The line every employer brand leaves out

Competitive salary appears in 4% of European employer value propositions. In North America, 4%. In Asia-Pacific, 6%. Those are Universum’s figures, from its Employer Branding NOW 2026 work, and the ranks that travel with them are 24th in Europe, 20th in North America, 21st in APAC.

Universum’s own headline is divergence - three regions taking distinct paths. The data supports it. Europe’s most common EVP attribute is learning and development at 47%, then a collaborative work environment at 44%, then a meaningful mission at 38%. North America leads with innovation at 53%, twenty-three points above the global rate. Asia-Pacific leads with career growth at 45%. These are not the same documents. They are barely the same genre.

Underneath the divergence sits a convergence nobody wrote up. Three regions whose employer brands share almost no vocabulary, and whose hiring conditions differ sharply - more than 60% of APAC organizations expect theirs to worsen in 2026, the highest figure anywhere - nonetheless agree on one thing: almost none of them say what they pay. Four, four, six. That is not three regions making three decisions. It is one decision, made three times, by people who never spoke.

The obvious reply is that pay is not a differentiator, so nobody lists it. That deserves to be taken seriously, because it sounds right. It is also refused by this data.

If differentiation were the filter, the omission would move with the market - a tight engineering market pushing pay onto the page, a slack one pushing it off. Instead the number sits flat across three continents. And look at what did make the page. Learning and development appears in 47% of European EVPs: half the market is saying the same sentence. Whatever employers are selecting attributes for, it is not distinctiveness. They are perfectly content to say what everyone else is saying. So the differentiation argument cannot be why pay is absent, because it is not why anything else is present.

Here is a better explanation, and it is less comfortable. Read the European list again - development, collaboration, meaning, balance. Then ask which of those a new hire could disprove. Not at eighteen months. By the second Wednesday.

They cannot. Not one of them resolves to anything checkable. A salary band does. It is the only attribute on that list a reader can hold against an offer letter and find short. Which suggests the EVP attribute set is not a list of what employers offer. It is a list of what employers can say without anyone being able to come back on it.

I would not call that dishonesty. I would call it a selection pressure nobody chose. An EVP is approved by people who can commit to a tone and cannot commit to a number, and the attributes that survive that approval are the ones with nobody on the other end of them.

The timing sharpens it. Universum is rolling out its Most Attractive Employers 2026 rankings market by market as I write - Poland released, France posted, the global edition still marked coming soon. Employers are being scored on attractiveness, this month, against documents that by the same vendor’s count rank pay around twentieth.

Two limits worth stating. The regional percentages are published with no sample size, no method and no interval, so they are unfalsifiable as they stand, and I am reading them as a shape rather than a measurement. And Universum both sells employer-brand consultancy and operates the rankings, which is worth holding in mind when reading any finding of theirs about misalignment.

The shape holds anyway. And it suggests a test that costs nothing. Take your EVP and mark every line a new hire could prove false in their first week. If nothing is marked, you have not written a value proposition. You have written a mood.