Реферат: Classical Macroeconomists And Stabilisation Policies Essay Research

Classical Macro-economists And Stabilisation Policies Essay, Research Paper

The “policy inefficacy” principle is probably one of the most famous and controversial assertions of the New Classical School which is often regarded as a (counter)revolution in macroeconomic theory. We shall see, however, that despite the important insights and innovations of the approach, most of its conclusions including the invariance postulate are based on quite restricted and unrealistic assumptions and assertions which may make them of doubtful practical use. The NC approach is characterised by its “relentless drive for microfoundations” and in particular the desire to base their theories on the assumption of the Rational Economic Man as a reaction perhaps to such Keynesian theories as “animal spirits”, money illusion or the liquidity trap which seemed to imply “irrational” individual behaviour. Economic agents are, thus, seen as consistent and successful optimisers to the limits of their information which implies that agents are continuously in equilibrium. This coupled with the assumption of universal perfect competition ensures that markets always clear, unlike most “classical” – and monetarist – economists who considered equilibrium as a limiting case. NCs therefore work within a Walrasian General Equilibrium unlike Monetarists who often preferred a Marshallian partial equilibrium approach. As required by Walrasian equilibrium the Classical Dichotomy is also re-established and Aggregate Supply (AS) is assumed to depend only on relative prices. As we shall see NCs restated even more strongly than monetarists the neutrality and superneutrality of money. Another important feature of the NC school is its adoption of Rational Expectations (RE) first developed by Muth. RE assume that people engage in a cost-benefit analysis of whether it is worth searching for certain forms of information and then make efficient use of the information available to them by understanding how the economy works and continually revising expectations to take into account of any new information. NC theory focuses in particular in the efficient use of information aspect of RE, having little to say on how this information is obtained. Lucas and Prescot thus defined RE as the situation where the subjective probability distribution of future economic variables held at any time t coincides with the actual, objective conditional distribution based on the information assumed to be available at that time t. NC economists accepted and extended Friedman’s notion of the natural rate of unemployment which refers to the level of unemployment consistent with stable inflation and labour market equilibrium and is, thus, thought to be “voluntary”. For Friedman adaptive expectations meant that though the Long Run Phillips Curve is vertical on the natural unemployment rate – and correspondingly AS is vertical on the level of “permanent” output, the Short Run Phillips Curve and AS are downward and upward sloping respectively. This was so because inflationary expectations were slow to adjust and, therefore, workers and -sometimes- employers suffered from money illusion. Thus Friedman did allow for Demand Management (monetary policy only of course) to affect output in the short run. The introduction of RE, however, meant that any systematic govt policy will not be able to affect output because rational agents will – immediately – adjust their inflationary expectations. Keynesians were, thus, accused of overlooking the fact that structural parameters change with government policy. This inability of a feedback money supply rule to affect output was termed by Sargent and Wallace as the “superneutrality” of money. Temporary divergences from the natural rate could only occur if the authorities – or any real shocks – managed to surprise individual agents. Hence, Lucas “surprise” AS function is derived: Y = Yp + a ( P – Pe ) (1) where Yp is the mean or permanent level of output, Pe is the expected aggregate price level and a is a positive constant. The above equation can also be expressed in logs so as to represent rates of change of output and prices (i.e. inflation). Nevertheless, an explanation is required of why incorrect inflationary expectations alter output. The answer NCs give is imperfect information (this is often explained in terms of the “island parable”). Agents have information of the current price (PX) of the good X they produce but they cannot observe all prices in the economy and hence do not know the current aggregate price level (P). Thus, following a change in PX, agents have to distinguish between a change in the relative price of their good and a change in the aggregate price level, i.e.: they are faced with a problem of “signal extraction”. Typically they will interpret it partly as a relative and partly as an aggregate price movement as shown in (2): Pe = b + c PX (2) The larger c is, the greater is the proportion of changes in PX attributed to a change in the price level. Thus, the more a country varies its money supply growth to affect output, the greater will be the correlation between P and PX and, thus, c will get larger, making signal extraction more difficult. Another question is how is it possible that mistakes in expected inflation cause that big changes in output and employment. The answer given by Lucas here is a distinction between temporary and permanent changes in the relative price of labour (real wage). Permanent changes cause little changes in employment, perhaps even of unknown sign because leisure is a normal good. Temporary changes in real wages, however, can have much larger effects in employment because of the so-called intertemporal substitution of leisure: Workers prefer to work more when real wages are higher than normal and consume more leisure when wages are low. Any (perceived) change in real wages will again be typically partly thought as temporary and partly as permanent; here agents face a double signal extraction problem. Hence to the extent a monetary shock makes agents think a temporary change in their real wages has taken place, it will have a fairly big effect on employment and we, thus, have a downward sloping (very)Short Run Phillips Curve as shown in figure 1. Imperfect information and intertemporal substitution of leisure has been used by Lucas and others to form monetary explanations of the business cycle while alternative NC real explanations of the cycle are also based on this postulate. The important thing to grasp here is that both the movements along the SR Phillips Curve and the trade cycle are not thought to be indicators of (even temporary) disequilibria but optimising and hence equilibrium responses to the available – imperfect – information. This is why Lucas argued that Keynes distinction between voluntary and involuntary unemployment was a misdirected attempt to distinguish between trend and cyclical unemployment. A problem with this theory is that it can explain only very short run fluctuations in output and employment. Indeed, this model suggests that output should be random walk around this trend i.e. its deviations from trend should be serially uncorrolated since agents will be very quickly able to find out if the aggregate price level changed. Obviously output and employment show significant persistence so Lucas has tried to improve the explanatory power of his AS function with an additional term representing deviations of past output from the mean as shown below: Y = Yp + a ( P – Pe ) + g (Y-1 – YP ) (3) This was justified by assuming that money supply shocks are serially correlated so that the current shock carries information about the future path of money stock. Further more, real shocks are likely to be more persistent since it takes more time to find out if the changes caused are permanent or not. Investment and capital stock may also be slow to respond to relative price changes. Though this extension did help its performance, it is not very persuasively justified and has been accused of putting ad hoc elements to NC theory. As we have seen the major innovations and differences of the NC school concern the Supply side of the economy. Indeed their Aggregate Demand (AD) analysis is not very different from the Monetarists but it should be noted that not all – though most – NCs would subscribe to the Quantity Theory of Money. Many NC economists, have argued that many forms of government demand management will be ineffective in affecting even AD itself. As far as fiscal policy is concerned, the monetarist Crowding-out postulate and the more fundamental Ricardian Equivalence are such examples. Very briefly, the Ricardian equivalence, restated by Barro, asserts that debt-financed fiscal policy will be neutral because agents will realise that in the future extra taxes will be imposed to repay the debt. This is so provided agents have infinite lives (at least relative to when debt will be repaid) or if the utility of children is of some concern to the parents (overlapping generations model). Various criticisms have been levied to this such that it assumes perfect capital markets, requires too much foresight and it ignores income redistribution effects and that a certain constant budget deficit can be indefinitely maintained under certain conditions. Probably the most powerful criticism is why then do budget deficits attract so much attention. As far as monetary policy is concerned, Wallace, for example, argued that the composition of Central Bank’s portfolio is irrelevant so open market operations have no effects on AD something which contradicts the Quantity Theory. As I said before, though, the NC case for policy inefficacy is not based on AD analysis. Still, it has been argued that to the extent that monetary and fiscal policy affect the composition rather than the level of AD through such channels as Investment grants or effects of inflation in the real interest rate, this will affect the long run value of output. The response of NC to this is that to the extent to which such effects exist they will be harmful since they will cause inefficiencies. The obvious counter argument to that is the existence of externalities but we are now digressing to micro theory and welfare economics. The obvious and more fundamental Keynesian criticisms to the policy invariance principle centre on the basic tenets of NC macroeconomics, namely RE and universal market clearing. It is clear that once one of these two assumptions is even partly violated, the NC conclusions are invalidated. Again very briefly, RE in their weak form that people do their best given their information seems quite plausible but its “strong” NC form which assumes that agents actually know the correct model of the economy (and that this model is the NC one) is much less obvious. It also sometimes argued that once the implicit assumption of symmetric information of the government and the private sector is relaxed, systematic policy could be effective. NCs doubt that this is the case and argue that even if so, the best course of action would be the government to spread the information. As far as market clearing is concerned it is an open issue to what extent non-market clearing is consistent with optimising behaviour since not all of the numerous models which supposedly rationalise disequilibrium are in fact fully “rational” as for example implicit contract theories which, in general, assume risk averse workers as far as wages are concerned but not as far as employment is! Nevertheless, there is overwhelming evidence that markets are not continuously in equilibrium and as Laidler pointed out we should take that into account even if we cannot entirely explain it. Another, probably somewhat less powerful, criticism of the NC position is the question why NC economists oppose so vehemently AD policies. Why NCs oppose discretionary AD management aimed to “surprise” agents is clear since it will contribute to inflation and possibly cause trade cycle type fluctuations. It is much less clear though why NCs also oppose stabilisation policies to smooth the trade cycle since if they are right such policies are simply ineffective. Friedman’s answer to that question would be that such stabilisation policies are most likely to prove destabilising because of the long lags in the detection of a problem, subsequent decision making and implementation and in the policy variables affecting the target ones combined with uncertainty about these lags and the quantitative effect of an action. Various political considerations in Public Choice-type situations will only make matters worse. NC economists on the other hand will put more emphasis on the bad effects of discretionary policies on the private sector. As we had seen before, a greater volatility of the money supply will only make signal extraction more difficult for the individual agents by increasing the variance of output. Furthermore, the gradual adjustment of markets back to long run equilibrium does not necessarily imply market failure because in an uncertain world agents may rationally prefer to move slowly so as to reduce the cost of incorrect adjustment… Any government intervention with this optimal path of adjustment will cause inefficiencies. As Tobin pointed out, however, these costs are hardly more significant than the shoe-leather costs of inflation. This takes us to another issue we have to examine, namely what fixed money supply growth rate will NC propose. NCs have an even greater problem than monetarists in explaining why inflation is bad (though of course the NCs have not made inflation their prime enemy as monetarists did). Nevertheless they do seem to agree with Friedman’s X% rule, probably in the grounds that the inflation rate which disrupts the least the transmission of relative price signals is zero inflation, though some NCs like Sargent and Wallace would argue for simultaneous balanced budgets in order to be able to control the money supply in the long run. Actually there has been an extensive literature of NC models of optimal policy making which take into account endogenous private sector expectations, largely based on game theoretic formulations and such things as the “time inconsistency problems”, credible commitments and reputation building. Such models, however, have been accused of ignoring the importance of political and electoral considerations on policy making. Concluding, NC macroeconomics and their policy inefficacy conclusion are more impressive theoretical constructions than really useful models of how the economy actually operates. The fact that they do not perhaps fully explain their dislike of stabilisation policy is certainly not the weaker point in a theory that takes for granted universal market clearing. Nevertheless, they have managed to force on the theoretical and policy agenda numerous important issues such as RE, the importance of microfoundations and of rational agents and even the importance of credibility in government policy. Macroeconomics, and probably most notably Keynesian macroeconomics will never be the same again.

еще рефераты
Еще работы по на английском языке