Transcript Empirical Evidence of the Effectiveness of Monetary Policy

Chapter 28
Empirical
Evidence of
the
Effectiveness
of Monetary
Policy
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
Learning Objectives
• Understand the historical volatility in velocity
and money demand and the implications of that
volatility for Keynesianism and monetarism
• Explain the fiscal and monetary policy lags and
resulting difficulties in enacting these policies
• Analyze the empirical evidence relating
monetary and fiscal policies to GDP, investment,
consumption, and interest rates
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-2
Introduction
• Previous chapters dealt with theoretical concepts
of monetary theory on economic activity
• This chapter explores empirical evidence
– How are interest rates affected
– What categories of spending are most influenced
– Does monetary policy alter real economic
outcomes—output and unemployment—as
compared to changes in prices
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-3
Living with Velocity
• Much of Monetarist-Keynesian debate hinges on
the behavior of velocity
– Monetarists—relatively stable and changes are
highly predictable
– Keynesians—neither contention is true, velocity can
vary unpredictably
• While velocity is not constant, changes do not
appear to be obviously random or perverse
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-4
Living with Velocity (Cont.)
• If the Federal Reserve could discover the
underlying determinants of these changes, it
might still be able to coexist with such a moving
target
• Figure 28.1 plots the historical movements of
measures of velocity for two definitions of the
money supply—M1 and M2—over time
– Velocity of M2 has been relatively stable
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-5
FIGURE 28.1 Historical movements
in M1 and M2 velocities.
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-6
Living with Velocity (Cont.)
• Figure 28.1 (Cont.)
– Velocity of M1 appears highly unpredictable
• Reached a peak of about 4 at the beginning of the 1920s
• Fell almost continuously during the Great Depression and
World War
• Experienced a dramatic increase since the low in 1949
reaching almost 9 during the early 2000s
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-7
Living with Velocity (Cont.)
• Figure 28.1 (Cont.)
– Main reasons for the post-World War II rise in
velocity of M
• The relatively narrow definition of M1
• Increasing attractiveness of other categories of financial
assets as prudent/desirable places to invest excess cash
• Higher interest rates experienced during the 1970s and
early 1980s increased opportunity cost of holding cash
• This lead to an increase in velocity by inducing business
firms and households to economize on money
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-8
Living with Velocity (Cont.)
• Figure 28.1 (Cont.)
– However, since the mid-1980s, M1 velocity has
reversed the post-World War II uptrend forcing the
Federal Reserve to abandon setting M1 targets and
focus on the more stable behavior of M2
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-9
The Demand for Money
• The historical movements in the velocity of M1
are a result of many factors
– Movements in interest rates
– Technology
– Innovations in financial markets
• In order to be able to focus more precisely on
the relative importance of each, must be able to
disentangle their separate influences
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-10
The Demand for Money (Cont.)
• Formal application of statistical techniques to historical
data of economic variables will permit identification of
individual relationships
• Statistical studies show that interest rates and the level
of GDP influence money demand
– Higher rates reduce demand for cash balances
– Contradicts extreme forms of monetarism, which assume zero
interest-sensitivity of money demand
– However, none of the empirical studies isolate the Keynesian
liquidity trap
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-11
The Demand for Money (Cont.)
• More sophisticated approach focuses on the
stability of the money demand
– Recognizes that specific numbers are less important
than ability to forecasting future money demand
– If this relationship can be established, Federal
Reserve can gauge proper amount of money to
add/subtract to hit a target of economic activity
– If this relationship is unpredictable, it provides no
useful guidance to the Federal Reserve
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-12
The Demand for Money (Cont.)
• Empirical evidence
– Demand for money was quite stable until the mid1970s
– Deteriorated after the middle of 1974 with people
holding smaller money balances than the historical
relationship suggested
– The observations in the 1980s were the opposite—
holding of too much money reflecting the decline in
M1 velocity
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-13
Econometrics and Time Lags
• The simple velocity and money demand approaches
leave much to be desired as a guideline for Federal
Reserve policy-making.
• Ignores time lags between changes in monetary policy
and impact on economy
• Ignores more sophisticated computer driven
statistical methodology that can simulate the impact of
monetary policy via formal econometric modeling
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-14
Econometrics and Time Lags (Cont.)
• Lags in monetary policy
– Prompt recognition of what the economy is doing
can be difficult
– Available data are often inadequate and frequently
mixed
– The economy rarely proceeds on a perfectly smooth
course
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-15
Econometrics and Time Lags (Cont.)
• Lags in monetary policy (Cont.)
– Recognition lag
• Time associated with getting an accurate understanding of what is
happening or is likely to happen in the future
• Data suggests the Federal Reserve generally starts to tighten only a
few months after a business cycle has reached its trough, while the
move toward easing is somewhat more delayed
• This leads to the conclusion that the Federal Reserve is typically
more concerned with preventing inflation than with avoiding a
recession
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-16
Econometrics and Time Lags (Cont.)
• Lags in monetary policy (Cont.)
– Impact lag—time span from when the central bank starts to
use one of the monetary tools until an effect is evident on the
ultimate objective—aggregate spending in the economy
•
•
•
•
It may take weeks before interest rates change
Changes in credit availability also take time
Delay before actual spending decisions are affected
Once monetary policy starts to influence spending, it will most likely
continue to have an impact well into the future
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-17
Econometrics and Time Lags (Cont.)
• Econometric Models
– A mathematical-statistical representation that
describes how the economy behaves
– How do different economic actors (consumers and
businesses) respond to economic stimuli
– Once the relationships are formalized in
mathematical representations, data on past
experience in the real world are used to estimate the
precise behavioral patterns of each sector
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-18
Econometrics and Time Lags (Cont.)
• Econometric Models (Cont.)
– The computer model simulates the economy in action and
makes predictions based on the formal relationships
embedded in the model
– However, depending on the theoretical propositions used
to construct the model, the data can produce different
results
– Since Monetarists and Keynesians use different theory,
predictions of models based on their respective theories
would be significantly different
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-19
The Impact of Monetary Policy on GDP
• The Federal Reserve, working with several
universities, has developed econometric models
of behavior of economic aggregates in the U.S.
• These econometric models are an evolutionary
phenomena, constantly being revised and
updated to reflect new and different perspectives
about economic reality
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-20
The Impact of Monetary Policy on GDP
(Cont.)
• Outcomes of these models will vary with
specific conditions of current economic activity
• The Federal Reserve model articulates rather
carefully the impact of monetary policy on
various categories of spending
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-21
The Impact of Monetary Policy on GDP
(Cont.)
• Fed’s model shows that a 1% increase in the money
supply raises real GDP by about ½% after one year,
rising over the next two years to reflect the full 1%
increase
• Model demonstrates that monetary policy has a
significant impact even within a year
• However, because of various lags, the continued
impact may cause difficulties for policymakers in
subsequent years
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-22
Fiscal Policy and Crowding Out
• The models test the Monetarists-Keynesian
distinction of the effect of fiscal policy and
crowding out
– Monetarists—increased government spending will
merely displace private spending, leaving little net
impact of fiscal policy on GDP
– Keynesians—“crowding-out” effects are
incomplete, implying that fiscal policy generates
much of the traditional multiplier effect on GDP
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-23
Fiscal Policy and Crowding Out
(Cont.)
• The Federal Reserve model
– Holding the money supply constant
• This results in an increase in the interest rates caused by
additional government borrowing
• An increase in government spending by 1 percent of GDP
increases the level of real GDP by about 1 percent for at
least two years
• It takes more than three years for crowding out to reduce
the impact on real GDP
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-24
Fiscal Policy and Crowding Out
(Cont.)
• The Federal Reserve model (Cont.)
– Allowing the money supply to increase
• In this case, the Federal Reserve finances the additional
borrowing/spending by printing money and interest rates
do not increase
• In this case the multiplier effects on GDP are substantial
• The effects on GDP continue even after 4 years
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-25
Fiscal Policy and Crowding Out
(Cont.)
• These models suggest that fiscal policy has a multiplier
of about one without any help from monetary policy,
but an accommodating monetary authority can make
fiscal policy even more effective
• Empirical evidence on fiscal policy confirms the
crowding-out effect, but only if the contractionary
effects on private spending are given substantial time to
work themselves out
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-26
Interest Rates
• Monetarists and Keynesians believe the initial liquidity
impact of an expansionary monetary policy reduces the
level of interest rates
• Eventually both recognize inflationary expectations
generated by excessive expansion of money will raise
interest rates
• Difference between the two rests on how long it takes
for inflationary expectations to counter the initial
liquidity effect
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-27
Interest Rates (Cont.)
• Figure 28.2
– Illustrates that levels of short-term interest rates have been
closely related to actual movements in the rate of inflation
– However, the chart also shows substantial intervals of
independent movements in the level of interest rates
– Most econometric models report that interest rates decline
and remain below their original levels for six months
following an expansion of money
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-28
FIGURE 28.2 Interest rates move
with the rate of inflation.
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-29
Interest Rates (Cont.)
• Figure 28.2 (Cont.)
– Similarly, interest rates are above their original
levels for a similar period after an contractionary
monetary policy
– After a year, however, the initial liquidity effect is
reversed, and interest rates move in the opposite
direction
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-30
Interest Rates (Cont.)
• It should be emphasized that the response of
inflationary expectations depends crucially on
the initial state of the economy
– Responding very quickly when the economy is very
close to full employment
– Also, a quick response when there is high degree of
concern about possible inflation
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-31
Business Investment
• Theory would suggest interest rates and all types
of investment spending to move in opposite
directions
• However, historically the two almost always
move in the same direction
– This conflicting result stems from the fact that
business investment is dependent upon a number of
factors besides interest rates
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-32
Business Investment (Cont.)
– The model, therefore, violates the assumption of
holding all other influences constant and just
changing the rate of interest
– An increase in interest rates may inhibit investment,
but an increase in investment caused by other factors
may offset the interest rate effect
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-33
Business Investment (Cont.)
• Fed’s econometric model can sort out effects of
individual variables and does demonstrate the negative
relationship between interest rates and business
spending
• Most of the effects occur during the last year of a 3
year time period
• The time delay relates to the fact that current business
decisions are not executed usually until several years
later
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-34
Residential Construction
• The impact of monetary policy is felt more promptly
and powerfully on residential construction
expenditures—higher interest, lower residential
construction
• Part of this effect occurs through credit rationing
activities by financial institutions engaged in mortgage
lending
• Construction expenditures by state and local
governments also appear sensitive to the actions of the
monetary authorities
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-35
Consumer Spending
• The Federal Reserve model includes the wealth
effect of modern Keynesians
• Lower interest rates raise the value of financial
securities which increase the level of consumer
spending
• According to the Fed’s model, the importance of
the wealth effect in the overall impact of
monetary policy is quite substantial
Copyright © 2009 Pearson Addison-Wesley. All rights reserved.
28-36