business cycles closed economy summaries

The AS-AD model determines the short-run equilibrium values of output and inflation as the
point of intersection between the upward-sloping short-run aggregate supply curve (the
SRAS curve) and the downward-sloping aggregate demand curve (the AD curve). The model
also determines the long-run equilibrium levels of output and inflation as the point of intersection
between the vertical long-run aggregate supply curve (the LRAS curve) and the AD
curve.
2. When expectations are static, the expected inflation rate for the current period equals the
actual inflation rate observed during the previous period. Under this assumption the AS-AD
model is globally stable, converging gradually towards the long-run equilibrium where output
is at its natural rate and inflation is at its target rate. The adjustment to long-run equilibrium
takes place through successive shifts in the SRAS curve, as economic agents gradually
revise their inflation expectations in reaction to observed changes in the actual inflation rate.
During the adjustment process the economy moves along the AD curve, as the central bank
gradually adjusts the real interest rate in reaction to the changes in the inflation rate.
3. With plausible parameter values the AS-AD model suggests that it will take about four years for
the economy to complete half of the adjustment towards the steady state and about 13 years to
complete 90 per cent of the adjustment. Hence output and in flat on may deviate from their trend
values for quite a long time, once the long-run equilibrium has been disturbed by a shock.
4. The AS-AD model may be specified in deterministic terms or in stochastic terms. In the deterministic
versior of the model, the demand and supply shock variables are non-stochastic. In
the stochastic AS-AD model, the shocks to demand and supply are treated as random
variables.
5. The deterministic model may be used to study the isolated effects of a single temporary or
permanent shock to supply or demand. The stochastic AS-AD model may be used to generate
simulated time series for output and inflation, allowing a calculation of the variance,
covariance and autocorrelation in these variables. These statistics may then be compared to
the statistical properties of empirical time series to investigate how well the stochastic AS-AD
model is able to reproduce the stylized facts of the business cycle.

6. In the short run, a temporary negative supply shock generates stagflation, defined as an
increase in inflation combined with a fall in output. Even after the negative supply shock has
disappeared, inflation will remain above the target rate and output will remain below the
natural rate for many successive periods, because it takes time for inflation expectations to
adjust back to the target inflation rate.
7. A temporary negative demand shock generates a bust-boom adjustment pattern in output. In
the period when the negative demand shock occurs, there is a drop in output as well as inflation.
But when the negative demand impulse disappears, output rises above its natural rate
because the SRAS curve shifts down, due to a fall in expected inflation. In the subsequent
periods, output gradually falls back to its trend level, and inflation gradually rises towards the
target rate.
8. By quantifying the parameter values, the deterministic AS-AD model may be used to generate
impulse-response functions showing how output and inflation will evolve over time in
response to an impulse such as a temporary or permanent demand or supply shock.
9. The deterministic AS-AD model can explain the observed persistence (autocorrelation) in
P.r.onomir. limP. SP.riP.S, lh~t is, it r.~n P.Xf'll~ i n why P.VP.n ~ IP.m['lOr~ry shor.k QP.nP.r~IP.S rrotr~r.tP.cl,
long-lasting deviations of output and inflation from their trend levels. However, the deterministic
model cannot explain the observed recurrent cyclical fluctuations in macroeconomic
variables. But if the demand and supply shock variables are treated as stochastic processes,
the AS-AD model is able to generate irregular, cyclical fluctuations in output and inflation.
10. A calibrated stochastic AS-AD model with static expectations where all shocks take the form
of demand shocks is reasonably good at reproducing the statistical properties of the empirical
time series for output, but such a model generates an unrealistically high degree of persistence
ir the rate of inflation. A stochastic AS-AD model with static expectations where all
shocks originate from the supply side is incapable of reproducing the statistical properties of
the time series for output as well as inflation. This suggests that a model intended to explain
the business cycle must allow for demand shocks as well as supply shocks and that the
assumption of static inflation expectations may be too simple.
11. The hypothesis of adaptive expectations says that the expected inflation rate for the current
period is a weighted average of all inflation rates observed in the past, with more weight being
put on the experience of the recent past than on the more distant past. Static expectations are
a special case of adaptive expectations where the weight given to last period's observed
inflation rate is 1 00 per cent.
12. A calibrated stochastic AS-AD model with adaptive expectations allowing for demand shocks
as well as supply shocks is able to reproduce the statistical properties of the time series for
US output and inflation reasonably well. The stochastic AS-AD model offers an explanation for
the business cycle in line with the Frisch-Siutzky paradigm. In th is paradigm, business fluctuations
are initiated by random demand or supply shocks which are then propagated through
the economic system in a way that generates persistence in macroeconomic variables. In our
AS-AD model, the persistence in the macroeconomic time series stems from the fact that the
expected and actual inflation rates adjust sluggish ly over time. However, one can construct
other macroeconomic models with other persistence mechanisms, so our AS-AD model is
not the only possible explanation for business cycles.

13. Real business cycle (RBC) theory seeks to explain business cycles by fluctuations in the rate
of technological progress which is also driving the process of long-term economic growth. A
Solow model in which the rate of productivity growth is stochastic does a fairly good job of
reproducing the short-run dynamics of output and total hours worked as well as mimicking the
long-run facts of economic growth. The model generates endogenous persistence, since a
positive productivity shock increases output which in turn increases saving, feeding into an
increase in the capital stock which generates another round of increase in output and saving,
and so on.
14. Despite its theoretical attraction, basic RBC theory has a problem explaining the observed
fluctuations in aggregate employment as the outcome of intertemporal substitution in labour
supply. The theory postulates that workers voluntarily choose to work less when real wages
are relatively low, and vice versa. For this theory to fit the data, the elasticity of labour supply
must be much higher than prevailing empirical estimates. Yet RBC theory has made an important
methodological contribution by introducing the practice of setting up stochastic dynamic
general equilibrium models which can be compared with the stylized facts of the business
cycle.

Stabilisation policy
are economic recessions and depressions a social evil which economic policy
makers should try to prevent, or do they reflect the economy's optimal response to
exogenous shocks? During the 1980s and early 1990s several economists
actually defended the latter view. as we saw in the previous chapter. This school of real
business cycle theorists claimed that shocks to productivity are the main driver of business
cycles. and that the economy is reasonably well described by the general equilibrium
model of a perfectly competitive economy. In such an economy Pareto-optimality always
prevails, so even though policy makers may regret that economic activity falls in response
to a negative productivity shock. there is nothing they can do to improve the situation.
given that the economy always makes efficient use of the available resources and
technologies.
Although th is view of the business cycle still has its advocates. most economists
nowadays believe that the competitive real business cycle model is an implausible theory
of short-run business fluctuations. for the reasons given at the end of the previous chapter.
The dominant view is that business cycles reflect market failures and that social wellare
could be raised if business fluctuations could be dampened. A fundamental issue for
macroeconomics is whether policy makers can indeed smooth the business cycle through
macroeconomic stabilization policy.
Stabilization policy involves the use of monetary and fiscal policy to dampen fluctuations
in output and inflation. In this chapter we ofler an introduction to the problems of
macroeconomic stabilization policy. The llrst section of the chapter discusses the goals of
stabilization policy, asking why business fluctuations cause welfare losses, and how the
goals of stabilization policy should be dellned. In Section 2 we use our stochastic AS-AD
model from Chapter 19 to characterize the optimal rnonetary stabilization policy under
alternative assumptions regarding the goals of policy makers and the type of shocks
hitting the economy. Section 3 then analyses the optimaljlsca/ stabilization policy.
The formal analysis in th is chapter maintains our previous assumption that expectations
are backlvard-looking. i.e .. that the expected future rate of inflation is determined
solely by the actual rates of inflation observed in the past. In Chapters 21 and 2 2 we shall
analyse the scope lor stabilization policy when expectations are .forward-looking in the
sense that agents try to form the best possible estimate of future inflation, utilizing all the
relevant information available to them at the time expectations are formed, including
knowledge of the way the economy works.
The present chapter also assumes that policy makers can inunediately observe and
react to fluctuations in the current levels of output inflation. The complications arising
when policy can only react to movements in output and inflation with a lag will be discussed
in Chapter 22. That chapter will also consider the challenges for stabilization
policy emerging when there is uncertainty in the measurement of the output and inflation
gaps, and when the authorities have difllculties establishing the credibility of their commitment
to a low and stable inflation rate.

Stabilization policy is the active use of monetary and fiscal policy to influence the aggregate
demand for goods and services. The goal of stabilization policy is to minimize the social
welfare loss from the volatility of output and inflation.
2. The instability of output and the associated fluctuations in real income impose a welfare loss
on risk averse consumers who prefer a smooth to an uneven time path of consumption.
Problems of moral hazard and adverse selection prevent consumers from insuring themselves
fully against the unexpected temporary income losses caused by business cycles, and credit constraints often prevent unemployed consumers from borrowing against their expected
future labour income. Hence the unanticipated income losses generated by recessions force
some consumers into cutting their consumption, resulting in welfare losses which are not
fully offset by the welfare gains from higher consumption during economic booms, due to
diminishing marginal utility of consumption.
3. The employment fluctuations resu lting from business cycles also create welfare losses for
(some) workers. Market imperfections drive a welfare-reducing wedge between the marginal
product of labour and the marginal rate of substitution between income and leisure. This
wedge goes down when employment rises during a boom, but typically the wedge increases
a lot more when employment falls during the subsequent recession. For this reason the
average level of welfare could be increased if employment could be stabilized around its
average (natural) rate.
4. Empirical evidence indicates that the potential gains from stabilization of employment are
considerable. Empirical studies have also documented that the losses of income and consumption
caused by recessions are borne disproportionately by low-paid unskilled workers.
Hence business cycles tend to exacerbate the unequal distribution of income. This is another
reason why society may wish to stabilize output and employment.
5. Stabilization policy should aim at stabilizing output around its trend (natural) level. Due to
market imperfections, natural output is lower than the Pareto efficient output level, but if policy
makers were to target the efficient output level - seeking to keep unemployment permanently
below its ratural rate - the result would be ever-accelerating inflation. Policy makers may use
structural policies to move natural output closer to the efficient output level, while stabilization
policy should be used to keep actual output close to natural output.
6. Apart from stabilizing output and employment, another goal for stabilization policy is to minimize
the volatility of the rate of inflation around its target rate. Inflation fluctuations generate
welfare losses because they are typically unanticipated, causing relative prices (including real
wages and real interest rates) to deviate from the expected levels on which household and
business plans are based.
7. Even anticipated inflation creates welfare costs arising from an inefficiently low demand
for cash balances, 'menu costs' due to frequent price changes, relative price distortions
stemming from non-synchronized nominal price adjustments, and a distorted measurement
of taxab le income due to an unindexed tax system. Yet it is widely agreed that the target
inflation rate should be kept above 0 to enable the central bank to cut the nominal interest rate
signifi cant~ in a serious recession without hitting the zero bound for nominal interest, and to
facilitate the downward adjustment of real wages to a negative shock in labour markets with
downward nominal wage rigidity.
8. A fiscal or monetary policy rule prescribes how the instruments of stabilization policy should
be set, given the observed state of the economy. Following a fixed policy ru le may help policy
makers to establish credibility and to increase the predictability of economic policy so as to
minimize welfare-reducing expectational errors in the private sector. The alternative to fixed
policy ru les is discretionary policy. Under discretion, policy makers may conduct monetary
and fiscal policy in any way they believe will help advance the goals of stabilization policy,
taking account of any special circumstances which might prevail. Discretion allows more flexibility than ru les, but typically at the cost of reduced credibility and predictability. The evi·
dence suggests that the monetary policy of the most important central banks is fairly well
described by some form of Taylor rule, although policy makers never mechanically follow a
fixed rule.
9. When business fluctuations are caused solely by demand shocks, the central bank interest
rate should decrease sharply in response to a fall in inflation and output to minimize the social
welfare loss from the volatility of output and inflation. The interest rate reactions should be as
strong as possible, subject to the constraint that the nominal interest rate cannot fall below 0.
There is no dilemma between reducing the variance of output and that of inflation when
business cycles are driven by demand shocks.
1 0. If business cycles are generated exclusively by supply shocks, and monetary policy makers
are mainly concerned about minimizing the fluctuations in output, they should cut the real
interest rate aggressively in response to a fall in output and increase it very little in response
to a rise in inflation. By contrast, if a central bank faced with supply shocks focuses mainly on
keeping the inflation rate low and stable, it should raise the real interest rate sharply in reac·
tion to a fall in output and increase it aggressively in response to a rise in inflation. Whatever
the main policy objective, a lower variance ot output can only be achieved at the cost ot a
higher variance of inflation, and vice versa, when the business cycle is driven by supply
shocks.
11. In the general case where the economy is exposed to demand shocks as well as supply
shocks, there is always a trade-off between reducing the variability of output and reducing the
variability of inflation when stabilization policy is optimally designed.
12. A general feature of the optimal monetary stabilization policy based on the Taylor rule is the
so-called Taylor principle, according to which the real interest rate should always be raised
in response to a rise in the rate of inflation, regardless of whether shocks to the economy
originate from the demand side or from the supply side.
13. Fiscal stabilization policy may take the form of changes in public spending in response to
changes in the output gap and in the inflation gap. Whenever it is optimal to tighten (relax)
fiscal policy, it is also optimal to tighten (relax) monetary policy, so as long as the zero interest
rate bound on the nominal interest rate is not binding, fiscal stabilization policy cannot achieve
anything which could not be achieved through monetary policy. However, in a deep recession
where the short-term nominal interest rate has been driven down close to 0, expansionary
fiscal policy may help monetary policy makers to pull the economy out of the recession.

Rational expectations and stabilisation
Economic activity today depends crucially on expect.ed economic conditions tomorrow.
A drop in the economy's expected future growth rate will tend to reduce the
propensities to consume and invest by reducing the expected future earnings of
households and firms. Hence the aggregate demand curve will shift do'lllrn, causing an
immediate fall in current output. As another example, a change in the expected rate
of inflation will shift the aggregate supply curve by feeding into the nominal wages
negotiated by workers and firms . It may also move the aggregate demand curve through
its impact on the expected real rate of interest. The expected inflation rate is thus an
important determinant of current economic activity.
Conventional macroeconomic models often assume that the expected future values of
economic variables depend only on the past history of those variables. Indeed, in the previous
chapters we typically postulated that the expected inflation rate for the current
period is simply equal to the actual inflation rate experienced in the previous period. ThL~
assumption of backward-looking e).:pectations may be plausible in 'quiet' times when the
macroeconomy is not subject to signiHcant shocks. When people have no particular
reason to believe that the tightness of labour and product markets next year will be much
dill'erent from what it is today, it seems reasonable for them to assume th at next year's
inflation rate will be more or less the same as this year's. However. if the economy is hit by
an obvious and visible shock such as a dramatic change in the price of imported oil. or if
there is a clear change in the economic policy regime, say, due to a change of government.
it does not seem rational for people to assume that next year's economic environment
will be the same as this year's. Instead of just mechanically extrapolating the past into
the future, rational households and firms will seek to utilize all the relevant information
available to them when they form expectations about the future state of the economy.
In the early 1970s, some macroeconomists took this idea of'forward-looking expectations
to its logical limit by advancing the rational expectations hypothesis (REH) . 1 According
to the REH, people use all the available information to make the best possible forecasts of
the economic variables which are relevant to them. Moreover, tire available irrjimnatior1
includes ilifimnntion about the structure of tile economy. The idea is that, even though in
practice the layman may not know much about the way the economy works. the economic
forecasts produced by professional economists are available to the public through
the media. so in this way people have access to the most competent forecasts of. say. next
year's rate of inllation. Economists should therefore model the fonnation of expectations
as if people use the relevant economic model to predict inl1ation and other economic variables
which are important lor their economic decisions. In other words. rational expectations
are model-consistent expectations: they are identical to the forecasts one would
make by using the available knowledge of the structure of the economy. as embodied in
the relevant economic models. Another way of putting it is to say that economic analysts
should not assume that they are smarter than the economic agents whose behaviour they
are trying to predict. Instead, they should assume that agents form their expectations in
accordance with the analysts' own description of the economy. If they did not. and if the
analysts' model is correct, then agents would be making systematic expectational errors,
and presumably this would induce them to change the rules of thumb by which they form
their expectations until there is no discernible systematic pattern in their forecast errors.
This idea of rational expectations essentially revolutionized macroeconomic theory.
The REH is obviously a very strong assumption, and as 'Ne shall see. it can be criticized on
theoretical as well as empirical grounds. But before addressing these criticisms. this
chapter will explain the case lor the REH in more detail and derive some of its striking
implications. Our main purpose is to illustrate the importance of the way expectations are
formed. In particular. we will show how the effects of macroeconomic stabilization policy
may differ signillcantly depending on whether expectations are rational or backwardlooking.
The 11nal sections of the chapter will discuss the validity of the REH. drawing on
theoretical arguments as well as empirical evidence.

The assumption of backward-looking (static or adaptive) expectations is hard to reconcile
with rational behaviour because it implies that economic agents may make systematic forecast
errors.
2. As an alternative to backward-looking expectations, economists have developed the rational
expectations hypothesis {REH) according to which an agent's subjective expectation of an
economic variable equals the objective mathematical expectation of the variable, calculated
on the basis of all relevant information available at the time the expectation is formed.
3. The REH assumes that the information available to agents includes knowledge about the
structure of the economy. Hence rational expectations are model-consistent: they correspond to the predictions of the relevant economic model. This does not require that ordinary people
are able to solve economic models, since the average person may rely on the publicly avail·
able forecasts of professional economists.
4. Some macroeconomic models with rational expectations have led to the Policy Ineffective·
ness Proposition (PIP) which claims that systematic demand management policy cannot
affect real output and employment because the private sector will fully anticipate the effects
of systematic policy on the rate of inflation.
5. The PIP is nowadays considered unrealistic, since the central bank can typically change its
policy in reaction to new economic developments occurring after nominal wages have been
temporarily locked into existing contracts. Because nominal wages and prices only respond
to economic shocks with a lag, systematic monetary policy can affect output and employment,
even if the policy is fully anticipated by rational agents. In these circumstances the optimal
monetary stabilization policy is qualitatively similar to the optimal monetary policy response to
shocks under backward-looking expectations, although it is quantitatively different. Thus ratio·
nal expectations do affect the optimal policy, but do not make stabilization policy ineffective.
6. The REH has led to the Lucas Critique which says that an econometric macro model which
was estimated under a previous economic policy regime cannot be used to predict economic
behaviour under a new policy regime. The reason is that a c~ange in the policy regime will
affect private sector behaviour, including the way in which expectations are formed.
7. Under rational expectations the announcement of future changes in economic policy will influ·
ence the economy already at the time of announcement, even before the new policy is imple·
mented. In particular, the flexible prices of financial assets such as stocks will 'jump'
instantaneously at the time of announcement and will then gradually adjust towards its new
long-run equilibrium value as the date of implementation of the policy change comes closer.
8. The REH has been criticized for being unrealistic because economists differ in their views of
the workings of the economy, making it difficu lt for the average person to base their expecta·
tions on expert forecasts. Defenders of the REH argue that it is not safe to base economic
policy evaluation on the assumption that policy makers are systematically better informed than
the private sector. Hence policy makers should assume that the knowledge embodied in their
economic models is also available to the private sector, as implied by the REH.
9. When consumers with rational expectations seek to smooth their consumption over time,
private consumption will follow a random walk, changing only as new information about future
incomes becomes available. The random walk hypothesis implies that the current consump·
tion level is the best forecast of future consumption (adjusted for underlying trend growth) and
that consumption changes only in response to unpredictable changes in income. By contrast,
under static expectations the change in consumption corresponds to the change in the con·
sumer's total income. Empirical evidence suggests that a large part of aggregate consump·
tion follows a random walk, consistent with the REH, but at the same time many consumers
behave as if they have static expectations.
10. In an economy where some consumers have rational expectations and others have static
expectations, the average expected inflation rate may be written as a weighted average of the
central bank's inflation target and last period's inflation rate. The weight given to the inflation
tarQet may be interpreted as the fraction of aQents informed about the monetary policy target even if these people do not have all the information needed to form strictly rational expectations
regarding the short run.

Limits to stabilisation
monetary and fiscal policy makers have the ability to stabilize the macro
 economy. thereby reducing the social costs ofbusiness cycles? In the two previous
 chapters our answer to this basic question in macroeconomics has been: yes.
However, our analysL~ was based on some important simplifying assumptions. First of all.
we assumed that policy makers have perfect informatioll about the current state of the
economy. Second. we assumed that they can react immediately to this information and
that their policy actions have predictable and well-known quantitative ejfects. Third, in the
case with forward-looking agents we postulated that any policy rule announced by policy
makers is always considered fully credible by the public implying, for example. that policy
makers never have any problem convincing the public that they will stick to an antiinfl
ationary policy. Taken together, these assumptions are very optimistic and not very
realistic. In this chapter we shall study the problems of stabilization policy when these
strong assumptions are replaced by more realistic ones.
We will start by studying the problems of establishing the credibility of an antiinfl
ationary monetary policy. This part of our analysis will show how our AS-AD model
combined with the hypothesis of rational expectations may provide a theoretical case
for delegatio11 of monetary policy to an independent cep·rtr-a! bank, as a lot of countries
have actually done in recent years. Other parts of the chapter investigate how uncertainty
about the current state of the economy and time lags in the implementation of policy
limit the scope lor stabilization policy. In the llnal section we focus particularly on Hscal
policy, investigating whether llscal policy makers have historically been able to dampen
the business cycle.


With rational expectations monetary policy may suffer from a credibility problem when the
socially desired output level exceeds the natural level and the central bank can act in a
discretionary manner after the private sector has formed its expectations. If monetary
policy makers announce that they will keep the inflation rate down to a certain target level, the
private sector may not consider such a statement to be credible when the central bank can
boost output and employment by generating unanticipated inflation. The problem is that a
ru le-based policy of price stability is not time-consistent.
2. In a time-consistent rational expectations equilibrium with discretionary monetary policy, rational
private agents anticipate the central bank's incentive to stimulate output by generating
surprise inflation. The expected and actual inflation rate then becomes so high that the central
bank does not wish to generate additional (unanticipated) inflation. The end result is an equilibrium
with inflation above the target rate and an output level equal to the (suboptimally low)
natural rate. Hence monetary policy suffers from an inflation bias under discretionary policy. A
socially superior outcome could be achieved if the central bank could credibly commit to pursuing
a ru le-based policy of price stability.
3. If the policy maker cares sufficiently about the future, he may not have an incentive to generate
surprise inflation in the short run, because he realizes that this will cause an unfavourable
upward shift in the expected inflation rate in the longer run. In such a case the inflation bias in
monetary policy may be held in check by the policy maker's desire to build up a reputation as
a defender of price stability. The inflation bias may also be reduced by delegating monetary
policy to an independent and 'conservative' central bank which puts more emphasis on price
stability than the government itself.
4. Errors in the measurement of macroeconomic data such as the output and inflation gaps
imply that monetary policy makers sometimes set the 'wrong' level of interest rates which is
not adequately tuned to the true state of the economy. In this way measurement errors contribute
to macroeconomic instability. The greater the average magnitude of the measurement
errors, and the higher the degree of persistence in these errors, the smaller is the optimal
interest rate response of the central bank to changes in the estimated output and inflation
gaps.
5. In practice, it is also a serious problem for macroeconomic stabilization policy that there may
be considerable lags from the time the economy is hit by a shock to the time when the economic policy response attains its maximum impact. In the meantime the state of the
economy may have changed so that the policy change turns out to be inappropriate. The
inside lag is the period from the time an economic disturbance arises until the time when a
change in the economic policy instrument has been implemented. The outside lag is the
period from the time the policy instrument is changed until it achieves its maximum impact on
the economy.
6. Typically it takes about two years before a change in the interest rate attains its full impact on
the rate of inflation. To cope with th is outside lag, a central bank which has been given an
inflation target should aim at a forecast for the inflation rate two years ahead. If the inflation
forecast is above the target inflation rate, the interest rate sho·Jid be raised until the inflation
forecast is on target, and vice versa. One can show that such inflation forecast targeting leads
to a version of the Taylor rule for interest rate policy where the coefficients on the inflation gap
and on the output gap reflect the way these variables affect the future inflation rate.
7. Many economists believe that it is easier to stabilize the economy through monetary rather
than fiscal policy because monetary policy is generally subject to much shorter inside lags.
The empirical evidence shows that in many Western European countries fiscal policy has
in fact tended to be procyclical in recent decades, thereby amplifying the business cycle. At
the same time there are indications that several countries have shifted from destabilizing
procyclical to stabilizing countercyclical fiscal policies during the last decade.