Saturday, 15 June 2013

The Consolidated Government Sector

The key analytic technique that MMT uses that sets it apart from most others, is that it uses a consolidated government sector in its analysis (although apparently the mainstream is slowly catching on). This allows it to cut through the obfuscating political constructions between the various government departments and institutions and concentrate on the essence of what is happening

This is entirely consistent with accepted accounting practice, using a technique known as group accounting - which produces consolidated financial statements (income, balance and cash) amongst a related group of entities. The international accounting standard for that is IFRS 10 'Consolidated Financial Statements' which requires that entities under common control present a consolidated set of accounts so that external users can obtain a 'true and fair view' of the actual underlying economic transactions.

The Central Bank in all sovereign jurisdictions falls under the definition of control by the Treasury - often de facto by the operation of law (Bernanke: "Our job is to do what Treasury tells us to do"), but also de jure, e.g in the Sterling area HM Treasury actually owns the entire shareholding of the Bank of England. The control model in IFRS 10 is elaborate to try and catch all those little tricks that entities use to avoid having to consolidate accounts and is worth studying to see the various 'Wizard of Oz' methods that control can be imparted even though the public face is supposedly independent.

Given the control relationship, consolidated financial statements are entirely appropriate and correct accounting which reveals the essence of the underlying transactions. Therefore in your model you should be able to swap out the detailed entities and replace them with the consolidated entity and nothing about the response should change. If it does then it is likely your model is wrong.

In Information Systems this is known as white box and black box testing. With white box you can see the internals and with black box you can't - requiring you to conform to that modules interface in your testing.

So let's look at the white box of the UK government sector

Loads going on here with flows flying every which way from both the main government entities and that's without breaking the sub departments out - the Issue and Banking department at the Bank of England, the Asset Purchase Facility and the Debt Management Office. Much is drawn from the current operations document at the Bank of England, HM Treasury's remit to the Debt Management Office but also this rather nice description of the Central government spending flows which has a very useful picture in it:

Importantly what this arrangement of accounts tells you, along with the description of the way Reserve accounts work at the Bank of England, is that spending is disconnected from the debt management operation. All entities operating at the Bank of England have an interest free overdraft available during the day to absorb flow differentials, and this effectively means that they spend on their accounts and then back fill (or under fill) to the target account balance using standard asset and liability optimisation mechanisms.

Very likely some of the government spending flow out to the private sector stock accounts earlier in the day will flow back later in the day to the National Loans Fund in exchange for Treasury bills or Gilts as a completely separate transaction. Spending therefore demonstrably comes first. Moreover it is in the Debt Management Office's interest for that to be the case since it ensures there will be healthy demand at the funding auctions.

Once you apply group accounting techniques to the government sector you get the black box view - which is remarkably simple.

All the government sector does is allocate tokens to the non-government sector (often in return for some real provision to the government) and then swaps those tokens around for different tokens which attract different income flows and have different liquidity characteristics and holding criteria. The elements this simplification reveals are quite interesting. For example, if I have a young child and qualify, then that child attracts Child Benefit. If I then sell that child to the gypsies, then the gypsies start getting the Child Benefit instead. There is remarkably little difference between a Gilt and a child from the financial systems point of view - a difference in the income stream attached and a difference in liquidity, holding cost, collateral value and issuing criteria :)

With this model you can see that QE is a token exchange - Gilts for Bank Reserves. The income flow goes down - because entities holding Bank Reserves receive less government spending than entities holding Gilts. So you will get a portfolio reconfiguration - since there is less income to go around and the mix of tokens has changed. Bank Reserves have a narrow holding criteria than Gilts and therefore are likely to create offsetting commercial bank deposits that have a wider holding criteria.

The detail is important to understand how the government sector works internally and to understand the actual operations. But it is very easy to end up in a situation where you struggle to see the wood for the trees. Hence why the consolidated government sector is an important modelling tool.


Frances Coppola said...

Point about disconnection of spending from funding is well made. However, you can't assume therefore that spending comes first. It may not. Governments may choose to pre-fund, issuing debt on the basis of planned spending. That's why governments need central bank deposit facilities.

strowger said...

Love the line about gilts and children. It isn't quite true of course - child benefit entitlement is not negotiable or heritable. It is also far more likely than a gilt to be defaulted or arbitrarily subject to confiscatory taxation.

Give me war loan any day ;-)

Neil Wilson said...

"However, you can't assume therefore that spending comes first."

I can because the document above written by the government shows that is what they do.

Even if they choose to pre-fund there got to be something out there to fund with, which will come from repo at the central bank.

So at the consolidated level the pre-fund disappears - the central bank issued the money.

So my point about swapping in and out the consolidated government sector into your model stands. If you'd done that then you wouldn't have made that mistake.

Neil Wilson said...

" child benefit entitlement is not negotiable or heritable."

Neither are premium bonds - hence my point about issue criteria.

Lars said...

Neil, your analysis is not the way it's usually done in the USA. Most government bond auctions involve repos where there dealer borrows from a bank intra-day using the t-bond as collateral with the money markets as the other side. So the funding source is actually the private sector in most cases. The repo very rarely comes from the central bank as MMT shows.

Maybe it's different in the UK, but that doesn't mean it's always the case. It's definitely not.

Charles Hayden said...

The currency issuer must spend/lend the currency before it can collect taxes in that currency. Simple point of logic.

y said...

When did Bernanke say "Our job is to do what Treasury tells us to do" ?

y said...

"That's why governments need central bank deposit facilities"

Huh? I don't understand this comment.

Are you saying central banks are not part of the government? Or are you saying something else?

Because "that's why governments need [government] deposit facilities" wouldn't make much sense.

Neil Wilson said...

"Most government bond auctions involve repos where there dealer borrows from a bank intra-day"


You are confusing the micro with the macro. Let's make sure we're working at the same level of abstraction.

Please be clear what I am saying.

If there are spare reserves *anywhere* in the system then that came from some form of previous government sector injection. The intra bank cannot clear at the end of the day without those spare reserves existing. QED.

If there are no spare reserves in the system then you cannot purchase government bonds without the central bank providing some. The central bank is part of government.

Nearly all the so called debates I've seen on this topic are down to definitions, understanding levels of abstraction and failing to follow transactions to their transitive conclusions.

Neil Wilson said...

"When did Bernanke say "Our job is to do what Treasury tells us to do" ?"

In testimony to the Joint Economic Committee on May 22, 2013

Lars said...


You're assuming there are reserve requirements to begin with. That's not the case in all banking systems and many systems operate without the central bank ever having to inject reserves to ensure settlement.

As I said, you seem to be extrapolating the case of the UK out to all systems to prove a point that is simply not always true. There is no real need for reserve injections in all systems. They certainly help at times, but they're not a necessary part of the settlement process as you imply.

Neil Wilson said...

"You're assuming there are reserve requirements to begin with."

No I'm not. Reserve requirements are irrelevant. There are *no* reserve requirements in the UK. You simply are not understanding how the accounting works at central banks.

If you want to continue then you need to provide accounting journals and/or the relevant legislation that allows an entity with a central bank account to retain an overnight overdraft on its account at the central bank.

Neil Wilson said...

On that Testimony, the point you are looking for is at 1:35:35 in the video. It's in relation to a question about the effect of the debt ceiling on payments.

y said...

Lars doesn't seem to understand that bonds are bought from the Treasury with reserves.

When treasuries are paid for, bank reserve accounts are debited and the Treasury's account at the CB is credited.

Primary dealers might borrow from banks or money markets or whatever, but this makes absolutely no difference to the fact that treasuries are paid for with reserves. It is very simple.

"many systems operate without the central bank ever having to inject reserves to ensure settlement".

Complete nonsense. In countries with no minimum reserve requirements the CB constantly supplies reserves through overdrafts, loans, repos etc.

This is how it works Lars. It's very simple. All else being equal, if a customer at bank A makes a $100 payment to a customer at bank B, bank A has to transfer $100 in reserve balances (currency/base money) to bank B.

"you seem to be extrapolating the case of the UK out to all systems to prove a point that is simply not always true"

You seem not to understand anything about the monetary system. It doesn't matter whether it's the UK or the US.

"There is no real need for reserve injections in all systems. They certainly help at times"

What on earth are you talking about? Reserves are money.

"but they're not a necessary part of the settlement process as you imply"

Reserve balances are used to make payments between organizations that have accounts at the central bank. Basic fact.

Ralph Musgrave said...

One could argue that the technical or legal question as to who actually owns a central bank is not all that important: the really important point is the EFFECT of a central bank, or how it behaves.

For example, the Bank of England was not nationalised till 1946. But ever since its foundation in 1694, it BEHAVED in relatively non-commercial manner didn’t it? I.e. it sort of played the part of “nanny” and “umpire”: e.g. acting as lender of last resort. In short, it acted AS IF it were part of the government machine.

Neil Wilson said...

de facto is always what is important for effect. You have to lift the veneer and see what is actually happening.

Just like a magician. It is nice to watch the trick. It is nicer to understand how it is done.

Lars said...

I see what you're doing here. You're implying that the reserves are the actual source of funding. That's all well and good except that it's not what's actually happening.

For instance, if I buy a Treasury Bond in a Treasury Direct account the actual funding source for the bonds and the government's spending starts with bank deposits that I transferred to my bank. The fact that the bank settles with the Treasury in reserves doesn't change that. And it doesn't mean the funding source started with the bank and not my Treasury Direct account. You're skipping a step in the process which is convenient for proving your point, but very wrong.

Also, Y, if a transaction settles between customers the same bank there is no change in reserves. They never even enter the equation. So stop spouting nonsense and telling other people they don't understand. You don't even know the basic facts here.

Neil Wilson said...

"I see what you're doing here"

I see no evidence of that in your response.

Tom Hickey said...

Ralph: "For example, the Bank of England was not nationalised till 1946. But ever since its foundation in 1694, it BEHAVED in relatively non-commercial manner didn’t it? I.e. it sort of played the part of “nanny” and “umpire”: e.g. acting as lender of last resort. In short, it acted AS IF it were part of the government machine."

The BoE was founded because financiers and business people did not trust the king to be disciplined and so it took over the job itself in order to ensure monetary discipline. It was an early form of "central bank independence" that this still enshrined today, even when central banks are government agencies. The people with the money don't trust kings of politicians to be disciplined, so they demand to set at the top of the hierarchy. Boils down to who holds the power. The king lost monetary power in 1694 owing to earlier issues over discipline, which induced the money powers to take that power away from the crown.

y said...

"if I buy a Treasury Bond in a Treasury Direct account the actual funding source for the bonds and the government's spending starts with bank deposits that I transferred to my bank".

You literally have no understanding of how the monetary system works. It's hilarious.

"if a transaction settles between customers the same bank there is no change in reserves"

That's blindingly obvious. I never said anything to the contrary.

DkN said...

Lars, I think you should direct your inquiries to find out where is the source of the money you transfer to your bank, look how it came to you from the time of its gestation in a computer of the central bank. And, please, read the central bank balance sheet. Everything is written there


y said...

"The fact that the bank settles with the Treasury in reserves doesn't change that"

Do you even know what the word "settlement" means?

y said...


the link:

"this rather nice description of the Central government spending flows"

doesn't work.

Neil Wilson said...

Ta. Fixed that - and added in the link to HM Treasury's operational brief to the DMO.

y said...


might be worth doing a post explaining in detail why the stuff Cullen Roche writes these days is nonsensical gibberish.

I mean look at this, for example:

If it wasn't so sad it would actually be quite funny.

ScottTheDot said...

Pretty pictures do not change the reality that we are being forced to live in austerity because some people THINK we owe a lot of inescapable debt.
"The rich rule over the poor, and the borrower is slave to the lender." - proverbs 22:7 Bible New International Version.
OK, what did they know?
But then why, in God or whatever's name, are we paying interest on money we could simple create ourselves, and supporting the wealthiest rent-seeking generation in history on bond payments for doing nothing?
And why are we doing this when the Constitution, SCOTUS (Julliard v. Greenman), Lincoln, and so much in our own and world history shows the benefit of true sovereign money?

Nick Edmonds said...


I like this post and I think it's a good point you make about the use of the consolidated view as an analytical tool.

However, I wondered if you could clarify for me the point you are making about spending preceding borrowing.

In terms of overall government net indebtedness including the OPG account (if we are not consolidating the BoE) or private sector reserve balances (if we are), presumably the spending and the borrowing must happen simultaneously, by the simple operation of debits and credits.

So, I'm guessing when you talk about borrowing, you don't mean in those terms, but are referring to the management of the form that net indebtedness takes. So, we are talking here about the bill issuance and money market operations carried out by the DMO as part of the on-going liability management.

I think the question then is whether you are saying that the precedence is one of timing or just causation. In other words, are you saying that the DMO waits until the government departments have processed their payments and then goes out into the market to issue bills and gilts. Or are you simply saying that the government departments tell the DMO what they expect to spend and receive and the DMO then uses this information to run the debt issuance, even if this means the debt issuance happens before the spending.


Neil Wilson said...

Because the BoE offers unlimited intra-day overdrafts nobody operating those accounts needs to co-ordinate the balances until the end of the day.

Dynamically it is better for all instrument sellers in the system for the overdrafts to be at their highest points before they start offering their own instruments into the market. Since that would guarantee the best bid.

What I'm saying is that the DMO does what it can to get the best deal for the Treasury, and the best is likely to be achieved if it offers after it believes those overdrafts have expanded - since there will be more bids in the market and they would get a better price.

I would expect serious treasury department in any organisation to know when they get the best price for their placements.

So I would suggest it irrational behaviour for a treasury department of any organisation to pre-fund if they can get away with it.

Nick Edmonds said...


So, I think you're saying that the DMO will aim to fund at the point when the excess of spending over revenue is at its highest, which will typically be at the end of the day. Is that right? I'm not that familiar with how the DMO in practice - I'll have to read up on it.

I found your last comment a little surprising. If an organisation knew it was going to be making a big payment near the end of the day, I would have thought they would want to fix their funding well beforehand. The money markets get more illiquid as the day goes on. It was my impression that bank dealers, for example, are generally quite nervous of being clobbered on price if they have to position late in the day.

Neil Wilson said...

"The money markets get more illiquid as the day goes on."

Yes they do - hence my 'get away with it' comment.

The shift in the traditional building societies during the 1970s from having an amount of money to lend per month to being able to issue loans pretty much on demand is an example of 'getting away with it' - due to using treasury operations.

And don't forget that the reserve market is a closed shop. The *only* place the excess can go in aggregate is to HM Treasury or to the OMO actions of the Bank of England, and the Bank and HM Treasury co-ordinate to make sure they're not bidding against each other - in order that the BoE can maintain its target interest rate.

The DMO's operations are dynamic throughout the day and there are fixed points when gilts, etc issued and notified well in advance. Loads of co-ordination techniques used to reduce 'risk' and 'cost' over time in what is really an elaborate game.

There's lots of good stuff on the DMO site where it details that the DMO targets a weekly cash balance in their BoE account to allow for daily buffering.

Nick Edmonds said...

Thanks Neil,

There is indeed some useful stuff on that site. Interestingly, I saw on there a box on "a day in the life of a government cash dealer" as part of the article from The Treasurer.

I think this answers my question about timing, because there are various comments suggesting that they try to do the majority of their financing early in the day. It mentions at one point that "by 8am, we may have lent or borrowed many hundreds of millions of pounds". This sounds to me very similar to what I would expect for a dealer at a major commercial organisation.

Neil Wilson said...

But bear in mind that they operate to a weekly buffer - which means that early in the morning can absorb the buffer spend from the previous day.

Working to a figure is just an overdraft system using a number greater than zero.

Nic said...

In the link that you posted it says that departments make forecasts of the next days spending and incoem and then this is used by the DMO to decide hopw much is need in funding. That imnplies that spending is not diconnected from the DMO or that it precedes income, but that all spending is pre-funded. The other thing that your analysis ignorees is that most departments use commercial banks and hold singificant cash deposits.

Neil Wilson said...

Desperate stuff Nic.

You can try and shoehorn things into your world view as much as you like, but the simple fact of the matter is that UK government cheques don't bounce - because there is nobody that is prepared, or even empowered, to say 'no'. And that simple fact changes the entire drive of the operation.

Running at a weekly cash target is precisely the same as running at a target of zero. It wobbles below and above that, but nothing ever bounces and nothing is ever stopped. The buffer is an operational irrelevance - designed to stop reports of using overdrafts upsetting those simple souls who believe those sort of things are akin to devil worship.

Budgetary practice is an internal control function to keep things in line with budgets. It is has no operational limitations.

That's why the UK government could nationalise Northern Rock at the drop of a hat.

Very simply you can't do a reserve drain until you've done a reserve add. And no amount of wishing very hard will make it otherwise.

spatchcock said...

Hi Neil,

Really interesting post. Thanks. I am trying to reconcile what I've learned from MMT with more mainstream descriptions of the same processes. One thing that is getting me confused is base money growth and involves the non-consolidated accounts of the treasury and the BoE.

I am trying to understand how the BoE grows base money, and in particular I am interested in any interplay between monetary and fiscal policy - MMT often states that fiscal deficits drive base money growth. Since, fiscal deficits are by rule neutralised in practise (in the UK), I'm trying to understand how fiscal and monetary policies combine to produce the same outcome.

All the publications I have read so far talk about expanding and shrinking reserves during open market operations and tend to imply a short term process but don't comment on the longer term, net growth. Does M0 simply grow by cumulative net adds via OMO's? If so, one would expect the BoEs balance sheet to grow through time, which I think it has done, although interestingly (to me) it doesn't appear as though the BoE tends to hold a lot of domestic government debt unlike the US Fed (I am thinking about non-QE times). The implication, as I understand it however, is simply that the government deficit spends, neutralises this with bond issuance, and the BoE buys back bonds or other assets to accommodate M0 growth. In a consolidated sense, some of the Treasury liabilities are now balanced by assets at the BoE (bonds or otherwise) and therefore one could infer that some of the real resources purchased in the deficit spend were financed by new money creation.

One thing that has thrown me a little is a publication by Simon Gray (The Management of Government Debt: which discusses how the Treasury tries to avoid issuing more bonds than are required if some are to be bought back by the BoE to accommodate M0 growth. This publication describes a financing calculation where the amount of securities required to accommodate M0 growth is subtracted from the gross amount required to neutralise the deficit and only the net quantity of bonds are issued. This seems coherent to me but it implies that basically some of the government deficit is allowed to be non-neutralised, that is, government spending by money creation (in line with monetary goals of course). This would also mean that reserves increase but assets at the BoE do not (no bonds bought). In a consolidated sense, this amounts to the same thing as above but in this case the BoE balance sheet looks different. In publications from the DMO I've read the opposite too - that the BoE might ask the DMO to issue more bonds than it needs for fiscal purposes. So it seems like the monetary authorities sometimes use the fiscal authorities to smooth their operations?

This might seem like a subtle difference but I'm really keen to get it right. I suppose I'm looking for evidence that what MMTers say about spending and money creation being constrained by monetary goals rather than income is in fact what is going - albeit obscured by the accounting and policy rules.


Neil Wilson said...

You're probably over thinking it. In system terms for there to be a constraint there has to be somebody who can say 'no'.

There isn't anybody in this process who can say 'no' and make it stick - legally or politically. The whole thing flexes dynamically to accommodate what HM Treasury has been instructed to do by government and parliament.

Also don't forget that the BoE can 'borrow' Treasury bills to issue to the market. That's how Funding for Lending works for example. In this process the Treasury expands its balance sheet and lends the bills to the BoE which then discounts mortgages with the same bills.

That borrowing process is 'off balance sheet'.

There is no real separation between the Bank of England and HM Treasury. They are two arms of the same creature and act in unison. Any separation is there as a political illusion for those that want to see such a thing.

In effect HM Treasury spends government bonds and the BoE provides the liquidity service that allows that process to happen.

spatchcock said...

Thanks Neil.

I didn't know that about Funding for lending - something else to dig into deeper.

Interesting that you say 'spends bonds'. I think Warren Mosler argues that bonds should just be considered as base money.

I take your point about over thinking it. I guess I've had so much resistance from friends, colleagues, Twitter, etc. to some of the ideas of MMT that want to be able to point at actual numbers and say "LOOK!". The consolidation seems to be pretty key, and it's unfortunate than many people I speak to have misunderstood the idea of BoE "independence".

Neil Wilson said...

Get them to point out the person that can say no to an elected government without causing a constitutional crisis along the lines of the 1910 constitutional crisis that prevented the House of Lords saying 'no' to a Finance Bill.