Contents
- The 30-second summary
- What you will find here
- Where the 62% comes from
- What happened over the following decade
- The inverse experiment: introducing a fee
- The effect nobody expects: free entry sells tickets
- What the mobility data shows
- The fairness objection almost nobody raises
- How to handle this, practically speaking
- Frequently asked questions
- What we do not know
- References
In 2001 the United Kingdom abolished admission charges at its national museums, and in 2002 the department announced 62% more visits in seven months. It is one of the most successful cultural policy figures of recent decades and it gets cited whenever free admission is discussed in any country.
What almost never accompanies that figure is the question a market researcher asked himself the moment he read it, and which turned out to be the decisive one: were a lot more people visiting museums, or were the same people visiting more often? The survey he ran to find out showed that, although the number of people had grown significantly, the increase was greatest among those groups who had always gone to museums, and much lower among those who might be described as socially excluded.
And the symmetrical natural experiment, run in Sweden, turns the coin over: even a very low entrance fee produces a significant reduction in several of the target groups the museum has a policy mandate to reach. Charging a little pushes out the people you wanted to attract, and charging nothing does not bring them in.
The 30-second summary
It is important to differentiate visits from visitors across every single row, since this is precisely where the gap between the main headline and the social impact is determined.
| Finding | Result | Study and year |
|---|---|---|
| Increase announced after charges were scrapped | 62% in seven months, per the UK department | Martin, Cultural Trends (2002) |
| Where that increase concentrated | Among the groups who already went regularly | Martin, Cultural Trends (2002) |
| Composition of the audience after ten years | More middle class and more middle aged than the population | Davies, Cultural Trends (2005) |
| Effect of a very low fee | Significant reduction in several target groups | Lampi and Orth, Kyklos (2009) |
| Relationship between free and paid visits in Italy | Free visits raise subsequent paid ones | Cellini and Cuccia, OEP (2018) |
| What explained China’s museum boom | Mostly expenditure and space; free entry also significant | Zhang and Courty, JCE (2021) |
| Size of Milan’s mobility dataset | 330,929 unique users in 2019 | Lerouge, Biferale and Bruno, JCE (2026) |
| Profitability of adding a free day | Would not be profitable, per the estimate | Steiner, JCE (1997) |
| Fairness objection to universal free entry | Consumes resources without targeted benefit to the disadvantaged | Hume, JAP (2025) |
| What drives repeat visits to a free museum | Content renewal, more than visitor demographics | Corona, MMC (2025) |
What you will find here
- Where the 62% comes from and what the survey that tested it actually asked.
- What happened to the social composition of the British audience over the following decade.
- The counter-experiment evaluating the impact of implementing a fee instead of abolishing one.
• Why in Italy free visits increase paid ones.
- What the mobility data of 330,929 people in Milan shows.
• The fairness objection nobody raises: why free entry can be regressive.
Where the 62% comes from
The context matters. The then Secretary of State for Culture, Media and Sport’s commitment to ensuring free entry for all visitors to national museums and galleries by the end of 2001 left many institutions with considerable uncertainty about what was coming.
The questions they were asking, according to Andy Martin — who was then working for the research firm MORI and published the analysis in Cultural Trends in 2002 — were three: what impact «going free» would have, whether those who might be described as socially excluded would be encouraged through the doors, and whether the money visitors saved on entrance fees would be spent in the shops and restaurants.
The initial question met with an extraordinary response: at the dawn of summer 2002, the agency reported a 62% surge in «visitor numbers» across the seven-month period following the abolition of admission fees.
And here comes the methodological observation that turns a press release into a research problem. Martin notes that the department tended to use the terms «visitors», «people» and «visitor numbers» to refer to visits per se. Hence his two questions:
• Did these figures mean there were actually a lot more people visiting museums and galleries, or were the same people visiting more frequently?
• Was the boost in visiting restricted to the national museums and galleries, or were more people visiting museums and galleries generally?
To answer them, MORI placed four questions about the British public’s museum-going habits on its omnibus study in August 2002. The results: although the number of people visiting museums has increased significantly since 2001, the increase is greatest among those groups who have traditionally always gone to museums and galleries, while the increase among groups who might be described as socially excluded is much lower.
The policy worked: there were more visits and more visitors. But it did not do what it was said it would do.
| Study | What it measured | What it found | What it cannot establish |
|---|---|---|---|
| Martin, Cultural Trends (2002) | An omnibus survey conducted after fees were eliminated | Growth was concentrated among the habitual audience | A single survey wave lacking extended follow-up |
| Davies, Cultural Trends (2005) | Nationwide data spanning 1994 to 2004 | The demographic became increasingly middle class and older | A timeframe only partially influenced by the initiative |
| Cowell, Cultural Trends (2007) | The ministry’s 23 indicators | Requires separating the impact of capital investment | An evaluation of existing metrics rather than original measurement |
| Lampi and Orth, Kyklos (2009) | Surveys administered prior to and following the introduction of a charge | Even a modest fee pushes out target groups | Restricted to a single Swedish museum instead of nationwide policy |
| Cellini and Cuccia, OEP (2018) | Monthly Italian statistics from 1996-2015 | Complimentary access stimulates later paid admissions | Macro-level figures lacking institution-specific breakdown |
| Lerouge and others, JCE (2026) | 330,929 users tracked via location metrics | Scale, geographical origin, and cultural voraciousness | Confined to one city over twelve months featuring sporadic free admission |
What happened over the following decade
Stuart Davies released a decade-long tracking study in Cultural Trends back in 2005, building upon his initial 1994 analysis of the galleries and museums sector through nationwide statistics. Traditional inquiries drive his research: the annual volume of visitors, the demographic profile of those who attend, and the underlying motivations for their visits.
His conclusion about the period, which includes the early phases of the free admission policy, is harsh: at best, the total number of visits grew only a little, and those who visit became more middle class and more middle aged, rather more than the population did itself in these years.
And the formulation of the underlying problem: museums and galleries might maintain their audiences, but they were not expanding them or broadening their social appeal.
Davies is even-handed in assessing the policy: that alone, he says, was sufficient justification for the government’s measures, whose early stages seem to have been successful. Which is to say, free admission was answering a real problem; what it did not do was solve it.
Ben Cowell added in 2007, in the same journal, the missing piece: an examination of how the government measured the impact of its own policy, based on the 23 performance indicators the department used in its funding agreements with each museum.
His verdict is a call for caution that still stands: further work is needed to isolate the precise impact of free admission, particularly on attracting new audiences. And he flags the confounder that almost never gets discounted: the effect of new capital investment. In those same years, British museums were refurbished with considerable funds, and separating the effect of free entry from the effect of having a new museum is not trivial.
The inverse experiment: introducing a fee
The most informative design in this field did not remove admission: it added it. Elina Lampi and Matilda Orth, of the University of Gothenburg, published in Kyklos in 2009 a study of a state-funded Swedish museum that was about to start charging.
Their sequence is what gives the work its value. While entrance to the museum was still free, they conducted a survey to collect information about visitor characteristics and used the contingent valuation method to measure visitors’ willingness to pay for a visit. After the fee was introduced, they conducted a second survey.
That yields a pair of distinct outcomes. The initial finding, concerning the impact: even a very low entrance fee level results in a significant reduction in several target groups that the museum has policy directives to reach.
The second, methodological and infrequent: the comparison between the results predicted by contingent valuation and the observed change in visitor composition implies that the method does predict a majority of the changes successfully. It is a validation of the method in a cultural good, which is unusual and which lends credibility to the first result.
Set beside the British finding, the picture is this: a small fee pushes the target groups away, and removing it does not bring them back. The price barrier is real in one direction and not symmetrical in the other, which suggests what keeps those groups out is not only — nor principally — money.
Marine Legall-Ely, Caroline Urbain and three co-authors reached a compatible conclusion in France with a multi-angulation design published in 2007: free admission alters the public’s perceptions of museums and monuments, making it a secondary consideration in planning and implementing a visit. Secondary. Their closing recommendation is that the public’s involvement in the visit needs to be increased and the reasons for free admission explained.
| Price movement | What happens to the habitual audience | What happens to the target groups |
|---|---|---|
| Removing admission | Visits more, and more often | A much smaller increase |
| Introducing a very low fee | Keeps visiting | Significant reduction |
| One free day a month | More paid visits afterwards, in Italy | Changes volume, provenance and voraciousness |
| Episodic free day | Different profile and group behaviour | Documented, without quantifying the broadening |
| Permanent free entry, at ten years | Audience maintained | Social appeal not broadened |
| Reading them together | Price is not their barrier | Price does hold them back, but is not the only thing |
That asymmetry is the most robust finding in the set and the one least used when designing policy. A small fee is enough to push the target groups away, and removing it is not enough to bring them in. Which implies free admission is necessary but not sufficient, and that the rest of the work — programming, community outreach, opening hours, transport, mediation — is where the intended effect actually sits.
The effect nobody expects: free entry sells tickets
There is a well-documented counterintuitive result in the Italian case. Roberto Cellini and Tiziana Cuccia published in Oxford Economic Papers in 2018 an analysis of Italian state museums and monuments using monthly data aggregated at the national level from January 1996 to December 2015.
Their finding: within a multivariate analysis that takes into account the seasonal structure of the data, they document a positive influence of the number of free visits to museums and monuments on subsequent charged visits.
And they analysed a specific policy change: the extension of free admittance approved in July 2014. The new rule entailed an increase in both free and charged visits, as well as a stronger link between the patterns of the two.
The plausible mechanism is the experience-good one: the free visit works as a product trial and generates subsequent demand. That makes partial free entry something other than a forgone revenue: it can be an investment in future demand.
It is worth contrasting that, however, with Faye Steiner’s classic estimate in Journal of Cultural Economics in 1997, which examined the impact of free admission on museum revenue and evaluated, from an income-maximising perspective, the desirability of an additional free day. She used regression analysis to estimate marginal shop and restaurant revenue for both art patrons and marginal consumers, and her result was clear: empirical estimation shows that an additional free day would not be profitable.
The two results do not necessarily contradict each other: Steiner measures museum revenue and Cellini and Cuccia measure paid visits. A museum can sell more tickets and still lose money if the cost of serving the free visitors exceeds the margin from the new payers.
What the mobility data shows
The most recent study uses the kind of data this field needed. Romain Lerouge, Lorenzo Biferale and Michele Bruno published in Journal of Cultural Economics in 2026 an analysis of the «Domenica al Museo» initiative — the Italian ministry’s policy giving free entry to state museums on the first Sunday of each month — exploiting high-frequency location-based data from 330,929 unique users in Milan in 2019.
The authors situate their contribution in a specific gap in the debate: while some advocate free admission to increase public engagement, others argue that such measures may not significantly broaden participation due to the relatively inelastic demand for museum visits. A newer approach, offering free access on select days, has emerged but remains underexplored in the academic literature.
Their findings encompass three facets once financial obstacles are eliminated: visitor volumes, their spatial provenance, and what the authors call «cultural voraciousness». That final aspect proves the most compelling and represents the exact metric aggregate data could never capture: the sheer quantity of museums a single individual tours during one day when admission is completely free.
The theoretical framing of the article also counts, as it clarifies the motivation behind this research: cultural policy has increasingly been shaped by goals of promoting social cohesion, with a central focus on broadening access to the arts and cultural heritage, particularly in urban settings where spatial inequalities, such as the core-periphery divide, can limit access to cultural amenities.
Along the same line of occasional free days, Cheryl Bowman, Adrianna Adkins and six co-authors analysed in Museum Management and Curatorship in 2020 the differences in visitor characteristics and experiences on episodic free days against standard days, with data from 370 people at a mid-sized natural history museum. Their contribution is practical: they document significant differences in visitor characteristics, group composition and behaviour, and interest in the exhibit, which allows the decision to add or remove a free day to be made on data rather than intuition.
The fairness objection almost nobody raises
There is an argument against universal free entry that comes not from accounting but from distributive justice, and it was made by Jack Hume, of University College London, in Journal of Applied Philosophy in 2025.
His thesis: free admission can be a surprisingly bad way of promoting cultural opportunities for disadvantaged groups. And the reasoning, focused on geographic inequalities and per-capita museum spending: free admissions consume resources without necessarily providing targeted benefits to disadvantaged groups and addressing background inequalities.
The proposal that follows is uncomfortable and carefully bounded: if we are aiming to prioritise the interests of less advantaged groups, we should be in favour of charging at London’s major museums, to finance «levelling up» across regions, and more direct access-promoting measures in targeted outreach, collaboration and programming.
Hume is explicit about the scope of his critique, which makes it stronger: it does not apply to all museums. It applies to the largest and most popular in the most advantaged areas, which can expect to keep drawing significant numbers while charging.
That reading fits the data. If the 62% increase concentrated among the audience that already went, then forgoing that audience’s revenue is subsidising people who could pay, while the target groups remain outside for reasons a zero fee does not touch.
The most recent evidence on what does move participation in a free museum points the same way. Laura Corona analysed in Museum Management and Curatorship in 2025 57 Finnish museums plus 70 visitor interviews at the Helsinki City Museum, and her central finding on repeat visiting is: content renewal drives repeat visitation in a free museum more than visitor demographics, and social motivations are central to engagement.
Her general conclusion orders the debate: admission policies are not mere financial tools but strategic instruments affecting the wider social role of museums. And she documents the practical divergence: free museums focus more on educational and participatory experiences, whereas ticketed museums pursue international visibility through diversified funding sources.
Maike Althaus, Stefanie Mueller and Dennis Kundisch supplied in International Journal of Cultural Policy in 2023 the map of intermediate options the binary debate ignores, with a taxonomy built from the pricing policy characteristics of 149 museums and a cluster analysis of 117 in the same region. Their three groups: free admission pricing policy, simple admission pricing policy and comprehensive admission pricing policy. Public discussion tends to present only the two extremes.
| Pricing model | What characterises it | Which objective it serves |
|---|---|---|
| Free admission | No fee, with an educational and participatory focus | Access and social use of the facility |
| Simple pricing | One fee and few exceptions | Sustainability with light administration |
| Comprehensive pricing | Bands, passes and segmented exemptions | Targeted reach without forgoing revenue |
| Periodic free day | Free entry bounded in the calendar | Product trial and subsequent demand |
| Free entry with charging at the majors | Charging where demand is inelastic | Financing levelling up across regions |
| What public debate offers | Only the first and the second | Which is why the discussion stalls |
| Data point | Figure |
|---|---|
| Increase announced in seven months after free entry | 62% |
| Where that increase was greatest | Among groups who already visited |
| Performance indicators used by the department | 23 |
| Period of the Italian monthly data | January 1996 to December 2015 |
| Unique users in Milan’s mobility data | 330,929 |
| Growth in Chinese museum visits, 1995-2016 | Eightfold |
| Finnish museums compared | 57, plus 70 interviews |
| Museums classified in the pricing taxonomy | 149, with 117 clustered into three groups |
One final piece for calibrating the real weight of free entry against other factors. Feng-Hua Zhang and Pascal Courty analysed in Journal of Cultural Economics in 2021 the Chinese museum boom, where visits grew eightfold between 1995 and 2016. Their decomposition: growth in museum expenditure and space contributed to most of the increase in visits, although the free admission policy rolled out in 2008 also had a significant impact.
Demand factors did not have a major impact, with the possible exception of the increase in urban population. Which is to say: what moves visits most is building and resourcing museums, and free entry helps. That is an order of priorities the debate about fees tends to invert.
How to handle this, practically speaking
- Count visitors, not visits. The distinction Martin raised in 2002 remains what separates a headline from a social effect, while press releases continue to conflate them.
• Do not expect free entry to broaden the audience on its own. The increase concentrated among those who already went, and after ten years the British audience was more middle class and older than the population.
- Avoid the token fee if there is a social reach mandate. The Swedish study found even a very low level significantly reduces several of the target groups.
- Consider the free day instead of all or nothing. Across Italy, complimentary access drives subsequent paid admissions, while the classification of 149 museums outlines three distinct models rather than two.
• Discount capital investment before claiming credit. It is the confounder Cowell flags and the Chinese evidence confirms: expenditure and space explain most of the growth in visits.
• Renew content if you want repeat visits. In the Finnish study, that drives repeat visiting in a free museum more than the audience’s demographics.
Frequently asked questions
Do admission removals boost attendance?
Yes. The British department announced 62% more in seven months, and the independent study confirmed that museum attendance had risen substantially, rather than just the frequency of regular visitors.
Do people who did not go before start going?
Scant. The rise proved most substantial within demographics that traditionally exhibited high attendance rates, whereas it remained considerably more modest among cohorts often labeled as socially excluded. A decade later, the spectator base had evolved to be older and possess a more middle-class profile compared to the general public.
What if a small amount is charged?
It is worse than it sounds. The Swedish study found that even a very low fee level produces a significant reduction in several of the target groups the museum was mandated to reach.
Does a museum lose money opening free for a day?
The two available measurements point in different directions because they measure different things. The classic estimate concludes an additional free day would not be profitable in revenue; the twenty-year Italian analysis finds free visits raise subsequent paid ones.
Is universal free entry fair?
There is a serious, carefully bounded objection: free admissions consume resources without necessarily providing targeted benefits to disadvantaged groups, and its author proposes charging at the major museums in advantaged areas to finance regional levelling up and community outreach.
What drives visitors back to a complimentary museum?
According to the study of 57 Finnish museums and 70 interviews, content renewal drives repeat visitation more than visitor demographics, and social motivations are central to engagement.
What drives more attendance, free admission or the museum itself?
The museum. In the Chinese case, where visits grew eightfold in twenty-one years, growth in expenditure and space explained most of the increase, although the 2008 free admission policy also had a significant impact.
What we do not know
• How much of the British 62% was free entry and how much was capital investment. Cowell flags it explicitly as the undiscounted confounder, and it remains unisolated.
- Why the price barrier is not symmetrical. A low fee pushes target groups out and removing it does not bring them in; nobody has identified which other barriers operate.
• Whether the Italian effect on paid visits holds museum by museum. The series is a national aggregate and does not distinguish between institutions.
• Whether universal free entry is regressive in quantitative terms. The fairness objection is a well-built philosophical argument, not an estimate of distributive incidence.
• Determining which of the three pricing models suits a specific goal best. The framework outlines the trio of categories without assessing their relative outcomes.
References
1. Martin, A. K. «The impact of free entry to museums». Cultural Trends, 2002. 10.1080/09548960209390329
2. Davies, S. «Still popular: Museums and their visitors 1994-2004». Cultural Trends, 2005. 10.1080/09548960500164509
3. Cowell, B. «Measuring the Impact of Free Admission». Cultural Trends, 2007. 10.1080/09548960701479375
4. Lampi, E. and Orth, M. «Who Visits the Museums? A Comparison Between Stated Preferences and Observed Effects of Entrance Fees». Kyklos, 2009. 10.1111/j.1467-6435.2009.00425.x
5. Cellini, R. and Cuccia, T. «How free admittance affects charged visits to museums: An analysis of the Italian case». Oxford Economic Papers, 2018. 10.1093/oep/gpy011
6. Zhang, F.-H. and Courty, P. «The China museum visit boom: Government or demand driven?». Journal of Cultural Economics, 2021. 10.1007/s10824-021-09410-x
7. Lerouge, R., Biferale, L. and Bruno, M. «The effect of consumption incentives on museum attendance: a case study on Milan through human mobility data». Journal of Cultural Economics, 2026. 10.1007/s10824-026-09581-5
8. Steiner, F. «Optimal Pricing of Museum Admission». Journal of Cultural Economics, 1997. 10.1023/a:1007432901540
9. Hume, J. «Openness, Priority, and Free Museums». Journal of Applied Philosophy, 2025. 10.1111/japp.70001
10. Bowman, C., Adkins, A. and others. «Differences in visitor characteristics and experiences on episodic free admission days». Museum Management and Curatorship, 2020. 10.1080/09647775.2019.1578992
11. Corona, L. «What free really costs: museum practice and policy in Finland». Museum Management and Curatorship, 2025. 10.1080/09647775.2025.2583522
12. Althaus, M., Mueller, S. and Kundisch, D. «What price culture? – a taxonomy of the admission pricing policy at museums». International Journal of Cultural Policy, 2023. 10.1080/10286632.2023.2217198
13. Legall-Ely, M., Urbain, C. and others. «An Exploratory Study of the Implications of Free Admission to Museums and Monuments: The Public’s Perceptions and Effects on Their Visiting Behavior». Recherche et Applications en Marketing, 2007. 10.1177/205157070702200202
This piece reviews peer-reviewed research regarding museum entry regulations and their impact on visitor numbers and demographic makeup. The data reflect the specific nations and timeframes under analysis and cannot be generalized to any individual venue. It serves neither as cultural policy guidance nor as an assessment of a particular gallery.
