What the lost revenue could fund: the loss in nurses, country by country

A billion is a number most people cannot picture. So here is a way to feel the scale without pretending it is a budget line. Take each country's estimated annual loss to corporate profit-shifting, and divide it by what that country pays a registered nurse, using its own published salary data. The result is not a spending plan and not a promise. The lost money is not sitting in a ring-fenced pot waiting to hire anyone. It is a sense of size, and on that measure the size is hard to ignore.

Read this first, because it is the honest frame. Every figure on this page is an illustrative equivalence, not a budget proposal. Two things have to be true for one of these comparisons to be worth anything, and we hold both:

• The loss figure is primary-sourced. Every country loss below comes from the Tax Justice Network's State of Tax Justice 2024, specifically the "corporate tax abuse, suffered" column of its published country-and-regional dataset, on 2021 OECD country-by-country reporting data. These are careful estimates built on the OECD's own data, not audited government accounts, and they carry real methodological uncertainty, which we explain at the foot.
• The unit cost is primary-sourced too, and we tell you when it is not from an official pay scale. The UK and US nurse salaries come from official sources (NHS Agenda for Change and the US Bureau of Labor Statistics). The Canada, Australia and India figures come from commercial salary aggregators or a constructed proxy, and we label each one, because an illustration built on a soft input should announce itself.

What this page does NOT do is restate the loss-suffered and loss-inflicted ledger; that lives on the league table with both cuts and all the rankings. Here we only use the loss-suffered figure, and only as the numerator in one vivid sum.

United Kingdom

Estimated annual corporate tax loss: about US$16.9 billion (roughly £13.3 billion). Unit cost: an NHS Band 5 registered nurse in England starts at £32,073 for 2026/27, the entry point on the Agenda for Change pay scales effective 1 April 2026.

The sum: £13.3 billion divided by £32,073 is about 415,000 entry-level nurses' salaries. For scale, NHS England's nursing workforce is in the region of 350,000 to 370,000, so the lost corporate tax is comparable in size to the entire NHS nursing pay bill in England. Illustrative only; the loss is not a hypothecated fund and would not automatically flow to NHS staffing.

United States

Estimated annual corporate tax loss: about US$32.6 billion. Unit cost: the median annual wage for a US registered nurse was US$93,600 in May 2024 (Bureau of Labor Statistics).

The sum: US$32.6 billion divided by US$93,600 is about 348,500 nurses at the median wage, more than 10% of all employed registered nurses in the United States. A separate and larger figure puts the loss the US inflicts on itself through its own multinationals at roughly US$271 billion over 2016 to 2021, which implies a higher annual run-rate once indirect effects are counted; that is a different and broader measure, and we keep it distinct from the US$32.6 billion direct figure used in the sum above. Illustrative only.

Australia

Estimated annual corporate tax loss: about US$22.1 billion (roughly A$33.4 billion). Unit cost: a registered nurse in Australia earns about A$82,500 on the market-rate average for 2025/26 (SEEK salary data, a commercial aggregator, labelled as illustrative not an official scale).

The sum: A$33.4 billion divided by A$82,500 is about 405,000 nurses, which exceeds Australia's employed RN workforce of roughly 360,000. There is a useful second anchor here: the Australian Taxation Office's own, more conservative large-corporate tax gap is A$3.7 billion for 2022 to 2023, and that alone would fund about 45,000 nurses. We show both so the illustration spans the full estimate range, from the cautious official gap to the broader NGO figure. Illustrative only.

Canada

Estimated annual corporate tax loss: about US$8.9 billion (roughly C$12.1 billion). Unit cost: a Canadian registered nurse earns about C$88,641 on a national salary average for 2025 (Glassdoor, a commercial aggregator; consistent with the Statistics Canada range of roughly C$60,000 to C$90,000, but not itself a StatCan benchmark).

The sum: C$12.1 billion divided by C$88,641 is about 136,500 nurses, roughly 40% of Canada's RN workforce of about 330,000 (Canadian Institute for Health Information). Illustrative only.

India

Estimated annual corporate tax loss: about US$21.4 billion (roughly ₹1.78 trillion). Unit cost: India has no single national nurse pay scale comparable to the NHS, so this one is explicitly a constructed proxy. The 7th Central Pay Commission places a government Nursing Officer (Pay Level 7) on a band that starts around ₹44,900 a month; we use an indicative annual figure of about ₹700,000 as a proxy, and flag it as such.

The sum: ₹1.78 trillion divided by ₹700,000 is about 2.54 million nurses' salaries, against an Indian registered-nursing workforce of roughly 3.4 million. The point is not the precise count, which the proxy makes soft. It is that the lost revenue is on the same order as a large fraction of the entire national nursing pay bill, in a country where nurse-to-patient ratios sit well below World Health Organization minimums. Illustrative only, and resting on a proxy salary, so treat the India number as the loosest of the five.

Why a nurse, and why "illustrative" matters

We picked one unit, the registered nurse, because it is concrete, it is something every one of these countries publishes a wage for, and it lets the five figures be read side by side. We could have used hospital beds, teachers or school places; the choice does not change the order of magnitude.

The label "illustrative" is doing real work, not hedging. The lost tax is an estimate, not a confirmed audit finding. It is not held anywhere; it is revenue that was never collected, so it could not simply be redirected to a payroll. And even if a government did collect it, it would face every competing demand on a budget at once. The honest claim is narrow and it holds: the scale of what corporate profit-shifting costs each country is, on its own official numbers, comparable to the cost of a very large part of that country's nursing workforce. That is a fact about size. It is not a manifesto.

A note on the loss estimates themselves

All six loss figures come from one source and one method, so they reconcile. The Tax Justice Network's State of Tax Justice 2024 uses OECD country-by-country reporting data for 2021 and a published profit-misalignment method. These are "direct" losses; IMF researchers estimate indirect spillover losses are larger again. The OECD's own, more cautious estimate of the global annual loss is US$100 to 240 billion, lower than the Tax Justice Network's; the two are measuring slightly different things and we report the range rather than a single magic number on our by the numbers page. The point of this page is scale, and the scale holds across the whole range.

Key facts

• UK: about US$16.9bn lost a year, about 415,000 NHS Band 5 nurses at £32,073 (2026/27). Illustrative.
• US: about US$32.6bn lost a year, about 348,500 nurses at the US$93,600 median wage (May 2024), over 10% of employed RNs. Illustrative.
• Australia: about US$22.1bn lost a year, about 405,000 nurses; the ATO's conservative large-corporate gap of A$3.7bn alone would fund about 45,000. Illustrative.
• Canada: about US$8.9bn lost a year, about 136,500 nurses, roughly 40% of the RN workforce. Illustrative.
• India: about US$21.4bn lost a year, about 2.54 million nurses on a labelled proxy salary; the loosest of the five. Illustrative.
• Every loss figure is from the Tax Justice Network State of Tax Justice 2024 ("corporate abuse, suffered"); UK and US salaries are official, the other three are aggregator or proxy and labelled.