In our research, we aim to shed light on the role of overdue debt in reinforcing poverty. This not only helps to better understand the dynamics of poverty trap induced by overdue debt but also enhances the rediscussing of current policy tools. Our research is based on data collected with targeted questionnaires in March and April 2019 by the Soreco Research Kft. Data were recorded with a personal question and answer method, by a so-called multi-stage stratified random sampling procedure. The data collection was anonymised and focused on the financial and liquidity decisions of households in small settlements in one of the most disadvantaged counties of Hungary, Borsod-Abaúj-Zemplén (BAZ) county. The sample is representative on the level of households living in small settlements. After cleaning the raw data, we have information from 504 households and 1794 individuals. 1196 individuals are of legal age, from whom 179 had overdue debt. We develop a theoretical model inspired by Akerlof (1978), Tirole (2006), and Mukherjee, Subramanian, Tantri (2019) to derive a feasibility condition for market-based debt relief programs. Our empirical analysis aims at investigating the role of overdue debt in creating poverty trap. With the help of statistical analysis and linear probability models we examine the impact of overdue debt on employment, on having a bank account, and on mental- and physical- health based on targeted questionnaires and in-depth interviews in the most disadvantaged regions of Hungary. We controlled for socioeconomic factors (e.g., gender, age, education level, ability to pay) and for settlement and county development indicators. In these regions, a significant part of the society has been the victim of financial exclusion well before the Covid 19 crisis, even under prospering economic conditions. Results: § The theoretical model shows that lenders have no interest to offer payment reductions if non-performing borrowers are few, have small debts, and are difficult to reach. In this situation, poor debtors serve better as deterrents, similarly if we put them into a pillory. § Calibrating model parameters to poor households struggling with overdue debts, we show that this might be the case on our sample, too. § As, in normal economic circumstances, private debt relief programs are typically not feasible, a state subsidy would be needed to consolidate the debts of the poor. State intervention can be justified both by positive externalities and moral considerations. § We find that many borrowers hide from debt collection as a consequence of overdue debt that has escalated to an unbearable level due to penalty rates. These borrowers are following the hiding strategy and take their decisions accordingly: to avoid deductions, they do not apply for registered jobs, do not open bank accounts and consequently, they are forced to live under constant stress. § To sum up the impact of overdue debt on social inclusion factors and according to our estimations, overdue debts reduce the likelihood of having a registered job by nearly 14 percentage points. Not having a registered job reduces the probability of owning a bank account by 22 percentage points and, in addition, overdue debts further decrease the probability by 5 percentage points. In addition, overdue debt also has a negative effect on the health of those living in the same household as the debtor, and this negative effect is greater than what a combined high school diploma and diploma could compensate for (0.4 versus 1.08-0.72 = 0.36). § Overdue debt, therefore, leads to a certain type of debt-trap mechanism resulting in significant loss for both the individual and the society. In this light, policy makers should pay more attention to addressing credit cycles and resolving non-performing debt obligations, especially in this fragile part of the society.