VoxEU Column Macroeconomic policy

Central Bank reserve creation in the era of negative money multipliers

Are central banks printing vast quantities of money? This column explains how money-multiplier economics (central banks create reserves that allow commercial banks to create money) no longer holds. Today, non-bank financial institutions play a pivotal role in money/liquidity creation, but hold no reserves. Their lending depends on “private reserves”, mainly highly liquid government securities. Creating more ‘public’ reserves by buying such ‘private’ reserves doesn’t trigger money creation – it just substitutes among reserve types. Open-market purchases only create money if they swap a monetary base for assets that are no longer accepted at full value as collateral in the market.

The phenomenal increase in bank reserves that has resulted from central bank responses to the current financial crisis has led to considerable anxiety regarding a potentially explosive and uncontrollable future increase in inflation. Virtually identical concerns within the Federal Reserve in late 1935 motivated large increases in reserve requirements during 1936 and 1937; actions widely regarded as contributing to the sharp 1937-8 recession (see Friedman and Schwartz 1963).

Have monetary economists and policymakers learned from that experience, or is a repetition of premature tightening possible owing to a misunderstanding of the role of bank reserves in the modern financial system and in the functioning of the money supply process?

Money neutrality

The title of Robert E. Lucas Jr.’s 1995 Nobel Prize lecture is “Money Neutrality”. Generations of graduate students in economics have been schooled in this theory and its impact on how people think about monetary policy cannot be overstated. Lucas summarised the implications of the theory as “The central predictions of the quantity theory are that, in the long run, money growth should be neutral in its effects on the growth rate of production and should affect the inflation rate on a one-for-one basis”.

The early canonical works on money neutrality pointed to the superiority of simple monetary policy rules that enable market participants to easily predict future money supplies. There was no discussion in these highly abstract models of the need for an institution—such as a “central bank”—that might be required to undertake actions to implement a rule-based policy, nor was a distinction made between direct liabilities of the central bank (the monetary base) and money more broadly defined (see Lucas 1972 and Kydland and Prescott 1977).

Conceptually, the gap between variables that are directly under the control of the central bank (a short-term interest rate or the monetary base) and the variables that are deemed to matter for inflation (M1, M2 or the term structure of interest rates) falls under that part of monetary theory known as the “transmission” mechanism.

Textbook presentations of the actual US monetary policy transmission mechanism frequently characterise it as follows:

  • The Federal Reserve uses open market operations to inject or withdraw commercial bank reserves.
  • These banks then create money via the “money multiplier”.

In short, banks respond to an injection of additional reserves by creating loans financed with monetary liabilities such as checking and savings deposits.

The money multiplier is an algebraic expression that predicts the final change in money supply that would result from a given change in the monetary base. So if the M2 multiplier were 5, an injection of bank reserves of $1 billion would be expected to expand M2 eventually by $5 billion. Regardless of the value of the money multiplier, provided it is stable, a given percentage increase in the monetary base would lead eventually to an identical percentage increase in money. Thus the “theory” of the money multiplier is a concise way of tying together a simple policy rule under the full control of the central bank with money and inflation.

Reserve creation without money creation

The application of these two concepts, money neutrality and the money multiplier, have led many contemporary observers of the Federal Reserve’s response to the current financial crisis to be concerned with the potential consequence on inflation. For example, in his Financial Times op-ed of 19 April 2009, Martin Feldstein said that “… when the economy begins to recover, the Fed will have to reduce the excessive stock of money and, more critically, prevent the large volume of excess reserves in the banks from causing an inflationary explosion of money and credit”.

So far, despite a several thousand fold increase in bank reserves, US inflation shows no sign of emerging from Pandora’s Box (see Table 1).

Table 1. US bank reserves and inflation

  2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Financial institution deosits at the Federal Reserve (US$ billion) 19.0 17.5 22.5 23.1 24.0 19.0 18.7 20.8 860.0 977.0 968.1 1562.3
Annual Change in FI deposits at the Federal Reserve (%) -20.8 -7.9 28.6 2.7 3.9 -20.8 -1.6 11.2 4034.6 13.6 -0.9 61.4
Change in the US CPI (%) 3.4 1.6 2.4 1.9 3.3 3.4 2.5 4.1 0.1 2.7 1.5 3.0


This suggests either that there is something wrong with:

  • the theory of money neutrality;
  • the theory of the money multiplier; or
  • how money is measured.

Alternatively, the market may expect the Fed to reverse its liquidity injection—which is another way of saying we are not yet in the “long run”.

New research

In Singh and Stella (2012) we propose two angles from which this seeming anomaly can be explained, both of which require a fundamental reassessment of the textbook money multiplier.

The first relies on a correct interpretation of the liquidity needs and management of a modern financial system which comprises not only conventional banks but also financial institutions operating in their shadows. Such non-bank financial institutions do not have access to monetary base as they hold no reserves at the central bank. Instead they rely on their access to the repo market predicated on their ownership of what is perceived to be highly liquid collateral.

Due to this shift, the liquidity fulcrum of the pre-crisis US financial market was composed only partly of central bank liabilities—and it was a very small part. More importantly, the magnitude of the liquidity fulcrum was determined not by the monetary policy authorities but instead by market practice. The nature and volume of assets determined by the market to be acceptable collateral is the key.

The ‘negative money multiplier’

Only this evolution of market-determined liquidity creation and management can explain these rather astounding facts:

  • Total credit market assets held by US financial institutions (excluding the monetary authorities) rose by $32.3 trillion from 1981 to 2006 (744%).
  • Commercial bank reserves held as deposits at the Federal Reserve fell by $6.5 billion during the same period1.

In fact, total commercial bank reserves at the Federal Reserve amounted to only $18.7 billion in 2006, less than the corresponding amount, in nominal terms, held by banks in 1951. So from the first angle it is fairly clear that not only have financial institutions not relied on an increase in reserves held at the Fed to expand credit, they have expanded credit by 744% while reserves fell. Thus the marginal money multiplier of increased bank reserves has been highly negative if not irrelevant. The relationship between bank reserves and inflation is also tenuous as shown in Figure 1.

Figure 1. Bank deposits at the Federal Reserve and inflation 1950-2007

Substituting public reserves for ‘private’ reserves

This second source of our sanguinity comes from the nature of the reserve creation. The enormous increase in bank reserves reflects a substitution of monetary base largely for highly liquid government securities in private-sector portfolios. But as government securities serve as collateral in the private-credit market, the effective size of the market liquidity fulcrum is unchanged by such operations. Little wonder then that market liquidity conditions remain tight despite the increase in bank reserves. That is, although quantitative easing which merely swaps bank reserves for US Treasury bills increases the textbook monetary base and “should” lead to an increase in market credit, in our view this accomplishes virtually nothing in terms of easing liquidity pressures. It merely changes the composition of assets within a given sized liquidity fulcrum.

In this view, a central bank open market purchase changes the size of the liquidity fulcrum only if it swaps monetary base for assets that are no longer accepted at full value as collateral in the market. Such moves go beyond a pure swap of ‘private reserves’ for public reserves and ease effective liquidity by increasing the overall market liquidity fulcrum.
Indeed, the ‘private reserve’ stock took a big hit with the evaporation of previously acceptable collateral asset classes (e.g. private label CMBS) and, in reaction, the market liquidity fulcrum collapsed during the 2008/09 Global Crisis. This collapse was been only partially offset by the substitution of monetary base for collateral no longer traded in the market.


Those schooled in the conventional money multiplier approach are understandably worried about the potential impact on inflation of expanding bank reserves. Nevertheless, there is good reason not to fear the money multiplier. Post-war US credit expansion – and, by inference, inflation—has not been dependent on an expansion of bank reserves and there is no reason to expect there will arise now a causal impact of the latter on the former. The liquidity fulcrum of a modern financial system is more complex than the monetary base and its size is determined by market conditions which continue to show signs of strain despite comparatively massive central bank injections of bank reserves.


Friedman, Milton and Schwartz, Anna J. (1963) A Monetary History of the United States, 1867-1960, Princeton University Press, Princeton, New Jersey, pp 520-21 and 543-544.
Kydland, Finn E. and Edward C. Prescott (1977) “Rules Rather than Discretion: The Inconsistency of Optimal Plans” Journal of Political Economy 85-3.
Lucas, Robert E. (1972) “Expectations and the Neutrality of Money”, Journal of Economic Theory 4.
Singh, Manmohan and Peter Stella (2012), “Money and Collateral”, IMF Working Paper 12/95.

1 See US Flow of Funds data for the corresponding years.


15,013 Reads