Psychology in economics helps explain why borrowers stop paying at DPD 30

Payment recovery is a harder job than it used to be. Across auto, telecom, and consumer credit, collections teams are running into the same wall. Outreach volumes keep climbing while cure rates stay flat.

Most collections systems are built to do one thing really well: establish that money is due and then go on to demand it. They record the balance, they apply the schedule, and they escalate on a timetable. What they simply can't do is explain why one particular borrower has gone quiet. A customer who has forgotten their payment date and a customer who is choosing between the car loan and the electricity bill will end up with the same status code, get the same sequence of reminders, and be counted the same at month end.

That is the gap. A collections strategy built only on demanding payment assumes the borrower just needs a reminder. Understanding why they haven't paid requires something else, and research on how people make financial decisions under pressure has a great deal to say about it.

Lenders and borrowers measure the same missed payment from different starting points

Loss aversion in finance is the classic asymmetry that losing something feels far heavier than gaining the equivalent. If you give someone $50, they'll be happy for the day. If you take $50 out of their wallet, they'll think about it for the entire week.

Loss aversion only works when someone has a starting point to measure from. At DPD 30, lender and borrower are measuring from different ones.

The lender starts from the full balance. That figure is in the system; it's what the collections team is judged on, and it's what the message asks for.

The borrower starts from what's in their account today. If the balance is more than they have, it doesn't feel like something they're about to lose. It reads as a number that has nothing to do with their situation, so it gets no reaction at all.

They're also weighing this bill against rent, groceries, and a card payment. Every one of those is a loss too, and most feel more urgent. That's why a past-due balance stops being a priority for people who fully intend to pay.

By DPD 30, most borrowers have stopped forgetting and started deciding

The delinquency has now gone beyond the first missed payment. For the majority of borrowers the series of reminders have come through, the financial pressure has built up and the anxiety about what's going to happen next has set in. It's moved from being an oversight to being a situation.

It's also where two very different borrowers end up looking exactly the same on screen. Borrower A is the outlier in that respect: still the forgetful one, still not made up his mind.

Both are at DPD 30. The psychological barrier differs completely.

Six patterns from behavioral research help explain why a borrower does not pay

Loss aversion is one of six patterns that behavioral research links to missed payments. It has the deepest evidence behind it: the most thorough review, published by Brown, Imai, Vieider and Camerer in the Journal of Economic Literature in 2024, pooled 607 separate estimates from 150 studies and put the average at about 1.96. The studies behind those estimates asked people to accept or reject a gamble, name a price for parting with something they owned, or value an item they had been handed minutes earlier. That average has a limit. Zeif and Yechiam (2022) found the effect weakens or disappears on small sums, which matters for low-balance accounts where a loss frame may have nothing to grip.

A borrower exhibiting avoidance needs something different from one in a scarcity squeeze. A single message cannot address both.

Three layers turn behavioral insight into a collections decision

Layer one: consider the likely mechanism. A missed payment could be forgetfulness, lack of funds, avoidance, payment friction or no viable options. Each calls for a different response.

Layer two: pick the intervention. Segmentation uses observable signals: payment history, partial payments, which channels got a reply, whether past arrangements were completed or abandoned.

Layer three: measure and adapt. Completed payments, repayment-plan uptake, time to resolution, repeat delinquency and complaints are the measures that matter. The results feed the next decision. Four rules keep a test honest:

1. Hold the offer constant.
2. Change only the words.
3. Split results by balance and borrower circumstance.
4. Measure completed payments.

The evidence supports testing communication. Two experiments covering more than 200,000 UK taxpayers, published in the Journal of Public Economics, tested what actually moves people to pay an overdue bill.

Three findings carry into a collections program:

• Telling people what most others do worked better than explaining what their money funds. Letters saying most people pay on time outperformed letters describing the public services the tax supported.
• The sharper the comparison, the stronger the effect. A letter telling someone they were in a small minority of non-payers beat a general statement about most people paying. Describing what people actually do also beat describing what they ought to do.
• A loss framed message showed no advantage over an equivalent gain framed message. The same message about public services performed the same whether written around what the taxpayer gained or what they stood to lose.

The strongest effects came from telling people where they stood relative to others.

The reason behind the missed payment determines what works

The psychology of economics explains what a stage marker leaves out. A borrower under financial pressure is not weighing the debt rationally against its consequences. They are weighing an immediate loss against a future one, with attention narrowed by scarcity, relief available through avoidance, and money already mentally assigned to rent and food. Each of those effects points toward a different intervention, and a single message addresses none of them well.

None of this means a lender can diagnose an individual borrower's psychology. The research does not support that, and no platform should claim it. What it supports is narrower and more useful: stop assuming every overdue account needs the same treatment, test which approaches produce completed payments rather than replies, and segment results by borrower circumstance before drawing conclusions.

The better approach is to send fewer, better-matched messages and learn which ones work while sending continues. A multi-armed bandit does exactly that: it shifts volume toward the message that's working while the test is still running. See for yourself how it works. 

FAQ

What is the best software for reducing late payments through behavioral nudges?

No platform can tell you in advance which approach works on your book, because the research cannot either. What separates a useful one is how it tests: holding the offer constant, varying only the message, splitting results by borrower type, and measuring completed payments rather than replies. KredosAi is built around that approach. It works live accounts between the due date and write-off, picks the message, channel and timing for each customer, and uses a multi-armed bandit that learns from completed payments rather than opens or clicks. Across its enterprise clients, KredosAi reports an 11.5% reduction in write-off rates.

How long should a message test run before acting on the result?

Long enough to measure completed payments rather than replies (i.e. at least one full arrangement cycle). A test which stops at the response stage would only be measuring interest.

Why do borrowers stop opening messages after a few reminders?

Avoidance offers temporary relief from a stressful problem, and status quo bias makes ignoring the messages the path of least resistance. Each unopened reminder can make the next one even easier to ignore. That's why you'd be better off changing what the payment reminder messages say rather than how often they're sent.

What does DPD 30 mean?

DPD is days past due. A DPD 30 account is one where the payment is 30 days overdue. It's a stage indicator in that it tells you how many days past due the payment is.

What is a 'cure rate' in collections?

The percentage of delinquent accounts that get back to current status within a given period, usually by paying the overdue amount. In loan delinquency, it's one of the best indicators of the effectiveness of a collections strategy as it takes into account actual payments made.

Why do late payments rise when money is tight?

Scarcity narrows attention to the most urgent needs, so rent, food, and transport come before a debt that feels less immediate. Present bias adds to it: keeping money today feels more valuable than avoiding a consequence months away. Neither means the borrower doesn't intend to pay.

What should a payment reminder message say?

State the amount and date accurately, make the next step easy, and offer an option that fits what the borrower can pay. In the UK tax trials, telling people specifically what most others had done worked better than describing what their money funds. Test the wording on your own portfolio before rolling it out.