Pay Transparency Statistics 2026: The 44-Point Gap Between What Employers Post and What They Tell Staff

Pay transparency is usually reported as a single trend line going up. It is not one trend. It is four separate things moving at different speeds and occasionally in opposite directions: what employers publish to strangers, what they tell their own staff, what the law requires, and what any of it does to actual pay.

This page assembles the verified 2026 dataset across all four. Every figure is attributed to a named study with its fieldwork period stated, because in this topic the date matters more than usual — the EU transposition count changed twice between June and August 2026, and several widely-quoted US numbers are from 2024 fieldwork being presented as current. Sources are listed in Methodology and sources. The figures that circulate widely but do not say what people think they say are in section 10, not repeated as fact.

Pay transparency benchmark cheat sheet

The numbers most often needed in a slide or a board paper, with the source and the period each one actually covers. Copy the period along with the figure.

1. How many job postings actually show pay?

The broadest measure comes from economists at the Federal Reserve Bank of New York, who analysed the Lightcast job-posting database covering January 2020 to March 2025. The monthly share of US online job postings containing pay information rose from an average of 15% before January 2018 to approximately 53% since January 2024. That is a tripling in six years and the single most defensible headline number in this dataset.

Platform-specific figures run higher, and the difference is methodological rather than contradictory. Indeed's Hiring Lab, which tracks the share of postings on its own platform, recorded 57.8% of US postings with salary information in September 2024, up from 52.2% a year earlier. Higher figures still — around 68% — circulate from vendor analyses of narrower samples. All are internally consistent; none are interchangeable, because a single platform's posting mix is not the national posting mix.

One sub-figure deserves more attention than it gets. About 15% of postings with pay are "point salary" postings — the advertised lower and upper bound are identical. That share has plateaued. A single number satisfies a disclosure requirement while telling a candidate nothing about the band they would be hired into, and it is the most common way to comply with the letter of a law and none of its intent. The other way is the very wide range, which Payscale's 2026 report flags directly: publishing a band so broad it excludes nobody is technically compliant and practically empty.

2. The inside–outside gap: 81% versus 37%

This is the finding that reframes the whole topic, and it comes from a single survey of 626 US employers by Aon, so the two figures are directly comparable rather than stitched together from different samples. 81% of those employers include salary ranges on job listings. 63% do not share salary ranges with their current employees at all.

An organisation is four times more likely to tell a stranger what a job pays than to tell the person already doing it.

That is a 44-percentage-point gap between external and internal transparency in the same organisations. And it narrows further under inspection: of the 37% who do share ranges internally, 61% disclose only where a law requires it. Multiply that through and roughly 14% of employers — about one in seven — share pay ranges with their own staff as a matter of policy rather than legal compulsion.

The practical consequence is an information asymmetry pointing the wrong way. An existing employee can read the range for their own job on a careers page and cannot get it from their manager. That is the mechanism behind the finding that 70% of organisations expect significantly more compensation questions from managers and 68% more from employees once ranges become visible — WTW's 2025 survey of 1,915 organisations. Over half (53%) expect more pay negotiations. The questions arrive whether or not the organisation has decided what to say.

This is also where reward strategy stops being a compensation problem and becomes a communication one. Our guide to communicating a rewards policy covers the mechanics; the data point worth carrying into it is that the questions are predictable and the answers can be written before the first one is asked.

3. Compliance: a quarter of covered ads still show nothing

Laws that require pay in postings are not self-executing. The NY Fed analysis found that as of January 2025, approximately 24% of job ads covered by a pay transparency mandate contained no salary information. The researchers checked whether this could be explained by small-employer exemptions and concluded it largely could not: roughly 87% of Lightcast postings originate from firms with eleven or more employees, and about 80.4% of vacancies belong to establishments with at least ten — while most state thresholds sit at 15 employees or below, and several at one.

The four early-adopter jurisdictions studied — Colorado (enforced from January 2021), New York City (November 2022), California and Washington (both January 2023) — together account for about 20% of all US job postings. So a quarter-share of non-compliance in those places is not a rounding error at national scale.

4. The US legal map, and why the state count is reported as 14, 18 or 19

There is no single correct count, and the discrepancy between trackers is definitional rather than factual. Jackson Lewis's 2026 review identifies 19 states with pay transparency obligations of some kind. Other widely-used trackers say 18 plus the District of Columbia, or 14 plus DC. The gap is created by three choices:

• Posting mandates versus on-request rules. Around eleven states require the range in the job advertisement itself. Connecticut, Nevada and Rhode Island require pay information on request or before an offer — a real obligation, but not a posting one. Trackers that count only posting mandates land near 14.
• Local ordinances rolled up into state counts. Ohio has no state-level requirement but Cincinnati, Cleveland, Columbus and Toledo do. Counting Ohio adds a state that does not, in fact, have a state law.
• Enacted versus in force. Delaware's obligation is enacted but takes effect in September 2027; California's SB 642 clarification took effect on 1 January 2026 and Rhode Island's written-notice requirement on the same date. A count of "states with laws" and a count of "states enforcing today" are different numbers.

Employer-size thresholds vary just as widely, which matters more than the headline count for anyone actually assessing exposure. Colorado and Washington DC cover employers with one or more employees; New York four; Vermont five; New Jersey ten; California, Illinois and Washington fifteen; Massachusetts twenty-five; Delaware twenty-six; Minnesota thirty; Hawaii fifty. A multi-state employer of thirty people is covered in some jurisdictions and exempt in others for the same role.

5. Europe: a 44-point spread across comparable economies

Indeed's Hiring Lab measured the share of job postings carrying salary information across European markets in March 2026. The spread between the most and least transparent large economies is enormous, and the ranking is not what a "northern Europe is more open" intuition would predict.

Two findings inside that chart are worth more than the ranking itself. Italy nearly doubled from around 20% at the start of 2025 to 36% in March 2026 — and Italy is also one of the handful of member states that actually transposed the EU directive on time. That is the clearest available evidence that national legislation, not the directive's existence, is what moves the number. Meanwhile the UK fell from nearly two-thirds to just over half, on a series with no posting mandate behind it, which is a reminder that voluntary transparency is reversible when the labour market loosens.

Indeed's own read is that progress across Europe has stalled since 2024–2025, and that "the broad intent of the EU directive alone appears insufficient to drive further adoption". Germany — the EU's largest economy — sits at 12% and, as section 6 shows, had not published a draft implementing bill at all as of mid-2026.

Outside the EU the trajectory is steadier. Canada reached 57% of postings with pay information by June 2026, up from roughly 20% six years earlier, with the growth concentrated in two provinces that legislated.

6. The EU directive: 4 of 27 at the deadline

Directive (EU) 2023/970 required all 27 member states to transpose it into national law by 7 June 2026. The European Commission confirmed on 18 December 2025 that the deadline would not move despite member-state pressure for postponement. It then moved anyway — in practice, if not on paper.

Littler's assessment dated 8 June 2026, the day after the deadline, found three member states fully transposed (Slovakia, Malta, Italy), three partially (Lithuania, Poland, Czechia), eight with published draft legislation, and the remainder with nothing published — including Germany and Austria, which had not published a draft implementing bill at all. A separate assessment by Trusaic in late July 2026 counted five fully transposed (adding Greece, and classifying Lithuania as complete rather than partial). The two disagree on Lithuania; both agree the on-time rate was between three and four in twenty-seven.

An on-time transposition rate of roughly 15% is the headline statistic of the EU's flagship pay equality instrument in its first compliance year.

What this does not mean is that the obligations have gone away. Directives bind member states as to the result to be achieved; the Commission can open infringement proceedings under Articles 258 and 260 TFEU, and workers in non-transposing states may in principle seek damages from the state itself under the Francovich line of case law. For employers the planning date has not shifted: the first gender pay gap reports are due 7 June 2027 on 2026 pay data, which means the data being generated right now is the data that will be reported.

The 5% rule is the one most often stated incorrectly. A 5% unjustified gap does not automatically trigger a joint pay assessment: the employer has six months to remedy it or justify it on objective, gender-neutral criteria. The assessment is the consequence of failing to do so. Our companion guide on what the directive means for bonuses and rewards works through how variable pay and non-cash rewards fall inside the definition of "pay" for these purposes — which is the part most reward programmes have not yet addressed.

7. What the research says transparency actually does to pay

Two peer-reviewed studies dominate the evidence base, and both produce results that are uncomfortable for the standard advocacy case.

On wage levels. Cullen and Pakzad-Hurson, publishing in Econometrica in 2023, modelled bargaining under two-sided incomplete information and predicted that transparency reduces individual bargaining power: an employer that cannot pay one worker more without triggering costly renegotiation with everyone else credibly refuses to pay it. Their event study of US state laws protecting workers' right to discuss pay with colleagues found wages declined by approximately 2%, with the smallest declines where individual bargaining power was already low.

On the gender gap. Bennedsen, Simintzi, Tsoutsoura and Wolfenzon, in the Journal of Finance in 2022, used Danish employer–employee administrative data to evaluate the 2006 Act on Gender Specific Pay Statistics, which required firms with more than 35 employees to report gender-disaggregated wage statistics. The gender pay gap fell by two percentage points, or 13% relative to the pre-legislation mean — and it fell primarily because male wage growth slowed, not because women's pay rose. Firm profitability was unaffected, with the lower wage bill offset by reduced productivity.

The gap all of this is aimed at, in EU terms, is 11.1% — Eurostat's unadjusted gender pay gap for the 2024 reference year, extracted February 2026 and still provisional pending benchmarking against Structure of Earnings Survey results due in December 2026. Estonia has the widest gap at 18.8%; Luxembourg records −0.8%, meaning men's average gross hourly earnings were marginally below women's on this measure. That negative figure is not evidence that Luxembourg has solved pay equity — the unadjusted gap reflects occupational and sectoral composition, not like-for-like pay.

8. What job seekers do with the information

Patriot Software surveyed 1,000 US adults through Pollfish in January 2026, screening for people who had applied for at least one job in the previous twelve months.

The 17% is the most damaging number for employers in the whole dataset. One applicant in six who saw a posted range then received an offer beneath it — which converts a transparency measure into an active trust liability. It also explains the 84%: the suspicion that pay is concealed strategically is not free-floating cynicism, it is calibrated to experience.

The Gen Z finding points somewhere more useful than the range itself. The practice most likely to make younger candidates apply and accept is not a wider band but a clear explanation of how pay is set. Process transparency is cheaper than pay transparency, is not mandated anywhere, and is the part almost no employer has written down.

9. Employer readiness and direction of travel

Aon found 75% of the 626 US employers surveyed were not prepared for pay transparency laws. Preparedness varied by sector: retail and e-commerce led at 33% feeling prepared, followed by financial institutions at 21%, manufacturing at 20% and professional and business services at 20%. Note that this fieldwork was released in December 2024 — it is the best available baseline, not a 2026 reading.

The intent data is more current and points one way. Payscale's 2026 Compensation Best Practices Report, drawing 3,413 responses between October and December 2025, found 49% of organisations targeting pay transparency either organisation-wide or publicly, up from about a third the year before. WTW's June 2025 survey of 1,915 organisations found 82% communicating, planning or considering communicating individual pay ranges with employees and 79% with external candidates. Asked why, 72% cited growing regulatory requirements — well ahead of corporate values (44%) and employee expectations (41%).

Regulation, in other words, is doing nearly all the work. Only 20% of organisations have publicly shared a commitment on pay transparency and 32% on pay equity, and just 56% use any metric to measure whether their transparency efforts have an effect. Meanwhile 47% of companies with significant European operations report concern about the EU directive specifically.

10. Pay transparency statistics to handle with care

11. What to actually do with this data

Methodology and sources

Every figure on this page is attributed to a named study, with sample size and fieldwork period stated where published. Where sources disagree — the US state count, the EU transposition count, the share of postings with pay — both figures are shown with the definition each uses. Derived metrics are labelled as derived and the arithmetic is shown. Figures we could not trace to a primary source are listed in section 10 rather than repeated as fact.

Key takeaways

• External disclosure has tripled since 2018 and is now the majority case in the US, the UK and Canada; internal disclosure has not moved with it, and the 44-point gap between the two is the defining feature of pay transparency in 2026.
• Compliance is the soft spot in the US regime — roughly a quarter of legally covered ads still show nothing, in jurisdictions accounting for a fifth of all postings.
• The EU directive missed its own deadline in 23 of 27 member states, but the 7 June 2027 reporting date on 2026 data did not move. The data being generated now is the data that will be reported.
• The strongest research finds transparency compresses pay differences rather than raising pay, and that Denmark's gender gap narrowed mainly through slower male wage growth. Business cases built on "transparency raises wages" are not supported by the literature.
• Regulation, not values, is the stated driver for 72% of organisations — which means the practices nobody mandates, above all a written explanation of how pay is set, are where the remaining differentiation sits.