The era of the retail great leap forward has ended: how should the banking industry face the battle over existing assets?

```

The retail myth of commercial banks, which had been surging for over a decade, is now hitting an invisible wall.

Recently, the central bank released financial statistics for the first quarter of 2026:

Of the total new loans in the first quarter, loans to enterprises and institutions increased by 8.6 trillion yuan, while household loans only increased by 296.7 billion yuan.

Within this less than 300 billion yuan household increment, the internal structure clearly signals contraction, with short-term household loans showing a substantial negative growth, a net decrease of 164 billion yuan.

At recent earnings meetings, many bank executives emphasized that developing corporate business has become a common choice for commercial banks.

Behind the contrast between the expansion of corporate business and the stagnation of retail is a shift in the underlying asset logic of the banking industry.

The slowdown in mortgage loan growth, which once served as a safety cushion, has changed the traditional asset allocation rhythm of banks; unsecured credit loans, lacking collateral support, have stalled under pressure from narrowing spreads and rising risks, making the old model—high yields covering bad debts—unsustainable.

As the closed loop of pricing is broken, the banking industry has entered a period of stock adjustment driven by the market. The difficult restructuring of the balance sheet has become a cyclical challenge the whole industry must face.

Shrinking Safety Cushion

The retail business of commercial banks used to be a finely tuned profit engine. The foundation of this was formed by personal mortgage loans, consumption and business loans, and credit card overdraft balances, sustaining the profitability of their balance sheets.

But now, the once coordinated machine is encountering systemic obstacles.

As the highest quality assets—longest term and lowest default rate—personal housing loans not only provided stable interest spreads over the past decade, but also, with low capital occupation and risk losses, were the deepest safety cushion for banks' retail business.

Statistics over the past 10 years show that from 2016 to 2022, the one-way rise of mortgage balances saw the six major banks (ICBC, ABC, BOC, CCB, BoCom, PSBC) reach a compound annual growth rate of 9.64%.

But in 2023, this established expansion trend came to an abrupt halt.

In the following three years, the compound annual growth rate of this asset reversed to -1.56%. In 2025, the contraction slope increased further, with a year-on-year decrease of 2.47%, an expansion of 42 basis points compared to the previous year. The aggregate housing loan balances of the six major banks, totaling 32.9 trillion yuan, have returned to pre-2021 levels.

The scale continues to retract.

In the first quarter of 2026, the weighted average interest rate of newly issued commercial individual housing loans nationwide fell to a low of 3.06%. However, the RMBS conditional prepayment rate index, reflecting the activity of early repayments, remains high, and the gap created by early repayments has not yet been filled by new demand.

The stagnation in mortgage scale has disrupted the banks’ original asset lineup—

The supply of low-risk foundational assets is in crisis, so the massive lending quotas and internal performance targets must seek new outlets, and the pressure has quickly shifted to personal consumption and business loans.

This reversal of supply and demand power is reflected in the product yield curve.

From 2021 to 2025, the average interest yield of personal loans at retail-focused joint-stock banks—Ping An Bank, Industrial Bank, and China Merchants Bank—declined by 270bps, 206bps, and 161bps respectively, down to 4.79%, 4.16%, and 3.94%.

At the start of 2026, banks generally adopted a combination of rate reductions, increased quotas, and extended terms to boost “opening red” performance; for some joint-stock banks, after adding discount coupons on “Quick Loans”, the actual annualized rate has fallen to 2.58%, even inverted with the 3.06% new housing loan rate over the same period.

Falling below the 3% pricing range is already a direct, homogenous struggle under asset scarcity.

In the ongoing squeeze on interest spreads, some borrowers even spotted arbitrage opportunities, using low-rate business loans to replace early high-rate mortgages, further intensifying the repricing of core bank assets.

More serious is that fierce competition on the supply side seldom receives a response from the demand side.

In the first quarter, China's total retail sales recorded a year-on-year increase of 2.4%, but residents’ willingness to expand leverage remains weak, with net short-term household loans down by 164 billion yuan—a rare quarterly negative growth in recent years.

Simultaneous stagnation in scale and declining prices are impacting the traditional bank credit model.

High Returns Broken

In the past, credit loans were a game of high return covering high risk; as long as returns outpaced bad debts, pricing could be self-sustaining.

But now, facing both scale and price declines as well as exploding risks, this logic is gradually failing.

Of all the unsecured credit assets, credit cards are the most sensitive indicator of this round of risk.

Take China CITIC Bank as an example: during the expansion phase from 2016 to 2022, its total cards issued grew from 37.38 million to 106.6 million, with a compound annual growth rate near 20%;

But after 2023, card growth slowed, and by 2025, the compound annual growth rate dropped to 4.88%.

The experience is more pronounced among large state-owned banks: Between 2023 and 2025, Bank of Communications had a net decrease of over 11 million cards; ICBC and CCB also saw scale reductions of several million cards.

The simultaneous peaking of scale and accelerated exposure of stored risks have arrived together.

Take Ping An Bank as an example:

After the retail reform in 2016, high-pricing products such as elite platinum cards enabled the bank to boost its net interest margin to lead its peers, with profit growth exceeding 25% from 2021 to 2022.

But with the macroeconomic downturn and a missed adjustment window, Ping An Bank’s credit card non-performing rate soared to 2.77% in 2023;

The new management made a “hard landing” adjustment to high-yield credit business; the retail department’s profit contribution fell from more than 70% at its 2019 peak to just 0.6% in 2024.

Other banks may not be as severe as Ping An, but they are not immune to cyclical waves—

Risk exposure of credit assets has a significant lag; previously embedded default risks, unable to be diluted by new assets in times of stagnation, will inevitably lead to a collapse in profitability and asset quality when they surface.

With both mortgages and credit cards losing steam, personal business loans have taken over as the relay supporting retail scale.

From 2020 to 2024, business loans maintained over 20% annual growth. By the end of Q4 2025, household sector business loans rose 4.0% year-on-year, becoming the only segment with positive growth in retail lending.

But as a credit extension also without collateral, business loans couldn’t escape the curse of the old model either.

As most small business owners cannot provide sufficient collateral, banks cannot truly grasp their operating conditions. Under low interest rates due to price wars, business loan profits may not cover costs, operations, and bad debts.

Coupled with pressure from the real economy, some borrowers have already turned credit into "lending to repay" capital maneuvers.

According to industry sources, many new business loan clients are not actually using the funds for operating turnover, but instead use Bank A’s low-cost funds to repay Bank B’s old debts. The cross-bank lending means the capital only circulates on paper, masking bad debts in the short term, but in the long run, business loans may become ballooning bundles of non-performing assets like a game of musical chairs.

Today's business loans face a situation similar to credit cards—mismatch between instant pricing and lagging risks.

Rates below 3% may affect current net interest margins, but real bad debt exposures lag by 12-18 months—this means the current low rates may not cover future risk exposure at all.

Zeng Gang, director of the Shanghai Finance and Development Laboratory, notes that today's credit market is highly abnormal: shrinking scale and falling prices.

"According to traditional credit pricing frameworks, lower prices should stimulate demand, leading to scale and price resonance, but the opposite is happening." Zeng points out, disorderly supply competition has nearly eliminated risk premiums, and structural shrinkage on the demand side means ultra-low-priced loans can't create effective demand.

The flexible space for risk pricing has been severely squeezed.

Zeng Gang believes the old pricing loop—expansion based on scale and high returns compensating for risk—has faced a structural break, not merely cyclical short-term pressure.

To maintain apparent asset quality stability, banks are forced to pay the high price of write-offs.

In the first quarter of 2026, the industry's reported NPL (non-performing loan) rate fell a nominal 1bp, but only due to accelerated write-offs of personal bad loans; massive write-offs continue to swallow up profits and provision resources, and relying on scaling up to dilute bad debt has been financially disproven.

Survivor Samples

With the old profit engine gradually sputtering out, commercial banks are being forced into a brutal market-driven reallocation of resources.

At the table of zero-sum games in stock competition, banks with different endowments have taken sharply divergent paths.

Many joint-stock banks once strong in retail have chosen to retreat.

From 2023 to 2025, Ping An Bank’s retail loan growth went from -3.4% to a low of -10.6% before narrowing to -2.3%; China Everbright Bank saw three years of negative growth, and Industrial Bank’s retail end also shrank by -3.41% in 2025;

In contrast, listed banks’ corporate loans continued steady growth, and have even become the main engine of scale growth.

This contraction is not proactive offense, but defensive, forced by bad debts.

Zeng Gang points out, some joint-stock banks fully exposed retail NPL peaks in the last cycle, and shifting toward corporate business is trading off short-term high returns for a phase improvement in asset quality—it’s a blood-stopping surgery that swaps time for space.

In the gap left by the retreat of joint-stock banks, state-owned big banks are leveraging their size advantage to harvest gains.

Over the past five years, Agricultural Bank of China and ICBC both saw loan increases over 50% and 40% respectively; while expanding rapidly, their NPL rates in Q1 2026 still held at safe levels: 1.25% and 1.31%, respectively.

In the new restructuring, although the six major banks' retail growth cannot match their corporate growth, compared to the widespread balance sheet shrinkage of joint-stock banks, they still firmly maintain positive growth.

The trump card for large banks is their low funding cost.

With abundant cheap capital, state-owned banks can accept asset pricing as low as 2.3% in lower-tier markets, steadily penetrating the domains of joint-stock banks.

But Zeng Gang warns, the “harvest” strategy of large banks is not without hidden dangers: continued downward penetration will accelerate the industry's overall pricing center downwards and may ultimately exert systemic pressure on the net interest margin system for all banks.

At the same time, Zeng points out that without a demand-side recovery, the size dominance of big banks and the asset retreat of small and medium banks is merely a positional switch in a stock game, not genuine incremental growth.

Beyond the competition among national and joint-stock banks, city and rural commercial banks are also accelerating.

For instance, Bank of Nanjing and Bank of Ningbo, which have expanded rapidly in recent years, have shifted defensive focus to consumer and micro business loans. Over five years, they maintained core retail asset expansion and kept NPL ratios in good shape.

Postal Savings Bank researcher Lou Feipeng notes that the moat for city commercial banks in Jiangsu and Zhejiang regions is their extremely high local concentration and flat decision chains.

Leveraging long-term, non-standardized “soft” information about local small and micro enterprises, these banks have established price-resistant defensive positions in fragmented, lower-tier markets, thus creating a locally stratified structure of coexistence.

Lou Feipeng believes this local advantage is sustainable. Given city banks’ clear differentiation, the industry structure will be more likely “layered coexistence” rather than zero-sum competition in the future.

After years of internal competition, data from Q1 2026 finally released a faint signal of stabilization: commercial banks nationwide achieved a slight net profit increase of 0.5% year-on-year, and the non-performing loan ratio temporarily stabilized at 1.59%.

With the continued recovery of the macro economy and improved household income expectations, Lou Feipeng thinks the pressure to expose retail loan risks will noticeably lessen in the future.

The multiple deposit rate cuts previously made are beginning to optimize funding costs, while the marginal stabilization of net interest margins gives banks a window to breathe and adjust.

This does not mean the crisis is over.

Zeng Gang points out that as the era of high-return-driven expansion ends, the core issue for commercial banks in restructuring their balance sheets must shift from “expanding the numerator” to “optimizing the structure”:

The key point is to return to relationship banking.

Zeng Gang notes, banks must rebuild comprehensive service capabilities centered around the customer, packaging loans, settlements, and wealth management to deliver as one, anchoring quality clients through stickiness, not price competition.

Second, asset allocation must be re-anchored to real economic activity.

“The credit system has a lot of idle funds. This not only drags down the real efficiency of social financing, but also leads to hidden risk accumulation in bank asset quality.” Zeng emphasizes that bank financing must penetrate industry chains via supply chain finance and tech credit, embedding money into real transaction scenarios to fundamentally reduce idle circulation.

Lou Feipeng adds that banks should strengthen scenario-based risk control by building a “document review–transaction verification–capital matching” three-in-one inspection mechanism, connecting merchants directly via API for closed-loop capital management, and using AI for real-time fund flow monitoring;

At the same time, differentiated pricing should be implemented, utilizing big data for fine-tuned segmentation to better match risks and returns.

Finally, active management of the liability side will become a key variable in the new era.

Zeng Gang believes that, with high-interest deposits expiring and net interest margins stabilizing, how well banks optimize their liability structure during this adjustment window will determine which become leading versus mediocre banks in the next phase.

In any case, the era of blindly relying on high-yield products for meteoric scale growth has come to an end. Shifting from “scale first” to prudent asset optimization may bring pain, but is an inevitable path for the long-term development of commercial banks.

Risk Warning and DisclaimerThe market is risky and investment needs caution. This article does not constitute personal investment advice and has not considered any specific user's investment objectives, financial situation, or needs. Users should assess whether any opinions, viewpoints, or conclusions in this article are appropriate to their particular situation. Investment decisions made based on this article are at your own risk. ```