The Broken Business of Repayments (#48)
Skewed collection incentives, multiple loans, and a passive business approach stymie repayments
Welcome to the 48th issue of Unit Economics. In today’s write-up, we dive into the less-talked-about problem of repayments in the Fintech industry. Dive in!
Once you start lending, it only gets tougher. It gets even tougher when you have to ask for the money back. First, you ask politely. No response? You get more people to ask with politeness. No response, still? You change your tone, and sometimes even issue soft threats. And if you still don’t get your money, you take it on the chin and move on.
But you can take it on the chin only so many times. If you consistently fail to collect the money you lend, you can’t lend for long. So, for a business – whatever may be your lending targets, collections keep you in the game. If you don’t collect well, the business bleeds until it faints.
It’s not really a game of skewed incentives as well despite how it would seem. The borrowers don’t have much to gain from defaulting. In fact, paying late or not paying at all only impacts their financial prospects. Some of the big personal and commercial decisions are made poorer with even the smallest unintentional mistakes of borrowers. House loans can get costlier, for example, and some other doors get completely shut down. Especially when all that you are judged by is a number and a sheet with your history of financial behavior. So, for borrowers too – unless not possible, paying back on time is always the optimal strategy.
But if the incentives on collections largely align for both lenders and borrowers, then why is that the lenders struggle to collect? Why is the 90 and 180-day portfolio at risk in India (PAR) as high as 3-6% for credit cards? and close to 10% for small-ticket loans?
I figure that we need to dive a little into the psyche on both sides to come out with answers here.
The Fintech Approach to Collections
I had written earlier about the conflict between product and underwriting teams that underlie the many lending decisions. But even the underwriting team can find solace when not asked to work on collections. I say this because of how collections are approached in the culture of ambitious disbursal targets.
The business team sets goals for how much to disburse first, drives product and risk policy to achieve it next, and approaches the collections at the last. Targets are set for collections, but the function is projected as ancillary and remarked rarely during product and risk policy decisions.
When collections miss the targets and the customer loans start turning up 0+ DPD, the focus slowly shifts. But the solution, industry-wide, stands largely reactive. The collections team resort to intrusive collection agents, IVRs, consistent messaging over SMSes, push notifications, e-mails, or even subtle threats to collect whatever possible. And the firefighting continues from one billing cycle to another. These strategies, of course, do their job of optimising collections before writing off part of the portfolio. But improvements in collection tactics make for small optimisations in the proportion of portfolio at risk. The much bigger issue here is instead the misalignment of product and underwriting design with collections.
This still does not explain why borrowers, who are financially capable of paying on time, pay late or default despite all their incentives suggesting otherwise.
The business models partly help explain the issue, and the rest are hypotheses drawn from the industry structure.
Why do we pay late or default at all?
If we were to explain the behavior through regression, there are likely tens of independent variables that we would find with high confidence. Similar studies, as you would guess, have been done before to some extent (one, two).
But in this write-up, I will instead choose the easier path and rely on a few hypotheses.
Lenders want to collect, but not on time
Credit cards and small-ticket loans (including BNPL, EMIs) usually come with an interest-free period and no-fees for timely repayments. But how do they make money on these loans then? Transaction fees or interchange is one source, but it doesn’t earn the business enough to cover the costs of lending and maintaining the business.
Instead, the model incentivises that the borrowers bounce. That is, the lenders benefit if the borrowers revolve and incur the late payment charges. These charges are hoped to be sufficient to subsidise other costs – and especially the cost of defaults. Now, if the business itself is dependent on borrowers paying later than the due date, then the product is built accordingly. Few nudges are added to enforce timely payments, and more focus is kept on maximising the loan recovery post the due date than before.
This business model brings about a chicken and egg problem. If the focus is on collecting past the due date, then the defaults are eventually higher due to the higher bounce rate. But if the nudges are enforcing timely collections, then the lender may not make enough to sustain their fee-free products.
Multiple, smaller credit products make it tough to track and prioritise repayments
What compounds the issues for borrowers is how easy it can seem to get credit today. The shifting focus towards NTC users and the high disbursal targets have gradually improved the underwriting rates in the industry – and borrowers who seek credit would find it tough to resist the pay-later, prepaid cards, and the credit cards on offer in the market.
What does this mean? An average borrower today has a higher number of credit products, and it can be tough to remember the dues and billing cycles for all – especially when the lenders don’t have the incentive to remind you of them. This is a problem similar to that of subscription management, but with much higher consequences.
If borrowers have a higher number of smaller-ticket size loans today, they are also likely to put lower importance on repayments for each than they would for one large loan. This closely ties up to another industry-wide issue.
Borrowers do not understand the long-term impact or costs of late repayments
For a set of users, it appears more attractive to default and to gain from free credit than to (1) pay high late payment charges or to (2) see their credit history being impacted. Both reasons come down to a high level of information asymmetry in the market.
Not many users understand the 3-4% monthly interest charges on outstanding dues and fewer realize the long-term consequences of poor repayment history. Details are accessible but kept almost hidden in the fine print. And as we know, there are incentives for lenders of interest-free products to keep it that way.
So, we end up with an industry where (1) the fintech companies approach collections passively, (2) incentives are steeped in to collect past due dates and to keep the revenues high from late payment charges, where (3) borrowers struggle to track and prioritise multiple loans and (4) to understand the consequences of defaulting or paying late.
How does the industry then solve the breaks in the repayments system? We can try to draw a few solutions by directly addressing the problem statements.
Repayments should not be an afterthought. The collections projections and preparation should be done at the point of loan origination, and not post-facto. For example, during onboarding, the lender can choose to seek information (E.g. alternate mobile number, mandate registration) to improve chances of loan recovery. In-app nudges can be built to further remind users of the due dates and costs of late repayments to seek early recovery. At a business level, however, this would mean closer interactions between the product and collections teams – something that does not happen as often in the industry.
The two objectives of high bounce rates and low loss rates are at crossroads with each other. Instead, stronger business decisions need to be taken to imagine business models where paying on time is encouraged and rewarded, and where late payment charges are not relied on to keep the business alive. Some of the BNPL businesses have succeeded in doing so by filling the gap through merchant revenue, but this model is still not as common across small-ticket loans. One way to encounter this problem can be to build the lending business on subscription premiums instead. Especially if you are incurring the risk and costs of lending, you should be able to charge customers for it. But then, you might have to pipe down your disbursal targets and chase a few other metrics instead.
Third, for the borrowers to seep over the problem of tracking and prioritizing repayments, the easiest solution would be to cut down on your credit mix. Focus on borrowing only when you can afford credit plus ones that serve an obvious need, not because the lender promises you registration and bank credit in two minutes.
Lastly, the information asymmetry on the consequences of late payments and defaults is much easier to overcome. This can be achieved through an industry-wide mandate/initiative to explicitly inform the user of (1) the credit bureau(s) being reported to, (2) the calculation of late payment charges, and (3) detailed impact on the user’s credit profile of late repayment or defaulting. The onus here lies on the regulatory bodies and on the business teams driving these lending products. You wonder that if subscriptions or other revenue sources could be found, the decision to distribute the above information to the user would be much easier to make.
Lenders have drastically improved the loan origination journey for borrowers in the last few years by taking the process completely online, and by focusing on minimising activation time. If the repayments are to appear as user-friendly, the next few years need an equal focus from the lenders on the user journey for collections. There is promise, no doubt. With the fintech industry fast moving towards greater lending dependence, there is an opportunity for many to question the current practices and to redefine repayments.
If you have any views or feedback to share on the topic, feel free to add a response below or to share your thoughts with me over Linkedin. In case you feel your friends or family would be interested in reading about payments, feel free to share the blog with them as well. See you in a couple of weeks!