1. Central banks in chaotic times Marc Lavoie The subprime crisis, monetary policy implementation, and changes in monetary theory

2. Motivation The financial crisis has forced central bankers to modify their procedures. It has created a communication problem, forcing them to explicitly reject some elements of textbook monetary theory. It also has some implications for heterodox (post-Keynesian) monetary theory. As a case study, I look at the Federal Reserve (IJPE, Fall 2010).

3. Outline Some preliminaries on monetary theory and central bank operations. Timeline: Evolution of operating procedures of the Fed since August 2007. Implications for PK monetary theory Implications for textbook monetary theory and the Public Relations problem of central banks today.

4. Preliminaries I: mainstream (textbook) monetary theory The central bank controls the amount of bank reserves or high powered money through open market operations. The purchases of bonds, from dealers, by the central bank, create new deposits (and reserves) in the banking system. This allows commercial banks to make new loans. This leads to a multiple creation of money deposits, through a money multiplier story. Excess reserves lead to an excess supply of money, which leads to price inflation. Click View then Header and Footer to change this text

5. Even Paul Krugman holds this mainstream view, in his NYT blog, 30 March 2012 “It’s obvious that many commenters don’t get the distinction between the proposition that banks create money — which every economics textbook, mine included, says they do (that’s what the money multiplier is all about) — and the proposition that their ability to create money is not constrained by the monetary base. Sigh. As I read various stuff on banking…I often see the view that banks can create credit out of thin air. There are vehement denials of the proposition that banks’ lending is limited by their deposits, or that the monetary base plays any important role; banks, we’re told, hold hardly any reserves (which is true), so the Fed’s creation or destruction of reserves has no effect.” Click View then Header and Footer to change this text

6. Paul Krugman continued “This [the PK view] is all wrong, and if you think about how the people in your story are assumed to behave — as opposed to getting bogged down in abstract algebra — it should be obvious that it’s all wrong. First of all, any individual bank does, in fact, have to lend out the money it receives in deposits. Bank loan officers can’t just issue checks out of thin air; like employees of any financial intermediary, they must buy assets with funds they have on hand. I hope this isn’t controversial… the recipient of the loan can and sometimes does quickly withdraw the funds, not as a check, but in currency. And currency is in limited supply — with the limit set by Fed decisions.” Click View then Header and Footer to change this text

7. Preliminaries 2: The post-Keynesian view of money creation Banks look for credit-worthy borrowers first. Then they try to find the reserves they need. Loans make deposits, deposits make reserves Reserves are demand-determined: Central banks provide the reserves (and the banknotes) needed by the system, at the rate of interest of their choice (the target rate). The money supply is demand-determined The money multiplier is an accounting identity –It plays no behavioural role. –It must be banned from all textbooks. If anything, there is a credit divisor. Click View then Header and Footer to change this text

8. THE TIMELINE Click View then Header and Footer to change this text

9. Source: Keister, Martin, McAndrews (2009)

10. The timeline at the Fed I Financial pressures started in mid-August 2007. Real financial pressures started at the end of December 2007. The Fed is forced to make use of its lending facilities, providing loans (liquidity) to banks. The expansionary effects of the central bank loans are neutralized by open market operations. Until 12 September 2008, the Fed is able to move the federal funds rate next to the FMOC target interest rate (2%). It is able to do so because the neutralizing operations of the Fed keep reserves at their approximate required level.

11. The balance sheet of the Fed until 18 September 2008 AssetsLiabilities Foreign reservesCash (banknotes) Credit to the domestic government (Treasury bills) Deposits of banks (reserves) (deposit facilities) Credit to the domestic private sector (lending facilities) Government deposits Central bank bills Click View then Header and Footer to change this text

12. Initially, the Fed keeps the supply of reserves in line with the demand for reserves, and the expected fed funds rate is the target rate Reserves Expected Fed funds rate Lending rate Deposit rate 0% Demand for reserves Supply of reserves Target Fed funds rate

13. Click View then Header and Footer to change this text Period 200120022003200420052006 2007-01-01 2007-08-08 Standard deviation 17.75.26.1 3.9 4.2* 7.15.42.9 Peak monthly value 50.1 (Sept: 9/11) 8.413.27.010.98.33.4 Period 2007-08-09 2007-09-14 2007-09-17 2007-12-31 2008-01-01 2008-09-14 2008-09-15 2008-10-08 2008-10-09 2008-11-05 2008-11-06 2008-12-05 2008-12-16 2010-04-30 Standard deviation 18.8 10.8 15.6* 9.062.816.919.44.1 Peak monthly value 23.5 (Dec) 11.05.2 Standard deviation of discrepancy between effective and target fed funds rate

14. Timeline at the Fed II On 15 September 2008, the Lehman Brothers declares chapter 11 bankruptcy (a buyer could not be found and Paulson/Bernanke declined to nationalize it). The interbank market freezes: banks don’t lend to each other. Most banks in a long position at the clearing house prefer to keep their excess funds as deposits at the Fed. Banks in a short position at the clearing house are forced to borrow from the Fed.

15. Timeline at the Fed II (bis) From then on, the Fed is unable to achieve its target (fed funds rate is at first too high, then too low) From 19 September 2008, the Fed decides to inject huge amounts of liquidities (provide huge lending facilities to the banks (and AIG)), without conducting compensating open market operations. This is when the balance sheet of the Fed starts exploding. Click View then Header and Footer to change this text

16. Timeline at the Fed II: Balance sheet explosion At first, the Fed asks the Treasury to issue securities, sold to the banks and dealers, and to deposit the proceeds in its account at the Fed, thus partially neutralizing the reserve-creating effect of granting advances to banks. But within a few weeks this is abandoned, as the Treasury approached its legal debt limit defined by Congress (of which we all became aware in the Summer on 2011). The Fed then gives up on its attempt to neutralize the reserve-creating effect of granting advances to banks. Click View then Header and Footer to change this text

17. The balance sheet of the Fed starting on 19 September 2008 AssetsLiabilities Foreign reservesCash (banknotes) Credit to the domestic government (Treasury bills) Deposits of banks (reserves) (deposit facilities) Credit to the domestic private sector (MBS, toxic assets …) Government deposits Central bank bills Click View then Header and Footer to change this text

18. Fed loses control of the federal funds rate: too high, then too low Reserves Target rate Lending rate Deposit rate = 0 Demand for reserves S Fed funds rate S’’ S’ Fed funds rate 15-16 Sept.

19. Timeline at the Fed III: Trying to regain control of fed funds rate On 6 October 2008, the Fed gets the authority to pay interest rates on (excess) reserves, thus setting up a corridor system, with a ceiling (the discount rate) and a floor (the interest rate on reserves). Despite this, the Fed does not regain control of the federal funds rate, with a target at 1.50% and fed funds rate hovering between 0.67% and 1.04%.

20. This is what should have happened with the corridor system: the fed funds rate should be at least equal to the deposit rate Reserves Target Fed funds rate Lending rate Deposit rate Rate of interest on reserves Demand for reserves S S’’ S’

21. Timeline at the Fed IV On 6 November 2008, the interest rate on all reserves at the Fed is set as the target fed funds rate (1%). Despite this, the fed funds rate hovers between 0.10% and 0.62%, getting ever lower. Finally, on 17 December 2008, the Fed announces a target between 0 and 0.25%, with a rate on reserves at 0.25%, and actual fed funds rate in 2009 (and still now) is between 0.10 and 0.24%.

22. The floor system: this is what should have happened with the target rate set at the deposit rate, and with a large amount of reserves : the fed funds rate is exactly equal to the deposit rate Reserves Target rate and Deposit rate Lending rate Demand for reserves S 1% S’

23. Why doesn’t the fed funds rate stay at the bottom of the corridor? Not all participants (GSEs: Freddie Mac, Fannie Mae) to the fed funds market are eligible to receive interest on their reserve balances. There are also foreign institutions that hold balances at the Fed that don’t get interest on reserves. They may thus lack bargaining power and being forced to lend their surplus funds at a rate below the floor.

24. IMPLICATIONS FOR PK MONETARY THEORY

25. The Decoupling Principle With the corridor system, achieving a higher target overnight rate did not require any change in reserves. With the target interest rate set at the floor of the corridor, central banks (FED, BOC) can now set the target rate at the level of their choice and simultaneously set the amount of reserves at the level of their choice. There is no relationship anymore between reserves and overnight rates. This is the decoupling principle ( Borio and Disyatat 2009 ) Central banks can have an interest rate policy divorced from a supply-of-reserves policy.

26. The floor system The floor system was recommended by Woodford (2000, p. 255), Goodfriend (2002, p. 3), Ennis and Keister (2008), and in more detail by Keister et al. (2008). It was also advocated by the post- Keynesian/Institutionalist Scott Fullwiler (2005) The floor system was adopted in New Zealand and Norway, before the crisis! Click View then Header and Footer to change this text

27. Implications for PK theory On a day-to-day basis, the supply of reserves is vertical (as represented here). A standard assertion of PK theory was that the supply of reserves is demand-determined, at the target overnight rate (the supply of reserves is said to be horizontal). With the target overnight rate set at the floor of the corridor, this is no longer true. The supply of reserves can exceed the demand for reserves. The central bank can maintain excess reserves (as long as banks don’t wish to pay back central bank advances, if there are any).

28. A point of controversy Some PK or NK authors believe that paying interest on excess reserves leads to more credit rationing and less economic activity (Palley 2010, Pollin 2010,Stiglitz 2010). Their argument is that the opportunity cost of holding reserves is reduced when interest is paid on reserves. This would induce banks to make less loans to firms or households. They propose to remove interest on reserves, and even to tax excess reserves, so as to induce banks to make more loans. Click View then Header and Footer to change this text

29. PK controversy These authors don’t seem to realize that, as pointed out by Fullwiler (2005), “banks cannot use reserve balances for anything other than settling payments or meeting reserve requirements; reserve balances do not fund additional lending”. Their argument is that other post-Keynesians don’t understand microeconomics and overly rely on macro laws (such as above). Each individual bank will try to get rid of excess reserves by making new loans and hence new deposits. Thus compulsory reserves will rise, and hence excess reserves will decrease. Their beliefs, ultimately, must be based on some implicit version of the money multiplier story, and on the belief that banks make a choice between loans to NFI and reserves. The Fullwiler-Lavoie answer is that taxing reserves will only lead to a reduction in the federal funds rate, as the alternative, holding reserves, now has a lower rate of return. The choice is between holding reserves or lending reserves to other banks. Click View then Header and Footer to change this text

30. Support for PK position by Bank of England deputy governor “The level of commercial banks’ reserves in aggregate is determined by the way we have funded the asset purchases, not by the commercial banks’ own decisions. The size of banks’ reserves cannot, as is frequently claimed, be a sign that they are “sitting on them”. No matter how rapidly or how slowly the economy is growing, or how fast or slow the money is circulating, the aggregate amount of reserves will be exactly the same. So it should be clear that the quantity of central bank reserves held by the commercial banks is useless as an indicator of the effectiveness of Quantitative Easing” (Bean 2009). Click View then Header and Footer to change this text

31. Quantitative easing ? It is sometimes said that the Fed was pursuing quantitative easing (QE1), with the Fed providing excess reserves in the hope that banks would then make more loans to the private sector. The analysis here shows that the Fed instead was pursuing credit easing, taking illiquid assets out of the market. The creation of excess reserves was simply the consequence of these credit-easing operations. Click View then Header and Footer to change this text

32. Credit easing, not quantitative easing “It is important to keep in mind that the excess reserves in our example were not created with the goal of lowering interest rates or increasing bank lending significantly relative to pre- crisis levels. Rather these reserves were created as a by- product of policies designed to mitigate the effects of a disruption in financial markets. In fact, the central bank paid interest on reserves to prevent the increase in reserves from driving market interest rates below the level it deemed appropriate given macroeconomic conditions”. (Keister and McAndrews 2009) Click View then Header and Footer to change this text

33. QE2 Quantitative easing operates through an attempt to lower interest rates on assets that go beyond short-term government securities. Instead of announcing the purchases of given amounts of securities and hoping this will have the forecasted effect on interest rates, shouldn’t the monetary authorities announce that they will purchase any given amount of government securities at a pre-announced price, thus setting the long-term interest rate on government securities? (Fullwiler and Wray 2010) What is the risk of such a policy: will the central bank be forced to purchase huge amounts of government securities? And if so, with a floor system, what is the problem? Or is it a solvency problem if bond prices drop in the future? Click View then Header and Footer to change this text

34. IMPLICATIONS FOR MAINSTREAM MONETARY THEORY

35. The problem of central banks today Standard theory says that reserves create money, and money creates price inflation. So if there are large excess reserves, this should lead to excess money supply or at least overly low interest rates, and hence inflation now or in the near future. But the decoupling principle shows there is no relation between reserves and interest rates, and hence no relation with prices. « There is a concern that markets may at some point, possibly based on the ‘wrong model’, become excessively concerned about the potential inflationary implications of these policies » (Bordo and Disyatat, BIS, p. 22). This means that the ‘market’ has the ‘wrong’ model. Gone are models of rational expectations within a single ‘correct’ model of the macroeconomy!

36. Central bank communications There is a big effort by central bankers in the US and elsewhere to convince financial experts that the correct monetary theory has changed: excess reserves do not lead to price inflation. As said by William Dudley (2009, p. 1), the President and CEO of the New York Fed, “it is not the case that our expanded balance sheet will inevitably prove inflationary. It is important that this critical issue be well understood” Keister and McAndrews, NYFRB (2008, 2009) claim that no inflationary pressures can arise when the target fed funds rate is the deposit rate. The reason given is that banks have no opportunity cost in holding these reserves, and hence will not try to use them by lending them. They reluctantly give some credibility to the multiplier story, but only when there is no floor system.

37. My critique to Keister (December 2009) “You seem to imply that the textbook multiplier still applies when reserves earn no interest. I think that this is a misleading statement.... There is never any money multiplier effect.”

38. Keister (NYFR) in personnal communication, January 2010 “I agree with you on the money multiplier, but I would state things in a slightly different way....I understand your comment to be that this mechanism is not the ‘money multiplier’ as commonly described. We decided to be more generous to the textbooks and say that this mechanism must be what they had in mind, even if they left out the part about interest rates to simplify things for the students. Importantly, I think we agree on the point that discussions of the money multiplier have done more harm than good in terms of helping people understand what is going on.”

39. Also an exploding balance sheet at the ECB The ECB has allowed fixed-rate tenders with full allotments at its regular main refinancing operations. Huge amounts have been borrowed by banks. Unlimited amounts of liquidity were also provided with the two 3- year longer-term refinancing operations (LTRO). Thus, as in the USA, there is a huge amount of reserves in the Eurosystem (but those that borrow the reserves may not be the same as those that hold them). How does this fit with the ECB pillar of maintaining price stability? Click View then Header and Footer to change this text

40. Vitor Constancio, vice-president of the ECB, 8 December 2011 “There is no acceptable theory linking in a necessary way the monetary base created by central banks to inflation. Nevertheless, it is argued by some that financial institutions would be free to instantly transform their loans from the central bank into credit to the non-financial sector. This fits into the old theoretical view about the credit multiplier according to which the sequence of money creation goes from the primary liquidity created by central banks to total money supply created by banks via their credit decisions. In reality the sequence works more in the opposite direction with banks taking first their credit decisions and then looking for the necessary funding and reserves of central bank money.” Click View then Header and Footer to change this text

41. Vitor Constancio, vice-president of the ECB again “In modern banking sectors, credit decisions precede the availability of reserves in the central bank. As Charles Goodhart pointedly argued, it would be more appropriate talking about a “Credit divisor” than about a “Credit multiplier”. Click View then Header and Footer to change this text

42. Conclusion The global financial crisis may have one good result: it may succeed in getting rid of both the erroneous money multiplier story and the misleading quantity theory of money.