Dynamic Response of Dependency Ratio on Government Expenditures in Indonesia

Abstract

The aim of this study is to see how government spending on education, health, and social security affects ratios in Indonesia. The third sector has a critical role to play in reducing the dependency ratio. It also aims to lower unemployment and poverty rates. This study uses the GMM panel data model. This model can determine the dynamic response of the ratio that comes from a number of variables. This study uses data from 33 provinces from 2010 to 2018. The results show that government spending in the education and health sectors has a positive effect on the dependency ratio, both in the short and long term. Social security has a significant effect on the dependency ratio in the long term, but not in the short term. Government spending in the education sector and health sector and social security sector have a positive and significant effect on disease and illness. The study's findings show a high level of poverty with a large standard deviation. The high ratio value is due to the large number of restrictions placed on a number of regions. Each province has made a significant contribution to overcoming these challenges, particularly in terms of the comparative ratio.

1. Introduction

Indonesia’s population growth in the last decade has shown an increase with a growth rate of 1.49 percent per year. The island of Java has the largest population density, with 57 percent of Indonesia’s 260 million people living there (Central Statistics Agency RI, 2016). This population is proportional to the number of people of working age (15–64 years). Indonesia currently has a relatively high productive age population, accounting for 66 percent of the entire population, with the remainder falling into the unproductive age category or being elderly. This situation indicates that Indonesia is in the midst of a demographic transformation. By 2035, Indonesia is predicted to reach the Demographic Dividend. According to census data from 2010, population growth in the under-15 age group is insignificant. This was not the situation throughout the 1970s and 1980s when the population of children under the age of 15 surpassed 60 million. In contrast, the productive age group (15–64 years) in 1970 amounted to 63 million people and experienced a significant increase. At the end of 2000, Indonesia’s population was 133 million, or which means an increase of 100 percent (Maryati, 2015).

The growth rate of the productive age population shows that Indonesia has experienced a demographic transition which has led to a demographic dividend. A demographic bonus or also called a demographic dividend or demographic gift means that economic benefits are caused by a decrease in the dependency ratio as a result of a reduction in the long-term fertility process (Painter & Lee, 2009). Dependence below 50 percent is an indicator of a demographic bonus (the number of productive age groups is higher than the number of non-productive age groups). Bloom and Williamson (1998), Ha and Lee (2016), and Lee (2003) conducted studies using reduced mortality and fertility as factors in the demographic transition itself. When the age group is shifted from non-productive to productive, investment that was previously used to fulfill the needs of the younger population can be diverted to meet the demands of economic development and family welfare (Bloom & Luca, 2016; Simon et al., 2012; Adioetomo, 2014).

The demographic bonus occurred in several countries in East Asia such as China, Japan, and Korea in the 1960–1990 period (Bloom & Luca, 2016; Bloom et al., 2011). This is indicated by the high income per capita in these countries. When the time period was extended (1960–2005), the data revealed strong demographic transition growth, except in Japan, where population growth was declining. Meanwhile, Asian countries such as the Philippines, Thailand, Singapore, and Indonesia have seen an increase in the working-age group during the 1980s, reflecting the impact of a bigger demographic transformation. In the years 1960–2005, population growth and the proportion of working-age people accounted for around 40% of Indonesia’s economic growth (Bloom et al., 2011). Mao and Xu (2014) stated that the demographic transition policy is an important policy made by the Chinese government, especially in social security which requires one child per family. Yoshino and Miyamoto (2017) conducted a study on how life expectancy affects economic performance along with effective fiscal and monetary policy in Japan. The findings revealed that there is a significant increase in the share of the population in productive age in response to increases in production, consumption, and investment, all of which enhance labor supply in the long run. Furthermore, a considerable increase in the proportion of the population of working age has an influence on wages and reduces the government’s social security cost. The efficiency of fiscal and monetary policy weakens as the proportion of retirees rises, according to this study (Heijdra & Romp, 2009; Gonzalez-Eiras & Niepelt, 2012; Dedry et al., 2016).

A high level of citizen independence and a low level of community interdependence are indicators of a successful demographic transition. The dependency ratio depicts the extent to which the elderly and children are reliant on the productive age group. The lack of financial autonomy and life fulfillment is the root of this dependency. The welfare of the family has a strong correlation with the reduction in dependency rates. The number of poor and unemployed people are the indicators. The high poverty rate will have an influence on the family’s ability to meet their daily necessities (Bjorvatn & Farzanegan, 2013; Dedry et al., 2016; Skirbekk et al., 2015). In the dependence ratio, employment opportunities are also at risk, which indicates that low employment opportunities (high unemployment) in society affect the level of dependency.

Indonesia’s high unemployment rate during the last two decades (2000–2019) has ushered in a demographic change. According to the 2010 population census, population growth among those under the age of 15 is insignificant. Between 1970 and 1980, there were 60 million people under the age of 15, and this population rose by no more than 10% till the year 2000 (approximately 3 million people). In contrast, the productive age group (15–64 years) in 1970 reached 63 million, and this figure increased to 100 percent by the end of 2000 (Bloom & Williamson, 1998). Increasing the productive age is an important component of economic strength, but it must be accompanied by a high level of education and skills. When looking at each province individually, the majority of the areas with high unemployment and dependency rates are in eastern Indonesia. Nusa Tenggara Timur, Sulawesi Tengah, Sulawesi Barat, Maluku, and Maluku Utara have the highest dependency rates. As indicated in Figure 1, the provinces with the lowest dependency rates are largely in Java, especially Jakarta, Yogyakarta, and Jawa Timur.

Figure 1: Dependency and Unemployment Rate in Each Province in 2010 and 2018

Figure 1 shows the dependency ratio (DR) and unemployment rate (UN) in each province in Indonesia for 2010 and 2018. In 2010, the unemployment rate (UN) in several provinces was above 10 percent. They are Jakarta, Jawa Barat, Banten, Sulawesi Utara, Sulawesi Selatan and Maluku. The high unemployment rate in Jakarta, Jawa Barat, and Banten originates from migrants from outside Java who are actually job seekers. This situation must be a top priority for the government, which should take various measures to create jobs so that Indonesia is prepared to deal with the demographic dividend, which is predicted to peak in 2028–2032. Job creation will benefit persons of working age that will increase due to demographic dividend. Fertility and death rates would be lower as a result of the demographic advantage (Edle et al., 2016; Ludwig & Vogel, 2010).

In the last decade, the dependency rate trend in Indonesia has decreased. During the years 2010 to 2018, the community reliance ratio was around 50% of the entire population on average. Aceh, Sumatra Utara, Sumatra Barat, Riau, NTB, Kalimantan Barat, Sulawesi Selatan, Sulawesi Barat, Sulawesi Tenggara, Maluku, and Maluku Utara are among the provinces with high dependency rates. Increased government spending, particularly for improving the quality of human resources, can help to reduce community dependency ratios. The Indonesian government’s Nawacita program (nine points on the development agenda) aims to improve self-reliance and minimize dependency ratios. Programs are focused on specific activities related to improving the quality of education, health (Ministry of Finance, 2018). In the last decade, the government has been aggressively facing this problem. This can be proven by the work program at the Ministry of Education, Health, and Society. Apart from this program, the government has also increased the budget allocation for these three sectors. The budget requirements in each province are carried out by looking at the supporting resources in each province. During the last decade, provinces with high population density have become the focus of this budget allocation. This is because provinces have high complexity compared to provinces with low population density.

Figure 2 depicts the allocation of the State Revenue and Expenditure Budget (APBN) for several sectors such as Education, Health, and Social Security in each Indonesian province between 2010 and 2018. In comparison to other sectors, the education sector received the highest part of the allocation. This is due to the government’s budget strategy, which compels the government to set aside 20% of the entire APBN for education. In short, this is an effort to increase education quality, with the primary goal of this budget being to improve education quality at the basic and secondary levels. Meanwhile, the healthcare budget is focused on enhancing the health of women and children, preventing and controlling diseases, and bolstering public health. The social security budget is aimed at reducing poverty by satisfying human needs. The amount of budget allocation is based on the needs and number of people. This results in the fact that the provinces that receive the largest APBN allocations are located in Java Island; Jakarta, Jawa Barat, Jawa Tengah, and Jawa Timur. Meanwhile, the regions that received the smallest budget allocations were eastern Indonesia. The APBN distribution imbalance generates a slew of concerns in various regions in eastern Indonesia that aren’t being addressed adequately, mostly in the areas of poverty, health, and educational possibilities. Indonesia is expected to increase the competitiveness of the nation in the international world by striving to accelerate national economic development. It is envisaged that by increasing the budget in these three areas, Indonesia will be able to improve its competitiveness in the worldwide market and accelerate national economic development.

Figure 2: Education, Health and Social Budget in 2010 and 2018

The demographic dividend is a very important source to achieve higher economic growth and prosperity in the future. To maximize the benefits of the demographic bonus, the federal budget should be allocated to improving the quality of education for Indonesians. Because of the demographic dividend, the function of the federal budget has been crucial for Indonesia so far. Therefore, it’s critical to look into the dynamic effect of the federal budget on demographic dividend returns. Therefore, this study is very important not only for the Indonesian government but also for providing empirical evidence for Indonesia as a developing country.

2. Literature Review

Shifting the structure of the public age is important in identifying the impact of demographic changes on economic performance (Central Statistics Agency RI, 2011). According to Bloom et al. (2011), the age structure of the population is an individual’s economic behavior, which differs at different stages of life. Throughout the life cycle, the age shift has a substantial impact on the performance of the national economy, particularly employment opportunities and savings levels. According to population projections, Indonesia would benefit from a demographic boost in 2030, with a population of roughly 180 million people of working age. Meanwhile, over 60 million people are in the non-productive age bracket. Thus, every 10 people of productive age will be responsible for 3–4 people of non-productive age. This situation affects people’s savings and prosperity (Frini & Muller, 2012). Data from the Central Statistics Agency RI (2016) states that a potential person to work is a person who can earn and earn at least 1 hour of income a week continuously. The population of working age in Indonesia is the population of productive age from 15 years to 58 years. The age of 58 years is the age limit for working and is the beginning of retirement. In developed countries, the working-age limit is up to 65 years. The United Nations has established a working-age group from 15 years to 64 years (Rodriguez & Brunswick, 1988).

On the other hand, most of the population of productive age tends to continue their education rather than working for future investment. Cervellati and Sunde (2015) stated the need for work demand balance so that work productivity can be stable. According to Pekarek (2018), while measuring the mortality ratio, the dependency ratio is closely tied to the birth rate and the life expectancy ratio. The government must improve the quality of human resources as a development agent by investing in education, health, communication skills, technology skills, and work skills. Aside from that, it’s also critical to ensure work opportunities (Ferreira & Dos Santos, 2013).

A study was conducted by Cervellati and Sunde (2015) using the variables of death, education, birth, and income and how they affect the economy. The results of this study indicated that a high life expectancy will have an impact on slow economic growth. This is solely due to an increase in government spending. Li and Zhang (2015) conducted a study on the dependency ratio of early childhood and elderly people in China. Their results showed that the early age group had a high level of dependence caused by their inability to work. Meanwhile, the elderly are categorized into two groups, namely the productive elderly group and the unproductive age group. Because of their experience and talents, the productive elderly group is still engaged and has a high income. The non-productive elderly group, on the other hand, tends to rely on the productive group for income, as their pensions and savings are their only source of income. Efforts to minimize the dependency ratio of the community have also occurred in industrialized countries. The reason is government spending on education, health, and social security is an effort to improve the quality of human resources so that they have the ability to work (produce) and earn better income (Simon et al., 2012). By increasing income, it is hoped that the people’s welfare will also increase. At the same time, it means the poverty rate is decreasing. For this reason, the government has an important effort in alleviating poverty. Another study revealed that the dependency ratio is closely related to economic growth.

Sinnathurai (2013) conducted research on the correlation of poverty, economic growth, and employment opportunities with dependency ratio in developing countries by taking 40 samples from countries in Asia, Latin, and Sub-Saharan Africa. The results showed that the age dependency ratio had a serious effect on poverty. Poverty has a relatively high effect on the level of dependency ratio. Although employment in the industry has a negative correlation with poverty, it does not have a significant correlation with poverty. Another result is that economic development, poverty, and employment in the industrial sector have a significant effect on age dependency rates in Model two, and can be carried out consistently using economic theory. Therefore, human resources are one of the important factors in accelerating national economic growth. The government in this case has a strategic role in increasing human resources, particularly in allocating budgets for education, health, and social security. Investment in human resources is believed to be able to increase people’s opportunities to earn income and create prosperity. Improving the quality of education, health and social security have an impact on human development targets. Human development itself has been prioritized in the fields of education, health, and social security as the main indicators of economic growth.

Another solution to lowering the people’s dependency ratio can be done by increasing the optimization of government spending. The optimization of expenditures continues to have a direct impact on society through budgeting in the fields of education, health, and social security. The health budget is intended to improve the quality of health services through the convergence of programs and safety for the elderly and pregnant women (Boldrin et al., 2015). Basuki (2020) stated that health spending in Indonesia has not been on target. The government has made the revitalization of the National Health Insurance a priority for citizens to receive quality treatments (De Meijer et al., 2013). The private sector must have a role in improving education quality, whereas the government’s role is primarily focused on satisfying elementary and secondary needs (Damodar et al., 2021). Apart from health services, concern in social security programs is actually an effort to meet the needs of the community, especially those who live in poverty. The budget in this sector plays a positive role for household savings and economic growth, such as in Germany for example. However, it has a negative impact on fertility. Meanwhile, the workers in Germany, especially women, prefer to pursue careers rather than have children. As a result, they do not have funds to raise children and do not receive a pension. In the end, the government can expand social security coverage for childbirth, baby/child care, as well as pensions, and elderly care (Cigno et al., 2003). The effect of public spending on the economy is also obtained from the study of Yoong et al. (2020), who stated that in the long run public spending has an effect on economic growth.

3. Research Methods and Materials

3.1. Research Model

This study uses a panel data GMM model to examine the dynamic response of the dependency ratio to government spending in Indonesia. Provincial-level data from 2010 to 2018 is used. The variables in this study are Dependency Ratio (DR), Education Budget (EDUF), Health Budget (HEAF), and Social Security Budget (SOSF), Number of Poor People (POV), and Unemployment Rate (UN). This study uses panel data from each province from 2010–2018. Descriptive ana-lytical and econometric models are used to determine how budgets in the fields of education, health, poverty, and unemployment in a descriptive and econometric way affect the dependency ratio of the people in relation to the demographic transition in Indonesia. The Generalized Method of Moment (GMM) approach is also used to address bias and inconsistencies in research. Arellano and Bond AR suggest the use of the GMM approach for two reasons: GMM is a general estimator for demonstrating a framework for valuation and as a simple choice compared to other estimators. Several tests with assumptions are required to produce consistent estimation analysis results. There is no valid correlation set of error terms, instruments, or limiting moments. Arellano and Bond AR tests are performed to detect whether or not there is an error term correlation series. In addition, Hansen and the Difference Test were conducted to determine the validity of the instrument and the moment conditions used in the study. This model is adjusted to the research objectives so that the estimated variables are government expenditures in the fields of education, health, social security, poverty, unemployment, and community dependency ratio. The general model of a dynamic data panel is illustrated in a simple and dynamic form as shown below:

Y_it=α_o+Y_it-1+X_it+ε_it       (1)

The first GMM model (1) in this study is:

DR_it=α₀+α₁DR_it–1+α₂LnEDUF_it+αLnHEAF_it+α₄LnSOF_it+α₅POV_it+α₆Un_it+ε_it       (2)

Then, the second GMM Model can be formulated as follows:

DR_it=β₀+β₁DR_it–1+β₂LnEDUF_it+β₃LnHEAF_it–1+β₄LnSOF_it–1+β₅POV_it+β₆UNit+ε_it       (3)

Condition:

DR_t = Public dependency ratio

EDUF_t = Budget in education

HEAF_t = Government spending in the health sector

SOSF_t = Government spending on social security

POV_t = Total poverty rate

UN_t = Unemployment rate

4. Results and Discussion

The dependency ratio affects government spending, both for household spending and government spending. Increased government spending on education, health, and social security is intended to improve public health and welfare (Ministry of Finance, 2017). The dependency ratio also requires a relatively large allocation of government spending. Increasing government spending is an effort to improve people’s quality of life.

This study on the dependency ratio on government expenditures attempts to identify policies that may be implemented to alleviate the high reliance on community and budget contributions in dealing with these constraints, particularly the dependency ratio in each province. In addition, the efforts of the working–age group to find work are also important, so that it will have an effect on reducing poverty and unemployment at the same time. It also needs to emphasize the government’s efforts to provide employment opportunities.

Table 1 shows the research variables that can affect the dependency ratio of the people in Indonesia. The research variable that is not stable in this study is the poverty rate with a high standard deviation value. The average score for the public DR variable in Indonesia is 51 percent with a maximum value of 70.6 percent and a minimum value of 37.4 percent. The high DR value is due to the high dependency ratio in several provinces, such as Nusa Tenggara Timur, Maluku, and Utara Maluku. Meanwhile, provinces in Java, such as Jakarta, have a low dependency ratio. EDUF, HEAF, and SOSF variables are important variables intended to create independence and reduce the dependency ratio of the community. The EDUF (education) variable is the variable that gets the most budget compared to other variables such as HEAF (health), SOSF (social security).

Table 1: Descriptive Statistics

The budget for education is around Rp. 3.6 trillion, the budget for health is around Rp. 985 billion, and the budget for social services is around Rp. 331 billion. An average of 853 thousand persons is affected by the variable poverty rate (POV), with a maximum of 5.52 million. The unemployment variable (UN) has an average of 5.55 percent and a maximum rate of 13.7 percent. This is due to disparities in job opportunities in various regions in Indonesia, especially between regions on the island of Java (except Banten) and other areas outside Java Island, especially provinces in eastern Indonesia.

Based on the results of the analysis, the estimated community dependency ratio (DR) in Indonesia is shown in Table 2. Meanwhile, the education budget (EDUF) and health budget variables have a negative coefficient on the community dependency ratio (DR).

Table 2: Estimation of the GMM Model

Note: () Prob. *sig 10%, **sig 5%, ***sig 1%.

This means that these variables will reduce DR when budget allocations can actually be increased and focused on underdeveloped provinces. Meanwhile, the social security budget (SOSF) variable has positive and negative coefficients. This means that DR can increase in the short term, and DR can decrease in the long term. These factors are caused by programs that are not running optimally and programs that are incidental. Meanwhile, the POV and UN variables have positive coefficients that can explain the increase in the poverty rate (POV) and the unemployment rate.

Table 2 shows 2 formula models that affect the dependency ratio of the community which systematically and differently have a significant effect with the following coefficient values:

Model 1; The previous model was developed with the result of the DR variable of 0.74 (74 percent) for the GMM system, and the Diff GMM value achieved is 0.62 (62 percent).

Model 2; developed by including the lag model, especially the health budget variable (LnHeaf (–1) and the social security variable (LnSOS (–1)). The obtained DR coefficient is 0.77 (77 percent) for the system and the Diff GMM value achieved is 0.56 (56 percent).

In Model 1, it can be concluded that the health and social security budget variables do not have a significant effect on the ratio of the burden of dependency of the community. Whereas for model 2 which uses lag, the health budget and social security variables have an effect on the decrease in the ratio of the burden of dependency of the community. This means that fiscal policy must be measured against the previous year’s benchmark to have an impact the following year. Fiscal sustainability is needed to reduce the dependency ratio of the people in Indonesia.

In addition, at the end of the analysis, it can be seen that the Hansen and Arellano Bond (AR2) statistical test rejects the null hypothesis in confirming the significant estimation results (without bias and autocorrelation) in identifying variable instruments. The results showed that with the GMM method, government spending on education is an important indicator in reducing the dependency ratio of the community as well as the government sector in the health budget. However, the social security budget does not have a significant effect on the dependency ratio of the community because the proposed program is limited to certain groups such as the elderly, people with disabilities, and the homeless.

Indonesia is in the midst of a demographic transition, with significant growth in the productive age group and insignificant growth in the non-productive age group. Indonesia is predicted to experience a demographic bonus in 2035. Facing this demographic dividend, the quality of human resources must be improved so that people have the ability to compete in the job market. This means that the government plays a role in various sectors such as education, health, and social security so that it can reduce poverty, unemployment, and at the same time reduce the dependency ratio of the community.

The community dependency ratio is one of the demographic indicators in measuring the demographic transition in Indonesia. From the results of the analysis of model 1, it can be concluded that the health budget and social security variables do not have a significant effect on the ratio of the burden of dependency of the community. The indicator is that so far the budget allocation has not been optimal in underdeveloped provinces, especially those that are lagging behind in terms of prosperity. This means that it does not have a significant effect on reducing poverty, unemployment, and the community dependency ratio. As for the results of the analysis of model 2 using lag, the variables of health and social security have influenced the decrease in the ratio of the burden of dependency of the community. In short, fiscal policy must be effective and can be measured from the previous year. At the same time, fiscal policy must also be sustainable. In other words, there needs to be a sustainable fiscal policy to reduce the dependency ratio of the people in Indonesia. Apart from that, the government must also pay attention to economic equality so that there is no disparity between provinces, or at least such disparity can be minimized. The Hansen and Arellano Bond (AR2) statistical test rejects the null hypothesis in ensuring a significant estimated value (without bias and autocorrelation) in identifying the variable instrument.

5. Conclusion

The results of the GMM method indicate that government spending on education is an important indicator in reducing the dependency ratio of the community as well as in the health sector. On the other hand, the social security budget has not had a significant effect on the community dependency ratio. Based on the findings of this study, the government should be able to conduct further research and establish optimal programs so that the budget may be spent effectively, benefiting all people in remote places and reducing the number of Indonesian provinces that are undeveloped. In addition, the education budget should be more focused on developing human resources, while the health budget should be more focused on social security and improving the quality and life expectancy of the community.