In less than one year, on 18 September 2014, the Scottish electorate will vote on a question of historic significance – should Scotland remain in the UK, or should it become an independent country?
But what would an independent Scotland look like? We think that one important question that has not received nearly enough attention is debt. How will the existing UK government debt be divided between an independent Scotland and the continuing UK – assuming the remaining home nations constitute the continuing UK (Tierney 2013)?
In fact, we argue that the initial amount of debt an independent Scotland would inherit is tightly linked to the choice of currency, and that it is impossible to say which currency arrangement would be best without a clear idea of Scotland’s initial fiscal position (Armstrong and Ebell 2013).1
Debt and division
If Scotland becomes independent, the existing UK assets and liabilities would have to be divided. While this is a matter for negotiation, a reasonable starting point is that assets are divided on a geographical basis (Scotland would have most of the oil and gas reserves) and liabilities are divided on a per capita basis.2 Using the Office for Budget Responsibility’s estimate of the UK’s public sector debt on a Maastricht basis in 2016/17 (when independence would take effect) implies that Scotland’s debt-to-GDP ratio would be approximately 86%, while for the continuing UK it would be 102%.3 For an independent Scotland to assume its obligations from the UK, there needs to be an agreement on a schedule of payments from Scotland based on its issuance of bonds.
Given this amount of debt to be issued by a newly independent state, our focus is on the implications for the credibility of the chosen currency regime.4 When a negative shock, say to tax revenues, hits the economy, some credible adjustment plan is necessary to return to the long-run equilibrium debt path. When a country has its own currency, this adjustment can either be fiscal – cutting spending and/or increasing tax rates, or monetary – cutting interest rates and depreciating the currency. If both types of adjustment are costly, it may be optimal to use both the fiscal and the monetary levers.5
In a currency union, a government’s only means of adjustment is fiscal, as some Eurozone countries have experienced. This implies that a government is restricted to a second-best fiscal-only adjustment, which will be more costly than an unrestricted combination of fiscal and monetary policy. This might call the credibility of the currency union into question – is it really better to pursue the more painful fiscal-only adjustment and remain in the currency union? Or is it preferable to pay the costs of leaving the currency union, and regain the flexibility to use both monetary and fiscal policy? As Velasco (1996) shows, the larger the shock – and the larger the initial debt level – the more attractive it is to devalue and the less credible is the fixed exchange rate embedded in the currency union.
As a result, highly indebted countries are more vulnerable, at the margin, to shocks. These shocks can lead to a self-fulfilling spiral away from the equilibrium adjustment path. Citizens and investors can doubt the feasibility of the adjustment path, leading them to sell domestic assets – including government debt, resulting in higher borrowing costs. These higher borrowing costs would further worsen the government’s fiscal position. If the sell-off of domestic assets – the capital flight – is serious enough, then doubts about the currency union could become self-fulfilling.
Two important questions
Hence, two important quantitative questions for the sustainability of the currency union arise.
- First, how high would Scottish borrowing costs be, and how sensitive would they be to fiscal policy?
- Second, what sorts of surpluses would Scotland need to run in order to converge to a long-run sustainable debt path?
The latter question is important, as the tighter is fiscal policy on the equilibrium path, the harder it would be to impose even more austerity in response to an adverse shock.
To estimate an independent Scotland’s borrowing costs, we first run a regression of Eurozone yield spreads over Germany between 2000 and 2012. This allows us to, in principle, eliminate exchange rate risk as one source of variation in the yield spread. We choose this approach because the Scottish government has expressed a desire to be within a formal sterling currency union. However, whether the UK would wish to share a central bank with an independent Scotland remains to be seen and so the spreads may be a lower bound.
The second step is to use the projected fiscal data for Scotland in 2016/17 together with the regression coefficients to estimate the Scottish yield spread over Germany. The estimated yield spread is 2.12%, which implies that over this period Scottish borrowing costs would be 5.75%.6 We can relate these estimates to the UK’s borrowing costs in two ways. Comparing fitted values for the UK and Scotland yields a spread of 0.72%. Alternatively, we could simply take the difference between Scotland’s estimated borrowing costs and the UK’s ten-year bond yields over this period to obtain a spread of 1.65%. Therefore we believe that the spread of Scotland’s borrowing costs over the UK will be between 0.72% and 1.65%.
One of the main sources of the higher borrowing costs is the small-country effect – the premium that Scotland would pay simply by virtue of having a less liquid bond market. The remainder of the spread over the UK comes from factors that could be influenced by policy – the budget deficit and debt levels as a share of GDP, as well as the volatility of tax revenue. Borrowing costs turn out to be quite sensitive to government deficits – a country whose deficits are on average one percentage point higher than Germany’s must pay a premium of 50 basis points on its debt. The more that borrowing costs respond to changes in the fiscal position, the more likely is a destabilising debt spiral of the kind recently observed in some Eurozone countries.
Once we have an estimate of the borrowing costs that an independent Scotland is likely to face, we can assess how it would be able to converge towards a sustainable debt burden. We assume that a long-run sustainable debt burden on a Maastricht definition is a 60% debt-to-GDP ratio. To achieve this in ten years, an independent Scotland would be required to run a primary fiscal surplus of 3.1% of GDP each year. In other words, having paid the interest on its current debts, the Scottish government would have to take 3.1% of GDP more in tax revenues than it spends. This is in contrast to the 2.3% primary deficit that Scotland ran on average between 2000 and 2012, even after attributing a geographic share of oil and gas reserves.7 The fiscal tightening of 5.4% is due to the high initial debt-to-GDP ratio and the high level of debt-servicing costs. This scenario assumes an inflation rate and real economic growth rate of 2% in every year.
The bottom line
Whether a currency union with the UK would be credible comes down to the beliefs of investors. The lower Scotland’s initial debt and debt servicing burden, the smaller the fiscal tightening necessary to return to a sustainable debt burden, and the less painful any further spending cuts or tax rises would be to the electorate. The less painful is fiscal adjustment, the more likely are markets to believe it to be a credible adjustment mechanism. If Scotland were to find itself with high debt and interest rates, and in the throes of an already painful austerity drive, then a further adverse shock might lead markets to question its commitment to remaining in the currency union. The greater are these doubts, the greater the likelihood of capital flight and the economic damage that entails. An independent Scotland might find that the financial stability advantages to having its own currency outweigh any disadvantages due to trade and transactions costs.
The economic consequences for the continuing UK are also important. If the oil and gas reserves are divided on a geographic basis, then the UK's debt to GDP ratio would rise by 12 percentage points. While the UK would now have a large claim on an independent Scottish government (to assume its share of existing UK debt), cross border agreements are notoriously difficult to enforce. One option we discuss is a debt-for-oil swap which may be economically beneficial to both sides of the border. This would reduce Scotland’s initial debt, and therefore its borrowing costs. At the same time, transferring a volatile asset income stream might be difficult to manage.
Armstrong, A and M Ebell (2013), “Scotland’s Currency Options,” Centre for Macroeconomics Discussion Paper 2013-2.
Laeven, L and F Valencia (2012), “Systemic Banking Crisis Database: An Update”, IMF WP/12/163.
Scottish Government (2013), Government Expenditure and Revenue Scotland 2011–12.
Tierney, S (2013), “Legal Issues Surrounding the Referendum on Independence for Scotland”, European Constitutional Law Review 3, forthcoming.
1 HM Treasury and the Fiscal Commission Working Group have also taken up the question of Scotland’s currency options, but they focus primarily on trade in an Optimal Currency Area framework.
2 The geographic basis is determined by the median line which implies that around 90% of the reserves would go to Scotland. Scotland’s First Minister suggested that a division of UK debt on the basis of population would be fair during a TV interview on 11th January 2012.
3 Scotland’s share of debt on this basis would be £153bn. TheOffice for Budget Responsibility's central estimate of tax income from remaining oil and gas reserves is £56bn (of which an independent Scotland might receive £50bn).
4 Much of this section draws on Velasco (1996).
5 In general, it will be optimal to use both the fiscal and the monetary levers if the costs to both are increasing and convex in the size of the adjustment.
6 We obtain this estimate by adding the Scotland to Germany spread to the long-run average German ten-year bond yield of 3.63%.
7 Historic Government Expenditure and Revenues Scotland, The Scottish Government.