163 countries of birth · Census 2021 · England and Wales

The fiscal impact of migration to Britain

This page estimates what people born in each country pay in tax and cost in public spending in a year, and what one person would pay and cost over a lifetime. It is built from UK sources — the census, HMRC, DWP, ASHE and the Treasury’s own spending figures — rather than by taking a Danish or Dutch result and applying it here. Where the evidence does not settle a question, the page gives the reader the control rather than picking an answer quietly.

What the model finds

On the assumptions currently set — with defence, debt interest and central government charged equally to everyone — the 9.9 million people in England and Wales born outside the United Kingdom come to about £634 a person a year, or £6.3bn in total. The UK-born, on exactly the same assumptions, come to −£4,984 a person. Neither figure is a surplus or a deficit in the government’s accounts: the country as a whole spends more than it raises, so most groups are negative once every cost is charged to somebody, and the useful comparison is between rows rather than against zero.

81 of the 163 countries are positive on these assumptions and 82 are negative, covering 5.8 million and 4.2 million people respectively. The single largest influence on which side of the line a country falls is not its origin but its age structure: groups that arrived recently are of working age and pay more than they cost, and groups that arrived decades ago have reached retirement and cost more than they pay. The same is true of the UK-born, who are the oldest group on the page.

A fiscal balance is not the economic effect of immigration. It counts tax paid and public money spent, and nothing else. It does not count what people produce, the businesses they start, the jobs they fill in the health service that would otherwise go unfilled, the effect on anyone else’s wages or rents, or anything that is not a transaction with the state. A person can be a net cost on this page and a large net benefit to the economy, and the reverse. Nothing here supports a conclusion about whether a person or a country’s migrants are good or bad for Britain, and the page does not draw one.

Checked against Denmark

The Danish Ministry of Finance publishes net contribution by country of origin, which makes it the only external check available on the rankings here. Put on the same footing — average costing, and each group charged its UK-born children — the two agree on the sign for 23 of the 35 countries both cover, with a rank correlation of 0.66. They diverge most for origins whose Danish population arrived as refugees. See the comparison →

How it is built

Everyone is classified by country of birth, which is what the census asks and what stays true for life. Not nationality, which changes when someone naturalises, and not ethnicity, which is a different question about a different thing. Children born in Britain to parents born abroad are UK-born and appear in the UK-born row, not in their parents’ row: a person’s balance belongs to them.

Income is not a single average wage. Each country’s residents are crossed with the occupations they actually work in, at three-digit detail, and each occupation carries the full pay distribution ASHE publishes for it. Tax is then computed by running the statutory schedule over that whole distribution, because tax is progressive and the tax due on the average wage is not the average of the tax due.

Spending is the Treasury’s own published total for each function, divided by a stated rule: health by age, education by the children of each school age, the State Pension by how many National Insurance years a person could have accrued since they arrived, benefits by DWP’s recorded caseload, housing by how many households actually rent from a council. The twelve lines reproduce their published totals to within a tenth of a per cent, which is the test that the parts add up to the whole.

Two things are deliberately absent. There is no assumption anywhere that a country of origin makes someone more expensive to treat, to police or to educate: health is allocated by age, policing per adult, and no offending or morbidity figure by nationality enters the model. And the whole capital value of a council house is never charged as an annual cost — the annual cost of social housing is its subsidy and its management, which is what the housing line contains.

The full method, every source, and what each figure can and cannot support →

What the evidence does not settle

These are the choices that change the answer. They are not details: moving the first one alone shifts every row by about £3,300 a person and changes the sign of several countries. Each is set to the more conservative reading by default.

Defence, debt interest and central government

About £198bn of spending does not rise when one more person arrives, but somebody pays for it. Charging none of it to a new arrival is marginal costing; charging everyone an equal share is average costing. The difference is roughly £3,300 a head and it decides the sign of several countries. Average costing is the default here because it is what the Danish Ministry of Finance does, which is what makes the two sets of figures comparable.

Children born here to a migrant parent

A child born in Britain is UK-born and is counted in the UK-born row. Denmark reports descendants as a category of their own, which its population register supports and the census does not: ONS never asks where a person's parents were born. Turning this on charges the schooling and children's services of the 2.3 million UK-born children living in a migrant-headed household to the parent's country of birth instead. It is a transfer, not a new cost — the UK-born row is credited the same £25bn — and it is one-sided, because the tax paid by the grown-up children of earlier migration stays in the UK-born row where it cannot be identified. It therefore reads worst for the communities settled longest.

How steeply health spending rises with age

The age profile of NHS spending is the model's weakest input: the published curves are behind bot protection and have not yet been obtained, so an interim shape stands in. Flat gives every age the same cost, which makes young populations look expensive; steep makes them look cheap. Move it and see how much of the answer depends on it.

If employment were different

The census counted employment in March 2021, in the third national lockdown. Shifting every group's employment rate by a tenth in each direction shows how much of the result rests on that measurement.

If pay were different

Pay comes from each group's occupations crossed with ASHE's distribution for those occupations, not from a survey of migrants. Shifting it a tenth either way tests how much that join carries.

Employer National Insurance

Counted here as revenue the job produces. Most economists hold that it is borne by the worker in lower wages, in which case it is not extra revenue from that person at all. Switch it off to see the other reading.

Corporation tax, capital gains, stamp duty and the rest

£200bn allocated in proportion to income, which is contestable: corporation tax in particular falls partly on shareholders, many of whom are not UK residents. Leaving it out is the most conservative reading.

Scaling revenue to published receipts

The model computes tax from the statutory schedule over ASHE's earnings distribution, which counts employee jobs rather than people, so it finds about a sixth less income tax than HMRC collected. Scaling each revenue line to its published receipt corrects the level; the same factor applies to every country, so it moves no country past another.

Inputs still standing in for a source. These are placeholders, stated as such, so the model can run and be checked rather than published as though settled:

The age profile of health spending. NOT A SOURCED PROFILE. A placeholder shape, stated so the model runs and so the sensitivity analysis has something to move. Every health figure, and so every net figure, is reported as a range across `sensitivity.health_gradient` until one of the canonical sources above is in hand. Do not quote the central health number as though it were measured.
Indirect tax as a share of disposable income. Shape taken from the published pattern in the ONS series; the exact current figures still need reading out of the ONS reference tables. Flagged rather than presented as final.

Country by country

Every figure is per person of that group per year, averaged over everyone born there — children, workers and pensioners alike — so a group with many children or many pensioners shows a lower figure than its workers alone would. Sort on any column. The confidence rating says how much is actually known: 56 of 163 countries have an administrative benefit figure that describes the same people the census counted, and the rest carry a regional stand-in, marked in the table and explained on the country’s own page.

75 of 163 countries. The UK-born row is −£4,984 a person a year on the same assumptions, and is the comparison every other row should be read against.

								Confidence
India	920,363	55.8%	£40,859	£18,513	£15,787	£2,726	£2,509	medium
Poland	743,082	76.1%	£32,393	£17,969	£15,107	£2,862	£2,127	high
Pakistan	623,553	44.6%	£31,403	£12,272	£15,119	−£2,848	−£1,776	medium · stand-in
Romania	538,842	73.6%	£32,605	£16,615	£14,986	£1,628	£877	high
Ireland	324,668	41.1%	£42,595	£17,062	£24,072	−£7,010	−£2,276	medium · stand-in
Italy	276,668	52.8%	£39,185	£16,591	£18,680	−£2,089	−£578	medium
Bangladesh	273,044	44.9%	£25,320	£10,218	£14,458	−£4,240	−£1,158	medium · stand-in
Nigeria	270,754	63.8%	£37,250	£18,805	£14,680	£4,125	£1,117	high
Germany	263,370	62.3%	£40,801	£20,724	£17,721	£3,003	£791	medium · stand-in
South Africa	217,173	69.2%	£44,925	£23,775	£16,052	£7,723	£1,677	medium · stand-in
United States	203,280	53.9%	£47,737	£19,979	£14,216	£5,763	£1,172	medium
China	182,380	45.6%	£37,492	£13,849	£12,666	£1,183	£216	medium
Spain	166,916	54.6%	£40,133	£17,042	£17,397	−£355	−£59	medium
Lithuania	164,153	75.4%	£32,507	£17,563	£14,845	£2,718	£446	high
Portugal	156,299	63.7%	£32,947	£16,034	£18,437	−£2,403	−£376	high
France	155,328	63.9%	£46,502	£22,391	£14,826	£7,565	£1,175	medium
Philippines	155,016	73.2%	£33,547	£19,030	£13,732	£5,298	£821	high
Bulgaria	150,262	71.9%	£31,782	£16,312	£16,689	−£377	−£57	high
Sri Lanka	144,303	60.7%	£33,098	£16,726	£16,646	£80	£12	high
Jamaica	142,149	46.0%	£30,795	£14,254	£22,250	−£7,995	−£1,137	high
Kenya	134,933	52.7%	£38,806	£18,247	£19,379	−£1,131	−£153	medium · stand-in
Ghana	132,581	67.2%	£30,766	£16,769	£15,650	£1,119	£148	high
Turkey	127,634	53.1%	£32,308	£13,561	£16,111	−£2,550	−£325	medium
Zimbabwe	124,144	71.3%	£39,027	£21,801	£14,280	£7,521	£934	high
Hong Kong	121,426	47.9%	£39,195	£15,558	£15,681	−£123	−£15	medium · stand-in
Australia	112,971	67.9%	£48,041	£24,708	£13,668	£11,040	£1,247	medium
Brazil	111,478	66.9%	£33,409	£16,360	£12,779	£3,581	£399	medium · stand-in
Iran	109,164	43.4%	£37,112	£13,900	£19,122	−£5,222	−£570	high
Somalia	108,921	45.6%	£24,996	£9,827	£16,348	−£6,522	−£710	medium · stand-in
Hungary	95,399	73.1%	£33,098	£17,935	£15,744	£2,192	£209	medium
Iraq	89,385	37.3%	£33,049	£11,109	£15,040	−£3,931	−£351	medium · stand-in
Latvia	88,583	70.1%	£31,787	£16,303	£15,450	£853	£76	medium
Afghanistan	85,696	42.3%	£27,671	£9,288	£14,119	−£4,831	−£414	medium · stand-in
Greece	80,121	64.6%	£44,662	£22,189	£14,969	£7,220	£578	medium
Nepal	78,628	63.7%	£30,834	£15,707	£14,067	£1,640	£129	medium
Slovakia	73,938	59.1%	£31,700	£14,240	£17,520	−£3,280	−£242	medium
Cyprus	72,557	41.7%	£37,740	£15,069	£21,579	−£6,510	−£472	medium · stand-in
Canada	69,096	63.1%	£46,548	£23,005	£14,641	£8,364	£578	medium
Albania	68,664	58.6%	£31,987	£13,319	£14,681	−£1,363	−£94	medium · stand-in
Netherlands	67,375	48.9%	£42,178	£16,169	£19,005	−£2,836	−£191	medium
Uganda	62,067	54.8%	£36,188	£17,410	£19,257	−£1,847	−£115	medium · stand-in
Malaysia	61,516	53.6%	£43,857	£19,620	£15,589	£4,030	£248	medium
Russia	56,055	61.4%	£44,000	£21,174	£13,244	£7,931	£445	medium
Moldova	55,386	72.7%	£32,133	£15,787	£14,872	£915	£51	medium · stand-in
Thailand	49,725	53.4%	£26,437	£11,704	£14,319	−£2,615	−£130	medium
New Zealand	49,145	71.1%	£48,380	£26,270	£14,034	£12,236	£601	medium
Czechia	46,169	59.8%	£32,142	£14,773	£16,835	−£2,062	−£95	medium
Egypt	43,519	41.6%	£47,932	£17,247	£18,848	−£1,601	−£70	medium
Syria	42,879	25.5%	£32,636	£7,553	£14,761	−£7,208	−£309	medium · stand-in
Singapore	40,925	54.4%	£42,896	£19,177	£15,837	£3,339	£137	medium · stand-in
Ukraine	39,510	69.4%	£37,215	£18,994	£12,898	£6,096	£241	medium
Mauritius	38,859	55.0%	£37,404	£17,957	£18,471	−£514	−£20	medium
Colombia	37,833	68.9%	£30,275	£16,583	£13,493	£3,090	£117	medium · stand-in
Sudan	37,390	34.7%	£35,405	£10,651	£20,110	−£9,459	−£354	medium
Vietnam	37,222	49.2%	£29,513	£12,613	£15,777	−£3,164	−£118	medium
Eritrea	36,370	41.0%	£27,810	£9,152	£20,035	−£10,883	−£396	medium
Japan	35,024	53.4%	£42,938	£18,003	£13,674	£4,329	£152	medium
Tanzania	34,861	50.2%	£36,276	£16,924	£20,608	−£3,683	−£128	medium · stand-in
Sweden	33,009	56.6%	£44,476	£19,170	£16,551	£2,620	£86	medium
Morocco	32,645	47.3%	£29,988	£12,433	£15,921	−£3,488	−£114	medium · stand-in
Saudi Arabia	31,387	33.5%	£45,448	£12,153	£13,571	−£1,418	−£45	medium · stand-in
Belgium	31,276	53.2%	£45,195	£18,617	£16,726	£1,891	£59	medium
Kosovo	30,515	62.1%	£32,487	£15,357	£15,160	£197	£6	medium · stand-in
Algeria	29,851	52.1%	£30,457	£13,119	£18,354	−£5,235	−£156	medium
Democratic Republic of the Congo	29,428	55.6%	£26,583	£12,237	£15,205	−£2,968	−£87	medium · stand-in
Zambia	26,252	69.4%	£40,797	£22,982	£16,109	£6,873	£180	medium · stand-in
Sierra Leone	24,799	63.6%	£32,050	£16,809	£16,989	−£180	−£4	low
Malta	24,476	47.3%	£38,897	£17,526	£21,853	−£4,327	−£106	low · stand-in
Switzerland	22,824	50.5%	£44,225	£17,670	£15,826	£1,845	£42	low
United Arab Emirates	22,604	33.6%	£44,709	£11,778	£15,017	−£3,239	−£73	low · stand-in
Israel	22,551	50.3%	£41,220	£16,720	£16,182	£538	£12	low
Kuwait	22,177	29.3%	£40,485	£10,057	£14,055	−£3,998	−£89	low · stand-in
Ethiopia	22,148	51.2%	£30,633	£12,572	£14,954	−£2,382	−£53	low · stand-in
Trinidad and Tobago	21,673	49.1%	£41,068	£18,325	£20,056	−£1,731	−£38	low · stand-in
Denmark	21,008	55.6%	£43,466	£18,998	£16,522	£2,475	£52	low

What this page cannot tell you

It is a photograph of March 2021. The census counted people on census day, before the Ukrainian arrivals and before the Hong Kong BN(O) route opened. Both countries are badly understated here and the ONS discontinued the quarterly series that would have updated them.
It is England and Wales. Scotland and Northern Ireland run their own censuses on their own classifications, and public spending totals are UK-wide, so the two are put on the same footing by scaling rather than by counting.
It is a static incidence calculation, not a forecast. It asks what these people would pay and cost under today’s rules. It does not model what happens to the economy if the population changes.
The confidence ratings are not decoration. A country with a few thousand residents has an earnings distribution built from a handful of census cells. Read the small countries as an order of magnitude and nothing finer.