The monetary & fiscal distinction revisited
Monetary policy involves changes in the supply of base money and is carried out by central banks. Fiscal policy involves taxation, spending and transfers carried out by national treasuries.
There are important and subtle distinctions which apply to these terms. Importantly, terms such as ‘tax’ and ‘transfer’ are not just purely economic concepts – they have legal and institutional characteristics too. For example, not paying an IOR on required reserves is often described by economists as a ‘tax’ on banks. In a purely economic sense this may be true. But legally it is not considered a tax.
Economists are rarely acutely focused on these distinctions. They are increasingly relevant – and we may be found wanting. Such distinctions are central to the next frontier of macroeconomic policy – the policies which will prevail when QE and lower policy rates become exhausted. We may already be there.
If you’re bored of the helicopter money debate, I apologise. It may help to think of it like this: how should we organise monetary and fiscal policy when further reductions in official interest rates, asset purchases and forward guidance have reached sensible limits? Colourful metaphors aside, that is the heart of the matter. If ‘helicopters’ upset you – think of it as post-interest rate monetary policy.
The somewhat illusive Mr JKH, who writes this blog, recently suggested that I am being dogmatic in insisting that helicopter money is ‘monetary policy’. As dogmatic, in fact, as those who say it is ‘fiscal’.
My position is that there does not need to be a fiscal component of helicopter money, but if there is one, it depends on the jurisdiction, and on the specific institutional arrangements used to implement it. In considering helicopter money, there are many important institutional choices to make.
Helicopter money can be purely monetary policy (and I suspect that this is institutionally optimal), but it could also be a combined action between the treasury and central bank. For example, under the definition of helicopter money as base money-financed transfers to the private sector which are not aimed at altering market interest rates, any central bank which pays different interest rates on different categories of reserves can engage in a helicopter drop. Consider a central bank which pays a higher interest rate on required reserves than on excess reserves, which affect market interest rates. The central bank can raise the interest rate on required reserves to whatever level it chooses and make a transfer to the banking sector. This is a purely monetary helicopter drop – a base-money financed transfer to the private sector using the legal and institutional tools of the central bank, something the Bank of Japan is already doing. (To argue that this has ‘fiscal effects’ in an abstract theoretical sense is not relevant – all monetary policy changes have fiscal effects, this does not make them fiscal policy – this is the ‘Bundesbank fallacy’).
Alternately, if a central bank agrees to coordinate with the fiscal authorities and says it will initiate a QE programme to purchase all of the bonds involved in a new fiscal stimulus, or will directly finance the new fiscal stimulus, this is clearly a combined monetary and fiscal operation (some economists will argue that a commitment about the future by the central bank is required to make this a ‘helicopter drop’, but that is a separate point).
Similarly, if by Act of Parliament, the UK government gives the Bank of England a new power to credit households with money financed by reserve creation in order to meet its inflation target, this would be deemed monetary policy in future. Is this delegating a combined fiscal-monetary operation to the central bank? I’m not sure, but at a certain point the operational distinction cedes to one of convention.
It seems clear then that helicopter money is always monetary policy – to the extent that it involves base money and, at a minimum, the cooperation of the central bank. Sometimes it is coordinated monetary and fiscal policy. This, in fact, is implicit in Milton Friedman’s 1968 AER address, where he points out that monetary policy is never redundant.
So we end up observing that the precise legal and institutions arrangements of future monetary policy will vary by jurisdiction; and within current constraints, different approaches are likely.
I suspect Bernanke’s institutional arrangement for the US, may be the only way the Fed can do helicopter drops – to households, at least. Here is what he suggests:
“So, how could the legislature and the central bank play their appropriate roles in managing a joint monetary-fiscal operation, without endangering central bank independence or falling down a slippery slope of unconstrained monetary finance of fiscal spending or tax cuts? A possible arrangement, set up in advance, might work as follows: Ask Congress to create, by statute, a special Treasury account at the Fed, and to give the Fed (specifically, the Federal Open Market Committee) the sole authority to “fill” the account, perhaps up to some prespecified limit. At almost all times, the account would be empty; the Fed would use its authority to add funds to the account only when the FOMC assessed that an MFFP [money-financed fiscal policy] of specified size was needed to achieve the Fed’s employment and inflation goals.”
I’m not convinced that this is an optimal arrangement for the Fed, because monetary policy’s effectiveness becomes contingent on the partisan and ideological make up of Congress. But it’s a reasonable suggestion and certainly better than nothing.
The institutional framework Bernanke proposes, is explicitly illegal in Europe. In the Eurozone, as I have argued elsewhere, helicopter drops can and must be purely monetary policy. The ECB will have to use its direct lending powers and operate without any involvement of national Treasuries. The UK, has yet to decide which way it will go – and the Bank of England should contribute more to this debate. It has the tools and institutional structure to take either, or both, approaches.
In conclusion, the distinction between monetary and fiscal policy can be carefully made. The next phase of macroeconomic policy will depend more heavily on legal, political and institutional arrangements. Its likely implementation will vary by jurisdiction, and dogmatic categorisation is both unhelpful and, in the main, false.
I agree that institutions matter. And words and definitions should be used carefully for framing and communicating the policy. That’s why I think it is helpful to sort out the use of the terms fiscal policy and monetary policy as best as possible. Some influential types like Michael Woodford use fiscal policy in the description of helicopter money from the get go. So the issue of semantics can’t really be avoided. If connection between fiscal policy and monetary policy is made as clear as possible, all the better.
If helicopter money ends up being done in a notable way, the effectiveness of it will have something to do with its announcement power and that the central bank is seen to have the authority to execute policy with something quite novel in its “tool-kit”. It is in that sense that I believe there is at least potential for it to have a greater effect than say – for example – an economically equivalent policy (some would say) of spending or transfer payments financed by Treasury bill issuance.
Central bank operations have a fiscal effect in their conventional scope. This arises from the profit (or loss) contribution resulting from a CB’s financial intermediation activity.
Unconventional operations such as qualitative or quantitative easing expand the scope of asset activity and increase the size of the monetary base beyond conventional trends.
In general, it is the asset and liability classes involved in these conventional and unconventional policies that define the scope of permissible financial intermediation activity.
Helicopter money differs from these conventional and unconventional monetary policies in the timing relationship between government spending or transfers and corresponding money financing. With either OMO or QE, , there is obviously a requirement for some type of standard fiscal action to have been undertaken in the past so as to create the budget deficit conditions for the issuance of government debt that is available to purchase from the market place. With QE, money financing supports spending or transfers that are essentially concurrent with the financing.
The proposals generally floated for helicopter money include several variations on operational arrangements. The one that seems to involve the central bank for the most part is the example of transfer payments made by a CB directly to households without operational involvement by Treasury. The payments are directly monetized to bank money and bank reserves. A different route involves Treasury at the operational level, with the central bank directly crediting the Treasury account (i.e. without the intermediation of bonds), from which Treasury spends or transfers funds that are subsequently monetized in the form of private sector bank money and bank reserves. A third option includes a central bank that acquires government debt newly issued to finance concurrent payments from the Treasury account, with those payments subsequently monetized in the form of bank money and bank reserves. This last option can be viewed as the logical limit of QE – the limit as the time lag between Treasury debt financing and central bank purchase of that debt goes to zero.
A consistent timing nexus is evident in all of these proposals. It is perhaps most clearly delineated in the third case of concurrent bond and money creation – seen as the limit of QE as the time lag between fiscal and monetary operations goes to zero. But in all three alternatives the same relationship holds between the timing of a standard type of fiscal action such as spending or transfer payments and the timing of money creation – it happens at essentially the same time. There is no lag. I think this timing consideration is the key to delineating a natural fiscal and monetary policy nexus for helicopter money. If fiscal policy is defined to include spending or transfers, and if monetary policy is defined to include monetary base creation, then HM has a dual fiscal and monetary policy characteristic.
Where does this leave us in terms of framing of fiscal and monetary policy vocabulary in the case of helicopter money?
I’m not a fan of the “already doing it” interpretation of standard central bank activity as it relates to helicopter money. For example IOR (including recently implemented tiered structures in a negative interest rate setting) is a matter of the interest expense structure, which is a subset of the profit determination which in turn becomes the standard central bank fiscal contribution in a structural sense. While it may be tempting to attach economic interpretations of tax-like properties in some cases, this is a structure that is determined by the central bank and therefore is part of monetary policy proper. Moreover, the motivation for the expense structure has much to do with operational requirements corresponding to interest rate policy objectives for the central bank. The same technical argument cannot be made for helicopter money – other than in a somewhat indirect way. Therefore, I see no persuasive interpretation of IOR as a precedent for central banks assuming the helicopter file as core monetary policy.
I am sympathetic to the idea that – if the central bank does it in fact – it’s monetary policy. Monetizing either financial assets or standard concurrent fiscal payments is monetary policy in that sense, be that OMO, QE, or HM. But that’s only one side of the transaction. The other side is the fiscal characteristic of the act of spending or transfer. And if it turns out that the two acts are coincident – as in the case if a central bank were to make a standard type of transfer payment to a household – then that coincidence is one of fiscal and monetary policy. And the central bank in that case executes both fiscal and monetary components. The policy implementation responsibility is a composite one – including a received delegation of authority to act in a standard fiscal capacity, which is clearly different than is the case for the standard financial intermediation contribution of the central bank.
However, such a delegated HM authority has a natural origin in an overarching fiscal policy framework that ideally should include the formulation of an overall spending and transfer capacity – adjusted appropriately for the risk implications of corresponding financing mix – with the central bank being delegated an authority to take on some of that capacity in the form of a dual-policy helicopter mandate. Such a “risk allocation” approach puts fiscal policy at the heart of HM.
a comment I made elsewhere regarding the relationship between HM authority and CB independence:
As a non-economist who has advised many executives and policy makers on my own technical specialism, I am bemused by these debates. Let me say what I see when I read them (not just on this blog, so my comments are not just about Eric’s view).
Firstly, economists do not appear to agree on the definitions of basic terms such as ‘monetary policy’ and ‘fiscal policy’ even though these terms were devised by and for economists. This is absurd and immediately makes non-economists wary of everything that follows. Economists discuss virtually everything in terms of badly-defined abstractions. They then declare war on each other when they talk past each other due to inconsistencies in the definitions of these abstractions.
Secondly, the term ‘helicopter money’ itself is badly-defined. It is obvious even to a non-economist that there are many variants of helicopter money. That means that you can have an unambiguous discussion only if you talk about each variant separately. That means you have to start by itemising, naming and defining each variant. That means that you need to ask what specification of a variant would be sufficient to make these definitions unambiguous. However, no-one does this and the result is endless repetitive discussion with everyone talking at cross-purposes and with no-one listening to anyone else.
In a recent post, Eric says that all of the following are helicopter money (my simplification of Eric’s wording):
The central bank changes the interest rates on reserves.
The central bank credits households with money.
The central bank buys government bonds used to finance government stimulus.
These are all different policies. Giving them the same name is daft. I might have been dreaming (a nightmare really) but I seem to recall a recent announcement by the governor of the BoE where he ruled out helicopter money. An economist then declared the announcement a victory for the proponents of helicopter money. The governor and the economist obviously meant different things by helicopter money and, as a result, appeared to contradict each other. Why does the economics profession think that anyone in wider society is going to give credence to this type of engagement?
Thirdly, Eric often writes that some economists say that ‘money is debt’ and he disagrees. However, ‘money is debt’ is a slogan rather than a definition. Eric blames accountants and others for the fact that (many) economists use such slogans and have not defined money in a way which fits with the real world, is consistent with accounting and can be agreed by all parties (by finance practitioners, policy makers and the general public as well as by economists). This is hopeless.
Economics has aspirations to be regarded as a science. However, science demands clear definitions before anything else in order that there can be fruitful debate.
And note that these comments don’t even begin to address the core issue here – which is about the limitations of the term ‘independent’ when applied to a central bank. Even if a central bank is independent of the government, the government is also (Euro area excepted) independent of the central bank. Economists like to think that central banks can constrain elected governments by adopting policies which subvert those of ‘misbehaving’ governments. However, as elected governments are also independent of central banks, they can subvert any central bank plan by making any central bank action illegal or by firing the governor. Central banks are independent only with the explicit agreement of the government, which is really no more independent than a traffic warden with delegated authority to fine illegal parking.
A final point. In the variant where ‘the central bank credits households with money’, there is an operational issue that neither the central bank, nor any of the commercial banks, has an inventory of all of the people in the country. There is no way that this policy can operate without such an inventory (unless you accept that people with accounts in multiple banks should receive multiple credits, and that people with no current bank account should receive nothing – which would not be acceptable politically and would also encourage fraud). The development of an effective inventory would cost billions and take years. No government would fund such a development when it already has infrastructure which is adequate for the distribution of money to the public – namely the tax and benefits systems. This is particularly true for a policy which operates only once on a blue moon when all other policies have failed.
Hi Jamie – I think we agree on the definitions, that is significantly the point being made in this post. I think economists should define ‘helicopter money’ ‘fiscal policy’ ‘monetary policy’ ‘debt’ and ‘money’, as clearly as possible, otherwise they will inevitably talk at cross purposes. I think however, that I have very clearly outlined the two types of policies which fall under the label ‘helicopter money’. I have also been very clear about the distinction between monetary & fiscal policy. I have also written several posts on the clear conceptual and empirical differences between ‘debts/liabilities‘ and ‘money‘. That said, there are also methodological challenges which are not trivial.
As regards, the ‘is economics a science?’ discussion, I have never heard anything insightful on this issue. Let’s just get on with the economics!
Finally, administering cash transfers to adult citizens: I think you’re too pessimistic. I have suggested using the banks – for example if all adult citizens are eligible for five hundred euros, banks would have to ensure citizenship and avoid fraud (multiple claims) – they already do this with anti-money laundering rules.
[…] is a leading proponent of HM. As he points out, the resolution of this issue could depend on the institutional regime for the central bank in question. For example, it may be quite different in the case of the ECB […]
How would you categorise a helicopter drop of bills by the central bank, either out of existing holdings or from stock borrowed (not purchased) from the national treasury? That doesn’t seem to fit either of your definitions.
It’s worth bearing in mind that the term ‘helicopter drop’ doesn’t really have clear definition. I have simply discerned two types of policy which people seem to describe as helicopter drops: i.e. transfers to the private sector financed by the CB, and the monetisation of increased fiscal deficits aimed at shifting beliefs. I am following the Wittgensteinian dictum of meaning residing in use! Now, I have not seen widespread advocacy of the policy you describe, nor anyone describing it as a helicopter drop. To that extent, it is not. You are, I suspect, raising a more interesting question – would its effects be any different? I think that depends. It might well make a difference in Greece. In the UK, it might not – although a lot of people wouldn’t be able to cash in gilts. I am very open to the view, however, that lot’s of policies look economically equivalent – but they remain importantly different, paticularly if repeated, and the similarities are often contingent. I hope that answers your question …
As an aside, I have advocated that CBs convert their bond holdings into equities and in fact distrubute them to the lowest 80% of the income distribution, as a sensible policy to tackle inequality – but I suggest structuring this so it is perceived as “wealth” rather than “income” (largely a matter of framing). Policies for tackling inequality Radical policies that might work.
Actually I was more interested in tying down your conceptions of fiscal policy and monetary policy. Those were the two definitions I was referring to. Sorry – that probably wasn’t clear.
That said I don’t personally think it very important whether helicopter drops are described as fiscal, monetary or a combination of both. The far more important question is about the effect and I agree that it is an interesting (and useful) question, whether what I describe would have very different effects from a normal helicopter drop.
As far as the UK goes, I suspect it would make little difference in current circumstances. It’s true that many households would find it awkward to dispose of gilts but of course in a regular helicopter drop they wouldn’t be given reserves. What would happen is that customer accounts at commercial banks would be credited and the central bank would transfer reserves (or in my scenario, bills) to those commercial banks. The only difference then would be in the assets held by the banks. At the moment, commercial banks are not holding marginal reserves for clearing purposes, but purely to meet liquidity coverage requirements. For this purpose, reserves and bills are pretty much equivalent.
Interestingly though, the PRA has recently made a change to exclude reserves from Leverage Ratio calculations. This brings back a distinction between reserves and bills although arguably this works in the other direction, making an injection of bills potentially more expansionary than an injection of reserves.
I think I understand Nick. To clarify: Can CBs ‘pay’ commercial banks with bills? Or under the process you are describing, initially bank deposits of households rise & simultaneously commercial bank reserves at the CB; then the CB issues bills to banks of an equivalent amount?
Either way, I’m not sure it would alter my thinking. I have made base money a defining characteristic of monetary policy and what distinguishes it from fiscal policy. If you are suggesting that bills & base money can become blurred, that is undoubtedly true. But I don’t think it undermines the distinction. If we were to give bills all of the functions of base money, bills would be base money and if the treasury issues the bills it would be doing monetary policy. Of course, in some jurisdictions, we also have bills issued by the CB …
It is clear, though, that there is a spectrum which has very distinct phenomena at the extremes: ‘pure’ money at one end and long-dated Greek govt bonds at the other, but the distinction starts to blur – at least in most developed economies as we approach one extreme – base money.
That is not grounds for equivalence: Paying IOR looks a bit like having the ability to issue bills, having bills contribute to reserves makes them more like money. But none of this shows that money is not a distinct thing, as is government debt (somewhat blatantly in the Eurozone). Most advanced economies want the creation of money to be separate to the issuance of government debt – I think with good reason. This is because the functions and phenomena (money, govt debt) are distinct, and the institutions which control both have different objectives and constraints.
I don’t know if that makes sense …
“Monetary policy involves changes in the supply of base money and is carried out by central banks.”
What if the central bank promises to convert demand deposits into currency at fixed exchange rate (just happens to be 1 to 1)?
I don’t think that changes anything.
“Safe assets are bunds, treasuries, gilts, JGBs and other government bonds issued by large developed economies.”
This from a different blog post. I could not find a way to post there. Did I miss it?
Also, aren’t currency and insured demand deposits safe assets?
True – list wasn’t meant to be comprehensive. But fair point!