For Loans, It Pays to Compare

In the economist’s eye, the simple explanation for this behavior is that many people aren’t willing to put in the time and energy to do that extra bit of research. Seeking rates at different banks can feel tedious, and comparing the APR, fine print, and hidden fees across products can be confusing. Studies have shown that consumers settle for a less-than-ideal rate because it’s easier than hunting for the best option. 

“But all of these studies assume that people know the interest rates that are generally available to them in the market,” says Sean Higgins, an associate professor of finance at Kellogg. Under that assumption, borrowers are able to weigh the benefits of searching for a better rate against the costs of doing so.  

Higgins and his colleagues Erik Berwart of Chile’s Comisión para el Mercado Financiero, Sheisha Kulkarni of the University of Virginia, and Santiago Truffa of the Universidad Adolfo Ibáñez questioned that premise. “We thought that people may not know how much they’re leaving on the table,” says Higgins. 

The researchers conducted a study in which they gave loan seekers explicit information about the rates given to other people who had similar backgrounds and circumstances as them. Seeing that information led the loan seekers to negotiate harder for better rates and walk away with more cost-effective loans. 

“The beliefs people hold about the loan market directly affect how much they search and negotiate,” Higgins says. 

Borrowers guess low 

The researchers ran their experiment in Chile, where the country’s financial regulator observes every consumer loan issued. It tracks the loan amount, rate, maturity, and lender, along with the borrower’s age, gender, income, and neighborhood at the time of application. While this highly granular data might be challenging to wade through for the average loan applicant , it allowed the researchers to uncover not just which rates exist, but which rates a specific kind of borrower could expect.  

Working with the Chilean regulator, the team recruited 112,063 people for their study, all of whom were looking into loans between November 2021 and June 2023.  

The study participants took a baseline survey on their finances and banking relationships. Three-quarters of participants took a more-extensive survey that assessed their expectations about the loan they might receive, including the rate they would get and the lowest and highest rates a bank could offer them.

Based on these initial surveys, 73.2 percent of loan seekers expected to get a lower rate than the one they ultimately received, and 74.8 percent underestimated how widely the available rate offers ranged. 

A percentage of participants were, further, randomly given access to a price-comparison tool, a different and simpler tool, or a short instructional video—and then responded to another survey about interest rates. 

The price-comparison tool showed the distribution of loan rates that borrowers like them had recently received for similar loans as well as the associated monthly and total costs in pesos. The researchers designed this tool using actual data from the 1.8 million loans granted to 1.2 million borrowers during the study period. The “simple tool” reported estimated savings that someone could get by requesting rates from one more bank. The video showed definitions of basic loan terms without providing information that would be useful for searching or negotiating for better rates. 

“The idea is that when you show actual rates that similar borrowers received for similar loans in the past, this should push people to update their beliefs on the kinds of offers they can expect and so update their beliefs on the potential benefits of search and negotiation,” Higgins says. 

As the researchers suspected, participants who used the price-comparison tool were more likely to update their initial answers about the interest rate they could get and what the spread of rate offers might look like, compared with those who used the other tool or saw the video. 

More negotiating, more shopping 

Then the researchers conducted a follow-up phone survey with a subset of 6,132 participants.  

Merely asking the participants which rates they expected to get had an effect on its own. It led them to search for loans at 3.9 percent more institutions and to obtain 7.1 percent lower rates, according to the survey data, and 1.2 percent lower rates based on a review of administrative data. 

What’s more, people who used the price-comparison tool were 38.7 percent more likely to negotiate for a better interest rate, without having to apply for loans at more institutions. These loan seekers received 13.1 percent more offers and a 3.5 percent reduction in interest rates. 

“This was an interesting contrast,” Higgins says. “On one hand, simply asking people their beliefs about interest rates on a survey caused them to search more, perhaps because they’d never really thought much about banks offering different rates. On the other hand, the effect we see on negotiation is driven by the comparison tool, which is changing what people know about the interest rates that different banks offer.” 

The price-comparison tool’s effect on people’s willingness to negotiate was especially notable among participants who had underestimated the range of loans available to them. This group was 78 percent more likely to negotiate, while those who had overestimated the range of rates didn’t negotiate more. 

The simple tool, in contrast, did not have a significant effect on people’s beliefs about loan rates and, likely as a result, had no effect on their searching or negotiating for better rates. 

The case for a public tool 

While it might seem like consumers could just get the same information that they got from the price-comparison tool from a Google search or a bank’s loan calculator, that’s usually not the case. In Chile, for example, rates advertised on Google ran below what borrowers actually received 73 percent of the time. The most popular third-party comparison platform gave similarly skewed information, with 74 percent of its quotes below the actual rate the same consumer got from the same bank.  

Banks, after all, have an incentive to depress quotes on an aggregator, then raise the actual rate once a customer applies. The rates shown on bank websites were unbiased on average but extremely noisy, landing far above or below reality. In practice, the only way to learn what a bank will charge on a loan is to apply for one. 

That underscores the potential for a policy remedy: an independent tool built on solid loan-level data that better informs borrowers about potentially available rates. If regulators required lenders to report loan terms alongside borrower characteristics, as Chile’s already does, then they could build a price-comparison tool themselves and share it with the public.  

Higgins and his coauthors are helping the Chilean government do exactly that. The politics surrounding this kind of initiative, however, are complicated.  

“Banks often push back on this type of tool, which would benefit consumers at some cost to bank profits,” Higgins says. “Based on our own experience, building this and now working to scale it up is not at all politically straightforward.”