By Philip Pilkington, a writer and research assistant at Kingston University in London. You can follow him on Twitter @pilkingtonphil. Originally published at Fixing the Economists
I have pointed out before that the meaning of the term ‘liquidity trap’ has today become completely altered — with said alteration mainly coming from Paul Krugman’s bizarre redefinition which seems tied up with his idea about a natural rate of interest and the central bank being unable to hit this natural rate due to their coming up against the zero-lower bound.
In actual fact, a liquidity trap occurs when people rush out of assets and instead hold money. This leads to a fall in asset prices and high interest rates which then do not respond to central bank action. We encountered a liquidity trap proper very briefly in late-2008 but due to unprecedented central bank interventions we had exited this liquidity trap by early-2009.
In this post I want to spell out exactly why Keynes thought his liquidity trap idea so important. In order to understand this we must read his short paper The Consequences to the Banks of a Collapse of Money Values which he wrote in 1931 and which is included in his book Essays in Persuasion.
Before he discusses what he will later call a ‘liquidity trap’ Keynes briefly mentions the phenomenon that is now known as ‘debt deflation’. Here is what he says,
We are familiar with the idea that a change in the value of money can gravely upset the relative positions of those who possess claims to money and those who owe money. For, of course, a fall in prices, which is the same thing as a rise in the value of claims on money, means that real wealth is transferred from the debtor in favour of the creditor, so that a larger proportion of the real asset is represented by the claims of the depositor, and a smaller proportion belongs to the nominal owner of the asset who has borrowed in order to buy it. This, we all know, is one of the reasons why changes in prices are upsetting. (p93)
Keynes treats debt deflation as if it were a perfectly well-known phenomenon. Many today, however, attribute the ‘discovery’ of this theory to Irving Fisher’s 1933 paper The Debt Deflation Theory of Great Depressions. It is usually one step from there to claim that silly old Keynes ignored the implications of debt deflation.
None of this is true. Keynes is not only aware of the debt deflation theory in 1931 but he also suggests that everyone is aware of it. One can only guess from reading the above paragraph that the idea was floating around in the ether at the time and was being discussed by upset bankers and their publications. Fisher then likely picked up on this and turned it into a theory of the depression.
But Keynes thinks that the “familiar” debt deflation theory is an enormous distraction. In the above cited paper he continues,
But it is not to this familiar feature of falling prices that I wish to invite attention. It is to a further development which we can ordinarily afford to neglect but which leaps to importance when the change in the value of money is very large—when it exceeds a more or less determinate amount.
Modest fluctuations in the value of money, such as those which we have frequently experienced in the past, do not vitally concern the banks which have interposed their guarantee between the depositor and the debtor. For the banks allow beforehand for some measure of fluctuation in the value both of particular assets and of real assets in general, by requiring from the borrower what is conveniently called a “margin.” That is to say, they will only lend him money up to a certain proportion of the value of the asset which is the “security” offered by the borrower to the lender. Experience has led to the fixing of conventional percentages for the “margin” as being reasonably safe in all ordinary circumstances. The amount will, of course, vary in different cases within wide limits. But for marketable assets a “margin” of 20 per cent to 30 per cent is conventionally considered as adequate, and a “margin” of as much as 50 per cent as highly conservative. Thus provided the amount of the downward change in the money value of assets is well within these conventional figures, the direct interest of the banks is not excessive;—they owe money to their depositors on one side of their balance-sheet and are owed it on the other, and it is no vital concern of theirs just what the money is worth. (p93)
Keynes clearly understood the financial system better than Irving Fisher. He knew that while debt deflation could transfer wealth from borrowers to creditors — a point he reiterated in the General Theory — he did not think that this was a particularly important phenomenon when considering the fragility of the financial system. Rather he emphasised something which seemed to him altogether a more pressing issue.
But consider what happens when the downward change in the money value of assets within a brief period of time exceeds the amount of the conventional “margin” over a large part of the assets against which money has been borrowed. The horrible possibilities to the banks are immediately obvious. Fortunately, this is a very rare, indeed a unique event. For it had never occurred in the modern history of the world prior to the year 1931. There have been large upward movements in the money value of assets in those countries where inflation has proceeded to great lengths. But this, however disastrous in other ways, did nothing to jeopardise the position of the banks; for it increased the amount of their “margins.” There was a large downward movement in the slump of 1921, but that was from an exceptionally high level of values which had ruled for only a few months or weeks, so that only a small proportion of the banks’ loans had been based on such values and these values had not lasted long enough to be trusted. Never before has there been such a world-wide collapse over almost the whole field of the money values of real assets as we have experienced in the last two years. And, finally, during the last few months—so recently that the bankers themselves have, as yet, scarcely appreciated it—it has come to exceed in very many cases the amount of the conventional “margins.” In the language of the market the “margins” have run off. The exact details of this are not likely to come to the notice of the outsider until some special event—perhaps some almost accidental event—occurs which brings the situation to a dangerous head. For, so long as a bank is in a position to wait quietly for better times and to ignore meanwhile the fact that the security against many of its loans is no longer as good as it was when the loans were first made, nothing appears on the surface and there is no cause for panic. Nevertheless, even at this stage the underlying position is likely to have a very adverse effect on new business. For the banks, being aware that many of their advances are in fact “frozen” and involve a larger latent risk than they would voluntarily carry, become particularly anxious that the remainder of their assets should be as liquid and as free from risk as it is possible to make them. This reacts in all sorts of silent and unobserved ways on new enterprise. For it means that the banks are less willing than they would normally be to finance any project which may involve a lock-up of their resources. (p93-94)
Here is the liquidity trap proper. It occurs when asset values fall so far, so fast that markets then herd into liquid assets — that is, money or money substitutes — and begin to dump all the rest of their financial assets. The end result becomes a situation wherein falls in asset prices beget falls in asset prices and banks rush for money and other money substitutes. But, in classic Keynes fashion, what works for the individual does not work for everyone: when all the banks dump assets at the same time their money value drops so far that the banks themselves become insolvent.
In the discussion that follows Keynes makes crystal clear that he means basically all assets: he discusses commodity prices, stock prices, bond prices and housing prices. For him a liquidity trap occurs when money rushes out of all these asset classes and into money or money substitutes. The effect is a rising interest rate on bonds and a fall in prices on other assets.
This, of course, is precisely what we saw in 2008. But it is also what Keynes saw in 1931. The difference was that in 2008 we had a reaction from the central banks which printed massive amounts of money to prop up asset prices while in 1931 we saw extensive bank failures. Funnily enough, this policy is today associated with Milton Friedman and monetarism but it is clear that Keynes’ idea of the liquidity trap was highlighting the exact same problem.
Keynes’ liquidity trap is almost certainly a more important phenomenon than debt deflation. It is the decline in asset prices that has extensive effects on the financial sector, not the rise in the value of money per se. But that said, the liquidity trap phenomenon is not all that important with regards to shortfalls in aggregate demand.
In the other essays Keynes wrote at the beginning of the depression it is clear that while he thought the liquidity trap phenomenon precipitated the fall in confidence which led to a drying up of investment and a massive shortfall in aggregate demand, solving the liquidity trap problem and stabilising asset prices would not solve the aggregate demand problem. That is a lesson we have relearned today in the wake of 2008. Central bank interventions can certainly stabilise the financial markets — although some, like housing, respond far slower than others like bonds, the stock market or commodities — but only large increases in aggregate expenditure can reflate the economy.
But, in classic Keynes fashion, what works for the individual does not work for everyone: when all the banks dump assets at the same time their money value drops so far that the banks themselves become insolvent.
Phillip, could you elaborate on this? My interpretation of the particular quote from which your comment stems is that banks become paralyzed into inaction by fear of taking losses, but you seem (to me) to be suggesting something a little different.
The quote is a description of the liquidity trap. I’m putting it in context by saying how that liquidity trap came to a head in 1931. I have in mind the paper he wrote the previous year where he described how an asset dump by one bank leads to greater solvency but an asset dump by all banks leads to mass insolvency.
In other words, by dumping assets and enhancing deflationary pressures banks can in the aggregate put themselves underwater.
Sure, that’s how the real estate market crashed…
Would banks just have waited it out, they could have recovered faster and better, but when they started to foreclose and throw people out of their homes since their debt was under water, and at the same time didn’t lend money to finance anything, the whole market got depressed and values crashed down like a stone.
In the case of a declining / crashing real estate market you have the additional problem that houses and neighborhoods decline rapidly once you have crossed a certain threshold of foreclosures, so that even those who would have waited otherwise could not recover, especially in cases where the houses stood empty for quite a while.
Would the banks have agreed to let people live in their homes after foreclosure (e.g. as tenants), this wouldn’t have happened, but they didn’t and ended up with ghost neighborhoods and obligations instead. In my opinion, they could even have paid the former owners for taking care of the houses and still have made more money as they did…
The problem is that banks and bankers are herd animals, they always tend to follow the flock, and don’t really dare to act anticyclical.
That’s the reason why the rules that should normally bring more security into banking (like the new equity and valuation rules) do exactly the opposite in a crisis, because they prevent banks from riding out smaller problems, and force them to act in a shortsighted way that will eventually produce bigger problems and damage more people.
Here is Fisher’s sequence of events, for reference:
Philip, notice that Keynes’s depiction of the debt deflation theory, by your cites, begins at (3), not at “O”. So your argument for his familiarity with it is not proved.
Those of us in the Soddy school would put (2) at the top of the sequence, and might phrase it more precisely thus: “Extinguishment of deposit currency as debts to banks are paid.” Emphasizing that these debts include not only loans but fees, and that in the aggregate they aren’t paid “off”, but paid as they grow.
But Fisher’s theory, rotted in deposit currency maths, is more is more probative than Keynes’s, rooted in the mass psychology of Sumo wrestling, where each agent waits for the other to make the first move for fear of surrendering the advantage.
is more rottedrooted
(Well, we tried to chime in for Fisher, anyway.)
Obviously I am not making the case that Keynes knew Fisher’s characterisation of the theory two years before he had written it!
The essence of the theory is that a deflation increases the real burden of debt. Keynes was well aware of this and thought it was of secondary importance. I agree with him.
Debt liquidation does not lead to distress selling. A move of money into highly liquid assets leads to distress selling (in fact, they are two sides of the same coin). That is the essence of the liquidity preference theory.
@Philip The essence of the theory is that a deflation increases the real burden of debt. Keynes was well aware of this and thought it was of secondary importance. I agree with him.
The essence of the theory is in the excerpt I cited. Any argument that Keynes trumps Fisher should engage the actual Fisher, not the straw Fisher. A depiction of debt deflation rooted in price drops invites the policy intervention of pumping up prices rather than addressing the burgeoning indebtedness engineered into deposit currencies.
But here you’ve let Keynes speak for Fisher as well as for himself. No sale.
Extremely odd statement given that Fisher advocated reflation which, presumably, would start with a reflation of asset prices.
But again, I didn’t “let Keynes speak for Fisher”. I said that the increase in real debt due to deflation — the key component of the Fisher debt deflation cycle — was, in Keynes’ mind of secondary importance. In reality it is of secondary importance too.
On the contrary; Fisher’s policy proposal was called 100% Money. He identified a problem with the very engineering of deposit currency, and devised a program to address it. Notice in his introduction what “deflation” is; it isn’t the fall of asset prices.
100% Money can be read in its entirety at:
Some excerpts, emphases mine.
This is a full reimagining of our money system, not a mere reflation. Philip, as you characterize Fisher’s analysis and proposals, I appeal to your sense of fairness to your subjects, as well as to your thoroughness as a scholar.
We may have avoided a Liquidity Trap, but only because speculation remains fertile ground for new lending. Indeed, banks now do little else but lend for speculation, while the Fed props up asset prices by engorging MBS and everyone pretends the banks are not insolvent, while the economy limps along and savers are savaged by ZIRP. What sort of trap do economists call this?
“But, in classic Keynes fashion, what works for the individual does not work for everyone: when all the banks dump assets at the same time their money value drops so far that the banks themselves become insolvent.”
This does not accord with what happened in 2008. Banks and financial institutions had illiquid assets dropping in value which they could not dump and which did render them insolvent. At this point, the Fed entered in to the picture. It agreed to exchange in various ways these illiquid assets for money, the ultimate liquid asset. This provided liquidity to the banks and helped bolster illiquid asset prices, effectively establishing a market, or at least a buyer, for them. It also loaned money to the banks at ZIRP. This also injected liquidity into the financial system and acted as a subsidy to banks. Banks could simply borrow money from the Fed, leave it in their reserve accounts at the Fed, and draw interest off it, the interest being the subsidy and worth billions to the banks. These efforts of the Fed unfroze the banking system and even sparked bubbles in certain asset classes, like equities and commodities. What they did not do was trickle down to the real economy where most Americans live. A bubble for the rich and banks that could access this liquidity but nothing for the rest of us. I am not a fan of Krugman but I think his use of “liquidity trap” refers to this failure of injections of liquidity making it down to the level of the real economy. That Keynes had a somewhat different take on “liquidity trap” does not really advance our understanding of what happened in 2008.
Did you finish reading the article?
“This, of course, is precisely what we saw in 2008. But it is also what Keynes saw in 1931. The difference was that in 2008 we had a reaction from the central banks which printed massive amounts of money to prop up asset prices while in 1931 we saw extensive bank failures. ”
For the details of how the Fed exited the liquidity trap see here:
I read an explanation of QE that said that all it did was change the composition of banks’ books, not add money to the real economy. Although this explanation (in the Washington Post, and it actually seemed pretty on the level) did go on to concede that somehow this “change in composition” (swapping banks’ T-bond holdings for bank reserves in their simplified diagram of the operations in play) could “leak out” and raise stock prices or other asset classes.
How does this “change in composition” leak out? Is it the interest paid on reserves that Hugh mentions that has inflated the stock market? Because bank reserves themselves are not money, right? Also, will any significant percentage of the money that “leaked” into asset classes ever “leak” into the real economy, if only in the form of high-end luxury goods and services? Or will the vast majority of it just keep moving around in financial assets?
Do you or anyone have favorite lay-person friendly (yet not over-simplified) explanations of QE operations you could recommend?
Yes, I did one a while back.
Sorry, but it isn’t what we saw in 2008, precisely because illiquid assets could not be unloaded, and because of the Fed bailouts. Keynes did not describe the 2008 situation. Krugman’s description of a liquidity trap does. It is kind of ridiculous to say that 1931 was just like 2008 except for all the differences.
This post reminds me I’ve thought about reading Keynes’ Essays in Persuasion. But I hesitate as well because I remember a damning critique of Keynes from Veblen (talking about Keynes’ Economic Consequences of the Peace), where IIRC Veblen painted Keynes as completely naive and clueless (or perhaps complicit, I forget.) Anyway it dropped Keynes a little lower on my Must Read Someday list. Somewhere I have also seen Keynes characterized as somebody who really sought after and craved exactly the sort of august reputation he has achieved posthumously; ie that there was a lot of vanity in his make-up, a degree of toadying to public (or at least the fashionable public?) opinion.
Wow. What a great article.
Having read a little Keynes, I always thought Krugman misunderstood what he was saying about liquidity trap, but I’m not smart enough to pinpoint where he went wrong.
Now I know.
More important, the article points out the flaws in the Fed’s policy which–as you say: “Central bank interventions can certainly stabilise the financial markets …but only large increases in aggregate expenditure can reflate the economy.”
Nicely put. I wish Yellen read this blog. it might help.
Re: “Wow. What a great article.”
I certainly agree. Thank you, Philip. Especially appreciated your concluding sentence.
And thank you, Yves, for posting this.
I think it’s important to differentiate credit contraction from a liquidity trap. When people start to realize that some of their speculative plays such as investments in emerging markets, fx, derivatives and ninja loans are getting too risky they might try to retreat into Treasuries. This is probably a necessary de-leveraging but very disruptive if done at once and ignoring fraud. Witness the Fed’s efforts to stabilize money market funds. And their current efforts to increase their authority in this area.
I tend to view liquidity trap as the trap where supplying liquidity alone (monetary policy) doesn’t work to get the economy moving again.
I liked Randy Wray’s discussion of a liquidity trap. He describes it from a somewhat different angle and in terms of what he thinks was the essence of Keynes thinking on the matter.
He describes it as ‘marginal efficiency of capital’ or easiest for me to understand ‘profit expectations’ which can actually turn negative making producers not want to invest in production as there are not enough people to buy their product (demand) so they would lose their money. (negative profit). The variable is not interest rates but profit expectations.
I like this reframing of a ‘liquidity trap’ as a ‘demand trap’. Even if several million dollars are available at low interest producers aren’t going to put that money into production unless there are enough employed people to buy the product. And this takes fiscal policy. (and we have a Congress that is out to lunch)
Some companies are starting to understand that federal deficits can correlate positively with company profits. The key is for these deficits to stimulate real demand rather than try and stimulate borrowing using the confidence fairy.
Our failure to understand our eusocial nature has resulted in the naivety of believing it is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own “interest” because in practice a collective failure to balance supply by demand will always stifle that “interest.”
Thanks FM, this makes clear sense to me; PP just left me in the dust. Capitalism became a force of nature by the logic of the market but by the MIC and the western world manufactured everything it could imagine. But the mantra that the market would take care of the details proved to be wrong. Global warming isn’t helping the market, clearly, nor is the destruction of the environment. The market needs something like military expenditures to jump start itself. If only it could be used to fix things instead of more of the same destruction. We put an impossible requirement on a “free market” (including ignoring the fact that it doesn’t exist) – expecting private banking and individual enterprise to drive it. The incentive for everyone becoming an entrepreneur was always profit and nothing else. So when such a narrow incentive falls victim to a demand trap it all stops working. We could use a little sober guidance going into the future. Monetary tweaks can’t replace policy. Money should be plentiful because politically it can always be had; interest rates should be ultra low because money is political; and profit should be capped so there is no frenzy to seek returns on money itself which has no intrinsic value. With the Bancor, the BIS would have charged surplus countries interest on their profits too… don’t know why that can’t be applied to domestic markets. Taxes don’t seem to be working. The only game in town right now is government subsidizing the big, almost useless, corporations and that can’t go on for long.
became a force of nature not by the logic of the market (also replying to FM)
“The market needs something like military expenditures to jump start itself”
Yes – isnt this how capitalism got turbo-charged in the 19th century, by imperialism and colonialism. It took massive state spending on these ventures to make capitalism.
“…The amount will, of course, vary in different cases within wide limits. But for marketable assets a “margin” of 20 per cent to 30 per cent is conventionally considered as adequate…”
Hmm, I wonder what allowed financial companies in the 1990s and 2000s to lever up at more than 3 – 4X.