‘Early warning signs of credit stress emerging’

Upgrade to High-Speed Internet for only ₱1499/month!

Enjoy up to 100 Mbps fiber broadband, perfect for browsing, streaming, and gaming.

Visit Suniway.ph to learn

April 20, 2026 | 12:00am

As oil shock builds

MANILA, Philippines — Early signs of financial strain are beginning to surface among Filipino borrowers as rising oil prices push up daily expenses, even as headline default rates remain stable, according to fintech firm Advance.AI.

Michelle Anne Chan, country manager of  Advance.AI, said lenders are already seeing behavioral shifts that typically precede broader credit stress, signaling that risks may materialize in the coming months.

“Repayments remain broadly stable; default rates unchanged,” Chan told The STAR, noting that these are lagging indicators and that stress usually surfaces “60 to 90 days post-shock.”

However, she emphasized that pressure is already building beneath the surface as higher fuel costs ripple through transport, airfare and daily household spending.

Chan said the clearest warning signs are emerging in how borrowers manage their finances, particularly in payment behavior and borrowing patterns.

These include an expected five to 10 percent increase in customers shifting from full repayments to minimum payments, as well as a rise in “loan stacking,” or borrowers accessing multiple lenders within short timeframes.

Applications for small-ticket loans, buy now, pay later (BNPL) and salary-based lending are also projected to grow by 10 to 20 percent over the next two to three months.

At the same time, digital transaction activity is beginning to decline, especially in fuel-sensitive segments, which Chan described as an early proxy for tightening household budgets.

“Credit stress begins with behavioral shifts, not defaults,” she said, adding that traditional credit metrics may lag the actual emergence of risk.

The impact is expected to hit certain borrower segments first, particularly those directly exposed to fuel costs. These include delivery riders and transport operators, informal sector workers and gig economy participants, as well as small business owners and low-income households.

Despite these pressures, Chan said the current situation differs from past crises such as the pandemic, as stress is more gradual and concentrated in specific sectors rather than across the entire economy.

Advance.AI expects a “two-speed dynamic” in the credit market, with rising loan demand driven by households bridging liquidity gaps, alongside tighter lending standards as more applicants show signs of financial stress.

“A higher proportion of those applicants are financially stressed and lenders will tighten borrowing criteria,” Chan said.

Lenders are responding by re-segmenting portfolios more frequently, acting on behavioral indicators before delinquency appears and shifting toward more targeted rather than blanket tightening.

Chan said the evolving environment is accelerating the adoption of artificial intelligence and alternative data in credit decision-making.

Traditional credit bureau data, which reflects past behavior, is increasingly being supplemented with forward-looking indicators such as e-wallet usage, telco payment patterns and cross-platform borrowing activity.

“Traditional bureau data answers what this borrower has done before. In today’s environment, the more important question is what will they do next,” she said.

These tools allow lenders to maintain risk controls while continuing to extend credit, avoiding the pitfalls of excessive tightening that could slow economic activity.

Looking ahead, Chan said a critical indicator for lenders will be sustained shifts from full repayments to partial payments, alongside rising loan stacking among previously reliable borrowers.

The situation could escalate into a broader credit cycle concern if these patterns persist beyond two to three billing cycles, delinquency rates exceed seasonal norms in vulnerable sectors, and loan applications surge while approval rates drop sharply due to tighter standards.