Financebotresearch desk研究台

Discovery › screen

Screen global vs us value

Date: 2026-07-29 Criteria: Look outside the US for value, then re-examine America. Anything worth owning? Method: Spiral search with geography/cap-tier as the distance axis (screen.md) Framework: analysis_notes.md §0 — and note that the trap test applies at the country level too. Markets are cheap for reasons.


Executive Answer

The cheapest equity risk on the board is in America, and it always was.

Ring Market Trailing P/E Off high
0 SPY (cap-weighted S&P) 26.2x −4.1%
1 RSP (equal-weight S&P) 22.3x −1.1%
1 VTV (large value) 21.5x −1.7%
2 IJH (mid) 21.6x −3.3%
2 IWM (small) 17.9x −4.7%
2 IWN (small VALUE) 14.9x −1.3%
3 EWJ Japan 17.8x · 3.81% yield −8.4%
3 EWG Germany 17.7x −6.1%
3 VGK Europe / EWU UK 18.4x −2.1%
3 EFA developed ex-US 18.8x · 3.24% −2.4%
3 EWC Canada / EWA Australia 21.3x / 21.5x −1.3% / −4.5%
3 EWL Switzerland 24.5x −4.2%
4 EWZ Brazil 11.4x · 4.24% −15.6%
4 EWW Mexico 12.9x · 3.28% −7.4%
4 EWY Korea 14.3x −34.7%
4 INDA India 22.2x −11.4%
4 EWT Taiwan 24.9x −20.7%

Three findings, in order of usefulness:

  1. US small-cap value at 14.9x versus the S&P at 26.2x — an 11.3-point spread inside a single market, with no currency, governance, jurisdiction or custody overlay to underwrite. This is the answer to "back to America."
  2. Nothing abroad is cheap enough to pay for its own structural discount — except Japan, and Japan's cheapness has migrated from the multiple into shareholder returns.
  3. Taiwan is the one to actively avoid: 24.9x trailing and −20.7% off its high. Expensive and falling — Korea's problem at a worse price.

Ring 2 — US Small/Mid Value ⭐ (the find)

Same market, different cap tier — the direct analogue of "same sector, different cycle" that paid off in the last screen.

Small/large forward P/E gap near its widest in two decades — S&P ~21-22x vs Russell 2000 ~14-15x = a 30-35% discount
Discount at the start of 2026 31%
Russell 2500 vs S&P 500 (fwd, 12/31/25) 18.5x vs 23.0x
Small-caps as % of Russell 3000 (6/30/26) 4.5% vs a 7.6% long-term average — unprecedented large-cap concentration
Consensus NTM earnings growth Russell 2000 +43% · Russell 2500 +18% · S&P 500 +11%

⚠️ The honest caveat, and it's material: this trade is already in motion. The Russell 2000 is up ~20% YTD and outpacing the S&P 500 in 2026. IWM sits only −4.7% off its high. You are not catching this early — you are joining a rotation that began months ago, and the +43% earnings-growth consensus is exactly the kind of number that gets revised down.

What this does not mean: more micro-caps. The portfolio already runs a deep speculative sleeve (Output/Microstocks/ — 24 names). Ring 2 here is quality small/mid at value multiples, which is a different asset from a pre-revenue story stock. Conflating the two would be the single easiest way to misread this screen.


Ring 3 — Developed ex-US: Japan Is the Only Real Case

Europe (18.4x), UK (18.4x), Canada (21.3x), Australia (21.5x) and Switzerland (24.5x) are all fairly valued with no identifiable catalyst and sit within 5% of their highs. Nothing to do.

Japan is different, and the reason isn't the P/E.

Metric Japan US
Forward P/E (TOPIX) 14.7x ~16% higher
Price / book 1.7x >2/3 higher
Companies below 1x book (top 2,000) ~40% ~10%
Cash / market cap highest among developed markets

The catalyst is capital returning, not multiples expanding:

  • Announced buybacks: ¥10.1T (FY2023) → ¥19.0T (FY2024) → a record ¥22.3T (FY2025). More than doubled in two years.
  • The Corporate Governance Code review lands June 2026 — the first in six years — explicitly targeting the cash hoarding that gives Japan the developed world's highest cash-to-market-cap ratio.
  • Cross-shareholding unwinds run through 2030, with megabanks and large insurers committed sellers. Proceeds fund buybacks.

Per §0's standing note that "growth comes in many forms — capital appreciation, dividends, buybacks, multiple re-rating", this is growth in the buyback-and-payout form, and it is contractual-ish rather than speculative.

⚠️ But be honest about what's already priced: Japan's P/B at 1.7x is a post-2008 high, and the Prime market's P/B has already expanded 35-40% since the 2022 TSE restructuring. The easy money was made. What's left is the second leg — the cash-release leg — which is slower and more grinding than the re-rating leg.

Japanese ADR candidates

Ticker Company Price %off High fwd P/E ttm P/E Yield Target Read
TM Toyota $192.84 −22.5% 12.2x 10.7x 3.36% +20% 🟢 The pick. World's largest automaker, cheapest quality name here, strong_buy.
HTHIY Hitachi $32.21 −17.4% 13.0x 29.8x 1.06% +20% 🟡 strong_buy, real restructuring story — but Hitachi Energy makes it partly an AI-grid derivative, which double-counts exposure already owned.
KUBTY Kubota $87.32 −19.1% 13.3x 17.5x 1.85% +4% 🟡 Ag equipment. Cheap, but target ≈ spot.
MITSY Mitsubishi Corp $599.80 −28.2% 15.9x 16.9x 2.41% +63% 🟡 Buffett trading house. Large implied upside, but no analyst coverage consensus (none) — verify the target before trusting it.
ITOCY Itochu $12.35 −18.2% 15.8x 2.16% 🟡 Same category, thinner data.
SONY Sony $23.27 −23.3% 19.6x 22.2x 0.70% +26% 🟡 Not cheap enough to be the reason to enter Japan.
TKOMY Tokio Marine $51.00 −1.0% 13.0x 29.8x 2.70% +16% 🔴 Already re-rated.
MUFG Mitsubishi UFJ $21.94 −5.6% 21.3x 16.9x 2.33% −5% 🔴 Target below spot. Rate-normalisation winner already priced — same pattern as the Korean banks.

🚩 Data-quality flags — three of twelve quotes are unreliable

  • MARUY (Marubeni) prints "−92.6% off high" and a 1.6x forward P/E. That is an ADR ratio artifact, not a crash. Do not act on it. I'm naming it rather than deleting it because it's exactly the kind of number a screen would otherwise surface as a spectacular find.
  • SMFG shows 60.9x forward against 16.5x trailing; MFG shows 7.1x forward against 16.1x trailing. Japanese bank ADR forward estimates in this data source are mismatched to ADR ratios. Use trailing only for Japanese bank ADRs, or price the local lines.

Ring 4 — Emerging: Cheap for Reasons

Market P/E Verdict
Brazil (EWZ) 11.4x, 4.24% yield, −15.6% 🔴 Cheapest board-wide — and the textbook §0 trap. Currency, political and rate risk are the reason for the discount, not an anomaly within it. Brazil has been "cheap" for a decade.
Mexico (EWW) 12.9x, 3.28% 🟡 The most defensible EM case — nearshoring is a real structural tailwind — but tariff/USMCA renegotiation risk is a live binary, and the portfolio has no mechanism to hedge it.
Korea (EWY) 14.3x, −34.7% 🔴 Already resolved in the prior screen: buying the index means buying ~60% Samsung + SK Hynix, i.e. buying the epicentre.
India (INDA) 22.2x 🔴 Expensive. Best structural growth story in EM, entirely priced.
Taiwan (EWT) 24.9x, −20.7% 🔴 Actively avoid. Expensive and crashing — semi concentration without Korea's valuation reset.

Nothing in ring 4 clears the bar. Every cheap market here is cheap for an identifiable structural reason that a US-based fundamental investor cannot underwrite or hedge.


Spiral Summary

Ring Distance Yield
0 · US megacap tech 🔴 Already over-owned; already screened
1 · US large non-tech / equal-weight 1 🟡 RSP 22.3x vs SPY 26.2x quantifies the megacap premium — no action
2 · US small/mid value 2 🟢🟢 IWN 14.9x — the find, but partly consumed (+20% YTD)
3 · Developed ex-US 3 🟢 Japan only — TM the single best name. Europe/UK/Canada/Australia/Swiss: nothing
4 · Emerging 4 🔴 All trap or expensive. Taiwan the one to avoid outright

Payoff peaked at ring 2 again — the same place it peaked in the tech-crash screen. Two independent spirals now agree that the second ring out is where value lives: close enough that the causal link is verifiable, far enough that the crowd isn't there.


Portfolio Fit

Current non-US exposure: TCEHY + BABA (China, structural-risk sleeve) · SAP (Germany ADR) · NVO (Denmark) · TRI (Canada) · ASML (watchlist, unowned)

Gaps this screen would fill: - Zero Japan exposure in either account — the one developed market with an identifiable, funded, multi-year catalyst - Zero Latin America — and after this screen I'd leave it that way - Beta is 78% technology and both accounts are US-centric

TM fills three gaps simultaneously — non-US, non-tech, and income-producing (3.36%) — which is unusual for a single position and is the main argument for it over the alternatives.


Recommendations

Add to Watchlist:

Ticker Sleeve Suggested entry Rationale
TM (Toyota) 📈 Income — Yield Tomorrow Attractive at $193; add $170-185 10.7x trailing, 3.36% yield, −22.5% off high, strong_buy, +20% target. Japan governance tailwind. ⚠️ Auto cyclicality + tariff risk are the real knocks; hybrid strategy vindicated as EV adoption slowed.
MITSY (Mitsubishi Corp) 🔧 Re-Rating Plays Verify target first Buffett trading house, −28.2%, 15.9x — but the +63% implied upside comes with no analyst consensus. Needs an /analyze before belief.

Do not add: MUFG/SMFG/MFG (rate normalisation priced, targets at or below spot), TKOMY (re-rated), MARUY (data artifact), anything in ring 4, EWT/Taiwan.

On US small-cap value: the asset class call is sound but I'm not recommending a name yet — screening quality small/mid value properly is its own run, and it must be kept distinct from the existing 24-name Microstocks sleeve. /screen quality small and mid cap value, profitable, low debt, non-speculative would do it.

Next steps

  1. /analyze TM — the one name here that clears both §0 tests and fills three portfolio gaps at once. My recommendation.
  2. /screen quality small/mid cap value — convert the ring-2 finding into names. Highest expected value of the three, but a fresh run.
  3. /analyze MITSY — only if you want trading-house exposure; resolve the target discrepancy first.

Bottom Line

  1. You did not need to leave America. US small-cap value at 14.9x against the S&P's 26.2x is the widest small/large gap in two decades, and it carries none of the currency, governance or jurisdiction risk that makes Brazil, Korea or Taiwan look cheap.
  2. But that trade is partly consumed — the Russell 2000 is already +20% YTD and outperforming. Joining is reasonable; calling it undiscovered is not.
  3. Japan is the only foreign market worth the effort, and the thesis is buybacks rather than multiples — ¥22.3T record repurchases, the first governance-code review in six years landing this June, and cross-shareholding unwinds funded through 2030. P/B at a post-2008 high means the re-rating leg is done; the cash-release leg is what's left.
  4. TM (Toyota) is the single best individual name found — 10.7x trailing, 3.36% yield, −22.5% off high, and it fills the Japan, non-tech and income gaps in one position.
  5. Three of twelve Japanese ADR quotes were unreliable (MARUY's "−92.6%" is an ADR artifact; SMFG/MFG forward P/Es are mismatched). Flagged rather than silently dropped, because the artifact would otherwise read as the best find on the page.

Screen run 2026-07-29, rings 0-4. Prices live. Related: screen-tech-crash-hidden-value-2026-07-29.md · watchlist-scan-2026-07-29.md

Sources: Japan governance reform & buyback records · Japan valuation vs US / sub-1x book · TOPIX forward P/E and P/B · TSE reform decade review · Russell 2000 vs S&P valuation gap · Russell 2000 outperformance 2026 · Small-cap market overview