ASEAS 15(1) | 23 The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia Agus Heruanto Hadnaa* & Media Wahyudi Askara a Universitas Gadjah Mada, Indonesia *corresponding author: hadna@ugm.ac.id Received: 23 June 2021 / Accepted: 24 May 2022 / Published: 30 June 2022 ► Hadna, A. H., & Askar, M. (2022). The impact of conditional cash transfers on low-income individuals in Indonesia. Advances in Southeast Asian Studies, 15(1), 23-41. Most governments claim that Conditional Cash Transfer (CCT) programs benefit poor people. This study aims to analyze the impact of conditional cash transfers on low- income individuals in Indonesia. This study used consumption expenditures as a poverty measure and found that the Program Keluarga Harapan (PKH) has significant impact on an individual's consumption. However, households in the lowest wealth quantile were found to not take advantage of those benefits due to the current CCT design. Moreover, the heterogeneity of the CCT can generate substantial inequality, as household incomes in the lowest quantile fall. Therefore, governments should be more generous to house- holds in the lowest wealth quantile, and carefully manage the program based on the needs of CCT beneficiaries. Keywords: Conditional Cash Transfer; Indonesia; Inequality; Lowest Wealth Quantile; Program Keluarga Harapan  INTRODUCTION In 2015, The State of Social Safety Nets, a World Bank study, concluded that the coverage of social safety net policies in developing countries represents only one-third of the total poorest quantile. As a result, many anti-poverty programs cannot lift people out of poverty and fail to result in more inclusive societies (Dollar & Kraay, 2002; Filmer, 1999; Honjo et al., 1997; Squire, 1993). Ravallion (2016) noted that anti-poverty policies in many countries, especially in devel- oping countries, have been ineffective because program interventions are not directly targeted. One of the preferred programs for reducing poverty is the Conditional Cash Transfer (CCT). Some studies found that CCT improves the welfare of the poor. In Indonesia, research conducted by Tim Nasional Percepatan Penanggulangan Kemiskinan (TNP2K) in 2018 showed that the CCT Program Keluarga Harapan (PKH) positively impacted per capita expenditures (PCE) in Indonesia by 4.8% Current Research on Southeast Asia w w w .s ea s. at 10 .1 47 64 /1 0. A SE A S- 00 67 mailto:hadna%40ugm.ac.id?subject= http://orcid.org/0000-0002-8716-1901 https://orcid.org/0000-0003-3635-0961 24 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia per month. The same Indonesian program raised PCE by 3% per month compared to the CCT program in the Philippines (Pantawid Pamilyang Pilipino Program; Chaudhury et al., 2013; Tutor, 2014). In Nicaragua, the effect of the Red de Proteccion was quite significant at 18% (Maluccio & Flores, 2005). Meanwhile, in Colombia, PCE grew by 15% through Familias de Accion (Attanasio & Mesnard, 2006). Research on the impact of CCT also yields positive conclusions on reducing inequality. A study conducted by Soares et al. (2009) concluded that CCTs helped governments in Brazil, Mexico, and Chile reduce inequality in welfare from the mid- 1990s to the mid-2000s. They found CCTs reduced inequality by 21% in Brazil and Mexico, and by 15% in Chile (Soares et al., 2009). Lagarda et al. (2017) implied that institutional weaknesses caused inequality in Latin American countries. Comparing stunting between rich and poor groups of children in Indonesia, Rizal and van Doorslaer (2019) found that PKH decreased the total rate of inequality among people with stunting. Almost all research examining the effects of CCT has revolved around issues in education, such as school enrolment (Behrman et al., 2005; de Janvry et al., 2006; Schultz, 2004), class attendance (de Janvry et al., 2006; Hadna & Kartika, 2017; World Bank, 2015), inequality of educational opportunities (Ham, 2014), student cognitive capacities (Ponce & Bedi, 2010), and academic outcomes (Juan et al., 2012; Schultz, 2004). Studies on the effects of CCT in improving health outcomes also exist. CCT programs have positively affected infant health, for example, in Mexico (Fernald et al., 2008; Millán et al., 2019), Latin American countries (Owusu-Addo & Cross, 2014), Zimbabwe (Robertson et al., 2013), and other Sub-Saharan countries (Garoma et al., 2017). CCT has also reduced chronic malnutrition (Farrington & Slater, 2006) and stunting prevalence in the Philippines (Kandpal et al., 2016) and Indonesia (Cahyadi et al., 2018). In terms of gender issues, CCT affected neither intra-household gender relations nor the relative position of women within house- holds in Indonesia (Arif et al., 2013) and Latin America (Franzoni & Voorend, 2012; Molyneux & Thomson, 2011). CCT also relates to social inclusion (Rawlings, 2006) and equality between rural and urban populations (Krishnakumar & Juárez, 2012). Despite those successes, there have been many criticisms of CCT programs, such as the argument that they are paternalistic, and of the ways low-income fami- lies spend the transfers (Ravallion, 2016). In addition, many fundamental questions remain unanswered about CCT effectiveness, including long-term welfare impacts, synergies between program components, trade-offs between transfer sizes and ben- eficiaries, and the balance between short-term transfer objectives and long-term human development objectives (Rawlings & Rubio, 2005). To date, research has not examined the impact of CCTs on the distribution of consumption expenditures among the poor (Kamakura & Mazzon, 2014), especially among each quantile of low-income groups targeted by CCT. This issue needs to be investigated more deeply because studies show that cash transfers, including CCT, do not automatically reduce poverty (Golan et al., 2017). Critics of CCT claim that most poor households spend cash transfers on overconsumption, encouraging new, unsustainable lifestyles (Kamakura & Mazzon, 2014). Consumption expenditure as a welfare indicator was mainly pioneered by the welfarist (utility) movement in eco- nomics, which measures the welfare of the people through their buying and selling ASEAS 15(1) | 25 Agus Heruanto Hadna & Media Wahyudi Askar behaviors on the market (Ravallion, 2016). However, a working paper published by the National Bureau of Economic Research in 2018 on PKH in Indonesia could not deci- sively measure the extent to which the program affects consumption expenditures. In conclusion, the objective of the CCT program remains to direct the spending of low-income families to be more focused on education and health, as a way to break the chain of family poverty (Cahyadi et al., 2018). Based on this research gap, this paper addresses the following question: What is the impact of PKH on the distribution of consumption expenditures among each quantile of lower-income groups? First, this question focuses on identifying and comparing the impact(s) of PKH on the distribution of consumption expenditures among the highest and lowest quantiles of poverty groups. This paper also dis- cusses the weaknesses of PKH in Indonesia, primarily in reducing inequality among low-income individuals. The PKH impact indicator of this study is consumption expenditure among the poor because it is easier to describe the welfare levels of those elements, and the results are more valid in terms of size. In addition, some experts have previously warned that poorly managed anti-poverty programs can disproportionately impact clients and lead to greater inequality (Aghion et al., 1999; Beaudoin, 2006; Bergh & Nilsson, 2014; Ravallion, 1995, 1997, 2016). Hence, research on poverty should look more closely at what happens among the poor when they partake in the program. In this case, however, conducting impact assessment studies by grouping the poor into quantiles with similar problems and characteristics can be problematic because the poor also have relatively sharp social and economic stratifications. Furthermore, this research assumes that households in the lowest wealth quan- tile may have the most difficulty obtaining consumption expenditure than the next higher quantile. This inability occurs because the CCT program does not fit all, nor do program materials and solutions that do not address the unique problems of the poorest groups. Consequently, CCT inadvertently maintains inequality of consump- tion expenditure among the poor. This study offers two significant contributions. First, substantially limited studies have sufficiently analyzed the impact of CCT on each quantile of recipients. Failure to capture this aspect may lead to bias as the models estimate only the conditional means of the responses. Second, most studies have evaluated the impact of CCT using Ordinary Least Square, Average Treatment Effect, or attempted to address endogeneity using the Instrumental Variable approach. While this paper follows the latter approach, we adopted an Instrumental Variable Quantile Regression (IVQR) approach because it is more suitable when capturing and comparing the consump- tion expenditure impacts of CCT on different quantiles. This paper closely examines the efficacy of PKH in Indonesia through a case study from 2019 in Yogyakarta. The poverty data is alarming because this province is the most impoverished region on Java Island. In 2019, the data from the Indonesian Statistics Bureau showed that the poverty rate was 10.62%. Moreover, the Gross Regional Domestic Product (GRDP) in Yogyakarta has always ranged in the lowest category on Java Island. Likewise, the level of inequality measured by the Yogyakarta Gini Ratio in 2019 shows the highest inequality at 0.42 points. 26 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia THEORETICAL FRAMEWORK Experts often compare the effectiveness of CCT with Unconditional Cash Transfer (UCT) (Baird et al., 2014; Schubert & Slater, 2006). They posit that it is easier to implement UCT than CCT (Haushofer & Shapiro, 2016): UCT does not reduce recip- ients’ work motivation (Banerjee et al., 2017). Also, UCT has a positive impact on the consumption behavior of beneficiaries, focusing on food consumption, health, and education rather than spending on luxury goods (Haushofer & Shapiro, 2016). UCT was found to not increase cigarette and alcohol consumption (Evans & Popova, 2014). However, long-term evidence showed that the impact of CCT on sustainable human development is more significant than that of UCT (Baird et al., 2014). Past research showed that a dollar unit spent on CCT is eight times more effective than a dollar unit spent on UCT (de Janvry et al., 2006). Compared to in-kind assistance, both models of cash transfer are more effective. Studies conclude that cash transfers provide more opportunities for parents to allocate money for their children's health and welfare (Kamakura & Mazzon, 2014 ; Miller & Neanidis, 2015). This study aims to examine CCT because it is considered the most prominent poverty alleviation program. The basic idea of CCT is that incentives given to households are to keep their children in school and diligently check their health (Ravallion, 2016). However, this strict requirement is seen by critics as a form of paternalism because it ignores poor people’s ability to identify the real needs of their families. On the other hand, CCT proponents believe that households remain poor because parents do not give chil- dren the right to obtain an adequate education. Instead, they prioritize helping their families meet life’s necessities (Ravallion, 2016). Studies on the effectiveness of CCT show a weakness in targeting or directed marketing. Initially, targeting suppressed program leakage due to incorrectly focused target groups (Ravallion, 2016). However, targeting has weaknesses, such as biases in determining target groups which is more profitable by local elites, and in deter- mining poverty lines to choose the target groups. Mirrlees (1971) highlighted these weaknesses regarding trade-offs between equity and efficiency of administrative capacity (Ravallion, 2016). Equity refers to the right of all poor people to fulfill their needs. In contrast, efficiency refers to the government strategy that prioritizes target groups due to limited resources. The paternalistic nature of CCT can be studied by comparing its impact across the distribution of consumption expenditures. The comparisons indicate how poor individuals can use the incentives to increase their consumption expenditures. The assumption is that spending on the education and health sectors is the most effective. However, there may also be more effective consumption expenditures outside those sectors. This raises the question of to what extent the poor use CCT incentives. The targeting issue is also valuable to study when comparing CCT use by different quantiles. The analysis technique measures the correlation of CCT use with con- sumption expenditure per quantile. The assumption is that the poorest low-income individuals should benefit the same as or more than those in the higher quantiles. Ravallion (2016) explicitly encourages the avoidance of paternalism in poverty allevia- tion programs. Instead, programs must be straightforward about various low-income people’s characteristics and understand the data limitations (one size does not fit all). ASEAS 15(1) | 27 Agus Heruanto Hadna & Media Wahyudi Askar Several approaches to measuring the level of people’s prosperity include welfare, utilities, abilities, and opportunities. The welfare approach essentially measures peo- ple’s interests through their ability to buy or sell goods in the market. This approach assesses their perceptions of several well-being indicators, such as fulfilling good nutrition. The question of the people's perceptions of fulfilling good nutrition are what can be achieved, what benefits them, and whatever people maximize in their own choices (Ravallion, 2016; Slesnick, 1996; Stutzer & Frey, 2004). The util- ity approach considers people as rational beings who seek to maximize profits for themselves (Ravallion, 2016). According to this approach, people will only choose to address those needs that fit their budgets and uses. Sen (1981; 1985) proposed the capabilities approach, considering the varying capacities of people to access available resources. Rather than commodities, he emphasized the minimum standard of living, such as life expectancy, nutrition, education, and health. The opportunity approach adds that unequal outcomes are not problems, as long as equal efforts are considered. Efforts are only possible if the circumstances support them (Ravallion, 2016). This study uses a welfare approach, which measures welfare with consumption expenditure by looking at income and consumption-based practices. The con- sumption-based method has been widely used, especially in developing countries (Cutler & Katz, 1992; Ravallion, 1992; Sen, 1981; Slesnick, 1996; Stutzer & Frey, 2004). Theoretically, consumption expenditure is assumed to be a suitable poverty meas- ure because it captures individual living standards via the consumption of goods and services. In addition, some researchers reported a more significant correlation between consumption expenditures and subjective well-being than income and sub- jective well-being (Meyer & Sullivan, 2011). Consumption expenditures are also more accurate than income when examining the living standards of the poor in the lowest wealth quantile (Meyer & Sullivan, 2011). This consideration is due to the nature of consumption expenditure questions, which appear to be less sensitive for the poor than income questions. Moreover, developing countries rely on the informal sector, often attempting to under-report income as tax avoidance (Schneider & Enste, 2000). Finally, money is not the salary for people who depend primarily on agriculture; instead, agricultural goods are the salary, though turning these goods into cash can be problematic (Coudouel et al., 2013). Regarding aggregation, measuring welfare should be at the individual level. However, numerous studies attempt to explain how every individual within a house- hold is likely to have different conditions (Duflo, 2003; Quisumbing et al., 1995). Therefore, the actual consumption of the individual is a better indicator of living standards, than the consumption of households, mainly because poor individuals can be part of wealthy households and vice versa (Falkingham & Namazie, 2002). In addition, household-based poverty measures may fail to capture individual living standards when financial shocks hit. For instance, sickness and job displacement may lead to substantial reductions in the well-being of individuals, even when they do not bring reductions in household welfare. When it comes to methodological issues, the endogeneity of the CCT must obtain robust estimates. As discussed in Zuluaga et al. (2020), the conventional econo- metrics model to estimate the impact of CCT cannot set aside issues of selectivity, endogeneity, and optimization. 28 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia Where 𝐶𝑇� is assumed to be exogenous, refers to a vector of instrumental var- iable that is related to 𝐶𝑇� , and uncorrelated to poverty, 𝑙𝑛𝑌� . Furthermore, 𝑋� 𝛽 refers to control variables, and 𝜀 is the error term. This study identifies the impact of poverty on different quantiles of our varia- ble interest to estimate the effect of PKH on poverty level. As in the Least Square Regression, conventional techniques are no longer helpful. They only allow us to cal- culate the link between our explanatory variables and the conditional means of the dependent variable poverty. On the other hand, quantile regression is an alternative approach that allows us to identify the link between explanatory variables and specific quantiles of dependent variables. In this study, we can estimate the impact(s) of PKH on particular groups of individuals, such as the lowest or highest consumption expenditure groups. As explained by Koenker and Hallock (2001), conventional Ordinary Least Square techniques allow us to estimate ∝, which refers to the impact of variables X on the conditional mean of Y. A Quantile Regression offers an estimation of beta by calculat- ing ∝(τ) for any quantile (τ)ε, (0,1), which reflects the impact of variables X on certain quantiles the distribution of Y. However, it is essential to note that the IVQR should be chosen when endogeneity is the issue. As Hansen (2005) described, IVQR can obtain robust estimates by address- ing endogeneity issues. It provides an estimate for any quantiles of the distribution of a dependent variable. Thus, this research goes one step further by calculating an alternative approach, namely the IVQTE. This technique can obtain robust estimates under endogeneity and, at the same time, provide an estimation of quantile treat- ment effect without functional assumptions. DATA AND METHODOLOGY Data The sources for this study were derived from research on poverty in Yogyakarta conduct- ed by the Center for Population and Policy Studies (CPPS) at Universitas Gadjah Mada in 2019. The dataset was collected by individuals who assessed respondent characteristics, financial activities, education, employment, and regional infrastructure. This study used this dataset to cover the most comprehensive PKH program, providing specific trends at the Indonesian regional level. There were 3,933 responses used in the research that focused only on low-income individuals. Nine sub-districts out of four districts select- ed as samples were chosen from the local government's policy for Yogyakarta, which determined the sub-districts as the poorest in Yogyakarta. The nine sub-districts include rural areas, namely Girisubo, Playen, Semin, Nglipar, Samigaluh, Girimulyo, Lendah, Pajangan, and Tempel. These areas have these characteristics in common: populations that rely on the agricultural sector, low education levels, geographic constraints, inade- quate infrastructure, and many residents who became labor migrants. ASEAS 15(1) | 29 Agus Heruanto Hadna & Media Wahyudi Askar The selection of the low-income individuals as samples in each sub-district was based on the Integrated Database (Basis Data Terpadu [BDT]) issued by the Indonesian Ministry of Social Affairs. The use of the BDT as a sampling frame guar- antees that the entire population in the BDT consisted of low-income individuals. The number of samples per sub-district was 450 with random techniques, and the total number of respondents obtained from random results was 4,050. This amount is approximately 20% of the total of low-income individuals that met the criteria as PKH recipients in the nine sub-districts. After conducting interviews according to a structured questionnaire, the total number of interviewed respondents was 3,933. The number of respondents decreased because some had moved residences, provided incorrect target group data, or died. In addition, some of the low-income people were already in more prosperous categories. As a result, almost 55% of the successfully interviewed respondents were PKH recipients, while the rest were still waiting for PKH assistance. Fifty-five percent of the group received the treatment for this cate- gory; the remaining 45% was the control group. This research used a set of variables that informed the socioeconomic inequality of low-income individuals. The first variables included head of the household, age, gender (Milazzo & van de Walle, 2015; Ravallion, 2016; Rosenhouse, 1994), and marital status (Ravallion, 2016). The second group of variables was the capacity of the household head, namely, the level of education (Abuya et al., 2011; Beal et al., 2018). Others were the number of dependent family members (Geberselassie et al., 2018; Mahmudiono et al., 2017; Olinto et al., 2013; Ravallion, 2016) and the employment of the household head (Handayani et al., 2017). Finally, the third group of variables was the behavior, namely the effort (Gans, 1995), of the household head and ownership of health insurance. Methodology This study investigates the relationship between PKH on poverty on specific, low-income individual quantiles, which allows us to identify the impact of PKH among the lowest quintile of poverty groups compared to the highest using Instrumental Variable Quintile Treatment Effect. The model was based on the conventional poten- tial (latent) framework. As Hansen and Chernozhukov (2005) discussed, a possible outcomes estimate influenced by variation of observation units is calculated against potential treatment called t ε, (0,1) and represented as Y�. The potential outcomes of Y� are latent, which reflects the chosen treatment t. Thus, the estimated values of poten- tial outcomes for every observational unit reflect only the impact of possible treatment. Y = Y� This research estimates the potential outcome of specific quantiles called τ-th under potential treatment t, conditional on some observed characteristics X, and can be described as follows: 𝑞(𝑡, 𝑋, 𝜏) Thus, the effect of quantile treatment is as follows: 𝑞(1, 𝑋, 𝜏) − 𝑞(0, 𝑋, 𝜏) 30 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia This formula represents the difference in the impact of treatment on each quan- tile of potential outcomes. As the treatment in this study is potentially endogenous, an IV identification strategy is employed to obtain robust estimates. Instrumental variable Z is to define potential treatment called 𝑇z under which the model can estimate the potential out- comes via the nonlinear conditional moment restrictions of quantile regression: 𝑃(𝑌 ≤ 𝑞(𝑡, 𝑋, 𝜏) ∥ 𝑋, 𝑍 ∥= 𝜏 Z reflects an instrument that is related to t but uncorrelated to potential out- comes Y. This study estimates exogenous variations in CCT programs in Indonesia. Z refers to an instrument that is rural municipality officials. Arguably, rural municipalities’ officials, who place higher engagement levels on their citizens, may increase the chances of individuals receiving cash under conditional transfer programs. Thus, the variable correlates with CCTs but is mainly beyond the respondents’ control and is exogenous to their activities and economic outcomes. RESULTS Table 1 describes the variables included in the model specification and reports the summary statistics. The data suggest that approximately 55% of low-income respondents received CCT programs. Some 43% of the respondents were male, and 17% were single. Most respondents had small families of less than three members. Only 10% of the respondents were gainfully employed. The education level of the sample was relatively low, with most respondents being elementary school gradu- ates and only 1% being university graduates. However, the ownership of healthcare accounts was very high in the sample, with most respondents claiming healthcare program membership. The IVQTR allowed us to obtain information on how the effects of PKH differed at various quantiles of the consumption expenditure distribution (see Table 2). The findings suggest that the treatment group tended to have a higher consumption expenditure than the control group, meaning a positive effect at every percentile of the distribution—the coefficient increased from the lowest to the highest quan- tile of the distribution regarding consumption expenditure. At the bottom quantile of consumption expenditure, the size of the effect was the smallest (1.596). It is not significant at the 1% level, indicating that PKH did not impact wealthier people. The impact for the highest quantile doubled the impact corresponding to the low- est quantiles. In other words, these results indicated that the poor with lower-level consumption expenditure were likely to obtain fewer benefits from the PKH initia- tives. As a result, the poorest people have less chance of improving their quality of life than the wealthiest. Furthermore, the results indicated that, despite being insig- nificant at the 1% level, individuals categorized as having paid work, high levels of education (university), and access to healthcare are likely to have high consumption expenditures. ASEAS 15(1) | 31 Agus Heruanto Hadna & Media Wahyudi Askar Variable Obs Mean Std. Dev. Min Max Poverty 3933 10.32 9.94 0 100 CCT 3933 0.55 0.50 0 1 Effort 3933 0.36 0.48 0 1 Age 3933 52.44 79.24 17 98 Gender 3933 0.43 0.49 0 1 Family members 3933 2.16 1.30 1 10 Marital status 3933 0.17 0.91 0 1 Paid work 3933 0.10 0.30 0 1 Unemployed 3933 0.60 0.49 0 1 Diffable 3933 0.09 0.28 0 1 Retired 3933 0.01 0.09 0 1 Student 3933 0.00 0.04 0 1 Housekeeping 3933 0.05 0.22 0 1 No school 3933 0.29 0.46 0 1 Elementary school 3933 0.34 0.47 0 1 Junior high school 3933 0.18 0.39 0 1 Senior high school 3933 0.17 0.38 0 1 University 3933 0.01 0.09 0 1 Healthcare 3933 0.97 0.18 0 1 Variables q10 q20 q30 q40 q50 q60 q70 q80 q90 (1) (2) (3) (4) (5) (6) (7) (8) (9) CCT 1.596 2.245* 2.839** 2.713** 2.761** 2.832** 3.297** 4.332** 5.951 (1.309) (1.255) (1.130) (1.143) (1.254) (1.260) (1.636) (2.149) (4.608) Effort -0.255 0.193 0.651 1.237 1.612 2.165 2.336 3.039 4.894 (1.267) (1.284) (1.318) (1.341) (1.585) (1.516) (1.876) (2.732) (3.492) Age -0.000 -0.000 -0.000 -0.000 -0.000 -0.000** -0.00** -0.000 -0.000 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Gender -0.042 -0.368 -0.324 -0.753 -0.883 -0.793 -0.842 -0.311 -0.790 (1.159) (1.136) (1.164) (1.211) (1.288) (1.354) (1.772) (2.368) (3.766) Family members 0.855 1.031 1.107 1.285* 1.401** 1.496*** 1.522* 1.892 2.556 (0.698) (0.635) (0.775) (0.757) (0.712) (0.520) (0.860) (1.151) (1.565) Marital status -0.236 -0.293 -0.518 -0.722 -0.963 -1.243* -1.542* -1.844* -2.874* (0.692) (0.572) (0.571) (0.561) (0.651) (0.639) (0.852) (1.071) (1.516) Paid work -1.928 -2.281 -2.197 -1.698 -1.194 -1.089 -1.255 -1.458 0.442 (2.687) (2.343) (2.365) (2.629) (2.825) (3.026) (3.749) (4.743) (7.562) Unemployed -0.401 -0.158 0.144 0.354 0.536 0.768 0.565 0.599 2.559 (2.169) (1.864) (1.819) (2.036) (2.112) (2.462) (3.084) (4.125) (4.827) Diffable -2.804 -2.660 -2.670 -2.915 -2.625 -2.784 -3.172 -3.742 -3.377 (2.254) (1.881) (1.808) (1.966) (2.164) (2.420) (3.019) (4.262) (4.548) Table 1. Descriptive statistics. (compilation by the authors) 32 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia Retired -2.826 -2.940 -3.441 -4.302 -3.907 -4.227 -5.236 -6.986 -7.787 (4.174) (4.294) (3.332) (3.922) (5.206) (4.703) (5.177) (5.630) (8.318) Student 10.615** 8.692** 7.594* 6.638 5.434 4.127 1.705 -2.858 5.554 (4.362) (3.802) (3.984) (4.125) (4.159) (4.495) (5.808) (6.080) (11.916) Housekeeping 2.999 3.055 3.326 3.426 3.878 4.826 5.116 2.881 -0.122 (11.514) (14.068) (13.871) (13.687) (21.956) (19.000) (12.526) (14.966) (11.774) Elementary school -0.145 -0.248 -0.478 -0.608 -0.597 -0.446 -0.734 -1.065 -0.430 (1.356) (1.396) (1.448) (1.446) (1.477) (1.620) (2.007) (2.532) (3.761) Junior high school 0.206 0.549 0.430 0.332 0.704 0.738 0.538 0.227 -0.065 (1.572) (1.660) (1.695) (1.639) (1.653) (1.703) (2.299) (2.728) (4.720) Senior high school 1.402 1.364 1.151 1.153 1.528 1.793 2.350 4.408 5.686 (2.100) (2.044) (2.164) (2.296) (2.414) (2.444) (3.563) (5.447) (11.377) University 5.077 3.801 2.568 1.475 -0.088 2.134 1.105 -0.026 -2.842 (16.656) (13.355) (13.619) (18.399) (17.048) (42.215) (51.145) (55.378) (44.248) Healthcare 1.202 1.577 2.131 2.560 2.572 2.364 2.582 3.109 4.326 (1.749) (1.533) (1.553) (1.675) (1.782) (2.184) (2.669) (3.072) (5.463) Observations 3924 3924 3924 3924 3924 3924 3924 3924 3924 Notes: A constant is included in regressions but not reported. Figures in parentheses are robust standard errors; *, ** and *** represent statistical significance at 10%, 5%, and 1%, respectively. Table 3 reports IVQTE for the distribution of selected consumption expendi- tures, further identifying the effects of PKH on the consumption expenditures share. Even though the program’s impact on main food is not statistically significant for all the quantiles of main food expenditure, the results suggested that the transfer was spent mainly on vegetables, except for the first (lowest) quantile. The program did not significantly affect the first (lowest) quantile of vegetable expenditure. However, the results turned out to be significant for the second through the ninth quantiles, suggesting that PKH increased the distribution of spending on vegetables. The ini- tial impact on vegetables ranged from 2.857% to 9.897% for the highest quantiles. However, the effects are not statistically significant on fruit expenditures. Although the data suggested an insignificant relationship between PKH and expendi- tures on vegetables and fruits, those in the lowest decile group did not necessarily have low consumption of vegetables or fruit. Overall, these group can provide at least two meals a day. However, they usually did not spend money on vegetables and fruits. Thus, the availability of vegetables and fruits, both self-grown and wild vegetables, is still enough to be consumed. Nonetheless, it is essential to note that the land-use change tendency, which reduces agriculture and forest areas, has threatened the food supplies of the most impoverished communities. Next, Table 3 shows that the program’s impact on education expenditure varied. While the effect is not significant for the first to the third quantiles, the program sub- stantially increased education expenditure from the fourth quantile. The program statistically increased education expenditure from 0.682 to 2.094 percentage points for the second-highest quantile. Again, many of the impacts were quite large and Table 2. Instrumental variable quintile treatment effect on the distribution of total consumption expenditure. (compilation by the authors) ASEAS 15(1) | 33 Agus Heruanto Hadna & Media Wahyudi Askar significant at the mean quantiles, suggesting that the program had more influence on individuals at the mean distribution in terms of education spending. Variable Dependent variables q10 q20 q30 q40 q50 q60 q70 q80 q90 (1) (2) (3) (4) (5) (6) (7) (8) (9) Main food CCT 0.572 0.463 0.494 0.787 1.136 1.384 1.481 0.746 0.297 (0.739) (0.990) (0.521) (0.629) (0.919) (1.013) (1.312) (1.106) (1.247) Fruits CCT 0.000 0.000 0.000 0.000 0.000 -0.000 1.153 1.750 2.544* (0.005) (0.011) (0.016) (0.021) (0.028) (0.100) (1.092) (1.071) (1.453) Vegetables CCT 2.204 2.857* 3.577*** 3.879*** 4.303** 5.150*** 5.611*** 7.562*** 9.897*** (1.378) (1.490) (1.385) (1.383) (1.996) (1.709) (1.946) (2.150) (3.775) Cigarettes CCT 0.000 0.000 0.000 0.000 -0.769 -1.419 -3.229 -5.894 -9.617* (0.022) (0.039) (0.056) (0.074) (0.763) (1.153) (2.649) (4.092) (5.177) Education CCT 0.000 0.000 0.000 0.682* 1.012** 1.303** 1.614* 2.094* 2.624* (0.058) (0.080) (0.373) (0.409) (0.493) (0.664) (0.948) (1.131) (1.832) Health CCT 0.000 0.000 0.000 0.000 -0.000 -0.000 -0.079 -0.543 -3.138 (0.012) (0.018) (0.025) (0.033) (0.162) (0.188) (0.234) (0.993) (4.048) Housing CCT -0.077 -0.117 -0.178 -0.312 -0.540 -0.666 -0.832 -0.639 -0.791 (0.168) (0.204) (0.257) (0.328) (0.382) (0.483) (0.634) (0.902) (1.013) Social activities CCT 0.000 0.042 0.063 0.133 0.250 0.098 0.373 0.290 -0.073 (0.020) (0.253) (0.344) (0.419) (0.648) (0.825) (1.096) (0.994) (1.619) Communication CCT 0.000 1.285 2.309* 3.028** 3.514*** 4.243*** 6.003*** 7.325*** 11.539*** (0.083) (1.334) (1.260) (1.266) (1.301) (1.628) (1.935) (2.207) (4.316) Religion CCT 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.762* (0.002) (0.004) (0.007) (0.009) (0.013) (0.016) (0.022) (0.082) (0.456) 34 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia Transportation CCT 0.000 0.002 1.155 1.386 1.782 2.379* 2.947 5.199 6.909** (0.047) (0.486) (1.035) (0.970) (1.256) (1.406) (1.967) (3.191) (3.053) Total expenditure CCT 1.535 2.055 2.435* 2.399* 2.271 2.308 2.291 2.128 3.182 (1.741) (1.583) (1.420) (1.412) (1.468) (1.445) (1.636) (2.346) (4.327) Notes: A constant is included in regressions but not reported. Figures in parentheses are robust standard errors; *, ** and *** represent statistical significance at 10%, 5%, and 1%, respectively. All regression in- cludes controls for effort, age, gender, family members, marital status, type of jobs, education, and health care. The data from the Indonesian Statistics Bureau in 2019 showed that only 14% of household members have not gone to school. This statistic indicates that the bur- den of low-income people for education expenditures, especially those in the lowest quantile, is minimal. The interpretation of the small percentage of spending on edu- cation is that the poorest people, namely those in the lowest quantile, do not have adequate education for their children. Despite government efforts to eliminate pri- mary and secondary education costs, some additional charges attached to children’s education, such as school uniforms, transportation, and learning aids, are still not included in the assistance, preventing the poorest group from accessing education. Even though health is a priority for the intervention of PKH in Indonesia, the pro- gram impacts on health expenditure were not statistically significant for all quantiles. The interpretation of the small percentage of health expenditure is that the lowest quantile had inadequate access to a health facility for their children or their pregnant wives. In addition, this study explored the impacts of PKH on the distribution of communication expenditure. Interestingly, the results significantly affected each quantile, except the first and the second-lowest quantile. The program’s impact on communication expenditure was statistically significant and ranged from about 2.309 percentage points for the third quantiles to 11.539 percentage points for the highest quantile. In 2018, the data from the Indonesian Statistics Bureau showed that the number of internet users in rural areas was 55.45% of the total population. This number was not significantly different from the people who do not own and use cell phones. Here non-internet users were the most disadvantaged group with little or no spending on communication. Regarding the program’s impact on transportation expenditure, the effect varied and was only significant for the sixth and the highest quantiles. It is also important to note that the program impacted selected expenditures such as cigarettes, hous- ing, social activities, and religion; these were not statistically significant. Overall, this study showed that the PKH program increased consumption expenditure. However, there were heterogeneous impacts of the PKH program. The results indicated that the program’s impact was not significant for the poor at the lowest level of expenditures. Table 3. Instrumental variable quintile treatment effect on the distribution of selected consumption expenditure. (compilation by the authors) ASEAS 15(1) | 35 Agus Heruanto Hadna & Media Wahyudi Askar DISCUSSION The findings of this study make two main contributions to the current literature. First, this study implies re-evaluating methods of identifying the ‘poor.’ Second, it suggests the existing poverty lines are ill-suited, as they force us to segregate the population into poor and non-poor groups. Currently, the essential criteria for deter- mining eligibility and registering beneficiaries rely on the regional poverty line and other additional measures (e.g., elderly, number of children, or a combination of categories). Then, the CCT program will be provided with the same amount to ben- eficiaries regardless of everyone’s levels of wealth. This approach may hinder CCT’s targeting efficiency by delivering the money to the poor households in the highest quantile of total wealth. This problem is related to the determination of poverty lines (Ravallion, 2016), which is often an obstacle in measuring poverty because it carries significant implications for the beneficiary group. Hence, there needs to be a better targeting process in CCT distribution to avoid inequality of CCT beneficiaries if the program’s criteria and characteristics depend on wealth levels among beneficiaries. Extra effort is needed, especially in this setting, by grouping the poor by wealth. By grouping the recipients (e.g., the lowest, middle, and highest quantiles), we can assign the different level priorities of the program, from the poorest to the wealthiest groups. Second, this research also implies the need to make CCT programs more agile and based on the needs for different types and forms of poverty. For example, an impov- erished household with low levels of education, which includes women and elderly, and has little money left should receive more cash to improve their survival strategy. Furthermore, the data showed that as many as 43.6% of the heads of these low-in- come households are elementary school graduates. It is also important to note that Yogyakarta had its highest percentage of elderly in Indonesia in 2018 (18.76%). The number of poor people in Yogyakarta over 64 years old (elderly) is 16%. This group can no longer generate substantial income and needs extra cash to fulfill its needs. Thus, households in the lowest wealth quantile may need to receive payments quickly. Education investments can be the second priority behind eating well for this group. As shown above, the CCT does not improve the share of expenditure on edu- cation among impoverished households. Besides, households in the lowest wealth quantile may never expect business to change their lives. Those in the top quantile may quickly expand their business and earn money, but households at the bottom of the quantile choose not to do so. This research suggests that CCT beneficiaries should not all receive the same amounts of cash. Low-income individuals in the highest wealth quantiles may receive less money than those in the lowest. Other incentives, such as training vouchers, education vouchers, and transportation incentives may be more suitable for these groups. Although they still need some support to avoid further income decline, they are no longer vulnerable to food insecurity. This finding is related to the view of Ravallion (2016) that the CCT program design is highly paternalistic. This outlook places the poor as a CCT object because it cannot understand their needs. The CCT should provide an excellent opportunity for the poor to determine their own family’s needs. 36 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia This research finding also suggested that we need a new strategy for assisting the poor. Assistance programs for the poorest beneficiaries may focus on improving access to health, education, sanitation, housing facilities, and abilities to optimize financial management. In terms of food security, it is more appropriate to develop programs that are not providing cash and emergency food aid. Assistance for the poorest beneficiaries is an excellent program because the poorest group spends a large share of their income on food. The data showed that the consumption of the food expenditure per capita of the lowest 10% of the population was the lowest on average (IDR 210,000). However, it was the highest in terms of the proportion of expenditures on food (65.4 %). Also, the value of per capita non-food spending in the lowest quantile was only IDR 111,000 or approximately 34.6%, compared to the non-food expenditure per capita in the highest quantile, which was IDR 3,350,000 or approximately 76.3%. Furthermore, evaluating the program’s effectiveness for the poorest beneficiar- ies should also be carried out in shorter periods of three to six months, given their limited abilities to optimize their minimal resources. On the other hand, for CCT beneficiaries in the highest wealth quantile, the CCT mentoring scheme may raise productivity and employment capacity. Moreover, these CCT beneficiaries can use technology to apply long-term outcome evaluation and mentoring activities. Third, this research showed an interesting trend in consumption patterns. The target group of the CCT program is comprised of vulnerable communities, with the hope of a substantial impact on school enrollment. Indeed, this study showed an increase in the share of education. Still, some extra spending also increased (e.g., communication and transportation), especially among households in the highest wealth quantile. This study indicated that the probability of wasting money from the CCT program is relatively higher among the upper poor quantile. This data demon- strated that CCT is likely to contribute to increased inequality among low-income individuals. Thus, it is essential to note that the CCT should not pursue the number of CCT beneficiaries, but rather strengthen assistance to the extremely poor groups, and at the same time, increase the amount of funding. A smaller and more precise targeting scheme can potentially increase the effectiveness of CCT. Together, the evidence presented in this study suggests the significance of CCT programs similar to those described in previous studies (Attanasio & Mesnard, 2006; Lagarda et al., 2017; Maluccio & Flores, 2005; Rizal & van Doorslaer, 2019). However, this study shines a new light on these debates by highlighting the different impacts among CCT beneficiaries. For example, poor people in the lowest wealth quantile may receive fewer benefits than those in the highest, potentially increasing inequality. Hence, improving targeting mechanisms used by CCTs should be the main pri- ority for policymakers. This improvement requires political will and revolutionary change in the existing tool. Unfortunately, political will has often failed to emerge due to limited scientific evidence on this issue. This study reminds governments that the poor have problems with their welfare levels. Therefore, governments must serve the poor, especially in developing countries, who often experience limited social spending. In this regard, governments should be more generous to households in the lowest wealth quantile, and carefully manage the program based on the needs of CCT beneficiaries. ASEAS 15(1) | 37 Agus Heruanto Hadna & Media Wahyudi Askar CONCLUSION This study fills the gap in the literature by providing an empirical analysis of the nexus of CCT and inequality in Indonesia. Estimations using the IVQTR ensure the robustness of the findings. The analysis showed that CCT has a positive and statis- tically significant impact on an individual’s consumption. CCT has indeed increased the consumption expenditure of low-income individuals. However, it is essential to note that the coefficient is substantially higher among the highest quantile of the distribution. The lowest quantile of low-income individuals is the slowest to respond to these changes. This means that the program is less effective for relatively more impoverished individuals. The finding of the impact of CCT on inequality among the poor provides a similar picture of policies at the macro level, which impacts the high Gini Ratio in Yogyakarta. The more prosperous groups have a higher rate of change than those below them in this region. In the end, the novelty of this study is that the current CCT design provides opportunities for gaps between the poor themselves. If the most destitute benefit less from the program and are left behind, the inequalities already experienced will exacerbate. Given the vital role of CCT in improving an individual’s economic welfare, the policy alternative to solving this problem would be readjusting the amount of cash, with the poorest receiving more money than the wealthier recipients. The CCT should not pursue the number of CCT beneficiaries but rather strengthen assis- tance to the extremely poor groups. The Government of Indonesia may also need to develop tailored economic and social assistance programs for the lowest quantile of CCT recipients. These recipients are likely to have low levels of education and less money in their pockets; most of them are women as well. This research only examined the impact of CCT in rural areas, and thus, has limi- tations in terms of scope. Therefore, future research should look at the same issue in urban areas, and make general conclusions regarding the relationship between CCT and the inequality of outcomes among program beneficiaries.  REFERENCES Abuya, B. A., Onsomu, E. O., Kimani, J. K, Moore, D. (2011). Influence of maternal education on child immunization and stunting in Kenya. Maternal and Child Health Journal, 15(8), 1389–1399. Aghion, P., Caroli, E., & Garcia-Penalosa, C. (1999). Inequality and economic growth: The perspective of the new growth theories. Journal of Economic Literature, 37(4), 1615–1660. Arif, S., Syukri, M., Isdijoso, W., Rosfadhila, M., & Soelaksono, B. (2013). Is conditionality pro-women? A case study of conditional cash transfer in Indonesia. SMERU Research Institute Working Paper, February, 2013. https://smeru.or.id/sites/default/files/publication/conditionalcashtransfer.pdf Attanasio, O., & Mesnard, A. (2006). The impact of a conditional cash transfer programme on consumption in Colombia. Fiscal Studies, 27(4), 421–442. Baird, S., Ferreira, F. H. G., Özler, B., & Woolcock, M. (2014). Conditional, unconditional and everything in between: A systematic review of the effects of cash transfer programmes on schooling outcomes. Journal of Development Effectiveness, 6(1), 1–43. 38 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia Banerjee, A. V, Hanna, R., Kreindler, G. E., & Olken, B. A. (2017). Debunking the stereotype of the lazy wel- fare recipient: Evidence from cash transfer programs. The World Bank Research Observer, 32(2), 155–184. Beal, T., Tumilowicz, A., Sutrisna, A., Izwardy, D., & Neufeld, L. M. (2018). A review of child stunting deter- minants in Indonesia. Maternal and Child Nutrition, 14(4), 1–10. Beaudoin, S. M. (2006). (Ed.). Poverty in world history. Routledge. Behrman, J. R., Sengupta, P., & Todd, P. (2005). Progressing through PROGRESA: An impact assessment of a school subsidy experiment in rural Mexico. Economic Development and Cultural Change, 54(1), 237–275. Bergh, A., & Nilsson, T. (2014). Is globalization reducing absolute poverty? World Development, 62, 42–61. Cahyadi, N., Hanna, R., Olken, B. A., Prima, R. A., Satriawan, E., & Syamsulhakim, E. (2018). Cumulative impacts of conditional cash transfer programs: Experimental evidence from Indonesia. NBER Working Paper Series, May, 1–23. Chaudhury, N., Friedman, J., & Onishi, J. (2013). The World Bank Philippines conditional cash transfer program impact evaluation 2012. Human Development Sector Unit East Asia and Pacific Region, Report Number 75533-PH, 1–80. http://documents.worldbank.org/curated/en/537391468144882935/pdf/75533 0REVISED000Revised0June0402014.pdf Chernozhukov, V., & Hansen, C. (2005). An IV model of quantile treatment effects.  Econometrica,  73(1), 245–261. Coudouel, A., Hentschel, J. S., & Wodon, Q. T. (2013). Poverty measurement and analysis. In Munich Personal RePEc Archieve (Issue 45362). MPRA Munich. https://mpra.ub.uni-muenchen.de/45362/1/ MPRA_paper_45362.pdf Cutler, B. D. M., & Katz, L. F. (1992). American Economic Association rising inequality? Changes in the distribution of income and consumption in the 1980's. The American Economic Review, 82(2), 546–551. de Janvry, A., Finan, F., Sadoulet, E., & Vakis, R. (2006). Can conditional cash transfer programs serve as safety nets in keeping children at school and from working when exposed to shocks? Journal of Development Economics, 79(2), 349–373. Dollar, D., & Kraay, A. (2002). Growth is good for the poor. Journal of Economic Growth, 7(3), 195–225. Duflo, E. (2003). Grandmothers and granddaughters: Old-age pensions and intrahousehold allocation in South Africa. World Bank Economic Review, 17(1), 1–25. Evans, D. K., & Popova, A. (2014). Cash transfers and temptation goods: A review of global evidence. The World Bank Policy Research Working Paper, May(6886), 1–3. http://documents.worldbank.org/curated/ en/2014/05/19546774/cash-transfers-temptation-goods-review-global-evidence Falkingham, J., & Namazie, C. (2002). Measuring health and poverty: A review of approaches. http://eprints. soton.ac.uk/35016/ Farrington, J., & Slater, R. (2006). Introduction: Cash transfers: Panacea for poverty reduction or money down the drain? Development Policy Review, 24(5), 499–511. Fernald, L. C. H., Gertler, P. J., & Neufeld, L. M. (2008). Role of cash in conditional cash transfer programmes for child health, growth, and development: An analysis of Mexico's Oportunidades. Lancet, 371(9615), 828–837. Filmer, D. (1999). Child mortality and public spending on health: How much does money matter? Policy Research Working Papers. The World Bank. https://doi.org/10.1596/1813-9450-1864 Franzoni, J. M., & Voorend, K. (2012). Blacks, whites, or grays? Conditional transfers and gender equality in Latin America. Social Politics, 19(3), 383–407. Gans, H. (1995). The war against the poor: The underclass and antipoverty policy. Harper Collins Publishers. Garoma, D. A., Abraha, Y. G., Gebrie, S. A., Deribe, F. M., Tefera, M. H., & Morankar, S. (2017). Impact of conditional cash transfers on child nutritional outcomes among sub-Saharan African countries: a systematic review protocol. JBI Evidence Synthesis, 15(9). https://journals.lww.com/jbisrir/ Fulltext/2017/09000/Impact_of_conditional_cash_transfers_on_child.12.aspx Geberselassie, S. B., Abebe, S. M., Melsew, Y. A., Mutuku, S. M., & Wassie, M. M. (2018). Prevalence of stunting and its associated factors among children 6-59 months of age in Libo-Kemekem district, Northwest Ethiopia: A community based cross sectional study. PLoS ONE, 13(5), 1–11. Golan, J., Sicular, T., & Umapathi, N. (2017). Unconditional cash transfers in China: Who benefits from the rural minimum living standard guarantee (Dibao) program? World Development, 93(1306), 316–336. ASEAS 15(1) | 39 Agus Heruanto Hadna & Media Wahyudi Askar Hadna, A. H., & Kartika, D. (2017). Evaluation of poverty alleviation policy: Can conditional cash transfers improve the academic performance of poor students in Indonesia? Cogent Social Sciences, 3(1), 1295548. Ham, A. (2014). The impact of conditional cash transfers on educational inequality of opportunity. Latin American Research Review, 49(3), 153–175. Handayani, F., Siagian, A., & Aritonang, E. Y. (2017). Mother's education as a determinant of stunting among children of age 24 to 59 months in North Sumatra Province of Indonesia. IOSR Journal of Humanities and Social Science, 22(06), 58–64. Hansen, C., & Chernozhukov. (2005). An Instrumental Variable Model of Quantile Treatment Effects. Econometrica, 73(1), 245-261. Haushofer, J., & Shapiro, J. (2016). The short-term impact of unconditional cash transfers to the poor: Experimental evidence from Kenya. The Quarterly Journal of Economics, 131(4), 1973–2042. Honjo, K., Verhoeven, M., & Gupta, S. (1997). The efficiency of government expenditure: Experiences from Africa. IMF Working Papers, 97(153). https://doi.org/10.5089/9781451922400.001 Indonesia Statistic Bureau. 2019. Potret Pendidikan Indonesia-Statistik Pendidikan 2019. Katalog BPS 4301008. https://www.bps.go.id/publication/2019/11/29/1deb588ef5fdbfba3343bb51/potret- pendidikan-statistik-pendidikan-indonesia-2019.html Juan, M.-A., Saavedra, E., Garcia, S., Saavedra, J. E., & García, S. (2012). Impacts of conditional cash transfer pro- grams on educational outcomes in developing countries: A meta-analysis. Labor and Population, WR-921-1, 1–64. https://www.rand.org/content/dam/rand/pubs/working_papers/2012/RAND_WR921-1.pdf Kamakura, W. A., & Mazzon, J. A. (2014). Measuring the impact of a conditional cash transfer program on consumption behavior with propensity scoring. SSRN Electronic Journal. https://link.springer.com/ content/pdf/10.1007/s40547-015-0037-0.pdf Kandpal, E., Alderman, H., Friedman, J., Filmer, D., Onishi, J., & Avalos, J. (2016). A conditional cash transfer program in the Philippines reduces severe stunting. The Journal of Nutrition, 146(9), 1793–1800. Koenker, R., & Hallock, K. F. (2001). Quantile regression. Journal of Economic Perspectives, 15(4), 143–156. Krishnakumar, J., & Juárez, W. C. F. (2012). The impact of Oportunidades on inequality of opportunity in rural and urban areas in Mexico. SSRN Electronic Journal. https://papers.ssrn.com/sol3/papers. cfm?abstract_id=1938284 Laabas, B., & Limam, I. (2004). Impact of public policies on poverty, income distribution and growth. In A. A. G. Ali & S. Fan (Eds.), Public Policy and Poverty Reduction in the Arab Region (pp. 1–36). The Arab Planning Institute. https://ebrary.ifpri.org/utils/getfile/collection/p15738coll2/id/34354/ filename/34355.pdf Lagarda, G., Linares, J., & Gallagher, K. P. (2017). Capital openness and income inequality: A cross-coun- try study and the role of CCTs in Latin America. SSRN Electronic Journal. http://doi.org/10.2139/ ssrn.2849463 Mahmudiono, T., Sumarmi, S., & Rosenkranz, R. R. (2017). Household dietary diversity and child stunting in East Java, Indonesia. Asia Pacific Journal of Clinical Nutrition, 26(2), 317–325. Maluccio, J., & Flores, R. (2005). Impact evaluation of a conditional cash transfer program. International Food Policy Research Institute. Meyer, B. D., & Sullivan, J. X. (2011). Measuring the well-being of the poor using income and consumption. Canadian Journal of Economics, 44(1), 52–87. Milazzo, A., & van de Walle, D. (2015). Women left behind? Poverty and headship in Africa. Policy Research Working Paper, 7331. https://doi.org/10.1007/s13524-017-0561-7 Millán, T. M., Barham, T., Macours, K., Maluccio, J. A., & Stampini, M. (2019). Long-term impacts of condi- tional cash transfers: Review of the evidence. World Bank Research Observer, 34(1), 119–159. Miller, S. M., & Neanidis, K. C. (2015). Demographic transition and economic welfare: The role of in-cash and in-kind transfers. Quarterly Review of Economics and Finance, 58, 84–92. Mirrlees, J. A. (1971). An exploration in the theory of optimum income taxation. The Review of Economic Studies, 38(2), 175–208. Molyneux, M., & Thomson, M. (2011). Cash transfers, gender equity and women's empowerment in Peru, Ecuador and Bolivia. Gender & Development, 19(2), 195–212. 40 | ASEAS 15(1) The Impact of Conditional Cash Transfers on Low-Income Individuals in Indonesia Olinto, P., Beegle, K., Sobrado, C., & Uematsu, H. (2013). The state of the poor: Where are the poor, whe- re is extreme poverty harder to end, and what is the current profile of the world's poor? World Bank Economic Premise, 125, 1–8. Owusu-Addo, E., & Cross, R. (2014). The impact of conditional cash transfers on child health in low- and middle-income countries: A systematic review. International Journal of Public Health, 59(4), 609–618. Schultz, P. T. (2004). School subsidies for the poor: Evaluating the Mexican Progresa poverty program. Journal of Development Economics, 74(1), 199–250. Ponce, J., & Bedi, A. S. (2010). The impact of a cash transfer program on cognitive achievement: The Bono de Desarrollo Humano of Ecuador. Economics of Education Review, 29(1), 116–125. Quisumbing, A. R., Haddad, L., & Pena, C. (1995). Gender and poverty: New evidence from 10 developing countries. FCND Discussion Paper 9. https://ageconsearch.umn.edu/record/97310/?ln=en Ravallion, M. (1992). Poverty comparisons: A guide to concepts and methods. http://documents1.worldbank. org/curated/en/290531468766493135/pdf/multi-page.pdf Ravallion, M. (1995). Growth and poverty: Evidence for developing countries in the 1980s. Economics Letters, 48(3–4), 411–417. Ravallion, M. (1997). Can high-inequality developing countries escape absolute poverty? Economics Letters, 56(1), 51–57. Ravallion, M. (2016). The economics of poverty: History, measurement and policy. Oxford University Press. Rawlings, L. B. (2006). Examining conditional cash transfer programs: A role for increased social inclusion? http://hdl.handle.net/10986/20204 Rawlings, L. B., & Rubio, G. M. (2005).  Evaluating the impact of conditional cash transfer programs. Oxford University Press on behalf of the World Bank. https://openknowledge.worldbank.org/ handle/10986/16406 Rizal, M. F., & van Doorslaer, E. (2019). Explaining the fall of socioeconomic inequality in childhood stun- ting in Indonesia. SSM - Population Health, 9, 100-469. Robertson, L., Mushati, P., Eaton, J. W., Dumba, L., Mavise, G., Makoni, J., Schumacher, C., Crea, T., Monasch, R., Sherr, L., Garnett, G. P., Nyamukapa, C., & Gregson, S. (2013). Effects of unconditional and conditional cash transfers on child health and development in Zimbabwe: A cluster-randomised trial. Lancet, 381(9874), 1283–1292. Rosenhouse, S. (1994). Identifying the poor: Is 'headship' a useful concept? LSMS Working Paper, Second Issue No.58. The World Bank. https://documents1.worldbank.org/curated/en/920711468765037968/ pdf/multi0page.pdf Schneider, F., & Enste, D. H. (2000). Shadow economies: Size, causes, and consequences. Journal of Economic Literature, 38(1), 77–114. Schubert, B., & Slater, R. (2006). Social cash transfers in low-income African countries: Conditional or unconditional? Development Policy Review, 24(5), 571–578. Sen, A. (1981). Public action and the quality of life in developing countries. Oxford Bulletin of Economics and Statistics, 43(4), 287–319. Sen, A. (1985). Commodities and capabilities. North-Holland. Slesnick, D. T. (1996). Consumption and poverty: How effective are in-kind transfers? Economic Journal, 106(439), 1527–1545. Soares, S., Medeiros, M., Osorio, R. G., Soares, F. V., & Zepeda, E. (2009). Conditional cash transfers in Brazil, Chile and Mexico: Impacts upon inequality. Estudios Economicos, 16 (1), 207–224. Squire, L. (1993). Fighting poverty. The American Economic Review, 83(2), 377–382. Stutzer, A., & Frey, B. (2004). Reported subjective well-being: A challenge for economic theory and economic policy. Journal of Applied Social Science Studies, 124(2), 191–231. Tutor, M. V. (2014). The impact of the Philippines' conditional cash transfer program on consumption. The Phillipine Review Economics, 51(1), 117–161. World Bank. (2015). The state of social safety nets 2015. https://openknowledge.worldbank.org/ handle/10986/22101 ASEAS 15(1) | 41 Agus Heruanto Hadna & Media Wahyudi Askar Zuluaga, B., Castellanos, D., & Borda, C. (2020). Conditional cash transfers and gender constructions: A unique dataset on women in Buenaventura, Colombia. Data in Brief, 32, 106-290. ABOUT THE AUTHORS Agus Heruanto Hadna is an Associate Professor at Universitas Gadjah Mada. Currently, he serves as the Head of the Doctoral Program of Leadership and Policy Innovation, Universitas Gadjah Mada, Indonesia. Hadna is also a lecturer at the Department of Management and Public Policy, Faculty of Social and Political Sciences, and a senior researcher at the Center for Population and Policy Studies at the same university. He received his PhD in Public Administration from the University of Duisburg-Essen, Germany, in 2007. His study and research interests are poverty, policy innovation, leadership, and public service. ► Contact: hadna@ugm.ac.id Media Wahyudi Askar is a lecturer at Universitas Gadjah Mada. He completed his undergraduate degree at Universitas Gadjah Mada, Indonesia and holds a master’s and a doctoral degree from the University of Manchester, United Kingdom. His professional expertise varies in financial development, poverty, digital economy, small medium enterprises, and economic inequality. Most of his research sees fiscal justice as key to the fight against extreme economic inequality. ► Contact: media.askar@ugm.ac.id ACKNOWLEDGEMENTS The authors are grateful for the data provided by the Center for Population and Policy Studies, Universitas Gadjah Mada. DISCLOSURE The authors declare no conflict of interest. 42 | ASEAS 15(1)