Avoiding Debt: Balancing Psychological Burdens and Financial Consequences

  1. References:

Alejandro Martínez Marquina (Klarman Fellow, Cornell University Department of Economics) shares experimental evidence around aversion to debt and reluctance to borrow money. This study was supported by CEGA’s Psychology and Economics of Poverty Initiative (PEP).

micheile dot com | Unsplash

The presence of debt is ubiquitous in our lives. Some of our most consequential financial decisions, like obtaining higher education or buying a house, often require access to credit. In low- and middle-income countries (LMICs), many worthwhile investments require credit-constrained households to borrow, which they might be reluctant to do.

Financial markets and many institutions often assume that people will borrow to invest if the benefits outweigh the costs. While previous research has shown people might borrow too much (for example, by using credit cards with high interest rates and being unable to fully pay down balances), in this project, I focus on when people do not borrow enough.

Do people in debt make optimal repayment decisions?

In an online study with 638 US participants over the age of eighteen, I presented individuals with a risk-free investment decision by providing them $5 to invest across four different virtual accounts. These accounts generated considerable interest returns every two days for a week. Some of these accounts had a positive balance and hence a positive return (savings accounts), while others had a negative balance that generated negative returns (debt accounts). The main decision participants faced is how to invest $5 across the accounts. For one group of participants, all four available accounts were savings accounts, while in another group, participants had two savings and two debt accounts. To maximize returns, participants should have always invested all $5 in the account with the highest interest rate (in this study, this was always a savings account) while disregarding the balances and any debt.

Instead, 1 out of 3 participants focus on repaying their outstanding debt at the cost of missing out on higher returns from savings. Despite correctly identifying which account would provide the greatest returns, most participants switched to a strategy of maximizing returns only once they paid their outstanding debt entirely. When asked, participants reported wanting to eliminate any debt before maximizing returns.

This figure represents how much participants allocate to the second highest interest account. Focusing on this account lets us measure the choices made by different groups: in the control group, this account has a positive balance and in the treatment group, it has a negative balance. The dashed line represents the amount that cancels the debt in the treatment group.

Additionally, when debt balances are such that participants can never repay them in full by the experiment’s end, some participants no longer try to repay them and instead focus on maximizing their returns. This evidence suggests that participants have a “debt-free premium,” and for some, this is enough to make them stop repaying their debt (if they think they can’t pay it all off).

Are people willing to borrow for a profitable investment?

Finally, I explore whether individuals miss out on worthwhile investments when they require going into debt. To do so, participants in the study are allowed to borrow from two additional accounts with low-interest rates to increase their initial $5. Those accounts have a positive balance in one condition (borrow savings) and borrowing reduces a positive number that cannot go below zero. In the other condition (borrow debt), they start with a zero balance and any borrowing causes their balances to go negative.

Compared with the control condition, participants in the borrow debt group are half as likely to borrow for investment. In the borrow savings group, almost 70% of participants borrow the maximum amount, with nearly everyone using the extra funds to obtain the highest possible return. In contrast, less than 30% of participants do the same when borrowing requires going into the negatives. Again, these differences are not driven by confusion or participants’ inability to identify how profitable borrowing can be. Even when we restrict to the subsample of subjects that invest the original $5 optimally, we still see a significant difference between groups, similar in relative terms (a 50% reduction).

Lessons and policy implications

Overall, our study shows that in many instances, access to credit might not be enough to encourage the uptake of good investments if individuals are debt-averse. Indeed, individuals may face a tension between a desire to avoid debt if it is psychologically burdensome and the potential benefits of high-return investments made possible by borrowing. These results might help explain some of the investment puzzles in LMICs (see Liu and Roth (2022) and Berkouwer and Dean (2019)). At the same time, they suggest additional benefits for debt forgiveness programs and income-based repayment schemes, which could provide an avenue for individuals to simultaneously pay down debt and make investments when they offer a high return.


Berkouwer, Susanna B., and Joshua T. Dean. “Credit and attention in the adoption of profitable energy efficient technologies in Kenya.” (2019).

Liu, Ernest, and Benjamin N. Roth. “Contractual restrictions and debt traps.” The Review of Financial Studies 35.3 (2022): 1141–1182.