For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for Mizuho Financial Group Inc. (MFG) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-08-19): Designation Watch · Gem Score +22 (−100…+100 Quality+Value blend) · Quality 20 · Value 24 · Sentiment -11 (timing only, not weighted) · Composite fair value $12.21 vs $10.69 at analysis
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
Mizuho Financial Group Inc.
MFG NYSEMizuho Financial Group Inc. ADR represents a prominent player in the financial services sector, offering comprehensive banking, trust banking, securities, and leasing services. Based in Tokyo, Japan, Mizuho Financial Group is one of the largest financial institutions in the country and has substantial international reach. As an American Depositary Receipt (ADR), it provides U.S. investors a way to invest in Mizuho Financial Group without dealing with direct foreign investments, offering convenience by trading in U.S. dollars on the New York Stock Exchange. This enhances the group’s visibility and accessibility to American investors. Mizuho Financial Group serves a wide array of clients, including individuals, corporate clients, and governmental entities, playing a crucial role in facilitating international trade and investment between Japan and global markets. The company's influence spans various industries, such as real estate, retail, and technology, by providing essential financial solutions and expertise. Its market significance is underscored by its contribution to both domestic and international financial landscapes, driving economic growth and stability.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
This company does not file structured financial statements with the U.S. SEC, so quarterly figures aren't available from our filings-based data engine. Annual figures shown here come from the sources that do cover it.
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 0.64
Total Equity: $72.85B
Shares: 12,413,823,272
Total Debt: $132.26B
Cash: $397.36B
EBITDA: N/A
Total Debt: $132.26B
Cash: $397.36B
Revenue: $55.82B
Revenue: $55.82B
Revenue: $55.82B
Total Equity: $72.85B
Tax Rate: 22.2%
Equity: $72.85B
Total Debt: $132.26B
Cash: $397.36B
Current Liabilities: N/A
Long-Term Debt: $132.26B
Total Debt: $132.26B
Total Equity: $72.85B
Shares: 12,413,823,272
Shares: 12,413,823,272
CapEx: -$1.95B
Shares: 12,413,823,272
Stock Price: $10.70
Net Income: $7.35B
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 5, 2026 9:38am (63d ago)| Metric | 2022 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Revenue | $13.4B | $27.1B | $54.0B | $51.9B | $55.8B |
| Cost of Revenue | — | — | — | — | — |
| Gross Profit | — | — | — | — | — |
| Operating Expenses | $3.9B | $4.4B | $4.8B | $5.2B | $6.1B |
| Operating Income | — | — | — | — | — |
| Net Income | -$664.7M | -$88.9M | $5.8B | $3.8B | $7.3B |
| EBITDA | — | — | — | — | — |
| EPS | $0.27 | $0.28 | $0.34 | $0.44 | $0.64 |
| EPS (Diluted) | $0.27 | $0.28 | $0.34 | $0.44 | $0.64 |
Balance Sheet (Annual)
Last updated: Aug 5, 2026 9:38am (63d ago)| Metric | 2022 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Cash & Equivalents | $331.1B | $431.5B | $470.4B | $466.1B | $397.4B |
| Total Current Assets | — | — | — | — | — |
| Total Assets | $1.5T | $1.6T | $1.7T | $1.8T | $1.9T |
| Current Liabilities | — | — | — | — | — |
| Long-Term Debt | $79.8B | $94.5B | $103.3B | $94.7B | $132.3B |
| Total Liabilities | $1.4T | $1.5T | $1.7T | $1.7T | $1.8T |
| Total Equity | $59.9B | $61.7B | $66.2B | $67.0B | $72.9B |
| Retained Earnings | $16.9B | $15.5B | $19.8B | $21.2B | $25.8B |
Cash Flow (Annual)
Last updated: Aug 5, 2026 9:38am (63d ago)| Metric | 2022 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Operating Cash Flow | $34.9B | $6.4B | -$22.0B | $374.7M | -$11.3B |
| Capital Expenditure | -$864.7M | -$1.2B | -$1.6B | -$2.1B | -$1.9B |
| Free Cash Flow | $34.0B | $5.2B | -$23.6B | -$1.8B | -$13.3B |
| Acquisitions (net) | — | — | — | — | — |
| Net Debt Issued / (Repaid) | $1.8B | $10.9B | $667.6M | -$7.6B | $32.4B |
| Dividends Paid | — | — | — | — | — |
| Stock Buybacks | -$12.2M | -$14.7M | -$21.5M | -$653.2M | -$2.6B |
| Net Change in Cash | $21.8B | $100.4B | $38.8B | -$4.3B | -$68.7B |
Growth Trends (YoY %)
Last updated: Aug 5, 2026 9:38am (63d ago)| Metric | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|
| Revenue Growth | +102.3% | +99.0% | -4.0% | +7.6% |
| Gross Profit Growth | — | — | — | — |
| Operating Income Growth | — | — | — | — |
| Net Income Growth | +86.6% | +6,613.5% | -35.0% | +95.2% |
| EBITDA Growth | — | — | — | — |
Dividend History (Last 20)
Last updated: Aug 5, 2026 9:38am (63d ago)| Date | Dividend | Declaration | Record | Payment |
|---|---|---|---|---|
| 2025-03-31 | $0.10 | — | — | — |
| 2024-09-30 | $0.09 | — | — | — |
| 2024-03-27 | $0.07 | — | — | — |
| 2022-09-29 | $0.06 | — | — | — |
| 2022-03-30 | $0.06 | — | — | — |
| 2021-09-29 | $0.07 | — | — | — |
| 2021-03-30 | $0.07 | — | — | — |
| 2020-09-29 | $0.07 | — | — | — |
| 2020-03-30 | $0.07 | — | — | — |
| 2019-09-27 | $0.07 | — | — | — |
| 2019-03-28 | $0.07 | — | — | — |
| 2018-09-27 | $0.07 | — | — | — |
| 2018-03-28 | $0.07 | — | — | — |
| 2017-09-28 | $0.07 | — | — | — |
| 2017-09-27 | $0.07 | — | — | — |
| 2017-03-29 | $0.07 | — | — | — |
| 2016-09-28 | $0.07 | — | — | — |
| 2016-03-29 | $0.07 | — | — | — |
| 2015-09-28 | $0.06 | — | — | — |
| 2015-03-27 | $0.06 | — | — | — |
Deep Analysis
Risk : Reward — upside vs downside from this company's own quarters
Not computed yetNarrative Economics
market-narrative step).
Claude Reading
Looking at the raw numbers first: FY2026 revenue of $55.82B with NI of $7.35B is a genuine step-change from the FY2023 print of $27.15B with a small loss, but the trajectory is not what it looks like. Japanese megabank ADR revenue is heavily distorted by yen translation and by how trading/securities gains flow through the top line — the "1.7% revenue CAGR" and "95% earnings YoY" are both artifacts more than fundamentals. What matters for a Japanese megabank is NIM expansion off BoJ normalization, credit costs, and fee income — none of which is legible from this file. ROE of 10.1%, ROA of 0.39%, and P/B of 1.82 are the numbers that actually matter, and they tell a coherent story: the market is already paying up for BoJ normalization optionality. A Japanese megabank historically traded at 0.5-0.7x book; 1.82x is not a discount, it's a premium reflecting rate-hike enthusiasm.
The operating cash flow of -$11.31B is a nothingburger — for banks, OCF is dominated by changes in loans/deposits and is not a "quality" signal the way it is for industrials. The synthesis flagging "poor cash flow quality" and the secondary signals piling on this is a category error; the classification engine correctly tagged this as a mature earner but the downstream cash-flow scoring is misapplied to a bank. That said, the Market Forces call — that earnings recovery is being flattered by reserve releases and Mizuho is the weakest of the three megabanks — is the sharpest observation in the stack and deserves weight. Mizuho has consistently underperformed MUFG and SMFG on ROE and cost/income ratio; being the #3 in a three-horse race with structural share loss is a real overhang.
The synthesis verdict of fair value with a $12.21 composite / $11.61 signal-adjusted vs $10.70 spot (+8.6% upside) is directionally defensible but the confidence is overstated. A contrarian would argue three things: (1) 1.82x book for a 10% ROE bank in a country with structurally lower terminal ROEs than US peers is not cheap — MUFG trades at similar multiples with better returns, so relative value inside Japan favors peers; (2) the BoJ rate-hike thesis is largely priced in — the sector has already rallied hard on this, and further upside requires actual NIM delivery, not more anticipation; (3) the 1.78% dividend yield is thin for what is essentially a low-growth financial — you can get 4%+ from US regionals with arguably better fundamentals if you're indifferent to Japan exposure. The narrative layer correctly identifies this as "fundamentals doing all the work" with minimal story premium, which I agree with — there is no cult, no myth, just a slow bank at a reasonable price.
I partially dissent from the fair-value synthesis and lean mildly overvalued. The 8.6% upside to composite fair value is inside the error bars of the FX translation and reserve-release noise; you're not being paid to take Japan-specific risk here (BoJ policy reversal, yen strength wiping out ADR returns, corporate lending share loss to SMBC/MUFG). The insider activity is all routine share awards — no signal. My fair-value anchor is closer to $10-11 (roughly current price), not $11.61, once you haircut for the reserve-release quality issue Market Forces raised and Mizuho's structurally lower ROE vs peers. If you want Japanese bank exposure on the rate-normalization thesis, MUFG or SMFG are cleaner expressions of the same trade. Mizuho is a hold at $10.70, not a buy — the +8.6% signal-adjusted upside is not enough compensation for being long the weakest of three peers in a mature market where the primary catalyst (BoJ) is largely reflected. Would get more interested below $9 (roughly 1.5x book), or if a specific ROE-improvement catalyst emerges (cost restructuring, capital return acceleration).
GPT Reading
At $10.7, the market is valuing Mizuho at about $130.4B against $72.9B of equity, or 1.82x book, for a bank producing just 10.1% ROE and 0.39% ROA. That is the first thing that jumps out: the stock is not cheap on the balance-sheet metric that matters most for banks. A 10% ROE bank can justify roughly book to modestly above book, but paying nearly 1.8x book usually requires either structurally higher returns, confidence in sustained ROE expansion, or superior capital distribution. The earnings rebound is real in the reported numbers—net income rose from $3.77B to $7.35B in 2026, up 95%, on revenue growth from $51.85B to $55.82B—but the longer history is messy: net income was negative in 2022 and basically flat to negative in 2023 before snapping back. That is not the profile of a premium-multiple compounder.
The quality of those earnings is where I part company with any simple “mature earner” comfort. Operating cash flow was negative $11.31B and free cash flow negative $13.25B in the latest year. Yes, bank cash-flow statements are often noisy and less decision-useful than industrial companies’ because deposit and funding flows swamp operating CF, so I would not mechanically treat negative FCF as distress. But when a bank trades at 16.7x earnings and 2.38x sales, I want a cleaner profitability story than “ignore the cash flow, trust the accounting rebound.” Revenue also tells a strange story: annual revenue went from $13.42B in 2022 to $27.15B in 2023 to $54.02B in 2024, then dipped to $51.85B before recovering to $55.82B. That doubling and redoubling likely reflects accounting or translation effects more than a true economic step-change, which makes revenue-based valuation metrics close to useless here and undermines confidence in model outputs built on those figures.
The balance sheet is not alarming in the way a non-financial would be—$397.36B of cash against $132.26B of debt sounds strong, but for a bank these are operating structure numbers, not excess liquidity one can capitalize at face value. What matters is whether the franchise can convert its enormous asset base into better returns, and 10% ROE says “adequate,” not “premium.” A 1.78% dividend yield also does not compensate investors for paying a rich multiple into a rate-cycle thesis that is still largely macro-driven. If this were at 1.1x-1.3x book, I could underwrite normalization and modest upside; at 1.82x book, much of the Japan-rate-normalization optimism looks already embedded. I think fair value is lower than the model’s $11.6-ish synthesis because that framework appears too forgiving about the disconnect between reported earnings momentum and underlying earning power.
The best pushback is straightforward: earnings did not just improve, they nearly doubled, and revenue reached a new high at $55.82B; if Japan is finally exiting a decades-long ultra-low-rate regime, large domestic banks like Mizuho can see meaningful net interest income tailwinds without needing heroic loan growth. On that view, a 10% ROE may be a waypoint, not an endpoint, and 1.82x book may be tolerable if ROE moves toward the low-to-mid teens. The 2025-to-2026 jump from $3.77B to $7.35B in net income is exactly the kind of inflection investors try to get ahead of, and the stock is not outrageously expensive on headline P/E if earnings continue to rise. I take that seriously. But I weigh it differently because the historical profit volatility is too severe, the accounting presentation is too noisy, and there is not enough evidence in this dataset that the current earnings level is durable enough to underwrite a premium to book this high.
What would change my mind is evidence that 2026 was not a one-year earnings spike. Specifically, if Mizuho can sustain or exceed roughly $7.5B-$8.0B of annual net income while lifting ROE from 10.1% toward 12%-13% without a deterioration in equity, I would be more open to the current valuation and even modest upside above $11. Conversely, if the next annual result slips back toward the $4B-$5B range, or if book value growth stalls while the stock remains around 1.8x book, the shares should derate materially. For me, the key proof points are sustained ROE expansion and cleaner evidence that reported earnings are translating into durable shareholder value, not just another cyclical or accounting-driven bounce.
Grok Reading
Mizuho’s reported trajectory from the raw annuals is a clear rate-cycle and accounting rebound story, not a structural growth compounder. Revenue climbed from $13.42B in fiscal 2022 to $55.82B by fiscal 2026, with net income swinging from a $665M loss to $7.35B; the latest year alone nearly doubled NI from $3.77B and produced a 13.2% net margin and 10.1% ROE on $72.85B of equity. At $10.7 and a $130B market cap that equates to 16.7x trailing earnings and 1.82x book—multiples that look full for a third-place Japanese megabank yet still sit modestly below the synthesis fair-value range of $11.61–$12.21. The balance sheet is fortress-like on a cash basis ($397B cash versus $132B debt), which is why EV-to-revenue prints negative and should be ignored; ROA of 0.39% is simply the arithmetic of a giant, low-margin balance sheet and is normal for the sector. What stands out is the contradiction between the momentum layer’s 1.7% revenue CAGR and the actual five-year step-change in the income statement: the CAGR figure is either mis-specified or diluted by earlier NIRP years, while the recent 7.7% revenue and 95% earnings jumps are the numbers that matter for the current earnings power.
Cash-flow quality is the loudest red flag in the briefing and cannot be dismissed lightly. Operating cash flow of –$11.31B and FCF of –$13.25B in the same year the bank reported $7.35B of net income is a classic bank working-capital distortion (loan growth and deposit flows swamp the P&L), yet the secondary signals correctly flag it as poor and note that reported earnings may embed reserve releases rather than core franchise strength. Combined with the market-forces commentary that Mizuho is losing share as the weakest of the three megabanks, the 10% ROE looks more cyclical than durable. Dividend yield of only 1.8% further undercuts any pure income thesis; investors are essentially paying 1.8x book for Japan rate-normalization optionality and little else.
The strongest opposing case is straightforward: at 16.7x earnings and 1.82x book the stock already discounts a permanent step-up in Japanese net interest margins, leaving no margin of safety if the BOJ pauses, the yen weakens further, or credit costs normalize higher. Negative free cash flow, below-benchmark sector positioning, and the explicit warning that “market prices in more growth than projected” all argue the $10.7 price is closer to fair than cheap; a smart skeptic would simply note that the entire upside to the $11.61 signal-adjusted value is only 8–9% before Japan macro and competitive risks, which is not enough compensation. I weigh those points seriously but still come out modestly constructive because the absolute earnings base has reset higher, tangible book is well-covered by cash, and Japanese megabank ROEs have historically compressed far below 10%—so sustaining even high-single-digit returns would justify today’s multiple.
What would flip the view is a pair of clean quarters showing NI falling back toward the $4B run-rate, ROE compressing below 7–8%, or continued double-digit negative operating cash flow without offsetting loan growth. Conversely, a sustained 11%+ ROE with positive FCF and any dividend hike above 3% would make the stock clearly cheap below $11.
Qwen Reading local experiment · off-panel
The single most important thing in this data set is that the revenue series is almost certainly broken. Going from $13.42B in FY2022 to $55.82B in FY2026 is a 4.16x jump in four years, yet the momentum model reports a 1.7% revenue CAGR — a number that is arithmetically impossible given those endpoints. What I suspect is a mix of IFRS reporting changes, ADR consolidation adjustments, and yen/USD translation swings (the yen fell from roughly 115 to 150+ over this window, which alone would inflate USD-reported revenue by 30%+), layered on top of what may be a genuine expansion in fee and trading income as BOJ rates normalized. The 7.7% recent revenue YoY is the only number in that series I'd trust, and even that is suspect given the 2025 dip to $51.85B before the 2026 rebound to $55.82B. The "Moderate Revenue Confidence" and "insufficient_data" quarterly flags confirm the data pipeline is struggling here. Any DCF or multiple-based valuation built on this revenue line is built on sand, and the negative EV/Revenue of -2.37 is a pure artifact of subtracting $397B of bank "cash" (really, deposits and central bank reserves) from market cap — it tells you nothing.
What I can actually anchor on is the earnings and balance sheet. Net income went from a $664.7M loss in FY2022 to $7.35B in FY2026, and the 95.2% YoY jump from $3.77B to $7.35B is real, but it's off a base that was still depressed by pandemic-era provisioning and the tail end of negative-rate drag. The 10.09% ROE is the number that matters, and it's unremarkable — a U.S. money-center bank would be embarrassed by it, but for a Japanese megabank it represents a genuine step up from the 2-4% ROE territory of 2020-2022. At 1.82x book, the market is paying a modest premium to the historical 0.7-1.2x book range Japanese banks traded at for two decades, and that premium is justified only if ROE holds above 10% as BOJ continues to normalize. The 16.7x P/E is in line with the other two megabanks (MUFG and SMFG trade in a similar band), so Mizuho is not commanding a multiple premium or discount to its domestic peers. The 1.78% dividend yield is thin — this is not an income play.
The prior models' "Poor Cash Flow Quality" flag and the -$11.31B operating cash flow / -$13.25B FCF are, in my view, a category error. Applying a non-financial free-cash-flow framework to a bank where the "cash" line is $397B of deposits and reserves, where "capex" of $1.95B is IT and branch spend, and where the operating cash flow statement is dominated by loan origination and deposit flows, produces a number that is technically correct and analytically meaningless. I would not let that flag drag my valuation. What I would worry about, if I were being adversarial, is the "Market Forces" model's specific claim of "unsustainable reserve releases" driving the earnings recovery. If a meaningful chunk of the $7.35B net income is the unwinding of pandemic-era loan-loss provisions rather than genuine NIM expansion, then the 10% ROE is a peak, not a floor, and 1.82x book is rich. I can't verify that from this data set, but it's the single most dangerous assumption in the bull case.
The strongest case against my "fairly valued" read is the combination of the "Below Sector Benchmarks" flag, the "Macro Headwinds" tag, and the insider-transaction data. The insider line shows ten transactions on 2026-07-01, all option exercises and returns-to-issuer — routine tax-cover mechanics, not directional signals, and the "No Insider Transactions" secondary signal is simply a processing error that contradicts the raw data. But the sector-benchmark miss is harder to dismiss: if Mizuho is the third-place megabank losing share to MUFG and SMFG on the commercial lending book, and to digital lenders on the retail book, then the 1.82x book multiple is paying for a franchise that is slowly eroding. The "steady-compounder" narrative archetype with "minimal" intensity and a "low" cult coefficient is accurate — there is no story here, no disruption angle, no activist overhang. You are buying a 10% ROE Japanese bank at 1.8x book and hoping BOJ doesn't reverse course. That's a reasonable position, but it's not a compelling one at $10.70.
What would flip me: a quarterly print showing ROE above 12% with NIM expansion (not reserve releases) as the driver, which would make 1.82x book look cheap and push my target toward $13-14. Conversely, a BOJ pause or reversal that stalls NIM improvement, combined with a peer (MUFG or SMFG) printing 13%+ ROE, would make Mizuho's 1.82x book look like the expensive one in the group and I'd trim. A yen move above 140 would also mechanically improve the ADR earnings line and could re-rate the stock 5-8% on translation alone. None of those are in the data in front of me, so I'm left with a middle-of-the-pack Japanese megabank at a middle-of-the-pack multiple, with a revenue data series I can't fully trust and a cash-flow statement that doesn't apply to the business model.
Big-3 Panel — where each AI stands
Claude — Second Response · overvalued changed direction
VERDICT: Mildly overvalued at $10.70; fair value $9.50-10.50 anchored on 1.3-1.5x book for a durable 10% ROE, with sub-$9 the entry point and MUFG/SMFG the better expressions of the same BoJ trade.
GPT's argument on the P/B math is the one that made me re-examine, and I want to name it precisely: paying 1.82x book for a 10% ROE bank with a volatile earnings history (negative in 2022, near-zero 2023, $3.77B in 2025, $7.35B in 2026) requires believing the 2026 print is a new base, not a peak. That is exactly the reserve-release quality question I raised in my first reading, and GPT frames it more sharply than I did — the Gordon-growth math (P/B ≈ (ROE-g)/(r-g)) supports ~1.0-1.3x book at sustainable 10% ROE with Japanese cost of equity around 8-9%, not 1.8x. So 1.82x is pricing in ROE convergence toward 12-13%, which is a forecast, not a fact.
Grok's counter — that Japanese megabank ROEs have historically compressed *below* 10%, so sustaining 10% would itself justify the multiple — is the argument I need to answer, and I don't find it fully persuasive. "Better than a bad historical baseline" is not the same as "cheap"; MUFG and SMFG are hitting similar or better ROEs at comparable multiples, so the relative-value case inside Japan still favors peers, and Grok never really engages with the #3-of-3 structural point. His constructive lean rests on the earnings base having reset higher, but that is precisely what the reserve-release critique disputes.
Net: I hold overvalued but tighten the range. GPT's $8.50-9.50 is too punitive — it under-weights that Japanese cost of equity is genuinely lower and BoJ normalization is a real (if partially priced) tailwind. My first-read $9 entry level and $10-11 fair-value anchor survive both critiques intact. I'm not drifting toward Grok's constructive view because his ROE-durability argument doesn't clear the peer-comparison hurdle, and I'm not drifting all the way to GPT's deeper discount because the rate cycle is real.
GPT — Second Response · overvalued revised
VERDICT: Modestly overvalued at $10.7; fair value is closer to $9.50-$10.50 today, with upside above $11 requiring sustained 11-12%+ ROE rather than a one-year rebound.
Claude’s critique moved me more than Grok’s optimism, but not enough to change direction. The key point is not just that bank cash flow is noisy — I already said that — but that the *relevant* valuation frame is relative returns on equity and book, and on those numbers Mizuho still looks stretched. A 10.1% ROE at 1.82x book is hard to defend for the structurally weaker #3 megabank, especially if MUFG/SMFG offer cleaner exposure to the same BoJ normalization theme. Claude’s point that the apparent revenue/earnings “step-change” is likely flattered by translation, securities accounting, and possibly reserve-release effects reinforces my original skepticism that the 2026 print should be capitalized at face value. That narrows the argument from “overvalued because cash flow is bad” to the stronger one: overvalued because the market is paying a premium multiple for a still-middling return franchise.
What did *not* move me in Grok’s case is the idea that “earnings power has reset higher” enough to justify ~$12. That could be true, but the evidence in this file is too thin. If the upside to fair value is only 8-9% before FX, policy, and competitive risk, that is not a compelling undervaluation case; it is within normal estimation error for a bank whose reported earnings history includes losses, volatility, and unclear top-line comparability. I do agree with Grok that the negative OCF/FCF should not be treated as a standalone red flag, so I would de-emphasize that from my first read. But once that is stripped out, the conclusion barely changes: this is still a roughly 10% ROE bank being valued like a meaningfully better one. I’ll revise only in degree, not direction — from clearly overvalued toward modestly overvalued / rich-fair, with fair value centered around $9.50-$10.50 unless ROE proves sustainably above 11-12%.
Grok — Second Response · fairly valued changed direction
VERDICT: Fairly valued at $10.7; ~$10.5-11.2 range after haircutting #3 position and already-priced BoJ gains
The colleagues’ emphasis on price-to-book discipline is the argument that shifted me. Claude’s observation that Japanese megabanks historically cleared 0.5-0.7x book, and that 1.82x today already embeds BoJ normalization for a 10% ROE franchise, is sharper than I initially weighted; GPT’s parallel point that a 10% ROE name normally supports only book-to-modest-premium valuation (absent clear path to 12%+) further undercuts the ~$12 earnings-power anchor I used. Both correctly flag that the 8-9% model upside sits inside FX, reserve-release, and competitive-share noise, especially given Mizuho’s persistent #3 status versus MUFG/SMFG. I underweighted that relative-value and structural-position drag.
Cash-flow negativity remains a non-issue for a bank, as Claude notes, so that red flag does not move me. The absolute earnings reset to $7.35B and fortress cash position still provide a floor near current levels, preventing me from chasing GPT all the way to an $8.50-$9.50 target. Yet once the P/B premium and thin 1.8% yield are squared against limited further catalysts, the stock looks closer to fairly valued than modestly cheap; any residual upside is optionality, not margin of safety. I would still watch for sustained ROE above 11% or a sub-$9 entry, but at $10.7 the risk-reward no longer skews positively enough to call it undervalued.
Advanced Analysis Forensic deep-dive · separate lenses
Mizuho is one of Japan's three megabanks, and most of the 'forensic red flags' here are artifacts of applying industrial-company screens to a bank. Altman Z at 0.09 is meaningless for a deposit-taking institution (banks always look distressed on Z because they are structurally leveraged by design), and FCF is not a useful metric for a bank whose 'cash flows' are loan book expansion, deposit flows, and securities positioning. The 397B of liquid cash vs 130B market cap simply reflects that this is a bank holding company - customer deposits and reserves dominate the balance sheet.
Verify before trusting this (7)
- CET1 capital ratio and regulatory capital cushion vs Basel III requirements
- Non-performing loan ratio and loan-loss reserve coverage
- Net interest margin trend and sensitivity to BOJ policy normalization
- Cost/income ratio trajectory (Japanese megabanks historically 60-70%)
- Cross-shareholdings unwind progress and unrealized securities gains/losses
- Dividend policy and payout ratio stability
- Exposure to US commercial real estate and offshore dollar funding gaps
The valuation math is tight and unheroic: anchored-PE lands at $12.21, the composite fair value at $12.21, and the signal-adjusted FV at $11.61, versus a $10.69 price. That is a 9-14% gap - a real discount, but well inside the noise band for a Japanese megabank whose earnings are sensitive to BoJ policy and yen moves. The Solid quality read and Adequate/Mixed earnings quality do not argue for pushing deserved value higher than the anchor already implies.
Verify before trusting this (5)
- BoJ rate path and its pass-through to Mizuho's NIM in the next two quarters
- Segment detail on fee income vs net interest income durability
- Any one-off gains inflating the recent earnings recovery
- Dividend and buyback pace vs guidance
- FX translation impact on USD-quoted ADR
MFG is a classic sentiment-neutral name right now. The tape is modestly risk-on (+46) but with a 0.38 beta and a steady-compounder archetype, very little of that lift actually reaches the stock; risk-on rallies typically bypass Japanese megabank ADRs in favor of higher-beta growth. At the same time, there is no active bear narrative pressing on the shares - narrative intensity is minimal, cult coefficient is low, and the story is durable but thin. The market is neither cheering nor punishing this name. Macro cross-currents are the only real forces. Higher US rates (10y 4.63%) and a stretched market PE (27.7) are a mild headwind for all equities, but for a Japanese bank the more relevant macro axis is BoJ policy and the yen - both ambient uncertainties, not acute pressures today. Recent price action (7.7% vs 1.7% long-term CAGR) shows a quiet drift higher, consistent with income-seeking flows rather than narrative-driven buying. Net: a mild positive drift from momentum and risk-on tape, offset by the absence of any story to attract capital and by structural skepticism toward Japanese bank profitability. Balanced, leaning very slightly constructive.
Verify before trusting this (4)
- Any BoJ policy shift or rate-normalization signal that would flip the Japanese-bank narrative
- USD/JPY direction - a weaker yen hurts ADR translation regardless of underlying performance
- Whether analyst tone or target revisions start clustering (currently thin coverage energy)
- Rotation into Japanese financials as a diversification theme vs continued US mega-cap concentration
This lens hasn't been run for this ticker yet.
When we made this prediction on Aug 7, 2026, MFG was $10.69. We expect it to be $11.45 by Feb 2027, and we consider it great value under $9.50. This is an early model (v0.6.0) — the direction is more reliable than the exact price. Made Aug 7, 2026.
Blue is our prediction, starting the day we made it. Grey is a slower route to the same place — the same destination, taking longer. Black is the actual price, so you can see how we are doing.