q. In the post-crisis period, the utilization of debt in the capital structure has a significant positive effect on firm performance measured by market performance. Additionally, this result is also confirmed from the regression depicted in Table 6. Regarding these regression tests, new variables of posttdta, poststdta, and postltdta are created, measured by postcrisis dummy (postcrisis = 1; crisis = 0) multiplied by value of tdta, stdta, and ltdta, respectively. The terms posttdta, poststdta, and postltdta examined the null hypothesis H0: βpostcrisis = βcrisis. Most of the results of posttdta, poststdta, and postltdta in Table 6 showed significant results and confirming the test comparison between the regression test of pre-crisis and crisis period in Table 4 that H2c is accepted. For the control variable regression results, company size and most sales growth showed a significant positive effect on firm performance measured through accounting and market performance. Sales growth shows six of the nine results of the proxy of sales growth regression has no influence on firm performance. It can be interpreted as sales growth in the post-crisis period is not relevant in improving firm performance.
The first and second hypotheses in this research are proven. Most of the regression results found that the utilization of debt will negatively affect accounting performance. The results of this study are in accordance with previous studies, the results of the study prove that there is a negative influence between capital structure on firm performance. In the pre-crisis period, the use of non-optimal debt caused the benefits received were not commensurate with the number of costs incurred. While in the crisis period, the impact of high debt use had a greater negative effect on firm performance than before the crisis. This is evidenced by slowing sales growth in addition to an increase in interest rates during the crisis period. For the post-crisis period, the use of debt has a significant positive effect on market performance. Moreover, the effect of debt uses on accounting performance in the post-crisis period produced a significant negative effect on the use of smaller debt compared to the crisis period.

CONCLUSIONS, IMPLICATIONS, AND LIMITATIONS
This research is intended to investigate the effect of capital structure on firm performance in Indonesia. It is also aimed that this research will compare the magnitude in pre-crisis, crisis, and post-crisis that occurred in the event of the Global Financial This research utilizes two performance approaches acting as its dependent variables comprising of accounting performance which uses ROA and ROE and market performance which uses PER and Tobin's Q as the proxies. In conclusion, capital structure significantly and negatively affects accounting performance. This result supports several studies conducted by Ahmad  Vătavu (2015) which posits that the more debt a company has, the more that firm performance will decrease because the debt-tax shield obtained is not sufficient to cover interest costs. Indonesia has a regulation that limits the deductibility of interest expense in the calculation of fiscal profit. Based on the Regulation of Ministry of Finance of Indonesia, which is outlined in Peraturan Menteri Keuangan (PMK) 169 of 2015, the rule essentially limits the amount of tax-deductible borrowing cost arising from the debt to a maximum Debt to Equity Ratio of 4:1. Any excess of the debt interest expense incurred will be accounted for as a non-deductible expense so that firms will not obtain any tax benefit from the utilization of debt. This negative impact is further supported by the fact that the majority of debt is used to fulfill working capital needs instead of investment in capital that brings more economic benefit to the firm.
In the pre-crisis period, the utilization of debt leads to a decrease in firm performance measured by ROE. It is caused by the costs incurred by the firm is superior compared to the benefits obtained from the utilization of debt. Whereas if ROA and Tobin's Q is used as the proxy, the capital structure does not affect firm performance. This can be explained by the large variation in total asset data that causes the relationship not to be linear. In the crisis period, the worsened off capital market condition causes firms to opt for debt in terms of financing, shown by the increase in leverage ratio during the crisis. According to the pecking order theory, firms will be faced with a higher risk of interest rate by using debt compared to internal financing so that firms are burdened by a higher interest cost. Not only firms should be able to meet their obligation from the utilization of debt, but firms should also allocate funds for operational purposes. It is found in this research that the usage of debt during crisis poses a greater impact on the deterioration of firm performance.
Diversely, in the post-crisis period, it is found that firms tend to show an increase in performance. The decrease in debt utilization is shown by the leverage ratio in the post-crisis period, which is inferior to that of the crisis period, which in turn causes firms to have more free cash flow. More free cash flow means that management of the firm can opportunistically make investment decisions so that the investments yield high returns that can be useful to increase firm performance. The regression results show that debt utilization in the post-crisis period posits a positive impact on firm performance measured by market performance. When firm performance is measured by accounting performance, the utilization of debt in the period after the crisis poses an inferior negative impact on firm performance than it is in a crisis period. Principal tend to use more debt to control agents as the agents are being more opportunistic. In line with the agency theory, the usage of debt will affect agents to act conservatively for they should take the high-interest cost of debt into account while the decision taken at the same time does not yield profitable returns to the principal that it causes firm performance not being maximized.
The limitation of this study is that it does not distinguish between the types of debt currencies in the three periods (before the crisis, the crisis and after the crisis) in Indonesia. Further researchers are suggested to examine this topic of the influence of capital structure management on the company performance in three periods of crisis time in Indonesia by comparing the types of debt currencies due to hedging elements.
It is expected that this research can be beneficial to firm management to optimally manage the firm's capital structure and anticipate the consequence of overutilization of debt during the crisis period. This research manages to prove that the utilization of debt during crisis leads to firms' performance being deteriorated the negative impact of capital structure on performance is the largest than that of in the pre-crisis and post-crisis. Besides, this research should come in useful as a perusal for creditors in issuing credit, therefore creditors can gain another perspective on assessing prospective debtors' ability to pay to prevent the unexpected hike in NPL ratio. It is also expected that this research will come in handy for investors as a perusal prior to opting to revoke investment. Investors are expected not to revoke their investment in a hurry in an event of crisis happening for it is proven that the utilization of debt leads to the downturn in firm performance, even worse than it is on pre-crisis and post-crisis period, which will, in turn, affects the decline in returns to shareholders.