Evidence Generation
Asia
Evidence Generation
Asia
Read the working paper here.
Consumer credit markets around the world are characterized by high levels of price dispersion: Borrowers with similar profiles routinely pay very different interest rates for comparable loans. In Chile, the mean and median annual interest rates on consumer loans were 25.8 percent and 23.9 percent, but the rates available to any one borrower can vary substantially across lenders.1 Conducting extensive searches of available credit at banks enables consumers to secure more favorable loans, but searching can be costly; negotiating may also enable borrowers to obtain better terms without having to search extensively. Not all borrowers search or negotiate, however, which can result in taking out loans with high interest rates that are costly to pay off.
Most research has focused on the costs that prevent consumers from searching more, such as the time required to visit multiple branches, high rejection rates for less creditworthy borrowers, or having to compare complex financial products.2 However, less is known about whether consumers have accurate beliefs about what rates are available to them in the first place. This evaluation examined whether a tool providing accurate, personalized rate information could change borrower perceptions and loan-seeking behavior.
Researchers partnered with the Comisión para el Mercado Financiero (CMF)—Chile’s financial regulator—to conduct a randomized evaluation to measure whether an online interest rate comparison tool helps consumers search for more favorable loans. This tool, which the researchers designed using loan and borrower data from the CMF, shows consumers the distribution of interest rates that similar borrowers received for similar loans in the past six months.
First, a larger group of 112,063 loan seekers were randomly assigned to either receive or not receive questions regarding their expectations about the interest rates that banks would offer them.
Following this, 46,051 loan seekers were randomly divided into one of three groups, where they viewed:
IPA provided research support in gathering survey data from the loan seekers to assess if they secured loans, at what interest rate, from which financial institution, how they searched for their loan, whether they negotiated with the lender, and their financial well-being.
Results show that simply asking consumers to report their beliefs, without showing them any tool, had meaningful impacts. Consumers whose beliefs were elicited searched at more institutions and obtained interest rates 7.1 percent lower than those who were not asked.
Most loan seekers held inaccurate beliefs about the interest rate market. Nearly three-quarters of consumers expected to receive a lower interest rate than they actually obtained, and a similar percentage underestimated the dispersion of rates they would face across lenders; but a sizable share also overestimated the rate dispersion.
With the price comparison tool, consumers significantly revised their expectations, expecting to receive an interest rate 16.2 percentage points higher than those in the comparison group (from 29 percent), and estimated interest rate dispersion to be 15.9 percentage points wider (a 68 percent increase). While consumers with the tool did not visit or apply to more lending institutions, they were 39 percent more likely to negotiate with their lender. As a result, they obtained 13.1 percent more formal loan offers, were 4.7 percent more likely to take out a loan, and paid 11.9 percent lower interest rates.
In a follow-up study funded by IPA’s Consumer Protection Research Initiative, researchers will conduct a survey with 1,200 additional consumers to assess whether a broader population has similarly biased perceptions about the interest rate market and rate variations as those in this evaluation. Using this survey, they will also measure what fraction of consumers search for consumer loans online to determine whether scaling up the tool causes consumers more broadly to see the tool and update their beliefs.
Results will be available in 2027.
1. CMF administrative data
2. Argyle, Bronson, Taylor Nadauld, and Christopher Palmer. "Real effects of search frictions in consumer credit markets." The Review of Financial Studies 36, no. 7 (2023): 2685-2720.
Agarwal, Sumit, John Grigsby, Ali Hortaçsu, Gregor Matvos, Amit Seru, and Vincent Yao. "Searching for approval." Econometrica 92, no. 4 (2024): 1195-1231.
Galenianos, Manolis, and Alessandro Gavazza. "Regulatory interventions in consumer financial markets: The case of credit cards." Journal of the European Economic Association 20, no. 5 (2022): 1897-1932.
