Pence, Euros and One Portfolio: How the Europe Undervalued Screen Works

A note on authorship: The research, analysis, and opinions in this article are the author's own. Claude (Anthropic's AI) assisted with drafting and editing the prose.

📱 Prefer this on your phone? Get the free Android app.

The other five screens each live in a single market and a single currency — the ASX 300 in Australian dollars, the S&P 500 in US dollars, the NIFTY 500 in rupees — and never need to ask what a number is denominated in. Europe breaks that assumption. What follows is how the ranking stays currency-free, the one place conversion is unavoidable, and what the resulting list actually tells you about where value sits in Europe right now.

What the screen actually covers

The universe is 200 companies across 11 exchanges, spanning the eurozone and the United Kingdom. Roughly half of it is London-listed:

An honest disclosure: unlike the ASX 300 or Nikkei 225 — real indices someone else maintains — no single published index covers "the eurozone plus the UK" at this scale. This universe is hand-assembled: the full FTSE 100, DAX 40 and CAC 40, backfilled with the largest names from the AEX, FTSE MIB, IBEX 35, BEL 20, ISEQ 20, ATX, OMX Helsinki 25 and PSI-20. Switzerland, Norway, Sweden, Denmark and Poland are excluded — European, but neither the eurozone nor the UK, and each would add another currency regime for little gain. The screen is called "Europe" because that's what people search for; it is not a replica of any real index.

The pence problem

The two codes differ by a single character and by a factor of one hundred. Any calculation that reads GBp and treats it as pounds will be wrong by 100× — and, the dangerous part, wrong silently. Nothing crashes; you simply get a plausible-looking number that is completely false. Half of this screen's universe is quoted this way, so getting it right wasn't optional.

Why the ranking itself needs no currency at all

Now the part that makes the whole thing tractable. The screen ranks by margin of safety: how far below its estimated intrinsic value a stock trades, as a percentage. Both halves of that — the intrinsic value from the DCF or dividend discount model, and the market price — are in the same currency for any given stock. Divide one by the other and the currency cancels out.

The one place conversion is unavoidable

Every undervalued screen carries a simulated paper portfolio — a hypothetical stake spread equally across the list, rebalanced each time the screen refreshes. On the Europe screen that stake is €10,000, and that has to be one number in one currency: you cannot add a position denominated in pence to one denominated in euros and get anything meaningful. So conversion happens in exactly one place — the moment the portfolio rebalances — and nowhere else on the page.

The screen rebalances weekly, and we intend to keep following this exact list — same rules, same €10,000 starting stake, no cherry-picking — against the STOXX Europe 600 as the benchmark. A handful of weeks tells you nothing; the question worth answering is whether patiently buying what a conservative DCF or DDM calls cheap actually beats simply owning the broad index over years, not whether it looks good on any given Monday.

Two exchange rates, not one

The obvious approach — grab today's GBP/EUR rate, convert every UK holding, done — is wrong, and the reason is the design decision worth the most attention here. Between two rebalances, a UK holding's contribution to a euro-denominated pool changes for two independent reasons:

• the stock's own price moved, in pence; and
• the pound itself moved against the euro.

A single-rate conversion captures only the first. Convert both last period's and this period's holdings at today's rate and the currency movement mathematically cancels out — you'd report the stock's local-currency return while presenting it as a euro return. That is not what actually happened to a euro-based investor's money: they earned whatever the stock did in its own market, and whatever sterling did against the euro while they held it. Global fund fact sheets routinely split "local return" from "currency return" apart for exactly this reason — the two can point in opposite directions.

Dividends are income, and income has a currency too

The paper portfolio credits real dividends paid by its holdings, not just price movement — which matters more here than on the other five screens. Six of the twenty stocks that currently qualify are banks and insurers valued by dividend discount model, meaning the entire investment case for holding them is the dividend, not price appreciation. A screen that only tracked price return would be quietly ignoring the return those six names are actually built to deliver.

What the valuation mix tells you

Past the mechanics, the 20 names that currently qualify say something about where value sits in Europe right now.

Discount rates on individual names range from 5.0% to 8.9% WACC — wider than the region-level 3.2%/5.0% risk-free split alone suggests. Banks and insurers sit highest (BNP Paribas at 8.9%, Hannover Rück at 5.9%): funding-dependent balance sheets and regulatory capital carry a structurally higher cost of capital than an industrial or consumer-staples name with predictable, contract-backed cash flow. A wide margin of safety cleared against a high hurdle rate is a stronger signal than the same margin cleared against a low one.

Terminal value dominates the DCF names — 73% to 87% of estimated intrinsic value comes from cash flows beyond the explicit forecast horizon, not the near-term numbers the model can actually pin down, but a perpetuity built on one terminal growth assumption. That's a standing criticism of DCF generally, not a defect of this screen: nudge the terminal growth rate half a point and the "cheap" verdict can move a long way, which is exactly why the DCF calculator exposes that assumption for you to stress-test rather than burying it.

The list is concentrated, not diversified — eight of the twenty qualifying names, 40%, are Financial Services. That reflects where cheap valuations currently sit rather than a deliberate sector call, but it means the paper portfolio illustrates the screen's output, not a template for a diversified European allocation. Treat the list as a shortlist to research individually, not a ready-made basket.

Two currencies on the same page, deliberately

Each row in the ranking table shows its own native currency — p for a UK stock, € for a eurozone one, never converted — while the portfolio total is always €. That looks inconsistent and is intentional: a row should match what you'd see looking the stock up anywhere else, whereas the portfolio is a euro pool by definition.

The second half needed a specific fix. The instinctive way to pick the portfolio's currency symbol is to read it off the top-ranked stock — which would make the total flicker between p and € week to week depending on which company happened to rank first. The pool's currency has nothing to do with the leaderboard.

Discount rates: fixed, and honestly so

Every DCF needs a risk-free rate. The Australian, US and Indian screens fetch theirs live from a government bond yield. For the UK and the eurozone, no reliable live 10-year Gilt or Bund yield ticker is available through the data source the screens use — the same situation Korea and Japan already faced. So these are fixed constants:

The UK's higher risk-free rate means British companies are discounted more heavily than eurozone ones — a genuine reflection of the rate environment, but it does structurally raise the bar UK names must clear. More importantly: these are point-in-time estimates and they will go stale. A fixed 5.0% Gilt proxy is reasonable today and may be poor in a year — which is why every number here is a starting point rather than a conclusion, and why the DCF calculator lets you change all of them yourself.

The first run, in numbers

The screen went live on 3 August 2026. Its first full run shows how aggressively the filters cut:

Most large European companies are not cheap — 114 of 200 came out above fair value. That's normal; a screen returning a long list every week would have its thresholds set too loosely. The 80% cap removed another 23: a margin that high almost never means a spectacular bargain, it means a broken input, a one-off distorting cash flows, or something the market knows and the model doesn't.

Six of the twenty were valued by dividend discount model rather than DCF — BNP Paribas, Hannover Re, Munich Re, AXA, Intesa Sanpaolo and Aviva. Free cash flow isn't meaningful for a bank or insurer, so those are valued on dividends; the engine picks per company, using the same logic as the live tools.

And despite half the universe being British, exactly one UK name qualified — Aviva, twentieth, at a slender 1.3% margin of safety. The higher UK discount rate is part of that; so is the FTSE 100's strong run. Whether it persists is exactly what the paper portfolio exists to record.

The live GBP/EUR rate on that first run was 1.169, and the top of the list was Edenred (EDEN.PA, Paris) at a 76.2% margin of safety.

What this screen can't tell you

Everything above is machinery. It says nothing about whether any of these companies is a good investment.

A margin of safety is a model's output, and a model is only as good as its assumptions — several of which are fixed constants that will drift out of date. The screen has no view on management quality, competitive position, regulatory risk, or the possibility that a stock looks cheap precisely because the market has priced in a problem the numbers haven't caught yet. Cross-border investing adds another layer: withholding tax, currency exposure and brokerage access vary by country and appear nowhere in this calculation. Narrowing 200 companies down to 20 worth a closer look is useful work — but it is the beginning of the process, not the end.