Determinants of Economic Growth, Poverty, and Unemployment: A Path Analysis Study

The main objective of this study is to reveal the role of regional original income, development expenditures, and regional minimum wages for economic growth, poverty, and unemployment in this research. A quantitative approach with path analysis was used in this study, which was in Cities and Regencies in Bali Province for the 2015-2021 period. A sample of 62 was analyzed using multiple linear regression methods. The results of the study state that Regional Original Income (PAD), development expenditures, and regional minimum wages have a significant and positive effect on economic growth. PAD and regional minimum wages have a significant and positive effect on poverty, and development spending has a significant and negative effect. PAD does not affect unemployment. Development spending and regional minimum wages significantly and positively affect unemployment.


INTRODUCTION
Diverse job opportunities will be realized through an increase in the economy of a region. Collaborative partnerships that are sought between local governments and communities in developing the potential and existing regional resources will be able to minimize existing autonomy constraints (Menajang, 2019). In the end, it will overcome inequality between regions that have not experienced optimal development (Badrudin, 2012). Regulations that guide the implementation of regional autonomy (2001) and Law no. 22 (1999), where the government made various changes in anticipation of various obstacles. Furthermore, the financial balance between the center and the regions is explained in Law No. 25 (1999) which was replaced with No. 23 (2014) and No. 33 (2004). The main goal is to try to bring the government closer to the people. In line with government services to be implemented more efficiently and effectively which leads to increased welfare and community services.
The economic development of a region can be said to be successful by looking at the level of economic growth. This increase is reflected in the main targets of development that can be achieved which then have an impact on people's welfare and productivity. reducing the number of unemployment and poverty. Todaro et al. (2006) revealed the primacy of economic growth as one of the conditions for achieving economic development, but not limited to statistical figures that describe the rate of growth, but also the value that can be created through this economic growth. In fact, only a few of the components of society feel it. This indicates that economic growth is at least hampered by efforts to reduce poverty and inequality. Conversely, if the components of society participate in helping to increase economic growth, then poverty can be reduced to a minimum, and further narrow the gap between the lower and upper classes of society.
One of the efforts that can be made to increase regional economic growth is through autonomy, as is the case with fiscal decentralization in the form of the availability of regional revenue sources that can be optimized independently. This can be interpreted potentially in accordance with the capacity possessed. His conceptual view describes that measuring regional independence can be done by seeking Regional Original Revenue (PAD). One of these sources is from regional taxes, levies, the results of regionally owned enterprises, and the processing of other separated assets as well as legitimate regional income (Taryono & Ekwarso, 2012). Another view states that a strong focus on financing in governance and regional development at the provincial, district, and city levels should be taken from regional self-sufficiency. This source is mainly from Regional Original Revenue (PAD). Kuncoro (2006), based on his findings, revealed that the ability of PAD to finance regional government spending is a maximum of only 20%. The inequality that occurs in funding makes it possible for the government to overcome it at the central level through transfers of funds that are a balance for each region through various forms, namely Revenue Sharing Funds, General Allocation Funds, and Special Allocation Funds.
Another effort that can be made to increase economic growth is by regulating regional spending. Regulation of the Minister of Home Affairs Number 27 of 2013 states that regional spending must be used for the implementation of government activities which are the responsibility of the province and district (city) covering mandatory and optional affairs. Furthermore, it is regulated by statutory provisions (Paseki et al., 2014). One of the efforts to protect and improve people's welfare as well as fulfill regional obligations, priority is given to obligatory affairs. Its implementation is in the form of improvements in terms of essential services, education, health, social and public facilities, and creating a social security system. The implementation of mandatory affairs refers to the Minimum Service Standards (SPM) that have been previously determined, such as Personnel Expenditure, Interest, Subsidies, Grants and Social Assistance, Tax Revenue Sharing, State Assistance, and Unforeseen. The budgeting is done rationally by considering the realization in 2012, of the possibility of activities beyond the control and influence of the local government. Direct spending is poured in the form of activity programs, and the public can benefit from performance achievements to improve the quality of public services and support local governments (Ratnadi, 2017).
In addition, other efforts to increase economic growth are through increasing the minimum wage. Low income has had the effect of lowering savings and investment rates, which has had an impact on capital accumulation, limited employment opportunities, and increased unemployment. Of course, this will result in an increase in the number of poor people (Hariawan et al., 2015). The highest UMP increase occurred in West Sumatra, with an increase of 9.15% compared to the 2022 UMP amount of IDR 2,512,539 to IDR 2,742,476. Furthermore, there is also Jambi, the UMP in 2023 will be IDR 2,943,000, up 9.04% from the UMP in 2022 which is IDR 2,649,034. Meanwhile, the lowest UMP increases occurred in North Maluku and West Papua, which increased compared to the 2022 UMP respectively 4% in North Maluku and 2.56% in West Papua. Article 1 paragraph 30 of Law Number 13 of 2003 concerning Manpower emphasizes the minimum wage as a standard that will be used by employers to determine real wages for workers (laborers) in their company. The minimum wage is determined by the government (the Governor through recommendations from the provincial council and the regent or mayor). Every year it fluctuates according to the purpose of its determination (Humau, 2015). Setiyawati & Hamzah (2007) state that PAD has a positive effect on economic growth, while DAU is detrimental to economic growth. Then directly test the Effect of Economic Growth on Poverty and Unemployment, the results of which show a significant effect. However economic growth is detrimental to poverty and even has a positive impact on unemployment. Rori's research (2016) gives the result that an increase in local original income will result in an increase in regional economic growth and community welfare. This will be seen because an increase in economic growth in a region is expected to be able to absorb labor and reduce the unemployment rate, growth in the industrial and trade sectors, the service sector, and other related sectors. The originality aspect in this study lies in the replication of previous research conducted by Setiyawati & Hamzah (2007), namely by adding the minimum wage variable so that it becomes six variables namely PAD, development expenditure, regional minimum wage, economic growth, poverty, and unemployment. The expected objectives of this study lead to efforts to determine the role of regional own-source revenue, development spending, and regional minimum wages on economic growth, poverty, and unemployment by using the path analysis method.

METHODS
This type of research leads to a quantitative descriptive. Sugiyono (2017) explains that this method has scientific qualities that are useful in collecting data for specific purposes. According to Hair et al. (2020), researchers will describe the phenomena that occur related to current problems. From 2015 to 2021, this study was carried out in the province of Bali's cities and regencies. Total sampling was used in this study's sample collection, which resulted in 62 samples being collected over the course of seven years of research from 9 cities and regencies (Denpasar, Tabanan, Buleleng, Gianyar, Karangasem, Badung, Jembrana, Klungkung, and Bangli), including several data: regional own-source revenues, development spending, and regional minimum wages for economic growth, poverty, and unemployment. Multiple linear regression is useful in analyzing survey data after tabulation. This analytical method examines the influence between variables, dependent and independent (Sugiyono, 2019). The multiple linear regression equation model in this study is formulated as follows:

RESULTS AND DISCUSSIONS
Initial testing must be done before deciding "whether the hypothesis is accepted or rejected". The first tests performed were validity, normality, multicollinearity, heteroscedasticity, and autocorrelation. The general assumptions in this study must be met before finally moving on to the regression data analysis stage. Testing using multiple linear regression can be continued if the test results are positive. The following are the results of the classical assumption test performed: The normality test uses the regular P-Plot and One-Sample Kolmogorov-Smirnov graphs from the SPSS 25.0 for the Windows program. If the residuals produce a normal distribution and the significance exceeds 5%, the data can be said to meet the requirements (Ghozali, 2018). The test results can be seen in Table 1. The value obtained from the normality test using the Kolmogorov-Smirnov Z in Model 1 in Table 1 is 0.076. If it is interpreted that the values obtained have significance sequentially from models 1 to 3, namely 0.200; 0.171; 0.212 with a significant value obtained of 0.058 exceeding the specified conditional value. So it can be concluded that H0 can be accepted and H1 cannot be accepted. The conclusion of this analysis is that all residuals in Model 1, Model 2, and Model 3 can be said to be normally distributed. The multicollinearity test assumptions prove that a model's independent variables are not correlated. If VIF is greater than 10, there is a multicollinearity problem between this variable and the other independent variables. It can be seen in Table 2.  Table 2, show that Models 1, 2, and 3 of all the variables in this regression model do not have multicollinearity requirements. The heteroscedasticity test seeks to prove whether or not there is an unequal variance between the residual observations that appear different in the regression model. Homoscedasticity is said to be a condition where the residual variance from one observation to the next observation still exists; heteroscedasticity indicates a condition where the residual variance has differences. A proper regression model shows homoscedasticity or not. The best method that can be used to determine the possibility of heteroscedasticity is the Glejser test (Table 3): Based on Table 3, heteroscedasticity tests were carried out; it is known that the regression model of model 1, model 2, and model 3 can be seen that all data do not occur heteroscedasticity in the regression model.
Autocorrelation arises because of "sequential observations that are related to one another over time." "To test the presence of autocorrelation can be detected with the Durbini-Watson test." Based on the Model 1 autocorrelation test results, the DW value is greater than the upper limit sure of 1.6589, and the DW value is less than 2.4311". Model 2 obtained a DW value of 1.855 which is greater than the upper limit (du) sure of 1.6589, and a DW value of 1.855 is less than (3 -du) 3 -1.6598 = 2.4311. Model 3 obtained a DW value of 1.902 which is greater than the upper limit sure of 1.6589, and a DW value of 1.902 is less than (3 -du) 3 -1.6598 = 2.4311. So there is no autocorrelation. The results of the analysis prove that the answer to H1 has a strong influence value and the t-count value is below the t-table, so PAD is able to have a large and negative impact on economic growth in the Province of Bali.
H2 through the results of the study provides evidence that the Development Expenditure variable has a significant value where the t-count has a stronger effect than the t-table. Strong evidence of this hypothesis is that Economic Growth is greatly influenced by Development Spending.
Likewise, the results of the next study (H3) revealed that the Regional Minimum Wage variable had a significance value of 0.000. This means that the Regional Minimum Wage has a fairly large and strong influence on economic growth in the Province of Bali.
If it is interpreted that the results of H4 determine that the PAD variable has a significance value of 0.119, the understanding is that the t-count value is lower than the t-table. The results prove that PAD has a small impact on poverty.
H5 means that the development expenditure variable has a significance value of 0.021 which implies that investment in development in the Province of Bali has a sizable and beneficial impact on poverty.
H6 finds a strong impact from the Regional Minimum Wage variable with a significance value of 0.022. This again shows that regional minimum wages have a significant and beneficial impact on poverty.
H7 states that the PAD variable has a significance value of 0.857 with a significance value of 0.05. The conclusion obtained is that PAD does not have a strong influence on unemployment.
H8 reveals evidence that the Development Expenditure variable has a significance value of 0.092, and the t-count value is stronger than the t-table, namely 1.713 > 1.67022. So it can be said that investment in development has a large and beneficial impact on unemployment.
H9 finds that the influence of the Regional Minimum Wage variable has a t-count value below the ttable, namely 3.717 > 1.67022. This proves that the Regional Minimum Wage has a large and beneficial effect on unemployment.
The independent variable has the ability to explain the diversity and the dependent variable which is mentioned as the coefficient of determination in the analysis model using linear regression so it becomes the basis for an understanding that can be explained quantitatively, especially other forms of variables that are able to influence the dependent variable. Through the formulation of its simplest form, the correlation coefficient (R) is a measure to determine the result, so that it can be seen to what extent the independent variable X can explain the variation of the dependent variable Y. Changes in Y of course state that they cannot be described by X if the coefficient of determination is 0 (R2 = 0). R2 = 1, and vice versa, variable X can explain the variation of Y as a whole. The test results can be seen in Table 6.  Table 6 shows that the "R-square value in model 1 is 0.651 or 65.1%", the variable Economic Growth is influenced by the PAD, Development Expenditure, and Regional Minimum Wage variables, and the remaining 34.9% is influenced by other variables that are not present in the study This. The R-square value in model 2 is 0.423 or 42.3%, the Poverty variable is influenced by the PAD, Development Expenditure, and Regional Minimum Wage variables. The R-square value in model 3 is 0.624 or 62.4%, the unemployment variable is influenced by the PAD, Development Expenditure, and Regional Minimum Wage variables.

The Effect of PAD on Economic Growth.
This research aligns with the findings of Saraswati & Ramantha (2018) and Entrepreneur et al. (2018), which also show the PAD variable's adverse effect on economic growth. Economic growth provides quite a large opportunity in terms of increasing PAD so that later it can lead to positive economic growth. This role provides an opportunity for local governments to focus activities to optimally empower communities that have economic strength in the local scope to increase their growth in addition to issuing various regulations in terms of taxes and fees. The tendency for PAD to increase will encourage further better economic growth. The future goal is to maximize and prioritize superior programs that can support functions related to economic activities, both in the fields of industry and trade, services, and others. The situation behind the participation of all regional components is due to the increase in PAD. This situation makes it possible for the government to carry out the construction of facilities and infrastructure and provide justice for the welfare of the community, autonomy, and increase GRDP every year. An increase in PAD will trigger and spur regional economic growth to be better than the previous regional economic growth. An increase in PAD can also optimize and increase activity in sectors related to economic growth, such as the industrial and trade sectors, the service sector, and other sectors. One of the main goals of fiscal decentralization is the creation of regional independence. Local governments are expected to be able to explore local financial sources, especially through PAD. If PAD increases, the funds owned by the local government will be higher.
The Influence of Development Spending on Economic Growth.
The findings in this study agree with the disclosures from research by Istianto et al. (2021) and Yasin (2020), that there is a strong influence between development spending and economic growth. It is hoped that resource management through a synergistic partnership between stakeholders (government, community, and private sector) will open up more opportunities for employment and opportunities for improving the regional economy. This success is the main step in the management of existing resources. In fact, the progress of a region can be measured by the amount of economic growth, economic structure, and reduced inequalities in the community, region, and each component.
The Effect of Regional Minimum Wage on Economic Growth.
Regional minimum wages have a strong effect on economic growth. Consistent with what was expressed by Nasution (2022), explaining that an increase in X3 can result in an increase in the desire and ability to do business in the community which leads to an increase in the productivity of an area. The wider impact is increased economic growth in the Province of Bali. an increase in the regional minimum wage will also encourage the desire to open business fields and improve the competitive environment that is innovative and creative. The program launched by the government to achieve a minimum wage for decent living needs is expected to be achieved by the community in stages.
The Effect of PAD on Poverty.
Research that is relevant and has gaps is presented by Paat et al. (2019) and Paul et al. (2019), who revealed that the regional minimum wage variable has a positive and strong influence on poverty. The findings obtained from this analysis illustrate the importance of PAD and its contribution to overcoming poverty. The high PAD is the driving force for the successful implementation of regional autonomy, in addition to the quality of the regional government apparatus itself and the quality of public education. Poverty is the main source of problems in achieving economic development. There are three main issues namely: what happens to the poverty rate, what happens to unemployment, and what happens to inequality in various fields. The basic problem is that they influence each other or are reciprocal, where the amount of poverty is caused by high unemployment. This will have an impact on other sectors. If one of them experiences fluctuations, it will have implications for other factors as well.
The Influence of Development Spending on Poverty.
The study found that development spending has a significant and positive effect on poverty, as indicated by the X2 variable (X2). This finding aligns with previous research by Evita & Primandhana (2022) and Leonita & Sari (2019), indicating that spending on development is crucial for addressing poverty. Effect of Regional Minimum Wage on Poverty.
The Regional Minimum Wage variable significantly impacts poverty, as it aligns with studies by Ayu (2018) and Sari (2020), indicating that the local minimum wage positively impacts poverty. This finding supports previous research on the regional minimum wage variable. The Effect of PAD on Unemployment' The PAD variable has a significance value, indicating no significant effect on unemployment. This value aligns with previous research by Ali & Ningsih (2021) and Rizal & Wati (2021), which also found no significant effect on unemployment. This states that local revenue allocated for community welfare is not directly used to reduce the unemployment rate, but the allocation is more directed to community welfare through financing free education, medical treatment, and other benefits so that later educated jobs will be created.
The Influence of Development Spending on Unemployment.
The findings indicate that investing in development has a significant and favorable impact on unemployment. The Development Expenditures variable indicates that investing in development has a strong and favorable effect on unemployment. This finding is consistent with Rotinsulu et al.'s (2019) findings, which found that variable development spending had a large and beneficial effect on unemployment. Effect of Regional Minimum Wage on Unemployment.
It has a strong influence on unemployment, as the findings produced in this study. The strong significance value indicates a beneficial effect on unemployment. The findings are consistent or in line with other studies which have been uncovered by Priastiwi and Handayani (2019), Sembiring and Sasongko (2019).

CONCLUSION AND RECOMMENDATION
The underlying fact is that, increasing PAD results in increasing economic growth and autonomy in the region. However, it cannot be ignored that increasing the achievement of PAD can also result in inefficiencies in its management, so that the value or amount will be absorbed for operational financing, in the sense that it does not affect capital expenditure for investment purposes. the conclusion is that the regions do not have independence and rely on transferring funds from the center. It is hoped that the regency and city governments of the province of Bali will be able to overcome unemployment by issuing policies that minimize the occurrence of unemployment. Such as creating jobs, improving the quality of the workforce, and developing the transmigration program. In addition to reducing the Open Unemployment Rate so that economic growth increases and increases so that it can become an independent region and can manage its own regional household economy.