I’ve read some utter tosh on the state of the UK public finances in the last few days. Here is an attempt to correct some of the biggest misunderstandings.
Most importantly, government debt does not have to be ‘repaid’, only serviced. As long as the government can meet the interest payments (I’ll come back to the risks here later), maturing debt can simply be rolled over.
Indeed, this is what usually happens. The last time the UK government ran a budget surplus was in 2000-01, since when public debt has increased by more than £1,700 billion – and most of that long before the pandemic.
Of course, there are limits to how much debt can be comfortably financed in this way. The key to fiscal sustainability is to avoid a vicious spiral where higher interest payments require more borrowing, adding to debt and thus driving up interest payments even further. A higher stock of debt obviously increases this risk.
For this reason, it is usually considered desirable to stabilise the ratio of debt to GDP (national income), or reduce it, and big jumps are a concern. Nonetheless, there is no particular threshold for the debt/GDP ratio where debt sustainability becomes a problem.
The Office for Budget Responsibility’s latest central scenario (which actually now looks too pessimistic) assumes that debt will jump to 105% of GDP this year and then stabilise close to this level. But this ratio has been much higher in the past, particularly after the two World Wars, when it was then reduced by stronger economic growth and inflation. Even today, UK government debt is still lower than in many other countries, including the US, France, Italy, and Japan
Incidentally, this measure of net debt includes loans made by the Bank of England to the private sector under its ‘Term Funding Scheme’, which are not netted off as assets because they are illiquid. If these are excluded, UK public debt is actually still below 100% of GDP.
What’s more, the cost of financing this debt has actually fallen: the increase in the stock of debt has been more than offset by a fall in the average interest rate on this debt.
OK, what about the risk that interest rates take off? A significant rise in the interest rates on a high stock of debt could be a game-changer and it is right to worry about this. But there are three reasons not to panic.
First, the average remaining life of UK government borrowing the gilt market is around 13 years. This means that even if bond yields do rise soon, it will be a long time before this is reflected in the overall cost of borrowing.
This is not just ‘kicking the can down the road’ for future generations. In ten to fifteen years the economy should be larger and the burden of a given stock of debt relative to income will be smaller.
Second, the Treasury now gets a large chunk of the interest back anyway from the Bank of England which refunds profits from its gilt holdings, minus the interest that the Bank itself pays on the reserves created to purchase them.
Admittedly, the way in which the Bank of England has financed its purchases of government bonds under Quantitative Easing (QE) has introduced a new risk. In a nutshell, QE involves the purchase of assets, or the making of loans, using newly-created money. This money takes the form of central bank reserves, which are electronic IOUs from the central bank to commercial banks.
Typically, the assets bought are government bonds. This is because the markets for these bonds are relatively large and liquid, and their yields set a benchmark for asset prices more generally. Crucially, central banks do not buy these bonds directly from the government, nor is the money being ‘given away’ to ‘fund the deficit’
Instead, central banks buy the bonds from private financial institutions by crediting their accounts at the central bank itself. These accounts are interest-bearing deposits (reserves). In the UK, the Bank of England pays Bank Rate (currently just 0.1%) on these reserves.
As the Bank of England is itself part of the public sector, this is effectively an asset swap which shortens the maturity of public sector borrowing. Rather than the government having to pay the usual long-term rates of interest in the bond market, the cost of serving the debt is determined by the short-term rate set by the central bank.
At some point the Bank of England may want to sell its gilts back onto the open market. This would be a double threat, both because it could drive up bond yields and because the Treasury will then lose the repaid profits.
Even if the Bank of England keeps the gilts on its books (as seems more likely), there is the risk that the interest rate that it pays on the reserves (the Bank Rate) increases from its current very low levels.
Mainly as a result of swapping gilts for floating rate central bank reserves, the sensitivity of debt interest to a one percentage point rise in short-term interest rates has doubled from £6 billion (0.2% of GDP) to £12 billion (0.5% of GDP).
But this is where the third point comes in. If interest rates do rise, we also need to know why they are rising. Some worry that investors might lose confidence and demand a much larger risk premium, but there is no sign of this happening (either here or in countries with even higher debts).
It is far more likely to be because the economy is recovering more quickly than expected, or because inflation is higher, or both. In either case, nominal GDP would also be higher, thus reducing the budget deficit (excluding interest payments) and the burden of debt as a share of GDP. This is likely to more than offset any increase in debt interest payments.
In fact, I’d go further and suggest that the return of interest rates towards more normal levels would actually be a nice problem to have!
Over the long run, it’s reasonable to assume that nominal GDP grows by at least 3-4% (e.g. 1-2% real growth plus 2% inflation). This means that as long as the annual budget deficit is kept below 3-4% of GDP, debt should shrink as a % of GDP.
It is also helpful if the average interest rate paid on debt remains below 3-4%. If the government then balances the budget excluding interest payments (‘the primary balance’) debt/GDP will also fall.
Interest rates are of course now well below these levels – 10 year gilt yields are just 0.3% and the short-term interest rate set by the Bank of England is just 0.1%. The chances of either of these rates rising much above 2-3% in the foreseeable future look pretty small.
To be clear, even if the government doesn’t face the same financial constraints as a household, high public spending and borrowing has other costs, including the risk of the misallocation of resources – and inflation. But talk of ‘maxing out the credit card’ is claptrap.
2 thoughts on “How I Learned to Stop Worrying and Love the Debt”
This post is very helpful — I am not an economist (I work in IT) but I try to understand macroeconomic issues as far as I can.
1) The BoE asset purchase facility (QE?) totals £895 bn. If the government bonds purchased under the APF are from banks, is that a straight swap of bonds for bank reserves? As I understand it, bank reserves only circulate between banks and don’t appear in the wider economy. So is there a difference in macroeconomic effect if the BoE APF buys from non-banks, say pension funds, rather than banks.
2) Is not almost all money borrowed? I have read the following:
where it states right at the beginning: “the majority of money in the modern economy is created by commercial banks making loans”.
If so what is the different between a government borrowing and the private sector borrowing?
3) I read often something like: “interest rates are low now but what if interest rates rise?” I see from your post above that a one percentage point rise in short-term interest rates would lead to an additional £12bn in interest payments from the public accounts. But why is this necessarily a bad thing — interest payments are not money thrown away? Surely, interest payments are made to bond holders, say pension funds, so provide income for the private sector. Aren’t interest payments just a part of the circulation of money in an economy?