The Effect of Government Expenditure in Education, Health and Social Assistance Spending on the Poverty Rate of Yogyakarta Special Region Province

This study has the final results to determine the relationship between government spending in education, health and social assistance spending on poverty rates in the Special Region of Yogyakarta. This secondary data-based research has the form of a time series in 2012-2021, then the method used is multiple linear regression analysis with the OLS (Ordinary Least Squares) method. The results showed that programs on education implemented by the government of the Special Region of Yogyakarta through variable government spending in education were right on target and able to reduce poverty. but, variable government spending on health has not been able to reduce poverty and social assistance spending will increase poverty due to the effects of government dependence.

This study has the final results to determine the relationship between government spending in education, health and social assistance spending on poverty rates in the Special Region of Yogyakarta.This secondary data-based research has the form of a time series in 2012-2021, then the method used is multiple linear regression analysis with the OLS (Ordinary Least Squares) method.The results showed that programs on education implemented by the government of the Special Region of Yogyakarta through variable government spending in education were right on target and able to reduce poverty.but, variable government spending on health has not been able to reduce poverty and social assistance spending will increase poverty due to the effects of government dependence.

INTRODUCTION
Economic development is an important prerequisite of a developed country, but in the process of achieving economic development there are various obstacles that can hinder the process.Poverty is a complex problem that can interfere with the process of achieving economic development.Various approaches and theories that have been implemented to reduce poverty are currently still not perfect, this is because poverty is not only about the low economy of the community but there are multidisciplinary aspects that must be known (Arifin, 2020) According to (Suharto, 2014) If there is social inequality in a group of people or individuals to enjoy public goods it will give birth to poverty.Equality is essential in achieving economic development.Every individual has the right to receive proper medical facilities, education, and standard of living.If there is a difference in enjoying these sources of capital, it will certainly lead to poverty, Therefore, government intervention is needed to overcome this, one of the efforts made by the government is government spending specifically allocated to overcome differences in various accesses, so that equality can be created.
When viewed statistically, the Yogyakarta Special Region Province has the highest number of school expectation rates in Indonesia, this means that the Yogyakarta Special Region Province has quality human capital.but the Yogyakarta Special Region Province is still not free from the problem of poverty.The problem of poverty is still a concern for the government of the Special Region of Yogyakarta.The poverty rate in Java Island in 2012-2021 is presented as follows: Table 1 14,98 14,44 13,58 13,32 13,19 12,23 11,19 10,58 11,84 11,25 YOGYAKARTA S R 15,88 15,03 14,55 13,16 13,1 12,36 11,81 11,44 12,8 11,91 EAST JAVA 13,08 12,73 12,28 12,28 11,85 11,2 10,85 10,2 11,46  Source : Statistics Indonesia Table 1 shows that Yogyakarta Special Region has the highest poverty rate in Java, but statistically the poverty rate of Yogyakarta Special Region decreases every year.According to Statistics Indonesia, the number of poor people in Yogyakarta Special Region Province in 2012 reached 565.70 thousand people.Until 2019 the number of poor people in Yogyakarta Province amounted to 448.47 thousand people, which means that the number of poor people has decreased by 117.23 thousand people.However, the poverty rate in 2020 increased with the value of the development of the poverty rate by 4.96% and the number of poor people rose to 475.72 thousand people, this increase in poverty rate occurred due to the Covid-19 plague (BPS, 2022).
People who fall into the poor category find it difficult to achieve an increase in human resources because they have limited space for movement.The daily expenses of the poor tend to be used to meet food needs, so needs other than food are often less prioritized.Conversely, people who are categorized as capable who have more free space, so that they can get opportunities for education, health, and other potential opportunities.The existence of a gap between the poor and rich regarding access to public service facilities is a responsibility for the government, which if not overcome will lead to poverty.
The Regional Revenue and Expenditure Budget (APBD) in its implementation is a tool in public policy that can be the key to improving facilities and community welfare.The allocation of government expenditure or spending is very strongly related to improving the quality of human capital, especially in the context of reducing poverty (Riva et al., 2021).Through government budgeting such as in the education, health and social assistance sectors, it is expected to have an impact in efforts to equality and reduce poverty.
The government's efforts to improve the education sector can be seen through government spending specifically budgeted for education.The following are government expenditures in the education sector of the Special Region of Yogyakarta in 2012-2021 in thousands of rupiah:  Rp 121,610,779,193.00compared to the previous year 2014 amounted to Rp 295,423,870,512.00.In 2021, government expenditure in the education sector of Yogyakarta Province experienced the highest increase of 365.26% with total government expenditure in education of IDR 1,855,261,706,000.00,compared to the previous year, 2020 of IDR 398,847,843,000.00.
Health is closely related to poverty.When an individual has good health then his productivity will increase.So it can be said that health contributes to the increase in income of an individual.The government's efforts in improving public health and improving health facilities and infrastructure are reflected through government spending in the health sector.The following are government expenditures in the health sector of the Special Region of Yogyakarta in 2012-2021 in thousands of rupiah: Social assistance spending of the Special Region of Yogyakarta Province can be seen through Figure 3.In 2016 social assistance spending experienced the deepest decrease of minus 78.99% with a value of Rp 1,919,000,000.00,where the previous year 2015 social assistance expenditure issued amounted to Rp 9,134,490,000.00.Meanwhile, the highest increase in social assistance spending in Yogyakarta Province occurred in 2021 with a development value of 1,413.40% with a total social assistance expenditure of IDR 22,580,000,000.00,where the previous year 2020 social assistance spending was IDR 1,492,000,000.00.

LITERATURE REVIEW Poverty Rate
According to (Suharto, 2014) among the many theories and approaches to understand more about poverty, there are two paradigms of poverty, namely the neo-liberal paradigm and social democracy.
1. Neo-liberal paradigm In this paradigm, the primary targets in seeing poverty are mechanisms in the market and individuals.This approach places one's freedom as an important part of society.Thus, in understanding poverty, this approach provides an understanding that poverty is an individual problem caused by the choices of the individual himself.In this approach, market forces are the most important key to solving the problem of poverty.This is because high economic growth and expanded market forces will reduce poverty.
In looking at poverty, the neo-liberal paradigm does not place the poor as objects but only sees poverty in terms of income so that it becomes a weakness of this paradigm.Therefore, various forms of poverty that occur in the community receive less attention, especially poverty caused by social problems is less considered.But with this paradigm poverty can be measured easily.

Social democracy
This theory sees poverty as a structural problem rather than an individual one.Differences and inequalities in society in enjoying public goods can give birth to poverty.This approach considers poverty as a result of limited access for certain individuals or groups.Equality is emphasized more in this theory, which is essential for gaining freedom and independence.Equality can only be achieved if everyone has and is able to obtain potential resources for themselves, such as sufficient income, good education, and good health.Freedom here means that each individual is free to determine the path of their choice.Governments must work to ensure that everyone can participate.One of the weaknesses of this theory is the emergence of a large dependence on the government.However, in this theory looks at the other side of the problem of poverty, namely the structures and institutions that have prevented some groups in society from gaining access.

Government Expenditure on Education
Education budget or so-called government expenditure in the field of education is the allocation of government budget for activities in the field of education which must be issued at least 20% of the APBN and / or APBD as formulated in the 1945 Constitution article 31 paragraph 4. Education expenditures carried out by the government have benefits, including increasing school enrollment rates, improving the quality of education, increasing the length of schooling, and improving the smooth running of educational activities (Rahim et al., 2021).
Research conducted by (Hidayat & Azhar, 2022) The education budget is allocated to build educational infrastructure, increase the number of educators, and other educational expenditures.Large education budget expenditures by the government can help efforts to reduce poverty.
H1: Government expenditure on education has a negative influence on poverty rates

Government Expenditure on Health
Government expenditure in the health sector is a budget used by the government to pay obligations in the health sector with a health budget allocation of at least 10% of the regional budget in accordance with Article 171 paragraph (2) of Law number 36 of 2009 concerning Health.In Law Number 36 of 2009, it has been outlined that the government's obligations in the health sector include public services.Every Indonesian citizen has the right to access health services fairly in order to realize the highest level of welfare.
Basically, government spending is closely related to efforts to realize the development of the country.Starting from infrastructure development or human resource development.This government spending in the future will have a direct effect on economic growth.Among many expenditures, the government is required to fulfill various basic rights of the people among which the most important is health.The health sector is a field that has large-scale influence.The government is required to provide subsidies for the provision of medicines and vaccines, subsidize the implementation of public health facilities (puskesmas), and even recruit workers.Therefore, efficiency in this sector will have a significant impact on the efficiency of the national economy (Ridwan & Nawir, 2021).
H2: Government expenditure on health has a negative influence on poverty rates

Social Assistance Spending
In the 1945 Constitution and Pancasila, the state is responsible for protecting the entire Indonesian nation, improving people's welfare, and safeguarding it from social problems.To do this, the State Budget (APBN) will be used.The government has an instrument called social assistance spending in the state budget to improve social welfare and prevent social risks.
In the Regulation of the Minister of Home Affairs Number 39 of 2012 Article 1 paragraph 15, social assistance has the definition of assistance to the community by the government in the form of goods to individuals or social groups who have a short time and are selective.Meanwhile, the definition of social assistance according to the Regulation of the Minister of Finance of the Republic of Indonesia Number 81 / PMK.05 / 2012 concerning Social Assistance Spending at Ministries / Countries / Institutions is defined as spending in the form of money, goods by the central government or local governments to the community in order to protect them from social risks and improve welfare.Social risk itself is an event that can result in a person or group which, if not helped, can result in worse conditions and living in unnatural conditions.(Rarun et al., 2018) Stating that through social assistance carried out by the government will be able to reduce the economic burden of people who fall into the category of poor people, this social assistance can be through the form of money or goods by the government which will provide benefits to improve community welfare reduce poverty.
H3: Social assistance spending has a negative influence on poverty rates From the hypothesis described above, this study has the following conceptual framework: Figure 4. Conceptual Framework

Data Types and Sources
The research is based on secondary data, with the source data coming from publications of the Directorate General of Financial Balance of the Ministry of Finance of the Republic of Indonesia and the Statistics Indonesia.Researchers use data that has quantitative properties with the nature of time series data or data with a certain time span.This study uses annual data, from 2012-2021 with the variables used are Poverty Rate, Government Expenditure in Education, Government Expenditure in Health, and Social Assistance Expenditure in the Special Region of Yogyakarta Province.

Definition of Research Variables
The variety of variables in this study is grouped into two parts, namely: a. Dependent variable The dependent variable is a bound variable that will be affected as a result of the independent variable (Sugiyono, 2019).The Poverty Rate (Y) which is the dependent variable to be studied has a detailed value from the comparison of the number of people who fall into the poor category with the population of Special Region of Yogyakarta Province starting from 2012-2021 and is expressed in percent form.

b. Independent Variables
The independent variable is a source of effect that causes the dependent variable to change (Sugiyono, 2019).In this study, Government Expenditure in Education (X1), Government Expenditure in Health (X2), and Social Assistance Expenditure (X3) in Special Region of Yogyakarta Province is used as an independent variable with a period of 2012-2021 by stating data in the form of thousands of rupiah.

Data Analysis Methods
This research is a quantitative study, by analyzing how far the effect of independent variables on one dependent variable then the type of analysis used is multiple linear regression analysis.Yogyakarta Special Region Province was chosen as the object of research using time series data for ten years (2012-2021), while the software used to assist data analysis is SPSS 26 software.

Classical Assumption Test
Classical assumption testing is an important prerequisite before continuing regression analysis to detect the occurrence of symptoms in research data, consisting of:

Normality Test
Variables that are normally distributed or not will be detected using the normality test.Well disseminated data is an important prerequisite in research (Ghozali, 2018).The Kolmogorow Smirnov test can be used in determining the residual distribution, with the decision if the prob value is more than 0.05, the regression model distribution is normal, and if the prob value less than 0.05, the regression model distribution is abnormal.

Multicollinearity Test
The absence of relationships between independent variables is the goal of the multicollinearity test (Ghozali, 2018).The tolerance value and variance inflation factor (VIF) value are references in this test.If the tolerance value is more than 0.10 and the VIF value is less than 10, it is considered that there is no multicollinearity disorder.

Heterokedasticity Test
The heterokedasticity test is carried out to determine whether there are differences in variation between residuals from study to another study (Ghozali, 2018).The determination of heterokedasticity symptoms can be known by looking at the scatterplot graph, if it does not show a regular pattern then it is said to pass.

Autocorrelation Test
Autocorrelation symptoms are the relationship between the residual of one observation with the residual of another observation (Ghozali, 2018).The Durbin-waston test is used in this test.With reference to the following conclusions: 1. 0 < d <dl indicates H0 does not experience positive autocorrelation symptoms with conclusions not accepted.2. d1 ≤ d ≤ du showed that H0 did not experience positive autocorrelation symptoms with the results of the conclusion could not be decided.
3. 4-d1<d< 4 indicate H0 does not experience symptoms of negative autocorrelation with the conclusion not accepted.4. 4-du≤d≤4-d1 showed that H0 did not experience negative autocorrelation symptoms with the results of the conclusion could not be decided.5. du<d<4-du showed H0 did not experience negative or positive autocorrelation symptoms with the results of the conclusion accepted.

Multiple Linear Regression Analysis
Linear regression analysis aims to find out the amount of influence of more than one independent variable on one dependent variable (Ghozali, 2018).(Sugiyono, 2019) Stating the formula as follows: Y= α+β1X1+β2X2+ β3X3+ e Information: Y: Poverty Rate (percent) A, B1, B2, B3 : Regression Coefficient X1: total government spending on education X2: total government spending on health X3 :total social assistance expenditure e: error Coefficient of Determination Test (R²) (Ghozali, 2018) The R2 test is used as an explanation of whether the independent variable under study has explained the dependent variable to what extent, on a scale of 1 to 0.

F Test
The F test aims to explain whether all the independent variables together have an influence on the dependent variable.With the result of the decision probability F below 0.05 is considered influential (Ghozali, 2018).

t Test
The t test has the intention to explain whether the individual independent variable has an influence on the dependent variable.With the result of the decision probability t below 0.05 is considered influential (Ghozali, 2018).

Figure 5. Normality Test Results
Source: SPSS 26 results The Normality Test which was carried out with the intention of knowing whether the data spread normally obtained results with the Kolmogorow Smirnov method of 0.200>0.05.The test results show a significant value above 0.05 which indicates that the data is spreading normally.

Multicollinearity Test
Table 2 In the table above shows the output of the multicollinearity test, all independent variables namely Government Expenditure on Education (X1), Government Expenditure on Health (X2), and Social Assistance Expenditure (X3) have a tolerance value greater than 0.1 then all VIF values are less than 10.In this case, it can be concluded that there is no multicollinearity disorder by all independent variables.

Figure 6. Heterokedasticity Test Results
Source: SPSS 26 results In figure 2 regarding the results of the heterokedasticity test, it is clear that there are no shapes or patterns manifested in the picture.All points are scattered at the top and bottom of the x and y axes.So the conclusion is that in the regression model in the study there was no heteroscedasticity disorder.

Figure 7. Autocorrelation Test Results
Source: SPSS 26 results Based on figure 3 shows a DW value is 2.045.Then the dL value is 0.5253, dU 2.016, 4-dU 1.984, 4-dL 3.4747.So that the Durbin-Waston curve is created as follows: Figure 8. Durbin-Waston curve From the curve above, it shows that the Dw test value keeps the position between 4-dU and 4-dL, so it is concluded that there are no negative autocorrelation symptoms but are in the area where the results of the conclusion cannot be decided or doubted, so it is continued with the Runs Test: (2-tailed) is 0.314, which is greater than 0.05.So that it can be concluded that the regression model in this study does not experience symptoms of autocorrelation.

Multiple Linear Regression Analysis
In this study, the formula for multiple linear regression equations was created as follows: TINGKATKEMISKINAN = 36,783 -1,439PENGPENDIDIKAN + 0,565PENGKESEHATAN + 9,824BELBANSOS The R Square value is known at 0.914 which explains that the contribution of the influence of the variables Government Expenditure in Education (X1), Government Expenditure in Health (X2), and Social Assistance Expenditure (X3) on the Poverty Rate (Y) is 91.4% and the remaining 8.6% is another variable that is not explained in this study.It is known that the value of F count is 21.330 and F table is 4.76 then F count is 21.330 > F table is 4.76 and the probability or significant value shows 0.001 < α = 0.05.Therefore, it can be concluded that the variables Government Expenditure on Education (X1), Government Expenditure on Health (X2), and Social Assistance Expenditure (X3) have a simultaneous influence on the Poverty Rate (Y).Based on the output in the above results it is known: 1.The Sig. value for the variable X1, namely government expenditure in education is 0.002<0.05,indicating a partial influence on the poverty rate.2. The Sig. value for variable X2, namely government expenditure in the health sector is 0.243<0.05,indicating no partial effect on the poverty rate.3. The Sig. value for variable X3, namely social assistance spending is 0.002<0.05,indicating a partial effect on the poverty rate.

DISCUSSION
The Effect of Government Expenditure on Education on Poverty Rate in Yogyakarta Special Region Province In the test results above, it was found that government expenditure in the field of education issued by the Yogyakarta Special Region Province had a significant relationship and negatively affected the poverty rate.This means that government spending on education has been able to reduce poverty.supported by research by (Hidayat & Azhar, 2022) and (Ayu & Prabowo, 2021) which states that government investment in education will help poverty reduction efforts.
The underprivileged on average have less opportunities to get education and are constrained materially to continue their education to an advanced level.Through government programs with education spending, there will be many poor people who can improve the quality of human capital and get out of poverty.Government expenditure on education in the Special Region of Yogyakarta Province has been allocated through appropriate programs, which then have a major role in the economic welfare of the community and nation.

The Effect of Government Health Expenditure on Poverty Rate in Yogyakarta Special Region Province
In the results of the interpretation of the SPSS output above, it can be seen that the independent variable of government expenditure in the health sector has not had a significant influence on the poverty rate in the Special Region of Yogyakarta.These results are supported by research (Fithri & Kaluge, 2017) which both produce results that the effect of government spending in the health sector on poverty rates is positive, not significant.
Spending allocated by the government to the health sector should reduce poverty, but based on testing government spending in the health sector actually obtained the opposite result.This is due to the fact that government expenditure in the health sector in the Special Region of Yogyakarta Province which should get an allocation of at least 10 percent of the regional budget in accordance with the rules of Article 171 paragraph (2) of Law number 36 of 2009 concerning Health but get realization tends to be lower than 10 percent.(Sumiyarti, 2022) which concludes social assistance spending will increase poverty.
In its implementation, this social assistance program will have a negative impact on the social aspects of the community such as mental and behavioral.Social assistance programs will cause poor people to have a sense of dependency and reduce their efforts to get out of poverty without government assistance (Armoyu, 2013)

CONCLUSIONS AND RECOMMENDATIONS
1.The results of the research analysis found that the variable Government Expenditure in Education has a significant influence and has a negative relationship with the variable Poverty Level in the Special Region of Yogyakarta Province.This reflects that the provincial government of the Special Region of Yogyakarta has implemented education programs that are right on target and have a significant impact on reducing poverty rates in the Special Region of Yogyakarta. 2. From the results of the analysis, it shows that the variable Government Expenditure in the Health Sector does not have a significant effect on the variable Poverty Rate.In reality, government expenditure on the health sector in the Special Region of Yogyakarta Province has a smaller portion than it should be, namely 10% according to Article 171 paragraph (2) of Law number 36 of 2009 concerning Health.Therefore, Government Expenditure in the Health Sector has not been able to reduce the Poverty Rate in the Special Region of Yogyakarta. 3. Referring to the results of the study, it shows that the variable Social Aid Expenditure has a significant influence and is positively related to the variable Poverty Rate in the Special Region of Yogyakarta.This can happen because social assistance spending issued by the government impacts dependence on the poor and increases the burden on the government.

ADVANCED RESEARCH
1.It is expected that the provincial government of the Special Region of Yogyakarta will continue to maintain and increase government spending on education.Because this has been proven to be able to help and reduce poverty that occurs in the Special Region of Yogyakarta. 2. Increasing government spending in the health sector of Yogyakarta Special Region Province needs to be done.Because the budget in the health sector is still unable to contribute to efforts to reduce poverty rates in the Special Region of Yogyakarta.Improving health facilities and facilities will certainly have an impact on community welfare, which in turn can reduce poverty.3. The Provincial Government of the Special Region of Yogyakarta should pay more attention to social assistance spending.Because social assistance in the form of money, goods, and services will risk increasing dependence for the community.

Figure 1 .
Figure 1.Government Expenditures in the Education Sector of the Special Region of Yogyakarta in 2012-2021 Source: DJPK 2023, (processed)

Figure 2 .
Figure 2. Government Expenditures in the Health Sector of the Special Region of Yogyakarta in 2012-2021 Source: DJPK 2023, (processed)

Figure 11 .
Figure 11.Test Results t Source: SPSS 26 results . Poverty Rate in Java Island in 2012-2021 The value of the BELBANSOS Coefficient of 9.824 shows that Social Assistance Spending (X3) has a positive effect, it can be concluded when the variable Government Expenditure in the Health Sector (X2) increases by one rupiah can increase the Poverty Rate (Y) by 9.824 percent.

Table 4 .
Realization of Government Expenditure in the Health Sector of the Special Region of Yogyakarta in 2012-2021