Few subjects in British politics generate more heat and less arithmetic than immigration. The public conversation is conducted almost entirely in aggregate numbers, while the decisions that actually produce those numbers are made sector by sector, in response to specific and often long-standing shortages of specific skills.
That mismatch is the source of a great deal of mutual incomprehension. Employers experience a shortage of welders, radiographers, HGV drivers or bricklayers. Voters experience pressure on housing, primary school places and GP appointments. Both experiences are real, and neither is addressed by arguing about a single headline figure.
How shortage becomes policy
The formal mechanism is unglamorous. Independent advice from the Migration Advisory Committee assesses whether a given occupation is genuinely short of labour, whether wages have risen as economic theory would predict, and whether recruitment from overseas is a reasonable response. Where domestic training could plausibly fill the gap within a few years, the case for a visa route weakens considerably.
In practice, employers rarely wait for the training pipeline. A construction firm with a contract to deliver cannot pause for a three-year apprenticeship cycle, so it recruits where it can. Repeated across thousands of firms, that produces migration flows which then become the subject of political argument, disconnected from the capacity decisions that caused them.
The training pipeline problem
Britain has an unusual pattern of skills investment. Higher education participation is high by international standards, while intermediate technical training is comparatively thin. Successive reforms to further education funding, apprenticeship levies and technical qualifications have changed the labels more often than the outcomes. Work by the CIPD has documented how employer training spend per employee has drifted downwards over two decades.
The consequence is a structural dependence on skills formed elsewhere. That is not automatically a problem. Every advanced economy imports talent. It becomes a problem when it is unplanned, when it substitutes for domestic capacity rather than complementing it, and when the public is told the flow is temporary while the underlying cause is permanent.
What the population data actually shows
Migration statistics are estimates, revised frequently, and built from survey and administrative sources that were not designed to be combined. The Office for National Statistics has been candid about the uncertainty involved, particularly in the years since travel patterns and student behaviour changed sharply. Confident claims built on a single quarter of data are usually claims about politics rather than demography.
What the data does show consistently is composition. Students and their dependants, health and care workers, and skilled workers on employer-sponsored routes account for the large majority of arrivals. Each of these has a distinct policy logic and a distinct fiscal profile, and treating them as a single undifferentiated flow guarantees bad policy.
The fiscal argument, honestly stated
Forecasts from the Office for Budget Responsibility treat net migration as a significant input to the public finances, largely because working-age arrivals pay tax before they draw heavily on age-related spending. That is an accounting fact rather than a moral argument, and it cuts both ways: the fiscal benefit depends on earnings, and low-wage routes deliver much less of it.
Where the honest debate lies is in the interaction with infrastructure. A rising population raises tax receipts and demand for housing and services simultaneously. Whether that nets out positively depends almost entirely on whether the country builds. A system that welcomes people and refuses to permit homes has chosen the worst of both arguments.
A more useful frame
The productive question is not how many people should come, but which capacities Britain intends to build domestically and on what timetable. Answering that would let visa policy follow from an industrial and workforce plan rather than substitute for one. It would also make the numbers predictable, which is what most of the public appears to want.
None of this resolves the underlying disagreement about the pace of change in communities, which is real and not reducible to economics. But it would at least connect the argument to the levers that governments actually control.
What a credible skills settlement would look like
The uncomfortable truth for both sides of the immigration argument is that visa policy and training policy are the same policy, and Britain has spent two decades treating them as separate departments. Employers who complain that the skilled worker route is too expensive are often the same employers whose apprenticeship levy contributions sit unspent. Ministers who promise to reduce net migration rarely explain which sector will absorb the resulting vacancies, or over what timescale a domestic pipeline could realistically replace them.
A settlement that survived contact with reality would do three things at once. It would price visas to reflect the genuine administrative cost rather than as a revenue line, because a five-figure bill per hire is a tax on growth in exactly the firms the economy needs to expand. It would tie any sector-specific concession to a verifiable domestic training commitment with published outcomes, so that concessions become temporary bridges rather than permanent dependencies. And it would fix the data, because the current argument is conducted with statistics that are revised heavily months after the political moment has passed.
None of that is glamorous, and none of it produces a headline number. But the sectors under most strain — health and social care, construction, engineering, parts of hospitality — share a common profile: pay that has not kept pace with the skill demanded, training routes that collapsed after 2010, and a workforce that is ageing faster than it is being replaced. Immigration has been the shock absorber for those structural problems. Removing the absorber without addressing the structure simply transfers the cost from the visa system to waiting lists, delivery timetables and unbuilt homes.
