Excessive Liquidity Preference

Any recession by definition is associated with an excessive liquidity preference.  An ex ante excess supply of goods and services, i.e. the demand for goods and services falling short of the base output at the base prices corresponding to that output, which is what a recession is, must be associated with an ex ante excess demand for money at that output and prices.  This is simply Walras’ Law.  The present recession in world economy however is characterized by “excessive liquidity preference” in a somewhat different and far stronger sense, namely it is an ex ante excess demand for money at the base prices and output that is not overcome through larger supplies of money.

There is also a second aspect to the current “excessive liquidity preference”.  The fact that the ex ante excess demand for money does not disappear at the base prices and output even when the supply of money in the possession of economic agents increases, holds not just for the economic agents that are usually called “the public”1; it holds for banks as well, indeed for all economic agents including both the “public” and all manner of financial institutions.  The demand for money at the base output and base prices corresponding to it, in short, is infinitely elastic, so that all increases in money supply to the “public” are simply held without preventing a fall in output/prices.  What is more, if money, of the “high-powered” variety, is directly injected into the banks and other financial institutions, then it too is simply held by these institutions without causing any increase in the money supply with the “public”.

Of course, the impact of variations in the ex ante excess demand for money on the output/prices of the non-money goods and services is mediated through the interest rate.  It follows then that the “excessive liquidity preference” characterizing the current recession is one where at the prevailing interest rate there is an infinitely elastic demand for money both among the “public” and among the financial institutions, especially banks.  Keynes’ (1949) “liquidity trap” related only to a situation of infinitely elastic demand for money at the going interest rate by the “public”.  The present situation of an infinitely elastic demand for money by both “the public” and also by banks constitutes in Stiglitz’s words a new version of the liquidity trap.  We may call it “liquidity trap in the broad sense” as distinct from Keynes’ liquidity trap, which is “liquidity trap in the narrow sense”.

The hallmark of any liquidity trap is that an injection of liquidity by the monetary authorities has no impact on the interest rate, i.e. the interest rate is at a floor level.  In Keynes this was because everyone holding bonds at this floor rate of interest expected the bond prices to fall at a rate equal to this interest rate; that is, everyone holding bonds was at the money-bonds margin and nobody was a “bull”.  If we take an alternative interest rate theory where we do not have “two views” among wealth-holders but instead “a dense concentration at the margin”2, then a floor to the interest rate can be explained by the fact that the short-term interest rate is at its lowest possible level (almost zero) while the long-term interest rate consists almost exclusively of the risk-premium.  (A variation of this, apposite to the context of the present crisis, is that the rate on short-term government debt is zero and all other interest rates are in accordance with the relative risks associated with holding those particular debts).

There is however a misconception about the liquidity trap which should be clarified.  There is a tendency in the context of the current recession to say that “fresh credit has dried up because the economy is in a liquidity trap”.  This is wrong.  The economy’s being in a “liquidity trap in the broad sense” does not rule out banks or individual wealth-holders adding to their stock of bonds at a price lower than what corresponds to the liquidity trap.  Hence a liquidity trap does not mean a drying up of fresh credit; it only means that fresh credit becomes available at an interest rate higher than the floor rate.  Hence the concept of the liquidity trap, whatever the sense in which we use the term, refers only to a state where the interest rate cannot fall, not to a state where fresh credit has dried up.  The two things, i.e. the non-availability of fresh credit at the going rate of interest, and the non-availability of fresh credit per se, are not identical.  The liquidity trap refers to the former not the latter.

There is also a second possible confusion here.  The non-availability of fresh credit at the going interest rate in a liquidity trap does not mean that entrepreneurs are credit-constrained.  The liquidity trap does not refer to a case of rationing; it is not a rationing equilibrium.  The entire stock of private debt is held at a price corresponding to the going interest rate in the liquidity trap.  But additions to this debt are not held at the going interest rate. It is fresh credit that is given only at an interest rate higher than the going rate.  But this is not the same as saying that the state of the liquidity trap is a state of credit rationing.  It is an equilibrium state where the demand for and the supply of money are equal; and the same is true of the demand for and the supply of private debt.

To come back to the main argument, the current recession is marked by the fact that at the going interest rate there is an infinitely elastic demand for money both among the “public” and among financial institutions, because of which any increase in money supply by the monetary authorities does not overcome the ex ante excess demand for money, and hence the ex ante excess supply of goods.  There is excessive liquidity preference in this strong sense.


Let us see what are the possible ways in which such excessive liquidity preference may be overcome whenever it happens to arise.  The most obvious is penalizing liquidity preference by putting a tax on the holding of money.  And the most famous suggestion in this regard was put forward by Silvio Gessel, a non-Marxist socialist, who was much appreciated by Keynes and was the Finance Minister for a while of the short-lived post-first world war Bavarian Soviet.  Gessel’s idea was to make the acceptability of the currency dependent upon its carrying a stamp that had to be purchased and renewed periodically.  This introduced a carrying cost into money holding, discouraging liquidity preference.  Keynes, while appreciating the idea and even suggesting that the rate of carrying cost should be so adjusted that it equalled the difference between the interest rate and the marginal efficiency of capital corresponding to the desired level of employment, was quite skeptical of it for an obvious practical reason, namely that money came in different forms and that such a scheme, even if successfully applied to currency, would leave other forms of money untouched.  But Gessel-type ideas, such as restricting the magnitude of government security holding by banks, and thereby forcing them to hold private debt, which would stimulate private expenditure without curtailing government spending (since government debt would be easily held in lieu of money by the “public”), have surfaced from time to time during the current recession.

The second which monetarism has suggested all along and which constitutes the core of the counter-revolution against Keynes is to let both money wages and prices drop.  While monetarism does not talk in terms of any liquidity trap, its simple remedy to overcome ex ante excess demand for money, and the corresponding ex ante excess supply of goods, is merely to allow the price mechanism to function, i.e. to let the price of money in terms of non-money goods, which includes labour-power, increase.  If money wages and prices fell, i.e. if the price of money in terms of non-money goods increased, then, as in standard text-book demand theory, the demand for money would fall and of non-money goods rise, which would cure the economy of recession.  Recession in short can arise only because of price rigidity, or, if non-labour prices are flexible (as Keynes believed), then wage-rigidity.  In the absence of these rigidities, the market always functions to keep the economy in a state that precludes involuntary unemployment.

The problem with this argument is that in the case of money the simple dictum “when price rises demand falls”, which is supposed to eliminate involuntary unemployment, does not necessarily hold.  There are several reasons for this.  One obvious reason relates to expectations.  Since money is held as wealth, along with other forms of holding wealth, expectations about the future must enter into its demand.  If the rise in the price of money gives rise to expectations of further rise, then liquidity preference, far from getting discouraged, is further buttressed.3  A second reason relates to Irving Fisher’s “debt-deflation” syndrome (1933).  A rise in the price of money in terms of non-money goods raises the real burden of inherited debt which is fixed in money terms.  To reduce the debt burden economic agents start selling goods, so that the rise in the price of money increases, rather than decreases, the ex ante excess supply of goods (an argument that has nothing to do with holding money as a form of wealth).  A third reason relates to what we have been discussing till now.  If the economy is in a liquidity trap, so that a rise in the price of money cannot lower the nominal rate of interest further, then it necessarily leads to an increase in the real rate of interest.  Since it is the real rather than the nominal rate of interest that affects investment and expenditure decisions, such a rise in the real rate lowers the demand for goods and services, compounding the recession.  Putting it differently, a rise in the price of money, far from lowering the ex ante excess supply of goods and services, further increases it.  Thus, the standard monetarist perception of how excessive liquidity preference can be overcome is an erroneous one; the recession that corresponds to such excessive liquidity preference is a result not of constraints upon the functioning of markets, such as wage rigidity, but of the functioning of markets itself.

The third possible way of overcoming excessive liquidity preference is to lower the risks associated with the holding of private debt.  And since the perception of this risk is linked not only to the nature of the debt but also to the balance sheets of the creditors, notably the banks and other financial institutions, improving this balance sheet becomes a possible way of overcoming excessive liquidity preference and hence the recession.  This indeed is what the Obama administration, like the Bush administration before it, has been trying to do.  The Geithner Plan, like the Paulson Plan earlier, which seeks to “bail out” the financial system is concerned with this.

There are two quite separate problems which the U.S. financial system currently faces and which the U.S. administration has been trying to tackle.  The first relates to the fact that the fall in the value of assets of the financial system generally, and of banks in particular, together with the losses on loans, exceeds the total capital backing of the assets.  For instance, in the case of banks the following estimates have been made.  The total losses on loans by U.S. financial firms and fall in the value of their assets is estimated to peak at $3.6 trillion, of which about half or $1.8 trillion have to be borne by banks and broker dealers.  The total capital backing of banks’ assets on the other hand comes to only $1.4 trillion (prior to the various “bailout” plans).4  This means that the U.S. banking system is at present basically insolvent, requiring significant recapitalization just to cover the losses and further substantial capitalization to restore its capacity to lend to private borrowers.

The second problem relates to the fact that some of the assets, the so-called “toxic” assets, which the banks hold are not immediately saleable.  Their true market price at present is not even known, the estimates mentioned above being just rough ones.  Hence even if the banks are recapitalized adequately they would still be left with some assets that are not easily saleable.

Now, all “bailout” packages have to be basically concerned with both these problems, though both the Paulson and Geithner packages till now have been concerned solely with the problem of “toxic” assets, i.e. with improving the liquidity of the banks’ assets.  The Obama administration has yet to concern itself seriously with the more basic issue of recapitalizing banks.  Some have even argued that the Geithner plan, concerned as it is with the “toxic assets” issue, may, for that very reason, create hurdles in the way of capitalization, since the impression inevitably would arise that the administration is helping the banks twice over.5

We need not go into these plans here since they have been much discussed already, and since the basic issue with which we are concerned happens to be altogether different.  The basic issue is that even if the Geithner plan succeeds it will still not revive the economy.  Indeed even if the banks are rid of “toxic” securities, and even if they are adequately capitalized, that still would not revive the economy.  To see why, let us assume that the Geithner plan succeeds and also that the Obama administration succeeds in pushing through the legislature a capitalization plan over and above the Geithner plan.  The net result of these two measures would be to reduce the risks associated with lending to private borrowers by banks.  The banks in other words get out of their liquidity trap and are willing to buy private securities at current prices.  But the public would still be in a liquidity trap.  The economy in other words would have got out of the “liquidity trap in the broad sense” but would still remain in the “liquidity trap” in the narrow sense, i.e. in the Keynesian liquidity trap.

This means that banks, even though they may be willing to lower the interest rate, will not be able to do so, since even an infinitesimal lowering of the interest rate would place before them an extraordinarily large supply of securities from the “public”.  Putting it differently, even if banks get out of the liquidity trap, as long as the “public” remains within a liquidity trap, the interest rate cannot be lowered.  And if the interest rate is not lowered, then there is no reason why the ex ante expenditure stream and hence the demand for credit should increase.  In other words, the sheer fact that banks are willing to lend more at the going interest rate does not in itself stimulate larger expenditure (unless the marginal efficiency of capital schedule shifts outwards, for which there is no reason).  Banks’ greater willing ness to hold private debt, to be effective in stimulating expenditure, must lead to a lowering of the interest rate.  But such a lowering requires not just banks getting out of the liquidity trap but also the “public” getting out of the liquidity trap.  “Toxic-asset-purchase”-cum-capitalization packages for banks get them out of the liquidity trap but do not necessarily get the “public” out of its liquidity trap.  And if the “public” remains stuck in its own liquidity trap, then there is no question of the economy getting out of recession via this route.

Plans for rescuing the banking system, no matter how this rescue is done, can therefore constitute a way out of the crisis only if they simultaneously get the “public” out of its liquidity trap, which means affecting the expectations and risk perceptions of the “public” in a manner that makes it more willing to hold private debt at the going interest rate. This would happen if the “public” becomes more “bullish” about bond prices or if it considers bond holding less risky, and no doubt both these possibilities are strengthened when the banks become more willing to hold bonds.  But while these possibilities get strengthened, they do not necessarily get realized, if for no other reason then the sheer fact that the public’s perceptions take time to change even after banks have become more willing to hold private debt at the going interest rate.  In such a case, the persistence of the “liquidity trap in the narrow sense” will mean that the interest rate will not fall, expenditures will not rise, and the recession will continue, which will in turn justify retrospectively the public’s unwillingness to hold larger private debt at the going interest rate, even after banks become more accommodating towards such debt.

It follows that rescuing banks does not itself constitute a means of overcoming excessive liquidity preference and hence the recession.  At the very least it has to be combined with some other measure that acts simultaneously to convince the public to be more accommodative towards private debt, some measure that is directly counter-recessionary and hence acts directly to improve the prospects for the economy and hence the attractiveness of private debt.  It is obvious that this measure, which is the fourth in our list, must consist in the fact that even as there is excessive liquidity preference on the part of the public and the financial institutions, some agency should consciously set itself to do the very opposite, namely to move from money to goods, or to borrow money to buy goods.  Such an agency can only be the State.  Government expenditure on goods and services financed by borrowing constitutes the real antidote to excessive liquidity preference.

First of all it is direct and certain.  Even in the liquidity trap, government debt is considered to be as good as money, if not better, so that the government has no problems whatsoever in raising as much resources as it likes at the current interest rate on government debt.  The government of course can always print money if necessary, but the need for it does not even arise.  The very fact of the economy being in a liquidity trap ensures the ease of government borrowing.  At the same time government expenditure directly generates demand for goods and services; by contrast, overcoming the liquidity trap of the public, even if it could be done, could only generate additional demand for goods and services indirectly, via stimulating private expenditure whose sensitivity to the interest rate is itself uncertain.

Secondly, such a fiscal thrust itself has the effect of improving the prospects, and hence the value, of private debt and thereby the balance sheets of banks.  In the context of the Geithner Plan there has been much discussion in the United States about what the prices of the toxic assets which are supposed to be taken over by private buyers with government guarantees are likely to be in the coming months.  The prices of these assets, as indeed of private debt generally and of other assets, will depend very much on the strength of the recovery and hence upon the magnitude of the fiscal stimulus itself.  Thirdly, for exactly this reason, overcoming the Keynesian liquidity trap and getting the public to be more disposed towards holding private debt becomes easier when a fiscal thrust is undertaken.  It follows then that while the propagation to the real economy of an impulse given to the financial sector is dubious, the propagation to the financial sector of an impulse given to the real economy is far more certain.  The point of intervention for coming out of a recession must be a direct State-delivered impulse to the real economy.

When there is excessive liquidity preference in the strong sense, i.e. an ex ante excess demand for money at base prices and output is not overcome through larger injections of money into the system by the monetary authorities, then obviously the cause of the problem lies not on the money side but on the real side, namely the unwillingness of wealth-holders to hold real assets or claims on real assets.  Hence this is best overcome through a direct intervention by the government increasing the demand for the latter.


This has an important bearing for the functioning of markets.  A market system, where there are asset markets dominated by speculators who are concerned exclusively with asset price movements and not with the yields on the assets, can function only if there are inelastic price expectations in such markets (Patnaik 2008).  Inelastic price expectations amount to saying that excessive divergences of price in either direction from some central value tend to be self-correcting, i.e. that speculation itself acts as its own antidote.  The fact that excessive liquidity preference, whose counterpart is the ex ante excess supply of goods and services, does not get eliminated through a rise in the price of money relative to goods and services including labour power is proof that speculation cannot be its own antidote, that price expectations are not inelastic.  A market system requires for its functioning therefore an antidote to its instability that is necessarily “external” to itself.  The crucial question of our time is the terms on which capitalism would be able to negotiate for itself, if at all, the arrangement of such an “external” antidote; the answer to this question however depends on class struggle.



1  Analytically the term “public” refers to any agent whose liability does not constitute “money”.

2  The distinction between these two theories is discussed in Kahn (1954).  Kalecki (1954) may be taken as the classic exponent of this alternative theory to Keynes, though the basic idea of this alternative approach is contained in Kador (1964) and Hicks (1962).

3  This point is discussed in detail in Patnaik (2008).

4 These estmates are taken from Roubini (2009).

5  This point has been made by Martin Wolf (2009).



Fisher Irving (1933) “The Debt-Deflation Theory of Great Depressions”, Econometrica.

Hicks J.R. (1962) Value and Capital, Clarendon Press, Oxford.

Kahn R.F. (1954) “Some Notes on Liquidity Preference”, Manchester School, reprinted in Selected Essays on Employment and Growth, Cambridge, 1972.

Kaldor N. (1964) “Own Rates of Interest” in Essays on Stability and Growth, Duckworth, London.

Kalecki M. (1954) The Theory of Economic Dynamics, George Allen and Unwin, London.

Keynes J.M. (1949) The General Theory of Employment, Interest and Money, Macmillan, London.

Patnaik P. (2008) The Value of Money, Tulika Books, Delhi, also by Columbia University Press, New York, 2009.

Roubini N. (2009) “Time to Nationalize Insolvent Banks”, February 26, 2009.

Wolf Martin (2009) “Successful Bank Rescue Still Far Away”, March 24, 2009.

Prabhat Patnaik is an economist at the Centre for Economic Studies and Planning in the School of Social Sciences of Jawaharlal Nehru University in New Delhi.   This paper was delivered at the International conference on ”Re-regulating Global Finance in the Light of the Global Crisis” organized by International Development Economics Associates (IDEAs), School of Economics and Management, Tsinghua University, Beijing and School of Economics, Renmin University, Beijing, 9 -12 April 2009, Beijing, China.  It is made available by IDEAs in PDF at <networkideas.org/ideasact/feb09/Beijing_Conference_09/
>.  It is reproduced here for educational purposes.