The Transition from Monetary to Real Equilibrium Business Cycle
Theory
As we have seen in Chapter 5, the dominant new classical theory of the
business cycle during the period 1972–82 was the monetary surprise model
(MEBCT) initially developed by Lucas (1975, 1977). Since the early 1980s
the leading new classical explanation of aggregate instability has focused on
real rather than monetary shocks, and after the contribution of Long and
Plosser (1983) became known as real (equilibrium) business cycle theory
(REBCT). The best-known advocates or contributors to this approach are
Edward Prescott of the University of Minnesota, Finn Kydland of Carnegie-
Mellon University, Charles Plosser, John Long and Alan Stockman of the
University of Rochester, Robert King of Boston University, Sergio Rebelo of
Northwestern University and Robert Barro of Harvard University (see interviews
with Robert Barro and Charles Plosser in Snowdon et al. 1994).
In the early 1980s, in response to recognized weaknesses in the MEBCT
some new classical theorists sought to provide a rigorous equilibrium account
of the business cycle which is both free from the theoretical flaws of earlier
new classical models and would, at the same time, be empirically robust. The
result has been the development of REBCT, which replaces the impulse
mechanism of the earlier models (that is, unanticipated monetary shocks)
with supply-side shocks in the form of random changes in technology. The
propagation mechanisms of the earlier new classical models are, however,
retained and developed. Ironically it was Tobin (1980b) who was one of the
first to recognize this unlikely escape route for equilibrium theorists. In
criticizing the monetary misperception stories of Lucas, Tobin noted that the
‘real equilibrium of a full information model could move around, driven by
fluctuations in natural endowments, technologies and tastes’ and, if such
fluctuations were seriously persistent random processes, the observations
generated ‘may look like business cycles’. Meanwhile, around 1980, Kydland
and Prescott were working on just such a model and two years after Tobin
made his comments Econometrica published Kydland and Prescott’s paper
containing their prototype non-monetary equilibrium model.
Before moving on to consider real business cycle theory in more detail, it
is interesting to note the reaction to this second phase of equilibrium theorizing
of two of the leading pioneers of the new classical mark I approach. In
Robert Lucas’s view, Kydland and Prescott have taken macroeconomic modelling
‘into new territory’ (Lucas, 1987). However, Lucas’s initial reaction to
REBCT was to suggest that the exclusive focus of such models on real as
opposed to monetary considerations was ‘a mistake’ and argued the case for a
‘hybrid’ model as a fruitful way forward. Nevertheless Lucas warmly approved
of the methodology adopted by real business cycle theorists who have
followed his own earlier recommendation that an understanding of business
cycles is best achieved ‘by constructing a model in the most literal sense: a
fully articulated artificial economy which behaves through time so as to
imitate closely the time series behaviour of actual economies’ (Lucas, 1977).
Such artificial economic systems can serve as laboratories ‘in which policies
that would be prohibitively expensive to experiment with in actual economies
can be tested out at much lower cost’ (Lucas, 1980a). This is exactly what
real business cycle theorists established as their research agenda during the
1980s, and Lucas’s (1980a) paper is the reference point for the modern era of
equilibrium theorizing. As Williamson (1996) points out, ‘in Lucas one finds
a projection for future research methodology which is remarkably close to
the real business cycle program’.
By 1996 Lucas admitted that ‘monetary shocks just aren’t that important.
That’s the view that I’ve been driven to. There is no question that’s a retreat
in my views’ (see Gordon, 2000a, p. 555). Meanwhile Lucas has put forward
the view several times that he considers the business cycle to be a relatively
‘minor’ problem, at least at the level experienced since 1945 (Lucas, 1987,
2003). In his view it is far more important to understand the process of
economic growth if we are really interested in raising living standards, rather
than trying to devise ever more intricate stabilization policies in order to
remove the residual amount of business cycle risk (see Lucas, 1988, 1993,
2002, 2003, and Chapter 11).
By the late 1980s Robert Barro (1989a, 1989c) also declared that the
emphasis given by new classical economists during the 1970s to explaining
the non-neutrality of money was ‘misplaced’ because the ‘new classical
approach does not do very well in accounting for an important role for money
in business fluctuations’. By the mid-1980s, Barro regarded the contributions
of real business cycle theorists as ‘especially promising’ and representing
‘real progress’ (Barro, 1984). Furthermore, his own work had provided a
bridge between the mark I and mark II versions of new classical macroeconomics
(see Barro, 1981). In any case the lack of robust empirical success of
the earlier models does not invalidate the achievements of the new classical
theorists in the 1970s which in Barro’s view led to ‘superior methods of
theoretical and empirical analysis’ (Barro, 1984).
The three main new classical achievements identified by Barro (1989a) are
(i) the application of equilibrium modelling to macroeconomic analysis, (ii)
the adoption and promotion of the rational expectations hypothesis, and (iii)
the application of game theory to policy making and evaluation. The first two
contributions satisfy the objectives of building macro models on choicetheoretic
microfoundations, as well as providing an analytical framework
which can better withstand the Lucas (1976) critique. The third area relating
to dynamic games has drawn out the importance in policy analysis of the
roles of commitment, credibility and reputation as well as clarifying the
distinction between rules and discretion. The insights gained relating to the
time inconsistency of policy have now been applied to a wide variety of areas
other than the conduct of monetary policy. Although Barro remains enthusiastic
about real business cycle theory, he, like Lucas, began to redirect his
research work in the late 1980s primarily towards issues related to economic
growth (see Barro, 1991, 1997; Barro and Sala-i-Martin, 1995).
For detailed surveys of the evolution and development of REBCT, the
reader is referred to Walsh (1986), Rush (1987), Kim (1988), Plosser (1989),
Mullineux and Dickinson (1992), McCallum (1992), Danthine and Donaldson
(1993), Stadler (1994), Williamson (1996), Ryan and Mullineux (1997), Snowdon
and Vane (1997a), Hartley et al. (1998), Arnold (2002), and Kehoe and
Prescott (2002).
Theory
As we have seen in Chapter 5, the dominant new classical theory of the
business cycle during the period 1972–82 was the monetary surprise model
(MEBCT) initially developed by Lucas (1975, 1977). Since the early 1980s
the leading new classical explanation of aggregate instability has focused on
real rather than monetary shocks, and after the contribution of Long and
Plosser (1983) became known as real (equilibrium) business cycle theory
(REBCT). The best-known advocates or contributors to this approach are
Edward Prescott of the University of Minnesota, Finn Kydland of Carnegie-
Mellon University, Charles Plosser, John Long and Alan Stockman of the
University of Rochester, Robert King of Boston University, Sergio Rebelo of
Northwestern University and Robert Barro of Harvard University (see interviews
with Robert Barro and Charles Plosser in Snowdon et al. 1994).
In the early 1980s, in response to recognized weaknesses in the MEBCT
some new classical theorists sought to provide a rigorous equilibrium account
of the business cycle which is both free from the theoretical flaws of earlier
new classical models and would, at the same time, be empirically robust. The
result has been the development of REBCT, which replaces the impulse
mechanism of the earlier models (that is, unanticipated monetary shocks)
with supply-side shocks in the form of random changes in technology. The
propagation mechanisms of the earlier new classical models are, however,
retained and developed. Ironically it was Tobin (1980b) who was one of the
first to recognize this unlikely escape route for equilibrium theorists. In
criticizing the monetary misperception stories of Lucas, Tobin noted that the
‘real equilibrium of a full information model could move around, driven by
fluctuations in natural endowments, technologies and tastes’ and, if such
fluctuations were seriously persistent random processes, the observations
generated ‘may look like business cycles’. Meanwhile, around 1980, Kydland
and Prescott were working on just such a model and two years after Tobin
made his comments Econometrica published Kydland and Prescott’s paper
containing their prototype non-monetary equilibrium model.
Before moving on to consider real business cycle theory in more detail, it
is interesting to note the reaction to this second phase of equilibrium theorizing
of two of the leading pioneers of the new classical mark I approach. In
Robert Lucas’s view, Kydland and Prescott have taken macroeconomic modelling
‘into new territory’ (Lucas, 1987). However, Lucas’s initial reaction to
REBCT was to suggest that the exclusive focus of such models on real as
opposed to monetary considerations was ‘a mistake’ and argued the case for a
‘hybrid’ model as a fruitful way forward. Nevertheless Lucas warmly approved
of the methodology adopted by real business cycle theorists who have
followed his own earlier recommendation that an understanding of business
cycles is best achieved ‘by constructing a model in the most literal sense: a
fully articulated artificial economy which behaves through time so as to
imitate closely the time series behaviour of actual economies’ (Lucas, 1977).
Such artificial economic systems can serve as laboratories ‘in which policies
that would be prohibitively expensive to experiment with in actual economies
can be tested out at much lower cost’ (Lucas, 1980a). This is exactly what
real business cycle theorists established as their research agenda during the
1980s, and Lucas’s (1980a) paper is the reference point for the modern era of
equilibrium theorizing. As Williamson (1996) points out, ‘in Lucas one finds
a projection for future research methodology which is remarkably close to
the real business cycle program’.
By 1996 Lucas admitted that ‘monetary shocks just aren’t that important.
That’s the view that I’ve been driven to. There is no question that’s a retreat
in my views’ (see Gordon, 2000a, p. 555). Meanwhile Lucas has put forward
the view several times that he considers the business cycle to be a relatively
‘minor’ problem, at least at the level experienced since 1945 (Lucas, 1987,
2003). In his view it is far more important to understand the process of
economic growth if we are really interested in raising living standards, rather
than trying to devise ever more intricate stabilization policies in order to
remove the residual amount of business cycle risk (see Lucas, 1988, 1993,
2002, 2003, and Chapter 11).
By the late 1980s Robert Barro (1989a, 1989c) also declared that the
emphasis given by new classical economists during the 1970s to explaining
the non-neutrality of money was ‘misplaced’ because the ‘new classical
approach does not do very well in accounting for an important role for money
in business fluctuations’. By the mid-1980s, Barro regarded the contributions
of real business cycle theorists as ‘especially promising’ and representing
‘real progress’ (Barro, 1984). Furthermore, his own work had provided a
bridge between the mark I and mark II versions of new classical macroeconomics
(see Barro, 1981). In any case the lack of robust empirical success of
the earlier models does not invalidate the achievements of the new classical
theorists in the 1970s which in Barro’s view led to ‘superior methods of
theoretical and empirical analysis’ (Barro, 1984).
The three main new classical achievements identified by Barro (1989a) are
(i) the application of equilibrium modelling to macroeconomic analysis, (ii)
the adoption and promotion of the rational expectations hypothesis, and (iii)
the application of game theory to policy making and evaluation. The first two
contributions satisfy the objectives of building macro models on choicetheoretic
microfoundations, as well as providing an analytical framework
which can better withstand the Lucas (1976) critique. The third area relating
to dynamic games has drawn out the importance in policy analysis of the
roles of commitment, credibility and reputation as well as clarifying the
distinction between rules and discretion. The insights gained relating to the
time inconsistency of policy have now been applied to a wide variety of areas
other than the conduct of monetary policy. Although Barro remains enthusiastic
about real business cycle theory, he, like Lucas, began to redirect his
research work in the late 1980s primarily towards issues related to economic
growth (see Barro, 1991, 1997; Barro and Sala-i-Martin, 1995).
For detailed surveys of the evolution and development of REBCT, the
reader is referred to Walsh (1986), Rush (1987), Kim (1988), Plosser (1989),
Mullineux and Dickinson (1992), McCallum (1992), Danthine and Donaldson
(1993), Stadler (1994), Williamson (1996), Ryan and Mullineux (1997), Snowdon
and Vane (1997a), Hartley et al. (1998), Arnold (2002), and Kehoe and
Prescott (2002).
No comments:
Post a Comment