By Philip Pilkington, a writer and journalist based in Dublin, Ireland. You can follow him on Twitter at @pilkingtonphil
We want structures that serve people, not people that serve structures!
– Graffiti in Paris, May ‘68
Recently there has been a bit of debate about MMT knocking around the blogs. To a large extent it has been rather superficial. This was not the fault of those involved – despite a few displays of pomp and bluster from some of the cruder economic bloggers (no names!). The superficiality of engagement was mainly due to the nature of the medium itself. For all their accessibility blogs don’t allow, or should I say, don’t generally encourage the scholarship that is required before considering a new theoretical approach. Indeed, the very nature of blogs is restrictive in that it demands that authors engage with a new theoretical approach only in a cursory manner. After all, there will always be something new to write about tomorrow.
Better, I thought, to engage with one of the participants more directly. So I dropped Dean Baker an email and he was more than happy to engage in a dialogue on the topic. Dean gave me a great deal of time and so I thought it might be worth publishing the results of the discussion in the hope of moving a sometimes slow moving debate forward somewhat. While Dean and I ultimately disagree on mainstream economic theory, he’s otherwise a great economist who I’ve learned a lot from and have a great deal of respect for. Anyway, enough of the niceties – on to the models.
MMT vs. The Mainstream
The starting point, as it so often is, is the old ISLM model. IS = “Investment-Savings”, LM = Liquidity – Money Supply”. MMTers hold that the whole model is faulty. It doesn’t take account of a labour market and there are serious problems with having a downward sloping IS-curve. Dean and I only discussed the LM-curve in any detail, however, so we will stick to that for now.
It should also be noted that many MMTers consider the ISLM, or at least the LM-curve therein, to be a fairly accurate depiction of a gold standard or fixed exchange-rate regime. In truth it probably is, but in that it is only a rough approximation; even in such regimes the money supply is far more fluid than the ISLM depicts. In what follows we will consider only a floating exchange-rate regime.
In mainstream and standard Keynesian economic analysis the LM-curve is upward-sloping. Before the reader falls asleep I’ll quickly highlight the implications of this. Basically, it means that when the demand for money rises interest rates should rise too, unless there is intervention by the central bank. Think of it this way: the economy operates with a limited supply of funds or, put differently, the money supply in the economy is fixed. If more households and firms chase after a set amount of money they will bid the ‘price’ of that money up in line with the laws of supply and demand. Thus, the interest rate will rise if more people make demands on this fixed supply of cash.
The role of the central bank is to increase or decrease the supply of money in the banking sector by engaging in open market operations. They do this either to keep inflation subdued or to expand the economy. In the mainstream analysis the central bank is assumed to have pretty much full control over the supply of money. If the economy is growing slowly, they open the spigot and supply more funds into the banking system. If the economy is overheating they extract funds from the banking system. Turning on the tap drives down interest rates as so-called ‘base money’ in the banking system increases; pulling back funds drives interest rates up as base money in the banking system decreases. The effects of these increases or decreases of base money in the banking system are then basically mirrored in the economy at large through the effect of the money multiplier.
MMTers (and other post-Keynesians) disagree with this characterisation. They claim that the central bank has little or no control over the supply of money. They claim that the supply of money is ‘endogenously determined’ – i.e. set by the amount of economic activity taking place at a given time. Because MMTers, through their close study of actual banking practices, do not adhere to the idea of a ‘money multiplier’; they do not think that the central bank determines the supply of money at all. Instead it merely sets the price of money (that is, the interest rate) and allows the demand to adjust to this price.
According to MMT in a floating exchange-rate regime there is an unlimited supply of funds, but these are given a fixed price by the central bank. This is a bit like a monopolist with enormous excess capacity that can, for all intents and purposes, produce an infinite amount of cars. The monopoly firm sets the price of the cars that they sell and allow demand to adjust in line with that price. Ditto for money in the MMT understanding of how the banking system works. The central bank is effectively a ‘money monopolist’ that allows the production of an infinite amount of money at a set price.
There is an enormous difference here in how these processes are conceptualised. The mainstream approach deals almost exclusively with ‘stocks’ of money within the banking system, while the MMT approach deals with ‘flows’ and their rate of rate of expansion and contraction. The mainstream view is that a ‘stock’ of money is injected into the banking system by the central bank and this in turn creates another ‘stock’ of money in the economy through the process of the money multiplier. The MMT view, on the other hand, is that the central bank sets a price for money and allows the demand for this money to determine the ‘flows’ into the economy that result.
The mainstream ‘money stock’ approach is actually remarkably primitive when thought through in any depth and leads to misunderstandings about money and debt creation. This harks back to the fact that mainstream theorists use crude ‘static’ modelling straight out of 19th century engineering when trying to understand the economy – but this might be a bit too complex to get into here; let’s just say that the approach is a tad dated.
But in Theory…
Frankly, if you understand how banking works in a modern monetary system you cannot really deny the MMT conception because simply describes how things actually operate. However, you can claim that it remains valid to conceptualise the money supply as being fixed theoretically.
Here’s why: It all rests on your definition of the interest rate. For the most part when economists talk about ‘the interest rate’ they are referring to the short-term or overnight interest rate that is set by the central bank. This is the interest rate that Greenspan and Bernanke talk about all the time. However, there are other interest rates in the economy. One important interest rate is the long-term interest rate on 10-year treasury bills. This interest rate is usually set by the markets. (Note that there are important exceptions to this that are not dealt with in the ISLM and call it even further into question). The markets tend to bid this interest rate up when they see a potential for inflation in the economy or when they think that the central bank might raise rates due to their own inflation expectations. So, when the economy looks like it might be growing too fast the market bids up the long-term interest rate.
The central bank keeps an eye on the inflation rate too and adjusts the short-term interest rate accordingly. If inflation gets too high, or inflationary pressures are thought to be in the pipeline, the central bank will jack up the short-term interest rate.
If you adhere to this view you could say that the standard ISLM model is theoretically valid. As the economy grows and the demand for money increases the long-term interest rate on 10-year treasury bills increases in expectation of future inflation or interest rate hikes. At the same time, the central bank is also reacting to economic expansion and the demand for money by tracking the inflation rate. These reactions on the part of both the markets and the central bank simulate a fixed stock of money because interest rates rise as the economy grows ‘too fast’. (As an aside it should be noted that these expected movements are not firmly grounded in reality because in floating exchange regimes empirical evidence indicates that interest rates can climb at the same time as the money supply increases, so there is no direct link here.)
The dance that takes place between the variables is rather chaotic but there is a clear pattern. You can get a good idea of it by looking at the graph below. In red is the long-term interest rate, in blue the inflation rate and in green the short-term interest rate.
As the reader can see, the long-term interest rate is not actually a very good predictor of future trends at all. The correlation between the inflation rate and the short-term interest rate is very tight which is not surprising because the central bank is tracking the inflation rate and adjusting the short-term rate accordingly. The long-term interest rate, on the other hand, seems to follow the lead of the other two.
The Rise of the Robot Banker
Even though the long-term interest rate is clearly the dependent variable, those that adhere to the ISLM assume that the interaction between the three variables leads to a restricted ‘market for funds’ in that interest rates in general will rise as the economy and the demand for money expand. But it seems clear that the main agent of change here is not The Market at all, but instead the central bank which responds to the rate of inflation or the expectation of inflation.
In reality this is not some sort of automatic process. By assuming an upward-sloping LM-curve – that is, a fixed supply of funds – there is an implicit assumption that actions on the part of the central bank are somehow neutral. ISLM enthusiasts implicitly assume that the central bank is simply responding to some otherwise ‘equilibrating’ market conditions and adjusting its rates in line with this. The implicit bias toward self-equilibrating markets that is buried in the ISLM is what allows economists to maintain an illusion of a fixed supply of funds (upward-sloping LM-curve).
In actual fact, if we look closely at this ‘theoretical’ justification of a fixed supply of funds (upward-sloping LM-curve) we see that it is simply a self-justifying model. What the standard ISLM model does is bury within its assumptions that the central bank’s interventions are neutral. It buries the fact that the central bank is actually taking a specific stance on policy and then tries to pass off this stance as a sort of quasi-market response (i.e. as if there were a market for a fixed supply of funds). But the central bank’s policy stance is nothing of the sort. Instead it is a sort of a simulation of what a market response is thought to be. Thought to be by whom? By economists that adhere to models similar to the ISLM, of course!
This is real ‘through the looking-glass’ stuff. What the standard ISLM model does is neutralise or disappear the economist or central banker within his or her own theories.
“No need to look over here,” says the central bank economist and his followers in the profession. “We’re not taking any ‘action’ per se; we’re just humble subjects of The Market’s great laws. We’re just undertaking ‘automatic’ actions that are dictated to us by an abstract and unseen overlord.”
This is nonsense. Put simply, the moment you disagree with the abstract ‘laws’ by which the economists think the economy operates, you call into question the actions they undertake because, to give just one example, they may judge the so-called ‘natural’ rate of unemployment to be far too low and hike interest rates when there is no real threat of inflation. Then their actions start looking a lot less ‘automatic’ and a lot more ideologically driven.
Dean didn’t disagree. He told me that he points all this out to his students within the confines of the ISLM model and I have no doubt that he does. Nor do I have any doubt that his students come away better informed than others who go through the sometimes unfortunate rigmarole of training in economics. But still, I cannot help but feeling that there’s a sleight-of-hand taking place here at the level of intellectual construction.
Some might say that I’m taking a ‘normative’ view of all this. They would say that what I’m really griping about is that the central bank does not undertake the sorts of policy actions that I might favour. But I don’t think that this is the case at all, this is not a case of judging policy actions per se. In fact, I think it’s the MMTers who are being good positivists. We are describing how the system works. It is the mainstream models that plaster over the normative judgments made by central bankers by positing their ISLM model with its upward-sloping LM-curve.
By allowing for a fixed supply of funds they confuse reaction on the part of the central bank with a quasi-automatic reaction taking place within the economy itself. By claiming that the LM curve is flat and that the central bank sets a price and allows demand for money to adjust the MMTers are being far more realistic and far less convoluted in the way they represent the functioning of the central banking system.
ISLM enthusiasts will, of course, assure me that they don’t really believe central bank actions are neutral. But then why do they adhere to models that imply that it is? To those that would make this excuse I would warn them to take a look at history. Models can often take on a life of their own. I’m fairly convinced that monetarism – with its money supply targeting – arose quite organically out of the faulty ISLM model. I would also say that economists can do better than simply positing flawed models and then making excuses when they don’t work. Just drop the model altogether. The rest of us get on just fine without it.
Life in a Derelict Building
Models such as the ISLM – nearly all economic models, really – are self-justifying. One could say that they are almost tautological constructions. What they seek to do is dissolve the conceptions of the world held by the model-user into the models themselves. Buried deep within the model, most students don’t question these assumptions at all. They just assume that the world is as the ISLM says it is, while in truth the world is simply run by folks who think that the ISLM is a good representation of that world and thus a good guide to policy.
The ISLM, then, is really a model of action for neoclassical central bankers who wish to try to impose an upward-sloping supply of funds (LM curve) on the economy. It does not represent how the world works, but instead how the world should work according to such people. It is a normative model in the strongest sense of the term and frankly, all those who use it to make predictions or try to understand how the economy works are not really engaged in detached reflection on how the economy works at all. No, they are buttressing the status quo – within their own minds, no less. And even though the better of them can tear themselves away from the dictates of the model when it comes to policy recommendations, they still ultimately remain tied to its apron-strings.
Arguing against the implicit implications of this model from within the confines of the model is an uphill struggle to say the least. In order to do so all sorts of ‘ifs’ and ‘buts’ have to be dragged into the debate and the whole thing becomes terribly messy.
This is because models like the ISLM are classical or neoclassical constructions. (Remember the title of Hicks’ original paper introducing the model…). In truth these models are built on the back of a seriously flawed theoretical edifice. And that, I think, is the problem with pretty much every mainstream model: they are not intellectual constructions that promote thought; they are more akin to cages built out of the remnants of long dead assumptions that are used to entrap the minds of those they are handed to. It is only those of highly independent mind that can wrench themselves from such conceptions once taught them.
It is worth quoting Keynes in this regard, who knew well the dangers of leaving rotten buildings standing:
The difficulty lies, not in the new ideas, but in escaping from the old ones, which ramify, for those brought up as most of us have been, into every corner of our minds.
Since Keynes, most economists have stopped trying to escape the old ideas. The MMTers, on the other hand, together with their post-Keynesian colleagues, have done nothing less than construct a theoretical edifice that quite literally shakes off the old ideas completely. And that is what makes them seem so strange to their colleagues.