About the Author
Mitchell Palmer is an Economist at the Adam Smith Institute.
He previously worked as a special advisor to the Deputy Prime Minister of New Zealand, whom he advised on fiscal policy and microeconomics. He has also worked in economic consulting in New Zealand and at a think tank in Singapore.
Mitchell holds a first-class degree in History and Economics from the University of Oxford (New College). While at Oxford, he was made a Hayek Fellow of the Mont Pelerin Society. He also studied at Yale-NUS College in Singapore.
Executive Summary
The Adam Smith Institute’s new UK Fiscal Sustainability Model is a simple Excel model, resembling that used by the New Zealand Treasury to produce its four-yearly Statements on the Long-term Fiscal Position. It offers a tractable but sufficient way to forecast the UK’s public finances over the next 50 years.
The model reveals that the UK’s public finances remain in a perilous position, despite tax increases in the recent Budget. Without structural reform, we face a structural debt spiral in less than a decade, with debt reaching 330% of GDP by 2075.1 This is driven by rapid growth in health and pension spending, relatively slow economic growth, and high pre-existing debt.
Things could get even worse. For instance, if real interest rates are higher than forecast by the OBR, the result could be disastrous. If the post-2030 gilt yield averaged 4.8%, public debt would exceed 400% of GDP by 2075. A similar result emerges if productivity growth remains stuck at its post-2008 average.
From today, moderate fiscal adjustment could solve the problem, but only if action is taken quickly. Ceteris paribus, permanent spending cuts equal to 4% of GDP (or 9% of government spending) from 2026/27 could ensure that public debt remained below 120% until 2075.2 If the adjustment was delayed until 2037, cuts would need to be 15% larger. This would be difficult while protecting pension and health spending.
Rapid productivity growth – when coupled with policies to moderate health and pension spending – could also provide a solution and fund substantial tax cuts. If productivity growth returned to the level achieved during the Thatcher/Major governments, pensions were linked to CPI, and health spending growth was restrained to 1% above Baumol- and age-driven cost pressures, permanent tax cuts of 3.6% of GDP could be introduced from 2028/29, while ensuring net debt remained roughly stable.
In the near future, we recommend bold supply-side reforms that deliver growth at a minimal cost to the Exchequer. Spending cuts – especially to restrain future cost escalation – should also be used to reduce the immediate debt burden and cut the most damaging taxes. As growth returns, broader-based tax cuts should follow.
What Lies Ahead
The most important three numbers in public finance are the GDP growth rate, the interest rate on government debt, and the current debt-to-GDP ratio. If GDP growth is fast, interest rates are low, and the existing stock of debt is small, then running moderate deficits is acceptable. The actual burden of servicing the debt – i.e., the interest rate multiplied by the debt level, divided by the size of the economy – will be both manageable and shrinking, even with expenditure consistently above revenue. From year to year, the debt-to-GDP ratio will actually shrink.
Unfortunately, for Britain today, all three of those variables are currently pointing in the wrong direction. GDP growth is anaemic, interest rates are high, and the existing debt stock is almost 100% of GDP. Despite this, the government is still running – and forecast to run – a substantial cash deficit.
Even worse, forecasts from the ASI’s new UK Fiscal Sustainability Model suggest the situation will only deteriorate over time. Interest rates may moderate, but GDP growth is likely to weaken even further. This is because an ageing population necessarily means a smaller workforce. At the same time, it also means much higher government spending. Most obviously, this will be on pensions, but ageing will also drive up health and social care costs. The result will be ever-larger primary deficits which, when added to the natural growth of the public debt through interest, will mean an exploding public debt burden.
Table 1: Base Case Forecast
Assumptions are detailed in Table 2.
Table 2: Major Assumptions in the Status Quo Case
| Quantity | Assumption | Reasoning |
|---|---|---|
| Economic Variables | ||
| Demographic profile of the population | ONS principal projections | This projection represents HMG’s best guess as to the future size and age composition of the population. |
| Labour force participation and employment | Age-standardized rates remain as today | Today’s economy is relatively close to full employment, so it should represent the average level of labour force participation and employment for each age group. |
| Labour productivity growth | 1% | This is the OBR’s new forecast for trend productivity growth. |
| Interest rates on the stock of UK government debt | OBR forecasts until 2030, remaining constant at 3.9% thereafter | The OBR’s forecasts typically have the economy return to its equilibrium, including the equilibrium interest rate by the end of the period. |
| Inflation | 2% | We assume that on average over the period, the Bank of England hits its target. |
| Major Spending Groups | ||
| Pensions | Triple lock remains in place | This is the announced policy of all major parties. |
| Health | Inflation + population growth + ageing effect + Baumol effect (70% of real wage growth) + health residual (1.71%) | Health spending is assumed to grow in line with wage costs and ageing pressure, as well as an unspecified ‘health residual’, which represents the fact that health spending has grown much faster than these factors alone would predict. 1.71% is the average since 2003. |
| Exogenous fiscal policy | 0% | After 2030, fiscal policy is set by the automatic growth of the various categories (laid out in further detail in the technical appendix). No exogenous tax cuts, spending growth, or austerity is imposed on top. |
| Revenue | ||
| Total revenue | OBR estimates for tax-to-GDP hold until 2030 and thereafter remain constant. | This accounts for likely fiscal drag in the short-run but assumes that tax thresholds will eventually be adjusted to maintain the tax burden at a constant level. |
But why does this matter? According to Modern Monetary Theorists (MMT), sovereign governments do not face financial constraints. After all, His Majesty can always print pounds sterling. This is true, as far as it goes. But it does not go very far at all. If the economy is approaching full employment (as it usually is), then pensioners cannot be fed out of thin air. Any resources which are spent on pensions – or health or defence or anything else – are resources which are unavailable to the private sector. Even if the funds (or, in MMT terms, resources) are raised from abroad, this means that – in the fullness of time – the British private sector will have to export much more than it imports to repay these loans. Naturally, this will lower living standards.
If one accepts – as one should – that financial constraints on government are real constraints or (at least) are good proxies for the real constraints, then a borrowing explosion like the one forecast is terrible news. It inevitably means higher taxes in the future. It also means higher interest rates – for the government and for the private sector. These two factors together will further slow down growth and contribute to an even-worse-than-forecast doom loop. Only default, radical austerity, or hyperinflation could break the loop. The consequences – for British prestige, social order, and savers – would be unimaginably grim.
What If
In addition to our core ‘status quo’ forecast, we have also modelled a series of possible scenarios. These show that fiscal fatalism is the wrong answer to the disturbing forecasts presented above. There is much to be gained – but also much to be lost – from actively changing the UK’s economic settings.
The Bond Vigilantes Visit
In this scenario, we present the dangers of a higher long-run interest rate for UK public debt. In particular, we assume that all other variables and policies remain constant, but that the long-run interest rate after 2031 averages 4.9%, rather than 3.9%. This is in line with the heights reached by the 10-year gilt during the Global Financial Crisis and at points during the last three years. A real yield of 2.9% is not unreasonable if the bond markets lose confidence in the UK’s long-term fiscal settings.
In this scenario, government net borrowing exceeds 10% of GDP by 2050, which is a deeply unsustainable level. Such high interest rates would also depress private sector activity, meaning that this is an underestimate.
Growth Continues to Disappoint
In this scenario, we model the result of productivity growth of 0.6%. This is in line with very recent historical experience (i.e., since the Global Financial Crisis but excluding COVID), rather than the OBR’s forecast of 1% trend growth, which is in-line with the average since 1997.
As shown on the graph on the right, a lower productivity growth rate has a substantial impact on the level of real GDP per capita in the long run. This naturally impacts the tax take.
However, offsetting this is the fact that health and pension spending are closely tied to wages – and therefore labour productivity. Nonetheless, this offset is not sufficient to reverse all of the damage.
Austerity 2.0
This scenario shows how spending restraint on non-health and non-pension spending would struggle to deliver a sustainable fiscal scenario, especially if it is delayed. The same is true, to an even greater extent, of tax increases that would simultaneously depress economic activity.
To ensure that net public debt remained below 120% for the entire period, a permanent exogenous fiscal consolidation of 4% of GDP from the fiscal year ending 2027 would be required. If health, pension, and interest spending were protected, this would mean either an immediate – and sustained – 14% cut in all other government spending or a 10% increase in the tax take (which would have to be higher once dynamic effects were accounted for).
If fiscal consolidation were delayed until 2036, it would have to be 4.5% of GDP. This would represent 16% of unprotected expenditure or 12% of the tax take.
Growth Returns
This scenario models a goldilocks scenario, where productivity growth is delivered and fiscal responsibility prevents these gains from being consumed by ever-higher public spending. In particular, we assume that productivity growth returns to its CAGR under Thatcher and Major of 2.26%, residual health spending growth is restrained to 1%, and pensions are linked to CPI.
If left unchanged, this scenario would generate substantial cash surpluses from 2035 on. The Government might choose to use these expected surpluses to deliver substantial tax relief from 2028/29. They could afford to cut taxes by 4.2% of GDP or 11% of the tax take while keeping net debt permanently under 100% of GDP.
Policy Recommendations
Priority 1: Radical supply-side reforms to unlock productivity growth
The analysis above demonstrates that returning to economic growth is crucial to the ongoing solvency of the UK government. It should therefore be the first priority for the Chancellor.
The most important ways for the UK to enable growth are unlocking housebuilding and energy supply. We can also find growth dividends by removing smaller (but still material) impediments such as trade barriers, employment laws, and financial regulation. Almost all of these reforms would be fiscally neutral to fiscally positive for the Exchequer, making them achievable despite the straightened position of the books.
In line with these recommendations, a forthcoming paper from the ASI will propose a pathway to achieving a substantial ‘level shift’ in UK GDP. Many of the proposals in that paper would also increase the long-run growth rate of the economy by increasing dynamism.
Priority 2: Structural reforms to health and welfare provision, including pensions
The most important drivers of the UK’s forecast structural deficits are growth in spending on the NHS and state pension. Only substantial reforms are likely to be able to bend these curves.
Delinking the State Pension from earnings and abolishing the 2.5% floor will ensure that it does not grow inexorably relative to GDP. Instead, payments should be purely indexed to CPI or another estimate for the cost of living for pensioners (e.g., the Household Cost Index for retired people) to maintain its real value. Specific support for pensioners retiring without a mortgage-free or social home may also be required, but this could be targeted to highest-need groups.
Cost control in the NHS will require either political will to deny voters’ access to the most advanced modern healthcare or an abrogation of the principle of free-at-the-point-of-use. A model based largely on self-provision for routine healthcare with mandatory (possibly state-provided) catastrophic insurance for large expenses is likely to deliver the discipline required. Supply-side competition and innovation could also deliver productivity gains, but this will require a fundamental restructuring of the healthcare industry.
Priority 3: Tax cuts for the most damaging anti-growth taxes
There is very little ‘fiscal room’ to fund large immediate tax relief, despite the tax burden relative to GDP being at its highest level since the immediate post-war era. Nonetheless, spending cuts may be able to fund cuts to some particularly damaging smaller taxes. We offer three proposals for places to start:
Abolishing stamp duty – as the ASI proposed in a recent paper – would be a straightforward way for the government to unlock substantial housebuilding and other economic activity.3
Reforming the corporation tax base to allow firms to claim larger allowances for residential and commercial buildings under full expensing, in line with the existing allowances for plant and machinery investments, would further underwrite construction activity.4
Smoothing the £100,000 tax cliff at which many earners face exceptionally high effective marginal tax rates could encourage higher earners to keep working and have only small revenue consequences, as any revenue loss would be partially offset by reducing the use of tax-minimizing methods such as salary sacrifice and encouraging higher participation.
Later Budgets: Broad-based tax cuts
As the ‘growth dividend’ increases the size of the tax base and structural reforms begin to deliver substantial savings, there will be an increased ability for this Chancellor or any other to deliver tax relief.
Naturally, any such tax cuts should be initially focused on the most inefficient taxes on the books. In particular, taxes on savings and capital should be first in her sights, as should the assortment of over-complicating, small ‘nuisance taxes’ that abound across the UK tax system, including various ineffectual green duties that undermine energy investments.
Technical Appendix: Model Summary
The ASI UK Fiscal Strategy Model (UK-FSM) is a straightforward spreadsheet model of the UK’s public finances, forecasted until 2075. It aims to be as parsimonious and interpretable as possible. It resembles the New Zealand Treasury’s Long-Term Fiscal Model in its essential features.
It is constructed from three key building blocks:
- Treasury, OBR, and ONS out-turns from the 2024/25 financial
year, as the basic source of truth
- The ONS’s primary population forecasts for the age structure
of the economy
- A series of assumptions about how the economy and public finances will evolve over time. These are largely based on historical experience and announced government and party policies.
These three building blocks are combined to create estimates for the overall size of the economy, public sector revenue, and public sector expenditure. Naturally, these combine to give estimates for the fiscal aggregates, most importantly the current budget deficit, Public Sector Borrowing, and Public Sector Net Debt.
Economic Aggregates
The model estimates labour input in each year by assuming that the proportion of each age group (16-17, 18-24, 25-34, 35-49, 50-64, 65+) that are in paid work remains constant at its 2024 level over time. These proportions are then applied to the population forecasts to determine the size of the labour force and the employed population. These figures are then multiplied by the average number of hours worked per employee, which is assumed to remain constant, to calculate the total hours worked in the economy.
This labour input estimate is then multiplied by an estimate for labour productivity to determine the real size of the economy (which is then adjusted ex post to match measured GDP in 2023/24). In the base case, labour productivity is assumed to grow at 1% per year. This is the OBR’s forecast for trend productivity. This is very close to the average compound annual growth rate for labour productivity since 1996 was 1.05%. It is higher than post-GFC, pre-COVID average growth of 0.6%.
This real GDP forecast is made nominal by applying a GDP deflator. Implicitly, inflation measured through the GDP deflator is assumed to equal CPI inflation. CPI inflation is forecast to be at the Bank of England’s 2% target consistently over the period.
Public Sector Revenue
In the base case, the tax-to-GDP ratio is expected to evolve as predicted by the OBR until 2031. This accounts for likely fiscal drag. After 2031, the tax-to-GDP ratio is kept static at 38% of nominal GDP. Government revenue from other sources is forecast in the same manner.
Public Sector Expenditure
Total public sector spending is made up of three categories. RDEL in PSCE is made up of day-to-day spending by Government departments that is managed over the spending review cycle. AME in PSCE consists of non-departmental spending that is determined annually and is largely comprised of welfare benefits, interest payments, and devolved/local government spending. Finally, PSGI consists of capital investment by government.5
The nominal RDEL forecast is taken from the OBR’s forecast until 2031. This is to account for announced government spending plans. After 2031, it is composed of bottom-up estimates for each major category of spending: Health, education, defence, law and order. Spending outside of the major categories is assumed to remain constant as a share of GDP.
The nominal AME in PSCE forecast is taken from the OBR’s forecasts until 2031. After this it is built from bottom-up estimates for spending on interest, pensions, and working-age welfare. All other components are assumed to grow in line with GDP.
After 2024/25, Scottish spending is considered AME in line with the changes made by the Treasury. It is constructed using bottom-up estimates.
PSGI is assumed to follow the OBR’s forecast ratio with GDP until 2031 and thereafter to remain constant as a share of GDP (5%).
Healthcare
Real per-capita health care costs are forecast by assuming that the relative cost of treating each age group of person remains unchanged over time. That is, the average healthcare costs for 60-year-olds will continue to be 2.5x those of 30-year-olds. As the population ages, this means real per-capita health costs necessarily increase.
Further escalation comes from the Baumol effect – whereby health costs rise in step with productivity in the non-health sector, because most of the inputs are labour – and the so-called ‘health residual’. This accounts for the fact that (historically) real per-capita health spending grows faster than demographics and the Baumol effect would predict. This may be because healthcare is a normal good (i.e., demand increases more than proportionately after an income increase) and/or because quality increases and drives up prices.
In the base case, this health residual is assumed to be 1.71%. This is the average amount of unexplained spending growth per-year since 2003. It is higher than the average between 2010 and 2019 of 0.36%.
Pensions
In the base case, per-pensioner spending is expected to grow in line with the triple lock – which, given the other assumptions made, is equivalent to an earnings link. The State Pension Age, which (combined with the ONS’s population forecasts) determines the number of pensioners, is assumed to follow its announced path (i.e., increasing to 67 by 2028 and further increasing by one year around 2045).
Working-Age Welfare
Welfare benefits per recipient are expected to grow in line with CPI. The number of recipients is determined by the size of the proxy population (i.e., under 65s that are economically inactive).
Education
The participation rates and relative rates of spending between the various stages of education are expected to remain constant at their 2024 levels. Thus, the change in overall education spending is determined by wage growth and the age structure of the population.
Defence
Defence spending is assumed to grow linearly to meet the Government’s objective of 3.5% of GDP spent on defence (under the NATO definition) by 2035 and thereafter to hold at that level, relative to GDP.
Law and Order
Spending on the Home Office (which is dominated by police spending) and Ministry of Justice are expected to grow in line with wages and the predicted level of crime, based on the age structure of the population and rates of crime by age.
Interest Expense
The annual interest expense is based on the forecast closing Central Government Net Debt position for the previous year (to avoid circularity). Until 2030, the interest rate is adjusted based on the OBR’s forecast for the future path of the interest expense, as a share of the outstanding stock. After 2031, it is assumed to remain flat.
Scotland
The Scottish Government’s current expenditure on health, devolved welfare, and education is forecast to grow at the same rate as UK expenses. The remainder of Scottish spending is assumed to remain flat as a share of UK GDP.
Fiscal Aggregates
These forecasts for public sector revenue and expenditure feed directly into an estimate for the current budget deficit and public sector net borrowing. The Public Sector Net Cash Requirement also includes an allowance for financial transactions, for which the nominal forecast of the OBR is adopted until 2031 and zero is assumed thereafter. This – with further adjustments based on the OBR’s nominal forecasts – leads to the annual change in the stock of government debt.
Forecasting Performance
Despite the parsimony of the ASI UK-FSM, it matches well with the OBR’s Fiscal Risks and Sustainability Report. By the early 2070s, both models predict that pension spending will be around 7.5% of GDP and that net debt will exceed 270% of GDP. The advantage of the ASI UK-FSM is that it is much easier for external analysts to conduct scenario analysis and to understand how the various fiscal aggregates interact with each other.
In this paper, a ‘structural debt spiral’ means that the debt-to-GDP ratio grows at an accelerating rate.↩︎
This target is chosen because it is roughly the highest level currently sustained by an advanced economy (the US) without financial repression (ruling out Japan) or special circumstances (i.e., Singapore’s unusual pension schemes).↩︎
Mitchell Palmer, Stamped Out: The Economics of Abolishing Stamp Duty on Primary Residences, Adam Smith Institute (2025). https://www.adamsmith.org/research/stamped-out-the-economics-of-abolishing-stamp-duty-on-primary-residences↩︎
See Adam Smith Institute and PricedOut, Boosting Brownfield: Full Expensing for Brownfield Development (2023), and Mitchell Palmer, Britain’s Tax System is Blocking Builders, Adam Smith Institute (2025). https://www.adamsmith.org/research/boosting-brownfield amd https://www.adamsmith.org/research/britains-tax-system-is-blocking-builders.↩︎
RDEL in PSCE = Resource Departmental Expenditure Limit in Public Sector Current Expenditure; AME in PSCE = Annually Managed Expenditure in Public Sector Current Expenditure; PSGI = Public Sector Gross Investment.↩︎