What Would Keynes Say?
The most inflential economist of the 20th century still dominates debate about the financial crisis. His biographer Robert Skidelsky and Britain’s leading monetarist Tim Congdon discuss the relevance of Keynes
Daniel Johnson: We should begin with Keynes. Robert, you’ve just published an important new book, The Return of the Master, which has already had a great impact here and abroad. Why is this great economist, who died more than half a century ago, still in your view the key thinker for those who want to understand the present situation, and what part of his legacy do you think is the most important?
Robert Skidelsky: He still provides the best explanation of how economies can crash, and also of how they can then stay in a depressed state for a long time. He overturned the classical view of his time — which actually was then revived — that markets are always efficient and that economies are always self-correcting without very much trouble. The effect of his theory was to give governments a definite role in both preventing big upsets, and, if big disturbances occurred, in getting economies out of the holes they were liable to stumble into and remain in for a long time.
DJ: Tim, you too respect Keynes as an economist, but I think you take a slightly different view from Robert’s about what part of his legacy matters now.
Tim Congdon: Keynes wrote a lot about the Gold Standard, and indeed about many different currency standards — his first book was about the Indian monetary system, of all things — and he realised that in the 20th century we couldn’t base our monetary regime on gold, or indeed on any precious metal. We needed to manage the currency, and manage it with a view to achieving price stability and a stable economy which would generate, as far as possible, high employment. I agree with all of that entirely.
The debates are about the extent of his enthusiasm for fiscal policy, for high government spending and for a large state sector. Robert has written a wonderful biography of Keynes, which I think will be definitive. But even that biography shows that Keynes had second thoughts about whether it really was a good thing for governments to be trying actively to manage the economy through budget policy.
DJ: Robert, how much do you think the present stimulus policies that have been very widely adopted across the Western world owe to Keynes, and are they working? Was this the right way to go, in your view?
RS: The stimulus has come in two forms. There’s been a monetary stimulus, that is to say that governments have printed money; and there’s been a fiscal stimulus, that is that their budget deficits have expanded. And both are in the mould of Keynes. Without Keynes they might not have done any of these things, and the test of that is that before, in the actual Great Depression, governments did all the wrong things. And that’s one of the reasons why the economy slid down, all the way, 12 quarters of decline.
The effect of the present combination of policies has been to cut off the decline after four or five quarters. It hasn’t brought a definitive recovery yet. In Britain the last quarter’s figures still show a continuing decline, a small one, and we don’t really know how vigorous the recovery will be. So there’s still a lot to do.
What this shows is that, despite very substantial boosts to spending, when you get a shock like this, when the banking system implodes, you really are in a very deep hole, and it’s very difficult to get out of it. And it would have tested all of even Keynes’s ingenuity to know what the best method of escape is and what the restoration of normal levels of activity — that is trend rate of growth — would take.
TC: There are a couple of things there that I don’t agree with. Robert is a great economic historian, so I should defer to you . . .
RS: But you won’t!
TC: No, I won’t. The first thing is that it’s absolutely right that the problem in the 1930s was the banking system imploding. But I would say that you can interpret what happened in the 1930s very much in monetary terms, and that the problem, particularly in America, was the collapse in the quantity of money: a roughly 35-40 per cent collapse in the quantity of money, according to work done by [Milton] Friedman and [Anna] Schwartz in their great Monetary History of the United States. And actually, what turned things round wasn’t fiscal policy, it was an aggressively expansionary monetary policy that became possible when America left the Gold Standard in 1933.
And if you look at the numbers you’ll see that there was very rapid growth of US money from 1933 to 1936. The banking system had its problems and credit to the private sector was very weak. But the government borrowed on a vast scale from the banking system and created money. So this was a case of managing the currency to boost the economy in a recession, something I very much would have agreed with if I’d been there then. By the way, Keynes advocated similar things in the UK to those done in America, and I’m advocating a similar approach at the moment in this country. So that’s one area of disagreement.
The second area is that you say governments are printing money. If you look at the numbers, monetary growth in the last nine months in America, the eurozone and Japan has been nil. There’s been a bit of monetary growth here in the UK, but that’s because of the specific policies — quantitative easing, targeted on the quantity of money — the Bank of England has been pursuing. This is so, so important. There is a huge debate at the moment about the relevance of both Keynes and Friedman and monetary economics. Where I disagree with you is that I believe that it’s vital in these circumstances to emphasise the quantity of money and to argue for raising its growth rate. That’s what really matters, not budgetary policy, not fiscal deficits. And there we disagree, I’m afraid, almost totally.
RS: I would agree with one thing: of course, in the early 1930s being on the Gold Standard was a great constraint on monetary policy, and once America got off the Gold Standard then it became easier and monetary conditions eased. And that was also true of Britain after 1931. But I would say that the crucial thing in a recovery is not the growth of the money supply — that is the growth of bank money, money produced by the central bank — but the growth of aggregate spending, and the connection between these two is actually not a close one, especially not in abnormal times. In normal times, and this was one of the things Friedman’s book did establish, there is a stable relationship between the supply of money and, in the old equation, the speed with which it’s spent. And that of course does suggest that a simple increase in the quantity of money will cause output and prices to rise in the same proportion. But when things are as disturbed as they now are, and when both banks and companies are actually hoarding money and sitting on cash balances, then I think you have to establish a much stronger connection between money and velocity than Tim has been able to do.
The important thing is to get aggregate spending up, and the surest way to do this is by the government spending the money itself. And even in that case, some of the money might be saved and one can argue about the size of what’s called the multiplier, how much extra government spending induces additional private spending. One can argue about all that, but it’s the spending that’s the important thing, not the quantity of money in itself.
TC: This all goes back a very long way, with one of the best statements on the problem being by Irving Fisher in 1911. If the government and central bank take action to increase the amount of money, if the banking system then expands its deposit liabilities, the quantity of money grows. Suppose the quantity of money increases by, say, 50 per cent. Is there then a change in people’s and companies’ desired ratio of money to their income and wealth? The evidence is overwhelming, from all countries, from all periods, that the desired ratio does not change. Alternatively, the velocity of circulation is not affected by changes in the level or rate of growth of money. What that means is that if the quantity of money rises by 10 or 15 per cent, it is very likely that the equilibrium level of the national income will also rise by 10 or 15 per cent.
RS: In the long run.
TC: You’re going ask about lags — again this is Friedman’s work, which, by the way, I’ve spent 30 years looking at and trying to comment on the numbers…
RS: It doesn’t take that long.
TC: The lag between money and output is very short. In fact we had a very vivid demonstration of this in the middle of 2008. We had gone in this country from roughly a 12 per cent growth of money in late 2006 and early 2007 to nil in the middle of 2008. In fact, in mid-2008 companies’ money holdings were actually falling. And what happened to the economy? You went from quasi-boom conditions to an appalling recession, with hardly any lag at all. The facts speak for themselves. The critical thing about what the Bank of England has done with quantitative easing is to have tried to bring back a positive rate of monetary growth. That is what really matters to the economy. To a large extent the fiscal deficit is a distraction and of no importance.
RS: I don’t agree with that. I think you said in some correspondence we had, “find me an economy where a 10 per cent boost to real, broad money in six months has not boosted asset prices, demand etc” — in other words, talking about demand. The explanation lies in what happened to velocity. It was the fall in the velocity of circulation — of the demand for, and supply of, loans — which produced the initial fall in the money supply. There was a loss of confidence, banks stopped lending, companies stopped borrowing. And the recovery now depends on banks lending more and people spending and borrowing more, not on a simple increase in Bank money. In the long run, I agree that there is a stable relationship between the quantity of money and thevelocity of circulation, but in the long run, as Keynes famously said, we are all dead.
TC: There is a problem with the money numbers at the moment in the UK. Three or four years ago, banks were creating semi-bank subsidiaries, to bypass capital rules. Deposits have been held in those subsidiaries, and they have been doing all sorts of very odd things in the last year or two. Because of that, the Bank of England has correctly taken out these semi-bank deposits from the measure of money they and other economists look at. This measure is called M4x, with the “x” denoting that they’ve excluded these semi-bank deposits.
And the relationship between that M4x measure and what’s happened to demand in the economy shows that this was the driving force behind this cyclical episode, as it was behind the cyclical episode in the late 1980s, the boom-bust then, and the boom-bust in the early 1970s and mid ’70s, both of which I’ve written about at some length. If you’re running a business, how much money you have in the bank is critical. And if all the businesses in the country find that their cash balance — the amount of money they have in the bank — is being squeezed, then you will get a thumping great big recession. And I’m afraid the latest episode is dramatic confirmation of the relevance of money to macroeconomic outcomes. To end the recession you need more money in the economy: that is the critical point.
RS: Of course you need more money in the economy. It’s the relationship between money and the real economy that is crucial. I would agree — banks have to lend more, and it’s also the case that people have to want to borrow more, and that depends on companies’ profit expectations. What I think you’re missing from the whole thing is the issue of psychology and confidence.
Let me go back to a point that you made earlier, which is that fiscal policy is irrelevant. Certainly, the amount of discretionary fiscal stimulus has been quite small so far, though more is coming on. But that excludes the automatic stabilisers. In fact the government has been running a very large fiscal deficit. Despite this huge injection of money into the banking system, Mervyn King himself said: “The recovery hasn’t happened.” If this is the best that monetary policy can do, we certainly need fiscal policy as well.
TC: Let’s just think about what you just said.
RS: I always try to.
TC: Let’s just be clear about what you’ve just said. You said there has been a discretionary fiscal boost so far.
RS: Yes, of 1.4 per cent GDP.
TC: OK, and do you think that has really been critical, is boosting the economy?
RS: No. I’ve said that there have been two types of fiscal expansion: the discretionary boost, and the effect of the automatic stabilisers.
TC: OK, so now let me ask you, how do you explain the recession? What has caused it?
RS: The recession started before any of these recovery programmes. We had the sub-prime collapse. We had the financial innovations and a huge explosion of securitisation, which were all based on risk control models which Keynes would have said claimed far more than they could deliver. We had a big increase in oil and commodity prices. We had, in the background, the global imbalances between China and the United States — in a Keynesian analysis, excess saving in China and excess spending, based on very cheap money, in America, without really a very strong investment, so that a lot of it went into asset bubbles and consumer indebtedness.
A lot of these things come together. All great crashes are multi-caused. And the same is true of the Great Depression itself. Now I don’t know where this leads exactly, but its effect was a credit freeze, which started in September 2007, when banks stopped lending to each other and to their customers, and the real economy started to wind down. There was a big collapse in aggregate demand globally, and once that had started, it went on, because of the multiplied effects of the initial drop in aggregate spending.
TC: Keynes was a great monetary economist, and his favoured measure of money was one that included all bank deposits — that’s very clear from an explicit footnote in the General Theory and the rest of his work. And therefore we’d expect, if Keynes’s theory was right, that at least part of the background of this recession would be a collapse in the rate of growth of money. You’d also think that if you had read Friedman’s work. Then look at the data, and that’s exactly what we find: we find that in America, we find it here too.
So if you ask me what the dominant cause is I would say that the banking system got into trouble, pressure was wrongly put on the banking system to shrink risk assets, and the result was that the growth of money, which had already been fading from the middle of 2007, then collapsed last year.
We ended up with this ghastly recession. The discussion of this recession can be pursued in terms that Keynes would have found familiar in the 1930s, and would have had have no difficulty with, and in those discussions the quantity of money would have been basic.
RS: It wouldn’t. That’s where you’re wrong. Of course any recession is accompanied by a collapse in the amount of money in the economy. The question is what causes that collapse? And for Keynes it was always a decrease in the demand for loans that was the causative factor in the collapse in the supply of money.
TC: Now Robert, I’m going to cause you a problem: I want quotes for that, please.
RS: No you can’t — that is just a debating point.
TC: It’s quite important, because I’ve read Keynes, and you’re not right.
RS: I am right. There were two elements in Keynes. What he thought was that you could offset, up to a point, a collapse in the demand for loans by increasing the supply of money. And by collapse in the demand for loans he simply meant that when confidence fell, liquidity preference increased, and you could try and fight the increase in liquidity preference by flooding the banking system with money. But the basic cause of the collapse of the money supply was a collapse in the demand for loans, and you can follow that in his chapters on long-term expectations, and on the rate of interest.
And here’s another quote — you asked for a quote, and you supplied the quote yourself, let me read you your quote. You say: “If there is some tendency to a measure of long-run uniformity in the state of liquidity preference” — an important proviso — “there may well be some sort of rough relationship between the national income and the quantity of money, taken as a mean over periods of pessimism and optimism together.”
That is a long-run relationship, and it depends on an assumption that liquidity preference stays the same. And that assumption is not in Keynes, it’s not in the General Theory, it just isn’t.
TC: It’s certainly true that Keynes claimed that problems in the economy arise because of instability in the demand to hold money balances. That is certainly part of his theory. Empirically, he’s wrong.
RS: He’s not wrong. How do you explain today? Although there’s been substantial quantitative easing, the actual growth in the demand for loans has been small, and in fact, I quoted you these figures already: net lending in July fell in the UK at its fastest pace since records began in 1993.
TC: Look, bank deposits and bank loans are totally different things.
RS: Doesn’t matter, I know that. It’s the demand for loans that’s important.
TC: Oh no, it’s not.
RS: : Oh yes it is, of course it is. You can make credit facilities available, it’s the spending that matters.
TC: Let’s get this absolutely straight in terms of definitions. Liquidity preference is the demand to hold money balances. Is that correct or not?
RS: Liquidity preference is the demand to hold money balances, yes. And that can fluctuate.
TC: And national income is determined by the interaction between the quantity of money created by the banking system, which does indeed depend partly upon loans to the private sector, and the demand to hold money. But money can be created by the state if it borrows from the banks. Bank lending to the private sector is not necessary at all.
RS: I wouldn’t put it that way.
TC: That’s the way Keynes puts it.
RS: No it isn’t: The national income was made up of consumption and investment. Consumption was stable, investment was unstable. That’s why you had fluctuations in national income and output.
TC: At the end of chapter seven of the General Theory he states what I just stated — it is actually the opening quotation in my last book, which you very kindly commented on. That is Keynes’s basic theory of the monetary determination of the national income.
RS: It’s not, no Keynesian would agree with you. I mean you’re almost alone. Look, you should be on your own in this, and not bring in Keynes. You’ve got a view of things…
TC: I’ve given you a quote.
RS: But I’ve given you the whole theory, not just an odd quote. Anyone can extract quotes out of the context of the whole framework.
TC: There is a very nice summary of the General Theory that was done immediately after its publication by John Hicks where he said that Keynes actually has two theories. He has the theory that national income is a multiple of investment, and the multiplier is the reciprocal of the marginal propensity to save. And there’s also the monetary theory. And the monetary theory is one in which the demand to hold money balances (L) is equal to the amount of money created by the banking system (M) which goes into the LM curve in the IS-LM diagram. That is basic theory. The monetary element in Keynes is a very major element, and I have just stated what his theory was, and I am correct about that.
RS: One bit of his theory.
TC: That was his theory throughout.
RS: You’ve just said there were two theories — now you say there’s one theory.
TC: His monetary theory was the theory I’ve just described that didn’t change.
RS: But his General Theory…
TC: He then added this idea, which of course is the theory that you like and that the Keynesians like, that income is a multiple of investment. And from that they went on to say that, because capitalism is inefficient, unstable and so on, and when there isn’t enough private investment, the government should come in as well.
So the second element in Keynes, this multiplier theory, the fiscalist element, then led to the standard Keynesian argument that you need fiscal policy, you need a larger state sector, you need budget deficits, and all of this kind of stuff. I know it’s there, Robert — I don’t agree with it. However, I am correct that a much larger part of Keynes’s work is actually about money.
RS: You always say “I am correct.” Please allow yourself a moment or two of doubt. I think that most Keynesians, including Keynes himself, thought that the savings-investment analysis was the fundamental one. It relates to his view as to why economies collapse. The General Theory has two chapters on the consumption function, he has four chapters on the investment function, including the role of the rate of interest. The bulk of the book is about the relationship between saving and investment.
It is true that Keynes did not believe that you could recover from a deep recession simply by borrowing money from the public; in other words, by increasing the government’s debt. He didn’t think you could do that because he thought there was a danger of financial crowding-out, unless you increase the quantity of money. And he thought that increasing the quantity of money was an absolutely essential part of a recovery process — it was necessary because it would lower the cost at which government had to borrow. In other words, it was part of deficit finance.
And you say this in the article you wrote in January — you admit this. You say: “Money creation is a way for the government to finance its budget deficit.” Now, Keynes would agree with that, of course he would. But budget deficit is the point here. It’s the budget deficit that’s crucial, and increasing the money supply is a way of financing the budget deficit without driving up short-term rates of interest. Now I think he accepted that. But why, despite the fact that there was a slump going on, was the existing supply of money not enough? Because he did think that people were increasing their money balances. That was characteristic of the decline in investment.
TC: Robert, I don’t follow. You were saying the reason there wasn’t enough money was because people were increasing their money balances…
RS: Hoarding — they were increasing their hoarding, or their propensity to hoard. And that was why, if the government was trying to borrow money from the public in order to spend and people were increasing their hoarding or their desire to hoard at the same time, there would be a tendency for the rate of interest to go up. And it was to counteract that that he wanted the money supply increased. That’s all very clear, not in the General Theory so clearly, but in an article he wrote in 1937 when he conceded that there might be some financial crowding-out even though the resources were seriously unemployed.
TC: So in that period Keynes thought that an increase in the quantity of money was important to recovery?
RS: He did. He thought it might be a necessity but he never thought it would be a sufficient condition.
TC: So you’d agree that at the moment it’s a good thing to have an increase in the quantity of money?
RS: Yes, yes!
RS: I’ve never denied it. You know this idea that Keynes wanted larger governments, that he wanted an enormously bigger economic size of the state, is rubbish. He actually doubted whether government should spend more than 25 per cent of the national income. He didn’t want a permanent increase in the size of the state, but he made a big distinction between when you’re in a slump and when you’re in normal times. In normal times you want to balance the budget and have as low a rate of taxation as is compatible with economic growth and these kinds of things. This is all true — you can’t make Keynes out to be this monster who wanted the size of the state to grow and grow and grow.
TC: The Keynesians certainly thought like that, and that’s why I find your emphasis on the fiscalist multiplier side of Keynes very disappointing. But can I just pick up a couple of issues? The first is to explain why, despite the Bank of England’s enormous quantitative easing programme in 2009, we still have low growth of the quantity of money.
By the way, can I actually make clear that bank lending isn’t by itself terribly important to all this? The important variable is the quantity of money, the quantity of money in bank accounts. And what’s happened is that, since last autumn, the banks have been under huge pressure to raise their capital/asset ratios. This has meant that they’ve been reducing their risk assets, their loans and to some extent their holdings of risky securities. They’ve also been raising lots of capital rather than issuing deposits. So the result of those forces has been to deduct something like about £150-£200 billion from the quantity of money in the last year or so. Against that, quantitative easing had added about £150-£200 billion to the quantity of money.
If we hadn’t had quantitative easing the quantity of money in this country would have fallen in the last year by about five to ten per cent, which isn’t that different from the rate of money contraction seen in the American Great Depression. So quantitative easing has prevented a calamity. Money growth hasn’t risen very much, but it hasn’t collapsed and that is a very good thing. I’m pretty sure, by the way, that Keynes would not have disapproved of that.
RS: No, I agree with you.
TC: And that’s quite important — I think Keynes would have blessed quantitative easing. And further on that, my second point, which is crucial. The state can create money in at least two ways. One way is, if it’s running a budget deficit, to finance that budget deficit from the banking system, either the commercial banks or in the extreme the central bank. OK, every economist knows that. But there’s another way of doing it, which is to run a responsible fiscal policy and to monetise the existing public debt. And what I advocated in my pamphlet earlier this year on How to End the Recession was not to increase the budget deficit. I in fact put an increase in the budget deficit in the category of “bad and/or uninteresting ideas”. Instead my proposal was that the government and the Bank of England, working together, should monetise the existing debt.
How? What about the mechanics? The government borrows from the commercial banks, the government then has a new deposit created for itself, and the government then uses that deposit to buy back its own long-dated debt. The result is to create new deposits, new money, in private sector hands. Do you know where I got this idea from, Robert? I got it from Keynes.
RS: Yeah, yeah.
TC: And the polemical point is that quantitative easing isn’t actually very original. The problem is that this part of Keynes’s work has been neglected by the Keynesians, who are so obsessed by all the fiscal stuff that they have neglected his monetary work. And all through the 1930s, you’re correct, there was this fiscalist agenda in Keynes’s writings. In my view the agenda was mistaken and dangerous, but of course it was there. However, there was also, very strongly, advocacy of expansionary market operations, which Keynes called monetary policy à outrance. And all I’ve been doing in 2009 is advocating essentially the same policy.
DJ: Is there one measure that either of you would advise the present or a new government to take to lift us out of this recession?
TC: The most important thing is to use debt management policy, open market operations, interest rate policy — all the available weapons — to make sure that the rate of growth of money (really meaning bank deposits) remains positive. Not at an excessive rate, but at a rate that means that balance sheets, asset prices and so on remain satisfactory throughout the economy. So the most important thing is a positive rate of monetary growth, not too fast, not too slow, just a steady rate of monetary growth, about five per cent a year.
RS: I would say that you certainly have to try and use all your instruments to maintain a rate of monetary growth, but to me that depends on fiscal policy and not just monetary policy. I’ve always thought that it’s a matter of causation, and there’s a very complicated pattern of causation between money and the real economy, and you have to make sure that the real economy grows — that is the important thing. In other words, you have to overcome the output gap, which is projected about five per cent — the economy will have contracted by five per cent by the end of this year. You have to stop the growth of unemployment. Tim keeps saying you’ve got to ensure that money supply grows by five per cent and that’ll cure these other problems. I don’t think it will on its own, and that’s been the nub of the disagreement between us.
I think there’ll have to be another boost, I don’t think it’s going to be quite V-shaped, I think we’ll drag along the bottom in a very mediocre condition for some time to come, and I think to ameliorate this there has to be more government spending. And that is coming on line — there are very big projects. To my mind the trouble with fiscal policy has always been the time it takes — the lags are probably longer than in monetary policy, unless you just give people money to spend; poor people particularly will spend more of the money they’re given than wealthier people.
If I gave Tim £1,000, he might not go out and spend it all, particularly if he was a banker — he might think that if he lends it to people he might lose it all. So he might not lend it, and if it goes into buying assets, that’s one way of doing it. But it’s not clear how the increase in wealth flows into the real economy or to big sections of the real economy. I think those are the points we might disagree on.
TC: Can I say one more thing? I believe that in an article in the Sunday Telegraph you proposed that everyone should be given spending vouchers by the state, to encourage them to buy more…
RS: Spending on British-made goods, I should say.
TC: Never mind that. The British public has something like £1,000 billion worth of spending vouchers at the moment — they call them bank deposits. And British companies have a further £250 billion of spending vouchers, and financial institutions also have another £400 billion or so of spending vouchers.
Ask yourself, “What is so special about adding another £20-30 billion in spending vouchers?” And if you think that these spending vouchers are so important, then why are you lukewarm about an increase in the quantity of money? Because spending vouchers and bank deposits seem to me to be the same thing.
RS: I don’t think they are the same thing. It depends on who gets the spending voucher.
DJ: You would want people at the bottom of society to get them…
RS: …whose propensity to save is very low, who may not have bank accounts or have very small bank accounts. You’ll get a bigger bang per buck if you direct the new money to those people than if you direct it to people who have a lot to lose by their spending.