BNPL contains multitudes
Daily Data: what does BNPL say about the health of the consumer and/or the health of consumer fintech
If BNPL grows too quickly, it might be bad. If it grows too slowly, it also might be bad. So how is it growing?
👉👉👉Reminder to sign up for the Weekly Recap only, if daily emails is too much.
BNPL contains multitudes
If BNPL grows too quickly, what does it say about consumers?
There has been lots of speculation about the role that innovative credit products, like Buy-Now-Pay-Later (BNPL), may or may not be playing in the consumer economy.
It started around the holiday season, when there was some evidence that BNPL was extremely popular among holiday shoppers, in what was otherwise a booming holiday shopping season (bookended by somewhat muted shopping periods).
Was BNPL driving a consumer binge?
Were people turning to BNPL because regular credit card companies wouldn’t take them?
Was BNPL enabling stretched consumers to spend beyond their means and/or hiding stress and defaults, perhaps because BNPL companies themselves are under pressure to generate revenue?1
BNPL is growing? Is that bad?!
These are all valid questions, especially because BNPL leaves plenty of room for speculation.
Compared to traditional credit and debit payment options, BNPL is a relative blindspot for the data junkies.2
So while it’s somewhat harder to speculate about what consumers are doing with their credit cards, it’s very easy to speculate about what they’re doing with their BNPLs.
If BNPL isn’t growing quickly enough, what does that say about investors?
Now, here’s something unfair: it’s also the case that if BNPL wasn’t growing, that would also be bad. (Maybe) bad if it grows, (maybe) bad if it doesn’t grow.
Well, the premise (and promise) of these companies is that technological innovation would unlock lending where lending should exist, but didn’t.
From a pure credit standpoint, lenders would have no issue financing an $120 trip to the grocery store, but since they can’t put a loan officer at every checkout counter (at a cost that would make an $120 loan worth their while), they don’t.
Enter BNPL, who says “we can put a loan officer at every checkout counter at a cost that makes $120 loans worth our while.”
And voila, by reducing transaction costs through technology, productivity is unlocked.
It’s the sort of thing that investors get very excited about, but to validate their excitement, BNPL have to actually make good on that promise. Not all at once, of course, but if the claim is “there is land to conquer,” then they need to be conquering.
Otherwise, BNPL is just a niche cool idea that’s marginally better than a credit card, and not all that exciting.3
I mean, look at this comparison of incumbent fin services revenues to disruptive revenues (and just take it as face value):
Fintech revenues are barely a rounding error!
You could interpret this one of two ways: (a) fintechs are overhyped (booo!); or (b) fintechs still have so much more land to conquer (yaayy!).
In order for the yays to preside, fintechs (like BNPL) need to grow and grow without making a lot of bad loans. Investors can be demanding, from time to time.
If it turns out that BNPL companies cannot not grow in a higher rate environment, then it would be an “uh oh” moment, and more false promises from an overhyped tech ecosystem (again, so the story would go).4
Basically, there is plenty of opportunity to be negative, if you want to be.
Narrative is unfair like that sometimes.
It says everything is awesome
You weren’t expecting that, were you?
But it’s true. On both scores. Growth and credit quality.
Affirm is definitely growing, and somewhat remarkably (to me), their delinquency rates are not only fine, they’re getting better:
While everyone else is seeing rising (aka “normalizing”) delinquencies, Affirm is flat-to-down.5
It’s just one company and just one quarter, but that’s pretty encouraging news.6
If the good-times story is that (a) BNPL is going to unlock a whole lot of quality lending, where lending previously dare not go; and (b) is at the precipice of a fresh run of rising incomes and consumer spending, that specifically transpires on the internet more than ever, then it’s precisely what you’d like to see.7
Score one for the optimists.
Previously, on Random Walk
Random Walk is an idea company dedicated to the discovery of idea alpha. Find differentiated data, perspectives and people, and keep your information mix lively. A foolish consistency is the hobgoblin of small minds. Fight the Great Idea Stagnation. Join Random Walk. Follow me on twitter. Follow me on substack:
As Random Walk has previously observed, this is also the flipside of banking the “underbanked.” One reason that many people are underbanked is because they represent poor credit risks, such that banking them leads to losses. When there are low losses, but high underbanked, lenders are accused of not banking enough. They need to be more inclusive. When lenders bank more, and then losses get high, lenders are accused to banking too much (i.e. being predatory). Predation is often just the mean word for inclusion, that people use after-the-fact.
Why is it a blindspot?
With regular credit card purchases, it’s easy to divine aggregated consumer behavior, at very low latency—within a couple of days, we know generally how much people are spending at different merchants. That’s because credit card purchases are recorded with both the merchant name and the transaction size. So, if someone spends $112.46 at Walmart, then it shows up pretty clearly in the data.
BNPL purchases, on the other hand, generally exclude both merchant name and transaction size. If someone uses Affirm to spend $250.16 at Walmart, the data will show Affirm as the merchant, and the transaction size will be whatever fraction of the total the customer decided to “buy now” (as opposed to pay later, which will show up on subsequent statements). In that case, only Affirm (and the Merchant) know how much money was actually spent and where.
Finally, BNPL gets to be much more selective about reporting credit performance. While credit cards regularly report the bureaus, BNPL basically report whatever it is they feel compelled to report when earnings roll around.
In fairness, the total volume of consumer transactions is so massive, that even being marginally better than a credit card, can get you places. You still have to grow, however.
I mean, you could also tip your cap to prudence and care, and investors over the longer haul might do that, but if you’re trading quarters, you want to have it all.
Credit quality is improving, even while interest income increases substantially, i.e. they are charging their customers more, and their customers haven’t broken a sweat:
Another fintech lender, Upstart, reports later this week. I’m curious to see how their experience has been.
There’s some other questions that someone with more knowledge about the company might have, like “why are pay-in-three” loans excluded? Why does it look like new customer growth is negative (and all the growth came from existing customers)?
But, Random Walk is being optimistic here.