Overseas residents made 40.9 million visits to the UK in 2019, an increase of 0.6 million compared with 2018.
There were 93.1 million visits overseas by UK residents in 2019, an increase of 3% compared with 2018.
UK residents spent £62.3 billion on visits overseas in 2019, an increase of 7% compared with 2018.
Overseas residents spent £28.4 billion on visits to the UK in 2019, an increase of 7% compared with 2018.
The results presented in this article were produced using an improved methodology for 2019 and a revised series of estimates for 2009 to 2018.
This article analyses data from the International Passenger Survey (IPS), which collects data from overseas residents leaving the UK and UK residents returning home.
An improved methodology was used to produce the estimates presented in this article. This includes both the final results for 2019 and revised results for 2009 to 2018. The principles of the new methodology are explained in Section 5: Improved methodology for the estimates, while the impacts of the new method on the figures are shown in Section 6: Impacts of the new IPS methodology. The datasets accompanying this article present the main estimates produced using the new method for 2009 to 2019.
The revised data for 2009 to 2018 presented in this article and the accompanying datasets supersede figures published previously for this period.
International Passenger Survey and COVID-19
International Passenger Survey (IPS) interviewing was suspended on 16 March 2020 because of the coronavirus (COVID-19). It is not certain when it will resume.
Travel and tourism estimates for Quarter 1 (Jan to Mar) 2020 will be published in July 2020, making the best possible use of the available data. We expect that publishable estimates for March 2020 can be produced using the data collected up to 16 March 2020.
No IPS data will be collected for the period when the survey is not operational, and the usual travel and tourism outputs from the IPS will not be published for this period. However, the IPS team will publish information to help users to understand trends in total international travel, based on the available administrative data from the Civil Aviation Authority (CAA) and the Department for Transport (DfT). This will provide figures on numbers of international journeys arriving into and departing from the UK, but there will be no information about the characteristics of these passengers.
Initial figures will be produced in July 2020, relating to passenger numbers in April and May 2020. These will be published alongside the provisional travel and tourism estimates for Quarter 1 2020. Under the usual IPS publication schedule, estimates for Quarter 2 (Apr to June) 2020 would be published in October 2020. At this time, we will seek to provide more information on international travel patterns for the quarter, drawing on available data sources.
The IPS is an important input to Office for National Statistics (ONS) figures on international trade. More information about their plans to produce figures in the absence of the IPS can be found in our statement on trade, published on 21 May 2020. More information about the plans of international migration statistics can be found in the latest Migration Statistics Quarterly Report (MSQR).Back to table of contents
There were 93.1 million visits overseas by UK residents in 2019 (Figure 1), an increase of 3% compared with 2018. Overall, visits overseas have increased steadily over the last 20 years, except for a notable fall observed in 2009.
UK residents spent a record amount of £62.3 billion on visits abroad in 2019 (Figure 2), which was 7% more than in 2018. The average spend per visit increased from £642 in 2018 to £670 in 2019, contributing to the overall increase in spending.
Holiday was the most popular reason for UK residents to visit another country
There were 93.1 million visits overseas by UK residents in 2019, the highest figure recorded by the International Passenger Survey (IPS). The number of visits in 2019 was 3% more than in 2018, when there were 90.6 million visits.
Nearly two-thirds of UK residents’ visits abroad were for holidays; this was constant over the five-year period from 2015 to 2019. There were 58.7 million holiday visits abroad in 2019, a 2% increase on the previous year (Figure 3); most of these visitors travelled to countries within the EU (44.6 million).
Visiting friends or relatives overseas was also popular with UK residents. The number of these visits increased from 6.6 million in 1999 to 23.5 million visits in 2019. The number of business trips has varied relatively little over the last 20 years: the number of business trips fell slightly in 2019 to 9.0 million visits from a 10-year high of 9.3 million in 2018.
Spending abroad increased for all purposes except business in 2019. UK residents travelling abroad for holidays spent £43.4 billion in 2019; this is 70% of the total spent by UK residents abroad. Business spend reflected the fall in business visits, and spending on these trips fell by 9%, from £6.9 billion in 2018 to £6.3 billion in 2019.
Spain was still the most visited country by UK residents
There were 18.1 million visits to Spain by UK residents in 2019, which was a 1% increase on 2018. Visits to Spain were mainly for holidays (87% of the total).
Spain, France, Italy, the US and the Republic of Ireland, in that order, remain the top five most popular countries for UK residents to visit (Figure 4), accounting for 46% of all visits abroad and approximately 43% of total spend abroad. Overall, 72% of visits were to EU countries.
Among the top 10 most visited countries, Greece saw the largest percentage change between 2018 and 2019, increasing by 18% from 2.9 million to 3.4 million. Of these visits, 90% were for holidays.
Back to table of contents
Overseas residents made a total of 40.9 million visits to the UK during 2019 (Figure 5), an increase of 1% compared with 2018 but lower than the peak reached in 2017 (41.1 million visits).
A total of £28.4 billion was spent on visits to the UK by overseas residents in 2019, an increase of 7% compared with 2018 (Figure 6). The increase in spending was bigger than the increase in visits (1%). One factor in the larger increase in spending is the higher average spend per day (increasing from £91 in 2018 to £98 in 2019) and the overall average spend per visit increasing from £658 in 2018 to £696 in 2019.
Holidays were still the most common reason for visiting the UK
The rise in visits to the UK was driven by an increase in holiday visits (Figure 7). Despite a small fall in 2018, holiday visits have generally increased year-by-year, and in 2019 they accounted for 16.9 million visits and 41% of all visits; this was the highest number of holiday visits recorded.
Visits to friends and family also rose over the last two decades; however, the figures have been stable recently, with over 12 million visits reported for each of the years 2017, 2018 and 2019. Business visits saw a peak in 2016 of 9.4 million, but the numbers have fallen slightly every year since to 8.7 million in 2019. Numbers of visits to the UK for miscellaneous reasons have changed relatively little in the years since 1999, with 2.9 million estimated visits in 2019.
The three countries whose residents visited the UK most in 2019 were the US, France and Germany. This has been the same for the last five years. Residents of Germany have remained the third most frequent visitors during this time, while those from France and the US swapped first and second places in 2016; the US has been at the top of this list since 2017. There were 4.5 million visits to the UK by residents from the US in 2019, more than any other country.
Of the 10 countries whose residents visited the UK the most frequently in 2019, eight were EU countries (Figure 8). The non-EU countries were the US and Australia (1.0 million visits, the 10th highest total). The countries featuring in the top 10 list have not changed since 2010.
London, Edinburgh, Manchester and Birmingham each attracted over 1 million stays from overseas visitors
London attracted 21.7 million overnight visits in 2019, far more than any other town or city. More than half of the visits to the UK included a trip to London. Overnight visits to cities other than London are shown in Figure 9. Edinburgh (2.2 million), Manchester (1.7 million) and Birmingham (1.1 million) each received more than 1 million overnight visits.
Most of overseas residents’ visits to Scotland were for holidays
Overseas residents made 3.5 million overnight visits to Scotland in 2019, of which 61% were for holidays and 25% were to visit friends and relatives (Figure 10). There were a million overnight visits to Wales, of which 39% were for holidays and 42% to visit friends and relatives, while 51% of overnight visits to London were for holidays and 24% to visit friends and relatives.
Back to table of contents
We have made methodological improvements to the travel and tourism estimates from the International Passenger Survey (IPS). These relate to the survey’s weighting process. The improvements are described in this section in outline. A detailed technical paper will be published in due course, setting out the changes comprehensively. These changes affect only travel and tourism and not estimates of long-term international migration.
The estimates presented in this article were produced using the new estimation method. The new method was also used to produce final results for 2019 and a revised series for 2009 to 2018. The impacts of the new method, and how these differ from the old, are presented in Section 6: Impacts of the new IPS methodology.
These changes have been planned for some time and have been referenced in previous travel and tourism publications. A statement about the upcoming changes was published on 16 March 2020.
At the time the statement was published, the new adjustment method was still being improved, so some of the published results are slightly different to the indicative results presented in the statement.
Why we made these improvements
There was substantial evidence that the previous estimation method was not providing accurate results for certain groups, in particular visits by Chinese residents. Users of the data reported that these figures were likely to be substantial underestimates. The evidence came from comparisons with other sources, notably Home Office landing card data.
Another strand of evidence came from the fact that the survey interviewed more passengers at the starts of their visits than at the end, whereas the numbers would be expected to be approximately equal. For the IPS, we interview passengers both as they enter and leave the UK. Interviews conducted at the start of visits are very brief and establish only the passengers’ nationality and country of residence. Interviews conducted at the end of visits capture the detailed information relating to the visit, for example, how long they stayed and how much they spent. This is true both for UK residents visiting abroad and for people living in other countries visiting the UK. All published estimates from the IPS are based on interviews conducted at the end of visits, although all visits receive weights in IPS weighting.
This discrepancy between passengers interviewed at the start and end of their visits has been termed the IPS “imbalance”.
One outcome of the imbalance is that published IPS estimates did not align closely to the administrative data on passenger numbers used in the weighting process. These administrative data are produced by the Civil Aviation Authority (CAA), the Department for Transport (DfT), Eurotunnel and Eurostar.
Researching the issues
The Office for National Statistics (ONS) launched a programme of research into the issue, both to understand the causes and to develop an appropriate and robust adjustment method. This drew on the expertise of users of the survey, and it was conducted in collaboration with ONS Methodology and academic experts. Home Office landing card data, where applicable, were a central source in the development and evaluation of the new method.
The main findings of the research for overseas residents were that:
estimates of overseas residents, in general, showed slight under-representation, owing to a higher tendency not to respond to the survey; the effect was slight, and for most countries the survey represented visits by overseas residents accurately
residents of certain countries (notably China) showed a much higher tendency to be under-represented in the departures data
the proportions of overseas residents by country among total arrivals were correct; to recap, overseas passengers arriving in the UK (at the start of their visits) are included in the IPS weighting process but not in the published estimates, and this was a crucial finding in the development of the method
The main findings of the research for UK residents were that the IPS underestimated the numbers of UK residents returning to the UK. To understand why this happens, it is necessary to explain a little about how the survey sampling operates.
The IPS generally samples from passengers crossing a line at set intervals. For example, every 20th person leaving security or passport checks. Sometimes, at busy times, IPS interviewers cannot interview the selected person because the sheer volume of passengers mean that they are all already occupied interviewing other passengers. The survey compensates statistically for the passengers missed, but this process works correctly only if the passengers missed are missed at random, relying on the assumption that the missed passengers have, on average, the same characteristics as those who are interviewed.
The new research concluded that this crucial assumption did not hold up as well as had been expected since changes to the sample design were implemented in 2009. In fact, the busiest times, when interviews are more likely to be missed, occur when UK residents arrive back in the UK in large numbers. It happens less when overseas residents arrive in the UK. This means that a UK resident arriving back in the UK after a visit abroad is more likely to be missed than an overseas resident arriving at the start of their visit. Hence, missing interviews because of surges of high passenger flow leads to bias in data collected during arrivals, with proportionally fewer UK residents interviewed as a result. This happens only on arrivals, not departures, because of the big surges of passengers sometimes seen in airport arrivals. Traffic at departures is generally more manageable.
How the new method works
The new adjustment method was developed in collaboration with ONS Methodology and academic experts. The following is a description of the method; a complete technical explanation will be published in due course.
- Step one: a small uplift is applied to the total estimates of overseas visits to the UK; this compensates for the small general under-representation of overseas visitors identified by the research.
The next two steps rest on the essential finding that the arrivals data provide the correct proportions of overseas residents.
Step two: the new method compares the results obtained in the arrivals and departures data (specifically, the proportions by country, among overseas residents, are compared); this is because the research concluded that the IPS captured these proportions correctly in arrivals.
Step three: where this comparison provides robust statistical evidence that the departures figures for a particular country are underestimates, an adjustment is applied to bring the departures figures in line with the proportion by country, among overseas residents, obtained in arrivals; conversely, the comparison might reveal that the departures figure is an overestimate, in which case it can be adjusted downwards.
The adjustment criteria are complex and will be described fully in the upcoming technical paper.
- Step four: the previous steps compute the estimates of overseas residents visiting the UK, while administrative data provide total numbers of international travellers; these figures combined are used to calculate the numbers of UK residents visiting overseas.
The new method made published estimates of UK and overseas visitors consistent with administrative data on total passenger numbers entering and leaving the UK, collected by the CAA and DfT.
Further planned improvements
We are confident that the new imbalance adjustment method gives improved estimates of travel and tourism. However, we are keen to improve the estimates obtained “at source” by the IPS and to reduce the need for the adjustments. The new method will adapt to such improvements automatically, and the scale of the adjustments will reduce.
We are researching measures to increase response rates among those nationals who have proved they are less likely to respond to the IPS. We are assessing the reasons why visitors from some countries are, overall, less likely to take part, with a view to addressing the issues and improving survey response. We are also launching an investigation into passenger flows on arrivals at busy times and measures that can be taken to improve the results obtained at such times.
Collection of Home Office landing card data ceased in spring 2019. Landing cards were used extensively in the development of this new method, but the method is self-contained and does not require administrative sources to operate. It will be valuable to use other sources of administrative data to monitor IPS results and the continued efficacy of the imbalance adjustment method over time. We will monitor and evaluate such sources for the future.
Impact on other ONS outputs
This revised methodology will affect estimates of travel services, which feed into the monthly UK trade publication as well as the gross domestic product (GDP) first quarterly estimate and GDP quarterly national accounts releases. It will also affect tourism figures within household’s final consumption expenditure data within quarterly GDP publications. These changes will be introduced during 2021 alongside our annual updates to data and methodology as part of the Blue Book publication; until then, these statistics will continue to be produced under the existing weighting method.Back to table of contents
The figures presented in this article have been produced using an improved methodology, relating to adjusting the weights applied during the survey’s processing. The improved method has been used to produce the final travel and tourism estimates for 2019 and revised estimates for 2009 to 2018.
The background to this change, and the principles of the new method, are described in Section 5: Improved methodology for the estimates. This section presents the impact of the change on International Passenger Survey (IPS) estimates, giving comparisons of some of the main results computed using both the old and new methods for 2009 to 2018.
Total international visitor numbers are higher under the new estimates
The new methodology produces higher estimates of UK residents visiting abroad and slightly higher estimates, overall, of overseas residents visiting the UK (Figure 11).
Overseas residents’ visits to the UK were 4% higher for 2009 under the new method than the previous method (increasing from 29.9 million to 31.1 million), rising to a larger increase of 6% in 2018 (from 37.9 million to 40.3 million visits).
UK residents’ estimated visits abroad were 8% higher under the new method for 2009 (from 58.6 million to 63.5 million). Between 2010 and 2017, the new method increased estimates by around 20%. For 2018, visits abroad were 26% higher under the new method (rising from 71.7 million to 90.6 million).
The impact of the new method on estimates of spending showed similar patterns to visits but with larger increases for both UK residents and overseas residents (Figure 12). Overseas residents spent 6% more in 2009 and 16% more in 2018, when compared with previously published figures. This was a rise from £16.6 billion to £17.6 billion in 2009 and from £22.9 billion to £26.5 billion in 2019.
Spending by UK residents rose more steeply, increasing from £31.7 billion to £34.5 billion in 2009 (an increase of 9%) and from £45.4 billion to £58.1 billion in 2018 (an increase of 28%). This increase in spending abroad generally reflects the increased visits abroad.
The impact of the new method on estimates of visitors to the UK is not uniform, and it varies greatly by country of residence. It can also vary over time for a specific country. Estimates of visits from some countries have increased, notably China, while estimates of visits from a few countries have decreased, for example, Australia.
As expected, Chinese (excluding Hong Kong) residents’ visits showed a marked difference using the new method. Visits more than doubled in 2018 from 472,000 visits to 1 million (Figure 13). There was a smaller increase in 2009 from 128,000 to 192,000 visits, an increase of 50%.
Visits to the US under the new method were similar to previously published figures between 2009 and 2014, but they were higher under the new method for 2015 to 2018. The new method estimated that 4.6 million US residents visited the UK in 2018; this was 18% higher than the previously published estimate of 3.9 million (Figure 14).
Visits to Wales and Scotland were slightly higher in recent years using the new methodology
The new method estimated similar numbers of visits from overseas to Wales and Scotland, compared with the previous method, for 2009 to 2015. The new method produced higher numbers of visits for 2016 to 2018 (Figure 15). There were 3.7 million visits to Scotland in 2018 under the new method compared with 3.5 million visits under the previous method (5% higher under the new method). There were 990,000 visits to Wales under the new method compared with 940,000 visits under the previous method (5% higher under the new method).
Back to table of contents
The collection of the IPS data
International Passenger Survey (IPS) data are collected by a team of over 200 interviewers who are recruited and trained specifically to work on the IPS. Interviews are carried out at air and sea ports, on board vessels leaving or returning to the UK, and on board the Eurotunnel Trains. Interviews are carried out on all days of the year, apart from Christmas Eve, Christmas Day, Boxing Day and New Year’s Day.
Until recently, responses were initially recorded on paper forms. In September 2017, data collection on tablets started to be phased in, and this implementation was completed in April 2018. Now data are keyed directly into the collection program, which includes a series of electronic checks. The data are then transmitted to the Office for National Statistics’ (ONS’) headquarters where a series of further quality and accuracy checks are made on the data before processing and analysis. Analysis has shown that results obtained on tablets are comparable to those collected on paper forms. There is a reasonable change seen in a few series but the recommendation is these changes are not large enough or there is not enough evidence to attribute them to the introduction of the tablets.
IPS response rates
Sample surveys such as the IPS depend on achieving high levels of response from the public. Non-respondents often have different characteristics of travel and expenditure compared with those who do respond, and this can lead to biases being introduced into the results.
The response rates for the air, sea and the Channel Tunnel samples are shown in Table 1. These response rates relate to complete and partial interviews. The overall response rate in 2019 was 74.2%.
|Total IPS response rate||Arrivals||74.8||73.0||70.8||71.8||72.0||69.6||71.5||72.2||72.5||71.6|
|IPS response rate (Air)||Arrivals||73.2||71.7||68.7||70.6||70.7||68.3||69.9||70.0||70.9||70.2|
|IPS response rate (Sea)||Arrivals||86.8||78.1||83.1||82.2||80.6||76.3||87.4||81.9||83.2||78.4|
|IPS response rate (Tunnel)||Arrivals||82.3||83.6||82.8||73.6||75.9||78.6||75.2||88.0||79.3||80.9|
Download this table Table 1: International Passenger Survey response rates for 2018 and 2019 estimates, percentage of complete or partial responses.xls .csv
Accuracy of overseas travel and tourism estimates
The estimates produced from the IPS are subject to sampling errors that result because not every traveller to or from the UK is interviewed on the survey. Sampling errors are determined both by the sample design and by the sample size – generally speaking, the larger the sample supporting a particular estimate, the proportionately smaller is its sampling error. The survey sample size is approximately 70,000 per calendar quarter.
Table 2 shows the 95% confidence intervals for the 2019 estimates of the total number of visits, number of nights and expenditure for both overseas residents visiting the UK and UK residents going abroad. These represent the interval into which there are 19 chances out of 20 that the true figure (had all travellers been surveyed) would lie.
If, for example, the relative 95% confidence interval relating to an estimate of 10,000 was 5.0%, there would be 19 chances out of 20 that the true figure (if all travellers had been surveyed) would lie in the range 9,500 to 10,500.
The following guidelines are provided to aid in the interpretation of the estimates and to enable their reliability to be assessed:
percentage of confidence interval is below 10%: precise
percentage of confidence interval is between 10% and 20%: reasonably precise
percentage of confidence interval is between 20% and 40%: acceptable
percentage of confidence interval is over 40%: unreliable (these estimates should be used with caution for practical purposes)
(+/- % of the estimate)
|Overseas visitors to the UK||Number of visits ('000s)||40,857||2.5%|
|Total earnings (£million)||28,448||3.4%|
|Number of visitor nights ('000s)||289,628||3.9%|
|UK residents going abroad||Number of visits ('000s)||93,086||1.4%|
|Total expenditure (£million)||62,325||2.8%|
|Number of visitor nights ('000s)||904,905||2.6%|
Download this table Table 2: International Passenger Survey confidence intervals for 2019 estimates.xls .csv
Confidence intervals for purposes for visit, region of the world, regions of the UK visited, and individual country of visit to and from the UK are presented for easy access in the relevant datasets.
More quality and methodology information on strengths, limitations, appropriate uses, and how the data were created is available in the IPS QMI. This will be updated shortly to reflect recent changes to the survey’s processes.Back to table of contents
The figures relate to the number of completed visits, not the number of visitors. Anyone entering or leaving more than once in the same period is counted on each visit.
Trips that do not involve an overnight stay abroad by UK residents, as well as day trips to the UK by overseas residents, are included in the total figures for visits and expenditure, but figures presented at lower levels of geography relate to overnight says only.
An overseas visitor is a person who, being permanently resident in a country outside the UK, visits the UK for a period of less than 12 months. UK citizens resident overseas for 12 months or more coming home on leave are included in this category. Visits abroad are visits for a period of less than 12 months by people permanently resident in the UK (who may be of foreign nationality).
Visiting multiple countries
When a resident of the UK has visited more than one country, expenditure and stay are allocated to the country stayed in for the longest time.
Visits for miscellaneous purposes include those for study; to attend sporting events; and for shopping, health, religious or other purposes, together with visits for more than one purpose when none predominates (for example, visits both on business and on holiday). Overseas visitors staying overnight in the UK on their way to other destinations are also included in miscellaneous purposes.
Earnings and expenditure
Earnings refer to spending in the UK by overseas residents, whereas expenditure refers to spending abroad by UK residents.Back to table of contents
A major strength of the International Passenger Survey (IPS) is that it is the main source of information in the UK on international travel and tourism and associated earnings and expenditure. It has been running since 1961, so it provides a comprehensive time series of travel and tourism, which can be useful for identifying long-term trends and patterns or undertaking time series analysis and forecasting. In addition, interviewing at the principal air, sea and tunnel routes and the use of a dedicated field force gives the survey some uniqueness.
IPS survey data are subject to both sampling and non-sampling errors. About 90% of passengers entering and leaving the UK have a chance of being sampled in the survey. The remainder are either passengers travelling at night, when interviewing is suspended, or on those routes too small in volume or too expensive to be covered. Furthermore, the number of survey interviews on particular routes or for some main reasons for visit, such as playing sports or getting married, are sometimes small and consequently attract higher sampling errors. This also applies to visits to or from countries with low visit numbers.Back to table of contents
Contact details for this Article
Telephone: +44 (0)1633 455731