X Financial (NYSE: XYF) — When the Cheapest Stock in the Most Hated Sector Hits Reality
A 1x P/E Chinese fintech with 12% dividend yield, 0.17x book — but the regulator just changed the rules
Introduction & Investment Case
Today I want to talk about a stock I have been quietly following for years and that, on the surface, looks like one of the most absurd valuation setups in global equities right now: X Financial (NYSE: XYF). At today’s share price of around USD 4.97, this Chinese online consumer-credit platform trades at roughly 1x trailing P/E, around 0.17x book value, has a forward dividend yield north of 11% — and yes, the company is still profitable, still growing in volume terms over the last fiscal year, and still buying back its own shares.
If you stopped reading the data sheet there, you would think this is the most obvious ‘fat pitch’ since the Chinese ADR sell-off of 2022. But as always with situations that look this cheap, there is a reason. And in the case of XYF, the reason got materially worse when the company released its Q4 2025 numbers on March 25, 2026.
“The most hated stock in the most hated sector in the most hated country. That is exactly where multi-baggers come from — and total losses, too.”
The short version of the thesis
X Financial is a founder-led online consumer-credit facilitator in China that converted from peer-to-peer lending into a bank-funded loan platform. Full-year 2025 net income was RMB 1.46 billion (USD 209 million), down only marginally from RMB 1.54 billion in 2024 despite a brutal Q4. The market cap, however, has collapsed to roughly USD 192 million. That is what gets you to a 1x P/E ratio. The company holds about USD 1.12 billion of total equity, of which the cash and equivalents make the balance sheet look fortress-like for a business of this size.
The catalyst for the recent price collapse is real and matters: in April 2025, China’s NFRA issued Notice 9, which de facto enforces a 24% per annum cap on total borrowing cost for consumer loans. Q4 2025 then showed what that regulation, combined with deteriorating consumer credit quality, can do — net income collapsed 85% year over year, operating margin fell from 30.7% to 1.4%, and Q1 2026 loan origination guidance came in at RMB 14.5–15.5 billion, less than half of Q1 2025’s record RMB 35.1 billion.
So the question is not whether XYF is statistically cheap. It clearly is. The question is whether the earnings power that the cheap multiple references can survive the new regulatory and credit reality. My base case is that 2026 will be a deep trough year, that 2027 stabilizes, and that by 2028 we can see EPS recover to roughly USD 2.20 per ADS. With a Fair P/E of 8x — appropriate for a Chinese consumer-credit platform with capital return discipline — that gets to a fair value of around USD 17.60 versus today’s USD 4.97. That is roughly 50% expected annual return over the next three years, but the path will be ugly.
1. Product, Business Model, Brand & Moat
What does X Financial actually do?
X Financial operates an online consumer credit platform, primarily under its flagship product, the Xiaoying Card Loan. This is an unsecured consumer loan with terms of 6 to 12 months and ticket sizes typically in the range of a few thousand RMB up to roughly RMB 50,000 — in EUR terms, anywhere from a few hundred euros to roughly EUR 6,500. The target borrower is a younger, mass-affluent Chinese consumer who is underserved by traditional banks because they do not own real estate that can be used as collateral. The minimum interest rate the company quotes starts at around one-tenth of the headline ceiling, and the effective APR for higher-risk segments goes up to the regulatory cap that is now de facto 24%.
The original incarnation of the company was a peer-to-peer model, but that business essentially died in China the same way it died in the US and Europe — regulators forced platforms to either work with licensed institutional capital or shut down. So X Financial pivoted into what is now called the loan facilitation model: it sources the borrowers, runs the credit assessment with its proprietary risk engine WinSAFE, and then routes the loan demand to a network of bank funding partners that put up the capital and book the loans on their balance sheets.
How does the company make money?
Revenue comes from five buckets, of which three are dominant. Loan facilitation service revenue — the upfront fee X Financial earns when it routes a borrower to a funding partner — was RMB 3.84 billion in FY2025. Post-origination service revenue (servicing the loan over its life) was RMB 1.07 billion. Financing income (interest on loans the platform itself co-funds) was RMB 1.40 billion. Guarantee income, which is the fee the company charges for guaranteeing repayment to its bank partners, exploded from RMB 202 million in 2024 to RMB 637 million in 2025 — and that is where the entire business model gets interesting and risky at the same time.
Officially X Financial is asset-light: the loans sit on bank balance sheets, not on its own. De facto, however, the company guarantees the credit performance of those loans to its funding partners. That guarantee is what gets the banks comfortable with sub-prime and near-prime borrowers they would never originate themselves. The model is structurally similar to Upstart or Affirm in the US — they all claim to be tech platforms, but the moment loan returns deteriorate, the funding partners walk and the platform either takes the credit risk on balance sheet or shrinks dramatically.
The Q4 2025 numbers tell that exact story. Provision for contingent guarantee liabilities exploded from RMB 116 million in Q4 2024 to RMB 398 million in Q4 2025 — a 243% increase. That is the cost of standing behind loans that are not performing the way the model assumed. So no, X Financial is not a software company with no credit risk. It is a credit underwriter dressed up as a tech platform, and the disguise just slipped.
Brand
Brand is honestly not the strongest part of the case. Xiaoying is recognizable in its target segment in China, but it is not a top-of-mind consumer fintech brand the way Lufax or Ant Group are. What matters more here than brand recognition is funding-partner credibility. The company discloses partnerships with names like Citi Consumer Finance and a roster of mid-sized Chinese banks. None of these are household names internationally, but together they provide enough funding diversification that the platform is not single-counterparty dependent. That is the real franchise here — not the consumer brand, but the institutional plumbing.
Moat — and the lack of it
Let me be honest. The moat here is thin. The risk model (WinSAFE) is proprietary, but every Chinese online lender claims to have a proprietary model. The funding-partner relationships are real but not exclusive. The closest thing to a moat is the regulatory burden of operating in this space — China has effectively shut the door on new entrants for years by tightening licensing and capital requirements. That gives the surviving listed players, including X Financial, FinVolution, LexinFintech, and Qifu Technology, a kind of regulatory oligopoly position. But that protection cuts both ways: when the same regulator decides to cap pricing, the moat becomes the cage.
2. Market — Big, Cyclical, and Politically Captured
The Chinese consumer credit market is enormous. Outstanding consumer loans excluding mortgages run into the trillions of yuan. The structurally underserved segment that X Financial targets — mass-affluent younger borrowers without collateral — is hundreds of billions in annual originations. So market size is not the bottleneck. The platform’s FY2025 origination of RMB 130.6 billion (USD 18.7 billion) is a rounding error in the broader market.
Growth, however, is now decelerating sharply. The decade of unconstrained Chinese consumer credit growth ended somewhere between 2018 and 2021, and what we are watching now is regulatory normalization. PBOC is cutting policy rates, which is supportive for loan demand on the borrower side, but the regulator is simultaneously squeezing the supply side by capping pricing and forcing whitelisted bank-cooperator lists.
The cyclicality of this business cannot be overstated. Consumer credit losses in China track urban employment, real estate sentiment, and youth unemployment closely. Q4 2025 91-180 day delinquencies of 6.31%, more than 2.5x the 2.48% of a year ago, are telling us that the Chinese consumer is under genuine stress — not just XYF’s borrowers. Look across the listed Chinese consumer lenders: every single one is showing the same pattern in late 2025. This is a sector-wide credit cycle, not an X-Financial-specific blow-up.
On political risk, my view has shifted somewhat over the last twelve months. There are real signs of de-escalation — Germany was just included in China’s visa-free entry list, US-China commercial dialogue has resumed, and the Hong Kong listing pipeline has reopened for Chinese ADRs. The threat of forced delisting from US exchanges, which dominated the 2022 sell-off, has substantially receded under the PCAOB audit cooperation agreement. So the geopolitical tail risk that was justifying part of the discount in 2022-2023 has materially declined. What has not declined is the domestic regulatory risk — and Notice 9 is the proof.
3. Culture & Management — Founder-Led, Skin in the Game
This is, for me, the single strongest qualitative argument for the stock. X Financial is led by founder and chairman Justin Tang, who previously founded eLong, China’s number two or three online travel platform, which eventually merged with Tongcheng to become a profitable mid-cap travel business. So this is not a first-time founder running a credit shop — this is somebody who has already built and exited a meaningful internet business in China. He does not need this to work for personal financial reasons. He is doing it because he believes there is a business there.
President Kent Li and CFO Frank Fuya Zheng have both been with the company since the early days. The Q4 2025 earnings call language was unusually direct about the regulatory pressure, the asset quality stress, and the visibility issues into 2026 — that level of candor is rare in Chinese ADR communication, where the default mode is corporate optimism. When management says on a call that ‘the possibility of operating losses in future periods cannot be excluded,’ you have to give them credit for not pretending.
Capital allocation tells me they understand the situation. The company has consistently returned capital to shareholders even as the share price has been falling: a USD 100 million buyback program of which roughly USD 53.85 million has been executed by mid-March 2026, with USD 46.15 million still authorized through November 2026, and a semi-annual dividend of USD 0.28 per ADS just declared for May 20, 2026 payment. At the current price, that is more than 11% dividend yield, and on top a huge buyback yield!
The real question on management is whether they continue to underwrite credit conservatively under regulatory pressure, or whether they chase volume to defend the top line. The Q1 2026 origination guidance of RMB 14.5-15.5 billion — a 56% reduction versus Q1 2025 — tells me that they have chosen discipline over volume. That is exactly what I want to see from a founder-led credit business in a stress environment.
4. Financials — The Brutal Q4 and What It Means
Full year 2025 versus 2024
Headline 2025 looks strong: total net revenue of RMB 7.64 billion was up 30.1% YoY, loan origination volume of RMB 130.6 billion up 24.5% YoY. But the headline hides the trajectory. Q1 2025 was a blow-out (revenue +60% YoY), Q2 2025 was a record (revenue +66% YoY, RMB 39 billion in originations), and then Q3 and especially Q4 fell off a cliff. Net income for the full year was RMB 1.46 billion — actually slightly below the RMB 1.54 billion of 2024.
What killed the margin in Q4
The single biggest line-item move was the provision for contingent guarantee liabilities. RMB 398 million in Q4 2025 versus RMB 116 million in Q4 2024 — a tripling of the cost of standing behind the loan book. Provision for accounts receivable went from RMB 13 million to RMB 140 million, and provision for loans receivable from RMB 64 million to RMB 133 million. Combined, the credit-related provisions alone went up by roughly RMB 480 million quarter on quarter, which essentially equals the entire delta in net income.
On top of this, borrower acquisition and marketing spend was actually cut to RMB 212 million in Q4 from RMB 504 million the prior year, reflecting the pull-back on volume. So the cost-of-credit story is what matters, not the cost-of-acquisition story.
Asset quality — the uncomfortable truth
Both buckets more than doubled. The 91-180 day bucket at 6.31% is alarming — historically when this bucket gets above 6%, write-offs in the following two quarters tend to be substantial. The company has responded by tightening underwriting (hence the Q1 2026 guide cut), which is correct from a long-term perspective, but it means 2026 net income is going to be materially below 2025.
Balance sheet — the real reason this is investable
Total assets of RMB 14.67 billion (USD 2.10 billion). Total equity of RMB 7.84 billion (USD 1.12 billion). Cash and cash equivalents plus restricted cash of RMB 2.13 billion (USD 305 million). Short-term borrowings of just RMB 410 million. Net cash position is positive. Against a market capitalization of roughly USD 192 million, the price-to-book is around 0.17x and the company holds more than 1.5x its current market cap in cash alone.
Even if I assume that 30% of accounts receivable and 20% of loans receivable get written off in 2026 — a genuinely dire scenario — the remaining tangible book value still exceeds the current market cap by more than 3x. That is the floor that makes this investable. You are not buying earnings power here; you are buying assets at a deep discount to liquidation value with a free option on earnings recovery.
5. Valuation — Fair P/E and the QFMA Framework
I value all my positions through the Fair P/E and Quality, Fundamentals, Macro, Awareness (QFMA) framework. For X Financial, the inputs look like this:
Fair P/E components
Quality: Below average. The business is cyclical, regulator-captured, and credit-risk-heavy. I would normally only assign a P/E of 6-7 to a pure play of this kind.
Capital allocation: Above average. Founder-led, consistent buybacks, double-digit dividend yield, no equity issuance. This adds 1-2 points to the multiple.
Growth: Negative for 2026, recovery from 2027. Long-term low-to-mid single digits at best given regulatory pricing caps. Neutral to slightly negative.
Macro environment: Chinese consumer is under pressure, but rates are falling and there are signs of geopolitical de-escalation. Slightly positive over a 2-3 year horizon.
Awareness: Practically nil. Sell-side coverage is minimal, US institutional ownership is collapsed, retail awareness is zero. This is the textbook deep-value setup where awareness can only go up.
Putting that together, my Fair P/E for XYF is 8x. That sits well above the absolute floor for a cyclical credit business but below what I would assign to a software-platform fintech. It explicitly reflects the asset-light disguise being only partially true — the company carries credit risk, and the multiple needs to acknowledge that.
Earnings projection
Fair value calculation
EPS 2028 estimate of USD 2.20 multiplied by Fair P/E of 8x = USD 17.60 fair value per ADS. Versus today’s USD 4.97, that is upside of approximately 254% over a 30-month horizon, or roughly 50% annualized expected return for the next three years. On top of that, you collect a roughly 11% dividend yield while you wait, and the buyback continues to retire shares at well below book value, which is mathematically accretive to remaining holders.
“Fair value: USD 17.60 per ADS based on a 2028 EPS of USD 2.20 and Fair P/E of 8x. Versus today’s USD 4.97, that is roughly 50% expected annual return for the next three years.”
Sanity-check this against peers: FinVolution trades at a higher multiple but has shown similar Q4 2025 stress. LexinFintech trades at a comparable low single-digit P/E. Qifu Technology is the higher-quality name in the space and trades at 4-5x. So a Fair P/E of 8x for XYF is in line with the peer group’s higher-quality benchmarks, not aggressive. This is consistent with my view that XYF specifically is the cheapest of the listed Chinese consumer-credit platforms — and that it is cheap for valid but not catastrophic reasons.
6. Risks — The Reasons This Is Not An Obvious Buy
Regulatory risk — the Notice 9 problem
This is the dominant risk, and it is not theoretical. Notice 9, issued by the NFRA on April 1, 2025, requires commercial banks to strictly control total borrowing costs. In practice, this is being implemented as a 24% per annum cap on total borrowing cost — and as management explicitly stated on the Q4 call, the actual cap may end up below 24% for some institution types. The previous ceiling under the consumer protection law was 36%. Compressing pricing from up to 36% down to 24% or below mathematically destroys roughly one-third of gross take rate, and most of that drops directly to the bottom line.
Worse, the same notice mandates whitelist management for loan facilitation platforms — banks can only work with platforms on the whitelist. The implementation timeline and the criteria for whitelist inclusion remain opaque. If XYF is not on the whitelist of one of its key bank funding partners, that funding line evaporates overnight.
Credit cycle risk
Q4 2025 delinquencies are bad and they tell me that 2026 vintages — loans originated in late 2025 and Q1 2026 — will produce more losses than the company has reserved for. The provision build in Q4 was significant but probably not the last one. I expect another quarter or two of elevated provisions before normalization in late 2026 or early 2027.
Litigation risk
Following the Q4 2025 release, Johnson Fistel announced an investigation of potential securities claims against X Financial. This is a typical ambulance-chaser response to any sharp share price decline by a US-listed Chinese ADR, and most such investigations never produce a class action that matters. But it is worth flagging.
VIE structure risk
X Financial uses the standard variable interest entity (VIE) structure that all Chinese ADRs employ. Theoretically, the Chinese government could invalidate VIE structures and US shareholders would have no recourse to the underlying Chinese operating entities. Practically, this risk has been present since 2018 and has never crystallized for any major listed name. But it is on the menu.
Total loss scenario
In a worst-case combined scenario — Notice 9 implementation forces effective rates to 18%, delinquencies stay above 6%, and the company is forced to recognize impairments on accounts receivable and contingent guarantee liabilities — XYF could go from current profitability to multi-year losses. Tangible book value provides a floor, but a stock at 0.17x book can go to 0.10x book on sentiment alone. So a 40-50% drawdown from here is genuinely possible before the recovery thesis plays out.
7. Latest News & Earnings
Q4 2025 reaction
Following the March 25, 2026 earnings release, XYF dropped roughly 18.6% on the day, taking the stock from the $5 area to below $4. That is the move that created today’s 1x P/E setup. The market correctly read the Notice 9 disclosure and the asset quality deterioration as a structural reset, not a one-quarter accounting issue.
Next earnings catalyst
Q1 2026 earnings are scheduled for late May 2026. The key data points to watch: actual Q1 origination volume against the RMB 14.5-15.5 billion guide, the new delinquency cohorts, and whether management can give meaningful FY2026 guidance now that they have one quarter of operating data under the new pricing regime. If management announces a Hong Kong dual listing — something multiple Chinese ADRs in this size range are considering — that would be a meaningful re-rating catalyst by broadening the investor base.
Capital return execution
The semi-annual dividend of USD 0.28 per ADS pays out around May 20, 2026 (record date April 30). The buyback continues with USD 46.15 million authorized through November 2026. Given the current share price, every dollar of buyback retires roughly twice as many shares as it would have at the start of the program, which is exactly what value-disciplined buyback should look like.
8. Conclusion & Investment Case
X Financial is the cheapest stock I currently own in the most hated sector in the most hated country, run by a founder who already exited one successful business and has a real personal stake in this one working out. It trades at 1x trailing P/E, 0.17x book value, with a 11% dividend yield and an active buyback. The balance sheet is fortress-strong relative to the market cap. And it just took a 60% drawdown in twelve months on a regulatory shock that is real but not existential.
Catalysts for a re-rating
Q1 2026 earnings showing that operating losses are limited and asset quality stabilizes — even just stabilization at current levels would be a positive surprise.
Regulatory clarity on Notice 9 implementation — once banks and platforms know the actual rate caps and whitelist criteria, the uncertainty discount can compress.
Continued buyback execution at current depressed prices, retiring shares at well below book — pure mathematical accretion to remaining shareholders.
Hong Kong secondary listing — opens the stock to mainland Chinese investors via Stock Connect and broadens the buyer base.
Dividend continuity — if management confirms the semi-annual dividend through 2026 even with reduced earnings, that signals confidence and supports the equity story.
Geopolitical de-escalation — visa-free entry programs, US-China trade dialogue, audit cooperation — all incrementally reduce the China-ADR discount.
Sector consolidation — if regulatory pressure forces smaller competitors out of the market, surviving listed players get share gains and pricing discipline.
How I am positioned
X Financial is a small position in the Haas Invest4 Innovation Fund (invest4.net) s. The position is sized explicitly to absorb a total loss without compromising the broader portfolio. I . If Q1 2026 earnings confirm that 2026 is a trough year and not a permanent reset, I would consider adding. If asset quality deteriorates further or regulatory implementation comes in tighter than expected, I would re-evaluate.
“This is the kind of position where you make multiples or you take a hit. Both outcomes are real. Position sizing is everything.”
If I am right, this is the kind of setup where a five-bagger over three years is achievable. If I am wrong on the regulatory trajectory, this could absolutely go to zero. Both outcomes are inside the range of possible. That is exactly why position sizing is everything in this kind of trade. I would not put more than 1-2% of a portfolio into a name like this, and even less for investors who are not experienced with deep-value cyclicals or Chinese ADRs. There are simpler ways to make money in equities — but the asymmetry here is too wide for me to ignore.
Risk Disclaimer & Conflicts of Interest
This article is for informational and educational purposes only and does not constitute investment advice, an investment recommendation, or an offer to buy or sell any security. The author, Philipp Haas, holds an indirect position in X Financial (NYSE: XYF) through the Haas Invest4 Innovation Fund (invest4.net) and through Wikifolio products he manages. As a result, the author has a financial interest in the price development of XYF and a conflict of interest exists.
The Haas Invest4 Innovation Fund is a cost-efficient, broadly diversified innovation-focused investment fund managed by the author. X Financial is one of many positions in the fund and represents a small fraction of total fund assets, sized to reflect its high-risk profile.
Investing in Chinese ADRs, micro-cap stocks, and consumer credit platforms involves substantial risks including but not limited to total loss of capital, regulatory risk, foreign exchange risk, VIE structure risk, audit and accounting risk, and severe price volatility. Past performance is not indicative of future results. All financial figures are based on publicly available company filings and may contain errors. Readers must conduct their own due diligence and consult with a qualified financial advisor before making any investment decisions. The author and investresearch.net accept no liability for losses incurred from acting on information contained in this article.