During a recent discussion on this blog on how to account for the UK's so-called national debt, regular commentator and bind expert Clive Parry suggested he would offer some comments on this issue. His thoughts follow:
What are we trying to achieve when we account for the national debt? Suppose, as a comparison, I took out a £100,000, 25-year, mortgage 5 years ago – what questions might I reasonably want answers to?
I would suggest that they are “What are my repayments now?” and “How much do I owe now?”. I might also want to know the answers to these questions for the recent past, the near future and the further out future.
The recent past is easy. My payments will comprise interest and capital repayment. For example, £7,000 in the last year could be split as, perhaps, £4,000 in interest and £3,000 capital repayment1. This will reduce the “amount owed” from (say) £89,000 to 86,000.
The near future is easy, too. I know the interest rate on the mortgage and can project what interest and capital will be paid next year.
Further out things are much more uncertain. Interest rates might change when I “refix” my rate. I might remortgage and take on more debt to build an extension. I might also switch to an “interest only” mortgage because money is tight. In that case, we can make some projections and estimates but they will only be educated guesses. The point is, no answer is certain.
So, how does this “thought experiment” inform our thinking on the National Debt?
First, I will confine the discussion to debt accounting but I mention two other things in passing:
- The debt is there to finance a useful asset. We need to consider both sides of the balance sheet – particularly as we consider public ownership of utilities. It is wholly inappropriate to consider the liability in isolation; we incurred it for a reason.
- To whom do I owe the money? This matters because if, for example, the money is owed to the Bank of England Asset Purchase Facility it really is not owed to anyone outside government, in which case it is not debt. Being able to recognise when a liability really exists is important.
Second, I will assume the National debt exclusively comprises gilts. I will mention short term Treasury bills (“T-bills”) along the way but other components of the National Debt are small and/or easily dealt with, and so I will ignore them here.
In that case, here goes…
In a simple world where (say) I borrow £100 at 2% interest for 5 years the accounting is very straightforward. Interest is £2 a year and my debt is £100. The complications come because few debts are that simple, especially when it comes to the national debt.
First, I could have borrowed £90 with no interest and agreed to repay £100 in 5 years time, which is a (roughly) equivalent arrangement.
Or, second, with interest rates now much higher than 2% the market value of this loan (assuming it had one) will no longer be £100, but may is much lower. If it were (say) £90 - is the size of the debt £100 (what I will owe on maturity) or £90 the cost of buying the debt in the market today and retiring it?
Third, we could ask “what must I pay back when the bond matures?” for an individual bond but this has no meaning for the National Debt which is a portfolio of gilts with differing maturity dates that are continually refinanced by the issuance of new bonds as the old ones are redeemed, and everyone knows that this is going to happen and that the refinancing will definitely take place.
Fourth, this could be an inflation linked debt/bond with an unknown future redemption amount (in cash terms) 2.
I will deal with these points in turn.
By and large, conventional gilts are issued at prices close to par – ie. the amount repayable is roughly equal to the amount borrowed and the coupon (interest rate) is fixed for the life of the bond at the prevailing “going rate”. We can then take these sums as known and certain with overall small capital appreciation/depreciation and spread those gains/losses out over the life of the bond and consider them as “interest” 3. This works for conventional gilts because all these cash flows and rates completely known at the point of issuance and do not change at all over the life of the bond. This is well trodden accounting ground, even if a little complicated.
So, what is the “size” of the national debt? Well, adding up the face value of all bonds outstanding tells us what it would cost to repay each bond as/when it matures. But it does not quite answer “what would it cost to pay off the debt today?”. The answer to this is the market value of all debt outstanding. Gilts are highly liquid and the data is already collected and available. Overall, this is the best estimate that there is; it does not depend on what price the gilts (index linked or conventional) were originally issued at or at what price bonds will mature at (par for all conventional bonds but uncertain for inflation linked bonds), merely the price today.
Whilst using Market Value solves almost all the issues related to the size of the National Debt, there is still a question about how we account for “interest” on Inflation linked debt.
The coupon on an Index Linked (I/L) gilt is typically much lower than that of a conventional gilt because the face value is uplifted by inflation and interest paid is then paid on this uplifted amount. Clearly, this not a “free lunch” and we have to account for it, but how? What will the redemption amount be? The answer is, we just don't know.
The ONS approach is to call any uplift due to inflation “interest”. It is an attempt to mirror the approach with conventional gilts issued with low coupons at deep discounts where the discount, spread over the life of the bond is considered as interest. However, this approach fails on two counts. First, the uplift is included as interest in the year it occurs even though the cashflows resulting from it are spread over many years. Second, the approach only really makes sense where the future cashflows are known, which they are most definitely not with I/L gilts.
I am going to suggest a far simpler approach across the board.
- I will define “Market Value” as the Clean Price of each gilt multiplied by the face amount outstanding. 4 The clean price excludes any accrued interest owing on the bond because some time has elapsed since a payment was last made.
- I define “interest” as the coupon rate due on it i.e. the cash paid on each bond in the period plus the change in accrued interest over that period. 5
All that is then required are a few judicious footnotes to the “Market Value” number to explain it.
The change in market value of the gilt portfolio will be the result of:
- Net issuance which (give or take) represents the Budget deficit.
- The inflation uplift on existing I/L gilts.
- The change in value assuming unchanged yields (ie the “pull to par” on bonds at large discounts or premia).
- The change in value due to changes in yields.
1 and 2 are easily available; 3 and 4 (combined) are just the residual that is left… and the split between 3 and 4 is easily modelled.
I stress, total market value is the clean price of each gilt multiplied by the face value outstanding – which follows accounting ‘mark to market rules' and is, therefore both a known concept and a proven current figure, which is of importance when decision making. My points 1 to 4 explain changes in this market value, with that difference not being interest but a movement in the capital value of the gilts in issue – a concept quite explicitly recognized as a movement in equity in accounting, which it should also be here so that no confusion arises. The interest cost is, effectively, the cash flow cost of payments actually due in a period.
Finally, there are Treasury Bills. They are “discount instruments” and carry no interest and are short-term (issued for maturities from 1 to 364 days). Over a year lots of bills will have been issued and redeemed so I think it best to take the “T-bill Market” in aggregate as a single beast. For any point in time we can easily know the outstanding amount; for any period we can calculate the difference between issue price and par of each bill issued which (unlike my suggested treatment for gilts) should be considered “interest” 6.
I think that all the information we want is captured and it answers the questions posed by our mortgage borrower at the outset. In particular, it gives a much better picture by treating Inflation uplift on I/L gilts more sensibly.
This means that this year's accounts will match the cash cost during the last year. This number, with some minor adjustments (for budget deficit and changing interest rates) will be a good predictor for the cash cost of debt servicing in the next year. Using forecasts for the budget deficit, inflation and interest rates we can make reasonable projections for things further out.
Finally, I would add that all the data required to do this are available on the DMO website so I am sure that the bright sparks at ONS could whip up a spreadsheet that did all this and publish it tandem with their current “Internationally recognised treatment” of debt.
Footnotes
1Even if the interest rate is constant (in this case at 5%), this ratio will change; as capital is repaid interest is payable on a smaller principal that in turn. In year 1 it will be £5,000 interest, £2,000 capital; in year 25 it will be almost £7,000 capital and no interest.
2Equally, it is true that conventional gilts have an unknown future redemption amount in real terms. Given that tax revenue and government spending is largely inflation linked there are arguments for both approaches.
3 This not always true; in order to promote liquidity in each gilt issue the DMO will, rather than create a “new” gilt at par, issue a further tranche of an existing gilt at the prevailing market price… which might be quite far from par. Investors will buy a bond with a coupon below the “going rate” if the capital appreciation to par (when the bond matures) is sufficient.
4 Usually, Market Value would use Dirty Price but in this instance Clean is is needed because we are trying to separate out a figure for “interest”.
5 Accrued interest used in its technical, bond trading sense.
6 Treatment of bills is complicated and involves discussions about the relationship with CBRAs and, indeed, the nature of money and T-bills themselves. The amounts outstanding are also volatile; in the last 5 years the balance outstanding ranges between £35bn and £110bn (currently £85bn).
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
I have to say that in conversations with ignorant and sceptical folk about such matters, its when you get down to the ‘Who does the Government owe the money to?’ you then get total silence.
And then you tell them that the BoE is nationalised and point out the Audit and Credit Act plus the Act that set up the BoE. Further silence follows.
And to cap off the discussion, tell them about the Central Bank Reserve Account and ask them what they think to Rishi adding interest to it to make what is a state line of credit to risky bankers look like money that we owe to them – basically a clear case of false accounting as I see it – and then watch their reaction to that of course by comparing that line of credit to austerity.
Thanks
I take that approach with people I know. The silence is indeed stunning. Especially when I give them the rough breakdown which Richard supplied a few years ago in a video blog [Nat savings, underwriting pension funds, trading account for import/export etc].
It gets even better when I remind them of what assets the government owns in terms of buildings and equipment, with a reminder that their own mortgage is often lower than their asset, then ask if they are really ‘in debt’.
Sometimes I even get an ‘oh’.
On that note, Richard, can you give an up to date breakdown of the national [non]debt? It’s really helpful.
Sorry- but what do you mean by the national non debt? I did the national debt on 23/12 – just scroll back
By ‘[non]debt’ I’m referring to part of it being some bad accounting which you have referred to in this blog and that most of it is investment rather than debt. I hope I’ve understood you correctly.
If I’ve misundertood, I apologise.
See my piece on 23/12
They tried to make us believe that Government spending all came from taxes.
Thatcher claimed that there is no government money!
① In 1853 tax revenue was about £70 million
② By 2022 it had increased to about £700 billion
Where does the extra money keep coming from?
Bank loans must be paid back, so it must be the government.
Thank you Clive
This is a Holmesian “three pipe” piece of thinking. It demands some concentrated reading.
Indeed
Many thanks to Clive for putting this together. Really well explained and suitably supported by facts.
So, for me two questions arise:
First, why can’t the ONS follow such a straightforward approach?
Second, is it believable that all those ‘experts’ at the Treasury and BoE don’t know this and thus understand what the actual national debt really is?
The answers are: Of course the ONS could do this. And, of course the Treasury and BoE (and many others) know that how Clive explains the situation here is reality.
So why does nothing change?
The answer is simple.
The debt myth must be maintained at all costs because of the central and fundamental role it plays in a long running and powerful narrative that allows government to claim that its hands are bound tightly when it comes to government spending.
Maintaining this myth makes it far easier for governments to reject policy demands without ever having to enter into an open and honest debate about what policies should get preference and why. In other words, come clean on ‘who gets what, when and how’ (as the old policy adage used to go).
Of course, the debt myth is only one element of this (now long standing) approach to policy making. A tame and reliable mainstream media is another, as are political parties mainly consisting of ‘cowardly’ politicians (i.e. those who are in politics primarily for themselves), and thus realise that going against convention is not a good career move. This also applies to many of those who work in government related organisations, such as the BoE, higher civil service, ONS, etc).
Additionally, and crucially, the debt myth is now so central to contemporary government that opposition parties – and particularly those who expect to be in government at some point (i.e. Starmer’s Labour Party) – will not challenge it as they know that, in time, it will serve their purpose too (plus, while in opposition anyone who points out that the myth is in fact, nonsense, will suffer a major attack by the legion of supporters and promoters of the myth (e.g. from the MSM, etc).
And so the pantomime continues, for the benefit of a few and the cost to the many.
That’s been the nature of policy making both here and in the US (and increasingly elsewhere) since Reagan and Thatcher and the rise of neo-liberalism, and the subsequent demise of the post war ‘courageous state’.
Thanks Ivan
“Second, is it believable that all those ‘experts’ at the Treasury and BoE don’t know this ”
Yes, this is how a paradigm works. A typical economics degree teaches neoliberal neoclassical economics. This is why the group “Rethinking Economics” came about, they state:
“The problem we face is the total dominance of a single and outdated form of economics at universities, which promotes the marketisation of society, leading to increased inequality, injustice and significant harm to the natural world. We need students of economics to leave university with a tool-kit of skills they can use to help us face these challenges effectively.” https://www.rethinkeconomics.org/about/
Harvard professor Dani Rodrik writes:
“much of what goes on in an introductory course in economics is a paean to markets. It gives little sense of the diversity of conclusions in economics, to which the student is unlikely to be exposed unless she goes on to take many more economics courses …
Instead of presenting a taste of the full panoply of perspectives that their discipline offers, they focus on benchmark models that stress one set of conclusions.” — Economics Rules: The Rights and Wrongs of the Dismal Science (2015) https://amzn.eu/d/5ACT1dp
Agreed
Ian, I don’t doubt the power of those paradigms. Indeed, I experienced them first hand as an academic, and particularly during my time teaching public administration/government and politics to undergrads. I was also lucky enough to have been at university (as a mature student) when political economy was still being taught on politics and economics degrees, and also – sadly – when, a few years later, it was removed from the politics and economics degrees at the same university, never to return (as happened more widely across academia in the UK).
But I can tell you that’s only part of the explanation/story, albeit probably the most significant. I know from post-grad students I’ve taught from the civil service – including, importantly, some at the Treasury – that there are plenty who understand only too well the myths and falsehoods that are routinely used to maintain the neo-liberal grip on government policy making, and thus on politics. Indeed, I’ve read many assignment over the years that covered such ground. But as I mentioned in my initial comment, the same people also know that if they want a successful career in government they need to go with the dominant narrative – as indeed do a lot of the academics who teach the reality free tosh included in many an economics degree since the early 2000s.
So true, Ivan.
Ivan
Part of my OU degree back in the 70s was on Soviet Government and politics.
I spent some of the autumn reading modern accounts of the Cold War. Your words
“thus realise that going against convention is not a good career move. This also applies to many of those who work in government related organisations, such as the BoE, higher civil service, ONS, etc).” remind me of the Soviet Union.
I am not saying we are as controlled as they were ‘back in the USSR’, but there are powerful similarities.
Quite depressing but I take some hope in that enough people knew what they were supporting was not reality and there came a tipping point when the whole thing fell apart. The crowd booing Nicolae Ceausescu and his realisation he had lost it, is the most striking image of it.
Counselling people from the local govt sector I came to the conclusion that the pathology was as much in the system as individuals. Also that most people didn’t subscribe to the ideology but felt they had to do so to survive. If they realised they were the majority, my fantasy they all said “no” at the same time and sanity emerged again.
Good to hear you’re an alumnus of the OU, where I spent many years as an academic, Ian. And I share your wish that at some point people will wake up to what’s going on in politics around them. But no doubt you’re aware of the concept of hegemony as a form power and control, and that is, of course what we have in the UK – as opposed to the more obvious forms of control practiced in what was the USSR, and now back with a bang in Putin’s Russia and Lukashenco’s Belarus, and many parts of the former Soviet Union (plus China, North Korea and so on). It’s so much more difficult under a hegemonic system to get people to realise that they are being controlled and the degree to which that’s the case. Indeed, I often smile wryly to myself when I hear people on the right banging on about ‘wokeness’ and how it’s impacting society when its actually the right that’s the primary – perhaps sole – beneficiary of the hegemonic system that applies (to a greater or lesser extent) in the UK. Of course, Putin and other dictators would kill (and frequently do) to have such a system, whereas most politicians on the right in this country fail to see the built in (and now age old) political advantage they have.
Agreed
If gilts that mature are mostly then refinanced with issuing new gilts, then can’t we just effectively say that the National Debt contribution of such perpetually reissued gilts, is just the interest on the gilts – not in an accounting sense of course, but in a political point making sense?
A very good point….
Gilts were originally issued, typically (and for centuries) as “perpetuities” undated (no redemption date guaranteed at issue); some were issued as annuities. The last perpetuity, I understand was only finally terminated in the UK, around 2015. Since most Gilts are rolled-over there is an inherent quasi-perpetuity in the character of the continuing debt.
Notice that economists love to address sovereign debt in the form of a ratio or percentage (GDP/Debt); and not the debt quantum (economists are uncomfortable with the problem of money). This leads to a distortion of their perspective. The ratio may rise or fall; but almost invariably it is wrong to translate that idea into the thought that in these “falls”, the debt itself is falling. In fact, the effect over time of economic growth or of inflation, or both generally means that while the GDP/Debt ratio may fall; it is very unlikely that the debt itself will fall. Almost invariably (save in periods of real deflation or Depression), the debt relentlessly rises. Sovereign debt in most States is, I submit ‘de facto’ the operation of an aggregate, rising, quasi-perpetuity.
It is, as you say, an aggregate, rising, quasi-perpetuity.
So, how do we get those perrenial debts issued again, because I think we need them.
I think Clive Parry will disagree.
No problem with the idea that the National Debt is “quasi-perpetual” – ie. all redemptions are refinanced with new issuance with debt levels continually rising in cash terms.
No problem with perpetual bonds, they could certainly be part of the debt mix. But, as you have observed, the “National Debt” is “savers’ safe savings” and whilst it may suit the State to issue perpetuals it may not be what savers want.
In general, I think the State should issue debt in the form most attractive to investors for two reasons; first, investors will “pay up” for a structure they like (reducing overall interest payments) and second, the State is much more able to manage the risk than savers.
So, perpetuals, conventionals and index linked bonds all have their place.
Noted
Thanks
Well, I confess I like the way this discussion is moving!
Yes indeed needs a very concentrated read.
Is there a version of this set out as a series of equations with all the esoteric terms as defined variables etc ?
Presumably some fundamental public sector accountacy sources/textbooks?
Richard seems to have done some of it with his calculations showing outstanding public debt being 1 trillion short of offical estimates?
That could be done
This issue is virtually unknown in academia
I have a draft paper in development now hitting the conference circuit
I wish to reflect further on Clive’s invaluable contribution rather than comment; save to make a general observation.
A general question is established, to frame the problem. Not, ‘the size of the national debt’, but rather: “what would it cost to pay off the debt today?”
This is a very odd question; because I can think of no set of circumstances in which this can be turned into a meaningful prospect, still less an outcome. It is a very odd question. What does it actually mean? The thought experiment was to establish the size of the national debt, but redemption prices offer higgledy-piggledy dates; meaningless. what would it cost to pay off the debt today, provides a single date, but I would submit is equally meaningless. I say this because, if such a purchase was aired (this, remember is a though experiment – I am entitled to ask, under the rules set); I suggest prices would immediately collapse. The supposed market price describes an environment, and a market that would not , and could not exist if the circumstance arose that allowed the question to be posed.
Having said that, I can think of case where something quite close in principle (but nowhere near the scale, even at the time, applied) was actually delivered. The bond issued to emancipate the Caribbean slaves in 1835 was calculated to produce compensation to the slave owners consistent with a colony-specific market price for each slave, supposedly at the point of emancipation. The issue was colossal in terms of Government national debt (but only a portion of total debt). The problem is, the compensation was paid because the colonies and slave owners knew the slave economy model, wtih a market for chattel slaves, was over. Most of the major absentee slave and plantation owners wanted to cash-in, and a way out. The exercise, however was a complete fiction; but a fiction that served the interests of both abolitionists and slave owners, in a mutual quandary. Really it was a political fudge, and in spite of a big, immediate impact on the national debt with the bond, the national debt did not really matter; and my point is, ‘at the touch’, it never does. Had the calculation been carried out, based on real market prices, knowing emancipation had arrived, the supposed ‘market price’ on which the emancipation bond was calculated, would immediately have collapsed.
The same applies here. What would it cost to pay off the debt today(?), is not just a thought experiment; it is a fiction. What is the size of the national debt (?) is a question that is extremely elusive and recondite, at least for all practical purposes.
In accounting we would call this mark to market accounting.
You are right with your concern: most mark to market accounting works on the assumption that a) the entity is no longer a going concern when it clearly wants to be so and b) there is a willing buyer for what might be sold – which may be true in the marginal case i.e. the daily traded volume for gilts for example, but not in aggregate. Clive’s model does assume aggregate sale at marginal prices when the aggregate will never be available for sale. I will muse on that.
Thank you for the acute appreciation of my argument.
I think you raised a very good point.
@John S Warren “what would it cost to pay off the debt today?”
This is surely impossible. The cost to pay off the Government Debt is surely the size of the Government debt. The money is in the economy, and would mean literally stealing everyone’s savings.
The only way it could be done is by some accountancy “fiddle” like the US $Trillion coin, https://en.wikipedia.org/wiki/Trillion-dollar_coin
But even so, the money in the economy will always be there, and accounted for, it’s just that we no longer pretend that it is a “debt”.
Agreed
“What would it cost to pay off?” is, as you correctly observe, a nonsense question with respect to National Debt; it cannot, will not and should not ever be “paid off”.
However, people DO talk about the size of the debt and the interest cost of servicing the debt and all I am trying to do is offer a simple way to calculate it in a way that is as similar as is possible to the way that folk might talk about their own debts. IE. Interest is as closely linked to actual cash flows as possible; size in “market value”
Do size and interest cost matter? Yes, they certainly do…. but not in the way most people think (ie. can I afford the monthly payments? will I ever be able to pay it back? etc.)
Size matters only in that it is (size) x (interest rate) that determines the cash spend by government (to pay gilt holders). This money can always be paid but it is a cash injection into the economy that may need to be drained by taxation/(more) gilt issuance if it is deemed inflationary. If done by taxation this could easily end up being a transfer from workers to rentiers unless action is taken.
I think the focus on capacity to pay is key – and you have been useful there.
As I understand it, selling and buying govenment bonds has no actual effect on government spending and does not constrian in any practical way.
That depends on your belief system