research
My research statement is available here.
publications
FinTech Lending to Borrowers with No Credit History (with Laura Chioda, Paul Gertler, and Paolina Medina). NBER Working Paper 33208. Forthcoming, Journal of Financial Economics.
-
Abstract
Despite the promise of FinTech lending to expand credit access to populations without a formal credit history, FinTech lenders primarily lend to applicants with a formal credit history and rely on conventional credit bureau scores as an input to their algorithms. Using data from a large FinTech lender in Mexico, we show that alternative data from digital transactions through a delivery app are effective at predicting creditworthiness for borrowers with no credit history. Using account-by-month level data on revenues and costs, a machine learning model predicting profits generates similar profits as a model predicting default.
- Ungated PDF – replication package
- Media coverage: MIT Technology Review; Financial Alliance for Women; Center for Effective Global Action
Financial Technology Adoption: Network Externalities of Cashless Payments in Mexico. American Economic Review 114(11), 3469–3512, 2024.
-
Abstract
Do coordination failures constrain financial technology adoption? Exploiting the Mexican government's rollout of 1 million debit cards to poor households from 2009--2012, I examine responses on both sides of the market, and find important spillovers and distributional impacts. On the supply side, small retail firms adopted point-of-sale terminals to accept card payments. On the demand side, this led to a 21% increase in other consumers' card adoption. The supply-side technology adoption response had positive effects on both richer consumers and small retail firms: richer consumers shifted 13% of their supermarket consumption to small retailers, whose sales and profits increased.
- Ungated PDF – Online Appendix – presentation slides – replication package
- Summary: VoxDev; IPA; World Bank Development Impact Blog
- Video: SFS Cavalcade (25 minute presentation, plus discussion and Q&A)
- Media coverage: Kellogg Insight
- Best Paper in Corporate Finance, SFS Cavalcade North America
- PBCSF Award for the Best Paper in FinTech, WFA
- Best Paper Award, GSU-RFS FinTech Conference
How Debit Cards Enable the Poor to Save More (with Pierre Bachas, Paul Gertler, and Enrique Seira). Journal of Finance 76(4), 1913-1957, 2021.
-
Abstract
We study an at-scale natural experiment in which debit cards were given to cash transfer recipients who already had a bank account. Using administrative account data and household surveys, we find that beneficiaries accumulated a savings stock equal to 2% of annual income after two years with the card. The increase in formal savings represents an increase in overall savings, financed by a reduction in current consumption. There are two mechanisms. First, debit cards reduce transaction costs of accessing money. Second, they reduce monitoring costs, which led beneficiaries to check their account balances frequently and build trust in the bank.
- Ungated PDF – Online Appendix – replication package
- Summary: VoxDev
- Media coverage: Kellogg Insight; Market Watch
Digital Financial Services Go a Long Way: Transaction Costs and Financial Inclusion (with Pierre Bachas, Paul Gertler, and Enrique Seira). American Economic Association Papers & Proceedings 108, 444-448, 2018.
-
Abstract
Debit cards reduce the travel distance to access bank accounts and can increase financial inclusion. We show that in Mexico, cash transfer beneficiaries who already received their transfers in bank accounts and subsequently received debit cards reduce their median distance to access the account from 4.8 to 1.3 kilometers. They also report being less likely to forgo important activities (childcare, work) to withdraw their transfers. Using account level data, we find a strong correlation between the reduction in travel distance and financial activity: beneficiaries facing the largest reductions in distance increase both their number of withdrawals and their savings balances.
- Ungated PDF – replication package
- Summaries of AEA session I organized: CEGA; NYU Wagner Financial Access Initiative
Can a Poverty-Reducing and Progressive Tax and Transfer System Hurt the Poor? (with Nora Lustig). Journal of Development Economics 122, 63-75, 2016.
-
Abstract
To analyze anti-poverty policies in tandem with the taxes used to pay for them, comparisons of poverty before and after taxes and transfers are often used. We show that these comparisons, as well as measures of horizontal equity and progressivity, can fail to capture an important aspect: that a substantial proportion of the poor are made poorer (or non-poor made poor) by the tax and transfer system. We illustrate with data from seventeen developing countries: in fifteen, the fiscal system is poverty-reducing and progressive, but in ten of these at least one-quarter of the poor pay more in taxes than they receive in transfers. We call this fiscal impoverishment, and axiomatically derive a measure of its extent. An analogous measure of fiscal gains of the poor is also derived, and we show that changes in the poverty gap can be decomposed into our axiomatic measures of fiscal impoverishment and gains.
- Ungated PDF – replication package
- Media coverage: Washington Post
working papers
Using Lotteries to Attract Deposits (with Paul Gertler, Aisling Scott, and Enrique Seira). NBER Working Paper 31529. Revise and resubmit, Journal of Finance.
-
Abstract
Despite the importance of deposit financing for lending, banks in developing countries struggle to attract deposits. In a randomized experiment across 110 bank branches throughout Mexico, a lottery incentive based on net monthly deposits caused a 36% increase in the number of accounts opened and a 21% increase in the number of deposits during the lottery months. Nearly all new accounts (96%) were opened by households previously unbanked at any bank. The temporary two-month incentive had a persistent 2-3 year impact on the flow of deposits and stock of savings, and increased the present value of branch profits by 6%.
- Summary: VoxDev; VoxEU
- Media Coverage: Kellogg Insight
- RCT registration
Do Behavioral Frictions Prevent Firms from Adopting Profitable Opportunities? (with Paul Gertler, Ulrike Malmendier, and Waldo Ojeda). NBER Working Paper 33387. Revise and resubmit, Econometrica.
-
Abstract
Firms frequently fail to adopt profitable business opportunities even when they do not face informational or liquidity constraints. We explore three behavioral frictions that explain inertia among individuals---present bias, limited memory, and distrust---in a managerial setting. In partnership with a FinTech payments company in Mexico, we randomly offer 33,978 firms the opportunity to pay a lower merchant fee. We vary whether the offer has a deadline, reminder, pre-announced reminder, and the size of the fee reduction. Reminders increase take-up by 15%, suggesting a role of memory. Announced reminders increase take-up by an additional 7%. Survey data reveal the likely mechanism: When the FinTech company follows through with the pre-announced reminder, firms' trust in the offer increases. The deadline does not affect larger firms, implying limited or no present bias, but does increase take-up by 8% for smaller firms. Overall, behavioral frictions contribute significantly to explaining profit-reducing firm behavior.
- Media coverage: Kellogg Insight
- RCT registration
Search and Negotiation with Biased Beliefs in Consumer Credit Markets (with Erik Berwart, Sheisha Kulkarni, and Santiago Truffa). NBER Working Paper 35605. Submitted.
-
Abstract
How do biased beliefs about the interest rate distribution affect search, negotiation, and loan terms in consumer credit markets? In collaboration with Chile's financial regulator, we conducted a randomized controlled trial with 112,063 loan seekers. Randomly eliciting beliefs about interest rates led participants to search more and obtain lower rates. Most participants underestimated both interest rate levels and dispersion. Showing them a price comparison tool we built using administrative data on the universe of consumer loans caused them to update their beliefs, negotiate more, obtain lower rates without searching more, and be more likely to take out a loan.
- Summary: IPA
- RCT registration
Towards a Cashless Economy? Evidence from the Elasticity of Cash Deposits of Mexican Firms (with Pierre Bachas and Anders Jensen).
-
Abstract
Policies that increase the cost of using cash seek to promote the transition towards cashless and traceable transactions, but those that operate through the banking sector may also discourage participation in formal banking. We study this trade-off in the context of Mexico’s tax on cash deposits (2008–2013), which levied a monthly tax on cash deposits while leaving non-cash deposits untaxed. Using account-level data on firms' deposits at a large commercial bank, we find that firms' cash deposits are highly elastic to the tax rate, while substitution towards non-cash deposits is incomplete, leading to a decrease in overall bank deposits. The elasticity of cash deposits is lower among firms in high-robbery areas, consistent with the frequent on-premises robberies reported by firms, and with high crime rates leading firms to prefer not to hold cash on site. We use firm census data to study real effects, and find that more-exposed firms reduce their economic activity in response, consistent with trying to limit the cash they hold on premises due to crime risk. Contrary to its intention, taxing cash caused financial disintermediation and had negative real effects on firms.
Mental Accounting for Credit: Evidence from a Large-Scale Credit-Limit Experiment (with Ray Kluender).
-
Abstract
We develop a model of mental accounting for credit in which consumers budget according to non-fungible spending categories that govern their borrowing behavior. We test our model in a randomized controlled trial with Cashea, Venezuela's largest consumer credit provider. Cashea provides two distinct credit lines to all of its borrowers: one for grocery stores, pharmacies, and restaurants, and one for all other retail merchants. We randomized 0%, 50%, or 100% credit-limit increases on one or both credit lines for 209,996 active users. For both credit lines, we estimate large same-category monthly marginal propensities to consume (MPCs) of 8-10 cents per $1 of additional credit, with near-zero cross-category MPCs, consistent with mental accounting. We present five additional supporting results: (1) the results persist in every post-treatment month; (2) the same-category MPCs represent an increase in total consumption; (3) the results hold separately for both constrained and unconstrained borrowers; (4) we fail to identify any subgroup that treats the relaxation of the budget constraint fungibly; and (5) the results hold separately for each sub-category of consumption within each credit line. We conclude that consumers form mental accounts for credit, with economically large effects on borrowing and the allocation of spending across categories. This has important implications for the regulation of financial technology and the effects of monetary policy.
- RCT registration
- Marie Curie Prize for Most Creative Paper, Advances with Field Experiments Conference
work in progress
Government-Backed Small Business Loans During Economic Crises Prevent Job Loss and Increase Sales and Profits (with Paul Gertler, Ana María Montoya, Eric Parrado, and Raimundo Undurraga). Email for slides; draft coming soon.
-
Abstract
We use a randomized controlled trial with 10,072 small and medium enterprises (SMEs) to examine the impact of a government-backed loan program for small businesses in Chile during the COVID pandemic. In collaboration with a commercial bank in Chile, we randomized half of these 10,072 SMEs to receive loan offers in March-June 2021 during Chile’s most widespread lockdowns. Take-up of the loans was 27%, and the total amount of credit that treated firms (compared to control firms) had across all lenders increased one-for-one with the loans from our experiment. Using linked employer-employee administrative data, we find that while control firms fired employees during lockdown, treated firms did not, leading to 5% higher employment in treated firms over the first year of the experiment. Treatment effects on employment persisted for at least 8 months after firms permanently exited lockdown. In data from a short-term follow-up survey 1-2 years after treatment, treated firms have 24% higher sales and 27% higher profits.
- RCT registration