On the opposition to market institutions on moral grounds

The paper examines why some market institutions elicit moral opposition despite being based on voluntary exchange. Building on debates about repugnance, moral norms, potential crowding out of prosocial motivation, distributional and relational harms, and paternalism, the authors investigate moral opposition to trading in several domains (organs, sex services, surrogacy, CO2 permits, goods from poor countries, and food from hunger-stricken countries). They hypothesize that opposition varies by context and that a key moral dimension—aversion to imposing risks on others for one’s own benefit—correlates with resistance to market institutions and policy preferences related to risk, equality, and redistribution.

The study connects to work on repugnant markets (McGraw and Tetlock; Roth; Bénabou and Tirole; Roth and Wang), concerns that markets erode moral values or crowd out prosocial motives (Falk and Szech; Bruni and Sugden; Storr and Choi; Ziegler et al.; Bartling et al.), and fairness/distributional critiques of markets (Sandel; Helpman et al.; Konow and Schwettmann). Satz’s framework on “noxious markets” highlights vulnerability, weak agency, and harms to individuals and society. Paternalism literature (Perri; Parker; Thaler and Sunstein; Ambuehl et al.) provides a rationale for limiting choices to prevent risky behavior. Ethical analyses of risk imposition (Hansson; Gardoni and Murphy) motivate the central measure of attitudes toward imposing risks on others. Prior work on demographic correlates of morality and attitudes (age, gender, socioeconomic status) also informs the study’s covariates.

Design: Online survey embedded in a broader experiment, implemented via SoSci Survey and fielded to U.S.-based Amazon Mechanical Turk workers in December 2020. Participants: 902 recruited; 854 completed and were analyzed. Average payment: USD 7.45 (survey answers not incentivized). Measures: Moral opposition to trade across seven fields: human body parts (opp_body), sexual services (opp_sex), food from countries with widespread hunger (opp_hunger), emission permits for countries (opp_cntrCO2) and firms (opp_firmCO2), goods from poor countries (opp_poor), and surrogacy (opp_surrogate). Ordinal scale 1 (morally acceptable) to 5 (morally unacceptable). Binary indicators for strong opposition (oppose_field = 1 if ordinal=5). Policy support (World Values Survey-style items): equality as a goal (pol_equality), redistribution via taxing rich/subsidizing poor (pol_redistribute), unemployment benefits (pol_unemploy), and restricting choices to prevent extreme risks (pol_riskprevent), each 1 (disagree completely) to 5 (agree completely). Explanatory variables: Risk preferences via Dohmen-style self-assessment (RiskAverseMe, higher values indicate more risk aversion). Social preference measure toward imposing risk on others for own benefit (RiskOnOthers, higher values indicate more aversion to imposing risks). Demographics: age, gender (female), university education (eduni), economics training (econ), political orientation (conservative scale 1 liberal to 10 conservative). Analysis: Descriptives, pairwise comparisons (Wilcoxon signed-rank tests), correlations across fields, and OLS regressions predicting opposition (ordinal and binary versions) from explanatory variables. Also OLS regressions for determinants of RiskAverseMe and RiskOnOthers, and for policy support (pol_*). Summary stats (Table 3): RiskAverseMe mean 2.92 (sd 1.14), RiskOnOthers 3.51 (1.27), conservative 5.67 (2.92), age 40.04 (11.74), female share 0.45, eduni 0.65, econ 0.52.

• Opposition varies by context: average opposition (ordinal mean) and strong opposition shares (binary) from Table 4: body parts 3.40; oppose 33%. Sex services 3.09; oppose 25%. Hunger-related food imports 3.26; oppose 23%. CO2 permits (firms) 3.06; oppose 15% (similar to countries 3.05; oppose 15%). Goods from poor countries 2.65; oppose 9%. Surrogacy 2.22; oppose 6%.
• Over half (53%) strongly oppose trade in at least one field; 21% in exactly one; 1% oppose all fields. Opposition measures are pairwise highly correlated at the individual level (p<0.002).
• Pairwise comparisons show significant differences across most fields; sex vs. hunger not significantly different in strong opposition.
• Regression results (Tables 7–8): Aversion to imposing risks on others (RiskOnOthers) is a strong positive correlate of opposition to trade across all contexts except surrogacy (where overall opposition is very low). University education and economics training associate with lower opposition. Females are more opposed to sex trade. Conservatives are more opposed to sex trade but less opposed in other domains.
• Determinants of risk measures (Table 9): Higher age and being female associate with greater aversion to risk (self and others). University education, economics training, and being conservative are associated with lower aversion. R-squared: 0.15 (RiskAverseMe), 0.22 (RiskOnOthers).
• Policy preferences (Table 4 averages): unemployment benefits 3.88, redistribution 3.55, equality 2.84, risk prevention 2.58, with all mutual differences highly significant (p<0.001). OLS (Table 10): Greater willingness to impose risks on others (lower RiskOnOthers) predicts less support for redistribution, equality, and risk prevention; no significant relation with unemployment benefits. Economics training positively relates to support for unemployment, redistribution, and risk prevention; conservatism negatively relates to support for equality and redistribution.

Findings show that moral opposition to market institutions is context-specific yet strongly interrelated within individuals, suggesting an underlying moral disposition. A key correlate is aversion to imposing risks on others for personal gain, indicating that paternalistic and relational concerns (vulnerability, weak agency, harms) shape resistance to certain markets. The alignment between this aversion and support for equality, redistribution, and risk prevention policies further links individual social preferences to policy attitudes. The weaker link for unemployment benefits and the counterintuitive negative correlation with risk-prevention policies are explained by the distinct nature of the measured preference (hesitation to impose risks on others for one’s own benefit) versus policies that restrict others from taking risks themselves. Demographic and ideological patterns (e.g., higher opposition among females to sex markets; conservatives opposing sex trade but supporting other market exchanges) highlight heterogeneity that maps onto known moral and political dimensions. Overall, the results address the research question by identifying a measurable moral preference—risk-on-others aversion—that helps explain opposition to various market institutions and associated policy views.

A substantial share of respondents strongly oppose trade in specific domains, notably body parts, sex services, and food from countries with widespread hunger. Opposition is systematically related to an aversion to imposing risks on others, suggesting that moral concerns about potential inducement into risky choices drive resistance to markets. The newly applied measure of aversion to exposing others to risk also predicts support for redistribution, equality, and risk-prevention policies. The study contributes by integrating multiple morally charged market contexts with a unifying behavioral measure that links market opposition to policy preferences. Future work should use survey experiments to causally identify determinants of opposition, for instance by exogenously varying information about the impacts and distributional consequences of trade opportunities.

• Observational survey design limits causal inference; results are correlational.
• Sample restricted to U.S.-based MTurk workers (December 2020), which may limit generalizability.
• Evidence of multicollinearity among regressors (e.g., RiskAverseMe and RiskOnOthers correlate with demographics), potentially inflating standard errors.
• Very low opposition to surrogacy reduces statistical power in that domain.
• Socioeconomic status proxied only by education; income not measured.