Авторы: 147 А Б В Г Д Е З И Й К Л М Н О П Р С Т У Ф Х Ц Ч Ш Щ Э Ю Я

Книги:  180 А Б В Г Д Е З И Й К Л М Н О П Р С Т У Ф Х Ц Ч Ш Щ Э Ю Я


5.4 Equilibrium Business Cycle Theory

Before Keynes’s (1936) General Theory many economists were actively engaged

in business cycle research (see Haberler, 1963). However, one of the

important consequences of the Keynesian revolution was the redirection of

macroeconomic research towards questions relating to the level of output at a

point in time, rather than the dynamic evolution of the economy over time.

Nevertheless, within mainstream macroeconomics, before the 1970s, the main

approach to the analysis of business cycles after 1945 was provided by

Keynesians and monetarists (see Mullineux, 1984). During the 1970s a new

approach to the study of aggregate fluctuations was initiated by Lucas, who

advocated an equilibrium approach to business cycle modelling (Kim, 1988).

Lucas’s equilibrium theory was a significant departure from Keynesian business

cycle analysis where fluctuations of GDP were viewed as disequilibrium

phenomena. Keynesian macroeconomic models are typically characterized

by various rigidities and frictions that inhibit wage and price flexibility.

Consequently, in the short run, markets fail to clear and GDP can depart

significantly from its potential level for extended periods of time. Milton

Friedman also criticized Keynesian models for their downplaying of the

importance of monetary disturbances as a major source of aggregate instability.

The Friedman and Schwartz (1963) study proved to be highly influential

to a whole generation of economists. In particular Friedman and Schwartz

argued that the Great Depression was ‘a tragic testimonial to the importance

of monetary factors’. While Lucas was very much influenced by Friedman’s

monetarist ideas, he preferred to utilize a Walrasian research methodology

rather than build on Friedman’s Marshallian approach when analysing business

cycles (see Hoover, 1984).

The foundations of Lucas’s approach to business cycle modelling can be

found in his seminal Journal of Economic Theory paper (Lucas, 1972a),

where his objective is clearly stated in the opening paragraphs:

This paper provides a simple example of an economy in which equilibrium prices

and quantities exhibit what may be the central feature of the modern business

cycle: a systematic relation between the rate of change of nominal prices (inflation)

and the level of real output. The relationship, essentially a variant of the wellknown

Phillips curve, is derived within a framework from which all forms of

‘money illusion’ are rigorously excluded: all prices are market clearing, all agents

behave optimally in light of their objectives and expectations, and expectations

are formed optimally … In the framework presented, price movement results from

a relative demand shift or a nominal (monetary) one. This hedging behaviour

results in the nonneutrality of money, or broadly speaking a Phillips curve, similar

in nature to that we observe in reality. At the same time, classical results on the

long-run neutrality of money, or independence of real and nominal magnitudes,

continue to hold.

Lucas demonstrated that within this Walrasian framework, monetary changes

have real consequences, but ‘only because agents cannot discriminate perfectly

between monetary and real demand shifts’ so ‘there is no usable

trade-off between inflation and real output’. In Lucas’s 1972 model ‘the

Phillips curve emerges not as an unexplained empirical fact, but as a central

feature of the solution to a general equilibrium system’.

Building on this insight, Lucas proceeded to develop an equilibrium approach

to the analysis of aggregate fluctuations. Lucas (1975) defines business

cycles as the serially correlated movements about trend of real output that

‘are not explainable by movements in the availability of factors of production’.

Associated with fluctuations in GDP are co-movements among different

aggregative time series, such as prices, consumption, business profits, investment,

monetary aggregates, productivity and interest rates (see Abel and

Bernanke, 2001). Such are the regularities that Lucas (1977) declares that

‘with respect to the qualitative behaviour of co-movements among series,

business cycles are all alike’ (the Great Depression being an exception). To

Lucas the ‘recurrent character of business cycles is of central importance’. As

Lucas (1977) explains:

Insofar as business cycles can be viewed as repeated instances of essentially

similar events, it will be reasonable to treat agents as reacting to cyclical changes

as ‘risk’, or to assume their expectations are rational, that they have fairly stable

arrangements for collecting and processing information, and that they utilise this

information in forecasting the future in a stable way, free of systematic and easily

correctable biases.

Building on his path-breaking 1972 and 1973 papers, Lucas (1975, 1977)

provides a ‘new classical’ monetarist explanation of the business cycle as an

equilibrium phenomenon. As Kevin Hoover (1988) observes, ‘to explain the

related movements of macroeconomic aggregates and prices without recourse

to the notion of disequilibrium is the desideratum of new classical research

on the theory of business cycles’. As Lucas (1975) puts it, ‘the central

problem in macroeconomics’ is to find a theoretical framework where monetary

disturbances can cause real output fluctuations which at the same time

does not imply ‘the existence of persistent, recurrent, unexploited profit

opportunities’ such as occur in Keynesian models characterised by price

rigidities and non-rational expectations.

Hayek (1933) had set forth a research agenda where ‘the crucial problem

of Trade Cycle Theory’ was to produce a solution that would allow ‘incorporation

of cyclical phenomena into the system of economic equilibrium theory,

with which they are in apparent contradiction’. By equilibrium theory Hayek

meant that which had been ‘most perfectly expressed by the Lausanne School

of theoretical economics’. While Keynesian economists regarded the quest

for an equilibrium theory of the business cycle as unattainable, it is one of

Lucas’s most notable achievements to demonstrate that it is possible to develop

an equilibrium account of aggregate instability. Although initially Lucas

claimed some affinity, via the notion of equilibrium theorizing, with the work

of Hayek on business cycles, it is now clear that new classical and Austrian

theories of the business cycle are very different. While the Austrian theory

views business cycles as an equilibrating process, in new classical models the

business cycle is viewed as a ‘continuum of equilibrium’ (Kim, 1988; see

also Chapter 9; Lucas, 1977; Hoover, 1984, 1988; Zijp, 1993).

Lucas’s monetary equilibrium business cycle theory (MEBCT) incorporates

Muth’s (1961) rational expectations hypothesis, Friedman’s (1968a)

natural rate hypothesis, and Walrasian general equilibrium methodology.

With continuous market clearing due to complete wage and price flexibility

the fluctuations in the MEBCT are described as competitive equilibria. But

how can monetary disturbances create fluctuations in such a world? In the

stylized classical model where agents have perfect information, changes in

the money supply should be strictly neutral, that is, have no impact on real

variables such as real GDP and employment. However, the leading and procyclical

behaviour of money observed empirically by researchers such as

Friedman and Schwartz (1963), and more recently by Romer and Romer

(1989), suggests that money is non-neutral (ignoring the possibility of

reverse causation). The intellectual challenge facing Lucas was to account

for the non-neutrality of money in a world inhabited by rational profitmaximizing

agents and where all markets continuously clear. His main

innovation was to extend the classical model so as to allow agents to have

‘imperfect information’. As a result Lucas’s MEBCT has come to be popularly

known as the ‘misperceptions theory’, although the idea of instability

being the result of monetary-induced misperceptions is also a major feature

of Friedman’s (1968a) analysis of the Phillips curve. In Lucas’s (1975)

pioneering attempt to build a MEBCT his model is characterized by: prices

and quantities determined in competitive equilibrium; agents with rational

expectations; and imperfect information, ‘not only in the sense that the

future is unknown, but also in the sense that no agent is perfectly informed

as to the current state of the economy’.

The hypothesis that aggregate supply depends upon relative prices is central

to the new classical explanation of fluctuations in output and employment.

In new classical analysis, unanticipated aggregate demand shocks (resulting

mainly from unanticipated changes in the money supply) which affect the

whole economy cause errors in (rationally formed) price expectations and

result in output and employment deviating from their long-run (full information)

equilibrium (natural) levels. These errors are made by both workers and

firms who have incomplete/imperfect information, so that they mistake genThe

eral price changes for relative price changes and react by changing the supply

of labour and output, respectively.

In neoclassical microeconomic theory the supply curve of an individual

producer in a competitive market slopes upward, indicating that the supplier

will produce more in response to a rise in price. However, this profit-maximizing

response is a reaction of producers to a rise in their relative price.

Therefore, individual suppliers need to know what is happening to the general

price level in order to make a rational calculation of whether it is

profitable to expand production in response to an increase in the nominal

price of the good they supply. If all prices are rising due to inflation, suppliers

should not increase production in response to a rise in price of their good

because it does not represent a relative (real) price increase. And yet the data

reveal that aggregate output increases as the general price level increases; that

is, the short-run aggregate supply curve slopes upwards in P–Y space. This

must mean that the aggregate response of thousands of individual suppliers to

a rise in the general price level is positive and yet profit-maximizing individuals

should not be reacting in this way. How can that be? Rational agents

should only respond to real variables and their behaviour should be invariant

to nominal variables. The answer provided by Lucas relates to agents (workers,

households, firms) having imperfect information about their relative prices

(Lucas, 1972a). If agents have been used to a world of price stability, they

will tend to interpret an increase in the supply price of the good (or service)

they produce as a relative price increase and produce more in response.

Therefore an unexpected or unanticipated increase in the price level will

surprise agents and they will misinterpret the information they observe with

respect to the rise in price of their good and produce more. Agents have what

Lucas (1977) refers to as a ‘signal extraction problem’, and if all agents make

the same error we will observe an aggregate increase in output correlated

with an increase in the general price level. Since Lucas’s model is ‘monetarist’,

the increase in the general price level is caused by a prior increase in the

money supply and we therefore observe a positive money-to-output correlation,

that is, the non-neutrality of money.

Consider, for example, an economy which is initially in a position where

output and employment are at their natural levels. Suppose an unanticipated

monetary disturbance occurs which leads to an increase in the general price

level, and hence individual prices, in all markets (‘islands’) throughout the

economy. As noted above, firms are assumed to have information only on

prices in the limited number of markets in which they trade. If individual

firms interpret the increase in the price of their goods as an increase in the

relative price of their output, they will react by increasing their output.

Workers are also assumed to have incomplete information. If workers mistakenly

perceive an increase in money wages (relative to their expected value) as

an increase in real wages, they will respond by increasing the supply of

labour (Lucas and Rapping, 1969). In contrast to Friedman’s model (see

Chapter 4), where workers are fooled, Lucas’s model does not rely on any

asymmetry of information between workers and firms. Both firms and workers

are inclined to make expectational errors and respond positively to

misperceived global price increases by increasing the supply of output and

labour, respectively. As a result aggregate output and employment will rise

temporarily above their natural levels. Once agents realize that there has been

no change in relative prices, output and employment return to their long-run

(full information) equilibrium (natural) levels.

The Lucas model emphasizes monetary shocks as the main cause of aggregate

instability and the whole story is based on a confusion on the part of

agents between relative and general price movements (Dore, 1993; Arnold,

2002). In the MEBCT, the supply of output at any given time (Yt) has both a

permanent (secular) component (YNt) and a cyclical component (Yct) as shown

in equation (5.9):

Yt YNt Yct (5.9)

The permanent component of GDP reflects the underlying growth of the

economy and follows the trend line given by (5.10):

YNt t (5.10)

The cyclical component is dependent on the price surprise together with the

previous period’s deviation of output from its natural rate, as shown in equation


Yct Pt E Pt t Yt YNt [ ( |1)] ( 1 1 ) (5.11)

The lagged output term in (5.11) is to recognize that deviations in output

from the trend will be more than transitory due to the influence of a variety of

propagation mechanisms, and the coefficient > 0 determines the speed with

which output returns to its natural rate after a shock. Because the combination

of the rational expectations hypothesis and the surprise supply function

implies that output and employment will fluctuate randomly around their

natural levels, further assumptions are required to explain why during the

business cycle output and employment remain persistently above (upswing)

or below (downswing) their trend values for a succession of time periods.

The observed serially correlated movements in output and employment (that

is, where output and employment levels in any one time period are correlated

with their preceding values) have been explained in the literature in a number

of ways. These explanations (propagation mechanisms) include reference to

lagged output, investment accelerator effects, information lags and the durability

of capital goods, the existence of contracts inhibiting immediate

adjustment and adjustment costs (see Zijp, 1993). For example, in the field of

employment firms face costs both in hiring and in firing labour: costs associated

with interviewing and training new employees, making redundancy

payments and so on. In consequence their optimal response may be to adjust

their employment and output levels gradually over a period of time following

some unanticipated shock.

By combining equations (5.9), (5.10) and (5.11) we get the Lucas aggregate

supply relationship given by equation (5.12):

Yt t Pt E Pt t Yt YNt t [ ( |1)] ( 1 1 ) (5.12)

where t is a random error process.

Although the actions of agents in Lucas’s model turn out ex post to be nonoptimal,

they are in a rational expectations equilibrium doing the best they

can given the (imperfect or incomplete) information they have acquired. As

Lucas (1973) demonstrated, this implies that monetary disturbances (random

shocks) are likely to have a much bigger impact on real variables in countries

where price stability has been the norm. In countries where agents are used to

inflation, monetary disturbances are unlikely to impact in any significant way

on real variables. Let represent the fraction of total individual price variance

due to relative price variation. Thus the larger is , the more any

observed variability in prices is attributed by economic agents to a real shock

(that is, a change in relative price) and the less it is attributed to purely

inflationary (nominal) movements of the general price level. We can therefore

modify equation (5.12) and present the Lucas aggregate supply curve in a

form similar to how it appeared in his 1973 paper ‘Some International Evidence

on Output–Inflation Trade-Offs’:

Yt t Pt E Pt t Yt YNt t [ ( |1)] ( 1 1 ) (5.13)

According to (5.13) an unanticipated monetary disturbance that takes place in

a country where agents are expecting price stability will lead to a significant

real output disturbance. In (5.13) we observe that output (Yt ) has:

1. a permanent component = + t;

2. a component related to the impact of a price surprise = [Pt – E(Pt | t–1)];

3. a component related to last period’s deviation of output from permanent

output = (Yt–1 – YNt–1); and

4. a random component = t.

Thus, in the Lucas model business cycles are generated by exogenous monetary

demand shocks that transmit imperfect price signals to economic agents

who, in a world of imperfect information, respond to price increases by

increasing supply. The greater is the general price variability (the lower the

variation in price attributed to relative price variation), the lower will be the

cyclical response of output to a monetary disturbance, and vice versa. A

major policy implication of the MEBCT is that a benign monetary policy

would eliminate a large source of aggregate instability. Thus new classical

economists come down on the side of rules in the ‘rules versus discretion’

debate over the conduct of stabilization policy.

We now turn to consider the main policy implications of the new classical

approach to macroeconomics in more detail.