Why are there no intraday money markets

1
Why are there no intraday money markets?

• Antoine Martin Jamie McAndrews
• Federal Reserve Bank of New York
• The views expressed in this presentation are
  those of the authors and do not necessarily
  represent the views of the Federal Reserve Bank
  of New York or of the Federal Reserve System.

2
Why is the question interesting?

• Typically, CBs provide
• overnight reserves through a market
• intraday reserves through a standing facility
• But historically many CBs provided overnight
  reserves through a standing facility
• If CBs chose markets for overnight reserves
  provision, why not adopt it intraday?

3
What we do

• In the paper we present arguments for and against
  an intraday market for reserves
• We provide no definite answer
• This presentation focuses on the relationship
  between intraday and overnight reserves
• Should there be an overnight market for reserves?

4
Why focus on this relationship?

• The relationship is important to think about how
  reserves should be supplied
• We view potential benefits of markets as small
• If, contrary to current practice, cost of
  overnight reserves is small, there is no need for
  intraday reserves. Benefits may be larger

5
Outline

• Current practice Positive cost of overnight
  reserves and zero cost of intraday reserves
• Alternative possibility Zero cost of overnight
  reserves and its impact on overnight market

6
Definitions

• Standing facility refers to institutions that
  provide reserves directly from the central bank,
  such as an overdraft facility
• We use the term market when most of the
  liquidity is obtained through other market
  participants

7
Definitions

• Overnight reserves refer to the reserves that
  are held on a central bank account at the end of
  the day
• Intraday reserves refer to additional reserves
  that are supplied during the day

8
Supply of intraday and overnight reserves

• Intraday reserves are typically supplied by a
  standing facility
• U.S. Uncollateralized at a small cost
• ECB, BoE, others collateralized at zero cost
• Overnight reserves are typically supplied through
  a market

9
The role of overnight and intraday reserves

• Overnight and intraday reserves typically play
  different roles
• Intraday reserves serve to make payments between
  banks and payments to ancillary systems
• Overnight reserves serve to satisfy requirements
  or contractual agreements with the CB
• The amount of reserves is much higher intraday
  than overnight

10
Why the difference?

• Monetary policy implementation often relies on
  positive opportunity cost of reserves overnight
• Since overnight reserves are costly at the
  margin, banks minimize the amount they hold
• In contrast, intraday reserves have a very low
  cost so banks use them liberally

11
A symmetric channel
Overnight rate
lending rate
policy rate
deposit rate
Nonborrowed Reserves
0
target supply
12
Why are intraday reserves inexpensive?

• Research in payment economics suggests that
  intraday reserves should be supplied at very low
  cost
• Costly reserves could lead to inefficient delays
  (Angelini, Bech and Garrat, Kahn and Roberds)
• Central bank can provide insurance against
  liquidity shocks (Green, Kahn and Roberds,
  Martin, Zhou)
• An application of the Friedman rule (Millard,
  Speight, and Willison, Bhattacharya, Haslag, and
  Martin)
• We view current CB practice as consistent with
  these arguments

13
To summarize

• Intraday and overnight reserves have different
  roles and are supplied through different means
• Overnight reserves are supplied through a market
  and have a high opportunity cost
• Intraday reserves are supplied through a standing
  facility, have a low opportunity cost

14
Where are we now?

• Current practice Positive cost of overnight
  reserves and zero cost of intraday reserves
• Alternative possibility Zero cost of overnight
  reserves and its impact on overnight market

15
What cost for overnight reserves?

• A positive opportunity cost of overnight reserves
  is not necessary for monetary policy
  implementation
• When this cost is close to zero a large supply of
  intraday reserves is no longer necessary
• Overnight reserves can be used intraday

16
How to do it in practice?

• The CB can pay interest on reserves at the policy
  rate
• When opportunity cost of overnight reserves is
  zero, there is no incentive to economize
• The policy rate becomes independent of the supply
  of reserves

17
Divorcing money from monetary policy
Overnight rate
lending rate
Supply of reserves is not linked to policy rate
policy rate
or
Nonborrowed Reserves
0
target supply
target supply
18
Is zero cost of reserves a good idea?

• The argument for zero marginal cost is the same
  as the argument for the Friedman rule
• Reserves are not a scarce resource, they can be
  produced for free by the central bank
• Opportunity cost of reserves should reflect their
  opportunity cost to society

19
The case of New Zealand

• New Zealand started paying interest on reserves
  at the policy rate in October 2006
• Supply of (overnight) reserves increased from NZD
  20 million to NZD 8 billion
• Intraday reserves no longer available from the
  RBNZ

20
Japan during quantitative easing

• Quantitative easing was not a new monetary policy
  implementation framework
• During quantitative easing, opportunity cost of
  overnight reserves was close to zero
• Supply of reserves greatly increased

21
Impact on overnight market?

• Hard to tell in New Zealand
• In Japan, overnight market shrunk
• Uncollateralized call market fell from 29.3
  trillion (1995) to 7.6 trillion (2005)
• Euro-Yen swap market fell from 23.6 trillion
  (1995) to 2.9 trillion (2005)
• Has market suffered from quantitative easing?

22
Should there be an overnight money market?

• Supplying overnight reserves at the policy rate
  may or may not impair overnight market
• If it does, how big of a concern is it?
• With large supply of overnight reserves central
  bank intraday exposure is reduced ?Market may be
  less useful

23
Conclusion

• Supply of intraday reserves should be viewed in
  context of monetary policy implementation
• Cost of overnight reserves not necessary for
  implementation of monetary policy. Why impose
  such cost?
• More research is needed to understand the
  benefits of markets for reserves both intraday
  and overnight

24
Conclusion (cont.)

• Should we focus on the benefits of markets?
• ? Maybe intraday money markets are desirable
• Should we focus on the benefits of a low
  opportunity cost of reserves?
• ? Maybe overnight markets are not that important