Key points
- 01The central benefit of international investing is access to distinct industries, currencies, and economic exposures—not merely a larger stock count
- 02The same factor has a different meaning across accounting regimes, rates, industry mixes, and trading costs, so local peer groups are necessary
- 03A global company should be tested through business, financial, market, and execution lenses, including disclosure, FX, withholding tax, and liquidity
- 04A concentrated ten-stock result may be driven by one winner and should be reported with equal-weight comparisons, attribution, and drawdown
- 05No allocation method is a permanent winner; a strategy that survives adverse periods is more useful than the highest fragile backtest return
Chapter 1. Use criteria, not intuition, to search thousands of stocks
Why consider international stocks
Global investing can add businesses and economic exposures missing from a domestic portfolio while reducing dependence on one policy regime and currency. The SEC also identifies additional risks: currency, information access, political events, liquidity, legal remedy, and higher costs. A local-market gain can become a loss after currency conversion.
Different domiciles do not guarantee diversification. Companies tied to the same customers, dollar funding, semiconductor supply chain, or AI capital-expenditure cycle may move together.
Four lenses for each company
The business lens examines the customer problem, market, moat, pricing power, and reinvestment. The financial lens tests cash conversion, debt, dilution, and capital allocation. The market lens ranks value, growth, momentum, quality, and estimate revisions within local peers. The execution lens checks language, reporting lag, currency, tax, custody, trading hours, and liquidity.
Comparing a U.S. multiple directly with an emerging-market multiple can confuse country risk and local rates with company-specific cheapness. Rank locally first, then account explicitly for country capital costs.
Chapter 2. Validate a live portfolio
January selection: ten names chosen by factors
A genuine January test uses only data available at that month's start, preserves the then-current universe including later delistings, and screens accessibility and turnover first. Each row should retain component scores, F-score, value range, currency, country and sector exposure, next review date, and falsification condition.
This article intentionally avoids a hindsight list of ten past winners. Reproducibility is more valuable than a list contaminated by survivorship and look-ahead bias.
Concentrated performance and lessons from failure
Score concentration can beat equal weighting when the model is right, but it also magnifies model error, industry crowding, trading halts, and FX shocks. Report total return with top-one and top-three contribution, drawdown, worst month, beta, and currency attribution.
Failures should be classified as signal failure, stale or incorrect data, business-thesis failure, or execution-cost failure. Outcomes and decision quality must remain separate because sound decisions can lose and poor decisions can profit.
What excess return over the S&P 500 and Nasdaq means
The S&P 500 is a float-adjusted market-cap portfolio of 500 U.S. large companies. The Nasdaq-100 is concentrated in large non-financial Nasdaq listings. Their industry and concentration profiles differ, so a strategy's alpha changes with the benchmark. A multi-country portfolio also needs a global investable benchmark, a cash return, and a common reporting currency.
Chapter 3. Factor weights and five allocation strategies
A June scorecard that can be audited
Monthly reporting should separate local-currency and base-currency return, dividends, FX, withholding tax, and trading cost. Cumulative performance needs drawdown, turnover, holding period, and factor exposure beside it. A January-to-June win can reflect seasonality or one event; a frozen forward-test period is essential.
| Method | Allocation | Advantage | Primary risk |
|---|---|---|---|
| Equal weight | Same weight per stock | Simple and less sensitive to score error | Volatile small names can dominate risk |
| Score proportional | Weight by COMP distance | Expresses the model directly | Small scoring errors become sizing errors |
| Capped conviction | Concentrate with stock and sector caps | Balances edge and ruin control | Correlations can jump together |
| Risk balanced | Equalize estimated risk contribution | Reduces single-name risk dominance | Past volatility can understate future risk |
| Barbell | Broad index core plus selected sleeve | Retains market winners and a selection edge | Hidden overlap between the two sleeves |
High return is not always a good strategy
A useful strategy persists after cost, keeps losses tolerable, and can be explained and followed. An investor who abandons a strategy during its expected weak cycle will not realize its long-run average.
Contrarian craft: moving against momentum
Momentum contrarianism is not buying every decline. It selectively reassesses a company whose long-run quality and value thesis remains intact while short-run flows weaken. If estimates, cash flow, and credit risk deteriorate with price, the trade may be opposition to evidence rather than independent judgment.
A monthly operating rule
Archive finalized data and FX at month-end, calculate liquidity and ranks on the next business day, manually review top candidates, set sizing and order limits, then record actual cost and failed fills. Review attribution and failures quarterly without changing weights merely because recent performance was weak.
Global investing first requires making unlike data comparable
More countries create more data definitions. Fiscal year-ends, accounting standards, depositary ratios, holidays, and FX timestamps can make a precise score hide a false comparison.
A first global portfolio is easier to validate in two or three liquid, disclosure-rich markets than across every exchange. Separating an index core from a selected sleeve and applying currency, country, sector, and stock caps makes the sources of risk more visible.
The ten-stock, January, and June labels describe the structure of a reproducible operating ledger. They are not reported historical performance or security recommendations, and local tax and legal eligibility require separate review.
Sources
Related notes
Direction of MoneyThe craft and philosophy of finding market leadersDesigns a reproducible process for reading market, sector, and stock strength and ranking candidates with percentile-based multi-factor scores
Direction of MoneyReading the market through an investor's eyesExplains how a few long-run winners create most market wealth and how value, growth, momentum, and quality can frame a search for future leaders
Direction of MoneyTesting the financial health of market leadersConnects Piotroski's accounting signals with composite rankings, valuation ranges, and an integrated sell disciplineThis is a personal research note, not investment advice