I started writing a long thread on NHS funding this morning before getting distracted into this one, on the absurdity of the accounting for interest on national debt, which has reared its ugly head, yet again:
Central government borrowing was, supposedly, £27.4 billion in December 2022. £17.3 billion of this was supposedly extra borrowing costs. This is utter nonsense. The numbers are wrong. A thread….
First the numbers are wrong because most of this supposed cost was on index-linked bonds. Most of this supposed cost won't be paid in cash for 18 years as a result, when these bonds will on average be repaid, which is when this extra cost might then be due.
Second, the absurd cost is in that case due to the insistence of the Office for National Statistics that we recognise a cost to be paid in 18 years time as an expense now when we should be spreading it over the next 18 years i.e., over the remaining life of the bonds in question.
Third, anyone writing about this has to appreciate that crap government accounting for a cost from which no one directly benefits right now is just crap accounting, and not a reflection of reality.
Fourth, there is also no real clue at present whether this cost might ever be paid. Deflation between now and repayment date could eliminate this expense. But even if that does not happen, the restoration of normal low interest rates means that over that period the real cost will be small.
So, fifth, those writing about this should say that crap accounting for a cost to be paid in 18 years time is no reason to panic now, or impose austerity now, because it does not in any way suggest our government has a finding crisis.
And, sixth, those commenting should ask a) why the government does such misleading accounting and b) why it issued bonds that could be so ridiculously expensive in the short term?
And seventh, they might point out that these index linked bonds were issued to help the pension industry that predominantly serves the interests of the wealthiest 10% of the UK so if any cuts are now called for as a result these will be to punish the poorest to help the richest.
And eighth, they should say ‘never again' should we set out to help the richest at potential cost to the poorest, because that's the real scandal in what is going on right now. As state services collapse what is being protected are the savings of the best off.
Ninth, then, these facts need to be laid by the media before ministers and they need to be asked to justify the crap accounting, the use of index linked bonds and the resulting upward flow of resources in society at a time of need.
These things are the scandal. The national debt is not an issue of concern at all, not least because the data that is being issued deliberately overstates it as an excuse to fuel austerity. That overstatement is the scandal. Talk about it.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Accounting for Inflation linked bonds is tricky for most people to understand…. but it matters because it is being used dishonestly to drive policy.
I think the simplest way to observe the absurdity of the ONS/Government position is to see it from the investors’ side. Have they received this extra £17bn in interest? They will answer (truthfully) NO – in which case the government has not paid it.
Now, there is a lot more detail and subtlety in accounting for inflation linked bonds but unless the ONS can “show me the money” they claimed was spent then I claim it was not spent.
Agreed
” Ninth, the facts need to be laid by the media before ministers…”
Not just the ministers it seems (from the BBC website “Energy Bill helps push borrowing to thirty year high”)
quote But Ruth Gregory, senior UK economist at Capital Economics, said the borrowing figures “provided more evidence that the government’s fiscal position is deteriorating fast”.
She said borrowing was well above what economists had expected, debt interest payments were at an “eye-watering” level, government spending was high, and there were “pressures from the weakening economy”. end of quote.
Jeremy Hunt then says ‘eye watering cuts must be made ti get the public finances back on track to be fair to future generations.
The policy of ‘budget balancing’ always involves cuts and IMHO diminishes future growth and productivity.
Is the suppression of the alternative view due to ignorance or is it deliberate to further an agenda?
Both seem plausible.
At the risk of boring for Britain on this topic……
It is clear that the claim that “interest costs” have risen by £17.3bn is nonsense so where does this number come from?
Well, it a confusion between “interest” – which is the periodic cost of servicing a debt and “capital” that total amount owed.
For Index Linked bonds the annual cash cost of paying coupons is quite stable (although slightly rising). What does change is the “amount owed”; an increase in RPI of 10% means that the amount owed (in nominal terms) is now 10% higher. This increase should (and IS) included in any statement of the size of the national debt (ie. money owing) – to include it as interest is wrong (and indeed counts the amount twice!).
So, the obvious follow up is “does the increase in the size of the debt matter?”… and the answer is “yes and no”. Yes – because it always matters that we account for things accurately. No – because concerns about the size of the National debt are misplaced both generally and in this case misplaced for very specific reasons.
Leaving aside MMT etc., (so as not to frighten the horses), the size of a debt matters in relation to the ability to service the interest. The state’s ability to service debt is linked to tax revenue which is linked to the size of the economy… so, if inflation is 10% then the economy (in nominal terms) is roughly 10% bigger and tax revenue rises by 10%.
Indeed, a one-off surprise burst of inflation (such as we have seen) actually reduces the debt burden as a proportion of GDP because the majority of debt is fixed rate in nominal terms.
Furthermore, any “market value” basis for determining National Debt has also fallen – just ask owners of gilts over the last year! (yes, rising rates reduces the value of the National Debt!! Weird but true).
…. and finally. (well, maybe)
Where I do disagree with you, Richard, is in your dislike for the issuance of Index Linked bonds because they benefit the already rich. In fact, I think the opposite is true.
A look at long term returns data for Index Linked and conventional gilts (adjusted for duration differences) show I/L gilts underperforming. This means that Pension Funds have “over-paid” for inflation protection… or, alternatively, Pension Funds value highly (and pay for) inflation protection that it is virtually costless for the government to provide (for arguments made in my last comment).
I think we will have to disagree on that
Again, it is another example of Neo-liberalism creating its ‘own reality’ as Mirowski has pointed out.
So, by issuing bonds, the Neo-libs have made something that did not have to be, be.
That is, that the Tories can say that it has gone to private individuals to raise money for its operations when it could have printed its own cash, like it did for the wanking bankers in 2008. It is simply a negation of government fiat money and sovereignty by people in government who do not believe in government. It gelds government power to help people.
A scandal? Well, OK, but I’d call it diabolical first and foremost. It’s Tim Snyder’s ‘the only idea is that there are no ideas’ concept in action. It is fascist sophistry, plain and simple.
This is good (for the Guardian at least):
https://www.theguardian.com/commentisfree/2023/jan/23/system-rigged-inequality-pandemic-despair-super-rich.
I still don’t understand why the ONS thinks the correct treatment is to recognise the uplift in the redemption amount of index linked gilts as “interest payable” in the month when the uplift is calculated.
I know this is what the international standards say, but it is stupid to abandon good sense to comply with a standard if the result is misleading.
This amount will not be paid for many years, and may not be paid at all. It would make more sense if this amount is accrued over the unexpired life of the bond, not recognised in one lump immediately.
They should, in my opinion, apply a true and fair override to the standard.
Richard, a note, interest paid on student loans RPI plus 3%. One of many elephants in the room.
Agreed
As we move to a situation where more than half of the younger age cohorts have tertiary education, so it becomes more the rule than the exception, it might make more sense to have a graduate tax, or a slightly higher income tax for everyone, or just fund further and higher education out of general taxation, than hang student debt over the heads of a generation.
But in the meantime, is there any justification for new student loans to be based on RPI rather than CPI? Why are they being gouged for excessive interest?
Abuse
You say that the pension industry mainly benefits the wealthiest 10% of the country. Can you help me reconcile this with what the ONS has reported?
“The workplace pension participation rate in the UK was at 79% (22.6 million employees) in April 2021”
The vast majority of funds in pensions are owned by the top 10%
The rest participate but at very low levels in most cases
Juxtaposition.
“Index linked bonds were issued to help the pension industry that predominately serves the interest of the wealthiest 10% of the UK.” Interest payed at RPI.
Interest is payed from day one of receiving a student loan. Many people working in the NHS, for example, will be paying back student loans, to privatised loan companies; interest on the loans being payed at RPI plus 3%. Profiteering?
One of your recent posts described an analysis you had made into which governments had actually made inroads to paying off debt and that despite having less years in office that Labour had actually paid off more.
So it would seem that the Tories use borrowing levels as justification for regressive policies.
However for much of my working life I have been conscious that “the working man is better off under the Tories” recognised by many who wouldn’t admit to voting for them.
Better off = less money taken out of the pay packet by government.
So has the history of Tory borrowing been about enabling indirectly the reduction of income tax and therefore maintaining a level of votes from this intrinsically selfish contingent. Who if they are getting something back are less inclined to ask where the rest of the money is disappearing to.
Most people
BBC distortion again – on radio 4 ‘More or Less’ recently Tom Harford – supposedly revealing the truth behind numbers bandied around – but instead , just climbed on the propagand bandwagon . Criticized some unions for using the ‘discredited’ RPI in their pay claims, – and quoting government saying so..
Didnt explain in any detail why RPI is discredited, nor that government itself uses it when fiddling its debt accounts, and apparently charging students on loans.
It would be interesting if you Richard, or someone, asked More or Less’ if they would do a programme on the numbers in these debt accounts.
No chance of that I fear
Not from the BBC…
Ben Chu also criticised the use PRI on Newsnight a few weeks ago. He said, as I recall that most economists use CPI as being more accurate.
Re. Andrew Broadbent- BBC distort again.
More or less- Teachers Pay 18th January 2023.
I have sent a little note to You and Yours in which I included several of their more dubious quotes.
From Josephine C” the teachers union and the IFS are using a different measures of inflation. IFS using CPI, and that’s the standard measure of inflation, but the NEU are using RPI.” Oh dear oh dear says Tim Harford ” RPI not approved of by statisticians”.
Josephine C agrees “Well known defects fallen out of favour – the ONS still publish RPI but are disparaging about it.”
The ONS said a few years ago ” our position on the RPI is clear we do not think its a good measure of inflation and discourage its use.” Luke, from the IFS agreed, he and lots of economists think that RPI over estimates inflation.”
I must admit I did mention in my little note” index linked bonds and student loans re. profiteering/ larceny?
So Tim, “RPI not approved by statisticians” – depends who you ask.
I am so glad you returned to this, Richard. Last time, I felt it was discussed here, I felt it was left as ‘unfinished business’.
Your argument is clear and persuasive, and Clive’s comments were of course – s usual – pertinent (I shall ‘body-swerve’ the debate you and Clive have on the value of issuing indexed linked gilts; intuitively I just don’t like them, because there are too many moving parts to be effectively managed, at lest from a DMO perspective).
I remain totally convinced by your accounting argument, against the current ONS position. Notably the ONS do not seem prepared to argue their corner (nor is the BoE as far as I know); which rather confirms your case. They need to ‘put up, or shut up’.
I have further objections. Basically, since we do not know the future; how can we possibly know what the actual redemption impact is on, say an ultra-long 50 year indexed Gilt, for a passing, ephemeral six month period of 10% inflation (say); which, let us speculate, could be followed soon after by twelve months of 10% deflation; with both impacts working into a redemption value, almost 50 years later?
We know from DMO consultation papers 2001- 2012 that various changes and adjustments are constantly being reworked for new indexed linked Gilts, for variables that keep affecting the investment attraction or value (to meet the challenge of real inflation proofing, which index-linking files – hence the consultation papers presumably); factors to adjust for the complexity include: the inflation index used; the indexation lag (8-month, 3-month etc); the length of time, short, meduim ultra-long (35-50 years), super-long (>50 year) Gilts; deflation floors and so on. If there is no deflation floor, what can you say about redemption payments for inflation that may be wiped out; or worse. We are also hazarding to take as an expense today something that may happen (if not overwhelmed by other contingencies) in 50 years or more time. This seems to me the ONS saying that it is the purpose of index linked Gilts to defeat the point of a sovereign country issuing debt – which is to defer capital costs far into the future, and pay for it by servicing the debt. In reality the long term future must take care of itself; because we know very little about it.
It seems to me the inflation element of the indexation cost of a redemption cash payment due in fifty years is in reality, and for today, presumably a contingent liability?
Allow me to provide a small example of the nature of the problem, in this quote from the 2001 DMO consultation paper on index linked Gilt redesign (the ongoing tinkering), which I think touches on the kernel of the complexity of the problem:
“Most notably, a deflation floor would reduce the value to HM Treasury of having index-linked gilts in its debt portfolio by reducing their deficit- smoothing properties in some circumstances. Inclusion of a deflation floor would also require primary legislation in order to amend Section 717 of the Income & Corporation Taxes Act 1998 (ICTA). In addition, it would rule out any possibility of index-linked coupon and principal strip fungibility if this were felt desirable at some point in the future.” (p.13).
‘Stripping’ opens a new complex market, in splitting a coupon-bearing bond into its individual cash flow, which then trades as a zero-coupon bond; and fungible refers to the capacity to improve market liquidity through allowing the ‘strips’ of coupons (an acronym: ‘Separate Trading of Registered Interest and Principal Securities’) to be interchangeable. I am sure others, more knowledgeable of the market are better placed to dicuss the detail. My point is to open the complexity that is being created, to view. We have had sufficient surprises of an unpleasant kind; the financial crash 2007-8, still affecting us indirectly though the following austerity obsession; or the pension crisis only weeks ago, with Liability Driven Investments (in the pension fund industry!) that few people understood, almost bringing another crash, within hours.
Crikey John, that all sound fascinating
Almost the basis of a paper….
“…to punish the poorest to help the richest.”
Precisely.
Richard, Surely in addition to index linked gilts, there must be actual interest being paid on the balances that the banks hold with the BoE as you have discussed at length in previous posts. Do we know whether that was included in the total and was it significant?. Thanks
They were
I need to look at the breakdown
What we know is most of the cost was on index linked bonds
Agree that BofE shouldn’t be writing off now what they don’t need to repay for many years, and most surprised that the ONS recommends it.
Most banks, and large corporates, hedge, in part or full (depending on risk appetite), their future risk, albeit at an extra cost, so they have a good estimate of the future and current value of assets, as well as debts. For inflation-based risk, there are a number of instruments (or combinations) with which to do this, including inflation swaps, futures, and options.
So, I’m just surprised that the BofE is even *allowed* to keep index-linked, or other, bonds in open /bare /unprotected /unhedged positions, particularly if they are trying to justify it by currently writing down a large proportion of the estimated future realised debt.
The National Debt is a Treasury, not BoE issue, and the BoE owns very little index linked debt
Your analysis misses the point
OK, with a third of National Debt held by the Bank Of England, I had imagined that the recent pension companies’ crisis had forced the BofE to buy a lot more index-linked gilts… -therefore, effectively, passing the risk to the BofE.
In any event, irrespective of who ultimately bears the risk, the cost of financial instruments allows an effective valuation of that risk, either by conversion to future known cashflows, or the valuation (/discounting) of those cashflows into present values.
Through these methods, any holder of the risk can determine what needs to be paid out, and when.
So, I am absolutely agreeing with your point that this debt doesn’t all need to be converted into current repayments, and am merely illustrating how that can be avoided. That is what I thought provided the strongest substantiation of your point, rather than missing your point.
See this morning’s post
And index linked binds hardly feature in the BoE purchases