Clientelism as Informal Social Protection in Unequal Democracies: Evidence from Thailand’s 2023 Election

Clientelism as Informal Social Protection in Unequal Democracies: Evidence from Thailand’s 2023 Election | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clientelism as Informal Social Protection in Unequal Democracies: Evidence from Thailand’s 2023 Election Stithorn Thananithichot This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9049351/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Electoral clientelism is often interpreted as a consequence of socioeconomic vulnerability in unequal democracies. Yet it remains unclear whether compliance with vote buying is primarily structured by demographic disadvantage or activated through situational incorporation into exchange networks. This study examines clientelistic behavior in Thailand’s 2023 general election using nationally administered post-election survey data. Distinguishing between structural indicators of vulnerability (education, age, gender, and residence) and direct exposure to inducements, the analysis constructs multidimensional measures of exposure, behavioral reciprocity, and coercive justification. The findings indicate that while exposure to vote buying remains a visible component of electoral competition, compliance is not independently stratified by demographic characteristics once exposure is taken into account. Nor do coercive expectations form a coherent driver of responsiveness. Instead, behavioral reciprocity is strongly associated with direct incorporation into inducement-based exchange networks. These results suggest that inequality shapes the structural conditions under which informal redistribution becomes meaningful, but exposure structures the behavioral response. By disentangling structural vulnerability from situational activation, the study advances debates on clientelism, informal social protection, and stratified political incorporation in middle-income democracies. Electoral Clientelism Vote Buying Informal Social Protection Political Inequality Thailand Figures Figure 1 Introduction Electoral clientelism remains a persistent feature of democratic politics in many unequal societies. Across Asia, Latin America, and parts of Eastern Europe, political actors continue to distribute cash, food, and material goods in exchange for electoral support (Aspinall, 2015 ; Carlin & Moseley, 2022 ; Mares & Young, 2019 ). A substantial body of scholarship interprets these practices as products of socioeconomic inequality, linking vote buying to poverty, limited education, and structural dependency (Della Porta & Vannucci, 2017 ; Jensen & Justesen, 2014 ; Scott, 1969 ). From this perspective, clientelism flourishes where vulnerability constrains voter autonomy and reinforces elite domination (Elliott, 2016 ; Kitschelt & Wilkinson, 2007 ). Poorer and less educated citizens are therefore commonly assumed to be more susceptible to inducements and more likely to reciprocate politically (Justesen & Manzetti, 2023 ). Although inequality undoubtedly shapes political opportunity structures, much of the existing literature implicitly assumes a relatively direct relationship between socioeconomic disadvantage and compliance with vote buying (Corstange, 2018 ; Hidalgo & Nichter, 2016 ). This assumption leaves an important theoretical and empirical question unresolved. Far less attention has been devoted to distinguishing whether clientelistic behavior is primarily driven by demographic vulnerability itself or activated through situational exposure to exchange networks that may cut across social strata. Put differently, it remains unclear whether clientelism is concentrated among disadvantaged voters or operates as a broader mechanism of electoral mobilization embedded within unequal yet competitive democracies. Understanding whether clientelism is demographically stratified or situationally activated therefore has implications not only for Thailand but for unequal democracies more broadly, where electoral exchange often coexists with expanding yet uneven systems of social welfare. This distinction is especially consequential for research on population dynamics and social stratification. If clientelistic compliance is strongly patterned by education, age, or place of residence, it may reinforce existing inequalities in political voice and representation. If, however, compliance is more closely associated with exposure to inducements rather than demographic characteristics, clientelism may operate less as a fixed hierarchy of dependency and more as a relational mechanism activated within specific electoral contexts. Clarifying this distinction is therefore essential for understanding how inequality intersects with democratic participation in societies undergoing demographic and institutional change. This study addresses this question by examining clientelistic behavior in Thailand’s 2023 general election. Thailand provides a theoretically relevant setting for such analysis. It is a middle-income democracy characterized by persistent regional inequality, rapid educational expansion, urban–rural migration, and long-standing public debates over vote buying and populism. These transformations raise a central question: in a society experiencing rising educational attainment and structural change, does clientelistic compliance remain concentrated among socioeconomically disadvantaged groups? This article advances the argument that clientelism in unequal democracies may function as a form of informal social protection. In contexts where formal welfare provision remains uneven or insufficient, material inducements can operate as short-term redistributive transfers delivered through political networks. Under this perspective, inequality shapes the structural environment that renders inducements meaningful, but compliance may be activated through situational incorporation into exchange relationships rather than predetermined by demographic vulnerability alone. To evaluate this argument, the study analyzes nationally administered post-election survey data collected following Thailand’s 2023 general election. It constructs multidimensional measures of clientelistic exposure, behavioral reciprocity, and perceived coercive justification. By distinguishing structural demographic indicators—education, age, gender, and residence—from direct exposure to inducements, the analysis directly tests whether socioeconomic disadvantage independently predicts compliance once exchange incorporation is taken into account. Rather than presuming that poverty or limited education mechanically translate into political dependency, this study systematically examines the demographic distribution and behavioral mechanisms of clientelistic exchange. In doing so, the study contributes to population and social studies scholarship in three ways. First, it refines the relationship between inequality and political participation by distinguishing structural vulnerability from situational activation. Second, it provides empirical evidence on how demographic characteristics intersect with exchange-based electoral practices in a middle-income democracy undergoing social transformation. Third, it contributes to debates on redistribution and social protection by highlighting how informal exchange mechanisms may coexist with—and in some contexts partially substitute for—formal welfare institutions. Distinguishing whether clientelism is demographically stratified or situationally activated carries important implications for policy and institutional reform. If compliance is primarily driven by structural vulnerability, long-term socioeconomic development may gradually weaken clientelist practices. If exposure-based mechanisms dominate, however, reforms targeting electoral incentives, campaign organization, and broker-mediated mobilization may be equally critical. By situating clientelism within broader processes of demographic change and social stratification, this study offers a more nuanced account of electoral behavior in unequal societies. The remainder of the article proceeds as follows. The next section reviews scholarship on inequality, clientelism, and electoral exchange and develops a theoretical framework distinguishing structural vulnerability from situational activation and reciprocity-based mechanisms. The subsequent section outlines the data, survey design, measurement strategy, and analytical approach. The findings section presents descriptive patterns of exposure and compliance, followed by multivariate analyses testing the competing hypotheses. The final section discusses the broader implications for inequality, social protection, and democratic participation and concludes with reflections on policy and future research. Theoretical Framework Structural Vulnerability and the Inequality Thesis A dominant strand of scholarship conceptualizes clientelism as a political response to structural inequality (Kitschelt & Wilkinson, 2007 ; Remmer, 2007 ; Stokes, Dunning, & Nazareno, 2013 ; Wantchekon, 2003 ). In settings where income distribution is highly skewed and formal social protection systems remain fragmented, discretionary, or exclusionary, targeted electoral transfers may substitute for universal welfare provision (Schaffer, 2004; Speck & Abramo, 2001). Under such conditions, electoral exchange becomes an adaptive mechanism through which citizens secure short-term material benefits in contexts of limited state reliability (Callahan, 2005 ; Bowie, 2008 ). From this perspective, vote buying is not merely electoral corruption but a distributive strategy embedded within unequal political economies (Jensen & Justesen, 2014 ). Within this framework, lower-income and less-educated citizens are presumed to be more susceptible to clientelistic inducements because immediate material gains carry greater marginal utility than distant or uncertain policy commitments. Clientelism thus reflects structural vulnerability rather than individual irrationality. Empirical studies in Southeast Asia—particularly from the Philippines—demonstrate that monetary and food-based inducements are frequently targeted toward economically precarious households (Canare, Mendoza, & Lopez, 2018 ), reinforcing the view that poverty shapes both the supply and demand of electoral exchange. The vulnerability thesis rests on two core assumptions. First, socioeconomic disadvantage reduces bargaining power. Individuals with fewer resources may be less able to reject inducements or demand programmatic commitments. Where public services are distributed arbitrarily, patronage networks may substitute for universal entitlements (Callahan, 2005 ). Second, material inducements possess greater relative value for economically precarious citizens. The short-term utility of cash or goods may outweigh the uncertain benefits of long-term policy platforms. Evidence from poor urban communities in the Philippines illustrates strong poverty-targeting patterns in monetary vote buying (Canare, Mendoza, & Lopez, 2018 ). If this framework holds empirically, structural indicators—such as education, age, income proxies, employment precarity, or rural residence—should significantly predict clientelistic compliance even after controlling for exposure to inducements. In other words, demographic vulnerability should exert an independent effect on behavioral responsiveness, beyond mere contact with inducement-based campaigns. However, comparative evidence complicates this deterministic expectation. Studies in Southeast Asia suggest that while exposure may be concentrated among poorer voters, compliance remains conditional rather than automatic (Aspinall, 2015 ; Owen, 2013 ). Voters frequently distinguish between accepting material benefits and relinquishing electoral autonomy, indicating that exchange may operate through negotiated reciprocity rather than strict economic coercion (Chattharakul, 2010 ; Cruz, 2019 ). Moreover, the marginal-utility logic need not be confined to the poorest citizens. In candidate-centered systems characterized by intense personal vote competition, inducements may be distributed broadly rather than narrowly targeted to disadvantaged groups (Hicken, 2011 ). Where turnout uncertainty shapes party strategy, clientelist mobilization may function as a generalized electoral tactic rather than a poverty-specific instrument (Singh, 2018 ). Inequality, therefore, may create the structural conditions under which clientelism emerges, but it does not necessarily determine individual-level compliance. As Bowie ( 2008 ) demonstrates in the Thai case, escalations in vote buying have coincided with institutional reform and rising electoral stakes rather than with static rural poverty. The central empirical question, then, is whether socioeconomic disadvantage independently predicts compliance once exposure is taken into account—or whether behavioral responsiveness is activated through situational incorporation into exchange networks. Reciprocity, Exchange, and Situational Activation An alternative perspective conceptualizes clientelism not primarily as a function of structural vulnerability, but as a form of reciprocal exchange embedded in everyday social norms and relational networks (Fergusson, Molina, & Robinson, 2022 ; Ravanilla, Haim, & Hicken, 2022 ). Rather than emerging solely from economic desperation or coercive domination, electoral inducements may activate culturally embedded expectations of return. Under this view, compliance is less determined by demographic status and more by situational incorporation into exchange relationships. In Thailand, broker-mediated networks (หัวคะแนน) operate through community ties, kinship structures, and localized trust relations (Callahan & McCargo, 1996 ; Chattharakul, 2010 ; Owen, 2013 ). Inducements are rarely isolated transactions; they are delivered through intermediaries who maintain ongoing social relationships. Monitoring is often social rather than technological, functioning through inference, visibility, and reputation rather than direct ballot surveillance (Owen, 2013 ). Such relational embeddedness lowers enforcement costs and reduces the need for overt coercion. Within this framework, compliance is activated through exposure. Once a voter receives a benefit—whether cash, goods, or assistance—a norm of reciprocity may be engaged. The exchange may be interpreted not strictly as bribery but as recognition, obligation, or pragmatic mutual support. Studies of hybrid voting behavior indicate that voters frequently distinguish between accepting inducements and relinquishing electoral autonomy (Birch, Cockshott, & Renaud, 2014 ; Taylor, 2017 ). Acceptance does not mechanically imply obedience; however, incorporation into exchange networks may nonetheless generate reciprocal responsiveness. This perspective shifts analytical focus from structural identity to situational activation. If reciprocity mechanisms dominate, exposure to inducements should strongly predict compliance across demographic strata. Education, age, and residence should exert weaker independent effects once exposure is accounted for. Behavioral responsiveness, in other words, should reflect incorporation into relational exchange rather than preexisting socioeconomic vulnerability. Importantly, this framework also clarifies why coercive perceptions and behavioral compliance may not align. Empirical research in Southeast Asia shows that voters may comply without expressing strong fear of retaliation (Canare, Mendoza, & Lopez, 2018 ). Historical analyses of Thailand similarly suggest that intensification of vote buying has coincided with heightened electoral competition rather than systematic intimidation (Bowie, 2008 ; Hicken, 2011 ). Exchange thus appears embedded within competitive institutional environments rather than enforced through uniform domination. Under a situational activation model, clientelism operates as a socially embedded mobilization mechanism. Exposure—not vulnerability alone—triggers behavioral responsiveness. The empirical implication parallels but contrasts with the vulnerability thesis: if incorporation into exchange networks drives compliance, exposure should emerge as the dominant predictor of reciprocal behavior, attenuating the independent effects of demographic disadvantage. Clientelism as Informal Social Protection Integrating structural inequality and reciprocity-based exchange perspectives allows clientelism to be reconceptualized as a form of informal social protection. In unequal democracies where formal redistribution mechanisms remain uneven, discretionary, or selectively administered, political actors may provide targeted material benefits that temporarily mitigate economic insecurity (Bardhan & Mookherjee, 2020 ). These inducements resemble short-term redistributive transfers delivered through electoral channels rather than institutionalized welfare systems. In Thailand, reliance on patronage networks has historically emerged in contexts where public services are perceived as arbitrarily distributed (Boossabong, 2025 ; Callahan, 2005 ). Moreover, escalations in vote buying have coincided with institutional reform and rising electoral stakes, which increased the distributive value of political office (Bowie, 2008 ). Clientelism therefore operates at the intersection of inequality, institutional incentives, and localized electoral competition. Within this integrated framework, inequality shapes the structural environment that renders material inducements meaningful, particularly where formal social protection is uneven or unreliable (Abdulai, 2021 ; Swamy, 2016 ). Institutional incentives simultaneously structure the supply side of exchange. In candidate-centered and highly competitive systems, targeted transfers may represent an efficient mobilization strategy (Ravanilla & Hicken, 2023 ; Singh, 2018 ). Once individuals are incorporated into broker-mediated exchange networks, exposure activates reciprocal norms embedded in ongoing relational ties (Hicken et al., 2022 ; Lawson & Greene, 2014 ). Compliance thus emerges less as an outcome of coercive domination and more as a pragmatic response within competitive political exchange. Conceptualizing clientelism as informal social protection shifts analytical attention from moral evaluation to distributive function. Inducements may serve as temporary, discretionary substitutes for formal welfare in contexts of uneven state capacity, linking electoral exchange to broader patterns of inequality and redistribution. Importantly, this framework yields two empirically distinguishable possibilities. If structural vulnerability predominates, demographic indicators—such as education, age, or residence—should independently predict clientelistic compliance even after exposure is taken into account. If situational activation predominates, exposure should emerge as the primary predictor of compliance, attenuating or eliminating the independent effects of demographic characteristics. Distinguishing between these mechanisms is central to understanding whether clientelism reinforces socioeconomic stratification in political participation or operates as a cross-cutting mobilization strategy embedded within competitive democracies. The empirical analysis that follows evaluates which of these pathways more accurately characterizes clientelistic behavior in contemporary Thailand. Multidimensional Clientelism: Reciprocity versus Coercion The framework further differentiates clientelism along two analytically distinct dimensions: behavioral reciprocity and coercive justification. Although often treated interchangeably in discussions of vote buying, these dimensions reflect different mechanisms of compliance and carry distinct normative implications. Behavioral reciprocity refers to the willingness of voters to support a benefactor—by voting for a candidate or adjusting turnout—in response to receiving an inducement. Under a reciprocity model, compliance reflects relational obligation embedded within exchange networks rather than fear of sanction. It is activated through incorporation into ongoing social relationships. Coercive justification, by contrast, refers to expectations of punishment, retaliation, withdrawal of benefits, or social sanction if reciprocity is not fulfilled. This dimension implies asymmetry of power and the possibility of domination. Under a coercion-based model, compliance is sustained by anticipated negative consequences rather than relational obligation. The empirical distinction is consequential. If clientelism operates primarily through coercion, coercive perceptions should form a coherent latent construct and significantly predict compliance behavior. Conversely, if clientelism operates primarily through exchange-based reciprocity, behavioral compliance may exhibit internal coherence even when coercive expectations are weak, fragmented, or statistically insignificant. Distinguishing between these dimensions therefore allows assessment of whether clientelism reflects structural subordination or pragmatic exchange embedded within competitive institutional environments. The preceding theoretical discussion identifies three mechanisms linking inequality and clientelist behavior: structural vulnerability, situational activation, and coercive domination. These mechanisms generate competing empirical expectations: H1 (Structural Vulnerability Hypothesis): Socioeconomic disadvantage is positively associated with clientelist compliance, net of exposure to inducements. H2 (Situational Activation Hypothesis): Exposure to vote buying is positively associated with clientelist compliance and attenuates the independent effects of socioeconomic indicators. H3 (Coercive Domination Hypothesis): Coercive expectations constitute a coherent construct and positively predict clientelist compliance. H4 (Reciprocity Dominance Hypothesis): Behavioral reciprocity forms a coherent construct and predicts compliance independently of coercive expectations. Together, these hypotheses enable the empirical analysis to adjudicate among competing pathways. Specifically, the analysis evaluates whether inequality shapes compliance primarily through demographic vulnerability, whether exposure activates reciprocal responsiveness across social strata, or whether compliance is structured by fear-based domination. The results thus speak directly to broader debates about how inequality intersects with electoral exchange in contemporary democracies. Methods Research Design and Data Collection This study utilizes data from a nationally administered post-election survey conducted following Thailand’s 2023 general election (14 May 2023). Fieldwork began in mid-June 2023 and was completed over a three- to four-week period. The timing of data collection is analytically advantageous. Respondents were interviewed shortly after the campaign period, reducing long-term recall decay while minimizing immediate campaign-period pressures that may heighten social desirability bias. The survey instrument was designed to capture electoral behavior, political attitudes, exposure to campaign inducements, and perceptions of political competition. In addition to standard demographic indicators, the questionnaire included a dedicated clientelism module. This module measured (1) exposure to material inducements, (2) the type and timing of inducements, (3) behavioral responses to inducements, and (4) perceived consequences of non-reciprocation. The post-election design enables the assessment of self-reported exposure and compliance behavior within a recent electoral context, while avoiding real-time campaign effects that may distort reporting. Sample The dataset comprises 1,200 respondents, with the final analytic sample varying modestly across models due to item-level non-response. Participants were drawn from multiple regions of Thailand, ensuring demographic heterogeneity across age cohorts, levels of educational attainment, gender, and urban versus non-urban residence. This variation enables systematic examination of how exposure to inducements and reported compliance are distributed across key demographic strata. Because the study centers on post-election exposure and behavioral responsiveness rather than turnout prediction, the analytic sample includes respondents who provided valid responses to both the clientelism module and the principal demographic indicators used in the analysis. Measures Clientelism is operationalized as a multidimensional construct comprising exposure to inducements, behavioral reciprocity, and coercive justification. This structure reflects the theoretical distinction between incorporation into exchange networks, reciprocal responsiveness, and fear-based compliance. Exposure to vote buying is measured using a binary indicator capturing whether respondents reported receiving or being offered money, food, or household goods during the 2023 campaign period (1 = exposed; 0 = not exposed). Responses coded as “Don’t know” or missing were excluded from the analysis. Behavioral compliance is operationalized as a composite index reflecting reciprocal responsiveness to inducement. The index is calculated as the mean of two binary items: (1) whether the respondent would vote for the candidate after receiving an inducement, and (2) whether the respondent would abstain from voting if no inducement were received. Each item is coded 1 for compliance tendency and 0 for rejection. The two items demonstrate extremely high internal consistency (Cronbach’s α ≈ 0.99), indicating a coherent underlying behavioral dimension. The resulting index ranges from 0 to 1, with higher values indicating stronger compliance tendencies. For robustness, supplementary logistic regression models employ a binary outcome based solely on willingness to vote for the candidate following inducement. Coercive justification captures perceived consequences of failing to reciprocate. The survey includes multiple items assessing expectations of punishment, such as loss of benefits, verbal threats, physical intimidation, or social sanction. Each item is coded 1 if the respondent expects a consequence and 0 otherwise. Reliability analysis indicates that these items do not form a coherent latent scale, suggesting that coercive perceptions are fragmented rather than structured as a unified construct. Accordingly, they are treated analytically as a distinct conceptual dimension rather than aggregated into a single index. The primary explanatory variable in the multivariate analyses is exposure to vote buying. Demographic controls include educational attainment (modeled as ordinal categories), age group (ordinal categories), gender (binary), and urban versus non-urban residence. These variables allow direct evaluation of the structural vulnerability hypothesis by testing whether socioeconomic characteristics independently predict compliance or moderate the effect of exposure. Analytical Strategy The empirical analysis proceeds in four sequential stages designed to adjudicate between structural vulnerability and situational activation mechanisms. First, descriptive statistics establish the prevalence and distribution of exposure to inducements, as well as patterns of behavioral compliance, providing a demographic baseline for subsequent multivariate analysis. Second, ordinary least squares (OLS) regression models estimate the association between exposure to vote buying and the Behavioral Compliance Index. The baseline model includes exposure only, while the fully specified model incorporates demographic controls—education, age, gender, and residence—to assess whether socioeconomic characteristics independently predict compliance net of exposure. Third, interaction models examine whether education moderates the relationship between exposure and compliance. This specification directly tests whether susceptibility to inducement varies across levels of socioeconomic status, as predicted by the structural vulnerability thesis. Fourth, logistic regression models estimate the probability of reciprocal voting behavior using a binary compliance measure. This alternative specification provides a robustness check against potential distributional constraints associated with the bounded 0–1 index and facilitates interpretation through odds ratios and marginal effects. All analyses employ listwise deletion of missing observations. Standard errors are computed using conventional OLS and logit estimators. Limitations Several limitations warrant consideration. First, the measures of exposure and compliance rely on self-reported responses and may be subject to recall bias or social desirability effects. Although the post-election timing reduces long-term recall decay, respondents may underreport exposure to inducements or overstate autonomous voting behavior. As with most survey-based research on sensitive political behavior, some degree of reporting distortion cannot be ruled out. Second, the cross-sectional design limits causal inference. While the analysis identifies strong associations between exposure and compliance, the data do not permit definitive claims regarding temporal sequencing or causal direction. Unobserved factors—such as prior partisan attachment, political efficacy, or community-level network embeddedness—may influence both the likelihood of exposure and behavioral responsiveness. Third, the substantial association between exposure and behavioral compliance raises the possibility of conceptual proximity between these measures. Although the survey instrument clearly distinguishes between reported inducement receipt and stated behavioral response, attitudinal consistency or post hoc rationalization may inflate observed relationships. Robustness analyses employing alternative operationalizations mitigate, but cannot fully eliminate, this concern. Finally, the study relies on individual-level survey data and does not incorporate community-level, broker-level, or constituency-level variation. Such contextual factors may shape patterns of inducement distribution, monitoring capacity, and relational enforcement in ways not captured by the present design. These limitations counsel caution in interpreting the results as definitive evidence of causal mechanism. Nonetheless, the post-election timing, multidimensional operationalization of clientelism, and demographic heterogeneity of the sample provide a rigorous empirical foundation for assessing how structural inequality and situational exposure interact in shaping reported compliance behavior. Findings Prevalence and Structure of Clientelistic Exposure Table 1 presents the prevalence of self-reported exposure to vote buying during Thailand’s 2023 general election. Among respondents providing valid responses, 29.19 percent reported that they had received or been offered money, food, or household goods during the campaign period, while 70.81 percent indicated no such exposure. These figures demonstrate that clientelistic inducements remain a visible component of electoral competition, yet they do not characterize the experience of the electorate as a whole. A substantial minority encountered material transfers, whereas a clear majority reported no direct contact with inducements. This distribution suggests that exposure to vote buying is patterned rather than universal. Although the survey design does not allow direct identification of targeting strategies or constituency-level campaign tactics, the fact that fewer than one-third of respondents report exposure indicates that inducement-based interactions are not evenly distributed across voters. Vote buying appears to coexist with non-clientelistic campaign practices rather than structuring the entirety of electoral engagement. Table 1 Exposure to Vote Buying During the 2023 General Election Category N Percent Not exposed 769 70.81 Exposed 317 29.19 Total 1,086 100.00 Note : Percentages are calculated among respondents providing valid responses. From a demographic perspective, the partial reach of exposure is analytically consequential. If incorporation into exchange networks is uneven, behavioral responsiveness cannot be inferred solely from socioeconomic characteristics. Instead, compliance must be examined in relation to whether individuals were directly exposed to inducements. Establishing this descriptive baseline clarifies two empirical premises for the multivariate analysis that follows: first, exposure to vote buying remains a non-trivial feature of electoral practice; second, its distribution is selective rather than comprehensive. The subsequent sections assess how this patterned exposure relates to behavioral compliance and whether demographic characteristics independently predict responsiveness once exposure is taken into account. Behavioral Compliance and the Structure of Reciprocity Table 2 reports descriptive statistics for the Behavioral Compliance Index. The index ranges from 0 to 1 and captures respondents’ stated willingness to reciprocate electoral inducements, either by voting for a candidate who provided material benefits or by adjusting turnout behavior in response to inducement. Across valid observations (N = 1,174), the mean compliance score is 0.395 (SD = 0.487), indicating that approximately 40 percent of respondents express some degree of reciprocal responsiveness. The distribution of the index exhibits substantial heterogeneity. The observed minimum (0.000) and maximum (1.000) values indicate polarization between respondents who reject reciprocal behavior entirely and those who report full compliance. The relatively large standard deviation further suggests meaningful variation in behavioral orientation toward inducement across the sample. Table 2 Behavioral Compliance Index: Descriptive Statistics Statistic Value N 1174 Mean 0.395 Std. Dev. 0.487 Min 0.000 Max 1.000 Note : The Behavioral Compliance Index ranges from 0 to 1, with higher values indicating stronger reciprocal responsiveness. Notably, the average level of compliance exceeds the prevalence of reported exposure (29.19 percent). This pattern indicates that expressed reciprocal responsiveness is not limited exclusively to those who directly experienced inducements. Instead, it may reflect generalized normative orientations toward exchange or hypothetical responsiveness under inducement conditions. This distinction underscores the analytical separation between exposure and compliance: exposure captures reported campaign experience, whereas compliance measures stated behavioral disposition toward inducement-based exchange. At the descriptive stage, these statistics do not reveal whether compliance tendencies are concentrated within specific demographic strata. The mean value suggests that reciprocal responsiveness is neither marginal nor universal, reinforcing the need for multivariate analysis to determine whether compliance is structured by socioeconomic vulnerability or more strongly associated with situational exposure. The next subsection turns to regression models to adjudicate between these competing mechanisms. Regression Evidence: Exposure versus Structural Vulnerability Table 3 presents the OLS regression estimates predicting the Behavioral Compliance Index. Exposure to vote buying emerges as a large and statistically significant predictor of reciprocal responsiveness (b = 0.886, p < .01). Given that the dependent variable ranges from 0 to 1, this coefficient implies a substantial shift in reported compliance among respondents who indicate exposure to inducements. Substantively, the magnitude suggests that exposure is associated with a near-complete movement across the bounded compliance scale. The model explains 69.1 percent of the variance in behavioral compliance (R² = 0.691), indicating that exposure accounts for the overwhelming share of observed variation in reported responsiveness. Table 3 OLS Regression Predicting Behavioral Compliance Variable Coefficient Std. Error Intercept 0.022 (0.048) Exposure 0.886*** (0.019) Education 0.005 (0.005) Age 0.005 (0.007) Gender 0.019 (0.016) Urban Residence 0.006 (0.010) Observations 1,016 R² 0.691 Notes : Standard errors in parentheses. * p < .10, ** p < .05, *** p < .01. In survey-based behavioral research, such explanatory power is uncommon and underscores the empirical centrality of direct inducement contact in shaping stated reciprocal behavior. By contrast, demographic indicators commonly associated with structural vulnerability—education (b = 0.005), age (b = 0.005), gender (b = 0.019), and urban residence (b = 0.006)—do not attain statistical significance in the fully specified model. Once exposure is incorporated, structural socioeconomic characteristics do not independently predict compliance. From the standpoint of the structural vulnerability thesis, this pattern is theoretically consequential. If demographic disadvantage were the primary mechanism driving clientelistic responsiveness, education or residence should retain independent explanatory power net of exposure. Instead, the results suggest that behavioral responsiveness is closely aligned with direct incorporation into exchange networks rather than with demographic position alone. To assess whether socioeconomic status conditions the magnitude of the exposure effect, the next subsection evaluates interaction terms between exposure and education. Interaction Model: Testing Educational Moderation Table 4 evaluates whether educational attainment moderates the association between exposure to vote buying and behavioral compliance. The interaction term between exposure and education is substantively negligible and statistically insignificant (b = − 0.001, n.s.), and overall model fit remains unchanged (R² = 0.691). The magnitude and explanatory power of the model are virtually identical to the baseline specification. These estimates indicate that the association between exposure and compliance does not vary systematically across educational strata. Higher levels of education do not attenuate the behavioral relationship between inducement exposure and reciprocal responsiveness. Substantively, the exposure effect appears stable across the educational distribution. Table 4 OLS Regression with Exposure × Education Interaction Variable Coefficient Std. Error Intercept 0.022 (0.048) Exposure 0.890*** (0.048) Education 0.005 (0.005) Exposure × Education –0.001 (0.010) Age 0.005 (0.007) Gender 0.019 (0.016) Urban Residence 0.006 (0.010) Observations 1,016 R² 0.691 Notes: Standard errors in parentheses. * p < .10, ** p < .05, *** p < .01. From the perspective of the structural vulnerability thesis, this finding is analytically consequential. If compliance were primarily concentrated among less-educated respondents, one would expect a significant negative interaction term indicating that the exposure effect weakens as education increases. The absence of such moderation suggests that once individuals are directly incorporated into inducement-based exchange, educational attainment does not meaningfully condition behavioral responsiveness. This pattern aligns more closely with a situational activation mechanism than with a stratified vulnerability account. Exposure appears to exert a comparable behavioral influence across socioeconomic groups rather than operating disproportionately among the least educated. Logistic Model and Predicted Probabilities To assess robustness under an alternative outcome specification, Table 5 presents logistic regression estimates predicting reciprocal voting behavior. The dependent variable captures willingness to vote for a candidate following receipt of an inducement. Consistent with the linear models, exposure to vote buying is strongly and positively associated with reciprocal voting. The estimated odds ratio for exposure is 587.45 (p < .01), indicating an exceptionally strong association between reported exposure and stated reciprocal voting behavior. The magnitude reflects substantial separation between exposed and non-exposed respondents in the binary outcome distribution. Importantly, the direction and statistical significance of the exposure effect closely mirror the OLS results. In contrast, demographic indicators—including education (OR = 1.078), age (OR = 1.085), gender (OR = 1.411), and urban residence (OR = 1.078)—do not attain statistical significance in the fully specified model. As in the linear models, socioeconomic characteristics do not independently predict reciprocal voting once exposure is incorporated. Table 5 Logistic Regression Predicting Reciprocal Voting (Odds Ratios) Variable Odds Ratio Std. Error Intercept 0.032*** (0.696) Exposure 587.449*** (0.520) Education 1.078 (0.067) Age 1.085 (0.092) Gender 1.411 (0.228) Urban Residence 1.078 (0.137) Observations 1,016 Pseudo R² 0.589 Notes: Standard errors in parentheses. * p < .10, ** p < .05, *** p < .01. To facilitate substantive interpretation, Table 6 reports average marginal effects (AMEs) derived from the logistic model. Exposure increases the predicted probability of reciprocal voting by approximately 0.451 (p < .01), holding other variables constant. In substantive terms, exposure is associated with a 45-percentage-point increase in the likelihood of reporting reciprocal voting behavior. By contrast, marginal effects for education, age, gender, and residence are small and statistically indistinguishable from zero. Table 6 Average Marginal Effects (Logistic Model) Variable AME Std. Error Exposure 0.451*** (0.040) Education 0.005 (0.005) Age 0.006 (0.007) Gender 0.024 (0.016) Urban Residence 0.005 (0.010) Notes: AME = Average Marginal Effects. Standard errors in parentheses. * p < .10, ** p < .05, *** p < .01. Figure 1 visualizes predicted probabilities of reciprocal voting by exposure status. The figure illustrates a pronounced divergence between exposed and non-exposed respondents, whereas variation across demographic categories remains comparatively limited. Taken together, the interaction, logistic, and marginal effects analyses reinforce a consistent empirical pattern. Exposure to inducement emerges as the dominant correlate of reciprocal behavior across model specifications, while demographic indicators associated with structural vulnerability do not independently predict responsiveness once exposure is taken into account. Moreover, educational attainment does not condition the exposure effect. Collectively, these findings provide stronger support for a situational activation mechanism than for a strictly vulnerability-based explanation. Behavioral compliance appears closely aligned with incorporation into exchange relationships rather than systematically stratified by socioeconomic position. Synthesis of Findings Across descriptive, linear, and nonlinear specifications, the empirical results converge on a consistent pattern. First, exposure to vote buying is substantial but not universal, reported by approximately one-third of respondents. Clientelistic practices therefore remain visible within Thailand’s electoral landscape, yet they do not structure the entirety of electoral interaction. The selective reach of inducements underscores the importance of distinguishing structural context from direct incorporation into exchange networks. Second, behavioral compliance displays meaningful variation across the sample, but this variation is overwhelmingly aligned with exposure. Both OLS and logistic models identify exposure as the dominant correlate of reciprocal voting behavior. The magnitude of the association remains large and stable across model specifications, alternative outcome measures, and interaction tests. Substantively, exposure accounts for the vast majority of observed variation in reported responsiveness. Third, demographic indicators commonly associated with structural vulnerability—education, age, gender, and urban residence—do not independently predict compliance once exposure is taken into account. Nor does education moderate the exposure effect. Behavioral responsiveness appears substantively similar across socioeconomic strata when individuals report inducement contact. These findings do not suggest that inequality is irrelevant; rather, they indicate that demographic position alone does not translate mechanically into differential compliance at the individual level. Fourth, coercive perceptions do not form a coherent latent dimension and do not emerge as systematic predictors of compliance. Reciprocal responsiveness appears more closely associated with exchange incorporation than with fear-based expectations of sanction. This distinction is central to adjudicating between domination-based and reciprocity-based interpretations of clientelism. In aggregate, the findings offer limited support for a purely structural vulnerability account in which demographic disadvantage independently predicts compliance. Rather, the empirical pattern is more consistent with a situational activation framework, whereby exposure to inducement emerges as the central correlate of reciprocal responsiveness. Nevertheless, the cross-sectional design and reliance on self-reported measures constrain causal inference. The results therefore document strong associations rather than temporal sequencing. These findings raise broader theoretical questions about how inequality interacts with exchange networks in shaping democratic participation. The following section situates these results within debates on informal social protection, distributive politics, and the demographic foundations of electoral behavior in unequal democracies. Discussion The foregoing evidence invites a reconsideration of how inequality, demographic structure, and electoral clientelism intersect in contemporary democracies. Rather than reinforcing a stratified model in which socioeconomic disadvantage independently produces compliant behavior, the empirical pattern points toward the centrality of exchange incorporation in shaping responsiveness. This suggests that the relationship between inequality and clientelism operates through more contingent and relational mechanisms than vulnerability-based accounts typically assume. Understanding this distinction is crucial for population and social studies scholarship, where demographic stratification is often treated as the primary channel through which inequality structures political behavior. Rethinking the Vulnerability Thesis Inequality-based accounts of clientelism rest on a clear expectation: socioeconomic disadvantage should independently heighten susceptibility to inducement. Under this logic, lower education, rural residence, or age-related precarity are not merely correlates of exposure but structural determinants of compliance. If vulnerability operates as theorized, demographic position should retain explanatory power even after direct contact with inducements is taken into account. The empirical pattern complicates this expectation. Once exposure is incorporated into the model, demographic indicators cease to predict compliance, and education does not condition the exposure effect. Reciprocal responsiveness appears broadly similar across socioeconomic strata among those who report inducement contact. This suggests that demographic vulnerability does not function as an independent driver of behavioral compliance in the manner vulnerability-based theories anticipate. A more plausible interpretation is that structural inequality shapes the opportunity structure of exposure rather than the decision to reciprocate. Inequality may sustain the demand for short-term material transfers and make inducements socially meaningful, yet it does not mechanically translate into differentiated behavioral responsiveness at the individual level. In demographic terms, compliance is not concentrated exclusively among the least educated or most structurally marginalized segments of the electorate. This reinterpretation refines the inequality–clientelism relationship. Rather than operating primarily through stratified compliance, inequality may serve as a contextual condition that renders electoral exchange viable without deterministically structuring who reciprocates. The mechanism appears less one of demographic subordination and more one of situational activation within unequal institutional environments. Situational Activation and Relational Incorporation If demographic vulnerability does not independently structure compliance, the explanatory weight shifts to the mechanism of situational activation. The consistently strong association between exposure and reciprocal behavior suggests that clientelism operates less through stratified predisposition and more through incorporation into exchange relationships. In this framework, exposure functions as the activating event: once individuals are directly engaged in inducement-based interactions, reciprocal norms become behaviorally salient irrespective of educational attainment or place of residence. This does not imply that structural inequality is irrelevant. Rather, it suggests that inequality shapes the conditions under which exchange becomes meaningful, while exposure structures the behavioral response. The mechanism, therefore, is sequential rather than purely stratified: structural inequality sustains the value of short-term material transfers, but compliance is triggered through relational incorporation into exchange networks. Clientelism thus operates as a mobilization strategy embedded in competitive electoral environments, where inducements activate norms of reciprocity across social strata rather than targeting only the most disadvantaged. The absence of a coherent coercive dimension further clarifies this mechanism. If compliance were driven primarily by fear-based domination, expectations of sanction would cluster systematically and independently predict behavior. Instead, coercive perceptions remain fragmented, while behavioral reciprocity exhibits strong internal coherence. This asymmetry reinforces the interpretation of clientelism as negotiated exchange rather than uniform coercion. Compliance appears to reflect relational obligation and pragmatic adaptation rather than systematic intimidation. Taken together with the previous subsection, the evidence points toward a dual-mechanism model: inequality conditions the environment of exchange, but exposure activates reciprocal responsiveness. This reconceptualization shifts the analytical focus from who is vulnerable to how and when voters become incorporated into exchange networks. Clientelism as Informal Social Protection The preceding analysis suggests that the vulnerability and situational activation mechanisms are not mutually exclusive but sequentially related within a broader political economy of redistribution. Reconceptualizing clientelism as informal social protection provides a framework for integrating these dynamics. In unequal democracies where formal welfare provision remains uneven, discretionary, or selectively administered, targeted inducements may function as short-term redistributive transfers delivered through electoral channels. Inequality thus shapes the demand-side conditions that render material benefits socially meaningful. At the same time, the behavioral activation of reciprocity appears to depend less on demographic stratification than on incorporation into exchange networks. Exposure, rather than structural identity alone, structures compliance. From a population perspective, this implies that inducement-based exchange need not be confined to the poorest citizens. In competitive, candidate-centered electoral environments, the distribution of inducements may extend across social strata, activating reciprocal norms beyond narrowly defined vulnerable groups. Compliance, in this context, reflects pragmatic adaptation within relational networks rather than passive subordination rooted in fixed demographic disadvantage. This interpretation complicates conventional narratives that treat vote buying solely as evidence of democratic pathology or as a straightforward product of poverty. While clientelism may undermine programmatic accountability, it also reflects how citizens navigate environments characterized by uneven social protection and competitive political mobilization. Inequality provides the structural terrain upon which exchange becomes viable, but exposure determines who becomes behaviorally incorporated into that exchange. Seen in this light, clientelism operates at the intersection of redistribution, relational obligation, and institutional competition. Implications for Inequality and Democratic Participation These findings carry broader implications for research on inequality and democratic participation within population and social studies. If clientelistic compliance is not independently stratified by education, age, or residence once exposure is taken into account, then socioeconomic development alone may be insufficient to weaken exchange-based electoral practices. Expanding educational attainment or reducing income inequality may alter the structural landscape of vulnerability, but such changes do not automatically disrupt relational exchange dynamics embedded in competitive electoral systems. Structural transformation may reduce the demand-side conditions that render inducements meaningful, yet the activation of compliance appears contingent on network incorporation rather than demographic position alone. At the same time, the selective nature of exposure underscores that clientelism does not uniformly structure political participation. Inducement-based exchange operates within specific networks, constituencies, and localities rather than encompassing the electorate as a whole. This pattern shifts analytical attention toward institutional and organizational incentives—campaign finance arrangements, candidate-centered competition, broker-mediated mobilization, and decentralized electoral strategies—that shape the distribution and strategic deployment of inducements. Demographic inequality may provide the terrain, but institutional design influences how and where exchange practices take root. More broadly, the evidence reinforces the importance of analytically separating structural conditions from situational mechanisms in the study of political behavior. Inequality creates the environment in which informal redistribution acquires social meaning, yet behavioral responsiveness appears closely tied to direct incorporation into exchange relationships. Clientelism in unequal democracies, therefore, cannot be reduced to poverty alone; it reflects a dynamic interaction between demographic structure, institutional incentives, and relational mobilization strategies. The concluding section considers how this reconceptualization informs debates about democratic development and the prospects for programmatic competition in societies marked by persistent inequality and evolving demographic change. Conclusion This study examined whether clientelistic compliance in unequal democracies is primarily structured by socioeconomic vulnerability or activated through situational incorporation into exchange networks. Drawing on post-election survey data from Thailand’s 2023 general election, the analysis distinguished structural demographic indicators from direct exposure to inducements and further differentiated reciprocity-based compliance from coercive justification. The evidence indicates that while exposure to vote buying remains a visible feature of electoral competition, compliance behavior is not independently stratified by education, age, gender, or residence once exposure is taken into account. Nor do coercive expectations emerge as a coherent driver of responsiveness. Instead, the observed patterns are more consistent with a situational activation mechanism in which behavioral reciprocity is linked to incorporation into exchange relationships rather than predetermined by demographic disadvantage. These findings refine how inequality should be understood in relation to clientelism. Structural inequality appears to shape the broader environment in which informal exchange becomes meaningful, particularly where formal social protection remains uneven or selectively administered. Yet inequality does not translate mechanically into differentiated compliance across demographic groups. In concise terms, inequality structures the demand side of clientelism, but exposure structures the behavioral response. Clientelism, in this context, operates less as a fixed hierarchy of domination and more as a relational mobilization strategy embedded within competitive electoral systems. For population and social studies scholarship, this distinction is consequential. If compliance were strongly concentrated among socioeconomically disadvantaged groups, clientelism would represent a clear mechanism reinforcing stratified political voice and democratic inequality. Instead, the evidence suggests that exchange-based responsiveness may extend across social strata once individuals are directly incorporated into inducement-based networks. This implies that socioeconomic development alone—while normatively desirable—may not be sufficient to weaken clientelist practices if institutional incentives and broker-mediated mobilization persist. More broadly, the study underscores the analytical importance of separating structural conditions from situational mechanisms in the study of democratic participation. Inequality shapes the context within which informal redistribution acquires social meaning, but incorporation into exchange networks appears central in structuring reported behavioral responsiveness. Understanding how these dynamics interact is essential for designing reforms aimed at strengthening programmatic competition and reducing reliance on discretionary electoral transfers. Future research incorporating longitudinal designs, community-level variation, and comparative cases across unequal democracies would help clarify whether the Thai case reflects broader regional dynamics or context-specific institutional configurations. In unequal societies undergoing demographic transformation, the study of clientelism requires attention not only to who is vulnerable, but to how and when citizens become embedded in exchange relationships. By disentangling structural vulnerability from situational activation, this study advances a more precise and demographically grounded understanding of how inequality intersects with electoral behavior in contemporary democracies. Declarations Funding Declaration This research received no specific external funding. The study was conducted as part of the author’s research activities at the Faculty of Political Science, Chulalongkorn University, Thailand. Ethics Statement The public opinion survey used in this study was reviewed and approved by the Institutional Review Board (IRB) of King Prajadhipok’s Institute, Thailand. The research was conducted in accordance with established ethical guidelines for social science research involving human participants. The survey data were collected anonymously, and no personally identifiable information was recorded. Participation in the survey was voluntary, and respondents were informed about the purpose of the study prior to participation. Consent to Participate Informed consent was obtained from all participants prior to their participation in the survey. Consent to Publish Not applicable. The manuscript does not contain any individual person’s identifiable data in any form. Author Contribution S.T. conceptualized the study, developed the theoretical framework, designed the research methodology, conducted the data analysis, interpreted the results, and wrote and revised the manuscript. Data Availability The datasets generated during and/or analysed during the current study are not publicly available because the survey dataset is maintained by King Prajadhipok’s Institute and contains confidential respondent information, but are available from the corresponding author on reasonable request. References Abdulai AG. Competitive clientelism, donors and the politics of social protection uptake in Ghana. Crit Soc Policy. 2021;41(2):270–93. ttps://doi.org/10.1177/0261018320945605. Aspinall E. Money politics: Patronage and clientelism in Southeast Asia. Routledge handbook of southeast Asian democratization. Routledge; 2015. pp. 299–313. Bardhan P, Mookherjee D. (2020). Clientelistic politics and economic development. The handbook of economic development and institutions , 84–102. Birch S, Cockshott P, Renaud K. Putting electronic voting under the microscope. Political Q. 2014;85(2):187–94. ttps://doi.org/10.1111/1467-923X.12071. Boossabong P. Patronage in Thailand: trust and reciprocity in policy networks. Handbook of Politicization and Political Patronage. Edward Elgar Publishing; 2025. pp. 404–20. Bowie KA. Vote buying and village outrage in an election in northern Thailand: Recent legal reforms in historical context. J Asian Stud. 2008;67(2):469–511. ttps://doi.org/10.1017/S0021911808000673. Callahan WA. The discourse of vote buying and political reform in Thailand. Pac Affairs. 2005;78(1):95–113. ttps://doi.org/10.5509/200578195. Callahan WA, McCargo D. Vote-buying in Thailand’s Northeast: The July 1995 general election. Asian Surv. 1996;36(4):376–92. ttps://doi.org/10.2307/2645405. Canare TA, Mendoza RU, Lopez MA. An empirical analysis of vote buying among the poor: Evidence from the Philippines. Asian Politics Policy. 2018;10(4):659–79. ttps://doi.org/10.1177/0967828X17753420. Carlin RE, Moseley MW. When clientelism backfires: Vote buying, democratic attitudes, and electoral retaliation in Latin America. Polit Res Q. 2022;75(3):766–81. ttps://doi.org/10.1177/10659129211020126. Chattharakul A. Thai electoral campaigning, vote-canvassing networks, and hybrid voting. Asian Politics Policy. 2010;2(1):67–89. ttps://doi.org/10.1177/186810341002900404. Corstange D. Clientelism in competitive and uncompetitive elections. Comp Polit Stud. 2018;51(1):76–104. ttps://doi.org/10.1177/0010414017695332. Cruz C. Social networks and the targeting of vote buying. Comp Polit Stud. 2019;52(3):382–411. ttps://doi.org/10.1177/0010414018784062. Della Porta D, Vannucci A. Corrupt exchanges: Actors, resources, and mechanisms of political corruption. Routledge; 2017. Elliott C. Clientelism and the democratic deficit. Stud Indian Politics. 2016;4(1):22–36. ttps://doi.org/10.1177/2321023016634915. Fergusson L, Molina CA, Robinson JA. The weak state trap. Economica. 2022;89(354):293–331. ttps://doi.org/10.1111/ecca.12399. Hicken A. Clientelism. Annu Rev Polit Sci. 2011;14:289–310. ttps://doi.org/10.1146/annurev.polisci.031908.220508. Hicken A, Aspinall E, Weiss ML, Muhtadi B. Buying brokers: Electoral handouts beyond clientelism in a weak-party state. World Polit. 2022;74(1):77–120. ttps://doi.org/10.1017/S0043887121000216. Hidalgo FD, Nichter S. Voter buying: Shaping the electorate through clientelism. Am J Polit Sci. 2016;60(2):436–55. ttps://doi.org/10.1111/ajps.12214. Jensen PS, Justesen MK. Poverty and vote buying: Survey-based evidence from Africa. Electoral Stud. 2014;33:220–32. ttps://doi.org/10.1016/j.electstud.2013.07.020. Justesen MK, Manzetti L. Poverty, partisanship, and vote buying in Latin America. Lat Am Politics Soc. 2023;65(3):1–19. ttps://doi.org/10.1017/lap.2022.66. Kitschelt H, Wilkinson SI, editors. Patrons, clients and policies: Patterns of democratic accountability and political competition. Cambridge University Press; 2007. Lawson C, Greene KF. Making clientelism work: How norms of reciprocity increase voter compliance. Comp Politics. 2014;47(1):61–85. ttps://doi.org/10.5129/001041514813623173. Mares I, Young LE. Conditionality & coercion: Electoral clientelism in eastern Europe. Oxford University Press; 2019. Owen N. Conceptualizing vote buying as a process: An empirical study in Thai provinces. Asian Politics Policy. 2013;5(2):247–65. ttps://doi.org/10.1111/aspp.12028. Ravanilla N, Hicken A. Poverty, social networks, and clientelism. World Dev. 2023;162:106128. ttps://doi.org/10.1016/j.worlddev.2022.106128. Ravanilla N, Haim D, Hicken A. Brokers, social networks, reciprocity, and clientelism. Am J Polit Sci. 2022;66(4):795–812. ttps://doi.org/10.1111/ajps.12604. Remmer KL. The political economy of patronage: Expenditure patterns in the Argentine provinces, 1983–2003. J Politics. 2007;69(2):363–77. ttps://doi.org/10.1111/j.1468-2508.2007.00537.x. Scott JC. Corruption, machine politics, and political change. Am Polit Sci Rev. 1969;63(4):1142–58. ttps://doi.org/10.2307/1955076. Singh SP. Compulsory voting and parties’ vote-seeking strategies. Am J Polit Sci. 2018;62(2):387–403. ttps://doi.org/10.1111/ajps.12386. Stokes SC, Dunning T, Nazareno M. Brokers, voters, and clientelism. Cambridge University Press; 2013. Swamy AR. Can social protection weaken clientelism? Considering conditional cash transfers as political reform in the Philippines. J Curr Southeast Asian Affairs. 2016;35(1):59–90. ttps://doi.org/10.1177/186810341603500103. Taylor JS. Autonomy, vote buying, and constraining options. J Appl Philos. 2017;34(5):711–23. ttps://doi.org/10.1111/japp.12205. Wantchekon L. Clientelism and voting behavior: Evidence from a field experiment in Benin. World Polit. 2003;55(3):399–422. ttps://doi.org/10.1353/wp.2003.0018. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 11 May, 2026 Reviewers agreed at journal 04 May, 2026 Reviews received at journal 03 May, 2026 Reviewers agreed at journal 03 May, 2026 Reviewers agreed at journal 03 May, 2026 Reviewers invited by journal 22 Apr, 2026 Editor invited by journal 21 Apr, 2026 Editor assigned by journal 20 Mar, 2026 Submission checks completed at journal 18 Mar, 2026 First submitted to journal 13 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9049351","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":633670940,"identity":"f03f5b99-a785-4450-a2b6-1a2b87d36f38","order_by":0,"name":"Stithorn Thananithichot","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYFACxgZmCIP5AJBhARPEo4ONsbEZwmBLAGqRgIgexKuFgRGqhceAOC3y85vbHxdU3MtjkO/5+LmwTUKegf3wA+aPO3BrMTgGdNiMM8XFDGy8m6VntkkYNvCkGTAcPINHC8gvvG0JiQ1svNuYedskgB7PATqsDY/D2uBaeJ6BtNg38L/Br4XhGEILG0hLYoMEAVsMjiU2zuY5k5DYxpZmLM1zTiK5TeKZwYGz+BzWfPzBZ56KhMR+5sMPP/OU2dj28yc/fFCJz2EwwIbMOECEhlEwCkbBKBgFeAAAtmlJuEiNvNsAAAAASUVORK5CYII=","orcid":"","institution":"Chulalongkorn University","correspondingAuthor":true,"prefix":"","firstName":"Stithorn","middleName":"","lastName":"Thananithichot","suffix":""}],"badges":[],"createdAt":"2026-03-06 10:24:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9049351/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9049351/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108414227,"identity":"8f378a54-6895-4cd4-a1a7-37168b815574","added_by":"auto","created_at":"2026-05-04 11:05:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":14182,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted Probability of Reciprocal Voting by Exposure Status.\u003c/p\u003e\n\u003cp\u003ePredicted probabilities derived from logistic regression estimates. Error bars represent 95% confidence intervals. Exposure to inducement is associated with a substantially higher probability of reported reciprocal voting behavior.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9049351/v1/4097e84069ab257436f04e29.png"},{"id":109067491,"identity":"044028a0-9d69-4865-b605-dedf9234cc2f","added_by":"auto","created_at":"2026-05-12 09:54:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":355620,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9049351/v1/1ae4c753-36f0-4c76-8492-9e430b2ef4b3.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Clientelism as Informal Social Protection in Unequal Democracies: Evidence from Thailand’s 2023 Election","fulltext":[{"header":"Introduction","content":"\u003cp\u003eElectoral clientelism remains a persistent feature of democratic politics in many unequal societies. Across Asia, Latin America, and parts of Eastern Europe, political actors continue to distribute cash, food, and material goods in exchange for electoral support (Aspinall, \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e; Carlin \u0026amp; Moseley, \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e; Mares \u0026amp; Young, \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). A substantial body of scholarship interprets these practices as products of socioeconomic inequality, linking vote buying to poverty, limited education, and structural dependency (Della Porta \u0026amp; Vannucci, \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e; Jensen \u0026amp; Justesen, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Scott, \u003cspan class=\"CitationRef\"\u003e1969\u003c/span\u003e). From this perspective, clientelism flourishes where vulnerability constrains voter autonomy and reinforces elite domination (Elliott, \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e; Kitschelt \u0026amp; Wilkinson, \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e). Poorer and less educated citizens are therefore commonly assumed to be more susceptible to inducements and more likely to reciprocate politically (Justesen \u0026amp; Manzetti, \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e). Although inequality undoubtedly shapes political opportunity structures, much of the existing literature implicitly assumes a relatively direct relationship between socioeconomic disadvantage and compliance with vote buying (Corstange, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e; Hidalgo \u0026amp; Nichter, \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e). This assumption leaves an important theoretical and empirical question unresolved. Far less attention has been devoted to distinguishing whether clientelistic behavior is primarily driven by demographic vulnerability itself or activated through situational exposure to exchange networks that may cut across social strata. Put differently, it remains unclear whether clientelism is concentrated among disadvantaged voters or operates as a broader mechanism of electoral mobilization embedded within unequal yet competitive democracies. Understanding whether clientelism is demographically stratified or situationally activated therefore has implications not only for Thailand but for unequal democracies more broadly, where electoral exchange often coexists with expanding yet uneven systems of social welfare.\u003c/p\u003e \u003cp\u003eThis distinction is especially consequential for research on population dynamics and social stratification. If clientelistic compliance is strongly patterned by education, age, or place of residence, it may reinforce existing inequalities in political voice and representation. If, however, compliance is more closely associated with exposure to inducements rather than demographic characteristics, clientelism may operate less as a fixed hierarchy of dependency and more as a relational mechanism activated within specific electoral contexts. Clarifying this distinction is therefore essential for understanding how inequality intersects with democratic participation in societies undergoing demographic and institutional change. This study addresses this question by examining clientelistic behavior in Thailand’s 2023 general election. Thailand provides a theoretically relevant setting for such analysis. It is a middle-income democracy characterized by persistent regional inequality, rapid educational expansion, urban–rural migration, and long-standing public debates over vote buying and populism. These transformations raise a central question: in a society experiencing rising educational attainment and structural change, does clientelistic compliance remain concentrated among socioeconomically disadvantaged groups?\u003c/p\u003e \u003cp\u003eThis article advances the argument that clientelism in unequal democracies may function as a form of informal social protection. In contexts where formal welfare provision remains uneven or insufficient, material inducements can operate as short-term redistributive transfers delivered through political networks. Under this perspective, inequality shapes the structural environment that renders inducements meaningful, but compliance may be activated through situational incorporation into exchange relationships rather than predetermined by demographic vulnerability alone. To evaluate this argument, the study analyzes nationally administered post-election survey data collected following Thailand’s 2023 general election. It constructs multidimensional measures of clientelistic exposure, behavioral reciprocity, and perceived coercive justification. By distinguishing structural demographic indicators—education, age, gender, and residence—from direct exposure to inducements, the analysis directly tests whether socioeconomic disadvantage independently predicts compliance once exchange incorporation is taken into account. Rather than presuming that poverty or limited education mechanically translate into political dependency, this study systematically examines the demographic distribution and behavioral mechanisms of clientelistic exchange.\u003c/p\u003e \u003cp\u003eIn doing so, the study contributes to population and social studies scholarship in three ways. First, it refines the relationship between inequality and political participation by distinguishing structural vulnerability from situational activation. Second, it provides empirical evidence on how demographic characteristics intersect with exchange-based electoral practices in a middle-income democracy undergoing social transformation. Third, it contributes to debates on redistribution and social protection by highlighting how informal exchange mechanisms may coexist with—and in some contexts partially substitute for—formal welfare institutions. Distinguishing whether clientelism is demographically stratified or situationally activated carries important implications for policy and institutional reform. If compliance is primarily driven by structural vulnerability, long-term socioeconomic development may gradually weaken clientelist practices. If exposure-based mechanisms dominate, however, reforms targeting electoral incentives, campaign organization, and broker-mediated mobilization may be equally critical. By situating clientelism within broader processes of demographic change and social stratification, this study offers a more nuanced account of electoral behavior in unequal societies.\u003c/p\u003e \u003cp\u003eThe remainder of the article proceeds as follows. The next section reviews scholarship on inequality, clientelism, and electoral exchange and develops a theoretical framework distinguishing structural vulnerability from situational activation and reciprocity-based mechanisms. The subsequent section outlines the data, survey design, measurement strategy, and analytical approach. The findings section presents descriptive patterns of exposure and compliance, followed by multivariate analyses testing the competing hypotheses. The final section discusses the broader implications for inequality, social protection, and democratic participation and concludes with reflections on policy and future research.\u003c/p\u003e\n\u003ch3\u003eTheoretical Framework\u003c/h3\u003e\n\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStructural Vulnerability and the Inequality Thesis\u003c/h2\u003e \u003cp\u003eA dominant strand of scholarship conceptualizes clientelism as a political response to structural inequality (Kitschelt \u0026amp; Wilkinson, \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e; Remmer, \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e; Stokes, Dunning, \u0026amp; Nazareno, \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e; Wantchekon, \u003cspan class=\"CitationRef\"\u003e2003\u003c/span\u003e). In settings where income distribution is highly skewed and formal social protection systems remain fragmented, discretionary, or exclusionary, targeted electoral transfers may substitute for universal welfare provision (Schaffer, 2004; Speck \u0026amp; Abramo, 2001). Under such conditions, electoral exchange becomes an adaptive mechanism through which citizens secure short-term material benefits in contexts of limited state reliability (Callahan, \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e; Bowie, \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). From this perspective, vote buying is not merely electoral corruption but a distributive strategy embedded within unequal political economies (Jensen \u0026amp; Justesen, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e). Within this framework, lower-income and less-educated citizens are presumed to be more susceptible to clientelistic inducements because immediate material gains carry greater marginal utility than distant or uncertain policy commitments. Clientelism thus reflects structural vulnerability rather than individual irrationality. Empirical studies in Southeast Asia—particularly from the Philippines—demonstrate that monetary and food-based inducements are frequently targeted toward economically precarious households (Canare, Mendoza, \u0026amp; Lopez, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e), reinforcing the view that poverty shapes both the supply and demand of electoral exchange.\u003c/p\u003e \u003cp\u003eThe vulnerability thesis rests on two core assumptions. First, socioeconomic disadvantage reduces bargaining power. Individuals with fewer resources may be less able to reject inducements or demand programmatic commitments. Where public services are distributed arbitrarily, patronage networks may substitute for universal entitlements (Callahan, \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e). Second, material inducements possess greater relative value for economically precarious citizens. The short-term utility of cash or goods may outweigh the uncertain benefits of long-term policy platforms. Evidence from poor urban communities in the Philippines illustrates strong poverty-targeting patterns in monetary vote buying (Canare, Mendoza, \u0026amp; Lopez, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). If this framework holds empirically, structural indicators—such as education, age, income proxies, employment precarity, or rural residence—should significantly predict clientelistic compliance even after controlling for exposure to inducements. In other words, demographic vulnerability should exert an independent effect on behavioral responsiveness, beyond mere contact with inducement-based campaigns.\u003c/p\u003e \u003cp\u003eHowever, comparative evidence complicates this deterministic expectation. Studies in Southeast Asia suggest that while exposure may be concentrated among poorer voters, compliance remains conditional rather than automatic (Aspinall, \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e; Owen, \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e). Voters frequently distinguish between accepting material benefits and relinquishing electoral autonomy, indicating that exchange may operate through negotiated reciprocity rather than strict economic coercion (Chattharakul, \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e; Cruz, \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). Moreover, the marginal-utility logic need not be confined to the poorest citizens. In candidate-centered systems characterized by intense personal vote competition, inducements may be distributed broadly rather than narrowly targeted to disadvantaged groups (Hicken, \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e). Where turnout uncertainty shapes party strategy, clientelist mobilization may function as a generalized electoral tactic rather than a poverty-specific instrument (Singh, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). Inequality, therefore, may create the structural conditions under which clientelism emerges, but it does not necessarily determine individual-level compliance. As Bowie (\u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e) demonstrates in the Thai case, escalations in vote buying have coincided with institutional reform and rising electoral stakes rather than with static rural poverty. The central empirical question, then, is whether socioeconomic disadvantage independently predicts compliance once exposure is taken into account—or whether behavioral responsiveness is activated through situational incorporation into exchange networks.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eReciprocity, Exchange, and Situational Activation\u003c/h3\u003e\n\u003cp\u003eAn alternative perspective conceptualizes clientelism not primarily as a function of structural vulnerability, but as a form of reciprocal exchange embedded in everyday social norms and relational networks (Fergusson, Molina, \u0026amp; Robinson, \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ravanilla, Haim, \u0026amp; Hicken, \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). Rather than emerging solely from economic desperation or coercive domination, electoral inducements may activate culturally embedded expectations of return. Under this view, compliance is less determined by demographic status and more by situational incorporation into exchange relationships. In Thailand, broker-mediated networks (หัวคะแนน) operate through community ties, kinship structures, and localized trust relations (Callahan \u0026amp; McCargo, \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e; Chattharakul, \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e; Owen, \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e). Inducements are rarely isolated transactions; they are delivered through intermediaries who maintain ongoing social relationships. Monitoring is often social rather than technological, functioning through inference, visibility, and reputation rather than direct ballot surveillance (Owen, \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e). Such relational embeddedness lowers enforcement costs and reduces the need for overt coercion.\u003c/p\u003e \u003cp\u003eWithin this framework, compliance is activated through exposure. Once a voter receives a benefit—whether cash, goods, or assistance—a norm of reciprocity may be engaged. The exchange may be interpreted not strictly as bribery but as recognition, obligation, or pragmatic mutual support. Studies of hybrid voting behavior indicate that voters frequently distinguish between accepting inducements and relinquishing electoral autonomy (Birch, Cockshott, \u0026amp; Renaud, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Taylor, \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e). Acceptance does not mechanically imply obedience; however, incorporation into exchange networks may nonetheless generate reciprocal responsiveness. This perspective shifts analytical focus from structural identity to situational activation. If reciprocity mechanisms dominate, exposure to inducements should strongly predict compliance across demographic strata. Education, age, and residence should exert weaker independent effects once exposure is accounted for. Behavioral responsiveness, in other words, should reflect incorporation into relational exchange rather than preexisting socioeconomic vulnerability.\u003c/p\u003e \u003cp\u003eImportantly, this framework also clarifies why coercive perceptions and behavioral compliance may not align. Empirical research in Southeast Asia shows that voters may comply without expressing strong fear of retaliation (Canare, Mendoza, \u0026amp; Lopez, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). Historical analyses of Thailand similarly suggest that intensification of vote buying has coincided with heightened electoral competition rather than systematic intimidation (Bowie, \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e; Hicken, \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e). Exchange thus appears embedded within competitive institutional environments rather than enforced through uniform domination. Under a situational activation model, clientelism operates as a socially embedded mobilization mechanism. Exposure—not vulnerability alone—triggers behavioral responsiveness. The empirical implication parallels but contrasts with the vulnerability thesis: if incorporation into exchange networks drives compliance, exposure should emerge as the dominant predictor of reciprocal behavior, attenuating the independent effects of demographic disadvantage.\u003c/p\u003e\n\u003ch3\u003eClientelism as Informal Social Protection\u003c/h3\u003e\n\u003cp\u003eIntegrating structural inequality and reciprocity-based exchange perspectives allows clientelism to be reconceptualized as a form of informal social protection. In unequal democracies where formal redistribution mechanisms remain uneven, discretionary, or selectively administered, political actors may provide targeted material benefits that temporarily mitigate economic insecurity (Bardhan \u0026amp; Mookherjee, \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e). These inducements resemble short-term redistributive transfers delivered through electoral channels rather than institutionalized welfare systems. In Thailand, reliance on patronage networks has historically emerged in contexts where public services are perceived as arbitrarily distributed (Boossabong, \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e; Callahan, \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e). Moreover, escalations in vote buying have coincided with institutional reform and rising electoral stakes, which increased the distributive value of political office (Bowie, \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). Clientelism therefore operates at the intersection of inequality, institutional incentives, and localized electoral competition. Within this integrated framework, inequality shapes the structural environment that renders material inducements meaningful, particularly where formal social protection is uneven or unreliable (Abdulai, \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e; Swamy, \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e). Institutional incentives simultaneously structure the supply side of exchange. In candidate-centered and highly competitive systems, targeted transfers may represent an efficient mobilization strategy (Ravanilla \u0026amp; Hicken, \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e; Singh, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). Once individuals are incorporated into broker-mediated exchange networks, exposure activates reciprocal norms embedded in ongoing relational ties (Hicken et al., \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e; Lawson \u0026amp; Greene, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e). Compliance thus emerges less as an outcome of coercive domination and more as a pragmatic response within competitive political exchange.\u003c/p\u003e \u003cp\u003eConceptualizing clientelism as informal social protection shifts analytical attention from moral evaluation to distributive function. Inducements may serve as temporary, discretionary substitutes for formal welfare in contexts of uneven state capacity, linking electoral exchange to broader patterns of inequality and redistribution. Importantly, this framework yields two empirically distinguishable possibilities. If structural vulnerability predominates, demographic indicators—such as education, age, or residence—should independently predict clientelistic compliance even after exposure is taken into account. If situational activation predominates, exposure should emerge as the primary predictor of compliance, attenuating or eliminating the independent effects of demographic characteristics. Distinguishing between these mechanisms is central to understanding whether clientelism reinforces socioeconomic stratification in political participation or operates as a cross-cutting mobilization strategy embedded within competitive democracies. The empirical analysis that follows evaluates which of these pathways more accurately characterizes clientelistic behavior in contemporary Thailand.\u003c/p\u003e\n\u003ch3\u003eMultidimensional Clientelism: Reciprocity versus Coercion\u003c/h3\u003e\n\u003cp\u003eThe framework further differentiates clientelism along two analytically distinct dimensions: behavioral reciprocity and coercive justification. Although often treated interchangeably in discussions of vote buying, these dimensions reflect different mechanisms of compliance and carry distinct normative implications. Behavioral reciprocity refers to the willingness of voters to support a benefactor—by voting for a candidate or adjusting turnout—in response to receiving an inducement. Under a reciprocity model, compliance reflects relational obligation embedded within exchange networks rather than fear of sanction. It is activated through incorporation into ongoing social relationships. Coercive justification, by contrast, refers to expectations of punishment, retaliation, withdrawal of benefits, or social sanction if reciprocity is not fulfilled. This dimension implies asymmetry of power and the possibility of domination. Under a coercion-based model, compliance is sustained by anticipated negative consequences rather than relational obligation.\u003c/p\u003e \u003cp\u003eThe empirical distinction is consequential. If clientelism operates primarily through coercion, coercive perceptions should form a coherent latent construct and significantly predict compliance behavior. Conversely, if clientelism operates primarily through exchange-based reciprocity, behavioral compliance may exhibit internal coherence even when coercive expectations are weak, fragmented, or statistically insignificant. Distinguishing between these dimensions therefore allows assessment of whether clientelism reflects structural subordination or pragmatic exchange embedded within competitive institutional environments.\u003c/p\u003e \u003cp\u003eThe preceding theoretical discussion identifies three mechanisms linking inequality and clientelist behavior: structural vulnerability, situational activation, and coercive domination. These mechanisms generate competing empirical expectations:\u003c/p\u003e\n\u003ch3\u003eH1 (Structural Vulnerability Hypothesis):\u003c/h3\u003e\n\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eSocioeconomic disadvantage is positively associated with clientelist compliance, net of exposure to inducements.\u003c/p\u003e \u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eH2 (Situational Activation Hypothesis):\u003c/h2\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eExposure to vote buying is positively associated with clientelist compliance and attenuates the independent effects of socioeconomic indicators.\u003c/p\u003e \u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eH3 (Coercive Domination Hypothesis):\u003c/h3\u003e\n\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eCoercive expectations constitute a coherent construct and positively predict clientelist compliance.\u003c/p\u003e \u003c/div\u003e \u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eH4 (Reciprocity Dominance Hypothesis):\u003c/h3\u003e\n\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eBehavioral reciprocity forms a coherent construct and predicts compliance independently of coercive expectations.\u003c/p\u003e \u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eTogether, these hypotheses enable the empirical analysis to adjudicate among competing pathways. Specifically, the analysis evaluates whether inequality shapes compliance primarily through demographic vulnerability, whether exposure activates reciprocal responsiveness across social strata, or whether compliance is structured by fear-based domination. The results thus speak directly to broader debates about how inequality intersects with electoral exchange in contemporary democracies.\u003c/p\u003e "},{"header":"Methods","content":"\u003ch2\u003eResearch Design and Data Collection\u003c/h2\u003e\u003cp\u003eThis study utilizes data from a nationally administered post-election survey conducted following Thailand’s 2023 general election (14 May 2023). Fieldwork began in mid-June 2023 and was completed over a three- to four-week period. The timing of data collection is analytically advantageous. Respondents were interviewed shortly after the campaign period, reducing long-term recall decay while minimizing immediate campaign-period pressures that may heighten social desirability bias. The survey instrument was designed to capture electoral behavior, political attitudes, exposure to campaign inducements, and perceptions of political competition. In addition to standard demographic indicators, the questionnaire included a dedicated clientelism module. This module measured (1) exposure to material inducements, (2) the type and timing of inducements, (3) behavioral responses to inducements, and (4) perceived consequences of non-reciprocation. The post-election design enables the assessment of self-reported exposure and compliance behavior within a recent electoral context, while avoiding real-time campaign effects that may distort reporting.\u003c/p\u003e\u003ch2\u003eSample\u003c/h2\u003e\u003cp\u003eThe dataset comprises 1,200 respondents, with the final analytic sample varying modestly across models due to item-level non-response. Participants were drawn from multiple regions of Thailand, ensuring demographic heterogeneity across age cohorts, levels of educational attainment, gender, and urban versus non-urban residence. This variation enables systematic examination of how exposure to inducements and reported compliance are distributed across key demographic strata. Because the study centers on post-election exposure and behavioral responsiveness rather than turnout prediction, the analytic sample includes respondents who provided valid responses to both the clientelism module and the principal demographic indicators used in the analysis.\u003c/p\u003e\u003ch2\u003eMeasures\u003c/h2\u003e\u003cp\u003eClientelism is operationalized as a multidimensional construct comprising exposure to inducements, behavioral reciprocity, and coercive justification. This structure reflects the theoretical distinction between incorporation into exchange networks, reciprocal responsiveness, and fear-based compliance.\u003c/p\u003e\u003cp\u003eExposure to vote buying is measured using a binary indicator capturing whether respondents reported receiving or being offered money, food, or household goods during the 2023 campaign period (1 = exposed; 0 = not exposed). Responses coded as “Don’t know” or missing were excluded from the analysis.\u003c/p\u003e\u003cp\u003eBehavioral compliance is operationalized as a composite index reflecting reciprocal responsiveness to inducement. The index is calculated as the mean of two binary items: (1) whether the respondent would vote for the candidate after receiving an inducement, and (2) whether the respondent would abstain from voting if no inducement were received. Each item is coded 1 for compliance tendency and 0 for rejection. The two items demonstrate extremely high internal consistency (Cronbach’s α ≈ 0.99), indicating a coherent underlying behavioral dimension. The resulting index ranges from 0 to 1, with higher values indicating stronger compliance tendencies. For robustness, supplementary logistic regression models employ a binary outcome based solely on willingness to vote for the candidate following inducement.\u003c/p\u003e\u003cp\u003eCoercive justification captures perceived consequences of failing to reciprocate. The survey includes multiple items assessing expectations of punishment, such as loss of benefits, verbal threats, physical intimidation, or social sanction. Each item is coded 1 if the respondent expects a consequence and 0 otherwise. Reliability analysis indicates that these items do not form a coherent latent scale, suggesting that coercive perceptions are fragmented rather than structured as a unified construct. Accordingly, they are treated analytically as a distinct conceptual dimension rather than aggregated into a single index.\u003c/p\u003e\u003cp\u003eThe primary explanatory variable in the multivariate analyses is exposure to vote buying. Demographic controls include educational attainment (modeled as ordinal categories), age group (ordinal categories), gender (binary), and urban versus non-urban residence. These variables allow direct evaluation of the structural vulnerability hypothesis by testing whether socioeconomic characteristics independently predict compliance or moderate the effect of exposure.\u003c/p\u003e\u003ch2\u003eAnalytical Strategy\u003c/h2\u003e\u003cp\u003eThe empirical analysis proceeds in four sequential stages designed to adjudicate between structural vulnerability and situational activation mechanisms. First, descriptive statistics establish the prevalence and distribution of exposure to inducements, as well as patterns of behavioral compliance, providing a demographic baseline for subsequent multivariate analysis. Second, ordinary least squares (OLS) regression models estimate the association between exposure to vote buying and the Behavioral Compliance Index. The baseline model includes exposure only, while the fully specified model incorporates demographic controls—education, age, gender, and residence—to assess whether socioeconomic characteristics independently predict compliance net of exposure.\u003c/p\u003e\u003cp\u003eThird, interaction models examine whether education moderates the relationship between exposure and compliance. This specification directly tests whether susceptibility to inducement varies across levels of socioeconomic status, as predicted by the structural vulnerability thesis. Fourth, logistic regression models estimate the probability of reciprocal voting behavior using a binary compliance measure. This alternative specification provides a robustness check against potential distributional constraints associated with the bounded 0–1 index and facilitates interpretation through odds ratios and marginal effects. All analyses employ listwise deletion of missing observations. Standard errors are computed using conventional OLS and logit estimators.\u003c/p\u003e\u003ch2\u003eLimitations\u003c/h2\u003e\u003cp\u003eSeveral limitations warrant consideration. First, the measures of exposure and compliance rely on self-reported responses and may be subject to recall bias or social desirability effects. Although the post-election timing reduces long-term recall decay, respondents may underreport exposure to inducements or overstate autonomous voting behavior. As with most survey-based research on sensitive political behavior, some degree of reporting distortion cannot be ruled out. Second, the cross-sectional design limits causal inference. While the analysis identifies strong associations between exposure and compliance, the data do not permit definitive claims regarding temporal sequencing or causal direction. Unobserved factors—such as prior partisan attachment, political efficacy, or community-level network embeddedness—may influence both the likelihood of exposure and behavioral responsiveness. Third, the substantial association between exposure and behavioral compliance raises the possibility of conceptual proximity between these measures. Although the survey instrument clearly distinguishes between reported inducement receipt and stated behavioral response, attitudinal consistency or post hoc rationalization may inflate observed relationships. Robustness analyses employing alternative operationalizations mitigate, but cannot fully eliminate, this concern. Finally, the study relies on individual-level survey data and does not incorporate community-level, broker-level, or constituency-level variation. Such contextual factors may shape patterns of inducement distribution, monitoring capacity, and relational enforcement in ways not captured by the present design. These limitations counsel caution in interpreting the results as definitive evidence of causal mechanism. Nonetheless, the post-election timing, multidimensional operationalization of clientelism, and demographic heterogeneity of the sample provide a rigorous empirical foundation for assessing how structural inequality and situational exposure interact in shaping reported compliance behavior.\u003c/p\u003e"},{"header":"Findings","content":"\u003ch2\u003ePrevalence and Structure of Clientelistic Exposure\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e presents the prevalence of self-reported exposure to vote buying during Thailand’s 2023 general election. Among respondents providing valid responses, 29.19 percent reported that they had received or been offered money, food, or household goods during the campaign period, while 70.81 percent indicated no such exposure. These figures demonstrate that clientelistic inducements remain a visible component of electoral competition, yet they do not characterize the experience of the electorate as a whole. A substantial minority encountered material transfers, whereas a clear majority reported no direct contact with inducements. This distribution suggests that exposure to vote buying is patterned rather than universal. Although the survey design does not allow direct identification of targeting strategies or constituency-level campaign tactics, the fact that fewer than one-third of respondents report exposure indicates that inducement-based interactions are not evenly distributed across voters. Vote buying appears to coexist with non-clientelistic campaign practices rather than structuring the entirety of electoral engagement.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab1\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExposure to Vote Buying During the 2023 General Election\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eCategory\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003ePercent\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNot exposed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e769\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e70.81\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eExposed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e317\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e29.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e\u003cb\u003eTotal\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e1,086\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e100.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003eNote\u003c/em\u003e: Percentages are calculated among respondents providing valid responses.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFrom a demographic perspective, the partial reach of exposure is analytically consequential. If incorporation into exchange networks is uneven, behavioral responsiveness cannot be inferred solely from socioeconomic characteristics. Instead, compliance must be examined in relation to whether individuals were directly exposed to inducements. Establishing this descriptive baseline clarifies two empirical premises for the multivariate analysis that follows: first, exposure to vote buying remains a non-trivial feature of electoral practice; second, its distribution is selective rather than comprehensive. The subsequent sections assess how this patterned exposure relates to behavioral compliance and whether demographic characteristics independently predict responsiveness once exposure is taken into account.\u003c/p\u003e\u003ch2\u003eBehavioral Compliance and the Structure of Reciprocity\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e reports descriptive statistics for the Behavioral Compliance Index. The index ranges from 0 to 1 and captures respondents’ stated willingness to reciprocate electoral inducements, either by voting for a candidate who provided material benefits or by adjusting turnout behavior in response to inducement. Across valid observations (N = 1,174), the mean compliance score is 0.395 (SD = 0.487), indicating that approximately 40 percent of respondents express some degree of reciprocal responsiveness. The distribution of the index exhibits substantial heterogeneity. The observed minimum (0.000) and maximum (1.000) values indicate polarization between respondents who reject reciprocal behavior entirely and those who report full compliance. The relatively large standard deviation further suggests meaningful variation in behavioral orientation toward inducement across the sample.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab2\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBehavioral Compliance Index: Descriptive Statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eStatistic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1174\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.395\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eStd. Dev.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.487\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMax\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003e\u003cem\u003eNote\u003c/em\u003e: The Behavioral Compliance Index ranges from 0 to 1, with higher values indicating stronger reciprocal responsiveness.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNotably, the average level of compliance exceeds the prevalence of reported exposure (29.19 percent). This pattern indicates that expressed reciprocal responsiveness is not limited exclusively to those who directly experienced inducements. Instead, it may reflect generalized normative orientations toward exchange or hypothetical responsiveness under inducement conditions. This distinction underscores the analytical separation between exposure and compliance: exposure captures reported campaign experience, whereas compliance measures stated behavioral disposition toward inducement-based exchange. At the descriptive stage, these statistics do not reveal whether compliance tendencies are concentrated within specific demographic strata. The mean value suggests that reciprocal responsiveness is neither marginal nor universal, reinforcing the need for multivariate analysis to determine whether compliance is structured by socioeconomic vulnerability or more strongly associated with situational exposure. The next subsection turns to regression models to adjudicate between these competing mechanisms.\u003c/p\u003e\u003ch2\u003eRegression Evidence: Exposure versus Structural Vulnerability\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e presents the OLS regression estimates predicting the Behavioral Compliance Index. Exposure to vote buying emerges as a large and statistically significant predictor of reciprocal responsiveness (b = 0.886, p \u0026lt; .01). Given that the dependent variable ranges from 0 to 1, this coefficient implies a substantial shift in reported compliance among respondents who indicate exposure to inducements. Substantively, the magnitude suggests that exposure is associated with a near-complete movement across the bounded compliance scale. The model explains 69.1 percent of the variance in behavioral compliance (R² = 0.691), indicating that exposure accounts for the overwhelming share of observed variation in reported responsiveness.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab3\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOLS Regression Predicting Behavioral Compliance\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eStd. Error\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e(0.048)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eExposure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.886***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e(0.019)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEducation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e(0.016)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eUrban Residence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1,016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eR²\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.691\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003eNotes\u003c/em\u003e: Standard errors in parentheses. * p \u0026lt; .10, ** p \u0026lt; .05, *** p \u0026lt; .01.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn survey-based behavioral research, such explanatory power is uncommon and underscores the empirical centrality of direct inducement contact in shaping stated reciprocal behavior. By contrast, demographic indicators commonly associated with structural vulnerability—education (b = 0.005), age (b = 0.005), gender (b = 0.019), and urban residence (b = 0.006)—do not attain statistical significance in the fully specified model. Once exposure is incorporated, structural socioeconomic characteristics do not independently predict compliance. From the standpoint of the structural vulnerability thesis, this pattern is theoretically consequential. If demographic disadvantage were the primary mechanism driving clientelistic responsiveness, education or residence should retain independent explanatory power net of exposure. Instead, the results suggest that behavioral responsiveness is closely aligned with direct incorporation into exchange networks rather than with demographic position alone. To assess whether socioeconomic status conditions the magnitude of the exposure effect, the next subsection evaluates interaction terms between exposure and education.\u003c/p\u003e\u003ch2\u003eInteraction Model: Testing Educational Moderation\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e evaluates whether educational attainment moderates the association between exposure to vote buying and behavioral compliance. The interaction term between exposure and education is substantively negligible and statistically insignificant (b = − 0.001, n.s.), and overall model fit remains unchanged (R² = 0.691). The magnitude and explanatory power of the model are virtually identical to the baseline specification. These estimates indicate that the association between exposure and compliance does not vary systematically across educational strata. Higher levels of education do not attenuate the behavioral relationship between inducement exposure and reciprocal responsiveness. Substantively, the exposure effect appears stable across the educational distribution.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab4\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOLS Regression with Exposure × Education Interaction\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eStd. Error\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.048)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eExposure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.890***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.048)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEducation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eExposure × Education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e–0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.016)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eUrban Residence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e1,016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eR²\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.691\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003eNotes: Standard errors in parentheses. * p \u0026lt; .10, ** p \u0026lt; .05, *** p \u0026lt; .01.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFrom the perspective of the structural vulnerability thesis, this finding is analytically consequential. If compliance were primarily concentrated among less-educated respondents, one would expect a significant negative interaction term indicating that the exposure effect weakens as education increases. The absence of such moderation suggests that once individuals are directly incorporated into inducement-based exchange, educational attainment does not meaningfully condition behavioral responsiveness. This pattern aligns more closely with a situational activation mechanism than with a stratified vulnerability account. Exposure appears to exert a comparable behavioral influence across socioeconomic groups rather than operating disproportionately among the least educated.\u003c/p\u003e\u003ch2\u003eLogistic Model and Predicted Probabilities\u003c/h2\u003e\u003cp\u003eTo assess robustness under an alternative outcome specification, Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e presents logistic regression estimates predicting reciprocal voting behavior. The dependent variable captures willingness to vote for a candidate following receipt of an inducement. Consistent with the linear models, exposure to vote buying is strongly and positively associated with reciprocal voting. The estimated odds ratio for exposure is 587.45 (p \u0026lt; .01), indicating an exceptionally strong association between reported exposure and stated reciprocal voting behavior. The magnitude reflects substantial separation between exposed and non-exposed respondents in the binary outcome distribution. Importantly, the direction and statistical significance of the exposure effect closely mirror the OLS results. In contrast, demographic indicators—including education (OR = 1.078), age (OR = 1.085), gender (OR = 1.411), and urban residence (OR = 1.078)—do not attain statistical significance in the fully specified model. As in the linear models, socioeconomic characteristics do not independently predict reciprocal voting once exposure is incorporated.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab5\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eLogistic Regression Predicting Reciprocal Voting (Odds Ratios)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eOdds Ratio\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eStd. Error\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.032***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.696)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eExposure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e587.449***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.520)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEducation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.067)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1.085\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.092)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1.411\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.228)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eUrban Residence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.137)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eObservations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e1,016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePseudo R²\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.589\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003eNotes: Standard errors in parentheses. * p \u0026lt; .10, ** p \u0026lt; .05, *** p \u0026lt; .01.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo facilitate substantive interpretation, Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e reports average marginal effects (AMEs) derived from the logistic model. Exposure increases the predicted probability of reciprocal voting by approximately 0.451 (p \u0026lt; .01), holding other variables constant. In substantive terms, exposure is associated with a 45-percentage-point increase in the likelihood of reporting reciprocal voting behavior. By contrast, marginal effects for education, age, gender, and residence are small and statistically indistinguishable from zero.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab6\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAverage Marginal Effects (Logistic Model)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eAME\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eStd. Error\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eExposure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.451***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.040)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEducation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.005)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.007)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.016)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eUrban Residence\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e(0.010)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003eNotes: AME = Average Marginal Effects. Standard errors in parentheses. * p \u0026lt; .10, ** p \u0026lt; .05, *** p \u0026lt; .01.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e visualizes predicted probabilities of reciprocal voting by exposure status. The figure illustrates a pronounced divergence between exposed and non-exposed respondents, whereas variation across demographic categories remains comparatively limited.\u003c/p\u003e\u003cp\u003eTaken together, the interaction, logistic, and marginal effects analyses reinforce a consistent empirical pattern. Exposure to inducement emerges as the dominant correlate of reciprocal behavior across model specifications, while demographic indicators associated with structural vulnerability do not independently predict responsiveness once exposure is taken into account. Moreover, educational attainment does not condition the exposure effect. Collectively, these findings provide stronger support for a situational activation mechanism than for a strictly vulnerability-based explanation. Behavioral compliance appears closely aligned with incorporation into exchange relationships rather than systematically stratified by socioeconomic position.\u003c/p\u003e\u003ch2\u003eSynthesis of Findings\u003c/h2\u003e\u003cp\u003eAcross descriptive, linear, and nonlinear specifications, the empirical results converge on a consistent pattern. First, exposure to vote buying is substantial but not universal, reported by approximately one-third of respondents. Clientelistic practices therefore remain visible within Thailand’s electoral landscape, yet they do not structure the entirety of electoral interaction. The selective reach of inducements underscores the importance of distinguishing structural context from direct incorporation into exchange networks. Second, behavioral compliance displays meaningful variation across the sample, but this variation is overwhelmingly aligned with exposure. Both OLS and logistic models identify exposure as the dominant correlate of reciprocal voting behavior. The magnitude of the association remains large and stable across model specifications, alternative outcome measures, and interaction tests. Substantively, exposure accounts for the vast majority of observed variation in reported responsiveness.\u003c/p\u003e\u003cp\u003eThird, demographic indicators commonly associated with structural vulnerability—education, age, gender, and urban residence—do not independently predict compliance once exposure is taken into account. Nor does education moderate the exposure effect. Behavioral responsiveness appears substantively similar across socioeconomic strata when individuals report inducement contact. These findings do not suggest that inequality is irrelevant; rather, they indicate that demographic position alone does not translate mechanically into differential compliance at the individual level. Fourth, coercive perceptions do not form a coherent latent dimension and do not emerge as systematic predictors of compliance. Reciprocal responsiveness appears more closely associated with exchange incorporation than with fear-based expectations of sanction. This distinction is central to adjudicating between domination-based and reciprocity-based interpretations of clientelism.\u003c/p\u003e\u003cp\u003eIn aggregate, the findings offer limited support for a purely structural vulnerability account in which demographic disadvantage independently predicts compliance. Rather, the empirical pattern is more consistent with a situational activation framework, whereby exposure to inducement emerges as the central correlate of reciprocal responsiveness. Nevertheless, the cross-sectional design and reliance on self-reported measures constrain causal inference. The results therefore document strong associations rather than temporal sequencing. These findings raise broader theoretical questions about how inequality interacts with exchange networks in shaping democratic participation. The following section situates these results within debates on informal social protection, distributive politics, and the demographic foundations of electoral behavior in unequal democracies.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe foregoing evidence invites a reconsideration of how inequality, demographic structure, and electoral clientelism intersect in contemporary democracies. Rather than reinforcing a stratified model in which socioeconomic disadvantage independently produces compliant behavior, the empirical pattern points toward the centrality of exchange incorporation in shaping responsiveness. This suggests that the relationship between inequality and clientelism operates through more contingent and relational mechanisms than vulnerability-based accounts typically assume. Understanding this distinction is crucial for population and social studies scholarship, where demographic stratification is often treated as the primary channel through which inequality structures political behavior.\u003c/p\u003e\u003ch2\u003eRethinking the Vulnerability Thesis\u003c/h2\u003e\u003cp\u003eInequality-based accounts of clientelism rest on a clear expectation: socioeconomic disadvantage should independently heighten susceptibility to inducement. Under this logic, lower education, rural residence, or age-related precarity are not merely correlates of exposure but structural determinants of compliance. If vulnerability operates as theorized, demographic position should retain explanatory power even after direct contact with inducements is taken into account. The empirical pattern complicates this expectation. Once exposure is incorporated into the model, demographic indicators cease to predict compliance, and education does not condition the exposure effect. Reciprocal responsiveness appears broadly similar across socioeconomic strata among those who report inducement contact. This suggests that demographic vulnerability does not function as an independent driver of behavioral compliance in the manner vulnerability-based theories anticipate.\u003c/p\u003e\u003cp\u003eA more plausible interpretation is that structural inequality shapes the opportunity structure of exposure rather than the decision to reciprocate. Inequality may sustain the demand for short-term material transfers and make inducements socially meaningful, yet it does not mechanically translate into differentiated behavioral responsiveness at the individual level. In demographic terms, compliance is not concentrated exclusively among the least educated or most structurally marginalized segments of the electorate. This reinterpretation refines the inequality–clientelism relationship. Rather than operating primarily through stratified compliance, inequality may serve as a contextual condition that renders electoral exchange viable without deterministically structuring who reciprocates. The mechanism appears less one of demographic subordination and more one of situational activation within unequal institutional environments.\u003c/p\u003e\u003ch2\u003eSituational Activation and Relational Incorporation\u003c/h2\u003e\u003cp\u003eIf demographic vulnerability does not independently structure compliance, the explanatory weight shifts to the mechanism of situational activation. The consistently strong association between exposure and reciprocal behavior suggests that clientelism operates less through stratified predisposition and more through incorporation into exchange relationships. In this framework, exposure functions as the activating event: once individuals are directly engaged in inducement-based interactions, reciprocal norms become behaviorally salient irrespective of educational attainment or place of residence. This does not imply that structural inequality is irrelevant. Rather, it suggests that inequality shapes the conditions under which exchange becomes meaningful, while exposure structures the behavioral response. The mechanism, therefore, is sequential rather than purely stratified: structural inequality sustains the value of short-term material transfers, but compliance is triggered through relational incorporation into exchange networks. Clientelism thus operates as a mobilization strategy embedded in competitive electoral environments, where inducements activate norms of reciprocity across social strata rather than targeting only the most disadvantaged.\u003c/p\u003e\u003cp\u003eThe absence of a coherent coercive dimension further clarifies this mechanism. If compliance were driven primarily by fear-based domination, expectations of sanction would cluster systematically and independently predict behavior. Instead, coercive perceptions remain fragmented, while behavioral reciprocity exhibits strong internal coherence. This asymmetry reinforces the interpretation of clientelism as negotiated exchange rather than uniform coercion. Compliance appears to reflect relational obligation and pragmatic adaptation rather than systematic intimidation. Taken together with the previous subsection, the evidence points toward a dual-mechanism model: inequality conditions the environment of exchange, but exposure activates reciprocal responsiveness. This reconceptualization shifts the analytical focus from who is vulnerable to how and when voters become incorporated into exchange networks.\u003c/p\u003e\u003ch2\u003eClientelism as Informal Social Protection\u003c/h2\u003e\u003cp\u003eThe preceding analysis suggests that the vulnerability and situational activation mechanisms are not mutually exclusive but sequentially related within a broader political economy of redistribution. Reconceptualizing clientelism as informal social protection provides a framework for integrating these dynamics. In unequal democracies where formal welfare provision remains uneven, discretionary, or selectively administered, targeted inducements may function as short-term redistributive transfers delivered through electoral channels. Inequality thus shapes the demand-side conditions that render material benefits socially meaningful. At the same time, the behavioral activation of reciprocity appears to depend less on demographic stratification than on incorporation into exchange networks. Exposure, rather than structural identity alone, structures compliance. From a population perspective, this implies that inducement-based exchange need not be confined to the poorest citizens. In competitive, candidate-centered electoral environments, the distribution of inducements may extend across social strata, activating reciprocal norms beyond narrowly defined vulnerable groups. Compliance, in this context, reflects pragmatic adaptation within relational networks rather than passive subordination rooted in fixed demographic disadvantage.\u003c/p\u003e\u003cp\u003eThis interpretation complicates conventional narratives that treat vote buying solely as evidence of democratic pathology or as a straightforward product of poverty. While clientelism may undermine programmatic accountability, it also reflects how citizens navigate environments characterized by uneven social protection and competitive political mobilization. Inequality provides the structural terrain upon which exchange becomes viable, but exposure determines who becomes behaviorally incorporated into that exchange. Seen in this light, clientelism operates at the intersection of redistribution, relational obligation, and institutional competition.\u003c/p\u003e\u003ch2\u003eImplications for Inequality and Democratic Participation\u003c/h2\u003e\u003cp\u003eThese findings carry broader implications for research on inequality and democratic participation within population and social studies. If clientelistic compliance is not independently stratified by education, age, or residence once exposure is taken into account, then socioeconomic development alone may be insufficient to weaken exchange-based electoral practices. Expanding educational attainment or reducing income inequality may alter the structural landscape of vulnerability, but such changes do not automatically disrupt relational exchange dynamics embedded in competitive electoral systems. Structural transformation may reduce the demand-side conditions that render inducements meaningful, yet the activation of compliance appears contingent on network incorporation rather than demographic position alone.\u003c/p\u003e\u003cp\u003eAt the same time, the selective nature of exposure underscores that clientelism does not uniformly structure political participation. Inducement-based exchange operates within specific networks, constituencies, and localities rather than encompassing the electorate as a whole. This pattern shifts analytical attention toward institutional and organizational incentives—campaign finance arrangements, candidate-centered competition, broker-mediated mobilization, and decentralized electoral strategies—that shape the distribution and strategic deployment of inducements. Demographic inequality may provide the terrain, but institutional design influences how and where exchange practices take root. More broadly, the evidence reinforces the importance of analytically separating structural conditions from situational mechanisms in the study of political behavior. Inequality creates the environment in which informal redistribution acquires social meaning, yet behavioral responsiveness appears closely tied to direct incorporation into exchange relationships. Clientelism in unequal democracies, therefore, cannot be reduced to poverty alone; it reflects a dynamic interaction between demographic structure, institutional incentives, and relational mobilization strategies. The concluding section considers how this reconceptualization informs debates about democratic development and the prospects for programmatic competition in societies marked by persistent inequality and evolving demographic change.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study examined whether clientelistic compliance in unequal democracies is primarily structured by socioeconomic vulnerability or activated through situational incorporation into exchange networks. Drawing on post-election survey data from Thailand\u0026rsquo;s 2023 general election, the analysis distinguished structural demographic indicators from direct exposure to inducements and further differentiated reciprocity-based compliance from coercive justification. The evidence indicates that while exposure to vote buying remains a visible feature of electoral competition, compliance behavior is not independently stratified by education, age, gender, or residence once exposure is taken into account. Nor do coercive expectations emerge as a coherent driver of responsiveness. Instead, the observed patterns are more consistent with a situational activation mechanism in which behavioral reciprocity is linked to incorporation into exchange relationships rather than predetermined by demographic disadvantage.\u003c/p\u003e \u003cp\u003eThese findings refine how inequality should be understood in relation to clientelism. Structural inequality appears to shape the broader environment in which informal exchange becomes meaningful, particularly where formal social protection remains uneven or selectively administered. Yet inequality does not translate mechanically into differentiated compliance across demographic groups. In concise terms, inequality structures the demand side of clientelism, but exposure structures the behavioral response. Clientelism, in this context, operates less as a fixed hierarchy of domination and more as a relational mobilization strategy embedded within competitive electoral systems. For population and social studies scholarship, this distinction is consequential. If compliance were strongly concentrated among socioeconomically disadvantaged groups, clientelism would represent a clear mechanism reinforcing stratified political voice and democratic inequality. Instead, the evidence suggests that exchange-based responsiveness may extend across social strata once individuals are directly incorporated into inducement-based networks. This implies that socioeconomic development alone\u0026mdash;while normatively desirable\u0026mdash;may not be sufficient to weaken clientelist practices if institutional incentives and broker-mediated mobilization persist.\u003c/p\u003e \u003cp\u003eMore broadly, the study underscores the analytical importance of separating structural conditions from situational mechanisms in the study of democratic participation. Inequality shapes the context within which informal redistribution acquires social meaning, but incorporation into exchange networks appears central in structuring reported behavioral responsiveness. Understanding how these dynamics interact is essential for designing reforms aimed at strengthening programmatic competition and reducing reliance on discretionary electoral transfers. Future research incorporating longitudinal designs, community-level variation, and comparative cases across unequal democracies would help clarify whether the Thai case reflects broader regional dynamics or context-specific institutional configurations. In unequal societies undergoing demographic transformation, the study of clientelism requires attention not only to who is vulnerable, but to how and when citizens become embedded in exchange relationships. By disentangling structural vulnerability from situational activation, this study advances a more precise and demographically grounded understanding of how inequality intersects with electoral behavior in contemporary democracies.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003cb\u003eFunding Declaration\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThis research received no specific external funding. The study was conducted as part of the author\u0026rsquo;s research activities at the Faculty of Political Science, Chulalongkorn University, Thailand.\u003c/p\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eEthics Statement\u003c/h2\u003e \u003cp\u003eThe public opinion survey used in this study was reviewed and approved by the Institutional Review Board (IRB) of King Prajadhipok\u0026rsquo;s Institute, Thailand. The research was conducted in accordance with established ethical guidelines for social science research involving human participants. The survey data were collected anonymously, and no personally identifiable information was recorded. Participation in the survey was voluntary, and respondents were informed about the purpose of the study prior to participation.\u003c/p\u003e \u003ch2\u003eConsent to Participate\u003c/h2\u003e \u003cp\u003eInformed consent was obtained from all participants prior to their participation in the survey.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent to Publish\u003c/strong\u003e \u003cp\u003eNot applicable. The manuscript does not contain any individual person\u0026rsquo;s identifiable data in any form.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eS.T. conceptualized the study, developed the theoretical framework, designed the research methodology, conducted the data analysis, interpreted the results, and wrote and revised the manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated during and/or analysed during the current study are not publicly available because the survey dataset is maintained by King Prajadhipok\u0026rsquo;s Institute and contains confidential respondent information, but are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbdulai AG. Competitive clientelism, donors and the politics of social protection uptake in Ghana. Crit Soc Policy. 2021;41(2):270\u0026ndash;93. ttps://doi.org/10.1177/0261018320945605.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAspinall E. Money politics: Patronage and clientelism in Southeast Asia. Routledge handbook of southeast Asian democratization. Routledge; 2015. pp. 299\u0026ndash;313.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBardhan P, Mookherjee D. (2020). Clientelistic politics and economic development. \u003cem\u003eThe handbook of economic development and institutions\u003c/em\u003e, 84\u0026ndash;102.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBirch S, Cockshott P, Renaud K. Putting electronic voting under the microscope. Political Q. 2014;85(2):187\u0026ndash;94. ttps://doi.org/10.1111/1467-923X.12071.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoossabong P. Patronage in Thailand: trust and reciprocity in policy networks. Handbook of Politicization and Political Patronage. Edward Elgar Publishing; 2025. pp. 404\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBowie KA. Vote buying and village outrage in an election in northern Thailand: Recent legal reforms in historical context. J Asian Stud. 2008;67(2):469\u0026ndash;511. ttps://doi.org/10.1017/S0021911808000673.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCallahan WA. The discourse of vote buying and political reform in Thailand. Pac Affairs. 2005;78(1):95\u0026ndash;113. ttps://doi.org/10.5509/200578195.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCallahan WA, McCargo D. Vote-buying in Thailand\u0026rsquo;s Northeast: The July 1995 general election. Asian Surv. 1996;36(4):376\u0026ndash;92. ttps://doi.org/10.2307/2645405.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCanare TA, Mendoza RU, Lopez MA. An empirical analysis of vote buying among the poor: Evidence from the Philippines. Asian Politics Policy. 2018;10(4):659\u0026ndash;79. ttps://doi.org/10.1177/0967828X17753420.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarlin RE, Moseley MW. When clientelism backfires: Vote buying, democratic attitudes, and electoral retaliation in Latin America. Polit Res Q. 2022;75(3):766\u0026ndash;81. ttps://doi.org/10.1177/10659129211020126.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChattharakul A. Thai electoral campaigning, vote-canvassing networks, and hybrid voting. Asian Politics Policy. 2010;2(1):67\u0026ndash;89. ttps://doi.org/10.1177/186810341002900404.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorstange D. Clientelism in competitive and uncompetitive elections. Comp Polit Stud. 2018;51(1):76\u0026ndash;104. ttps://doi.org/10.1177/0010414017695332.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCruz C. Social networks and the targeting of vote buying. Comp Polit Stud. 2019;52(3):382\u0026ndash;411. ttps://doi.org/10.1177/0010414018784062.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDella Porta D, Vannucci A. Corrupt exchanges: Actors, resources, and mechanisms of political corruption. Routledge; 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElliott C. Clientelism and the democratic deficit. Stud Indian Politics. 2016;4(1):22\u0026ndash;36. ttps://doi.org/10.1177/2321023016634915.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFergusson L, Molina CA, Robinson JA. The weak state trap. Economica. 2022;89(354):293\u0026ndash;331. ttps://doi.org/10.1111/ecca.12399.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHicken A. Clientelism. Annu Rev Polit Sci. 2011;14:289\u0026ndash;310. ttps://doi.org/10.1146/annurev.polisci.031908.220508.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHicken A, Aspinall E, Weiss ML, Muhtadi B. Buying brokers: Electoral handouts beyond clientelism in a weak-party state. World Polit. 2022;74(1):77\u0026ndash;120. ttps://doi.org/10.1017/S0043887121000216.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHidalgo FD, Nichter S. Voter buying: Shaping the electorate through clientelism. Am J Polit Sci. 2016;60(2):436\u0026ndash;55. ttps://doi.org/10.1111/ajps.12214.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJensen PS, Justesen MK. Poverty and vote buying: Survey-based evidence from Africa. Electoral Stud. 2014;33:220\u0026ndash;32. ttps://doi.org/10.1016/j.electstud.2013.07.020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJustesen MK, Manzetti L. Poverty, partisanship, and vote buying in Latin America. Lat Am Politics Soc. 2023;65(3):1\u0026ndash;19. ttps://doi.org/10.1017/lap.2022.66.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKitschelt H, Wilkinson SI, editors. Patrons, clients and policies: Patterns of democratic accountability and political competition. Cambridge University Press; 2007.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLawson C, Greene KF. Making clientelism work: How norms of reciprocity increase voter compliance. Comp Politics. 2014;47(1):61\u0026ndash;85. ttps://doi.org/10.5129/001041514813623173.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMares I, Young LE. Conditionality \u0026amp; coercion: Electoral clientelism in eastern Europe. Oxford University Press; 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOwen N. Conceptualizing vote buying as a process: An empirical study in Thai provinces. Asian Politics Policy. 2013;5(2):247\u0026ndash;65. ttps://doi.org/10.1111/aspp.12028.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRavanilla N, Hicken A. Poverty, social networks, and clientelism. World Dev. 2023;162:106128. ttps://doi.org/10.1016/j.worlddev.2022.106128.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRavanilla N, Haim D, Hicken A. Brokers, social networks, reciprocity, and clientelism. Am J Polit Sci. 2022;66(4):795\u0026ndash;812. ttps://doi.org/10.1111/ajps.12604.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRemmer KL. The political economy of patronage: Expenditure patterns in the Argentine provinces, 1983\u0026ndash;2003. J Politics. 2007;69(2):363\u0026ndash;77. ttps://doi.org/10.1111/j.1468-2508.2007.00537.x.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScott JC. Corruption, machine politics, and political change. Am Polit Sci Rev. 1969;63(4):1142\u0026ndash;58. ttps://doi.org/10.2307/1955076.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSingh SP. Compulsory voting and parties\u0026rsquo; vote-seeking strategies. Am J Polit Sci. 2018;62(2):387\u0026ndash;403. ttps://doi.org/10.1111/ajps.12386.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStokes SC, Dunning T, Nazareno M. Brokers, voters, and clientelism. Cambridge University Press; 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSwamy AR. Can social protection weaken clientelism? Considering conditional cash transfers as political reform in the Philippines. J Curr Southeast Asian Affairs. 2016;35(1):59\u0026ndash;90. ttps://doi.org/10.1177/186810341603500103.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor JS. Autonomy, vote buying, and constraining options. J Appl Philos. 2017;34(5):711\u0026ndash;23. ttps://doi.org/10.1111/japp.12205.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWantchekon L. Clientelism and voting behavior: Evidence from a field experiment in Benin. World Polit. 2003;55(3):399\u0026ndash;422. ttps://doi.org/10.1353/wp.2003.0018.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-global-society","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Global Society](https://www.springer.com/journal/44282)","snPcode":"44282","submissionUrl":"https://submission.nature.com/new-submission/44282/3","title":"Discover Global Society","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Electoral Clientelism, Vote Buying, Informal Social Protection, Political Inequality, Thailand","lastPublishedDoi":"10.21203/rs.3.rs-9049351/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9049351/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eElectoral clientelism is often interpreted as a consequence of socioeconomic vulnerability in unequal democracies. Yet it remains unclear whether compliance with vote buying is primarily structured by demographic disadvantage or activated through situational incorporation into exchange networks. This study examines clientelistic behavior in Thailand\u0026rsquo;s 2023 general election using nationally administered post-election survey data. Distinguishing between structural indicators of vulnerability (education, age, gender, and residence) and direct exposure to inducements, the analysis constructs multidimensional measures of exposure, behavioral reciprocity, and coercive justification. The findings indicate that while exposure to vote buying remains a visible component of electoral competition, compliance is not independently stratified by demographic characteristics once exposure is taken into account. Nor do coercive expectations form a coherent driver of responsiveness. Instead, behavioral reciprocity is strongly associated with direct incorporation into inducement-based exchange networks. These results suggest that inequality shapes the structural conditions under which informal redistribution becomes meaningful, but exposure structures the behavioral response. By disentangling structural vulnerability from situational activation, the study advances debates on clientelism, informal social protection, and stratified political incorporation in middle-income democracies.\u003c/p\u003e","manuscriptTitle":"Clientelism as Informal Social Protection in Unequal Democracies: Evidence from Thailand’s 2023 Election","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-04 11:05:04","doi":"10.21203/rs.3.rs-9049351/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-11T17:08:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"236085961051787546039422559966577974439","date":"2026-05-04T13:35:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-03T12:19:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"8164028729208323519588661401105363813","date":"2026-05-03T10:18:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"201354485550190402080396327735605129692","date":"2026-05-03T06:39:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-22T18:32:13+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-21T10:38:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-21T01:57:27+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-18T10:29:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Global Society","date":"2026-03-13T14:48:20+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-global-society","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Global Society](https://www.springer.com/journal/44282)","snPcode":"44282","submissionUrl":"https://submission.nature.com/new-submission/44282/3","title":"Discover Global Society","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2ea0cd2c-b8d8-4400-9c7d-d0010b4b1646","owner":[],"postedDate":"May 4th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-11T17:08:35+00:00","index":115,"fulltext":""},{"type":"reviewerAgreed","content":"236085961051787546039422559966577974439","date":"2026-05-04T13:35:44+00:00","index":107,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-03T12:19:15+00:00","index":100,"fulltext":""},{"type":"reviewerAgreed","content":"8164028729208323519588661401105363813","date":"2026-05-03T10:18:03+00:00","index":99,"fulltext":""},{"type":"reviewerAgreed","content":"201354485550190402080396327735605129692","date":"2026-05-03T06:39:50+00:00","index":97,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-04T11:05:04+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-04 11:05:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9049351","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9049351","identity":"rs-9049351","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}