The Nexus of Gender Inequality and Poverty Rate in Indonesia

- The phenomenon of feminization of poverty intertwines issues of poverty and gender as interconnected challenges. This research aims to elucidate the intricate relationship between gender inequality and the poverty rate in Indonesia. By utilizing panel data from 34 provinces during 2018 to 2022, Multiple linear regression modeling is conducted to analyze the impact of independent variables comprising Gender Inequality Index, Gender Development Index, Gender Empowerment Index, dependency ratio, and Gini ratio on the poverty rate. The findings reveal that Gender Inequality Index, Gender Empowerment Index, dependency ratio, as well as Gini ratio significantly positively affect the poverty rate in Indonesia. Meanwhile, Gender Development Index significantly negatively affects the poverty rate in Indonesia. Consequently, comprehensive and gender-aware policies are essential to alleviate gender inequality and poverty issues. This also suggest that the government should reduce gender bias paradigms in order to ensure women's empowerment efforts can effectively reduce the poverty rate.


INTRODUCTION
Poverty remains a significant challenge in developing countries like Indonesia, where over 25 million people (9.36%) live below the poverty line (Badan Pusat Statistik, 2023b).The complexity of this issue intertwines with various facets of life, rendering it a multidimensional problem (Todaro & Smith, 2020).This intricate problem is further exacerbated by gender inequality, notably impacting women disproportionately.Women face a higher likelihood of experiencing poverty compared to men, with women bearing a disproportionate burden within households (Bradshaw et al., 2017;Buvinic et al., 2009;McLanahan & Carlson, 2001).The global recognition of this phenomenon is underscored by the World Bank Report, which discloses that at least 2.4 billion economically active women lack equal economic opportunities (World Bank, 2022), leading to the concept of the 'feminization of poverty' (Chant, 2014;EU Parliamentary Assembly, 2007;Gökovalı & Danışman, 2010).Rooted in social and cultural factors (Rokhmansyah, 2016), gender inequality restricts women's access to education, healthcare, and politics, leaving them disempowered and vulnerable (Gerecke, 2013;Rietveld & Patel, 2022).Such disparity impedes poverty alleviation hindering both economic and human development (Anggraini & Faridatussalam, 2024;Nathan et al., 2022;Seguino, 2008).Empowering women through enhanced education and job opportunities, resulting in better income, is a pivotal factor in eradicating poverty (Pervaiz et al., 2011;Kanat et al., 2023).Moreover, Arifin (2018) argued that poverty reduction can be furthered by optimizing women's roles in economic development, with improved job opportunities contributing to reducing impoverished households.However, support in education and economics alone proves insufficient if persistent barriers impede women's access to healthcare, particularly basic health services and maternity care.The involvement of women in policymaking institutions is essential, as it can lead to more inclusive and gender-aware policies, thereby mitigating the impact of poverty on women.The United Nations Development Programme (UNDP) introduces comprehensive metrics to quantify gender inequality, including the Gender Development Index (GDI), Gender Empowerment Index (GEI), and Gender Inequality Index (GII) (UNDP, 2023).Indonesia's GII ranking of 110th globally and 7th in Southeast Asia, with a value of 0.444, highlights notable gender disparities in health, empowerment, and the labor market.Provincial variations in GII values, such as in Yogyakarta and Jakarta with lower values, and West Nusa Tenggara and West Papua with higher values, underscore nuanced patterns (Badan Pusat Statistik, 2023a).Prior research extensively explores the complex interplay between gender inequality and poverty.Nisak & Sugiharti (2020) investigated the impact of reproductive health, education, and women's income on poverty in Indonesia, while Direja & Paramitasari (2022) focused on women's education length, both revealing significant influences on economic deprivation.Aprilia & Triani (2022) affirmed GII's notable impact on poverty.Employing a qualitative approach, Septiadi & Wigna (2013) identified a link between gender inequality and the poverty rate in Cikarawang Village.Notably, investigations by Pardhan (2018) and Nurhidayati et al. (2023) revealed that GDI has a significant negative impact on poverty.However, when scrutinizing GEI, Adnan & Amri (2021) observed a substantial negative impact on poverty, while Saputri et al. (2023) reported no significant impact on poverty, highlighting the nuanced nature of these dynamics.In the light of the background above, this research endeavors to to elucidate the intricate relationship between gender inequality and poverty rate in Indonesia.It will examine the effects of the GII, GDI, and GEI as well as dependency ratios and the Gini coefficient as control variables on poverty levels.This research hopes to contribute a nuanced understanding of how gender dynamics intersect with poverty at both national and regional levels and to inform evidence-based policy formulations, fostering gender equality and reducing poverty in Indonesia.

RESEARCH METHODS
This study focuses on examining the influence of GII, GDI, IDG, Dependency Ratio, and Gini Ratio using the data from Indonesia Statistics Agency (BPS) publications.The panel data utilized in this study consist of data from 34 provinces in Indonesia spanning the years 2018 to 2022.The author utilizes a quantitative research methodology which involves a structured series of experiments/investigations into a specific phenomenon using relevant data, followed by measurement through mathematical or computational statistical techniques (Abdullah, 2015).The independent and dependent variables used in this study are outlined in Table 1.To evaluate the impact of independent variables on the dependent variable, a multiple linear regression analysis is employed.The data processing in this study is conducted using Stata 17 software.The model for multiple linear regression analysis of panel data for hypothesis testing is presented below.
In the panel data regression analysis, there are three common approaches that are commonly used: Common Effects Model (CEM), Fixed Effects Model (FEM), and Random Effects Model (REM) (Sihombing, 2021).To determine which approach is the most suitable, a model selection test is conducted first.REM is better than FEM FEM is better than REM Source: Gujarati (2004) Following the selection of the best model, a battery of classical assumption tests is carried out to make sure the model fulfill the criteria for assessing the effect among variables and predicting the dependent variable's values based on the independent variables (Gujarati, 2004).The tests for classical assumptions can be observed in Table 3.

RESULTS AND DISCUSSION
The discussion begins with a descriptive analysis of the data to obtain an overview of the characteristics of the variables in this study.As seen in Consequently, the initially selected FEM model was transformed using the Panel Corrected Standard Error (PCSE) model (Baltagi, 2005;Greene, 2018).Based on the hypothesis test presented in Table 9, the coefficient of determination value is 0.5402.This value signifies that 54.02% of the variation in the poverty rate can be accounted for by the independent variables within this model, while the remaining 45.98% is attributable to other variables outside the model.The F-test results reveal that all independent variables simultaneously exert a significant impact on the poverty rate.The statistical probability value F=0.00, smaller than alpha 0.05, indicates that the conducted modeling is appropriate.In the partial test, each variable demonstrates a noteworthy influence, with the probability value of the t-test being < alpha 0.05.The regression equation can be written as follows.
POVit = -70.09331 + 8.467487 GIIit -0.1461149 GDIit + 0.097932 GEIit + 0.8848961 DRit + 117.2335GINIit DISCUSSIONS 1.The Impact of GII on Poverty Rate GII influences the poverty rate positively and significantly.Based on the test results, the coefficient is 8.467487 with a t stat= 2.19 > t table=1.96,and the probability value is 0.029 < alpha = 0.05.These results indicate that every 0.01-point decrease in GII will reduce the poverty rate by 0.08467487 percentage points, assuming other factors remain constant.This finding aligns with research conducted by Septiadi & Wigna (2013), Alisjahbana & Pitriyan (2016), Direja & Paramitasari (2022), revealing that gender inequality significantly influences the poverty rate.This implies that enhancing women's health, education, employment opportunities, and political influence can reduce a nation's poverty rate.The decrease happens as women can maximize their potential, ultimately improving their living standards and that of their families (Rietveld & Patel, 2022;Todaro & Smith, 2020).

The Impact of GDI on Poverty Rate
GDI exhibits a significant negative impact on the poverty rate, with a coefficient of -0.1461149, t stat= 2.55 > t table=1.96,and a probability value of 0.011 < alpha = 0.05.This indicates that every 1-point increase in GDI the poverty rate is expected to decrease by 0.1461149 percentage points, assuming that other variables remain constant.This outcome is supported by previous studies by Pardhan (2018) and Nurhidayati et al. (2023), indicating that increasing GDI can enhance growth and reduce poverty.Therefore, besides equality in access, poverty eradication also requires genderequitable human development, especially in health, education, and income aspects.World Bank (2020) reported that investment in development programs that close the gender gap contributes to further reduction of poverty.

The Impact of GEI on Poverty Rate
The positive impact of GEI on the poverty rate is statistically significant, as indicated by a coefficient of 0.097932, |t stat|= 3.38> t table = 1.96, and the probability value is 0.001 < alpha = 0.05.This implies that raising GEI by 1-point results in an increase in the poverty rate by 0.097932 percentage points, assuming that other factors are constant.However, this result reveals that there is insufficient evidence to assert a significant negative impact of GEI on the poverty rate.This finding aligns with the research by Saputri et al.'s (2023) on the impact of GEI on poverty in five regencies in Central Java and the research by Adnan & Amri (2021) on the influence of GEI on poverty in eight provinces in the western part of Indonesia.
GEI measures women's active involvement in political and economic decision-making (Angraini et al., 2022).With the gender inequalities in education and health affecting Indonesian women, this participation has not effectively reduced poverty.Moreover, women's roles in strategic positions are still limited by gender bias in social norms that consider men as better leaders and decision-makers than women (UNDP, 2023).

The Impact of Dependency Ratio on Poverty Rate
The dependency ratio (DR) significantly contributes to the poverty rate in Indonesia, as evident from the coefficient of 0.8848961, |t stat|= 5.84> t table = 1.96, and the probability value is 0.0000 < alpha = 0.05.This result implies that a 1-point increase in the dependency rate leads to a 0.8848961 percentage points rise in poverty, assuming other variables remain unchanged.A higher dependency burden means a smaller share of income for the productive-age population due to supporting more individuals.This finding is consistent with Todaro & Smith (2020) assertion that a high dependency burden reduces savings and investment, slowing economic growth and leading to higher poverty.Numerous previous studies, including those by Wintara et al. (2021), Junaidi et al. (2017), andFirdaus et al. (2021) support the positive impact of the dependency ratio on poverty.Globally, Vijayakumar (2013) found similar conditions in the middle and low income countries in Asia, Latin America, and Africa, which corroborates the finding in this study.

The Impact of Gini Ratio on Poverty Rate
The regression outcomes indicate that the Gini ratio has a significant positive influence on the poverty rate as evidenced by a coefficient of 117.2335, t stat= 14.77 > t table = 1.96, and the probability value is 0.0000 < alpha = 0.05.Consequently, a 0.01-point increment in the Gini ratio is linked to a 1.172335 percentage point increase in the poverty rate, assuming other variables remain constant.This finding aligns with prior studies by Abdillah & Mursinto (2016), Atmojo (2017), Nurrizqi et al. (2022), Maulana et al. (2022), and Naufal & Fikriah (2023) highlighting the consistent positive impact of the Gini ratio on poverty in Indonesia.Similar trends are observed in high-income countries (Wagle, 2010) and developing countries (Ali et al., 2022), reinforcing the notion that higher income disparities among the population correspond to higher poverty percentages in a country (Lakner et al., 2022).

CONCLUSION
By using the Panel Corrected Standard Error (PCSE) transformed Fixed Effect Model this study revealed that all independent variables exhibit a statistically significant impact on the poverty rate.GII, GEI, dependency ratio, and Gini ratio are positively associated with poverty, whereas GDI demonstrates a significant negative effect on the poverty rate.Therefore, addressing poverty in Indonesia requires the government to not only confront issues related to inequality among the productive-age population and income disparities, but also to improve gender equality conditions.To achieve this, the government should prioritize comprehensive and gender-responsive policies aimed at enhancing women's access to health, education, job opportunities, and overall well-being.Moreover, reducing gender bias paradigms is crucial for ensuring women's empowerment efforts can effectively reduce the poverty rate.Although this study successfully addressed several aspects as summarized above, it has various limitations.The independent variables utilized in this study fall short of providing a comprehensive explanation for poverty, hence there remains a need for future research to expand on the determinants of poverty.Additionally, the reliance on secondary data underscores the importance of incorporating primary data in subsequent studies, particularly to delve into the intricacies of the feminization of poverty through surveys or interviews, thereby enriching the overall understanding of poverty dynamics.

Table 1 .
Research Variables

Table 2 .
Panel Model Selection Test

Table 3 .
Classical Assumption Test

Table 5 ,
the average of Indonesia poverty rate is 10.39%.The lowest poverty rate, 0.01%, occurred in the Bangka Belitung Islands in 2021.Meanwhile, the highest poverty rate, 31.7%,occurred in the East Nusa Tenggara in 2019.Based on gender-related indices, the average GII score in Indonesia is 0.48 points, with the highest value of 0.668 points in Bali in 2019 and the lowest value of 0.149 in Central Java in 2020.The average GDI score is 90.39 points, with the highest value of 94.99 points in D.I. Yogyakarta in 2022 and the lowest value of 79.59 points in Papua in 2020.The GEI score has an average of 69.59 points, with the highest value of 83.2 points in Central Kalimantan in 2019 and the lowest value of 51.04 points in West Papua in 2018.The dependency Ratio in Indonesia has an average of 47.51 points, with the highest dependency rate of 64.89 points occurring in East Nusa Tenggara in 2020.The lowest dependency rate in Indonesia occurred in DKI Jakarta in 2021 with a value of 38.96 points.The average income inequality ratio in Indonesia, measured using the Gini ratio, is 0.35 points, with the lowest ratio of 0.247 in the Bangka Belitung Islands in 2021 and the highest ratio of 0.459 in D.I. Yogyakarta in 2022.

Table 7 .
Based on the selection test, the Fixed Effect Model (FE) was chosen as the most suitable model to explain the influence among variables in the study.