So-called Modern Monetary Theory (MMT) has been attracting a great deal more attention than usual this year. I guess that isn’t overly surprising, in view of (a) the severe recession the world is now in, and (b) the passivity and inaction (and the ineffectiveness of what actions they do take) of central banks, those with day-to-day responsibility for the conduct of monetary policy.
Until about three years ago I had had only the haziest conception of what the MMTers were on about. But then Professor Bill Mitchell, one of the leading academic (UNSW) champions of MMT ideas, visited New Zealand, and as part of that visit there was a roundtable discussion with a relatively small group in which I was able to participate. I wrote about his presentation and the subsequent discussion in a post in July 2017. I’d still stand by that. (As it happens, someone sent Mitchell a link to my post and he got in touch suggesting that even though we disagreed on conclusions he thought my representation of the issues and his ideas was “very fair and reasonable”.) But not many people click through to old posts and, of course, the actual presenting circumstances are quite a bit different now than they were in the New Zealand of 2017. Back then, most notably, there was no dispute that the Reserve Bank had a lot more OCR leeway should events have required them to use it.
Among the various people championing MMT ideas this year, one of the most prominent is the US academic Stephanie Kelton in her new book The Deficit Myth: Modern Monetary Theory and How to Build a Better Economy (very widely available – I got my copy at Whitcoulls, a chain not known for the breadth of its economics section). Since it is widely available – and is very clearly written in most places – it will be my main point of reference in this post, but where appropriate I may touch on the earlier Mitchell discussion and this recent interview on interest.co.nz with another Australian academic champion of MMT ideas.
As a starting point, I reckon MMT isn’t particularly modern, is mostly about fiscal policy, and is more about political preferences than any sort of theoretical framework (certainly not really an economics-based theoretical framework). But I guess the name is good marketing, and good marketing matters, especially in politics.
The starting proposition is a pretty elementary one that, I’d have thought, had been pretty uncontroversial for decades among central bankers and people thinking hard about monetary/fiscal interactions: a government with its own central bank cannot be forced – by unavailability of local currency – to default on its local currency debt. They can always “print some more” (legislating to take direct control of the central bank if necessary). So far so good. But it doesn’t really take one very far, since actual defaults are typically more about politics than narrow liquidity considerations and governments may still choose to default, and the actual level of public debt (share of GDP) maintained by advanced countries with their own currencies varies enormously.
A second, and related, point is that governments in such countries don’t need to issue bonds – or raise taxes – to spend just as much as they want, or run deficits as large as they want. They can simply have the central bank pay for those expenses. And again, at least if the appropriate legislation was worded in ways that allowed this (which is a domestic political choice) then, of course, that is largely true. That means governments of such countries are in a different position than you and I – we either need to have earned claims on real resources, or have found an arms-length lender to provide them, before we spend. Again, it might be a fresh insight to a few politicians – Kelton spent a couple of years, recruited by Bernie Sanders, as an adviser to (Democrat members of) the Senate Budget Committee, and has a few good stories to tell. But to anyone who has thought much about money, it has always been one of the features – weaknesses, and perhaps a strength on occasion – of fiat money systems.
Kelton also devotes a full chapter to the identity that any public sector surplus (deficit) must, necessarily, mean a private sector deficit (surplus). Identities can usefully focus the mind sometimes in thinking about the economy, but I didn’t find the discussion of this one particularly enlightening.
It all sounds terribly radical, at least in potential. One might reinforce that interpretation with Kelton’s line that “in almost all instances, fiscal deficits are good for the economy. They are necessary.”
But in some respects – at least as a technical matter – it is all much less radical than it is sometimes made to sound. As a matter of technique and institutional arrangements, it is mostly akin to “use fiscal policy rather than monetary policy to keep excess capacity to a minimum consistent with maintaining low and stable inflation”. Supplemented by the proposition that advance availability of cash – taxes, on-market borrowing – shouldn’t be the constraint on government spending, but rather that the inflation outlook should be.
Quoting Kelton again “it is possible for governments to spend too much. Deficits can be too big”.
What isn’t entirely clear is why, as a technical matter, the MMTers prefer fiscal policy to monetary policy as a stabilisation policy. In the earlier discussion with Bill Mitchell, it seemed that his view was the monetary policy just wasn’t as (reliably) effective as fiscal policy. In Kelton’s book, it seems to reflect a view that using monetary policy alone there is inescapable sustained trade-off between low inflation and full employment (a view that most conventional macroeconomists would reject), and that only fiscal policy can fill the gap, to deliver full employment. Kelton explicitly says “evidence of a deficit that is too small is unemployment” – it seems, any unemployment, no matter how frictional, no matter how much caused by other labour market restrictions.
I can think of two other reasons. The first is quite specific to the current context. Some might prefer fiscal policy because they believe monetary policy has reached its limits (some effective lower bound on the nominal policy rate). Kelton’s book was largely finished before Covid hit – and US rates at the start of this year weren’t super-low – but it seems to be a factor in the current interest in MMT. The other reason – not really stated, but sometimes implied by Kelton – is that central bankers might have been consistently running monetary policy too tight – running with too-optimistic forecasts and in the process falling down on achieving what they can around economic stabilisation. Since 2007 I’d have quite a bit of sympathy with that view – although note that in New Zealand prior to 2007 inflation was consistently too high relative to the midpoint of the target ranges governments had set. But it is, at least initially, more of an argument for getting some better central bankers, or perhaps even for governments to take back day-to-day control of monetary policy, than an argument for preferring fiscal policy over monetary policy as the prime macro-stabilisation tool.
But in general there is little reason to suppose that fiscal policy is any more reliably effective than monetary policy. Sure, if the government goes out and buys all the (say) cabbages in stock that is likely to directly boost cabbage production. If – in a deep recession – it hires workers to dig ditches and fill them in again that too will directly boost activity. But most government activity – taxes and spending (and MMTers aren’t opposed to taxes, in fact would almost certainly have higher average tax rates than we have now) – aren’t like that. If it is uncertain what macro effect a cut in the OCR will have, it is also uncertain how – and how quickly – a change in tax rates will affect the economy, and even if governments directly put money in the pockets of households we don’t know what proportion will be saved, and how the rest of the population might react to this fiscal largesse. In principle, there is no particular reason why fiscal policy should be better, as a technical matter, than monetary policy in stabilising economic activity and inflation. But Kelton just seems to take for granted the superiority of fiscal policy, and never really seems to engage with the sorts of considerations that led most advanced countries – with their own central banks, borrowing in local currencies – to assign stabilisation functions to monetary policy, at arms-length from politicians, while leaving longer-term structural choices around spending and tax to the politicians.
These probably shouldn’t be hard and fast assignments. In particular, there are some things only governments (fiscal policy) can do. Thus, if an economy largely shuts down – whether from private initiative or government fiat – in response to a pandemic, monetary policy can’t do much to feed the hungry. Charity and fiscal initiatives are what make a difference in this very immediate circumstances – just as after floods or other severe natural disasters. And we consciously build in some automatic stabilisers to our tax and spending systems. But none of that is an argument for junking monetary policy completely, whether that monetary policy is conducted by an independent agency, or whether such agencies (central banks) just serve as technical advisers to a decisionmaking minister (as, for example, tended to be the norm in post-war decades in most advanced countries, including New Zealand).
The MMTers claim to take seriously inflation risk. This is from the Australian academic interest.co.nz interviewed (Kelton has very similar lines, but I can cut and paste the other)
“They should always be looking at inflation risk. Because when we say that our governments can never become insolvent, what we are saying is that there is no purely financial constraint that they work under. But there is still a real constraint. So New Zealand has a limited productive capacity. Limited by the labour and skills of the people and capital equipment, technology, infrastructure and the institutional capacity of business organisations and government in New Zealand. That limits the quantities of goods and services that can be produced there is a limitation there. Also it depends on the natural resources of a country,” says Hail.
“If you spend beyond that productive capacity it can be inflationary and that can frustrate your objectives, frustrate what you’re trying to do. So it’s always inflation risks that’s important. Within that productive capacity, however, what it is technically possible to do the Government can always fund. So yes, you can fund any of those things but there’s always an inflation risk and that inflation risk is not specific to government spending. It’s specific to all spending.”
There is a tendency to be a bit slippery about this stuff. Thus Kelton devotes quite some space to a claim that government spending/deficits can’t crowd out private sector activity. And she is quite right that the government can just “print the money” – so in a narrow financing sense there need not be crowing out – but quite wrong when it comes to the real capacity of the economy. Real resources can’t be used twice for the same thing. When the attempt is made to do so, that is when inflation becomes a problem – and the MMTers aver their seriousness about controlling inflation (and I take them at their word re intentions).
Partly I take them at their word because Kelton says “the economic framework I’m advocating for is asking for more fiscal responsibility from the federal government not less”. And it certainly does, because instead of using monetary policy, the primary stabilisation role would rest with fiscal policy. That might involve easy choices for politicians flinging more money around to favoured causes/people in bad times, but it involves exactly the opposite when times are good, resources are coming under pressure, and inflation risks are mounting. Under this model, a government could be running a fiscal surplus and still have to take action to markedly tighten fiscal policy because – in their own terms – it isn’t deficits or surpluses that matter but overall pressure on real resources. And they want fiscal policy to do all the discretionary adjustment.
Maybe, just maybe, that is a model that could be made to work in (say) a single chamber Parliament, elected under something like FPP, so that there is almost always a majority government. Perhaps even in New Zealand’s current system, at a pinch, since to form a government the Governor-General has to be assured of supply.
But in the US, where party disciplines are weak, different parties can control the two Houses, and where the President is another force completely. What about US governance in the last 30 years would give you any confidence in the ability to use fiscal policy to successfully fine-tune economic activity and inflation, while respecting the fundamental powers of the legislature (no taxation without representation, no expenditure without legislative appropriation)? In a US context, I’m genuinely puzzled about that. [UPDATE: A US commentator on Twitter objected to the use of ‘fine-tune” here, suggesting it wasn’t what the MMTers are about. Perhaps different people read “fine-tune” differently, but as I read MMTers they are committed to maintaining near-continuous full employment, and keeping inflation in check, and even if some like rules – rather than discretion – it seems to me frankly no more likely that preset rules for fiscal policy would successfully accomplish that macrostabilisation than preset rules for monetary policy did. “Successfully managed discretion” is what I have in mind when talking about “fine-tuning” in this context.]
But even in a relatively easy country/case like New Zealand using fiscal policy that way doesn’t seem at all attractive. It takes time to legislate (at least when did properly). It takes time to put most programmes in place, at least if done well – and don’t come back with the wage subsidy scheme, since few events will ever be as broad-brush and liberal as that, especially if fine-tuning is what macro-management is mostly about. And every single tax or spending programme has a particular constituency – people who will bend the ear of ministers to advance their cause/programme and resist vociferously attempts to wind such programmes back. And there are real economic costs to unpredictable variable tax rates.
By contrast – and these are old arguments, but no less true for that – monetary policy adjustments can be made and implemented instantly. They don’t have their full effect instantly, but neither do those for most fiscal outlays – think, at the extreme, of any serious infrastructure project. And monetary policy works pretty pervasively – interest rate effects, exchange rate effects, expectations effects (“getting in all the cracks”) – which is both good in itself (if we are trying to stabilise the entire economy) and good for citizens since it doesn’t rely on connections, lobbying, election campaign considerations, and the whim of particular political parties or ministers. And what would get cut if/when serious fiscal consolidation was required? Causes with the weakest constituencies, the least investment in lobbying, or just causes favoured by the (at the time) political Opposition. Perhaps I can see some attraction for some types of politicians – one can see at the moment how the government has managed to turn fiscal stabilisation policy into a long series of announceables for campaigning ministers, rewarding connections etc rather than producing neutral stabilisation instruments – but the better among them will recognise that it is no way to run things. It is the sort of reason why shorter-term stabilisation was assigned to monetary policy in the first place.
Reverting to Kelton, her book is quite a mix. Much of the first half is a clear and accessible description of how various technical aspects of the system work, and what does and doesn’t matter in extremis. But do note the second half of the book’s title (“How to Build a Better Economy”): the second half of the book is really an agenda for a fairly far-reaching bigger government – (much) more spending, and probably more taxes. There is material promoting lots more (government) spending on health, welfare, infrastructure, and so on – all the sort of stuff the left of the Democratic Party in the USA is keen on.
That is the stuff of politics, but it really has nothing at all to do with the question of whether fiscal or monetary policy is better for macro-stabilisation. I guess it may be effective political rhetoric – at least among the already converted – to say – as Kelton does – “cash needn’t be a constraint on us doing any of this stuff”. But – and this is where I think the book verges on the dishonest (or perhaps just a tension not fully resolved in her own mind) – the constraint, or issue, is always about real resources, which – per the quote above – can’t be conjured out of thin air. Resources used for one purpose can’t be used for others, and even if some forms of government spending (or lower taxes?) might themselves be growth-enhancing in the long run, that can’t just be assumed, and almost certainly won’t be the case for many of the causes Kelton champions (or that local advocates of MMT would champion).
I can go along quite easily with much of Kelton’s description of how the technical aspects of economies and financial systems work, but the really hard issues are the political ones. So, of course, we needn’t stop government spending for fear that a deficit will quickly lead to default and financial crisis, or because in some narrow sense we don’t have the cash available in advance. But we still have to make choices, as a society, about where government programmes and preferences will be prioritised over private ones – the contest for those scarce real resources, consistent with keeping inflation in check. And we know that rigorous and honest evaluation of individual government tax, spending and regulatory programmes is difficult to achieve and maintain. And we know that programmes committed to are hard to end, And that government failure is at least as real a phenomenon as market failure – and quite pervasive when it comes to many spending programmes. And so while Kelton might argue that, for example, balanced budget rules (in normal circumstances, on average over the cycle) are some sort of legacy of different world, something appropriate and necessary for households but not a necessary constraint for governments, I’d run the alternative argument that they act as check and balance, forcing governments to think harder – and openly account for – choices they are making about whose real resources will be paying for the latest preferrred programme.
Kelton tries to avoid these issues in part by claiming that “outside World War Two, the US never sustained anything approximating full employment”, and yet she knows very well that real resource constraints still bind – inflation does pick up, and was a big problem for a time. Hard choices need to be made – not by the hour (government cheques can always be honoured) but over any longer horizon.
There are perfectly reasonable debates to be had about the appropriate size of government. but they really have nothing to do with the more-technical aspects of the MMT argument. Even if, for example, one accepted the MMT claim that there was something generally beneficial about fiscal deficits, we could run deficits – presumably still varying with the cycle – with a government spending 25 per cent of GDP (less than New Zealand at present) or 45 per cent of GDP (I suspect nearer the Kelton preference).
This post has probably run on too long already. Perhaps I will come back in another post to elaborate a few points. But before finishing this post I wanted to mention one of the signature proposals of the MMTers – the job guarantee. There is apparently some debate as to just how central such a scheme is – that is really one for the MMTers to debate among themselves, although it seems to me logically separable from issues around the relative weight given to fiscal and monetary policy. I covered some of the potential pitfalls in the earlier post and I’m still left unpersuaded that the scheme has anything like the economic or social benefits the MMTers claim for it, even as I abhor the too-common indifference of authorities (fiscal and monetary to entrenched unemployment. In the current context, one could think of the wage subsidy scheme as having had some functional similarities, but it is a tool that kept people connected to (what had been) real jobs, and which works well for identifiable shocks of known short duration. That seems very different from the sort of well-intentioned job creation schemes the MMTers talk about. From the earlier post
It all risked sounding dangerously like the New Zealand approach to unemployment in the 1930s, in which support was available for people, but only if they would take up public works jobs. Or the PEP schemes of the late 1970s. Mitchell responded that it couldn’t just be “digging holes and filling them in again”. But if it is to be “meaningful” work, it presumably also won’t all be able to involve picking up litter, or carving out roadways with nothing more advanced than shovels. Modern jobs typically involve capital (machines, buildings, computers etc) – it accompanies labour to enable us to earn reasonable incomes – and putting in place the capital for all these workers will relatively quickly put pressure on real resources (ie boosting inflation). If the work isn’t “meaningful”, where is the alleged “dignity of work” – people know artificial job creation schemes when they see them – and if the work is meaningful, why would people want to come off these government jobs to take existing low wage jobs in the private market?
And much of Kelton’s idealistic discussion of the job guarantee rather overlooked the potential corruption of the process – favoured causes, favoured individuals, favoured local authorities getting funding. It is a risk in New Zealand, but it seems a near-certainty in the United States.