Read on Twitter
NEW-ish Paper Alert! The Effects of Disclosure and Enforcement on Payday Lending in Texas - with Kathleen Burke
TLDR: Disclosures work, enforcement matters. Come for the paper summary, stay for the disquisition on payday lending!
Paper:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3756496#
This story starts with Bertrand + Morse, who fielded an experiment on the effects of behaviorally-motivated disclosures on payday borrowing.
Their 2011 JF paper inspired Texas and Ontario to adopt similar disclosures, an amazing example of how research can inform regulation.
Their 2011 JF paper inspired Texas and Ontario to adopt similar disclosures, an amazing example of how research can inform regulation.
We look at the Texas regulation, and find that mandatory disclosures reduce borrowing by 13%.
It was not obvious that the disclosures would still work in equilibrium, especially when payday lenders have so nimbly stayed one step ahead of regulators over the last 30 years.
It was not obvious that the disclosures would still work in equilibrium, especially when payday lenders have so nimbly stayed one step ahead of regulators over the last 30 years.
Most disclosure papers find little to no effect, so I think it's still a mystery as to when disclosures work and why. Our work shows that they can work for payday.
There's no change at all in prices, so there's also an interesting story about pass-through.
There's no change at all in prices, so there's also an interesting story about pass-through.
Per my reading at least, there aren't a huge number of papers that estimate the pass-through of financial regulation to prices. This is a critical question for whether or not regulation ultimately benefits consumers, and I don't think we fully understand when pass-through occurs.
In the second part of the paper, we look at city ordinances in Austin and Dallas. Disclosures are a controversial policy tool, because a lot of people believe they don't work, and even worse, act as a fig leaf for bad practices. So city regulators put stricter rules in place.
In particular, we look at the interplay between regulation + enforcement + compliance
Most papers only look at regulation dates and the letter of the rules, but in this case we show that enforcement really matters.
And in the end, even city regulators can make a difference!
Most papers only look at regulation dates and the letter of the rules, but in this case we show that enforcement really matters.
And in the end, even city regulators can make a difference!
Enforcement and compliance are woefully under-studied in finance. My takeaway from our work is that you have to look in the data and do your homework.
You can't just assume a rule is fully enforced immediately, or that it's fully evaded. Especially in a market like payday.
You can't just assume a rule is fully enforced immediately, or that it's fully evaded. Especially in a market like payday.
We did a lot of legwork and talking to regulators in Texas to understand what was really going on, so hardcore payday junkies can check out the footnotes for deets.
Ok, that's it for the paper summary! Stay tuned for comments below on payday regulation in general.
Ok, that's it for the paper summary! Stay tuned for comments below on payday regulation in general.
Crash course on the state of federal payday regulation
In 2016, the CFPB proposed strict rules limiting payday lending. Full disclosure: I worked on this rule!
In 2020, the CFPB rescinded the main underwriting provision of the 2017 rule
The future:
https://www.consumerfinance.gov/payday-rule/
In 2016, the CFPB proposed strict rules limiting payday lending. Full disclosure: I worked on this rule!
In 2020, the CFPB rescinded the main underwriting provision of the 2017 rule
The future:
https://www.consumerfinance.gov/payday-rule/
Payday lending is very controversial. TBH, I think we are spending too much time on this. The entire industry lends about $40 bil / year
Credit cards, auto, and student loans are > 20X larger, but get the same or less attention from regulators, advocates, and researchers alike.
Credit cards, auto, and student loans are > 20X larger, but get the same or less attention from regulators, advocates, and researchers alike.
The original 2017 rule was nearly 1700 pages long, and got over a million public comments.
But I think this may be a case where, like so many issues of our day, more attention doesn't seem to clarify the problem or solutions. We're too entrenched in partisan arguments.
But I think this may be a case where, like so many issues of our day, more attention doesn't seem to clarify the problem or solutions. We're too entrenched in partisan arguments.
Too much of the debate centers around whether payday loans are "bad" or "good."
The academic literature has been stubbornly mixed on overall welfare effects. Here's the brief lit review from our Texas paper (sorry if we missed anything!).
The academic literature has been stubbornly mixed on overall welfare effects. Here's the brief lit review from our Texas paper (sorry if we missed anything!).
I have a hunch that taking a closer look at the actual rules and evasion, compliance, and enforcement across states might help reconcile some of these estimates, but they are also local to different outcome measures and contexts.
But also ... maybe it doesn't actually matter!
But also ... maybe it doesn't actually matter!
The canonical use case for a payday loan is that a family is going along just fine. A sudden emergency occurs, and the loan saves them from disaster. They gladly repay with interest using regular income.
This is not the scenario for the vast majority of loans.
This is not the scenario for the vast majority of loans.
Numerous studies show that about half of Americans lack a basic $400 buffer stock of savings. While payday borrowing tends to occur when households are particularly constrained, proceeds are mostly used for cash management and everyday expenses.
https://www.federalreserve.gov/publications/files/2018-report-economic-well-being-us-households-201905.pdf
https://www.federalreserve.gov/publications/files/2018-report-economic-well-being-us-households-201905.pdf
In this scenario, loans alleviate a short-run constraint but reduce consumption and tighten constraints even further when they have to be repaid. A payday loan, or ANY interest-bearing loan even at credit card interest rates, is just not a helpful solution to lack of savings.
Americans need buffer stocks, not loans. Payday loans may help in some cases, but often they just tighten the noose. But importantly, banning these loans doesn't help much either.
We're lending people whiskey when what they need is kale, and the policy discussion is too silo'ed.
We're lending people whiskey when what they need is kale, and the policy discussion is too silo'ed.
Abstracting from whether a loan of any kind is actually useful at all, too much of the policy debate is around whether payday loans should be allowed as is, or banned completely. Very little is about how to make the product better.
I think there is some low-hanging fruit here.
I think there is some low-hanging fruit here.
To me, the key failure of payday loans is not the price (per se), or the lack of underwriting (which has always been an awkward argument).
It's that the standard finance charge is interest-only. Consumer loans should always amortize. Period.
It's that the standard finance charge is interest-only. Consumer loans should always amortize. Period.
This idea builds on pioneering work by Madrian, Beshears, Choi, Laibson, and coauthors, showing that retirement saving is dramatically impacted by seemingly irrelevant factors such as the default contribution rates.
A growing literature on consumer debt shows similar effects. Consumers tend to pay the minimum payment on credit cards (Keys & Wang), take out the maximum amount of student loans (is there a paper on this?), and pass through housing wealth to total savings (Bernstein & Koudijs).
In our 2014 CFPB Data Point, we showed that like credit cards, payday loans tend to roll over indefinitely with interest-only payments. Even though borrowers almost always have the option to pay down principal, they rarely do.
https://files.consumerfinance.gov/f/201403_cfpb_report_payday-lending.pdf
https://files.consumerfinance.gov/f/201403_cfpb_report_payday-lending.pdf
This is the actual solution to the "debt trap" problem. Given that payday loans are only $300-500, it would only take an extra $50-100 every pay period, instead of a potentially unaffordable balloon payment, to pay down the debt pretty quickly.
To sum up: 1) we need to address the liquidity problems of American households more holistically, and focus on building savings instead of fighting about a relatively small debt market 2) conditional on taking out debt, contracts should be designed to amortize. The end.
