Money, Inflation and the Business Cycle

1
Money, Inflation and the Business Cycle

• The Taylor Rule
• Reverse Causation
• Segmented Markets
• Misperceptions Model
• Commitment Monetary Policy

2

• Readings
• Williamson, Ch 11
• Williamson, Ch 17

3

• Recall that money in the CE model is either
• (i) Neutral Superneutral in the ad-hoc or CIA
  model w/ exogenous income (Classical
  Dichotomy).
• (ii) Neutral but not Superneutral in CIA model
  w/ production.
• CIA model w/ production ? money growth leads to
  lower employment and output (countercyclical).
• BC fact is money is procyclical and leading.

4
Figure 3.13 Money Supply (black line) and Real
GDP (colored line) for the Period 19592003
5

• Three CE explanations of procyclical money
• (i) Reverse Causation ? procyclical neutral.
• (ii) Segmented Markets ? procyclical
  non- neutral.
• (iii) Misperceptions (Lucas-Friedman) Theory ?
  procyclical non-neutral.

6
The Taylor Rule

• Movements in money supply are often endogenous
  and react to economic events.
• Traditionally, FED has two primary objectives
• (i) Price Stability
• (ii) Output Stability
• The Taylor Rule (J. Taylor Stanford) quantifies
  the observed decisions of the FOMC as depending
  upon deviations of
• (i) inflation from a target level (p)
• (ii) GDP from potential or target y

7
FOMC Statement May 2000

• The Federal Open Market Committee voted today to
  raise its target for the federal funds rate by
  50 basis points to 6-1/2 percent.
• Increases in demand have remained in excess of
  even the rapid pace of productivity-driven gains
  in potential supply The Committee is concerned
  that this disparity will continue, which could
  foster inflationary imbalances.
• The Committee believes the risks are weighted
  mainly toward conditions that may generate
  heightened inflation pressures in the foreseeable
  future.

8
FOMC Statement March 18, 2008

• The Federal Open Market Committee decided today
  to lower its target for the federal funds rate 75
  basis points to 2-1/4 percent.
• Recent information indicates that the outlook for
  economic activity has weakened further
  Financial markets remain under considerable
  stress Inflation has been elevated, and some
  indicators of inflation expectations have risen. 
  Uuncertainty about the inflation outlook has
  increased.
• Downside risks to growth remain.  The Committee
  will act in a timely manner as needed to promote
  sustainable economic growth and price stability.

9

• Taylor Rule expressed in terms of a nominal
  interest rate rule
• 1 increase in p gt p ? 1.5 increase in R
• 1 decrease in y lt y ? 0.5 decrease in R
• Taylor Rule expressed in terms of a money supply
  rule
• where a1 ,a2 gt 0.

10
RBC View Endogenous Money(Reverse Causation
View)

• If money is neutral in a CE model, can it explain
  why the nominal money supply is procyclical and
  leading the business cycle?
• Yes. Focus on Feds price stability objective.
• Combine the Taylor Rule with RBC model Let y
  be the CE value of output and set yt yt
• where a1 gt 0.

11

• Example Productivity Shocks and Fed Reaction
• Today (t) Fed forecasts negative productivity
  shock ( )
• ? future prices (Pt1)
• ? p gt p
• ? Mts
• Future (t1) Productivity shock ? Yt1
• and p p.
• Conclusion Mts today and Yt1 tomorrow.
• ? money is procyclical and leading but
  neutral

12
Figure 3.13 Money Supply (black line) and Real
GDP (colored line) for the Period 19592003
13
Non-Neutral Money

• There is evidence that monetary policy is
  non-neutral
• (i) Reverse causation cannot completely explain
  procyclical money supply.
• (ii) Statistical Evidence Money ? Output
• (iii) The Taylor Rule implies FED operates as if
  it can affect real output
• Can this explained in CE models w/o assuming
  fixed prices (e.g. IS-LM)?

14
Segmented Markets Theory

• Motivates by the Bank Lending Channel of
  monetary policy emphasized by B. Bernanke and A.
  Blinder.
• Originates with works of Robert Lucas (1990) and
  Timothy Fuerst (1992).
• Sometimes called Limited Participation models
• Inventory Models of Money (Baumol/Tobin)
• Costly Portfolio Adjustment

15

• The traditional view of effect of monetary
  policy
• Short-Run ? Liquidity Effect
• Long-Run ? Anticipated Inflation Effect
• or Fisher Effect R r pe
• Assumptions
• (i) Segments goods and financial market.
• (ii) Monetary policy works through financial
  markets.
• (iii) Firms must use borrowed cash from
  financial market to pay their wage bill (and/or
  investment).

16

• Money Supply
• where Xt mtMts transfer of money to
  financial market and mt money growth rate
• average money growth rate, e random money
  growth shock.

17
M1 Money Supply, 2000-2010Growth Rate
18

• Flow Chart
• Timing in Period t
• (i) Households begin with Mt money, Firms have
  Kt capital stock.
• (ii) Portfolio decision - Household deposits Dt
  into financial market based on rational
  expectation of Xt. Deposits pay interest rate
  Rt
• (iii) Fed injects money Xt into financial
  market.
• (iv) Firms borrow Qt from financial market to
  pay wages, produces output from labor and
  capital, buys investment goods. Loans are
  charged Rt.
• (v) Households consume with cash Mt - Dt.
• (vi) All loans and interest on deposits are
  paid. Ends period with Mt1, Kt1.

19

• Key Portfolio decision (ii) happens before
  (iii)!
• Modified Cash-in-Advance constraints.
• Households
• Firms

20

• Household Deposits/Portfolio FOC
• expected marginal cost of deposits expected
  marginal benefit
• or expected marginal value expected marginal
• of cash in goods market value of cash in
• financial market
• Let L (Actual Marginal Value of Cash in Goods
  Mkt)
• -
• (Actual Marginal Value of Cash in Financial
  Mkt)
• L lt 0 ? More than expected cash in goods market
• L gt 0 ? More than expected cash in financial
  market

21

• Labor Supply FOC
• Firm FOC
• Investment
• Labor Demand
• Market-Clearing Conditions.
• Goods
• Money
• Labor
• Financial

22

• Household portfolio decision, i.e. money demand,
  based upon expected nominal interest rate (
  )
• Net-of-Deposits Money Demand
• The expected nominal interest rate obeys Fisher
  Effect
• The actual nominal interest rate follows

23

• Unexpected positive money shock ? Excess cash in
  financial market ? R lt Re.
• Unexpected negative money shock ? Excess cash in
  goods market ? R gt Re .
• Recall Labor Demand will be inversely related to
  R

24
Expected Money Shock

• Purely temporary (r0) and expected money shocks
  are neutral.
• An expected and persistent (r gt 0) positive money
  supply shock (e) ? Similar to CIA model w/
  production
• Increases expected inflation rate
• Increases nominal interest rates (R)
• Decreases ND and NS
• Decreases N, c, y
• Money growth is countercyclical!

25
Simulation Expected Monetary Shock in Period 5

• Nominal Interest Rate (R)

Money Growth Rate (m)
26
Simulation Monetary Shock in Period 5
Labor (N)

• Output (Y)

27
US DATA Unexpected Negative Monetary Shock
(Bernanke Gertler, 1995)
28
Unexpected Money Shock

• An unexpected positive money supply shock (e)
• Current Period
• (1) Excess cash in financial market ? R lt Re
• Lower R ? Firms borrow more to increase labor
  demand.
• ? Increases w and N
• (3) Higher N ? Shifts Output Supply right
• ? Increases y and decreases r
• (secondary effect on NS)
• This is often called the LIQUIDITY EFFECT

29
Simulation Unexpected Monetary Shock in Period 5
Segmented Markets Model

• Nominal Interest Rate (R)

Money Growth Rate (m)
30
Simulation Unexpected Monetary Shock in Period
5, Segmented Markets Model
Labor (N)

• Output (Y)

31

• Future Period
• (1) Excess cash is pulled out of financial market
  by households ? R Re
• (2) Persistence of money shock ? increase
  expected future money growth and inflation ?
  increase in R.
• The ANTICIPATED INFLATION EFFECT
• N,y, c will be temporarily lower than their
  long-run steady state values but eventually
  converge.
• Money is procyclical and non-neutral in the
  short-run.
• Shortcoming No persistence effect of monetary
  policy on nominal interest rates and real GDP.

32
Table 11.2 Data Versus Predictions of the
Segmented Markets Model with Monetary Shocks
33

• Non-neutrality of money is caused by imperfect
  (unexpected) information about monetary policy.
• Remarks about optimal policy
• (i) Generally inefficient to create unexpected
  monetary shocks to boost output (lowers
  utility).
• (ii) If the Fed can react faster to productivity
  shocks then a procyclical money supply may be
  welfare improving
• ? Optimal ND greater than expected but
  limited cash in financial market. FED adds
  liquidity to support ND and increase Y to its
  efficient level

34
Misperceptions Theory(Money Surprise Model
Williamson Ch 17)

• Origins M. Friedman (1968) The Role of
  Monetary Policy
• The Worker Surprise Model
• Formalized R. Lucas (1973) Some
  International Evidence on Output-Inflation
  Trade-Offs
• The Producer Island Model

35
Worker Surprise Model

• What should matter to workers are real wages w
  W/P.
• Labor Market with Nominal Wages.
• Workers confuse W and w.
• Recall W wP. It can increase because
• (i) positive productivity shocks ? increases w
• (ii) an increase in Ms ? increases P.
• An unexpected increase in P ?
• increases N and Y.

36
Lucas Island Model

• Supply of individual firms depends positively on
  local price relative to aggregate price level.
• Time-Line of Information
• (i) Period t-1 Form Et-1Pt (expected value of
  Pt given information up to time t-1).
• (ii) Period t Observe local price NOT Pt.
• Let 0 lt q lt 1 represent the correlation between
  local price and aggregate price Pt
• q 0 ? No Correlation
• q 1 ? 100 Positive Correlation
• 0 lt q lt 1 ? Some Correlation

37

• The AS Curve
• where y CE value for output (w/ perfect info)
• If q 1 ? y y
• If q lt 1 then
• Et-1Pt Pt ? y y
• Et-1Pt lt Pt ? y gt y
• Et-1Pt gt Pt ? y lt y

38

• The Lucas Critique Expectations about
  (monetary) policy affects the impact of the
  policy.
• Value of q is based on rational expectations
• High inflation countries ? q 1 ? AS steep
• Low inflation countries ? q 0 ? AS flat

39
Application Rational Expectations and Monetary
Policy

• T. Sargent and N. Wallace (U. of Minnesota)
• Consider the following reduced form macro model
  (let a 1-q).
• (AS Curve)
• (AD Curve)
• (Monetary Policy)
• Variables are in logs. y is the CE value of
  output w/ complete info.
• Tools of Fed
• e ? surprise shock to money supply.
• r ? anticipated (systematic) policy rule.

40

• Results
• Anticipated or systematic changes in monetary
  policy (r) have no effect on real output (y).
• Unexpected shocks matter for real output

41

• In this situation (exogenous) changes in money
  supply is procyclical ( non-neutral).
• Note that unlike CIA model w/ production, money
  is also superneutral (no anticipated inflation
  effects)
• Policy Evaluation a matters for the effect of
  monetary policy
• q 1 ? a 0 ? dy/dm 0

42

• Result became know as the Policy Ineffectiveness
  Proposition Anticipated changes in monetary
  policy are neutral and unexpected changes are
  non-neutral.
• Information regarding monetary policy and the Fed
  matters for the effect of money on real output.
• If goal is to minimize output fluctuations (yt
  y), ala the Taylor Rule, it is optimal for FED
  to set et 0. (monetarist constant growth rate
  rule)

43
Phillips Curve

• The Phillips Curve is a statistical
• (i) Positive relationship between inflation and
  real GDP.
• (ii) Negative relationship between inflation
  and unemployment rate.

44
Figure 17.2 The Phillips Curve, 19471959
45
Figure 17.3 The Phillips Curve, 19601969
46
Figure 12.1 The Phillips curve and the U.S.
economy during the 1960s
47
PC in Keynesian Model

• If nominal wages (W) are slow to adjust there
  will be a
• (i) Positive relation between p and Y
• (ii) Negative relation between p and
  unemployment rate u.
• where u is the natural unemployment rate
• Trade-off can be permanent
• Fed can exploit this trade-off and control UR by
  choosing p.

48
Some Historical Facts about US Inflation

• Stable and low inflation in 1950s and 60s
• High inflation and unemployment in 1970s
• Low inflation and unemployment in 1990s
• PC broke down since 1970s.
• Failure of Keynesian models to account for PC
  break down (see Mankiw article).

49
Figure 17.3 The Phillips Curve, 19601969
50
Figure 17.4 The Phillips Curve, 19701979
51
Figure 17.5 The Phillips Curve, 19801989
52
Figure 17.6 The Phillips Curve, 19902003
53
Figure 12.2 Inflation and unemployment in the
United States, 19702002
54
CE Model View of PC

• Both segmented markets misperceptions models
  imply expectations about monetary policy, prices,
  and inflation affect PC trade-off.
• Expectations augmented PC
• where u natural rate of unemployment.
• Natural Unemployment is due to worker-job
  mismatches caused by
• (i) Frictional reasons (search theory ? may be
  optimal)
• (ii) Structural reasons

55

• Generic macro policy cannot should not be used
  to affect natural unemployment. Only targeted
  policies
• Worker retraining, education, employment
  agencies, unemployment insurance reform, ect.
• There are costs of inflation in CEM high
  nominal interest rates, lower output. (recall the
  Friedman Rule)
• Robert Barro studies money supply and inflation
  across 83 countries (1950-90). Finds
• Median Inflation 8 (23 gt 10)
• Median Money Growth 11 (57 gt 10)
• WHY??

56
Central Bank Commitment

• If inflation is bad why do we see it in almost
  every country around the world?
• Two possibilities
• (i) Inflation tax.
• (ii) Central bank commitment problem
• Kydland and Prescott (2004 Nobel Winners)
  pioneered work on policy rules and discretion.

57

• Should Fed follow rules or discretion?
• The time consistency problem arises when the best
  plan is made and then there are incentives to
  abandon it at a future date.
• Hostages, Speeding, Exams.
• PC and central bank commitment.
• Fed preferences
• Credibility problem and pe

58
Commitment A Simple Example

• Assume that FED can directly set inflation rate
  p.
• FEDs Loss Function
• L p2 u2
• Phillips Curve
• Initially pe p 0 ? u u and L (u)2

59

• FED takes pe 0 as given and sets p to minimize
  loss function L
• p 2u/5 gt 0
• Public has rational expectations and understands
  the incentives of FED
• p pe 2u/5 gt 0.

60

• Rational expectations equilibrium w/ no
  commitment
• p pH 2u/5 gt 0
• u u
• L (pH)2 (u)2 gt (u)2.
• Rational expectations equilibrium with
  commitment
• p pH 0
• u u
• L (u)2.

61

• Solutions to commitment problem
• (i) Tough central banker
• (ii) Tougher consequences for not meeting
  inflation targets.
• (iii) Central bank independence.

62
Senate Testimony of Ben Bernanke, Fed Chairman
Nominnee (11/15/05)

• the Federal Reserve's success in reducing and
  stabilizing inflation and inflation expectations
  is a major reason for this improved economic
  performance.
• Monetary policy is most effective when it is as
  coherent, consistent, and predictable as
  possible.
• One possible step toward greater transparency
  would be for the FOMC to state explicitly the
  numerical inflation rate or range consistent
  with the goal of long-term price stability

63
Figure 14.10 Central bank independence and
inflation
64

• Another reason for expansionary monetary
  policies CE is wrong, markets fail, involuntary
  (cyclical) unemployment is a big problem, and
  there is a role for Keynesian stabilization
  policies.
• (i) Sticky price (non-market-clearing)
• (e.g., IS-LM model)
• (ii) Coordination failure