Buy now, stress later: The real cost of BNPL

 

IN parliamentary replies reported in early 2026, Malaysia’s buy now, pay later (BNPL) balances stood at RM4.9 bil as of Dec 31, 2025—about 0.3% of total household debt—with 7.5 million users and an overdue ratio of around 3.3%.

Strait of Hormuz

The same reports noted that BNPL usage is heavily concentrated among youth and lower-income groups, with modest ticket sizes often used for essentials such as food, groceries, transport and services.

These figures are often used to reassure the public that BNPL remains “contained”, which is not entirely inaccurate from a macro perspective.

The more important question, however, is not whether BNPL is large enough to threaten the financial system, but whether it is becoming a default coping mechanism for cashflow stress.

That is where long-term consumer harm accumulates quietly, even when national numbers appear small.

The metric that hides the real problem

(Image: NST)

BNPL is often compared to credit card debt, but this misses a crucial difference in how the product is experienced. Credit cards are typically applied for deliberately and then used repeatedly.

BNPL, by contrast, is embedded at checkout, presented at the moment when attention is on the product price rather than the cost of credit. The framing shifts from “borrowing” to “splitting”, making the decision feel smaller than it really is.

This is why BNPL can remain “small” in outstanding balances while still shaping household behaviour.

A system can generate millions of micro-instalment transactions tied to everyday consumption without producing a headline figure that triggers concern.

The risk lies less in total debt and more in fragility, which emerges when many small obligations accumulate across providers while the borrower experiences them as separate, manageable commitments.

When used mainly for discretionary purchases, BNPL functions as a convenience. When used for essentials, it begins to resemble salary smoothing, especially for workers whose expenses do not align with income timing and who have limited financial buffers.

In such cases, BNPL becomes an informal credit layer compensating for income volatility and rising living costs, normalising short-term borrowing for routine needs.

Why “stacking” is where harm concentrates

The most under-discussed risk in BNPL is “stacking”, where consumers hold multiple BNPL accounts across providers and merchants. Each appears small in isolation but becomes burdensome in aggregate when repayment dates cluster.

This is less a moral issue than a design and coordination challenge, as BNPL prioritises ease of approval while household budgets follow pay cycles, not instalment schedules.

This also explains why overdue ratios can remain modest overall while a segment of users experiences repayment stress. The problem concentrates among borrowers who are less visible to any single provider’s risk model.

The early warning signal is not total debt, but rising “instalment congestion”, where multiple obligations collide with rent, utilities and transport costs within the same period.

Malaysia is already moving BNPL providers towards stronger governance under the Consumer Credit Act 2025, including licensing requirements and responsible lending duties. This provides an important foundation.

The real test, however, is whether firms treat responsible lending as a compliance exercise or as a product design discipline that reduces fragility while preserving legitimate use cases.

Three market commitments would meaningfully improve outcomes:

First, BNPL providers should adopt an affordability-by-design approach rather than reacting after defaults occur. Limits and approvals should function as dynamic tools that respond to early stress signals, not as levers for growth.

Clear thresholds should guide when repeated late payments trigger cooling-off periods, when limit increases are paused, and when repayment schedules are adjusted to reduce clustering.

If BNPL is increasingly used for essentials, product design must prioritise repayment resilience, not just conversion rates.

Second, marketplaces and large merchants should treat BNPL placement as a consumer trust decision, not a revenue optimisation tool.

Checkout design should clearly present total repayment amounts, instalment schedules and due dates, while avoiding urgency cues that encourage reflexive use.

BNPL should not be pre-selected by default, and language should not imply “free money”, as fees and penalties may apply. Short-term gains from frictionless sign-ups can quickly erode trust if repayment experiences become difficult.

Third, the payments ecosystem should establish shared visibility standards to make stacking easier to detect and manage. Banks, e-wallets, payment gateways and credit bureaus should treat BNPL visibility as a consumer well-being feature.

(Image: CNBC)

An industry-standard dashboard showing upcoming instalments across providers, paired with consent-based data sharing, would help users see and manage obligations in aggregate. Fragmentation is precisely how small credit becomes quietly dangerous.

BNPL can be a useful financial tool when designed for resilience and transparency, functioning as a controlled budgeting aid rather than a substitute for income adequacy.

Malaysia’s BNPL debate should move beyond reassuring aggregate figures and focus on behavioural reality. A product embedded in checkout flows shapes spending decisions, and when used for essentials, reflects underlying household stress even when macro indicators appear stable.

If BNPL evolves into a form of engineered wage smoothing, it must also be engineered for stability.

Otherwise, the real cost will emerge later in the form of distrust, disputes and financial fatigue among the very consumers the digital economy depends on. 

Inv. Galvin Lee Kuan Sian FRSA is a PhD Researcher in Marketing at the Asia-Europe Institute, Universiti Malaya, and serves as Lecturer of Marketing and Economics & Programme Coordinator in Business at a Private College in Malaysia, specialising in the scholarship of teaching and learning via educational technology. 

The views expressed are solely of the author and do not necessarily reflect those of  MMKtT.

- Focus Malaysia.