Nick Rowe and his colleagues in Canada have a superb blog. If you’re interested in economics and don’t already follow it – take a look.
Recently, Nick and I have been discussing the relevance of ‘permanence’ to helicopter drops. In some ways our positions reflect polar opposites. Nick defines a helicopter drop as a permanent increase in base money. I think applying the term ‘permanent’ to a helicopter drop is meaningless. I don’t buy the analogy with tax cuts. I define helicopter drops as transfers to the private sector financed with base money. Period.
Ironically, I think Nick and I agree on the substantive point. My interpretation of what Nick is saying is that in order for a cash transfer from a central bank to households to be a genuine policy innovation it cannot be a loan. In this sense, he is arguing that a transfer payment financed by bond issuance is a loan, because at some point taxes will have to be raised to repay the bonds.
To this extent – and perhaps this is all that matters – we are in agreement. I think cash transfers to households are not loans, or if they are to be structured as loans they should be perpetuals which are never repaid – which would be legal in the Eurozone. So we both agree that helicopter drops are transfers to households which are never ‘repaid’. I put ‘repay’ in inverted commas, because it is not at all clear to me what ‘repay’ means in this context.
Part of where I think I diverge with Nick is over how people think. I’m an economist, and I simply don’t have any relevant expectations regarding the future path of the monetary base. Nor would my expectations change if I got a cheque from the Bank of England. Nor do I think I can extrapolate from changes in the monetary base to my personal economic circumstances. In this regard, I think I am a ‘representative agent’.
As an economist, I am also reasonably certain that a 2-5% of GDP cash transfer from the ECB to Eurozone citizens over 12 months would raise nominal spending and has a very good chance of shifting the Eurozone to a better, higher growth, equilibrium. And I simply don’t think that expectations about the monetary base or inflation are particularly relevant to this. I do think it would be a positive shock to real growth expectations.
I could leave it there, but unfortunately framing matters a great deal in economics and economic policy. And I think framing helicopter drops as ‘permanent’ is profoundly unhelpful – it is analytically unnecessary (even if defined in an internally consistent way, which I have yet to see), and practically confusing.
This selection from my exchange with Nick on his blog may help:
First Nick’s definition of a helicopter drop:
“Consider two policies:
A. I print $100 and give you $100. Period.
B. I print $100, and give you $100, and at the same time tell you (lead you to expect) that I will take $100 away from you next year, and burn it.
A is an outright gift of newly-printed money. It is helicopter money.
B is not a gift. B is an interest-free loan. The fact that you do not give me an IOU is irrelevant. The fact that I do not buy your IOU is irrelevant. If I know where to find you, and have no problems remembering that I lent you $100, and have no problems forcing you to pay me $100, I do not need little bits of paper with “IOU” written on them to help jog my memory and help me get the $100 back through the courts. But it is exactly as if I printed $100 and bought your one-year IOU for $100 with it. Or bought an IOU from you that someone else had signed. And bonds are IOUs.
B is an Open Market Operation, in which I print $100, buy $100 worth of bonds from you, and reverse that operation one year later. B is not Helicopter Money.”
EL: Nick – Let’s consider two regimes – (1) a depression, and, (2) steady growth at full employment. The ‘equilibrium’ stock of base money is much higher in the depression. Assume we are currently in regime (1) (or the Eurozone!), and the central bank says that it will print money and transfer it to households until it succeeds in shifting the economy to regime (2).
The central bank is also committed to bringing base money back to equilibrium – which really means it will be a permanent expansion if you stay in depression and a permanent reduction if you return to growth.
Given that this (probably) describes reality your permanent helicopter drop is only possible if it fails.
That doesn’t seem satisfactory.
So the answer must be as follows: a heli drop does not need to be permanent in order to work because in a depression people in receipt of cash payments really don’t care if they will be symmetrically taxed in a future boom. In fact, they would be quite happy with that.
NR: Eric: Helicopter Money only needs to be “permanent” *relative to the time-path the stock of base money would have followed otherwise*. It doesn’t matter if it’s permanent relative to where the current stock is now.
If it’s permanent in my sense, it has no debt or future tax implications.
Now, you might say that helicopter money in my sense is not needed, and that a temporary bond-financed transfer payment can do the job that’s needed, because people are borrowing-constrained for example. But that’s a different argument.
EL: Nick, in the example I have given, the stock of base money is lower than it would have been if the central bank had not pursued the helicopter drop.
We may be talking at cross-purposes here – you might have to be clearer about what exactly you mean by “relative to the time-path the stock of base money would have followed otherwise”. For example, the macro effects of a helicopter drop might raise the ‘equilibrium’ level of base money for very different reasons. So, for example, the stock of base money might be higher than it would have otherwise have been, but inflation lower, would this still be ‘permanent’ in your thinking? Or is ‘permanent’ fully defined by the intentions of the central bank and beliefs of the households? (If you define ‘permanent’ as ‘not expected by households to have any future tax implications’ – even if it does – I’d be fine with that.)
Part of my argument is undoubtedly due to uncertainty. We have *no* idea what “the time-path the of stock base money would have followed otherwise”, we don’t know what the impact of the helicopter drop is on the equilibrium amount of base money, and there are so many levers to affect demand for base money. Also, if the change in the stock of base money is not ‘permanent’ is it possible for the transfer to be ‘permanent’ if it has no future tax implications (or if the implications are lower future taxes).
So let’s say we cannot determine whether or not a cash transfer is ‘permanent’ – I can’t determine it, because I am not sure what you mean, and you can’t determine it because even though you know what it means, you can’t measure it. Do you think it makes any difference to the effect a cash transfer from the central bank to households would have?
Is it any different to a cash transfer by the government financed by debt which is bought by a central bank under a QE programme subject to an inflation target? It shouldn’t be … but framing seems to matter way more than logic when it comes to the national balance sheet, so in practice, ‘yes, it is’.
EL: I would also be interested to get your thoughts on the ‘contingent reversal’. There is surely a difference between being given $100 in one period, and having it taken back in the next, and being given $100 and having it taken back if your circumstances significantly improve.
NR: The clearest thought-experiment is the one I did in my old post, where there is a finite horizon, and all the money is redeemed in the year 2525 (remember that old song?).
Not sure if that helps.
But the current effects depend only on how it affects people’s current expectations of the future, not on what will actually happen in the future. If it is *perceived* as permanent then it is permanent.
The future affects the present only via people’s expectations of the future.
The contingent reversal is interesting. It is precisely those contingent policies that let central banks target things like inflation, the price level, NGDP.
NR: BTW: I do not know the time-path X(t). And I do not know the time-path X(t)+$100. But I do know that X(t)+$100 will be permanently higher than X(t).
EL: “The future affects the present only via people’s expectations of the future” – that is a very nice way of putting it. And that is really my point, I don’t think anyone in receipt of a check from the central bank would materially change their expectations of how this might change their future circumstances. If they were rational, empirically literate economists, they would expect the future to be brighter – that’s what all the evidence suggests. Beyond that, they have no relevant expectation – nor would they if the central bank said the change in monetary base was ‘permanent’. Identifying the extreme contingency of the relevant factors surrounding the future path of the monetary base – not to mention its impact on an individual’s circumstances – is another way of saying we have no relevant expectations in this regard.
Of course, the other way to square this circle is to say that individuals in receipt of check from the central bank will act as if it is ‘permanent’ having not really given it any thought … but I think we may be doing them a disservice!
I haven’t followed every twist and turn in this discussion, but I find it interesting, because my reaction has been the same as yours, roughly – although first in the case of how Paul Krugman has used the term “permanent” in a similar way on various occasions prior to this.
Permanence in base money projections is a slippery idea. There’s just too much going on otherwise.
First, the counterfactual scenario corresponding to the case without HD is not easy to identify.
Second, the future contingent response of the monetary authority under HD is not easy to identify. Nor is the marginal action of a growing currency base eating up bank reserve balances. Both have implications for the interpretation of permanence.
Third, there may be a contingent response associated with the fiscal authority.
As an unusual example of the last category, it would be possible for the fiscal authority to drain HD-created bank reserves by levying a tax and crediting the equity account of the central bank with the proceeds. That amounts to a partial/full recapitalization of what in accounting terms must have been an HD-created negative equity position. And it doesn’t require bonds. (It could also be done with bonds, at any time, which is the more usual case for capitalization/recapitalization.)
I think “contingency” is the most operative word.
As in contingent liability.
I think what I’m recalling is Krugman’s use of the term “permanent ” in connection with QE (in effect), rather than HD.
But the ambiguity problem is similar – particularly in view of the eventual dominance of growth in the stock of outstanding currency.
Totally agree JKH. It makes a difference in most models – but the models don’t map sensibly onto reality in this instance.
I think both authors miss an important point: money creation is yet an helicopter money for the banking system because the seigniorage of the capital is exposed as an actual fake liability in their balance. In the case of the central bank it is easy to spot that there is not a liability to anyone because today the National Treasury Departments are too stupid to understand that the seigniorage of the principal must be a liability in the book of the bank – yes- but finding the asset counterpart in the books of the state Treasury ! (the principal liquid asset in a sovereign country is the money the government can print out of DECREE).
While in commercial banks the fake liabilities are the deposits which should be segregated from the financial statement accounting (deposits must be kept out of the books).
Both the above accounting tricks (not recognizing the Treasury of the nation as the counterpart to which the banks are debtors) where designed to hide to the public the real seigniorage coming from money – i.e. everything that the bank get in exchange for the new money created.
The creation of irredeemable money – electronic money today – must be accounted for in the balances and in the cash flow statement of the banks as a newly created asset exchangeable for other assets of the market. Banks should not use the financial statement assuming that they don’t create the money they use and that they need the deposits which in reality they don’t use other than for book-cooking purposes…
So when above I read: “B is an Open Market Operation, in which I print $100, buy $100 worth of bonds from you, and reverse that operation one year later. B is not Helicopter Money.” you show that you really think that money is destroyed while only the accounting item is destroyed by the malpractice of the banks netting the cash flow account. The banks cash flow statement don’t take into consideration the hidden money creation process but assumes simply that the bank can overdraft on herself – while money thus created is circulating in the market forever. Or better, most of it is stored in secret offshore accounts as ‘shadow reflux money’ unaccounted for and missing in action. But ready to reappear in vulture funds to earn the free lunch nobody here is able to see….
Eric, I think I know where the problem lies.
First, let me define “helicopter money” (HM) as I see it. If you’re familiar with Adair Turner’s Overt Monetary Finance, that’s pretty much the way I understand HM (and I think Nick Rowe is with me here). HM should not be confused with traditional fiscal policy, although they have a lot in common: in both cases we can have “transfer payments” to households (crediting household bank accounts; tax cuts would of course have very much the same effect), and both show as an expenditure in government budget. Nothing new here. If we want to add some “windfall money” on households’ bank accounts, there is no need to use HM, or “helicopter drops”.
HM has nothing to do with the expenditure side. It’s all about how the budget deficit will be covered — and this is usually done by issuing government bonds. It is crucial to understand this, as this seems to be where much of the confusion lies.
What makes HM HM is that instead of government issuing bills and bonds as it always does, this time it would issue perpetual bonds with zero coupon. And the CB would buy these bonds (who else would?). The effect would be pretty much the same as if the government would mint, say, a small copper coin with $100 billion face value and sell it to the CB. The whole point is that the liability (a bond, or coin which is always a government liability) would not be any real liability — just a joke, explicitly.
The future path of the monetary base doesn’t matter here. What matters is that we know that some IOUs have been issued that don’t represent any real IOUs. And we know this because the one who issued those IOUs told us that the IOUs are worth nothing, and never will be. The whole point of HM is to be “a little bit reckless” (anti-Volckerian).
I’m happy to elaborate if something here is unclear.
Antti – what Adair Turner describes and you have outlined is nothing new (so no need to call it a ‘helicopter drop’). It is known as the monetisation of budget deficits. There is no need to introduce perpetual zero coupons either. The government just runs a deficit by creating base money – in the US this used to be administered by a treasury account at the Fed (I’m not sure of the current set up). Adair does not emphasis ‘permanence’ – wisely in my opinion. Nick, does – and I’m not convinced it is coherent or necessary.
What I am describing is very different, and new. The central bank makes direct money transfers to the household sector. I think it is an analytical error to deem this ‘fiscal policy’ in the abstract, and a legal/institutional category error in the Eurozone.
Before we even discuss helicopter money, we need to deal with a fundamental issue: What is money? I’ve read your text where you try to define money. If you stick to that definition, you cannot see helicopter money as I see it (or how Nick sees it, if I’m correct).
I hold you in high regard and I enjoy reading your posts. I often wonder how anyone can produce balanced blogs of such a high quality, every week. So I don’t say this lightly: you are wrong about money (though you’re in very good company).
In your definition of money, you write:
1. “Who cares if accountants and economists – wrongly – treat the money created by the government as a liability?”
2. “When, for example, I quizzed the renowned economist Paul Krugman, no less, on the implications for governments’ balance sheets of central banks buying government bonds by creating new money, he repeated accounting convention, taking it as assumed that money was a liability of the government”
(What follows is by no means common knowledge, so please read it carefully.) There is a difference between “money created by government being a liability” and “money being a liability of the government”. It all comes down to who is the ‘I’ in I.O.U. It is not “I, the government/CB” (we’re not in pre-revolutionary France). It should actually be W.O.U. in this case, as behind the ‘I’ are we, the people. Government issues IOUs on behalf of the people. Btw, I don’t think MMT does a good job in explaining how this all works, and I find them badly confused when they talk about net financial assets of the private sector — behind government spending lies a private sector liability, always (as I said, government issues IOUs on behalf of the people).
Likewise, cash or reserves — both of these are on the right-hand side of CB balance sheet — are not really central bank liabilities (or IOUs). But this doesn’t mean that there isn’t a real liability behind these. Have you ever considered what happens when the MBSs that the Fed now holds are repaid? Behind the MBS is a private sector liability, and someone in the private sector holds the corresponding asset (CB holds the MBS, the original asset, but CB is only a middleman). In this case the corresponding asset is a bank reserve (or cash; these are fungible, so we can never track down THE corresponding asset). When an MBS is repaid, both the private liability and a certain amount of reserves — the corresponding asset — are “destroyed”.
I could go on and on, but I’m afraid I’ve already lost you. So I’ll stop here for now, in hope we could find some common ground. This is very deep, fundamental stuff.
Thank Antti – I have written a lot on the issue of ‘is money a liability’. And it is v clear to me that it is not. You need to read my exchange with Randall Wray, and also the blog on the economics of language. Do you think language is a liability? If not – you don’t think money is! Btw, Nick Rowe & I are in broad agreement on this one. The example of the MBS being repaid makes no difference. Base money is the electronic equivalent of dollar bills. Dollar bills are ‘money’ not ‘debt’. The issuer of dollar bills repays nothing, the function of the dollar bill is distinct, as is the origin of its value. It really is a simple as that. Dollar bills and debts are completely different things. What’s interesting is why the confusion and why people don’t accept uniqueness of money.
Thanks for the replies, Eric!
I’ll get back to you later. Meanwhile, you might want to read my comment in Nick’s blog (I addressed it to both you and Nick):
I’ve gone through your discussion with Randy Wray. I found myself mostly in agreement with Randy. But I appreciate how hard it is for you to get on the same page with him. In this you’re right: you two don’t speak the same language.
Perhaps I’m wrong, but for an openminded, thoughtful guy like you, whose opinions most of the time seem very balanced, I find your position on “What is money?” very inflexible and dogmatic. You must know that the greatest economists throughout times have all been very humble in front of this question? It is a question that has remained unanswered in a fully satisfactory way.
Thanks Antti – that is a fair reprimand! It is important to have humility regarding these questions. However, I would distinguish between empirical issues or
policy recommendations, which are often very uncertain, and definitions – which should be clear. In engaging with Randy Wray, I don’t believe I am being dogmatic (in fact I agree with a lot of MMT). Rather, I am being clear about the meaning of “money”, “debt”, “liabilities” etc. The reaction of some MMTers can be defensive, when faced with clear definitions, because some of Randall Wray’s arguments are premised on obfuscation – as I am sure you could tell from our exchange.
Thanks again, Eric!
This is a huge subject, so let me try to focus on what I find most essential (we could call these “building blocks”): 1. Definition of “money” and “debt”. 2. Accounting.
I’ll start with the latter. In the comments to your previous post, you wrote me (when I asked which account the CB debits when it credits a deposit/reserve account): “When a CB buys a bond under QE – it creates a new deposit in the system. Money has not been transferred from one bank account to another. Same with a transfer to a household. There is no ‘debit’.”
I agree that money has not been transferred between two “bank accounts” (I assume that with “bank account” you mean “deposit account” or similar). But the CB has no other way to “create a new deposit” than to *credit* a reserve account. All this happens in the accounting realm. So, instead of talking about the CB creating “deposits”, we can talk about how the CB credits, or makes a credit entry on, an account. And if there is a credit, there must be a debit. You must be wrong if you think otherwise.
As far as I know, when the CB buys a government bond, it makes a debit entry on an account called “Treasury securities held outright” (or similar). This account is an asset account and a debit entry to an asset account always increases the balance. Guess what happens when a commercial bank buys a government bond from an investor who happens to have an account in the same bank? The bank credits the investor’s deposit account and debits an asset account (the name of the account might be “Investment securities” or similar, not far from the Fed’s account naming convention). No money was transferred between two bank accounts in this case, either.
So, if you suggest that the CB should credit an account (a reserve account/commercial bank’s deposit account with the CB), then you must be able to tell us what kind of an account the CB should debit. An account is an account, whether you call it “bank account” or not. The CB doesn’t just “create deposits”. “Creating deposits” is just the credit entry part of a transaction which requires two entries (hence, “double entry”). There is no “deposit creation” which happens outside the accounting realm.
Perhaps we could get this clear before we move to the definition of “money” and “debt”?
Antti – I think you need to read this before we can proceed. Also, have a look through the discussion.
I read that already two days ago, Eric.
I actually find myself in a fortunate position between you and Wray. When you say that money is not debt/an IOU/a liability, I’m actually ready to agree with you. As I’ve said elsewhere, I don’t see “base money” as a liability of the CB, nor government. To me, “base money” is a credit in CB books (this is a fact that can be confirmed by looking at the actual accounting). If the total nominal value of credit balances (claims) is X, then we must have somewhere debit balances (liabilities) for the same amount X. In the case of QE, commercial banks’ credit balances (“reserves”) are increased in the CB books, and we know that there are entities whose debit balances, liabilities, are increased in the same CB books (this can be Treasury or, for instance, an SPV behind an MBS). Like I’ve said, the CB must *debit* some account, and that debit to an account will either increase some entity’s liabilities recorded in the CB books, or alternatively decrease some entity’s — other than the CB — credits recorded in the CB books.
To get back to my previous comment, I’d still like to hear which account will be debited when the CB credits a “deposit account” (say, as part of People’s QE)? When I know this, I will know who are the agents whose liabilities will be increased (alternatively, credits decreased) — that is, who is supposed to ultimately bear the cost of this distribution of “windfall credits” on certain agents’ accounts.
(Despite my being an accountant, I hope you realize that my aim is not to prove any “accounting point”. I’ve built a whole macroeconomic theory on the premises I’m trying to explain here.)
It is easy to talk at cross purposes here, because people use terms differently. It is often helpful to start with physical cash. After all, reserves are the electronic equivalent. So, lets say all the base money is cash – so the CB has bought the bonds from the public under QE with physical cash. Let’s also assume the public holds all the new cash and doesn’t deposit it in the banking system. My main point is that whatever accounting system is used to describe this process must recognise the uniqueness of money, or it fundamentally misses something of huge importance.
Eric, what are you trying to prove? Why can’t you just answer my question about which account is going to be debited? It doesn’t matter if it’s cash or reserves we are talking about. If notes instead of reserves are issued, then the CB credits the account “Notes in circulation”.
And have you ever considered the possibility that reserves are not strictly speaking an electronic equivalent of cash? That “reserves” are instead just part of “pure accounting” whereas physical notes bring an additional layer to this accounting (a tool which serves an accounting purpose)? There is no “electronic cash” on accounts. No “electronic money” is transferred between accounts. Credit and debit entries are made, that’s all. Perhaps it’s been a mistake all along to think in terms of cash, when we should have been thinking in terms of accounting, pure and simple.
Antti, my entire argument is that accounting is convention, and the conventions applied to central banks are v misleading. I have shown how they are in part accidents of history and convenience. The important point is that a central bank can create money at will, and money is not a debt. The actual accounting treatment is not that important and could be re-written. In the case of QE, central banks are removing bonds and creating money. You and I can discuss how this *should* be treated in accounting, how it *is* treated does not change the nature of what is occurring.
Eric, what is occuring is that the CB makes a credit entry on an account in its bookkeeping system. This is what is observable. There is no other “money creation” occuring, and so it cannot possibly be that the nature of what is occuring differs from the accounting entries made. By convention, you call the creation of this credit entry “money creation”.
The CB cannot “create money”, i.e. make a credit entry on a reserve/”notes in circulation” account, without making a debit entry on another account. And the effect of this debit entry is either to increase the LHS or decrease the RHS of the CB balance sheet. On the LHS of the balance sheet there are mainly public and private liabilities and on the RHS there are public and private claims. Thus, making the debit entry either increases public or private liabilities or decreases public or private claims. And as I said earlier, I don’t view these non-CB liabilities as liabilities to the CB, nor do I view the claims as claims against the CB. Why don’t I? Because if I did, I would be stuck and couldn’t create a consistent picture of how the monetary system is linked to the real economy.
Eric, to really understand the monetary system you need to *embrace* accounting — not do the opposite. And you need to let go of “money”, no matter how unique and concrete, and comfortable, it seems — it’s only an illusion.
Dear Mr. Lonergan,
what a great article and subject. However, I respectfully disagree with your assessment. ‘Helicopter money’ (A) and ‘free lending to the masses’ (B) is extremely different.
We are broadly speaking in a liquidity trap. The last thing we need is another name for the option to borrow more money. The term ‘borrowing’ would point to the necessity of repayment. It would, /if/ it was spend, be used for long-term items that retain value (Housing!). The borrower would be prudent to do this, as he has to repay the money at some point. Ideal for blowing another housing/gold/stock bubble. Vitally, it will /not/ kickstart consumer confidence or non-investment spending.
The prospect of future costs/repayments is /exactly/ why people don’t borrow and spend, and /exactly/ the reason why we are in the deflationary quagmire. Helicopter money should be one-way, highly visible, non-structural and in smaller size than QE.
You can be fearful about helicopter money getting out of hand, but if I see the chaos created by QE, I think we have chosen the wrong route all along