Impact of public expenditure on private investment - Study in some Asian economies and policy implications for Vietnam

s and Vietnam Chapter 5: Conclusions and policy implications for Vietnam 6 THEORETICAL BASIS AND LITERATURE REVIEW ABOUT THE IMPACT OF PUBLIC EXPENDITURE ON PRIVATE INVESTMENT 2.1. Public expenditure 2.1.1. Definition of public expenditure In scope of this thesis, the term “public expenditure” is based on the view of International Monetary Fund (IMF). It consists of total expense and the net acquisition of non-financial assets (such as acquisition of fixed capital assets, strategic stocks, land, and intangible assets). These are all government expense financed by the state budget; including the expenditure of different levels of government (local and general government), from maintaining the public sector to supplying public goods and services. 2.1.2. Role of public expenditure Public expenditure plays an instrument to allocate resources effectively; creating socio-economic infrastructure, supplying public goods and services; guaranteeing a safe society. Adding gaps which private sector can not invest. Redistributing income, ensuring social benefits. Maintaining macroeconomic stability and accelerating the economy. 2.1.3. Characteristics of public expenditure 1- Public expenditure activities are associated with the state apparatus. 2- Having public characteristics, toward the common benefits of the community. 3- Complying with the principal of non-refund or non-direct refund. 4- Being related to policy lags, including internal and external lags. 2.1.4. Classifying public expenditure Categorization is based on many different criteria such as similarity of specific spending items, classification goals, benefits received by sectors in society, main function of the government, transferable feature, economic feature, so on. 2.2. Private investment 2.2.1. Definition of private investment According to WB (2019), private investment covers gross outlays by the private sector, including private nonprofit agencies on additions to its fixed domestic assets. More detail, it can be seen as private sector’s purchase of fixed asset with the expectation to produce more income and value in the future. 7 2.2.2. The impact factor on private invesment Interest rate: When the interest rate goes up, the lending cost will become more expensive, making it difficult for investors in accessing capital, so limiting investment. Capital source of other economic sectors - Public expenditure: Public expenditure is a special factor as both crowding in and crowding out investment of private sector. - FDI: Foreign Direct Investment (FDI): FDI can have positive or negative spillover effect on investment. Credit for private sector: The scarcity in funding sources is considered to be one of the elements hindering private investment. Hence, the more credit fund is, the more easily financial providing for investment occurs. Economic growth. The growth of the economy will bring higher income for people, leading to increase goods demand, thereby stimulating investors to expand production and provide more capital. Some other macroeconomic factors: Inflation. Human resources. Trade openness. Political stability. 2.3. The impact of public expenditure on private investment 2.3.1. The crowding out effect “The crowding out effect” is an economic item describing that rising intervention to the market by government spending drives down or negatively affects private sector investment. In economic schools, Keynesian economics researches the crowding out effect in closed and open economy while classical theory studies in short term and long term. The way that the crowding out effect occurs includes two categories: (1)- Direct and indirect channels, and (2)- impact through accessing production resources. 2.3.2. The crowding in effect Public expenditure can adversely affect private investment, creating the crowding out effect as mentioned above. However, this trend can also occur in the opposite direction, meaning that raising of public expenditure drives expanding of private investment. This phenomenon is called “the crowding in effect”. 2.4. Literature review 2.4.1. The crowding out effect The crowding out effect on private investment by public expenditure is discovered by many researchers (Argimon et al., 1997; Furceri and Sousa, 2011). For 8 instance, Furceri and Sousa (2011) figured out each increase in government spending drove down 1.2% of consuming and 0.6% of investment from private sector. The cumulative influence after 4 years was greater as those figures were 1.9% and 1.8% respectively. In other words, the longer the time was, the more the crowding out effect was. It was the same point with Argimon et al. (1997) despite changes in research method. 2.4.2. The crowding in effect Some other analysts confirm that public capital will crowd in investment of private sector. In detail, the Government provides capital to fields of basic research, public health, education, and infrastructure in order to build socio-economic and technical background. From the microeconomic perspective, this activity helps to reduce the initial investment costs, creating favorable conditions in attracting private capital. From the macroeconomic perspective, it brings long-term growth of potential output, making investors more optimistic about prospects of development, leading continue to invest (Aschauer, 1989; Keynes, 1936). This theory receives the support of many empirical studies such as Blejer and Khan (1984), Ghura and Goodwin (2000), Blanchard and Perotti (2002), Erden and Holcombe (2005), Gjini and Kukeli (2012). However, not any research shows a unique dimension impact of government spending on private investment. The influence of public expenditure on private consuming and investment do exist, but there are differences depending on economic context, development level. From view of geography, Furceri và Sousa (2011) realized that the crowding in effect happened only in Middle East zone whereas the other continents as Africa, Asia, Europe, North America, South America, and Western India were opposite. From view of development level, the crowding out effect was found in both of OECD and devevloping countries. In the context of Vietnam, the researchers find two dimensions of the impact of public expenditure on private investment, but the crowding in effect overcomes the crowding out effect. Dao Thi Bich Thuy (2014) recorded positive reactions of productivity in private sector to changes in public expenditure; meanwhile the crowding out effect by tax (which was used for finance public expenditure) rose, making private income and investment down. However, the consequence showed that the positive effect prevailed over the opposite effect on private investment. In other words, it cannot be asserted certainly that public spending make negative impact on private sector investment. Su và Bui (2016) used GMM to study panel data of Vietnam provinces and confirmed that two dimensions but the crowding in effect was greater. 9 2.4.3. Research gap on the impact of public expenditure on private investment The impact of public expenditure on private investment has been debated in many studies. Many foreign researches directly assess the connection between private investment and public expenditure by considering private investment as a dependent variable in model or putting it in economic growth model. The Vietnam researchers often study this relationship in growth model or the system of equations which are used to evaluate the impact of policies and shocks on economic variables. So, within the author’s knowledge, the partition into separate models to examine clearly each impact has not been done by domestic researchers. This is also the gap which the author myself can exploit. The author will conduct an empirical research on the influence of public expenditure on private investment in Vietnam. Besides, the remarkable researches in Vietnam often use time-series data. Panel data only happens in the scope of provincial data in Vietnam (Su and Bui, 2016). Meanwhile, if data and method are adjusted, results and analyses can change. Thus, instead of using time-series data, in this thesis, panel data which is on 14 emerging and developing economies in Asia will be employed; then analyzing the linkage with Vietnam with expecting to find the new results of the impact of public expenditure on private investment. Within author’s knowledge, 3SLS up to now has not been much used in the study on the impact of public expenditure on private investment in general and Vietnam in particular. It will create a difference of this research compared to existing ones. Furthermore, the economists argue that institution plays an important role in the economy. The the crowding out effect can take place in the periods or economies with less developed institution; will diminish and turn into the crowding in effect when institution tends to get better (Everhart and Sumlinski, 2001). Government spending even can be reversed by corruption (Mauro, 1998). So, in order to contribute to comprehensively assess the impact on private investment, the author add variable “Institution” to the model. This variable also has not appeared in research on the impact of public expenditure on private investment in Vietnam, except the latest research of Su and Bui (2017) which uses similar variable as Provincial Competitiveness Index. 10 MODEL AND RESEARCH METHOD 3.2. Research model 3.2.1. Research model Research model in this thesis consists of two model: (1) Equation 1 about the impact of public expenditure on private investment; and (2) Equation 2 about the determines of public expenditure. priit = α0 + α1pgexit + α2pfdiit + α3pcreit + α4ririt + α5git + α6infit + α7labit + α8sch1it + α9ptra1it + α10polit + uit (1) pgexit = β0 + β1pgrevit + β2lgdpr2it + β3ygap2it + β4age2it + β5inf2it + β6effe2it + β7cor2it + vit (2) In which: i: country, i = 1-14. α, β: estimated coefficients t: year, t = 2000-2018 u, v: residuals of model Table 3.1. Variables in model Sign Variable Sign Variable pri Private investment ptra Trade openness pgex Public expenditure pol Political stability pfdi FDI pgrev Budget revenue pcre Private credit lgdpr Real GDP size rir Interest rate ygap Output gap g Economic growth rate age Dependent population inf Inflation effe Government effectiveness lab Labour force cor Mức độ kiểm soát tham nhũng sch Human capital Source: Author’s proposal based on literature review 3.2.2. Hypothesis of equation on the impact of public expenditure on private investment - Hypothesis H1: Public expenditure has the crowding in effect (positive impact) or the crowding out effect (negative impact) on private investment. - Hypothesis H2: FDI has positive impact on private investment. - Hypothesis H3: Private credit has positive impact on private investment. - Hypothesis H4: Interest rate has negative impact on private investment. 11 - Hypothesis H5: Economic growth has positive impact on private investment. - Hypothesis H6: Inflation has negative impact on private investment. - Hypothesis H7: Human resource has positive impact on private investment. - Hypothesis H8: Trade openness has positive impact on private investment. - Hypothesis H9: Political stability has positive impact on private investment. 3.2.3. Hypothesis of public expenditure equation - Hypothesis H10: Budget revenue has positive impact on public expenditure. - Hypothesis H11: Output size and gap has positive impact on public expenditure. - Hypothesis H12: Dependent population has positive impact on public spending. - Hypothesis H13: Inflation has positive impact on public expenditure. - Hypothesis H14: Institution has negative impact on public expenditure. 3.3. Research methods 3.3.1. Quantitative method The structural equation model in this thesis is regressed by 3SLS. 3SLS quantitative method is employed to address a part of research objective, focusing on the second research question about the impact of public expenditure on private investment in some Asian emerging and developing economies. This method is performed using Stata. 3.3.2. Qualitative method Qualitative method is conducted to analyze some characteristics and the impact of public expenditure on private investment in some Asian economies in general as well as in Vietnam particularly. The main methods of qualitative method in this study cover process-tracing, deductive research, descriptive research, and case analysis. 3.4. Research data Thesis’ data is collected from some emerging and developing economies in Asia, including Vietnam, Cambodia, Indonesia, Malaysia, Myanmar, Philippines, Thailand, China, Mongolia, India, Sri Lanka, Bangladesh, Pakistan. Hence, this is panel data from 14 countries, between 2000 and 2018 (19 years), creating 266 observations. Those are secondary data retrieved from the reports of international organizations such as WB, IMF, etc. Besides, data of Vietnam is collected from the reports of General Statistics Office, Ministry of Finance, Ministry of Planning and Investment, etc. 12 ANALYZING THE IMPACT OF PUBLIC EXPENDITURE ON PRIVATE INVESTMENT IN SOME ASIAN ECONOMIES AND VIETNAM 4.1. Situation of public expenditure and private investment in some Asian economies 4.1.1. Situation of public expenditure Public expenditure in the connection with GDP in some emerging and developing economies in Asia from 2000 to 2018 as below: Figure 4.1. Public expenditure/GDP in some Asian countries Source: IMF data Most countries in Asia faced the common situation that public spending rose in the period of 2008-2009. This is the recession in the world, marked general decline and difficulties observed in many economies. Some countries even paid attention to stagflation. The Governments occurred a series of solutions to support, resolve recession, and preventing inflation. Vietnam, Laos, Cambodia, Myanmar, Thailand, etc. were not also out of this trend. 5 10 15 20 25 30 35 40 45% Việt Nam Bangladesh Trung Quốc Indonesia Ấn Độ Campuchia Lào Sri Lanka Myanmar Mông Cổ Malaysia Pakistan Philippines Thái Lan 13 4.1.2. Situation of private investment Figure 4.2. Private investment/GDP in some Asian countries Source: WDI, WB In general, the majority of private sector in developing countries in Asia is micro, small, and medium enterprises in which are mainly micro and small ones. The main problem of those enterprises is capital and technology constraints. As a result, they have to face weakness in competitiveness. Also, the advantage of cheap labor gradually diminishes when many businesses begin to strictly require on the professional qualifications and skills of the employees. In addition, some countries give preference for state-owned enterprises, that making it more difficult and unfair for the private sector. 4.2. Empirical research results about the impact of public expenditure on private investment in some Asia economies 4.2.1. Descriptive statistics Table 4.1. Descriptive statistics Variable Observations Unit Mean Std. Dev. Min Max pri 252 9.81 2.15 4.69 15.06 pgex 252 % 21.60 5.81 11.41 40.52 pfdi 252 % 3.52 5.25 -37.15 43.91 pgrev 252 % 18.17 5.40 8.49 33.92 0 5 10 15 20 25 30 35 40 45 50% Việt Nam Bangladesh Trung Quốc Indonesia Ấn Độ Campuchia Lào Sri Lanka Myanmar Mông Cổ Malaysia Pakistan Philippines Thái Lan 14 Variable Observations Unit Mean Std. Dev. Min Max g 252 % 6.43 2.74 -1.55 17.29 lgdpr 252 11.72 1.92 8.18 16.20 ygap 252 % 0.35 5.21 -14.00 39.97 rir 252 % 3.94 5.00 -18.73 27.40 pcre 244 % 53.48 41.87 3.12 161.14 inf 252 % 6.17 6.32 -1.71 57.07 lab 252 8.02 1.72 4.55 11.28 sch 252 year 6.75 2.20 3.10 11.00 age 252 % 54.75 10.98 35.59 88.49 ptra 252 % 81.28 49.08 0.17 220.41 pol 252 % 29.38 21.33 0.47 88.36 effe 252 % 43.06 20.01 2.39 85.85 cor 252 % 31.05 17.04 0.47 68.69 Source: Author’s calculation 4.2.2. Correlation matrix between variables Equation 1 describes the influence of public expenditure on private investment. The relationship between variables in the model and private investment, in general, have statistical significance, but variable of economic growth. Nevertheless, only 5 variables have the same correlation as expected in research hypotheses. Public expenditure, which is the most important in the model, has positive impact on private investment. The sign of private credit, interest rate, inflation, and human resource also exist as predicted. Equation 2 shows the impact factors on public expenditure and most of those relationship have statistical significance. It is predicted that there is no multicollinearity since the correlations between those variables are quite low. The temporary absence of multicollinearity in both equations - up to this step - will help the model avoid errors, increasing the reliability and exactness of regression results. 4.2.3. Regression result of the model about the impact of public expenditure on private investment After testing and robusting some errors of the system of equations, regression result observes p-value of F-test at 0,000, which is smaller than α = 1%, proving model has statistical significance when explaining the influence of independent variables on 15 private investment and public expenditure. The regression results of OLS and 2SLS are added for references. Table 4.2. Regression result of equation 2 Dependent variable 3SLS OLS 2SLS pgex (b/se) (b/se) (b/se) pgrev 0.9399*** 0.9406*** 0.9406*** (0.040) (0.040) (0.040) lgdpr 0.2043* 0.1714 0.1714 (0.117) (0.119) (0.119) ygap -0.0147 0.0026 0.0026 (0.032) (0.033) (0.033) age 0.0785*** 0.077*** 0.077*** (0.020) (0.021) (0.021) inf 0.0337 0.0289 0.0289 (0.028) (0.028) (0.028) effe -0.0967*** -0.1036*** -0.1036*** (0.028) (0.021) (0.021) cor 0.1519*** 0.1585*** 0.1585*** (0.022) (0.022) (0.022) Constant -2.9272 -2.3891 -2.3891 (2.336) (2.384) (2.384) Observations 266 266 266 F-test (p-value) 0.000 0.000 0.000 R-squared 0.7922 0.7924 0.7924 Note: * p < 0.1, ** p < 0.05, *** p < 0.01 b/se: beta coefficient / standard error Source: Author’s calculation 16 Table 4.3. Regression result of equation 1 Dependent variable 3SLS OLS 2SLS pri (b/se) (b/se) (b/se) pgex 0.0397*** 0.0195*** 0.0197* (0.0115) (0.0093) (0.0118) pfdi 0.0236*** 0.0294*** 0.0294*** (0.0075) (0.0089) (0.0090) pcre 0.0123*** 0.0125*** 0.0125*** (0.0016) (0.0016) (0.0016) rir -0.0084 -0.0072 -0.0073 (0.0092) (0.0096) (0.0098) g -0.0138 -0.0136 -0.0135 (0.0160) (0.0166) (0.0167) inf -0.0315*** -0.0304*** -0.0305*** (0.0071) (0.0072) (0.0073) lab 0.8856*** 0.9023*** 0.9021*** (0.0386) (0.0393) (0.0400) sch 0.2824*** 0.2975*** 0.2974*** (0.0214) (0.0213) (0.0217) ptra -0.0106*** -0.0098*** -0.0098*** (0.0013) (0.0013) (0.0013) pol -0.0072** -0.0063** -0.0064** (0.0030) (0.0031) (0.0032) Constant 0.5916 0.6587 0.6583 (0.3991) (0.4101) (0.4104) Observation 266 266 266 F-test (p-value) 0.000 0.000 0.000 R-squared 0.9205 0.9206 0.9206 Note: * p < 0.1, ** p < 0.05, *** p < 0.01 Source: Author’s calculation From the regression result, the most crucial variable “Public Expenditure” (pgex) happens as expected when the estimated coefficient has p-value = 0,000 < α = 5%, 17 hypothesis H1 is accepted, so positive impact of this variable on dependent one Private Investment can be confirmed. This conclusion once again affirms complementary effect of public expenditure on private investment like the previous researches on developing in Asia (Greene and Villanueva, 1991; Furceri and Sousa, 2011; Erden and Holcombe, 2005; Gjini and Kukeli, 2012; Dao Thi Bich Thuy, 2014). Also, this result looks like the consequence of preliminary review of the general trend of 14 countries in the sample. At the same time, most of the remaining hypothesis turned out to be as expected; but variable of Economic Growth, Trade Openness, and Political Stability. Among the variables that meet the expectation, only 4 out of 6 variables are statistical significance including FDI, Private Credit, Inflation, and Human Source. 4.3. Situation of public expenditure and private investment in Vietnam 4.3.1. Public expenditure in Vietnam Figure 4.3. Situation of public expenditure in Vietnam Source: Annual budget settlement report, Ministry of Finance Total budget expenditure has increased, starting at VND 130.8 trillion in 2001. By 2017, this spending got slower than the same period last year, although it was higher than estimation. Nevertheless, government spending as a share of GDP was still deeply high compared to many other countries, about 27.9% in 2018 while the average of observed sample at 23%; only lower than China (32.8%), Mongolia (34.4%), approximately India (27.5%); those figures were below 24% in the rest of other countries. -20% -10% 0% 10% 20% 30% 40% 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 % Nghìn tỷ đồng Tổng chi ngân sách (nghìn tỷ đồng) Tốc độ tăng chi ngân sách (%) 18 Besides, structure of public expenditure almost did not change in porportion while current expenditure still dominated being over 50% almost year. The burden of current spending mainly comes from both salary payments (including pensions) and administrative management payments which account for about 30% of total budget expenditure, more than half of total current spending. Another problem is the expenditure for the state-sponsored mass organizations. Meanwhile, though being considered as main funding for basic infrastructure, the porportion of spending on development investment is quite modest. In 2001, this spending made up 30.8% of total budget expenditure, being the highest of the whole period of 2000-2017. After that, this figure kept decreasing gradually. This decline is related to the restructuring of public investment in the direction of continuously reducing the rate of state investment. Limited budget spending capacity pluses uncertainty in effectiveness of this expenditure will hardly meet the needs of economic development. Thus, borrowing is unavoidable. Moreover, if the entire loan is not used to offset the budget overspending but instead being provided to development investment expenditure, the gap in public expenditure structure could be partially improved, in spite of still very low compared to the demand for development of the country. 4.3.2. Private investment in Vietnam Private economic sector plays a crucial role on a series of aspects as contributing to the socio-economic development, promoting to establish business ecosystem, creating jobs, etc. The aggregate capital and GDP of private sector are pretty good as consistently leading in the economy. Nonetheless, the deatail value is not so optimistic. Firstly, business size is not large. Secondly, the labor productivity of the non-state sector tends to fairly bleak. Thirdly, difficulties come about in accessing capital. Fourthly, the level of technology is still low. Last but not least, business costs such as wages, transportation costs, pressure on capital costs, barriers to market entry, so on. cause rather obstacles for private capital. 19 Figure 4.4. Capital and GDP structure by economic sector in Vietnam Source: Vietnamese Statistical Yearbook 4.3.3. The impact of public expenditure on private investment in Vietnam Preliminary review for the Vietnam case, public expenditure has positive impact on private investment; but when public expenditure exceeded 30% of GDP, it starts to take place effects which hinder private investment. In fact, this above ratio only existed for a few years when the economy turned down, inflation surged in the period of 2009- 2010, 2013. Hence, public expenditure, in the general trend, still positively affect private investment in Vietnam. Figure 4.5. The relationship between public expenditure and private investment in Vietnam Source: Author’s model analysis results 0% 10% 20% 30% 40% 50% 60% 70% - 100 200 300 400 500 600 700 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Nghìn tỷ đồng Vốn nhà nước Vốn ngoài nhà nước Vốn FDI %GDP Khu vực nhà nước %GDP Kinh tế ngoài Nhà nước %GDP Khu vực FDI 7 7. 5 8 8. 5 9 9. 5 Lo g (P riv at e in ve st m en t) 22 24