Table 2.

The effect of changes in debt structure on psychological functioning and economic decision-making

Variables	Error rate (flanker)	Response time (flanker)	Combined score (flanker)	GAD	Risk aversion	Present bias
Debt-relief amount	−0.026*** (0.006)	−0.043*** (0.016)	0.222*** (0.048)	−0.020 (0.018)	−0.110 (0.072)	0.029 (0.017)
Debt accounts paid off	−0.035*** (0.010)	−0.095*** (0.024)	0.353*** (0.086)	−0.113*** (0.025)	−0.220 (0.140)	−0.100*** (0.032)
Constant	0.170*** (0.007)	0.445*** (0.021)	6.337*** (0.065)	0.773*** (0.018)		0.430*** (0.026)
Observations	392	392	392	392	196	298
Number of participants	196	196	196	196	196	149

Each column reports a separate ordinary least-squares regression model with controls for individual fixed effects, except for “Risk Aversion,” which uses an interval regression model on first-differenced data. “Debt-Relief Amount” is in SGD thousands. The “Error Rate” dependent variable is the proportion of incorrect flanker trials (1.0 = all trials incorrect). The “Response Time” dependent variable is the log of median response time in seconds, based on all flanker trials for that participant. The “Combined Score” dependent variable is a 0–10 scale that combines flanker error rate and response time. The “GAD” dependent variable is 1 for GAD symptoms and 0 for no GAD. The “Risk Aversion” dependent variable is the change in the CRRA parameter interval pre- and post-debt relief. The “Present Bias” dependent variable is 1 for present bias and 0 for no present bias, and it excludes respondents with inconsistent time-discounting choices. SI Appendix, section 3 details the measures. Robust SEs are in parentheses. ***P < 0.01.