Does economic growth moderate the effect of fundamental values on the stock return of Indonesian infrastructure companies?

* Corresponding author, email address: dodyhapsoro@gmail.com ABSTRACT In recent years, issues of infrastructure development and economic growth have become very popular topics during President Jokowi’s administration. Infrastructure development is expected to have an impact on economic growth. The purpose of this study was to examine the effect of fundamental values on the stock returns of infrastructure companies listed on the Indonesia Stock Exchange in 2015-2017 with economic growth as a moderating variable. This research uses a purposive sampling technique. The analytical method used is partial least squares with WarpPLS software version 6.0. the results show that EPS has a positive effect on stock returns while DER, PER, and NPM do not affect stock returns. Furthermore, it also indicates that economic growth does not moderate the effect of EPS and DER on stock returns. However, the results of the study prove that economic growth can moderate the effect of PER and NPM on stock returns. This research implies that government policy that sets priorities for infrastructure development needs to be supported because it is proven that the government policy has a positive effect on the profits and stock returns of infrastructure companies.


INTRODUCTION
The 2014 was a political year in Indonesia because during this period there were major political events, namely presidential elections and legislative elections. General election affects the economy, especially national economic growth. The election had an ambiguous impact (positive and negative) on the Indonesian economy. On the one hand, the election can encourage the economy but on the other hand, it slows the national economy (Pepinsky and Wihardja, 2011).
Capital markets are important instruments in the modern economy. Capital markets are considered one of the effective means to accelerate the development of a country because the distribution of capital plays a very important role in the economy development (Samimi and Jenatabadi, 2014). Capital market conditions are very volatile therefore it is difficult to predict and it makes the investors very risky. Investment activity is an activity that has risks and it is difficult to predict the rate of return. Investors when making investment decisions in the capital market, they need various kinds of information, evaluation, and consideration. The strategy commonly used by investors is using financial ratios to determine the fundamental value of shares. Financial ratios allow shareholders to compare different information in making investment decisions (Singh and Schmidgall, 2002).
In addition to considering the benefits gained from stock investments, investors also need to pay attention to the risks that they will have. In this case, stock return is a very important factor them to invest their funds in the capital market. The higher the company's stock return, the more attractive investors will be to invest. The higher the return or profit obtained, the better the position of the company owner (Aslama, 2019).
There are several studies on the analysis of the effect of fundamental stock factors on stock returns. The previous study proved that the fundamental value of shares has a significant effect on stock returns (Putra and Herawati, 2018). The fundamental values they studied are return on assets (ROA) and price to book value (PBV). Also, Jasman and Kasran (2017) stated that changes in stock returns are affected by variables ROA, PBV, earnings per share (EPS), and exchange rates. Yet, the variable debt to equity ratio (DER) proved to have no significant effect on stock returns. Different results were shown by Manoppo (2015) stating that economic value added (EVA), ROA, return on equity (ROE), and return on sales (ROS) variables both simultaneously and partially do not significantly influence stock returns. For example, Al-Qudah and Laham (2013) found that the DER variable and stock beta have a significant effect on stock returns while the ROE, PBV, and EPS variables have no significant effect on stock returns. However, the results of the Sorongan study (2016), showed that DER and beta stocks do not significantly affect stock returns.
Inconsistent research results indicate that fundamental values are not enough to be used as a basis for investors to make investment decisions. EPS, PER, DER, ROA, and other similar ratios are fundamental factors originating from the internal company. Also, there are fundamental factors that come from external companies such as economic conditions, government policies, or strategic plans that also need to be considered.
President Jokowi stated that the current national development strategic plan focuses on infrastructure development (Nasional. Kompas.com., March 27, 2017). According to Estache and Garsous (2012), infrastructure is one of the pillars supporting national economic growth. Infrastructure development is an effort to strengthen the foundation for the creation of economic growth. Infrastructure is physical facilities that are developed or needed by public agents for government functions in water supply, electricity, waste disposal, transportation, and similar services to facilitate social and economic goals (Asian Development Bank, 2016). Meanwhile, infrastructure companies are business entities that function to manage physical systems to meet basic human needs in the social and economic sphere.
The government's seriousness in building infrastructure is reflected in the increase in the allocation of the State Budget (APBN) from 2014 to 2017. Figure 1 is the data taken from the Indonesian Ministry of Finance. It shows a significant increase in the infrastructure budget in the APBN over the past three years. APBN or government budget or state budget is a document prepared by the government and/or other political entity representing their anticipated tax revenues (inheritance tax, income tax, corporation tax, import tax) and proposed spending/ expenditure (health care, education, defense), roads, state benefit) for the coming financial year. During 2014-2017 there was an increase of more than 200% of the APBN budget in the infrastructure sector. Also, in the draft data on the Financial Note RAPBN 2018, it is stated that the APBN budget in the infrastructure sector reaches 409 trillion rupiahs. The budget increase is the evidence of the government's seriousness in improving infrastructure in Indonesia.
Infrastructure development is believed to be able to drive Indonesia's economic growth to reach 6% in 2019 (Waluyo, 2018). The government seems to signal to investors to participate in the success of national development. Infrastructure is one of the factors that influence economic growth, so that it is necessary to examine the effect of the trend of infrastructure development and economic growth on stock returns in infrastructure companies.
The fundamental value of shares of infrastructure companies listed on the IDX is one of the bases for decision making for investors. The fundamental value of shares can be reflected in financial ratios. In this study, the fundamental values include EPS, PER, DER, and Net Profit Margin (NPM). However, due to the results of previous studies that show inconsistent results, the present researchers need to add economic growth variable that is proxied by GDP (gross domestic product) as a moderating variable. This is intended to understand the effect of fundamental factors originating from external companies on stock returns.
Economic growth is predicted to be able to increase the effect of the fundamental values on stock returns of infrastructure companies. Infrastructure plays an important role in increasing economic growth. Higher economic growth is found in areas with sufficient levels of infrastructure availability (World Bank, 2014). Economic growth is believed to have a positive effect on stock returns (Ritter, 2012). On the contrary, research by Bayar et al. (2014) as a whole does not show a significant effect of economic growth on stock returns.
Based on the description above, researchers are interested in examining whether economic growth moderates the effect of the fundamental values on stock returns of infrastructure companies in Indonesia.

THEORITICAL FRAMEWORK AND HYPOTHESIS Fundamental Analysis
Fundamental analysis is one method used by investors to predict stock prices in the future (Venkates et al., 2012). It is done by looking at the company's equity value based on the analysis of published financial statements and other information without reference to the price of the company's shares in the capital market (Wafia et al., 2015).
The role of financial ratios is very important in fundamental analysis. For example, financial ratios can be used to compare companies that have the same or similar activity and size (Baresa et al., 2013). Factor that can be compared, according to Abardanell and

Figure 1 Increased State Budget Per Sector
Source: Ministry of Finance Bushee (1997), is macroeconomic condition that is GDP. This also needs to be considered because they relate to fundamental values and corporate profits. Another role of fundamental values is that it can also be used to estimate the company's stock returns in the future (Abardanell and Bushee, 1998).

Framework
This study aims to examine whether economic growth moderates the effect of the fundamental value of shares on stock returns of infrastructure companies in Indonesia. The research framework is shown in Figure 2.

Hypothesis Development The Effect of EPS on Stock Returns
Earnings per share (EPS) are an important measure used by investors to assess company performance (Wet, 2013). According to Islam et al. (2014), EPS shows the amount of the company's net profit that is ready to be shared with the company's shareholders. The number of the company's EPS can be known from the company's financial statement information.
Although some companies do not list the size of the EPS of the company concerned, EPS can be calculated based on the balance sheet information and the company's income statement.
The company's ability to generate net income per share is an indicator of the company's financial fundamental that the investors use for consideration in choosing the stocks. With accurate and accurate assessment, EPS analysis can minimize investment risk and help investors gain profits. Based on the description above, the first hypothesis can be formulated as follows: H 1 : EPS has a positive effect on stock returns

Economic Growth Weakens/Strengthens the Effect of EPS on Stock Returns
The study by Velankar et al. (2017) found that the effect of economic growth on EPS and stock returns is inconsistent. Therefore, economic growth has a positive or negative effect on EPS and stock returns. This evidence is the same as that by the Crestmont Research study in 2017. According to Samimi and Jenatabadi (2014), economic growth encourages companies to issue new shares because they are considered capable of increasing company profits. Coban (2014) stated that there is a positive relationship between profitability and economic growth. Profit affects earnings that automatically affect EPS. Based on the description above, the second hypothesis can be formulated as follows: H 2 : Economic growth strengthens the positive effect of EPS on stock returns

The Effect of DER on Stock Returns
According to Haque and Sarwar (2013), DER has a positive and significant effect on stock returns. The increase in DER will make the market react positively if the market tends to interpret that an increase in DER is considered a good signal about the company's prospects in the future (Degutis and Novickytė, 2014). This can occur because a high DER is considered capable of producing a higher return if it is used optimally (Acheampong et al., 2014). The company's risk level is reflected in the debt to equity ratio (DER). DER shows how much capital is owned by the company in meeting their obligations. Based on signaling theory, information about the company's financial statement is used by investors as a signal about the company's condition in the future. Every investor avoids investing in a company that has a high DER because it reflects a high level of risk (Çelik and Isaksson, 2014). De Luca (2017) stated that the greater the DER, the greater the risk of defaults faced by the company. The higher the DER the company must also pay higher interest costs (Roshan, 2009). Based on the description above, the third hypothesis can be formulated as follows: H 3 : DER has a negative effect on stock returns

Economic Growth Weakens/Strengthens the Effect of DER on Stock Returns
DER is related to the amount of company's debt. Although debt is one of the considerations of investors in determining investment decisions, it does not mean that the company's debt has caused the stock return to be bad. Debt is related to financial institutions (banks) lenders. The high and low rate of loan interest affects the company and economic growth. Manzoor et al. (2019) assessed that the reduction in lending rates could be the key to driving economic growth.
The lower the interest, the less the investment risk faced by investors. Also, a lower interest rate can encourage higher dividend payments. The source of corporate funding derived from debt has advantages, namely lower tax costs due to interest costs (Aslama, 2019). Based on the description above, the fourth hypothesis can be formulated as follows: H 4 : Economic growth weakens the negative effect of DER on stock returns

The Effect of PER on Stock Return
The price-earnings ratio (PER) is the ratio of price per share to earnings per share. This ratio shows the amount investors are willing to pay for each dollar (rupiah) reported by the company (Manzoor et al., 2019). The higher PER value indicates the prospect of the stock price being valued higher by the investor on the income per share. For that reason, the higher PER also shows the more expensive the stock is to the income per share.
The companies with a high PER have a high chance of growth rates. It causes investors to be interested in buying company shares which can then increase stock prices (Angelovska, 2016). The increase in stock prices that occur will be responded positively by investors because it affects stock returns. Thus, it can indicate that PER will have a positive effect on stock returns. This statement is supported by the results of research conducted by Karami & Talaeei (2013). Based on the description above, the fifth hypothesis can be formulated as follows: H 5 : PER has a positive effect on stock returns

Economic Growth Weakens/Strengthens the Effect of PER on Stock Return
According to MSCI Barra Research (2010), economic growth is assumed to affect shareholders in three stages. First, economic growth affects the growth of company profits. Second, the growth of aggregate income translates into an increase in EPS; third, an increase in EPS translates into an increase in stock prices. Increasing stock prices will affect stock returns.
PER shows how many times investors pay for each rupiah profit per share of shares produced by the company (Puspitaningtyas, 2018). The PER value is used by investors to understand how the market appreciates the company's performance as reflected by EPS. Companies with high growth rates usually have a high PER. This shows that the market expects earnings growth in the future. The development of the infrastructure sector and the increase in economic growth are expected to be able to influence the increase in profits and return on the stock of infrastructure companies. Based on the description above, the sixth hypothesis can be formulated as follows: H 6 : Economic growth strengthens the positive effect of PER on stock returns The Effect of NPM on Stock Returns Net profit margin (NPM) is used to calculate the amount of profit from each dollar (rupiah) of sales remaining after all costs are deducted, namely operating costs, interest, and taxes (Graham, 2006). According to Dita and Murtaqi (2014), NPM has a positive and significant effect on stock returns. The statement was supported by Martani et al. (2009) which first stated that NPM has a positive and significant effect on company stock returns. A high NPM value also signals investors to invest because it indicates the company's ability to increase net income (Dita and Murtaqi, 2014). Based on the description above, the next hypothesis can be formulated as follows: H 7 : NPM has a positive effect on stock returns

Economic Growth Weakens/Strengthens the Effect of NPM on Stock Returns
Economic growth is a macro variable that affects stock returns. In this study, economic growth can be seen from the size of the GDP. The increased GDP can be interpreted as an increase in the income and purchasing power of the people. Increasing people's purchasing power can affect company profits and investment levels. The greater the NPM, the more effective the company's performance will be. This can increase investor confidence to invest in a company . This ratio shows how much percentage of net profit earned from each sale. The higher the GDP, the higher the level of sales of the company is expected. Investment, in general, has a long-term relationship with economic growth (Li and Liu, 2002). One indicator that shows that GDP affects investment is because investment depends on the outputs obtained from all economic activities (Samuelson and Nordhaus, 2001). Low GDP indicates poor economic growth. Also, significant changes in GDP have a significant effect on the stock market (Sattar et al., 2018). This affects the high or low stock prices of companies that automatically affect stock returns. Also, economic growth increases investor expectations for future profits. Based on the description above, the hypothesis is stated as follows: H 8 : Economic growth strengthens the positive effect of NPM on stock returns

RESEARCH METHOD
This study used the population of 63 infrastructure companies listed on the Indonesia Stock Exchange. The sample was taken from the annual report of infrastructure companies in the 2015-2017 observation periods. They were taken using a purposive sampling technique, in certain terms and condition.

Types and Data Sources
This study used documentary data taken from the financial statements of infrastructure companies listed on the IDX and economic growth data from BPS in 2015-2017. They were taken from the IDX and the database of PT Mirae Asset Sekuritas.

Data Collection Method
The data were taken by means of documentation method. It was done by collecting financial statements of infrastructure companies on the IDX official website and the PT Mirae Asset Sekuritas database during 2015-2017. They were then processed to produce an overview related to the research variables.

Evaluate the Goodness of Fit Model
This study used a structural model (inner model) because all variables can be calculated directly and interpreted with exact numbers and standard formulas. The following is an evaluation of the structural model using WarpPLS:

Multicollinearity
According to Daoud (2017), the multicollinearity test aims to test whether, in the regression model, there is a correlation between independent variables. In a good regression model, there should be no correlation between among independent variables. If the independent variables correlate with each other, then the variables are not orthogonal. Orthogonal variables are independent variables whose correlation value between the independent variables is zero (0). Multicollinearity in the WarpPLS software can be known through the value of the average block variance inflation factor (AVIF). The ideal AVIF value is less than 3.3, but it can still be accepted if it is worth less than 5. The results of the multicollinearity test in this study can be seen on Table 1. On Table 1, the AVIF value is 1.739 < 3.3 (ideal). This shows that in this study there was no multicollinearity between independent variables.

Inner model
According to Sholihin and Ratmono (2013), the measures that can be used in assessing the inner model are as follows: Coefficient of determination R 2 values were 0.67, 0.33, and 0.19 indicating that the models were good, moderate, and weak (Chin, 1998). The value of R 2 of the test results using WarpPLS is as follows: Table 2 shows that the R 2 value of this research model is 0.257 (weak). The R 2 value shows that the dependent variable returns on infrastructure companies in Indonesia for the 2015-2017 period can be explained by the independent variables (EPS, DER, PER, NPM) and moderating variables of economic growth of 25.7%. The remaining 74.3% is explained by other variables, not in this research model.

Predictive relevance
Q 2 predictive relevance of structural models is a value that shows how well observation is generated by the model and also its parameter estimates. Q 2 is greater than zero (0) which indicates that the exogenous latent variable has a predictive relevance of the endogenous latent variables that are affected (Sholihin and Ratmono, 2013). The value of Q 2 of the test results using WarpPLS is on Table 2. Table 3 shows that the Q 2 value of this research model is 0.339. This value is more than zero (0), so it can be concluded that the exogenous latent variable has predictive relevance to the endogenous latent variables that are affected.

Effect Size
Effect size is a measure of the practical significance of research results in the form of a measure of the magnitude of the correlation or the effect of a variable on other variables (Sullivan and Feinn, 2012). The effect size can be found on the standard errors and effect size of the path coefficients menu in the WarpPLS software. Effect size according to Sholihin and Ratmono (2013) can be grouped into three categories, namely weak (0.02), moderate (0.15), and large (0.35). The value of the effect size in this study is presented on Table 4. Table 4 shows that the estimated effect size of the EPS variable is 0.051 > 0.02. The estimation result is considered medium so that EPS has a moderate contribution to increasing stock returns. The estimated effect size for the DER variable is 0.020 = 0.02. The estimation result is moderate so that DER has a moderate contribution to increasing stock returns. Estimated value of effect size for PER variable is 0.132 > 0.02. The estimation result is moderate so that PER has a moderate contribution to increasing stock returns. Estimated value of effect size for NPM variable is 0.008 < 0.02. The estimation result is weak. Therefore, it indicates that NPMs contribute low in increasing stock returns. Also, the estimated effect size for EPS, DER, PER, NPM variables on the dependent variable return on infrastructure companies in Indonesia for the 2015-2017 period with economic growth as a moderating variable shows a value of 0.007 (GDP*EPS), 0.003 (GDP*DER), 0.099 (GDP*PER), 0.034 (GDP*NPM). This value indicates that the contribution of GDP*EPS and GDP*DER variables to stock returns is classified as weak (<0.02). However, the contribution of GDP*PER and GDP*NPM to stock returns is classified as moderate.
In addition to the size above, the fit indices and P-values model also displays the results of  Kock (2013), the P-value of APC and ARS can be received if it is less than 0.05 (significant). Table 5 shows that the value of APC and ARS are 0.026 and 0.002 respectively. This value is acceptable because it is less than 0.05. Because APC and ARS < 0.05 and AVIF 1.739 < 3.3, it can be concluded that this research model fits the data used and is free of multicollinearity.

Coefficient of Determination (Adjusted R 2 )
The problem often encountered in the use of R-squared to assess the good or bad of a model is. The value continues to rise along with the addition of independent variables into the model. According to Daoud (2017), the coefficient of determination (R 2 ) is a measure of a model's ability to explain variation in the dependent variable. Adjusted R 2 serves to measure the level of confidence, adding the right independent variable to increase the predictive power of the model. The adjusted R 2 value will never exceed the R-squared value, it can even go down if there are additional independent variables that are not needed. Adjusted R 2 can have a negative value. The adjusted R 2 estimation results in this study are presented on Table 6.
The adjusted R 2 value was 0.188. It shows that the regression model used in his study can explain the magnitude of the effect of the independent variables EPS, DER, PER, and NPM on the dependent variable of stock return of infrastructure companies in Indonesia in the period 2015-2017 with economic growth as a moderating variable of 18.8%. Yet, the remaining 81.2% is explained by other variables that are not in the regression model.

Partial Significance Test
The partial test is used to examine the effect of each independent variable EPS, DER, PER, and NPM on the stock return of infrastructure companies. The researcher also examined the effect of each independent variable on stock returns with economic growth (GDP) as a moderating variable. Partial testing was done by comparing the P-values of each independent variable and the moderating variable to the dependent latent variable. The P-value of each variable is presented on Figure and Table. Explanation of Table 7 is as follows.

Effect of independent variables on the dependent variable Effect of EPS on Stock Returns
H 1 : EPS has a positive effect on stock returns From Table 7 it is known that the EPS path coefficient value is 0.216 with p-values of 0.013. P-values are lower than the significance value of 0.05. H 1 is supported if the p-values are less than 0.05. P-values 0.013 <0.05, then H 1 is supported. Therefore, it can be concluded that EPS has a positive effect on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. Return (Path Coefficient) 0,051 0,020 0,132 0,008 0,007 0,003 0,099 0,034 Source: WarpPLS output processed Table 5 Average Path Coefficient and ARS

Average path coefficient
Average R-squared P-values 0,026 0,002 Source: WarpPLS output processed Effect of DER on stock returns H 3 : DER has a negative effect on stock returns From Table 7 it is known that the value of the DER path coefficients is -0,146 with a p-value of 0,070. P-values 0.070 > 0.05, then H 3 is not supported. Therefore, it can be concluded that DER has no effect on stock returns of infrastructure companies in Indonesia for the 2015-2017 period.

Effect of PER on stock returns
H 5 : PER has a positive effect on stock returns From Table 7 it is known that the value of the path coefficients PER is -0.293 with 0.001 p-values. P-values 0.001 < 0.05, then H 5 is not supported. Therefore, it can be concluded that PER does not affect stock returns of infrastructure companies in Indonesia for the 2015-2017 period.

Effect of NPM on stock returns.
H 7 : NPM has a positive effect on stock returns From Table 7 it is known that the value of NPM path coefficients is -0,097 with p-values <0.165. P-values 0.165 > 0.05, then H 7 is not supported. Therefore, it can be concluded that NPM has no effect on stock returns of infrastructure companies in Indonesia for the 2015-2017 period.

Effect of independent variable and moderation variable on the dependent variable
The effect of EPS on stock returns with economic growth (GDP) as a moderating variable. H 2 : Economic growth strengthens the positive effect of EPS on stock returns From Table 7 it is known that the path coefficient value of EPS*GDP is 0.049 with p-values of 0.203. P-values 0.203 > 0.05, H 2 is not supported. Therefore, it can be concluded that economic growth does not moderate the effect of EPS on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. Before moderation, the effect of EPS on stock returns was positively significant, but after being moderated by GDP the effect of EPS on stock returns was insignificant. This shows that GDP does not have a significant ability to moderate the effect of EPS on stock returns.   Table 7, it is known that the path coefficients of GDP*DER is 0.083 with p-values of 0.313. P-values 0.313 > 0.05, then H 4 is not supported. Therefore, it can be concluded that economic growth does not moderate the effect of DER on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. Before moderation, the effect of DER on stock returns was insignificant and after being moderated by GDP the effect of DER on stock returns was also insignificant. This shows that GDP does not have a significance to moderate the effect of DER on stock returns.
The effect of PER on stock returns with economic growth (GDP) as a moderating variable. H 6 : Economic growth strengthens the positive effect of PER on stock returns From Table 7, it is known that the value of GDP*PER path coefficients is -0.217 (greater than -0.293) with p-values of 0.013. P-values 0.013 < 0.05, then H 6 is supported. Therefore, it can be concluded that economic growth strengthens the positive effect of PER on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. Before moderation, the effect of PER on stock returns was insignificant and after being moderated by GDP the effect of PER on stock returns was positively significant. This shows that GDP has a significant ability to moderate the effect of PER on stock returns.
The effect of NPM on stock returns with economic growth (GDP) as a moderating variable. H 8 : Economic growth strengthens the positive effect of NPM on stock returns From Table 7, it is known that the value of NPM*GDP path coefficients is 0.197 with p-values of 0.022. P-values 0.022 < 0.05, then H 8 is supported. Therefore, it can be concluded that economic growth strengthens the positive effect of NPM on stock returns on infrastructure companies in Indonesia for the 2015-2017 period. Before moderation, the effect of NPM on stock returns was insignificant and after being moderated by GDP the effect of NPM on stock returns was positively significant. This suggests that GDP has a significant ability to moderate the effect of NPM on stock returns.

CONCLUSION, IMPLICATION, SUG-GESTION AND LIMITATION Conclusion
It provides evidence that the fundamental values (EPS, DER, PER, and NPM) on stock returns, moderated by economic growth (GDP). It indicates that EPS has a positive effect on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. For investors, EPS is information that is considered the most basic and useful because it can describe the earnings prospects in the future. DER does not affect the stock return of infrastructure companies in Indonesia for the 2015-2017 period. Therefore, it shows that the company is unable to maximize its debt to generate profits. Investors will not be interested in companies that have low profits.
PER does not affect the stock return of infrastructure companies in Indonesia for the 2015-2017 period. Thus, it indicates that the company's shares have a low market price. The low PER value also indicates that the company is in bad condition and, therefore, it is risky for investors. NPM does not affect the stock return of infrastructure companies in Indonesia for the 2015-2017 period. Of the total 63 infrastructure sector companies studied, 23 of them had negative NPM values. The negative NPM value indicates that the company suffered a loss. Economic growth does not moderate the effect of EPS on stock returns of infrastructure companies in Indonesia for the 2015-2017 period.
Economic growth can trigger issuers to issue new stock or do stock splits. Too many shares will reduce the value of EPS and the delusion of shares. Economic growth does not moderate the effect of DER on stock returns of infrastructure companies in Indonesia for the 2015-2017 periods. It has an effect on decreasing credit interest. However, investors prefer to invest in the company stocks that generate higher profits than new companies that have the potential to earn higher profits because of the use of financial leverage.
Economic growth strengthens the positive effect of PER on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. In this case, it increases investors' expectations of company profits in the future. Investors will not mind buying shares at higher prices but have higher profit prospects and stock returns as a result of infrastructure development and economic growth. Economic growth strengthens the positive effect of NPM on stock returns of infrastructure companies in Indonesia for the 2015-2017 period. The GDP increase as a proxy for economic growth should be able to increase consumers' purchasing power of the company's products and services. The increase in sales or the company's earnings can increase the company's profits, implying that it can increase the investors' confidence in investing.
The novelty of the research is indicated by the finding that economic growth (GDP) can moderate the effect of NPM and PER on stock returns, but unable to moderate the effect of EPS and DER on stock returns. In other words, GDP has cannot fully moderate the effect of fundamental values on stock returns of Indonesian infrastructure companies. It is due to the economic growth impact on the stock returns of manufacturing companies, in which, it takes a long time so that future research agendas need to consider a longer time or research period.

Implication
This study proves that economic growth does not moderate the effect of EPS and DER on stock returns of infrastructure companies. However, economic growth can moderate the effect of NPM and PER on stock returns. It means, when investors will invest in the company, investors need to pay attention to economic growth factors.
Investors should begin to observe the potential for increasing stock returns that can be influenced by infrastructure development and economic growth. The effect of infrastructure development policies will not be felt shortly. However, in the long run, infrastructure development can boost the better economic growth. Therefore, government policies that set priorities for infrastructure development need to be supported because it has a positive effect on profit and return on the stock of infrastructure companies.

Limitation
This study has several limitations that can lead future research to obtain better results. These limitations include such as, first, the number of samples studied is limited because the focus of research is only on infrastructure sector companies. Secodnly, thsis study only uses four independent variables related to fundamental values, namely EPS, DER, PER, and NPM, and one moderating variable of economic growth which is proxied by GDP.

Suggestion
In connection with the limitations, researchers provide several suggestions for future research as follows. First, the next researchers are advised to increase the time of the observation period to obtain a more adequate number of samples so that more accurate results can be obtained. Secondly, they are also advised to add the number of independent variables related to stock returns.