Entering September, the banking sector has experienced a rare wave of "elephants dancing." Stocks like Bank of Nanjing, Bank of Jiangsu, Bank of Hangzhou, Bank of Chengdu, and China CITIC Bank have recently hit consecutive record highs, while major state-owned banks such as ICBC, China Construction Bank, and Bank of China also refreshed their records at the start of the month. Amid the A-share market's volatile adjustment, this collective surge by banking stocks seems to have once again elevated the "low valuation + high dividend" narrative as market gospel.

In recent years, the most compelling story for bank stocks has been "low valuation." When bank valuations continuously fell below net assets, the market typically focused on PB, dividend yield, and asset revaluation. However, by 2026, this narrative is changing. Data from the National Financial Regulatory Administration shows that in the first half of 2026, commercial banks achieved a cumulative net profit of approximately 1.24 trillion yuan, a year-on-year decrease of 0.6%. But breaking down the numbers reveals a stark "K-shaped divergence": large commercial banks saw a 1.6% year-on-year increase in net profit, and city commercial banks grew by 7.4%; meanwhile, joint-stock banks declined by 3.4%, and rural commercial banks fell by approximately 12.5%. The difference in half-year net profit growth between city commercial banks and rural commercial banks is nearly 20 percentage points.

More troubling is that this gap isn't confined to profits. As of the end of Q2 2026, the non-performing loan (NPL) ratio for rural commercial banks had risen to 2.83%, up 0.04 percentage points from the end of Q1; during the same period, the NPL ratio for city commercial banks was 1.87%, a quarter-on-quarter increase of 0.02 percentage points. Rural commercial banks aren't just experiencing the steepest profit declines; they also face greater asset quality pressure.

This "K-shaped" divergence in profits and asset quality has, in essence, torn apart the unified "low valuation" narrative that has enveloped bank stocks over the past few years. When the industry no longer shares a common cycle, bank valuation logic can no longer be simply packaged as a sector story of "low PB + high dividend." Instead, it must return to the individual level to answer a harsher question: Why are some banks already navigating through the net interest margin (NIM) downturn, while others are still waiting for their own cycle to pass?

Where the True Bottom Lies

Looking only at overall banking industry data, it's easy to reach a straightforward conclusion: the banking sector is bottoming out. In Q2 2026, the commercial banking NIM was 1.41%, up 1 basis point from Q1, marking the first quarter-on-quarter rebound since Q1 2022; quarterly net profit also showed improvement. However, the issue is that NIM stabilization doesn't mean all banks will benefit synchronously.

The true profit formula for a bank has never been just "NIM." Behind it lie questions of whether loans can continue to grow, whether the customer base is high quality, whether funding costs can be reduced, and most importantly—whether new loans will morph into future NPLs. City commercial banks, in this cycle, possess certain conditions that are difficult for rural commercial banks to replicate.

The first is regional advantage. Many leading city commercial banks are located in economic hubs, which are city clusters with high concentrations of industry, population, and capital—take Bank of Jiangsu, for example. In H1 2026, Bank of Jiangsu saw total assets reach 5.61 trillion yuan, up 13.79% from the start of the year; loans grew by 12.20%; operating income was 48.952 billion yuan, up 9.11% year-on-year; and net profit attributable to shareholders was 21.876 billion yuan, an 8.09% year-on-year increase. More critically, its NPL ratio was only 0.81%, down 0.03 percentage points from the end of last year, reaching one of the best levels since its listing.

A similar example is Bank of Chongqing. In H1 2026, Bank of Chongqing achieved operating income of 8.486 billion yuan, up 10.8% year-on-year; net profit attributable to shareholders was 3.518 billion yuan, up 10.3%. Total assets reached 1.11 trillion yuan, up 7.27% from the start of the year.

This isn't to say all city commercial banks are thriving; significant divergence exists within the listed city commercial banks themselves. As of mid-September disclosures, of the 29 A-share and Hong Kong-listed city commercial banks, 25 reported growth in H1 net profit attributable to shareholders. However, Bank of Lanzhou, Bank of Tianjin, Bank of Guiyang, and Jinshang Bank still saw declines; some banks even reported negative operating income growth. This indicates that the "7.4%" growth for city commercial banks isn't an average capability across the entire segment, but rather the result of leading banks benefiting from regional dividends and strong operational capabilities.

The Predicament of Rural Commercial Banks

The story for rural commercial banks is quite the opposite. Their strengths have always been clear—proximity to customers, deep network penetration, and familiarity with local industries and social structures. But this advantage is built on a premise: that the county-level economy can continuously generate enough high-quality credit demand. Once local industrial structures, the real estate market, consumer spending, and small and micro business operations all face pressure simultaneously, this model encounters a very practical problem: branches remain, customers remain, but the amount of money worth lending out diminishes.

Data from the end of Q2 2026 shows that the NIM for rural commercial banks was 1.59%, actually better than the 1.40% for city commercial banks during the same period. This suggests that explaining the operational pressure on rural commercial banks solely through "NIM disadvantage" is too simplistic. With the inversion of the NIM scissors gap, the core issues surface: on one hand, whether interest-earning assets can sustain effective growth; on the other, the risk costs that need to be provisioned for these assets.

Regulatory data shows that as of the end of June, the NPL ratio for rural commercial banks was 2.83%, the highest among major bank types, up 4 basis points from the end of Q1. Their provision coverage ratio was 156.66%, significantly lower than the 171.10% for city commercial banks. This data essentially tells the story. When profits decline by 12.5% while the NPL ratio continues to rise, banks can't simply rely on expanding asset scale to solve the problem. Every additional loan means potential future demands on more capital, provisions, and risk management resources.

This is why rural commercial banks have frequently resorted to clearing and transferring NPLs this year. Reports indicate that many rural commercial banks are intensifying efforts to dispose of non-performing assets. In Q2 2026, there were approximately 502 listed projects for NPL transfer transactions, involving principal and interest of about 101.1 billion yuan. Rural commercial banks are bearing higher risk disposal costs than other banks, and these costs ultimately reflect on their income statements.

Root Causes of the Gap

Writing the story as "city commercial banks good, rural commercial banks bad" would fall into another simplification trap, as significant divergence also exists within rural commercial banks. Changshu Bank is a prime example. In H1 2026, Changshu Bank reported operating income of 6.428 billion yuan, up 6.05% year-on-year; net profit attributable to shareholders was 2.178 billion yuan, up 10.64%. More notably, its NPL ratio was only 0.75%, down 0.01 percentage points from the start of the year, with a provision coverage ratio of 431.94%.

The bank's loan structure is also quite telling: personal loans accounted for 50.77% of total loans, with personal business loans at 34.87%; inclusive small and micro enterprise loans totaled 110.262 billion yuan, serving 219,700 borrowers. This set of data indicates that rural commercial banks don't inherently lack growth potential. The core difference lies in whether a bank builds its "down-market" presence into a true core competency or merely accumulates potential risks through down-market operations.

Consider another example—Chongqing Rural Commercial Bank. In H1 2026, the bank reported operating income of 15.892 billion yuan, up 7.81% year-on-year; net profit attributable to shareholders was 8.168 billion yuan, up 6.09%. As of the end of June, its NPL ratio was 1.05%, down 0.03 percentage points from the end of last year, and it has declined for six consecutive years; the provision coverage ratio was 357.47%.

Jiangyin Bank provides a similar answer. In H1 2026, the bank achieved operating income of 2.449 billion yuan and net profit attributable to shareholders of 866 million yuan; as of the end of June, its NPL ratio was 0.81%, down 0.01 percentage points from the start of the year, with a provision coverage ratio of 349.17%, up 19.19 percentage points from the start of the year. Additionally, Jiangyin Bank's total loans grew by 5.93%, and its deposit cost rate fell by 29 basis points year-on-year.

Similarly, Zijin Bank also achieved net profit of 943 million yuan in H1 2026, up 3.33% year-on-year, with its NPL ratio dropping to 1.33%, down 0.02 percentage points from the start of the year.

So, the real answer has surfaced: the issues facing rural commercial banks stem from different regions, asset structures, and risk control capabilities combining to produce entirely different balance sheets. The same license can yield Changshu Bank's 0.75% NPL ratio, or align with the industry's average of 2.83%. Likewise, among regional banks, Bank of Chongqing can achieve 10.3% net profit growth, while Bank of Lanzhou sees an 8.95% decline; Bank of Jiangsu grew profits by 8.09% in H1, while Bank of Tianjin's profits are falling. This is the most noteworthy change in the banking sector for 2026.

Stop Chasing Banks Based Solely on PB

In recent years, market discussions about bank stocks often circled a persistent question: why are banks so profitable, yet their stocks languish below book value for so long? But perhaps the question should now be reversed: if a bank has a very low PB, but profits are continuously declining, NPL ratios are rising, and provisions are being depleted, then "cheap" doesn't necessarily mean value. Conversely, if a bank can maintain asset growth, profit growth, and declining NPL ratios even amidst an industry-wide NIM downturn, it's understandable that the market would eventually assign it a higher valuation.

From this perspective, the H1 2026 banking earnings reports provide a valuable stress test. The city commercial bank segment saw overall net profit growth of 7.4%, but the true outperformers were those banks with stronger regional economies, customer bases, and asset management capabilities. The rural commercial bank segment saw overall net profit decline by 12.5%, yet Changshu Bank, Chongqing Rural Commercial Bank, and Jiangyin Bank still managed to achieve profit growth and improved asset quality.

Therefore, the so-called "rise of city commercial banks, fall of rural commercial banks" isn't the final answer. A more accurate description is: the banking industry is moving from "license-based pricing" to "balance sheet pricing." In the past, the market might first ask what type of bank you are. Now, the market increasingly needs to ask: Who did you lend to? What are your deposit costs? Does your regional economy have industrial support? Are your NPLs declining, or being offset by continuously newly generated NPLs? Are your provisions a reservoir of profit, or are they becoming a drain on profit?

From this viewpoint, "trading below book value" is merely superficial. Net profit, NPL ratio, provision coverage ratio, and asset structure are the true core of a bank. In H1 2026, the nearly 20-percentage-point gap in profit growth between city commercial banks and rural commercial banks might just be the beginning. The real divergence in bank stocks has shifted from valuations to balance sheets. For those investors still debating "when will bank stocks see value return," what they should focus on next isn't whether a bank trades below net assets, but whether it can first lift itself out of its own mire.