The Pre-Selected Option: The Most Consequential Unregulated Variable

Consumer financial regulation is overwhelmingly about disclosure — what must be told, in what form, at what moment. Almost none of it governs what happens when the customer does nothing, which is what most customers do most of the time. That leaves the variable with the largest effect on outcomes substantially outside the framework built to protect people from bad outcomes. And it isn't neutral across a customer base: overriding a setting costs time, attention, and confidence, all unevenly distributed — so a default operates as a decision imposed most firmly on the people least able to escape it.

Where defaults decide

Setting	Typical default	What it determines
Card payment amount	The minimum	Total interest paid, by a lot
Overdraft coverage	Varies, frequently on	Whether a purchase costs a fee
Subscription renewal	Automatic	Whether you keep paying
Statement delivery	Electronic	Whether you see it at all
Payment due date	Set by the lender	Whether it lands before payday
Data sharing	Frequently on	Who gets your information
Line increases	Automatic where permitted	Your available exposure
Credit insurance add-ons	Varies	Cost of the product

Row one is the largest single effect in consumer credit. Per our minimum payment analysis, the difference between paying the pre-filled minimum and a modestly higher figure is measured in years and in multiples of the balance — and it's presented as the standard option on every statement.

Row five is the cheapest to fix and least addressed. Per our timing analysis, a due date that falls before income arrives produces late payments from households that can afford the obligation — and the date was assigned by a system, not chosen by anyone.

The variable with the largest effect on outcomes is the one nobody had to justify.

Why they stick

Not inattention. Four separate barriers, each sufficient on its own.

1. Knowing the setting exists. Most people don't know a due date is changeable or that overdraft coverage was an election.
2. Knowing what the alternative does. Per our complexity analysis, evaluating an option requires understanding a product structure.
3. Having time to act. Per our time analysis, changing a setting frequently means a call during business hours.
4. Confidence that changing is right. The pre-selected option carries an implicit recommendation, and departing from it feels like assuming risk.

The fourth is the one that gets least attention and does real work. A standard setting reads as what the institution thinks you should do — so overriding it requires believing you know better than the provider, which most people reasonably don't.

Which means the common framing — that people are lazy or inattentive — is wrong in a consequential way. The barriers are informational and practical rather than motivational, and that changes what would fix them.

The regulatory gap

The structural observation, and it's striking once stated.

	Heavily governed	Largely not
What must be disclosed	Yes	—
How it must be presented	Substantially	—
The substance of some terms	In places	—
What is pre-selected	Scattered rules only	Mostly
How hard changing is	Rarely	Almost entirely

The bottom two rows are where outcomes are actually determined, and they sit largely outside the framework. There are exceptions — specific rules addressing specific settings exist, and some jurisdictions have moved on particular defaults — but they're targeted interventions rather than a general principle.

Why the gap exists is understandable. Disclosure regulation rests on a model where an informed consumer chooses, and within that model the starting position is incidental. Per our consent analysis, that model has been under strain for other reasons too.

The practical consequence: a provider can comply completely with every disclosure requirement while the pre-selection does the work. The minimum payment is disclosed accurately, the alternative is available, and the pre-filled figure determines the outcome — and nothing in the compliance framework asks why that figure rather than another.

Who overrides

The distributional finding, and it's the part that makes this more than a design observation.

All four barriers above are unevenly distributed.

• Knowing a setting exists correlates with financial familiarity and with having dealt with the problem before.
• Understanding alternatives correlates with education and with product experience.
• Time to act correlates with schedule control, per our time analysis — the most unequally distributed resource in the whole system.
• Confidence to depart from the standard correlates with everything above.

Which produces the finding: a default is a decision that applies most firmly to the people least equipped to escape it. Someone with slack changes the due date, turns off overdraft coverage, and pays more than the minimum. Someone without slack accepts all three — and the three together are worth a substantial amount of money over a few years.

Per our monoculture analysis and our place analysis, this is the third mechanism in this library where a neutral-looking design decision sorts on capacity rather than on need — and it's the same population each time.

Why take-up is the wrong metric

The measurement point, and it explains why organizations don't notice the problem.

A firm reviewing a default typically asks what share of customers changed it. Low change rates are read as evidence the default suits people.

That inference doesn't hold:

• Non-change is consistent with any of the four barriers and with genuine preference, and the metric can't distinguish them.
• Per our measurement analysis, what the alternative would have produced is unobserved — the counterfactual isn't in any dataset.
• Satisfaction surveys miss it, because people don't complain about a setting they don't know exists.

What would actually test it:

1. Randomize the default itself for new customers, and compare outcomes — not take-up.
2. Compare outcomes for those who changed against those who didn't, controlling for what you can.
3. Test an active-choice version where neither option is pre-selected.
4. Ask a sample whether they knew the setting existed, which is cheap and frequently decisive.

Item four is the one to run first. If most customers don't know a setting exists, the low change rate carries no information about preference at all — and that question can be answered in a week.

Defaults that work well

The constructive side, because the argument is for deliberate defaults rather than against defaults.

A default is unavoidable — something has to happen when nobody chooses — and a well-chosen one is genuinely protective, because it delivers a good outcome without requiring the time, knowledge, and confidence that most people don't have.

What distinguishes a good one:

• It's what an informed customer would mostly choose.
• It fails safe — the cost of being wrongly defaulted is small and reversible.
• It's stated, so the customer knows a choice was made for them.
• Changing it is easy, in the channel the customer already uses.
• It doesn't favour the provider where the interests diverge.

Examples where a default is doing protective work:

• Automatic payment of the full statement balance, rather than the minimum.
• Overdraft coverage off for card purchases, per our overdraft guide.
• Low balance alerts on rather than available.
• A due date set from observed income timing rather than from the origination date.
• Data sharing off, with an easy opt-in.

The fourth is available to any lender that sees transaction data and essentially nobody does it. Setting a due date a few days after income reliably arrives would eliminate a category of delinquency that has nothing to do with capacity — and per our cure analysis, that category is substantial.

A test for any default

Practical, and applicable in a product review:

1. Name the default explicitly. Many are inherited and nobody chose them.
2. State who benefits when the customer doesn't change it.
3. Ask whether you'd state the rationale publicly.
4. Establish what share of customers know the setting exists.
5. Estimate the cost of being wrongly defaulted, and whether it's reversible.
6. Check how hard changing actually is, by doing it.
7. Consider active choice where the options genuinely differ in cost or risk.

Step three is the sharp one. Where the honest answer is that the default performs better commercially, that's a finding rather than a justification — and it's the point at which a firm should decide deliberately whether it's comfortable, rather than discovering the answer later from someone else.

And step seven deserves more use than it gets. Active choice — requiring a selection with nothing pre-ticked — sidesteps the whole problem where the options materially differ, and it's defensible in a way that any pre-selection isn't.

The strongest objections

"Someone has to choose a default." Entirely agreed, and the report says so. The argument is that the choice should be deliberate, stated, and reviewed against outcomes — not that defaults should be abolished, which is impossible.

"Regulating defaults means regulating product design." True, and it's a real cost. Design regulation is harder to write, easier to arbitrage per our definition analysis, and risks freezing products. The narrower version — requiring active choice where options differ materially in cost — avoids most of that and is a much smaller intervention than governing what the default must be.

"You've offered no evidence of magnitude." Correct and it's the central limitation. The mechanisms are well established in the behavioural literature; what isn't established here is how much money is involved in consumer finance specifically, and the report offers no estimate. The randomization tests below are the way to find out, and a single lender could run them.

Testable implications

1. Randomizing a default should change outcomes substantially, more than any disclosure change does — the direct test.
2. Override rates should correlate with schedule flexibility and product familiarity, not with how much the change is worth.
3. Most customers should be unaware that most settings exist, measurable in a survey.
4. Active-choice framing should shift selections materially relative to either pre-selection.
5. Due dates aligned to observed income should reduce delinquency without any change in credit quality.
6. Defaults favouring the provider should show lower override rates than neutral ones, if friction is doing the work.

The fifth is the one worth running first, because it's cheap, it's clearly beneficial to both parties, and it's a clean test. Randomize whether a cohort's due date is set from origination date or from observed income timing, and compare delinquency. If alignment reduces missed payments, a share of the delinquency in every consumer portfolio is an artifact of a setting nobody chose.

The conclusion we'd hold: the framework built to protect consumers governs what they're told and leaves what happens by default largely alone, which puts the most consequential variable outside supervision. The fix isn't prescribing defaults centrally — it's requiring that they be chosen deliberately, stated plainly, and defended where they favour the provider.

Frequently asked questions

---

This report presents an analytical framework and the authors' interpretation; it is not financial, legal, or policy advice. Regulatory treatment of specific settings varies by product and jurisdiction and changes over time, and no magnitude is estimated here for any effect described; the implications identified as testable are hypotheses.