Determinants of Investment in the Jordanian Productive Sectors

Abstract

This paper aims to find out the main factors that are important in determining the size of investment in the Jordanian productive sectors. For this purpose, the study used panel data for four Jordanian productive sectors over the period 2000-2017. Also, fixed-effects modeling was carried out to identify the relationship between investment and its potential determinants. Empirical investigations of the four productive sectors reveal the following results: The real value of sector's production and the real value of credit facilities have a positive and significant impact on investment, while the real interest rate has a negative effect on investment in the Jordanian productive sectors. Also, at the sector level, agriculture was more responsive to changes in the real value of credit facilities, while other sectors were more responsive to changes in the real value of sector's production. According to these results, it seems that some policy actions should be taken to enhance the size and the role of investment in the economy. For example, policymakers should adopt a mixed policy and expand the provision of credit facilities, especially to the agricultural sector, to enhance agricultural activity in a manner that ensures the improvement of infrastructure and land reclamation.

1. Introduction

Investment is considered to be one of the main factors in the success of economic development plans through its role in increasing the level of employment, reducing the unemployment rate, increasing the level of production, and achieving high growth rates. Investment also plays an important role in increasing levels of exports and replacement imports, which improves the balance of payments. Moreover, economic decision-makers are aware of the role of fluctuations in investment in the performance of the economy. Therefore, the process of identifying the determinants of investment behavior is an important issue for the economy.

The importance of investment induces countries to concentrate on this variable and to improve their environment in order to be attractive to investment. Countries also attempt to reduce or eliminate the obstacles that can affect on the investment spending by reviewing their laws and regulations. In addition, some countries have established institutions that support and encourage investment, such as the Jordan Investment Commission, which was established in 2014. This is a lead agency for implementing the government’s investment promotion policies and activities, in order to enhance the competitiveness of the Jordanian economy, attract investment, facilitate investment success and eliminate bureaucratic constraints.

In Jordan, gross fixed capital formation amounted to an average of 23.3% of gross domestic product (GDP) during the study period. This was despite the great interest in investment, and the efforts expended to reduce and eliminate potential obstacles to investment in Jordan. Why these efforts have not succeeded in improving investment remains an important question that must be subjected to empirical investigation. Furthermore, the rate of gross fixed capital formation in the Jordanian productive sectors is low: the average percentage of fixed capital formation in the agricultural sector was 1.2% of gross fixed capital formation in Jordan during the study period;1.5% in the construction sector; 2.1% in the mining sector; and 8.9% in the industrial sector.

Therefore, this study aimed to contribute to the debate about the determinants of investment, using panel data on the Jordanian productive sectors over the period from 2000 to 2017, it is noted that there is an absence of this type of study in the existing literature despite the importance of the studies that take into account individual differences between sectors, and the responsiveness of each sector to variables used.

This study consists of four sections. The first section presented an introduction to the subject; the second section describes the theoretical framework and some previous studies that have been concerned with the determinants of investment; the third section presents the methodology and econometric analysis; the fourth section concludes and put forward recommendations.

2. Literature Review

Many economic theories have focused on studying the determining factors for investment, owing to the importance of this variable in achieving economic growth and development and many variables have been used in these theories and related models. One of the variables that have been most often employed in investment models and equations is the real interest rate, which was first used in the investment equation by Jorgenson (1963).

Jorgenson derived the required stock of capital as a function of real production and the opportunity cost of capital in a theory of investment behavior that developed from the neoclassical theory of optimal accumulation of capital. This theory posits that the required stock of capital correlates positively with production and negatively with the cost of capital, where the decrease in the real interest rate leads to a decrease in the opportunity cost of capital, which leads to an increase in investment spending.

On the other hand, others who were interested in the role of financial markets in the accumulation of capital, such as McKinnon (1973) and Shaw (1973), showed that there is a positive relationship between real interest rates and the size and quality of investment in financially-repressed economies through the role of real interest rates in generating an incentive to accumulate savings and providing the funds needed to implement any investment project. Therefore, an increase in the real interest rate leads to an increase in savings and then investment.

The effect of the interest rate is not confined to its impact on the size of investment; many economic models have dealt with the role of the responsiveness of investment to the interest rate in the efficiency of fiscal and monetary policy, whereby the responsiveness of investment to the interest rate can increase or decrease the efficiency of economic policies in achieving economic targets. Findlay (1999) indicated that fiscal policy is more efficient if investment responds less to the interest rate.

Another variable that have been used in investment models is total output, which reflects demand in the economy. This variable was used in the investment function for the first time in the context of the acceleration principle, which was developed by Clark (1917). The acceleration principle states that growing demand for consumer goods increases accumulation of the capital equipment needed to produce these consumer goods. Clark showed that there is a special and technical relationship between the demand for consumer goods and the demand for the capital equipment needed to produce these goods.

The acceleration principle shows how total output can induce additional investment spending. Based on this principle, a small change in national income or output will lead to a bigger change in investment. Clark explained that the increase in output often results in a more than proportionate increase in investment spending through the role of the demand for consumer and capital goods.

Other studies have used credit facilities to explain the variation in investment, as in a study by Wai and Wong (1982) that presented the relationship between credit facilities and investment through the following points: Firstly, the behavior of any sector in the economy depends at least partly on its ability to obtain financing in the easiest way. Therefore, the unavailability of projects financing acts as a short-term constraint. Moreover, short- and medium-term loans help in financing day-to-day business operations, which contributes to capital accumulation. This point is more in line with the situation of developing countries that are experiencing economic growth, and many emerging companies that require multiple financial resources. In contrast, established companies in developed countries tend to rely on their retained earnings.

Secondly, some recent financial theories of economic development have presented the relationship between money and capital in the developing countries as a complementary relationship. This complementary relationship is mainly due to policies such as the maintenance of high rates of inflation and low and disequilibrium interest rates, which create ‘financial repression’. In such economies, money works as a channel for capital accumulation, and an increase in the desirability of holding cash balance or borrowing would reduce the opportunity cost of saving internally for the purpose of capital accumulation.

Thirdly, for the countries that depend on importing a large proportion of machinery and production equipment, the importer needs import deposits to continue in the importing process. Therefore, the availability of credit facilitates is needed for the import process of this machinery and equipment, which positively affects investment.

Other explanatory variables have been used in applied studies interested in investment. For example, Fielding’s (1993) study showed that both monetary and financial integration play a significant role in determining investment. This study used an eclectic model of investment that was constructed using data from Kenya and Cote d’Ivoire.

The study by Ndikumana (2000) investigated the effects of financial development on domestic investment for a sample of 30 sub-Saharan African countries. This study used a dynamic serial-correlation investment model to arrive at its main finding, which is that there is a positive relationship between domestic investment and several indicators of financial development. This finding implies that financial development stimulates economic growth through capital accumulation.

Another attempt to investigate gross investment behavior was made by Salahuddin et al. (2009). They used panel data from 21 Muslim developing countries over the period 1970–2002. The results of this attempt suggest that lagged investment, growth rate per capita real GDP, domestic savings, trade openness, and institutional development have a positive and significant impact on investment. In addition, foreign aid and credit to the private sector were found to have significant positive effects on investment, but not a robust effect. Finally, foreign debt servicing was found to have a consistent negative effect on investment.

Also, Attefah and Enning (2016) analyzed the determinants of private investment using time-series data over the period 1980-2010. This study used a aid of multiple linear regression model to determine the factors that have a significant impact on private investment in Ghana. The main findings of this study are that public investment, credit to the private sector, external debt, the openness of the economy, corporate tax, and democracy have significant impact on investment. In contrast, GDP growth, real interest rate, inflation, and real exchange rate were not found to be statistically significant.

Rashid and Shakoor’s (2019) study identified the indicators that explain the behavior of investment in Asian countries over the period 1987–2017. The results showed that domestic savings, GDP per capita, and government expenditure have a positive and significant impact on domestic investment. Also, the study emphasized the role of these variables in attracting investment and determining its growth.

To investigate the impact of foreign direct investment on private investment, Tung and Thang (2020) used a sample of 49 developing countries in Asia and Africa during the period 1990–2017. The results confirmed a crowding-in effect of foreign direct investment on private investment, which implies that foreign direct investment induces private investment. In addition, the lag value of private investment has a positive and significant impact on itself in the next period, which reflects the inertia in the trend of private investment in recipient countries. In the full-panel sample, GDP per capita, trade openness, and electricity have a positive and statistically significant impact on private investment. Tung and Thang (2020) also estimated the investment function for subgroups and they found that trade openness and labor force have a positive and significant in Africa but not in Asia. Also, the domestic credit variable has a negative and significant effect on private investment only in Asian developing countries. Furthermore, the electricity variable has a positive and significant impact on private investment in Asia.

Githaiga (2020) investigated the impact of foreign remittances on private sector investment and the moderating role of banking sector development in 15 sub-Saharan African countries over the period 1986–2017. The main results of this study were that foreign remittances and banking sector development had a significant and positive impact on private investment in sub-Saharan Africa. Also, banking sector development significantly moderated the relationship between foreign remittances and the private sector investment.

Le and Kim (2020) tested how economic freedom affected firms’ investment in Vietnam, whereby the economic freedom, such as capital freedom and domestic credit freedom, allows firms to access external finance more easily, so that their investment depends less on internal cash flow. Le and Kim (2020) used unique firm-level data over the period 2006–2016, including data on listed firms on two major stock exchanges and unlisted firms in the Unlisted Public Company Market. The findings of this study indicated that capital freedom and domestic credit freedom played an important role in investments for Vietnamese firms. On the other hand, Le and Kim (2020) were unable to find evidence that overall economic freedom relaxed financial constraints on firms.

Finally, the study by Ayeni (2020) employed the ARDL co-integration method to analyze a long-run equilibrium model for private investment in Gambia. This study used exchange rate, credit to the private sector, external debts, real interest rate, real exchange rate, and inflation as explanatory variables in its investment model. The main findings of this study show that a high exchange rate increases the real cost of importing capital goods, which makes investment very costly. On the other hand, the financing of huge debts plays a constraining role in private investment in Gambia. Aggregate demand conditions, real interest rate, real exchange rate, and inflation all performed below expectations. Also, credit to the private sector did not induce private investment in Gambia due to insufficient credit.

3. Methodology

In order to achieve the objective of this study and identify the determinants of investment in the Jordanian productive sectors for the period 2000–2017, this study used panel data on four productive sectors: agriculture, manufacturing, construction, and mining. The methodology of this study was constructed based on the empirical methods used in the previous literature, with some modifications and variables that are consistent with the Jordanian economy.

3.1. Model Specification

The sectoral investment function was estimated as a function of several variables, which were determined from the theories of investment and some applied studies that have dealt with this topic. These variables reflect some macroeconomic conditions, such as the sector’s production, which is considered part of the total demand in the economy and measured by the level of economic activity in that sector. The real interest rate was used to capture the effect of the user cost of capital in determining the size of investment. Also included are real credit facilities were used to measure the degree of liquidity granted to each sector. Therefore, the model can be formulated as follows:

\(I=f(R \text { Pro, } \mathrm{RCR}, \mathrm{RR})\)       (1)

Function (1) was estimated after taking the logarithm:

\(\begin{aligned} \log \left(I_{i t}\right) &=\beta_{0}+\beta_{1} \log \left(R \operatorname{Pro}_{i t}\right)+\beta_{2} \log \left(\mathrm{RCR}_{i t}\right) \\ &+\beta_{3} \log \left(\mathrm{RR}_{t}\right)+\varepsilon_{t} \end{aligned}\)       (2)

Where:

log(I_it): logarithm of gross fixed capital formation in the sector i in period t.

log(Rpro_it): logarithm of the real value of production Pro it for sector i in period t.

log(RCR_it): logarithm of real credit facilities granted for sector i in period t.

log(RR_t): logarithm of the real interest rate in period t.

β₀ , β₁ , β₂ , β₃: coefficients to be estimated.

ε_t: error term.

3.2. Data

The data on gross fixed capital formation were used as a measure of investment as in previous studies, while the real value of production for the sector was calculated by dividing the sector’s production by the price index and assuming the base year 2010. The credit facilities granted to the sector were used as an indicator for liquidity, and were converted into real value by dividing it by the price index and assuming the base year 2010. The credit facilities to the private sector were also used, and the results were close. However, this study adopted the variable of credit facilities granted to the sector in order to take into account the specificity of each sector.

Finally, the real interest rate was calculated by subtracting the inflation rate from the nominal interest rate, where the interest rate for lending was relied upon, and the rediscount rate was also used to verify the results and they were close. The data for these variables were obtained from the Central Bank of Jordan website (www.cbj.gov.jo). The reason for ending the study period in 2017 is the lack of data on gross fixed capital formation for the last three years.

3.3. Empirical Methods

When using panel data, three models are most often used in estimating this type of data: the pooled model, the fixed- effects model and the random-effects model. Each of these models has its own characteristics and suitability for the specified data, and the appropriate model is chosen based on the set of econometric tests: the fixed-effects test, the random-effects test, and the Hausman test.

4. Results and Discussion

4.1. Stationary Test

Before starting with the estimation process, it must be noted that the first step in analyzing the data is to test whether the data satisfy the stationary condition or not, so that the econometric method that achieves the objectives of this study can be chosen. Therefore, the current study adopted the Levin, Lin, and Chu (LLC) test to check the stationarity of the variables. This test is considered one of the most popular tests used to examine the stationary panel data.

Levin, Lin, and Chu (2002) indicated that the main formula for this test was first drawn from the Augmented Dickey-Fuller (ADF) model and then it was developed to eliminate the problem of autocorrelation and be consistent with the nature of panel data. The null hypothesis of the test states that there is a unit root in the variable (H₀ : ρ_i = 0), while the alternative hypothesis indicates that there is no unit root. So that a decision can be made, in this study the calculated value must be compared with the critical value of the modified t-test.

According to this test, if the null hypothesis is rejected, then the variable satisfies the stationary condition at a level, and if all variables have the same characteristic, then the study can continue in estimating equation (2) using the appropriate model. But if the null hypothesis fails to be rejected, then the variable is not stationary, and the test must be repeated to find the degree of stationarity for this variable.

The results of the stationary test are presented in Table 1 and show that the null hypothesis is rejected at the 5% level for all variables, which means that all variables are stationary at a level. According to these results, traditional methods, including the ordinary least squares method, could be used to identify the determinants of investment (Gujarati, 2009).

Table 1: The Results of Levin, Lin and Chu Test

The Optimal Lags Periods were Determined Based on the Schwarz Information Criterion (SIC).

4.2. Results of Regression Analysis

To estimate the model specified in equation (2) using the panel data it is necessary to estimate this equation with the pooled model, the fixed-effects model, and the random-effects model, and then select the most appropriate model for the study data based on the following tests:

The random-effects test: Baltagi (2008) indicated that the Breusch-Pagan Lagrange Multiplier (BP-LM) test can be used to compare the random-effects model with the pooled model. The null hypothesis of this test states that there are no individual effects between sectors, which is formulated as follows: (\(H_{0}: \sigma_{\mu}^{2}=0\))

According to the results of this test, if the null hypothesis is rejected, then the random-effects model is more appropriate for the data used. After applying this test, the calculated value was 0.732 and the tabulated value was 7.81. Therefore, the null hypothesis fails to be rejected at the 5% level, which means the pooled model is more appropriate than the random-effects model.

The fixed-effects test: Baltagi (2008) saw that the F test of the restricted model can be used to compare the fixed-effects model with the pooled model in this case, it is assumed that the pooled model is the restricted model, and the fixed effects model is the unrestricted model. Here, the null hypothesis in this test states that all sectors have the same constant and the individual differences between sectors equal zero \(\left(H_{0}: \mu_{1}=\mu_{2}=\ldots=\mu_{N-1}=0\right)\)If the null hypothesis is rejected, then the estimates of the fixed-effects model are the most appropriate for the study data.

According to the results of this test, the calculated F value was 13.302 and the tabular value was 1.71 at the 5% level. Therefore, the test rejects the assumption of equality of the individual differences between sectors with zero, which means that there are individual differences between sectors and according to this result the fixed-effects model is more appropriate than the pooled model.

The Hausman test is used to compare the random-effects model with the fixed-effects model, through testing a null hypothesis that states that there is no correlation between errors and explanatory variables E(e_it/X_it))=0. If the null hypothesis is rejected, then the fixed-effects estimates are appropriate. The results of this test are invalid, due to the incompatibility of the data with the random-effects model. Therefore, the fixed-effects model is the more appropriate for the study data than the random-effects model.

The summary of previous tests indicates that the fixed-effects model is the most appropriate for the study data, and the following table (2) presents the results of this model:

4.3. Discussion

The fixed-effects model was estimated using the Feasible Generalized Least Squares (FGLS) method, in order to ensure that the estimated model corrects the problems of cross-section heteroskedasticity and contemporaneous correlation. This method works to weight the standard error and then change the calculated t-value without affecting the values of the coefficient, this method only affects the statistical significance (Green, 2012).

The results of the fixed-effects model indicate that there are positive effects for both the real value of sector’s production and the real value of credit facilities where increasing the real value of sector’s production by 1% leads to an increase in investment of 0.789%. This result is consistent with the accelerated theory of investment, which indicates that there is a positive effect of production on investment. The equation was also estimated using real GDP instead of the real value of sector’s production, and the real GDP coefficient was 1.13. This is consistent with the principle of acceleration, which indicates that an increase in output often results in a more than proportionate increase in investment.

According to the results presented in Table 2, the real value of the credit facilities granted to the sector positively affects on investment, since an increase of credit facilities by 1% leads to an increase in investment by 0.998%, which means that most of the credit facilities granted to the sector are directed toward investment. As in most developing countries, debt is considered to be one of the most important external sources of project financing, especially when projects are first established. Therefore, policymakers in Jordan must enhance efforts to facilitate access to credit and increase the ability of banks to grant credit facilities.

Table 2: The Results of Fixed Effect Model

The Values in the Parentheses are the Calculated t-Value

* Denotes Statistically Significant at 5% Level.

On the other hand, the results indicate that the real interest rate has a negative impact on investment: if the real interest rate increases by 1%, investment decreases by 0.067%. This result is consistent with the neoclassical model of investment, which sees the real interest rate as part of the cost of using capital. When the real interest rate increases, the opportunity cost of capital increases, which leads to a decrease in investment spending. Low elasticity of investment to the interest rate can also be seen, amounting to 0.067, which means that investment is inelastic to change in the interest rate. Therefore, fiscal policy is more effective in influencing economic variables.

The study also used a number of additional explanatory variables, such as inflation, tax, and a dummy variable to capture the effect of the 2008 financial crisis on investment, but these variables did not have statistical significance. Finally, it should be noted that the explanatory variables used in the fixed-effects model are jointly significant at the 5% level, according to the calculated value of the F test, which was 171.669. Finally, the explanatory variables explain 94% of the variation in investment, based on the value of the adjusted coefficient of determination(\(\bar{R}^{2}\)).

The final step in this study’s analysis is the attempt to estimate the individual equation for each sector. The results of this attempt indicate that investment in the agricultural sector is more responsive to the change in credit than it is to other variables. Therefore, it is necessary to work on extending the provision of credit facilities and expanding the circle of beneficiaries of credit facilities in this sector to enhance agricultural activity in a manner that ensures the improvement of infrastructure and land reclamation. It is noticed that the level of credit facilities granted to this sector in Jordan is low: it amounts to an average of 1.6% of the total credit facilities granted to all economic sectors during the study period. The cause may be banks’ fear of granting loans to the agricultural sector, because production in this sector is linked to high-risk factors such as climate change, pests, and diseases that affect agricultural production.

On the other hand, the response of investment in other sectors to change in production is greater than its response to other variables, and this is in line with the principle of acceleration, which indicates that growing production leads to an increase in demand for consumer goods and the capital equipment that provides these goods, and then increases investment.

5. Conclusion

The objective of this study has been to find out the main factors that are important in determining the size of investment in the Jordanian productive sectors. For this purpose, the study used panel data for four sectors over the period 2000–2017. Fixed-effects modeling was carried out to identify the relationship between investment and its potential determinants. A number of conclusions can be drawn from econometric analyses, which are summarized as follows:

The real value of sector’s production and the real value of credit facilities have a positive and significant impact on investment, while the real interest rate has a negative effect on investment in the Jordanian productive sectors. Also, at the sector level, agriculture was more responsive to changes in credit facilities, while other sectors were more responsive to changes in sector’s production.

In accordance with the results on the elasticity of investment to interest rate and the impact of credit on investment, the study suggests that policymakers adopt a mixed policy to enhance the size and the role of investment in the economy. Finally, it is necessary to work on expanding the provision of credit facilities and also the circle of beneficiaries of credit facilities, especially in the agricultural sector, to enhance agricultural activity in a manner that ensures the improvement of infrastructure and land reclamation.