I have been looking at the thorny issue of what makes up the public sector net debt, excluding public sector banks, that was announced last week to amount to £2,004 billion (£2.004 trillion). To do so I have used the mass of spreadsheet data published by the Office for National Statistics with that release, all of which spreadsheets are deeply frustrating because they have all the formulas removed from them before publication, which makes no sense at all.
Having undertaken this exercise I believe that the figure for debt is officially made up as follows:
To the point where debt is £1,809 billion, this makes sense, or at least I can guess the origin of most of it, albeit that it would be good to know what 'other sterling debt and foreign currency debt' really was, and what the liquid assets were would also be good to know. But there is a number I am truly baffled by, and that is the Bank of England c0ntribution to this debt.
The ONS said that in February 2020, when the Bank of England last issued accounts, that the Bank of England contribution to Public Sector Net Debt was subject to three issues that they noted:
The figure that they suggested the sum came to in February this year was not insignificant, at £174.1 billion. So I have tried to track the number down as at that moment because the ONS says it is based on the Bank of England accounts, which are available for that date, and the ONS's own estimates.
The Bank of England balance sheet reads like this in February 2020:
As will be noted, assets and liabilities are remarkable evenly matched: there are net assets of £5.8 billion. So to find liabilities of £174.1 billion looks to be hard work.
Thankfully, some is easily found. That's because, bizarrely, the Band of England does not publish one set of accounts for its activities, but three. Those for the Prudential Regulatory Authority can, however, be ignored: they are utterly immaterial to this issue. But those of the currency-issuing department are not. This is their balance sheet on that same date:
I think that locates £74.4 billion of liability: it is currency in issue. There is just £99.7 billion to find then. And since we are looking for liabilities only notes 11 and 12 should hide such a sum. These are the notes in question:
The deposits held on demand are clearly not what is being looked for: because of the relative size of them and the number being looked for it seems very unlikely that they are what makes up this figure.
I think we can also ignore the £66,552 owing by the issues department: it is an intra-group balance that should be cancelled on consolidation. That is confirmed by Note 4 to the Issues Department accounts.
So, we are left with a hotchpotch. The effective impact of the Asset Purchase Facility is in the loan to it, amounting at this date to £445 billion. But this is an asset. Of course, that asset helped create the central bank reserves of £479.4 billion. But why are they offset, if they are? Is it now acknowledged that the objective of QE failed and it just produced cash and not new investment? That is what the offset, if it is made, would imply. And yet, it seems that they must be, because otherwise those reserves are not reflected in debt and yet it is said by the ONS that the Asset Purchase Facility that is related to both does have an impact. In that case, then, does note 8 also come into play? This says:
However, I stress, once more, that this is an asset. So what this has to do with debt is hard to tell, and that would also seem to leave the ONS note as deeply confusing, at best.
I have then some simple questions to ask in that case:
- What makes up other sterling debt and foreign currency debt?
- What are the government's liquid assets and where are they held and how does this sum interact with the Bank of England?
- How is the Bank of England contribution calculated and how does it relate to the Bank's published accounts?
I genuinely have no clue as to the answers to these questions and the ONS data on the national debt does not answer them, and nor does my attempt to link the data to published source documentation do so. I have made Freedom of Inf0ormation requests to try to find out.
I have already made the point, time and again, that the above logic, which ignores debt subject to quantitative easing is wrong. It is not credible to ignore QE simply because it is deemed that the Bank of England subsidiary company that owns this debt is considered to be outside the government sector by the Office for National Statistics when that company is wholly owned and controlled by HM Treasury, for whom it acts as an agent, making a mockery of what the ONS assert, but I now have another concern.
In my opinion, if the Office for National Statistics is going to issue data that it knows will grab headlines - and it clearly and knowingly did that last week - then it has a duty to also put out an unambiguous explanation as to where its data comes from, and to show how it can be reconciled to it, if appropriate.
Try as I might I cannot explain the national debt based on the figures they published, and I am a fairly informed reader of accounts and statistics. There is, then, a problem. And given the significance of this issue this needs to be resolved. Too much hangs on this number for it to be made of data of unknown origin and credibility.
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:
This blog is an act of pure public service as far as I am concerned.
I have always felt that public accounts are no better than so many private ones.
As things are these accounts can be interpreted in any number of ways that getting to the truth of the matter is impossible (I remember trying this myself a number of years ago as a student and just giving up because it made no sense).
How can any competing claim on fiscal policy ever gain credence with so much
misinformation around? Therefore I think that what we have here is nothing but deliberate obfuscation of the situation. Obfuscation that can be abused.
And it is appalling.
I would have thought the £72,295 billion is the bank notes in circulation, though you may be right that it is part of the BoE contribution to debt of £194 billion. However, it seems strange as a BoE contribution to debt would suggest that the BoE has an accumulated loss or some sort of other deficit.
I am not an accountant but since the BoE is part of the Treasury Group then do you not have to cancel out all items between the Treasury, BoE and the APF to get a consolidated Group Account and thus the actual overall position from the external view point?
If Keynes was alive today I’m sure he wouldn’t be saying the government failing to set out its accounting straight forwardly was “a masterly manipulation” as he did with the Bank of England’s sleight of hand in raising money for the First World War effort:-
“… to cover its tracks, the Bank made advances to its chief cashier, Gordon Nairn, and his deputy, Ernest Harvey, who then purchased the securities in their own names with the bonds then held by the Bank of England on its balance sheet. To hide the fact that the Bank was forced to step in, the bonds were classified as holdings of ‘Other Securities’ in the Bank of England’s balance sheet rather than as holdings of Government Securities (Wormell, 2000).”
https://bankunderground.co.uk/2017/08/08/your-country-needs-funds-the-extraordinary-story-of-britains-early-efforts-to-finance-the-first-world-war/
Richard, I think I can help / explain the Bank of England mystery – though I’m sure ONS will come back to you.
The important thing to note is that it is a net debt concept – so gross debt (using a specific definition) minus some, but not all, of the banks assets – the liquid ones.
So, most of the liabilities of the Banking Department and the currency in circulation liability of the issue department are probably included – this comes to about £650b.
But you then have to offset the liquid assets – this includes the £445b loan to the APF, because the APF is (I think) ultimately consolidated here and so the APF holding of gilts (which are, I guess, considered liquid assets) are offset.
This gets you to £205b. Which is in the ballpark. You can probably get to the precise number by consolidating the Banking and Issue Department (there are some cross holdings here) and some other adjustments.
Now, as we discussed, it would probably be helpful and more transparent for ONS to break out the gross debt and offsetting liquid asset components for BoE, or explain this component better, but it is explicable – I’d agree it could be made clearer (as we see with the other components of PSNDex) though even here ONS could explain some components better (eg the impact of classification of housing associations or RBS / LLoyds on nonfinancial corporations debt and public sector banks)
I’ll be interested to see the answers to your other questions!
Phil
On reflection the issues department cannot be added on gross, because £66bn is already in liabilities in the banking department balance sheet. So that would be double counting as it turns out
And if you offset the £445 all the rest should, be offset too as liquid assets
So we get to about £6bn in the balance sheet net liabilities plus another maybe £8bn grossing up notes in issue
I don’t see tha answer in your suggestion, but let’s see what is said
If anything I am further apart now….
It’s the wrong amount, but there’s still a balance of £370m on the Ways & Means Advances to HM Treasury, which is an asset for BoE, but also will have created a liability for BoE and for HM Treasury.
How is that accounted for?
As it’s an intra-government balance it should cancel out, but thanks
Hi Richard,
Its a mystery isn’t it? I’ve spent an hour playing around with balance sheets for the Banking Department and the Issue Department and consolidated numbers to try and figure out the method. My guess is that they are (wrongly) not consolidating the 66bn deposit asset / liability between the Banking Department and Issue Department. Instead, they are recording currency in circulation from the issue department and maybe most, but not all of the “deposit” liabilities of the Banking Department as gross debt of about 642bn.
I suspect they are then offsetting as liquid assets the BEAPFF loan (they must be offsetting this, as its the only way to get anywhere near the net debt contribution) and some of the other assets (probably the various securities). Using this method I can get to 178billion.
Its wrong, but I think its plausibly what they are doing – obviously things would be much simpler for ONS if the BoE produced consolidated accounts and if ONS showed the gross positions.
One final, fascinating thing. I was interested to see if I could get closer to the ONS numbers using an older set of Bank of England Accounts, so looked at 2009 – what is fascinating is that from 1997/98 through to 2007/08 the Bank of England contribution to PSND is negative (and again in 2011/12). I.e. they are including a greater amount of assets than liabilities in net debt for these periods. Its a total mystery to me how this is the case!
I’ll be fascinated to find out more about their precise methodology!
I tried all that too….it’s like a jigsaw with some seriously missing pieces and maybe one or two from another box
I will be pursuing this one
If but does not get answered by FOI there are other routes…..