In contrast to the UK government, where borrowing is constrained only by how much and at what price the financial markets are willing to lend and any formal borrowing limits are self-imposed, the borrowing powers of the Scottish and Welsh Governments are limited by UK legislation. Similarly, there are also limits on how much money they can place in reserves and draw down in later years.
Capital borrowing powers
Both governments are able to borrow to top up their budgets for capital spending – including investment in buildings, equipment, infrastructure and research.
The Scottish Government can, as of 2025–26, borrow up to £472 million a year, up to a total cap of outstanding debt of £3,145 million, for capital purposes. These limits are updated each year in line with forecast inflation, and as of the end of 2025–26 the Scottish Government is forecast to have outstanding capital debt of £2,534 million – 81% of its limit.
The Welsh Government can, as of 2025–26, borrow up to £150 million a year, up to a total cap on outstanding debt of £1,000 million, for capital purposes. As of the end of 2025–26, £550 million – or 55% – of this will have been utilised. These limits for Wales will be increased by 10% in 2026–27 and in line with forecast inflation from 2027–28 onwards. This would still be lower as a share of its capital budget than in the case of Scotland.
Both the Scottish and Welsh Governments have, so far, borrowed via the UK government – rather than directly from the financial markets. But they are both able to issue their own bonds if they wish, and the Scottish Government plans to start borrowing via its own bonds from 2026–27.
Borrowing to cover forecast errors
Both governments are able to borrow to address forecast errors for devolved taxes, benefits and associated block grant adjustments which have negative financial impacts on their funding (see our explainer on the block grant adjustments). For example, if the revenues actually collected from a tax are lower than forecast when budgets are set, or the associated block grant adjustment higher, the devolved governments can borrow to address the shortfall in funding this creates.
The Scottish Government can, as of 2025–26, borrow up to £629 million in any one year, subject to a total cap on outstanding debt of £1,834 million. These limits are updated each year in line with forecast inflation, and as of the end of 2025–26 the Scottish Government is forecast to have outstanding forecast error debt of £219 million – or 12% of its limit.
The Welsh Government can, as of 2025–26, borrow up to £200 million a year, subject to a total cap on outstanding debt of £600 million. The Welsh Government has yet to borrow to address forecast errors. These limits are not increased in line with inflation, but are higher as a share of devolved tax revenue and benefit spending than in the case of Scotland.
Borrowing to address forecast errors must be undertaken via the UK government.
Financial reserves
Both governments are able to pay money into reserves for use in later years, with separate reserves for capital (investment) funding and resource (day-to-day) funding. The Scottish Government is able to hold and carry forward up to £734 million in reserves as of 2025–26, with this limit increased in line with forecast inflation each year. There are no limits on its ability to draw down reserves. The Scottish Government started 2025–26 with £571 million in reserves and, after planned drawdowns of £421 million, is currently forecast to have £150 million of (resource) reserves remaining for use in future years.
The Welsh Government is able to hold and carry forward up to £350 million as of 2025–26. At the start of 2025–26, the reserve limit was used in full, and with £40 million planned to be used in-year, this would imply a figure of £310 million in reserves at year-end. The £350 million limit will be increased by 10% in 2026–27 and by forecast inflation thereafter. As with the Scottish Government, there are no longer any limits on how much the Welsh Government can draw down from reserves in any given year.
Why not enable the Scottish and Welsh Parliaments to set their own borrowing limits and rules?
The Scottish Government has called for powers to set its own borrowing limits under a so-called prudential regime: it would determine how much it thinks it could prudently service given its various funding sources.
Given the relatively small size of the devolved governments’ budgets compared with the size of the UK government’s overall budget (and the UK economy), such borrowing would unlikely have much effect on overall fiscal sustainability. It could, in principle, mean the difference between a UK fiscal target being met or not. But in practice this eventuality seems unlikely given the scale, uncertainty and volatility of UK government borrowing. Even if the eventuality did arise, a marginal difference in whether a deficit or debt limit is breached has very little economic significance (although its political significance could be higher). And the UK government could define its fiscal rules to exclude borrowing by the devolved governments if it wished, without diminishing their credibility.
The more significant concern is one of fairness. The argument here is that all borrowing undertaken by the UK government implicitly benefits the devolved nations – it helps fund the UK government’s UK-wide responsibilities (such as defence or the state pension) and/or England-only responsibilities (such as health and education) and so generate funding for the devolved governments via the Barnett formula. It would therefore arguably be unfair to England if the devolved governments were able to borrow large amounts of money ‘on top of’ UK government borrowing (particularly because relatively high levels of public spending mean the devolved nations benefit disproportionately from existing UK government borrowing).
One might argue that as long as the costs of devolved government borrowing are borne uniquely by the devolved governments and devolved taxpayers in future, then the equity concerns do not provide a strong justification against devolved government borrowing. On the other hand, if devolved governments did choose to borrow additional funds – despite the associated debt servicing costs in future – this shows that that borrowing is valued, and the fact that the option to capitalise on that is only available to residents of the devolved nations and not England could be seen as unfair. While voters in England could elect a government that was willing to borrow more, such borrowing would either be for UK-wide spending or generate additional funding for Scotland, Wales and Northern Ireland via the Barnett formula, meaning it would not be possible for them to benefit from additional borrowing on top of UK government borrowing without more fundamental reform of the UK’s fiscal architecture.
For this reason, Bell, Eiser and Phillips (2021) conclude that providing the devolved governments with unrestricted borrowing powers would be undesirable at the present time. Instead, they recommend providing the devolved governments with modest borrowing powers, equivalent to about 1–2% of their annual funding for resource purposes other than addressing forecast errors, subject to total caps as with existing borrowing powers. This would allow the devolved governments to borrow to cover: temporary reductions in revenues that are forecast in advance (perhaps due to a particular slowdown in the Scottish economy); temporary increases in benefits spending that are forecast in advance (for similar reasons); and expected or unexpected increases in other resource spending requirements (such as higher-than-expected public pay increases). These are things that the UK government can and does borrow to cover (although in doing so, generating funding for the devolved governments too). But with modest limits, any fairness concerns vis-à-vis England would be relatively minor.










