The new prime minister has promised a cut in business rates for the country’s public houses as well as other social venues. Whether this helps pubs’ prospects of survival depends on how much of the pressures on these local institutions are really about costs, rather than changing consumer tastes.
Despite a decade of rising costs and headlines about pub closures, data show only a modest change, and potentially an increase, in pub numbers since 2016. But while cities such as Birmingham and Liverpool have seen the number of pubs grow, rural areas have experienced stagnating and declining numbers.
Pubs matter for more than pints: as a ‘third place’, they are linked in economic research with social cohesion, community wellbeing and even economic mobility. Whether recent cuts to business rates help struggling pubs to survive depends on how much of the pressure is really about costs, rather than changing consumer tastes.
What’s been happening to British pubs?
It has been a rough decade for the UK hospitality industry. Six years ago in the early stages of the Covid-19 pandemic, lockdown orders were issued, effectively closing all pubs. This was followed by active recovery policy, but also inflation, increases to employer national insurance contributions, and uplifts to minimum wages.
This has led to headlines reporting that many pubs have succumbed to the pressure, with a pub closing every day in England and Wales (The Guardian, 2025). Are we seeing the decline of pubs in the UK?
Why should we care about pubs?
Pubs are one of several types of local amenities known as third places (after home and work), which can make a neighbourhood and a city, town or village, better. A good pub becomes part of the community fabric (Jacobs, 1961) and can encourage social mixing (Heine et al, 2025). A study of rural parishes in Northern England finds that pubs are significantly correlated with higher levels of social cohesion (Mount and Cabras, 2016).
This kind of local social infrastructure has attracted growing interest from economists more broadly. One piece of research shows that pubs in rural Ireland support both community cohesion and local economic activity (Cabras and Mount, 2017), while another finds that the presence of a pub is capitalised into higher local house prices across Great Britain (Sohns et al, 2020) – a ‘revealed preference’ indication of how much residents value them.
More broadly, work using large-scale social network data links the kind of everyday social interaction fostered by places like pubs to economic outcomes: American research shows that social capital, measured through friendship networks, is strongly associated with rates of upward income mobility in the United States (Chetty et al, 2022).
How many pubs are there in the UK?
The exact number of pubs in the UK will depend on how a pub is defined and therefore varies by data source. Here we focus on two data sources. The first consists of data extracts from the Inter Departmental Business Register, provided by the Office for National Statistics (ONS) and made available through the Nomis website – referred to here as the ONS data.
These data have the advantage of being easy to access in a user-friendly format. The data can also be broken down into smaller geographies. But the smallest geographical levels available are still relatively large (middle super output areas). Further, the reported pub counts are rounded to the nearest multiple of five, which can lead to considerable miscounting for analysis at smaller levels of geography. These data are publicly available each year from 2010.
The second data source is Ordnance Survey Points of Interest data, referred to here as the POI data. The main advantage of the POI data is the extraordinary level of geography, as longitude-latitude coordinates are available for local units. As these data track individual businesses, they can be used to study place-specific turnover. Further, data are updated quarterly, allowing for relatively frequent and up-to-date analysis.
The primary disadvantage of these data is that they are not in a ready to use format for most purposes: considerable work is required to prepare them for analysis. These data are publicly available from 2015. Clark et al, 2023 provide an excellent overview of the validity of the POI data, comparing it to the ONS data, with respect to identifying pubs.
The ONS data report 37,805 pubs in the UK in 2025, with 34,500 in England and Wales, 2,565 in Scotland, and 755 in Northern Ireland. For comparison, the British Beer and Pub Association reports 44,650 pubs in 2025 (BBPA), but do not provide a geographical breakdown.
POI data report 37,351 pubs in England and Wales as of June 2025, slightly more than the figure reported by the ONS data, but considerably less than reported by BBPA even after making a reasonable adjustment for Northern Ireland and Scotland.
Figure 1: Pub density across England and Wales, and Greater London
Source: POI business counts calculated using data from Ordnance Survey Points of Interest data as of September 2024.
Note: The maps show the number of pubs per 10,000 residents by lower super output area (each of which has a residential population of between 1,000 and 3,000).
Figure 1a plots the number of pubs per 10,000 residents across neighbourhoods in England and Wales. Counter-intuitively, the density of pubs is lower in more populated areas, such as London, and higher in less populated areas, as in the North.
Figure 1b provides a detailed look across the Greater London area, which has some of the highest pub densities in the country, but where they are concentrated in relatively few areas, such as central London. The remainder of the city is sprinkled with relatively modest pub density.
Are pub numbers declining?
The different data sources show slightly different pictures of what has been happening to the country’s pubs. Clark et al, 2023 provide an excellent overview of the spatial distribution of pubs in England using POI and ONS data.
Figure 2: The number of pubs over time in England and Wales by data source
Source: ONS UK Business Counts: local units by industry and employment size band, provided by the ONS through Nomis (accessed 13 July 2026.) POI business counts calculated using data from Ordnance Survey Points of Interest data.
Note: Values indicate the number of local units for England and Wales.
From 2010 to 2025, the ONS data show a 15% decline in the total number of pubs across England and Wales – approximately 6,000 establishments (see Figure 2).
But the change in pub numbers over time depends on which data source we consider. At the beginning of 2016, pub counts were almost identical between the two data sources. But by 2025, they had diverged, with the ONS data showing a decrease in pubs of about 4%, but the POI data showing an increase of about 3%.
These differences are likely to reflect differences in how establishments are characterised: some establishments listed as licenced clubs in the ONS data may be categorised as pubs in the POI data (Clark et al, 2023).
Either data source is consistent with, at most, modest changes in the number of pubs over the last decade. This is somewhat surprising considering the significant social and economic changes over this time period, including a pandemic that led to a notable change in how much time workers spend at home versus the office (Matheson, 2020; Meyrick, 2023; De Fraja et al, 2025).
While the period from 2010 to 2016 does seem to show a considerable decline in the number of pubs across England and Wales, the decline appears to have stabilised, and perhaps even reversed, in the subsequent decade.
Figure 3: The change in the number of pubs in England and Wales by local authority type
Source: POI business counts calculated using data from Ordnance Survey Points of Interest data.
Note: This figure shows, by local authority type for March, June, September and December of each year, the number of pubs divided by the total pubs in March 2016. Local authority types (not including Greater London) are defined by residential population density: Rural is less than 300/km squared; Small is 300 to 1,000/km squared; Medium is 1,000 to 4,000/km squared; Large is greater than 4,000/km squared.
Even considering the relative optimistic view of the industry from the POI data, the growth in pubs was not evenly realised across England and Wales (see Figure 3):
- The increase was concentrated in medium and large cities: Birmingham, Blackpool and Liverpool all realised pub growth more than 20%.
- In contrast, rural local authorities experienced a decline in pubs overall: Copeland, North Warwickshire and South Norfolk all experienced pub losses of 20%.
What are the factors affecting pub numbers?
The supply side
The last decade has seen the cost of running a pub increase from several different sources. They include higher minimum wages, the increase in employer national insurance contributions and removal of pandemic-era business rates relief (Commons).
While there is no shortage of anecdotal evidence suggesting that these costs make running pubs unsustainable (for example, ‘We can’t increase prices any more’: UK hospitality firms hit by cost triple blow, The Guardian, 2026) evidence of a causal link is more difficult to find.
But it is notable that the cost of a pint of draught lager has increased by 67% between January 2010 and January 2025 (RPI: ave price – draught lager, per pint – ONS), or an average of 3.6% per year. This is significantly higher than the overall inflation rate for the period, during which prices increased by 54% overall or 2.9% annually (see Figure 4).
Figure 4: Price index, pints versus the CPI
Source: ONS, CPI Index 00 and RPI: ave price – draught lager, per pint. Price indices adjusted to a base value at January 2010.
The demand side
We cannot rule out the possibility that the lack of growth we have seen in the pub industry over time reflects changing consumer tastes. There are a couple of potential drivers of this.
The most recent large social change has been the shift to working from home, which we have observed since the Covid-19 public health measures in 2020 (De Fraja, Matheson and Rockey, 2021; De Fraja, Matheson, Mizen, and Rockey, 2021; De Fraja et al, 2025; Goken et al, 2026). This could affect pubs in two ways:
- First, it changes where people consume: an after-work pint may be purchased near the home rather than near the office.
- Second, it changes the consumption options: if a pub is not conveniently near my home, or if I just want to save money, I can consume alcohol at home, or perhaps not consume at all.
It is also possible that we are observing a change in preferences for alcohol. This could be driven by health trends, shifting population demographics, or a broader shift towards non-alcoholic establishments. A recent YouGov poll shows a modest increase in the share of British adults who say they do not consume alcohol. This is mirrored by modest declines in alcohol consumption reported in the Health Survey for England.
Consistent with changing preferences for alcohol, we can use the POI and ONS data to look at what has happened to non-licenced cafes over time. Cafes provide an alternative to the pub as a third place, where people can meet to work and socialise.
The rise in cafes since 2010 has been substantial, regardless of which data we use (see Figure 5). Between 2010 and 2025, the number of cafes across England and Wales increased by 135%, or more than 18,000 establishments.
Figure 5: Pubs versus cafes over time in England and Wales
Source: ONS UK Business Counts: local units by industry and employment size band, provided by the ONS through Nomis (accessed 13 July 2026.) POI business counts calculated using data from Ordnance Survey Points of Interest data.
Note: Values indicate the number of local units for England and Wales.
Will recent government policies help?
On 23 July, shortly after taking over as the new prime minister, Andy Burnham announced a 20% cut to business rates for pubs, social clubs and live music venues (BBC, 2026). This follows a 15% cut given to pubs in April.
But both of these relief policies follow the April 2026 removal of pandemic-era relief and re-evaluation of business rates (Commons). It is therefore unclear if the net change will be a decrease or an increase in costs for pubs relative to what was paid in 2025.
In addition, by how much a reduction in costs for pubs will affect their survival depends on the importance of supply-side versus demand-side factors. Although we lack hard evidence, it is altogether possible that the lack of growth in pubs over the last decade reflects a change in preferences. If this is the case, then supply-side policies are likely to have little impact on the trend.
That said, if the goal is to create a profitable economic environment for pubs to thrive as a third place, then cuts to operating costs probably will not hurt.
Where can I find out more?
- Remote working and the new geography of local service spending: this 2025 study in Economica calculates the potential loss in local services spending due to the post-pandemic shift to remote working.
- An exploration of recent trends in the number of British pubs and how these vary by neighbourhood type: this 2023 study in the Geographical Journal provides an excellent overview, with geographical analysis, of different data sources available to study pubs in the UK.
- How has coronavirus affected pubs, cafes and restaurants? and How is the UK’s night-time economy coping? These two Economics Observatory articles provide an overview of how pubs and other local services were affected by the Covid-19 pandemic from 2020.
