This article argues that imperfections in the working of individual (nonfinancial)markets are not a clear source of macroeconomic instability and that betterregulation of these markets is not the way to lize the economy. Improvement heremeans both increased surveillance and removing government-fostereddeficiencies. The basic arguments are (1) the long-standing disarray thatdistinguishes macroeconomics from microeconomics greatly increased startingin the 1970s, (2) great dispute exists over the workability of different "classic"macroeconomic measures, (3) despite at least seven decades of advocacy ofmicroeconomic measures for macroeconomic goals, no defensible case exists, (4)support for microeconomic measures rests on a belief in a high degree of marketfailure about which microeconomists have become more skeptical, (5) theseweaknesses imply that no clear macroeconomic benefits offset themicroeconomic drawbacks of regulations of individual markets, and (6) extensivederegulation would produce substantial microeconomic and macroeconomicbenefits, but the microeconomic case is much stronger.

JEL classification: K00, E00

Keywords: Macroeconomics, Keynes, Law and Economics

1. Introduction

This article examines contentions that imperfections in the working of individual(nonfinancial) markets are a source of macroeconomic instability and that betterregulation of these markets is the most feasible way to offset theseimperfections. Improvement here means both increased surveillance andremoving government-fostered deficiencies. The overriding theme is theapproach has little support.

The basic arguments are (1) the long-standing disarray that distinguishesmacroeconomics from microeconomics greatly increased starting in the 1970s,(2) great dispute exists over the workability of different "classic" macroeconomicmeasures, (3) despite at least seven decades of advocacy of microeconomicmeasures for macroeconomic goals, no defensible case exists, (4) support formicroeconomic measures rests on a belief in a high degree of market failure aboutwhich microeconomists have become more skeptical, (5) these weaknesses implythat no clear macroeconomic benefits offset the microeconomic drawbacks ofregulations of individual markets, and (6) extensive deregulation would producesubstantial microeconomic and macroeconomic benefits, but the microeconomiccase is much stronger.

Dealing with these points requires much effort. The prior two paragraphs, notonly make sweeping assertions, but use numerous ambiguous terms. The articlethen seeks to undertake four tasks. (1) The terms are clarified. (2) Selected partsof the debates over macroeconomics are reviewed. As the literature review belowseeks to suggest, the presentation necessarily cannot even fully cover everyidea encountered. Choice is limited to those that seem more relevant. Others can(and indeed in the refereeing process did) stress different viewpoints. (3) Themicroeconomics of market behavior are sketched. (4) Perspective is provided onthe history of proposals to use microeconomic policies to secure macroeconomicgoals.

2. The Nature of Macroeconomics Recalled

While economists occasionally use the term macroeconomics to describe anyhighly aggregative analysis, the more usual concepts relate to economy-wideinstabilities, particularly in unemployment rates but also involving inflation andbalance of payments problems. This clearly is the stress of the manymacroeconomics texts. Since the rise of extensive formal studies ofmacroeconomics, the traditional concerns with the allocation of resources inmarkets became microeconomics.

Actually, both branches deal with the total economy. Macro concentrates onhow the combined behavior produces instabilities. Micro stresses how well eachcomponent of the economy performs the task of making useful goods availableto consumers.

Nothing in economics is neat, and this is true of the borders between macroand microeconomics. The banking system is both a micro and a macro concern.The role of banks in money supply is a basic concern of macroeconomics. Therole of banks in serving individuals involves employing the standard tools ofmicroeconomics. In practice, a further fuzzing arises from conventions adoptedin standard texts. Economic growth, at least as conventionally modeled, is clearlya microeconomic problem, and any correction is by policies affecting individualmarkets. To be sure, more applied discussions recognize the impacts ofalternative government tax and spending policies. For example, U.S. politicaldebates present, albeit in the overly loose fashion necessarily adopted inpolitical debates, a choice between a Democratic model focusing on growthpromoting government spending and a Republican view focused on makingmore money available for private sector investment. Conventionally, micro textsat both the advanced undergraduate and graduate level ignore economic growth.(A major exception is Mas-Colell, Whinston, and Green, 1995.) Macro texts, incontrast, typically cover growth.

Unemployment is central because it is harder to explain or correct thaninflation and balance of payments problems. The basic causes and cures of thelast two were established by the eighteenth century (as in David Hume's essaysthat relate to economics). In contrast, neither what causes unemployment norwhat if anything can be done about it is well determined. Many explanationsexists, but all have severe (and widely examined) drawbacks. This article arguesthat indeed these defects are so severe that any public policies must assume thatthey are dealing with a mysterious disease whose origin and untreated responseare unknown.

What is most important here is that macroeconomics concentrates onmonetary and fiscal policy. Increasing or decreasing control of individualnonfinancial markets has a decidedly secondary role. Macroeconomic theoristswho contend that problems in nonfinancial markets or in their regulation are themain cause of instability often do not advocate cures involving directly alteringcontrol of these markets. One has to resort to nonconformist writings to findstrong arguments for regulatory approaches to instability. Kelman's (1993) effortto examine the issue tries valiantly to find good rationales but only identifiesdrawbacks.

3. The Tools to Consider

The central policy distinction made here is between the monetary and fiscalpolicies that are the focus of traditional macroeconomics and regulatoryinitiatives. Monetary policy relates to control by various means of the supplyof money in the economy. (Such financial innovations as mutual funds thatinvest in short-term securities and allow owners to withdraw funds by writingchecks and credit and debit cards have increased the always difficult problem ofdistinguishing money from other financial assets. This problem is not relevanthere.) Fiscal policy relates to the effects of government tax and spending policy.

Regulation here means those government policies that control the behaviorof individual firms and households in the economy. The concept consideredhere is somewhat broader than that used in other discussions of regulation. Inparticular, a key element proves government policies governing thecompensation and rights of workers and the treatment of trade unions."Regulatory" economists tend to leave most of these issues to labor economists.Stress is on good market controls; the only labor policies treated are health andsafety regulation. Regulatory policies are the focus of many articles in thisencyclopedia. For that reason and because of space limitations, I neitherdelineate in detail nor evaluate the microeconomic problems of such policies.The formidable difficulties of producing predictable results from such policiesis taken as proved elsewhere.

A more critical consideration is that the scope of proposed regulations maybe greater when macroeconomic considerations are added to microeconomicconcerns. The macroeconomics literature talks of incomes policy that involvescontrols of prices and wages throughout the economy. Microeconomic practiceemphasizes concentrating on sectors in which monopoly power exists or marketsfail to internalize all costs. This distinction can be interpreted in at least threedifferent ways. First, the two arguments may be equivalent ways of stating thesame view; economywide may really only mean monopoly sectors. Second,macroeconomists may see more monopoly than do microeconomists. Third,macroeconomists may feel that many sectors that cause no microeconomicproblems are macroeconomic threats. Some indications arise that enthusiasts ofwage and price controls, in fact, see a greater prevalence of extensive monopolypower than do microeconomists specializing on monopoly problems.

4. An Overview of the Macroeconomic Policy Debates

Stress here is on conflict between "Keynesians" and classical economists. Bothapproaches encompass many different, often mutually incompatible specificanalyses. For present purposes, examples of each position that seemed mostgermane are presented. The Keynesian position is identified with the view thatreal economies have features that produce large, undesired, and undesirableinstability and that feasible "active" public policies exist to improve onunregulated performance. Active means closely viewing economic behavior andreacting to it.

The classical position involves numerous criticisms of the Keynesianoutlook. In particular, an influential new classical group of macroeconomists hasdramatically expanded the demonstration of the impediments to successfulgovernment programs to stabilize the economy. A key unfluence on acceptanceof the case is rejection by economists from all viewpoints of the 1930s belief thatdeep extended depressions are an ever present danger. The arguments are notconclusive. However, they have enough plausibility that serious considerationmust be given the possibility that feasible active policies are ill-defined, if notnonexistent.

Even if this conclusion is rejected, it still can be inferred that whatever isdone must be limited by implementation problems. Behind the bitter debates mayonly be a minor quarrel about exactly how small is the scope for action. Theanalysis here focuses on this narrowing of the ambitions of active monetary andfiscal policy from the more ambitious proposals of the 1936 to 1968 period. Tothe extent they support anything, modern Keynesians advocate restrainedintervention that has been called coarse tuning (Lindbeck, 1993, p. 154) (inobvious response to the excessive prior claims that one could fine tune theeconomy). Much effort was devoted to arguing that active policies weredesirable without indicating whether the activism significantly differed fromfollowing policy rules. Attention turns to arguing that, given these limitationsand the widely known microeconomic drawbacks, increased regulation ofnonfinancial markets is not an attractive alternative stabilization policy.

Unfortunately, reasonably treating these supposedly limited questionsrequires examination most of the thorniest issues of both macroeconomics andmicroeconomics. In particular, the proposition that regulating individual marketsis a desirable macroeconomic policy presumes that actual economies are bestdescribed by theories of imperfect markets. Acceptance of the empiricalrelevance of such theories, moreover, does not suffice to justify marketregulation. It must also be shown that such regulation is a desirable way tooffset the effects of market imperfections. In particular, the intervention mustreduce unemployment and not produce other effects, such as increased inflationor undesirable impacts on the regulated markets, that cause harms that outweighthe employment benefits.

The issues have been major concerns in economics for nearly sevendecades. All of the key questions remain unresolved and indeed often not evenclearly raised. The characteristics, causes, and cures of economic instability andhow best to analyze them are all bitterly debated. A growing stress on theoreticprowess may have caused analysts inadequately to consider the empiricalrelevance of the theories. At least three issues arise about macroeconomics. Thefirst is what comprises the theoretically sound models of instability. The secondis which of these models best explains reality. The issue about theory choicestressed here is whether market imperfections are the primary causes ofeconomic instability. It must be shown that the theoretically possible alternativemechanism prevails in practice.

The third concern stressed here is what the realistic models say about thecorrectability of behavior. They must show that the characteristics of theeconomy also allow effective stabilization policy.

5. Viewpoints

The combination of possible viewpoints produces a mass of alternativepositions about problems, solutions, and the best ways to analyze them. Evenwithout considering the many variant positions on how to analyze the issues,at least five policy postures can be delineated. Several different ways exist toreach each of the policy outlooks. Given the underlying complexity, thecategories are devised as epitomes to make the discussion manageable. (Theclassification initially was designed to recognize distinctions made by Kelman(1993) and overcome their drawbacks, particularly his failing to distinguishbetween the two radically different branches of new classical economics.) Theclassification is among traditional Keynesians, microinterventionists,deregulators, microkibbitzers (or Kennedy Keynesians), and skeptics aboutintervention.

To examine alternative routes to these policy postures, the valuableambitious survey by Snowdon, Vane, and Wynarczk (1994) distinguishes amongclassical Keynesians, new Keynesians, Post-Keynesians, monetarists, marketclearing classicists, real business cycle advocates, and Austrians. As discussedbelow, the last four each develop somewhat different cases against activestabilization policy.

The position associated here with Keynesians, as noted, is that activemonetary and fiscal policy is effective and preferable as an anti-unemploymenttool. This certainly is the classical Keynesian position, and a large part of newKeynesian economics is devoted to defending against new classical criticism.Some post-Keynesian economists advocate wage and price controls.

Monetary and fiscal policy can take many forms. Thus, general approval ofactive stabilization involves support of many different specific practices. Whatis feasible is ill-discussed. The present treatment stresses the problems of anyforms of activism and does not examine exactly what might be feasible. Forreasons discussed below, labeling these views Keynesian is common but notnecessarily universal.

The microinterventionists and deregulators believe that better governmentsupervision of individual firms throughout the economy can contribute toreducing unemployment or at least allow the reduction to occur with lessinflation than if only monetary and fiscal policy are used. Microinteventionistsbelieve that unemployment is a serious problem originating from inherent marketfrictions and most appropriately cured by increasing regulation of individualmarkets. Deregulators, who tend to doubt the severity of economic instability,see government as creating the critical barriers to good performance and wantto decrease regulation.

The increased regulation outlook had its height in the 1930s. The extensivethrashing about for explanation of the profound economic collapse in thatdecade produced many theories. A number stressed the role of rigidities in theeconomy. The supporters differed considerably in what they proposed.Suggestions were then made for either extensive national economic planning toregulate private market behavior or radically restructuring the economy byvigorous application of U.S. antitrust laws (see below). Some of the advocatessurvived long after World War II but attracted little intellectual support.Politicians, to be sure, have acted on acceptance of the belief. In contrast, theoverregulation thesis is largely noted in passing by the most avidantigovernment economists. (Kelman gives the macroeconomic elements ofthese arguments greater prominence than they ever have secured in theeconomic literature.)

The failure of massive depressions to emerge since World War II lessened,but did not eliminate, concerns. The question of whether more directlyconfronting market or government imperfections was an effective strategypersisted but never dominated. Support for such measures was limited.

At least two episodes arose in the United States. They were inspired largelyby the persistent problems of taming inflation generated by the Korean and VietNam wars. Towards the end of the Eisenhower presidency, much discussionarose of cost inflation. The Kennedy administration devised a response, but itwas the alternative option discussed next. The Nixon administration wasenduring inflationary problems reflecting lingering effects of the Viet Nam Warand then was hit with the 1973-74 oil price shocks. The administration adoptedfirst a set of general price controls and then initiated what proved an extendedconcentration on regulating energy for many reasons including allegedmacroeconomic benefits. The much explored energy experience (see Bradley'smassive 1995 effort to sum up the experience) illustrated the severemicroeconomic problems that can arise.

Many of those involved during the 1980s in developing more acceptabletheories of how rigidities cause economic instability are cautious about makingpolicy suggestions. The point is made in the editors' introduction to Mankiwand Romer's anthology of U.S. writings developing such models. One author inMankiw and Romer (Bryant, v. 2, p. 28) notes "almost anything can be modeledas optimizing behavior". The Winter 1993 issue of the Journal of EconomicPerspectives had a symposium on the work in which the developers Romer,Greenwald and Stiglitz, a leading old Keynesian-Tobin, and a new classicaleconomists-King all express reservations about the empirical relevance of thetheories. Robert J. Gordon (1990) (from a distinguished family of economists towhich I am not related) has separately discussed the drawbacks. All sides arguethat it is still not established what rigidities are most critical and how theyoperate. Thus, the latest work on imperfections at most is the basis for eventualdevelopment of policy advice.

When John F. Kennedy was elected, his official and unofficial economicadvisors adopted a compromise between Keynesianism and microintervention.Recognizing the drawbacks of explicit controls, the advisers proposed a halfwayhouse approach of wage-price guideposts to indicate to industry what would benoninflationary wage and price changes. Estimates were provided of the ratesof wage and price increases consistent with overall price stability. Industry andtrade unions were exhorted to keep wage and price changes within theseguideposts. This was a curious, justly largely forgotten episode since economicanalysis tends to scorn reliance on unenforceable pressures. Indeed, the onlyreason that this approach is mentioned is that Mankiw and Romer included apaper by Okun advocating among other things a return to guideposts (andproving a link, badly missing elsewhere in the anthology, to precursor work from1935 to 1970).

6. The Skeptics

The skeptical view stressed here is that high unemployment is transitory, ofill-understood origins, and cannot be alleviated by public policy. Skepticism, asdiscussed below, involves various challenges to belief that instability iscorrectable. Objections are both theoretical and practical. It is argued thattheories stressing instability have questionable theoretic basis and littleempirical relevance. Another contention is that instabilities are so difficult torecognize and so transitory and policy responses are so slow and variable thatintervention is likely to be harmful.

Skepticism, as treated here, is an extension of the Milton Friedman positionand its development by the macroeconomists using, among other things, theassumption of rational expectations. That last assumption is compatible withmany different further premises; the new classics initially concentratedconcentrate on models in which policy ineffectiveness is likely. The maindifference of the present discussion is its greater stress on the absence of aconvincing explanation for the causes of unemployment.

The absence of such an explanation strengthens skepticism. Thus, thisarticle stresses the defects of business cycle theories with a clear message ofany type on both causes and policies. I read Friedman as saying only thatpassive monetary is the least worst available alternative. The wildly implausibleidea that all problems will vanish under a monetary rule has its supporters, butFriedman does not seem one of them. Similarly, I note with no enthusiasm effortsof real business cycle theorists to suggest business fluctuations are not a policyissue because they arise from voluntary changes in the willingness to work.

7. The Rise of Macroeconomics

Historically economics concentrated on attainment of efficiency in themarketplace. Disturbances such as unemployment, inflation, and problems withinternational balances of payments were considered secondary. Textbooks priorto the end of World War II hardly mentioned the issues. Writings existed, butthey apparently commanded limited attention. (The impression about earliertextbooks is based on sampling made over many years. Here, Samuelson'scomplaints (e.g. 1977, p. 770) that the issues were ignored when he was anundergraduate in the 1930s should be contrasted with the mass of material citedin Haberler's Prosperity and Depression.) The persistence of massiveunemployment in North America and Europe in the 1930s shattered theprevailing beliefs. Economists struggled to devise better explanations ofeconomic instability. The result of all this was the emergence ofmacroeconomics.

The new realm of macroeconomics long was dominated by models inspiredby the work of John Maynard Keynes in the General Theory of Employment,Interest, and Money. Keynes presented an exposition on the causes ofinstability (or at least of a depression as prolonged as that in the 1930s) thatsecured wide acceptance as the needed answer. Given the ambiguity of Keynes'swork and the need to deal with circumstances radically different from those ofthe 1930s, enormous explanation, refinement, and revision has occurred.

Keynes-inspired economics has at least three major components-a frameworkfor analyzing economic instability, a vision about the nature of the instability,and views on stabilization prospects given the instability. Much controversyappropriately arises about what is the correct view of each area. A more unusualaspect of the debate is the preoccupation about whether a given interpretationis one advocated by Keynes. This contrasts with the excessively short memoryof many microeconomists who often write as if general equilibrium theory leapedfrom Walras to Arrow and Debreu without the intermediate assistance of evenHicks and Samuelson. For most purposes, the debate about interpreting Keynesis a distraction from the fundamental questions about what is the most usefulway to model the economy.

The difficulty of determining what Keynes meant arises from the loose, oftenambiguous exposition in the General Theory. Examining the mass of Keynes'sother writings apparently increases the uncertainty about his beliefs. Similarly,those who differ with some possibly quite substantial part of the Keynesianposition as defined here may or may not still identify themselves as a Keynesianin a looser sense.

The greatest consensus exists about the basic analytic framework. That usedgenerally is an extension of J.R. Hicks's pioneering 1937 effort to provide asystematic formulation of what Keynes meant. The accord, however, consistsonly of agreement to employ a few equations dealing with among other thingsthe supply and demand for four goods-production as a whole, labor, money, andan interest bearing security.

Considerable variation prevails in the specification of equations in differenttheoretic exercises. In many cases, these alternatives are presented as oneproviding a superior view of actual economic conditions. Others are simplyefforts to test the sensitivity of results to the assumptions. (Hicks's version ofKeynes was one of the earliest to argue that the General Theory did not assumeimperfect competition. Hicks concentrates on differences between Keynes andthe classics about the sensitivity to interest rates and the level of income ofdemands for money and investment goods and the supply of savings. The laborand good production market are absent from Hicks's models but appear inModigliani's widely cited 1944 effort to summarize the debate.)

As discussed more fully below, many macroeconomics models of persistentunemployment long have assumed some departure from the (idealized)assumptions of the microeconomic theory of competitive markets. Someobservers, notably Patinkin (1965), correctly argue that this approach trivializesand misrepresents Keynes and thus that an alternative view of Keynes wasneeded. (Patinkin makes this case tacitly.) The implications of major departuresfrom the core assumptions of the theory of competitive markets are too obviousto justify terming the consideration of imperfections a revolution. The attacks onclassical economics in General Theory are deemed to suggest less obviousflaws in the old analysis than neglect of market imperfections.

Patinkin focused on an issue that then and now traditional theory cannottreat, the speed of adjustment. He suggested that this adjustment might be tooslow, where too slow means slow enough that governmental intervention couldeffectively speed the process. This viewpoint, however, has received fewadherents. Economic rigidity models of various forms discussed below remainthe mainstay of theories of persistent unemployment. Keynes's exposition is toofuzzy to resolve definitely whether he postulated a more subtle problem,overstated the newness of his theory, or both. Validating Patinkin's view ofKeynes or the similar contentions of Clower (1965) and Leijonhufvud (1968) isnot feasible or essential. The observation that the implications of marketimperfections is (better) treated by classical economics remains valid whateverKeynes believed.

Haberler's study of economic instability largely written before the appearanceof the General Theory anticipates the concern over stress on marketimperfections:

It might therefore just as well be maintained that rigidity of our economicsystem, or its financial or monetary organisation, or particular features of thelatter, are the causes of the cycle as that inventions or crop changes or changesin demand are responsible (p. 6).

By itself, this debate over what comprises the most relevant macroeconomicmodels and how consistent that model is with that of Keynes could be and oftenis a mere intellectual exercise. The germane concern is about the practicalimplications of whatever is the right model. In particular, the central issue iswhether the characteristics of actual macroeconomics justify and permit effectivecountercyclical policies.

8. Alternative Views on the Stability Problem

While many different models exist, the number of possible outcomes is limited.The basic possibilities are stagnation (unemployment that inspires no marketgenerated corrections), a business cycle sufficiently prolonged that publicpolicy can lessen its impacts, and a business cycle so transitory that publicpolicy cannot lessen the impacts. Feasibility depends upon the comparativelength of the cycle and of policy response.

The stagnation idea was one that arose in the depression of the thirties anddied with the boom after World War II. Whatever Keynes believed, some ofthose whom Keynes inspired initially argued that unemployment was chronic.The original Keynesian approach, therefore, stressed the existence of inherentlycorrectable persistent demand deficiencies in the economy. Fears of massivepersistent unemployment faded and with them interest in equally massive andslow moving programs of government spending.

The alternative view that temporary periods of unemployment, the problemoften called the business cycle, arise now predominates. The "Keynesian"position is that public policy can fruitfully counteract these transitorydownturns. The "classical" approach stressed here is essentially an attack onthe legitimacy of this extension to short recessions of claims plausible topersistent depressions. (In fact, such concerns have been raised within theKeynesian traditions, see Phillips (1950, 1954, 1956) and Allen (1956).)

Inflation proved a more persistent problem than many Keynesians (but notKeynes himself) expected. Reflecting these changes, Keynesian economicsgreatly altered its focus from 1936 to its "crisis." in the 1970s. The revised beliefwas that the fluctuations that did occur had characteristics that should andcould be counteracted by active government policies. The initial successes inthe 1960s produced faith (indeed in the worst cases arrogance) that this viewhad been vindicated.

9. The Return of Classical Macroeconomics

During the years immediately after World War II, only a few economists, mostnotably Milton Friedman, argued that the problems and the ability to deal withthem had been overstated. To be sure, a substantial number of older economists,of whom Haberler is an example, who were unimpressed by Keynes's analysiswhen it appeared and never recanted. Paul Samuelson's fascination (e.g. 1977,p. 878) with Kuhn's concept of scientific revolutions undoubtedly involvedappreciation of Kuhn's concept of a recalcitrant old guard who refused torecognize the truth. Long before Kuhn provided supporting rationales, theKeynesians were quite successful at dismissing older critics as obdurate. Therevival of more classical approaches implies the dismissals may have beenpremature. The old school lived to see their earlier doubts gain support thatattracted new disciples. Snowdon, Vane, and Wynarczk (1994, p. 21) suggestthat indeed Kuhn's analysis is generally inappropriate for economics because nooutlooks are abandoned.

Stagflation (high unemployment, inflation, and slow growth) in the 1970scaused greater attention to Friedman's views. A group of economists termed thenew classical economists developed models that raised the possibility thatFriedman's criticisms understated the drawbacks of intervention.

Friedman and similar critics stress the existence of long and unknown lagsin the recognition of problems, design of response, and reaction to the response.Given all these lags, active public policy cannot clearly produce improvementand might even be harmful. This position has been enormously influential. Manyfully accept the arguments. More importantly, those who defend activestabilization policy advocate actions far less ambitious than those proposed inthe 1960s.

After the disastrous performance in the 1970s, Friedman's argument thatinstability was not necessarily inherent and definitely was not correctablereceived increased attention. This encouraged economists too young to beinfluenced by the experiences of the thirties to examine what lay behind theearlier beliefs. In particular, economists such as Lucas and Sargent developedan alternative, more formal analysis to deal with Friedman's assertions. (At leastthree major anthologies are available on new classical economics: Lucas andSargent (1981), a large collection of pieces by many writers who deal witheconomics principles, statistical techniques, and empirical evidence; Lucas, acollection of work authored or co-authored by Lucas; and Miller (1994) acollection stressing later work, all associated with the Federal Reserve Bank ofMinneapolis.) The new analysis showed that under some conditionsintervention was totally ineffective.

10. Market Clearing Rational Expectations Models

Since this result depended on introducing expectations into macroeconomicmodels and the developers relied specifically on a theory of expectationsdeveloped by Richard Muth, one name given the application to macroeconomicswas based on the term, rational expectations, Muth used to describe hisapproach.

Since the theory also insisted that it should be assumed that markets clearrapidly, the alternative name of market clearing macroeconomics was also used.The meaning of market clearing used by the group seems widely misunderstood.Lucas's formulation seems to indicate nothing more than an analytic strategy(1981, p. 226). He felt that assuming perpetual disequilibrium was too permissive.However, he adds that equilibrium states are not always desirable. One way ofinterpreting him is by recalling Alfred Marshall's period analysis. In Marshall,markets always clear but not all the ultimately desirable responses are madeimmediately. Reality consists of a successions of ever better adjustments tounderlying conditions. Nevertheless, the new classical economists tend to seea rapid full adjustment.

Finally, the approach was dubbed new classical since it in many wayssuggested that classical competitive theory was still the most practically relevantmodel. One of the critical ways was to assert that prior models lacked"microfoundations." By this was meant that no plausible microeconomic theoriesexisted to justify many parts of the prior interventionist macroeconomic models.This applied at least to the assumptions about expectations, the specification ofequations in econometric models of the economy, and the assertions of pricerigidities.

In the most dramatic applications, the approach showed cases in whichactive policy would be ineffective. (Lucas's work provided a starting point, butthe fullest developments were in papers in Lucas and Sargent, notably Sargentand Wallace.) These new classical economists contended that Keynesianarguments, particularly those associated with the Lipsey (1960) andSamuelson-Solow (1960) developments of Phillips's 1958 analysis of an inflationunemployment trade-off, depended on assuming that the public wasirredeemably stupid. The new classical economists argued that, if the publicanticipated that monetary expansion would occur, they would neutralize itsoutput effects by inflating prices by the amount of the monetary expansion. Theresult is an immediate consequence of the key assumptions of competitivemarkets that adjust slowly but not interminably and of decision makers capableof anticipating public policy. The essence of the case is that the models assumeenough flexibility in the economy to rule out any output of monetary expansionexcept that resulting from failure to anticipate.

Another key contention was that statistical analysis could not establishwhich theory was correct. The new classical argument was extended to remindmacroeconomists of the classic econometric problem of observationalequivalence (see, e.g., Sargent, 1976). In many cases including those of concernhere, observable behavior can be consistent with many radically differenttheories. (The basic idea is that we only observe the interaction of supply anddemand. An approach called identification is used to move backward from theobserved results to determining the equations that produced the equilibrium.However, models tend to be overidentified, i.e., consistent with an infinitenumber of conditions.)

The new classicists also adopt Friedman's concerns about the enormous dataneeds for successful stabilization. Much effort has been devoted to showing theimperfection of the models and how to develop models in which activestabilization works. This seems an inadequate response. The new classicists arechallenging the empirical validity of faith in active instability. A response mustinclude convincing factual evidence to the contrary.

While the new classical economists described their work as supportingFriedman's arguments, he was not pleased. He joined many Keynesians withconcerns over the allegedly high degree of rationality assumed (see hiscomments to Snowdon, Vane, and Wynarczk, 1994, p. 175). He felt that hisarguments were more plausible than the assumptions of rationality made by thenew classical economists. These complaints may exaggerate the differences.Friedman had suggested the idea that real effects of monetary policy werepossible only if people were temporarily fooled.

The new classical economists made the sensible extension of specifyingwhat was required to eliminate surprise. The analysis leaves some room fortemporary surprise. More critically, however, the approach at least implicitlywarns that all theories of the business cycle, particularly the monetary, rely tovarious degrees on the assumption of transitory, potentially ultimatelycorrectable confusion over what the monetary authorities are doing. Lucashimself relies on such a lag. The minimum contribution of a rational expectationsbenchmark is warning that the magnitude and duration of effects depends uponprecisely what economic analysis cannot treat, the precise degree to whichreality differs from idealized conditions.

A further implication is that, as is surely true in 1997, after decades of activemonetary policy, the private sector will better understand and anticipate theactions of the monetary authorities. In short, monetary policy can createdisturbances, but their nature is too unclear and so subject to eventual erosionto permit counteraction. Attacks on rationality, moreover, have overstated whatone must know. What is involved is knowledge about what the monetaryauthorities are doing and what it implies. The leading newspapers and televisionnetworks have managed to grasp such information.

Houthakker's (1957) research on the operation of futures markets suggestedthat the few full-time professionals with knowledge of market conditionsdominated the markets and made them efficient. (This is an implicit applicationof Hayek's arguments about how markets economize on knowledge acquisition.)Similar specialists could operate in other markets to ensure they two were guidedby the best possible knowledge.

11. Further Challenges

Friedman in collaboration with Anna Schwartz (1963, 1982) added anotherelement to the anti-intervention case. They examined the monetary record in firstthe United States and then the United Kingdom and observed a high correlationbetween money supply movements and economic instability. They concludedthat monetary policy seemed the cause rather than merely reacting. A major partof the analysis was a detailed description of policies before and during the greatdepression. The review led them to attribute the depression to overly aggressivemonetary restriction in the late twenties and failure to expand money in the 1930sby the U.S. Federal Reserve Board. Two leading subsequent commentators,Kindleberger (1986) and Temin (1989), disputed Friedman and Schwartz'sexplanation of why the Federal Reserve acted improperly but agreed that badmonetary policy was the critical cause. A symposium in the Spring 1993 Journalof Economic Perspectives (C. Romer, Margo, Calomiris, and Temin) reinforcedthis viewpoint.

These conclusions dispute the central premise of Keynesian economicsstressing inherent, correctable instability in the economy. Clearly, as Friedmanand Schwartz know, strong associations simply suggest linkages. Ultimately, werely on our judgments about the plausibility of different explanations. IfFriedman and Schwartz are right, the Keynesians identified the disease (badpolicy) as the cure. This is an ultraclassical (almost Austrian) conclusion that inmacroeconomics as well as microeconomics government is better thought of asa threat than as an all knowing benevolent force. This, in turn, hits at a mainweakness of the neoclassical synthesis, the inconsistencies between its viewsof government in microeconomics as opposed to macroeconomics. It is possiblethat macroeconomic problems are so much more serious or more tractable thatgovernment can be trusted more in macroeconomics than in microeconomics. Italso is possible that the drawbacks that arise in microeconomic problems alsooccur in macroeconomics. These conclusions become muted but still relevant ifless classical explanations are more valid. We are left with concerns that, asalready noted, all responsible observers share, about the severe limits to activeintervention.

12. Real Business Cycles

Some new classical economists, as noted, responded to discontent with theexplanations given of business cycles proposed new real (in the economic senseof non-monetary) theories of the business cycle in which variations inemployment were voluntary responses to changing market conditions such assudden changes in the productivity of the economy. This is a very differentapproach that, if valid, would be the ultimate challenge to Keynesian economics.Countercyclical policy would be interfering with efficient behavior.

Real business cycle theory to date is severely (and properly) criticizedbecause it is as unconvincing as all the prior simple explanations. In particular,no plausible justifications were provided to justify belief in developments thatwould inspire massive but temporary voluntary decisions of some workers tostop working. (Even, the advocates recognize the difficulty of identifying realphenomena that explain historical experience. The summer 1989 issue of TheJournal of Economic Perspectives presents papers by Plosser-a developer andMankiw-a skeptic.)

Snowdon, Vane, and Wynarczk note the thread in real business cycleanalysis that emphasizes the cycles are random deviations from trends. Thediscussion fails to cite Slutsky's classic paper that deals with possibility that byits nature economic evolution will have random deviations from trends. Thisearlier work relies on the broad view that the complex processes involved cannotbe expected to operate smoothly. This seems more satisfactory than the viewthat the unevenness has a cause.

13. The State of Policy

New classical economics, Austrian economics, and Friedman all effectively arguethat Adam Smith's concerns about the difficulties in determining the public goodapplies as much to stabilizing the whole economy as to regulating individualmarkets. "Keynesian" belief in the opposite presumption is challenged toprovide a convincing counterargument.

The arguments were essentially a formalization of a key tacit premise inmacroeconomics. It was an act of judgment, heavily influenced by the greatdepression, that stabilization policy was needed. Keynesian economics and thebroader neoclassical synthesis in which it was embedded effectively argued thatthe dangers of unemployment were so severe and readily correctable that it wasobvious that economists should suspend their traditional distrust ofintervention. The continuation of this approach after 1945 involved a giant leapthat was never well justified. What was true for a chronic depression was clearlyless justified for a world of milder business cycles.

As long as the policy seemed to work, it escaped challenge. Leaps of faithseemed good enough. Once the program faltered, economists felt emboldenedto restate the traditional doubts about intervention. These new classicistschallenged the prevailing evaluation criteria by suggesting many reasons whythis approach was suspect.

Given the strong commitment among macroeconomists to active stabilizationpolicy, these views have been resisted vigorously. Stress is on complaints thatthe models assume a implausibly high degree of rationality, postulateunrealistically well functioning markets, and inadequately explain the businesscycle. A striking contrast between the new classical and new Keynesiananalyses is that the former involved extensive consideration of the theory andpractice of statistical testing of hypotheses while the latter concentrates on puretheorizing. As suggested above, these complaints inadequately considerwhether the flaws suffice to justify active stabilization policy. As discussedbelow, to date, the most that the Keynesians can claim is that such justificationmight exist. Thus, the attacks on new classical economics seem incomplete.

The Keynesians might have explicitly countered with what they seem tobelieve, that the risks of inaction were too great to allow reliance on simple rules.Probably because they recognize that the new classical economics also warnsthat action also poses possibly greater dangers, Keynesians have tried to refutethe criticisms.

Unfortunately, the situation precludes a conclusive resolution. The situationis standard in economic policy debates. We have two radically differentjudgments of what constitutes the prudent strategy. One side is concerned withthe perils of inaction and urges continued effort. The other side concentrates onthe evils of intervention and urges radically altered less intrusive controls. Thisobviously cannot be settled here. What is critical is that the doubts are sufficientto justify restraint about adopting radical new measures.

14. Confronting Cyclicality

Keynesian and other widely used macroeconomic models have a flaw of dealingonly with equilibrium at a specific moment. A long tradition has at least tacitlyrecognized that this approach ignores the expansion of the economy and thevariations in expansions traditionally termed business cycles. Early Keynesiansexpressed their recognition by developing models explicitly treating cycles.More classical economists directly attacked neglect of cyclicality as a fatal flawof Keynesian economics.

In its most general and useful sense, the business cycle idea suggests thatincreased unemployment occurs in an ever changing economy and thateconomy may spontaneously generate measures to eliminate the unemployment.To understand the debate, we must recall now abandoned work in economics.

Staring in the 1930s, various economists, most notably the very young PaulSamuelson, tried to devise macroeconomic models that accounted for thesecyclical forces. Samuelson and those who followed him concentrated on theexisting theory most readily reduced to a manageable model. In particular, hetreated an accelerator model in which investment depended on the most recentchange in national output. (By modern standards, this is a hopeless naive modelsince it ignore anticipations.) This effort reached a dead-end in the 1950s (seeAllen, 1956). The models had become unmanageable long before they produceda plausible analysis of cyclicality. Macroeconomists abandoned, not only thedevelopment, but the recollection of cyclical analysis. In the process ofabandoning this approach, a critical insight was weakened. Studies, particularlythose of Phillips (1950, 1954, 1956, 1957) , of stabilization policies in a cyclicaleconomy showed the difficulty of producing desirable results in a steadilymoving economy. A basic problem of any interventionist strategy is that longeffort has failed to produce a convincing analysis of cyclicality. It can be arguedthat the analysts have lost sight of the difficulties involved.

Cycle theory was followed by steady-state growth theory that traced theeffects of steady increases in the ability to produce. Here too, the difficulties ofanalyzing complex cases long hindered progress; some signs of revival havearisen but is not treated here. However, the pioneering models in this realm arewidely cited in later macroeconomic models.

Shortly after the appearance of the General Theory, the League of Nationspublished a survey of business cycle economics. The review was by GottfriedHaberler, an Austrian trained economist long associated (particularly in his longcareer after the book was published) with anti-intervention views on theeconomy. (Haberler revised the book twice for the League and finally in 1957 theHarvard University Press issued a fourth edition. The last was used here.According to that edition, every edition had two main parts, the survey ofdifferent theories and a "Synthetic Exposition Relating to the Nature and Causesof Business Cycles." The fourth edition omitted much material from earliereditions, retained a chapter on Keynes added to the second edition, and addedreprinted of later articles.) The discussion provides a richer view of proposedexplanations and their drawbacks than appears in subsequent work.

Haberler begins by discussing the problems of explanation. He suggests, forexample, that some supposedly single-cause models at least tacitly postulateadditional influences. Supply or demand shock based models makepresumptions about the adaptability of the economy and particularly themonetary system to shocks. Monetary theories may implicitly concern theresponse of the monetary authorities to external forces. His basic posture is thatthe theories are at best incomplete and often indefensible. Subsequent work hasneither added plausible new explanations nor overcome the severe deficienciesof prior concepts.

15. The Causes Considered

Haberler's survey of views distinguishes several categories of causes - purelymonetary theory, (nonmonetary, monetary, and demand change)over-investment theories, changes in costs, horizontal maladjustments,overindebtedness, underconsumption, psychology, and harvests. However,over half his discussion of explanations is devoted to over-investment.

Haberler's monetary theory is devoted entirely to the views of R. W.Hawtrey. The latter's theory holds that business cycles are due entirely tounwise fluctuations in the money supply. Emphasis is on the effects on realinterest rates. (This is a stronger and thus less convincing argument againstactive monetary policy that the later view of Milton Friedman that stable growthhas better effects than active monetary management. Hawtrey as interpreted byHaberler argues that active monetary policy is cause rather than cure oraggravating force.) As proves true of many of the models discussed, this theoryinvolves precisely the dependence on failures of anticipation about which therational expectations approach warns.

The alternatives to a purely monetary approach tend to involve a genericproblem of imbalances among supplies, demands, and prices. Haberler'sdiscussion of specific theories thus often expands upon his initial warning thatwhatever problem a given theorist stresses, the analysis typically involves theinteraction of several forces.

Monetary over-investment theories such as Wicksell's (and Hayek asepitomized by Snowdon, Vane, and Wynarczk) differ from Hawtrey's "pure"monetary theory by stressing the influence on investment of changes in realinterest rates. Haberler starts his review of nonmonetary over-investmenttheories by noting that the lack of explicit consideration of money is a flaw. Thetheorists "are compelled to assume an elastic currency or credit supply in orderto prove what they wish to demonstrate" (p. 73).

Haberler then turns to a central ideal in business cycle theory, theacceleration principle, used in Samuelson's model (1966 v. 2, p. 1107-1124) andits extensions. This theory builds on the fact that investment either replacesretired capital goods or adds to productive capacity. A simple rule is thatinvestment to add new capacity is proportional to the change in output in therecent past. While the level of economic output changes slowly, the changesand thus investment can move radically. Haberler ends his discussion ofqualifications to the theory with consideration of expectations.. He notes "...anyinvestment, directly or indirectly is looking forward to, and is made in theexpectation of, a future demand for consumers goods" (p. 97). His concern ismore qualified than that of the 1970s. Haberler talks only of "types of investmentwhich look forward to their utilisation to a very distant future...." (p. 98).

The cost change theory that Haberler stresses is that as a boom proceeds,firms start operating on a steeply rising portion of their cost curves. Haberlerrecognizes that this would not be problematic unless real wages and prices failedto adjust to these changes. Horizontal maladjustments are lack ofcorrespondence between demand and supply in goods and factor markets.According to Haberler (p. 115), the overindebtedness theory is most satisfactoryin warning that existing debt hinders response to downturns and lesssatisfactory in suggesting that the rise in burdens restrain activity and producedownturns. The financial organization explanation is that valuing financial assetsin nominal money reduces price flexibility.

Haberler views overconsumption as a grab-bag of illformulated, often invalidtheories. The valid theories, moreover, are often restatements of theories bestclassified under his other categories. He indicates that many theories involveeconomically unsound ideas that consumption is inherently unable to expandas rapidly as productive capacity. Haberler indicates that valid theories involveshort-run imbalances among consumption, savings, investment, and productivecapacity. In some of these models, stress is on income distribution effects. Ashift of income from wages to profits undesirably reduces consumption.

Haberler views psychological theories as attempting to postulate swingsarising from optimism and pessimism distinct from the fluctuations observed inother theories. Haberler notes the difficulties of distinguishing these effects butdoes not deny their existence.

As suggested above, Haberler's leaves us with many ideas and no clear wayto determine whether any of those that are theoretically sound are empiricallyrelevant. In short, he documents the case that a satisfactory explanation isunavailable. It is argued above that little progress has been made in the last sixdecades at overcoming these problems.

16. Imperfect Competition and Unemployment: The MicroeconomicBackground

One critical controversy in macroeconomics centers on whether the traditionalflexible competitive market outlook or alternative imperfect market or defectivegovernments models best describe actual economies. The macroeconomicdebates are echoes of the microeconomic debate over what is the empiricallymost valid model of the economy. Since the relevance of the associatedmacroeconomic models depends on the underlying microeconomic work, thenature of the latter must be considered.

The history of the microeconomics of imperfect markets is convoluted. Aparticular concern is that theory and practice may have radically diverged. Thecase for regulating markets arose in ill-formulated work in the thirties. GardinerMeans (1939, 1972) became the quintessence of the process due to hislong-standing, diverse and often discussed efforts. His most persistent idea wasthat industrial prices were excessively rigid. This view was, in typical Meansfashion, derived solely from observation of data. Reaction to arguments lackingeither theoretic basis or proper skepticism about the problems of datainterpretation clearly were a major inspiration for Edward S. Mason's efforts toencourage sounder work on the issues (as is clearly shown in the germanearticles in the anthology of his writings (1957) and in his later (1982) essayreflecting on economics at Harvard in the 1930s).

The results were development of both sounder theories and better empiricalanalyses. The theory has produced an overwhelming number of models. Bysheer count, those models that postulate market imperfections seem to dominate.Even the term is loaded in this direction. Imperfect, in practice, means a departurefrom the idealized assumptions used in general equilibrium models that are theprimary tools of modern economic theory. However, the essence of models isdeliberate simplification. Reality thus is more complex than the model. Theoristsoften fail to consider how these complications alter what is optimal. In additionto the attention to market failure, important work has been done on developingmodels suggesting the microeconomic drawbacks of regulation, governmentfailure.

An important aspect of the evolution was the emergence starting in the 1970s(and still continuing) of theories inspired by the formalism of post World WarII economic theories. This new work displays both the strengths andweaknesses of formalism. The inadequate microfoundations displayed by thoselike Means have been banished. However, theory only establishes what mighthappen. Empirical verification is needed to determine the practical relevance ofalternative visions. A tendency exists relentlessly to follow every idea withoutconcern about relevance. The emphases in the theorizing seem better explainedby the greater ease with which market imperfections can be analyzed.

The literature is too vast to treat adequately here. Among the more importantoverviews are the essays in Schmalensee and Willig, the treatises by Tirole,Krouse, and Spulber (who uniquely does reflect on the limits to applications),and the texts by Scherer and Ross (1990) and by Carlton and Perloff (1994). Thecritiques of Schmalensee and Willig by Peltzman (1991), Fisher (1991), andKlevorick (1991) all reflect the concern expressed here over inadequateconsideration of empirical relevance. This, in turn, reflects a broader discordabout the practical importance of monopoly that these critiques neglect.

Applied economics and even public policy has moved towards discontentwith regulation that is so widespread that its nature can only be sketched here.Both regulation of individual markets and the efforts to use broad "antitrust"policies to regulate competition are under severe challenge.

With antitrust, the "moderate" position is that long-standing proposalsmassively to restructure industry lack justification for action and that policiesover mergers, price fixing, and other trade practices must be administered withgreater concern for promoting competition. (The Bork (1978) and Posner (1976)books on antitrust well state all but the first of these points. The best indicatorof what has happened with restructuring is the lack of efforts since the late 1960sto promote the approach.) However, the Chicago position on flawed regulationhas regularly led to conclusions that satisfactory administration is intrinsicallyimpossible. Failure to apply this proposition to antitrust invited and ultimatelyreceived challenge (see McChesney and Shughard, 1995).

The views on more specialized policies ranges from condemnation (e.g., formost agricultural, transportation, and energy intervention) to desires for radicalreform (e.g., for environmental regulation). The challenges involve a mixture ofskepticism over both whether any of the market imperfections cause significantproblems and whether public policy can make things better. No one book can doall this justice, but Viscusci, Vernon, and Harrington covers much of the criticalground from a careful analytic view. The Cato Institute, a vigorous advocate oflimited government, prepares more sweeping but much less analytic surveys ofinterventions that seem unsound.

One polar extreme is the Chicago position that monopoly is rare andtransitory unless supported by government policy. Some economists (e.g.Scherer and Ross, p. 541 talking of Chicago being "fervent in advancingsimplistic theories of economic behavior" and Bresnahan in his comments onFisher (p 227) talks of "the too-hasty Chicago consensus") profess scorn for thisoutlook. However, this does not produce great concern over monopoly orenthusiasm for regulation. Scherer and Ross's attacks on Chicago, for example,preface sections absent from earlier versions extensively discussing previouslyneglected Chicago arguments.

The point here is that the ability of market regulation to accomplish anythingis under severe challenge. Thus, a case for regulating on macroeconomicgrounds must overcome three problems - (1) proving possible imperfectionsexist, (2) showing that they have important macroeconomic effects, and (3)devising workable policies to overcome the problem.

17. The Effort at Application

The case here is that work to date falls far short of fulfilling the requirements justset. It is suggested that all that has been done yet is produce theories that showhow imperfections might produce major macroeconomic effects. Empiricalverification has not been provided. Moreover, considerable evidence exists thateven the developers of the theory are concerned about this limitation.

Stress here is on a group of U. S. economists who have termed themselvesnew Keynesians and Europeans who have emphasized labor market imperfectionissues. New Keynesian analysis is the effort of a newer generation ofeconomists to evaluate the attacks on the Keynesian faith of the leadingacademics (notably Samuelson in his textbook and Tobin in his extensivewritings on macroeconomics) of the immediate post-World War II years. U.S.new Keynesians come from leading universities, particularly those in theforefront of advancing the Keynesian revolution, and publish in the major longestablished journals.

The focus is governed by the high accessibility of the work, particularly inthe Mankiw and Romer anthology noted above. The term new Keynesian mightsuggest that active macroeconomic is feasible. However, even the advocates ofthe concepts are more cautious. They admit that the work only suggestsarguments that might lead to revival of belief in active monetary and fiscalpolicy. Mankiw and Romer (in their introduction) add that some of thedevelopers of the models reach the new classical conclusion that it is infeasiblein practice to implement such policies.

Whatever their influence on microeconomics, the models of imperfectmarkets were seized upon (in many cases by their developers such as Carlton,Stiglitz, and Shapiro, all also contributors to Schmalensee and Willig) in the1980s as the basis of more sophisticated models of the macroeconomicsimplications of violations of core assumptions. In the process, newer forms ofdisfunction were discovered (see below).

This methodology can be contrasted with older postKeynesian movement.As one proponent of the group (Sawyer, 1991, p. 202) notes, the term postKeynesian evolved from a umbrella for all reviews of Keynes to a description ofa specific approach to macroeconomics that includes concern over the impactsof market imperfections. The postKeynesians, however, seem to revel inunorthodoxy and separation from the rest of the macroeconomics profession(see Snowdon, Vane, and Wynarczk, 1997, p. 367-380).

18. New Keynesian Models

New Keynesian economics is a collection of more traditional analyses of marketimperfections. Many, but probably not all, of the proponents sought to refutethe new classical approach. Stress was on models with sound microeconomicbases in which instability was inherent. Usually, these models providemicroeconomic justifications for assuming good or factor price rigidities that leadto persistent unemployment. However, the models include some withcharacteristics that can preclude the existence, uniqueness, or stability ofequilibrium. The details are not critical here. Bryant's observation strongly holds.With enough assumptions about market imperfections, theorists can developmodels that inevitably lead to persistent unemployment. Mankiw and Romerdivide the contributions into seven parts - costs of price change, the role ofcontracts, good market monopoly, "coordination failures", labor marketbehavior, imperfect credit markets, and cyclical behavior of the goods markets.

Costs of price changes, infelicitously termed menu costs by Mankiw(Chapter 1 in Mankiw Romer), are the expenses of making, reporting, andimplementing a price change. (Except as noted, all the ideas presented here comefrom Mankiw and Romer and the attacks on them come from the articles notedabove.) As pioneering work by Barro (1972) (a new classical economist) showed,when price changing is expensive, it is no longer optimal to respondinstantaneously to price changes. The price change should occur only when thetransaction costs are repaid by higher profits from operations. Between pricechanges, the price should be a (time-weighted) average of the prices that wouldclear the market under flexible pricing. A series of articles (e.g. Mankiw andAkerloff and Yellen - chapter 2 in Mankiw Romer) examined how pursuit of suchpolicies might increase macroeconomic instability by inadequately respondingto changing market conditions.

As is standard in this literature, doubts have been raised about thetheoretical and empirical plausibility of the models (see the articles noted aboveand Snowdon, Vane, and Wynarczk). Even the menu metaphor is viewedskeptically. It is pointed out that announcing price changes often is cheap.Cheap changes such as scribbled entries or mimeographed inserts were possiblelong ago - for example, when I worked in my father's restaurant in the 1950s;modern technology allows cheap production of formal looking updates. The newKeynesian literature often cites an unpublished study that recognized that thelarge U. S. mail order retailers issue new catalogs far more often than theychange prices. It is suggested that better explanations must be provided aboutwhy frequent price changes are undesirable.

Another concern is that reliance on long-term contracts can be destabilizing.Those presenting such models clearly recognize that the problem with suchcontracts is that they inevitably involve incorrect specification of futureeconomic conditions. The first work in this realm took the existence of contractsas given, and only later was a transaction cost justification developed. Thetheories stress the difficulties of changing contracts. Stanley Fischer's 1977article (chapter 7 in Mankiw Romer and also in Lucas and Sargent) on the subjectexamines the role of adjustment clauses in contracts but only notes that the"right" clause would be more complex than those actually chosen. This seemstoo pessimistic. Actual contracts have explicit provisions to increase responseto changing conditions. Bad contracts are renegotiated. (This criticism emergedfrom my experiences with energy contracts rather than from the literature.)

Another type of model simply considers the macroeconomic consequencesof oligopoly. Most of these models examine how with monopoly labor supply,macroeconomic policy can offset the supply restriction and expand output.Examples are noted below.

Another, particularly complex family of models deals with what are termedcoordination failures. These involve situations in which the characteristics of theeconomy produce multiple equilibriums, the free market choice may not beoptimal, and public policy can move the economy to a more efficient equilibrium.An article (Cooper and John, chapter 16 in Mankiw and Romer) synthesizing thework suggests that the problem is best viewed of one of an economy operatingunder the principles of the noncooperative games so beloved by new industrialorganization theorists. At a minimum, this raises the problem, on which the gametheorists have lavished much attention, that noncooperative behavior is notgood for the decision makers and should be (but not necessarily is) abandonedwhen interactions regularly reoccur. Those skeptical of game theory would arguethat its microeconomic limitations make it a poor basis for macroeconomics.

Another area is alternative labor market models that postulate conditionsunder which wages should be less flexible than traditional theory seems tosuggests or reexamine the nature of unemployment. American new (and old)Keynesians focus on such ideas as efficiency wages that postulate that effortis increased by paying higher wages. A European alternative due to Lindbeckand Snower (1988) is of a conflict between insiders (the already employed) andoutsiders (those not employed). Without unionization the insiders have theadvantages of training, experience, and ability to "harass" newcomers.Unionization furthers insiders' advantages. Another European analysis byLayard, Nichells, and Jackel (1991) relies on less traditional concepts such asmark-up prices. As discussed below, these European analyses have a muchheavier policy content than models in Mankiw and Romer.

Attention also is given to why capital markets may be imperfect. Anotherform of imperfection involves the alleged cyclicality examined by Haberler in therelationship between price and marginal cost.

Two eminent older Keynesians, Hahn and Solow (1995) produced another,much less successful effort to revitalize the case for intervention. They only usetheoretical models (which are less transparent than those in Mankiw and Romer).Thus, Hahn and Solow too are subject to the prior criticism of inadequatetesting. They conclude, "But whatever the answer, imperfect information is notof itself an argument for inaction (p. 152)." This seems at best an ex cathedraargument for supporting active policy. It does not even address the question ofhow much activity is feasible. Their case, moreover, is not helped by the strategyemployed. The bulk of their book develops a series of difficult-to-comprehendmodels in which instability arises and can be cured by the right policy. Thesemodels have the drawbacks of lack of clear difference in substance betweenthose in Mankiw and Romer and considerably lesser clarity.

Snowdon, Vane, and Wynarczyk's survey of the work notes the work "hasbeen, until recently heavily biased towards theoretical developments (p. 328),""yielded numerous elegant theories which are often unrelated"(p. 328), and hasnot produced uniform policy conclusions (p. 326). On the last, Snowdon, Vane,and Wynarczyk add that "most" new Keynesians see a need for governmentactions for "coarse-tuning - policies designed to offset or avoid serious macrolevel problems." (p. 326). (Coarse tuning was Lindbeck's term for limitedintervention, 1993, p. 154.) At least as defined by Mankiw and Romer, the newKeynesians are necessarily too diverse to form a clear approach. At least two ofthose included in Mankiw and Romer, Solow and Okun, are veteran traditionalKeynesians. At least two others, Hall and Taylor, are skeptical aboutintervention.

The new Keynesians have only demonstrated something that the newclassicists never attacked, the existence of more theoretically sound models ofmarket imperfections that cause economic instability. In the worst cases, thesuccess of active intervention depends upon knowing what model characterizesthe economy. At best, the theories have established that in principle rationalexpectations need not by themselves render policy ineffective. The nature andcorrectability of actual economic instability remain where they always have been,uncertain and controversial.

19. The Policy Failure Alternative

A curiosity here is that so little attention has been given the possibility that badregulation is a major cause of instability. Aspects of a bad policy model were (asecondary) part of the early classical approach to instability and at least one newtheory revives this approach. The traditional arguments noted how trade unioninsistence on rigid nominal wages was the major barrier to effectivemacroeconomic adjustment. Union success was blamed on the willingness ofgovernments to support union activity. While many expressions of this viewappeared, that from Mises (1966, p. 769-779) seems to be the only one still widelyavailable.

This is not a central point of Austrian macroeconomics. The epitomes bothby Mises (1966, p. 538-586) and by Snowdon, Vane, and Wynarczk suggest thatthe crux is a monetary theory involving uneven receipt of the increased moneysupply that produces incorrect perceptions and distortions in the behavior ofreal variables. Mises (p. 580) talks of "the futile attempts to explain the cyclicalfluctuations by a nonmonetary doctrine,..." Snowdon, Vane, and Wynarczk (p.363) view wage and rigidity arguments as elements added on to explain theseverity of the great depression. It is presumably the existence of such viewsthat caused Patinkin to criticize interpreting Keynes as postulating only marketimperfections.

Among the newer work, only the European models stressing labor marketdefects have a strong government failure component. The microeconomicdefects of regulation are clear. Skepticism about macroeconomic explanationimplies that one cannot and thus should not add macroeconomic justificationsto deregulation. However, glimmers of concern do arise. Snowdon, Vane, andWynarczyk (p. 327) note that the insider-outsider approach produced numeroussuggestions to increase labor market flexibility. These include easing up onexisting policies such as job security legislation, laws encouraging unionization,and unemployment insurance. However, new programs are proposed to retrain,increase labor mobility particularly by making the housing market function better(without discussing whether this means more aid or removal of governmentcontrols), and profit share. Lindbeck and Snower (p. 260-268), Lindbeck (p.150-169), and Layard, Nickell, and Jackman all examine some of thesealternatives.

Goff (1996) has surveyed the role of regulation in the U.S. economy andattempted to measure its impact. Such a process is hindered by lack of a clearmeasure of the extent of regulation. Goff, therefore, employed a standard methodfor dealing with the problem, statistically generating an artificial measure that isthe weighted function of observable indicators such as that perennial - the sizeof The Federal Register, the U.S. government's report on its new proposed andadopted regulations, the budgets of regulatory agencies, the ratio of lawyers toengineers and scientists in industry, litigation, and state governmentemployment. Given this index, Goff seeks to test whether changes in regulationaffected performance. His tests are avowedly simple. He examines theassociation between changes in the trend in key measures of macroeconomicactivity are correlated with increasing regulation. He is well aware that otherforces may be at work, but at the simple level at which he worked, he can onlyintroduce two alternative variables, oil prices and non-defense governmentspending. In any case, he concludes that regulation reduced growth in grossnational (sic) product by .9 percentage points.

20. Conclusions

The present discussion has argued that considerable debate prevails inmacroeconomics about the nature of economic instability and whethergovernments can act to counteract them. The policies advocated range fromones that heavily limit government flexibility to those that encourage extensiveaction including and possibly centering on regulation of individual industries.What seems to be the "moderate" position at the end of the twentieth centuryis that governments should cautiously alter monetary and fiscal policy inresponse to changing circumstances. Caution is the minimum responsenecessitated by new classical criticisms. Market regulation remains suspect byboth new classicists and the believers in the neoclassical synthesis. Thissuspicion arises from the presence of much evidence of the drawbacks ofregulation and the absence of evidence that regulation is stabilizing. Efforts oversix decades have only strengthened these doubts.

21. Background to the Article

The author is a specialist in applied microeconomics who examinesmacroeconomics mainly to determine its relevance to efforts to justify control ofindividual markets on macroeconomic grounds. My research activity is devotedentirely to study of markets and their regulation. This article is heavilyinfluenced by the universal occurrence of severe government failure ineverything that I have studied. My teaching included long teaching a graduatetheory course. All this has required persistent examination of the literature inmicroeconomic theory. I have reviewed macroeconomics to evaluate proposalsto regulate energy markets in the interest of macroeconomic stability. Priorwritings that caused the invitation to write this article were inspired byrecognition that energy economists were accepting macroeconomic argumentsfor intervention that were not well supported in the macroeconomic literature.

I am Professor Emeritus of Mineral Economics, College of Earth and MineralSciences, The Pennsylvania State University, University Park Pa, 16802. SinceI was born in 1934, I am too young to have first hand knowledge of the thirties,but old enough to be aware of prior work inadequately considered by the newKeynesians and their inspirations among the new industrial organizationtheorists.

My examination of literature, in any case, is selective. This article extendsresearch undertaken for my 1994 book, Regulation and Economic Analysis. Mystarting points were the then current editions of several undergraduate texts inmacroeconomics, the very advanced survey by Blanchard and Fischer, and fouranthologies noted in the text of major articles on macroeconomic. Thissuggested much additional reading including return to some old favoritesremaining in my personal library. This reading has unearthed far more referencesthan can readily be mastered. A referee alerted me to a particularly valuablesurvey of the work (Snowdon, Vane, and Wynarczyk) and European work onimperfect labor markets.

22. Bibliographic Note

Thousands of books and articles on these issues have appeared. The booksinclude many comprehensive textbooks and a few more specialized surveys.Despite their mass, no single book fully covers all the issues or adequately citesthe literature.

The articles appear in the many journals directed at professional economists,the various periodicals produced by banks, particularly the regional FederalReserve Banks in the United States, and symposium volumes originating fromamong others again the regional Federal Reserve Banks and research institutessuch as Brookings and the National Bureau of Economic Research. Numerousanthologies, some devoted to a topic and others to a specific author, haveappeared. Some major articles have multiple incarnations. Among theparticipants in the debates for whom anthologies of relevant writings exist areKeynes, Hicks, Samuelson, Tobin, Friedman, and Lucas.

Each item consulted adds references to more publications. The listings hereare deliberately limited. I have selected anthologies, texts, and treatises thatcollectively include or discuss a large part of macroeconomics from the 1930 tothe 1990s. Much of the material contained is not explicitly cited. Thus, whilesome classics are not listed, they often are contained in what is cited. Anotherlimitation is that no pretense was made exhaustively to examine the many textsdesigned for the second (in U.S. academic jargon, intermediate) undergraduatecourse in macroeconomic theory. My citations of texts consist of two long extantones that are frequently cited and one, recommended by a colleague, that Ifound a good alternative; the citations whenever possible are to the latestedition of which I was aware rather than to the edition actually examined.

In the process of preparing my 1994 book, I began compiling a masterbibliography of works cited in summary volumes on microeconomics andmacroeconomics consulted. Part of the effort was to overcome the irritatingfailure of many authors to provide an integrated bibliography. My bibliographybecame unmanageably massive as I extended its reach. The germane literaturein macroeconomics alone is too extensive to treat adequately and the literatureon modern industrial organization, while smaller, is still overwhelming.

The books cited here include anthologies that contain at a minimumimportant parts of the literature and at their best helpful guides to the rest of theliterature. Miller alone of these books has a unified bibliography. Mankiw andRomer have an extensive bibliography (broken into separate parts by topicareas), but it falls far short of presenting everything cited in the book. Amongsurveys, Snowdon, Vane, and Wynarczk uniquely provides a unified andextensive bibliography. Blanchard and Fischer prepare separate bibliographiesfor each chapter. Even when a bibliography exists, it does not necessarilyprovide all the critical information. In particular, often the reprinting inanthologies is not noted. Generally, I have provided both the original citationand at least one of the reprints of which I was aware for articles. However, I havenot attempted to examine every writer's collected works. The main exception isSamuelson. So much of what is cited comes from inaccessible original sourcessuch as symposium volumes that his collected works are the best source.

Bibliography on Law and Macroeconomics (0620)

Akerlof, George A. and Yellen, Janet L. (1985), "A Near-Rational Model of the BusinessCycle with Wage and Price Inertia", 100 Quarterly Journal of Economics, supplement,176-213. Reprinted in Mankiw and Romer (1991), v. 1, 43-58.

Allen, R. G. D. (1956), Mathematical Economics, London, Macmillan and Co. and NewYork, St. Martins Press

Bain, Joe S. (1956), Barriers to New Competition: Their Character and Consequences inManufacturing Industries, Cambridge (MA), Harvard University Press.

Bain, Joe S. (1968), Industrial Organization, 2d ed., New York, John S. Wiley & Sons.

Barro, Robert J. (1972), "A Theory of Monopolistic Price Adjustment", 39 Review ofEconomic Studies, (January), 17-26.

Blanchard, Olivier Jean, and Fischer, Stanley (1989), Lectures on Macroeconomics,Cambridge (MA), The MIT Press.

Bork, Robert H. (1978), The Antitrust Paradox, a Policy at War with Itself, New York, BasicBooks.

Bradley, Robert L., Jr. (1995), Oil, Gas & Government.: The U. S. Experience, Lanham,Rowman & Littlefield.

Bryant, John B. (1983), "A Simple Rational-Expectations Keynes-Type Model", 98Quarterly Journal of Economics, (August), 525-528. Reprinted in Mankiw and Romer(1991), v. 2, 25-29.

Burstein, M. L. (1963), Money, Cambridge (MA), Schenkman Publishing.

Calomiris, Charles. W. (1993), "Financial Factors in the Great Depression", 7 Journal ofEconomic Perspectives, (Spring), 61-85.

Carlton, Dennis W. and Perloff, Jeffrey M. (1994), Modern Industrial Organization, 2nd ed,New York, Harper Collins.

Cato Institute (1997), Cato Handbook for Congress: 105th Congres, Washington (D.C.),The Cato Institute.

Clower, Robert W. (1965), "The Keynesian Counter Revolution: An Appraisal", in Hahn,Frank and Brechling, F. (eds.), The Theory of Interest Rates, London, Macmillan.Excerpted in Clower ed. (1969), 270-297.

Clower, Robert (ed.) (1969), Monetary Theory, Harmondsworth, Penguin Books.

Dornbusch, Rudiger and Fischer, Stanley (1993), Macroeconomics, 6th ed., New York,McGraw-Hill Book Company.

Fellner, William and Haley, Bernard F. (eds.) (1951), Readings in the Theory of IncomeDistribution, Philadelphia, Blakiston.

Fischer, Stanley (1977), "Long Term Contracts, Rational Expectations, and the OptimalMoney Supply", 85 Journal of Political Economy, (February), 191-206. Reprinted inLucas and Sargent (1981), 261-275 and in Mankiw and Romer (1991), v. 1, 216-231.

Fisher, Franklin M. (1991), "Organizing Industrial Organization: Reflections on theHandbook of Industrial Organization", Brookings Papers on Economic Activity:Microeconomics 1991, 201-240. (Comments, 226-238)

Friedman, Milton (1948), "A Monetary and Fiscal Framework for Economic Stability", 38American Economic Review, (June), 245-264. Reprinted in Friedman (1953), 117-156and in Lutz and Mints (1951), 369-393.

Friedman, Milton (1953), Essays in Positive Economics, Chicago, University of ChicagoPress.

Friedman, Milton (1956). "The Quantity Theory of Money: A Restatement ,"in MiltonFriedman (ed.), Studies in the Quantity Theory of Money, Chicago, University of ChicagoPress. Reprinted in Friedman (1969), 51-67 and Friedman (1987), 285-303.

Friedman, Milton (1968), "The Role of Monetary Policy", 58 American Economic Review,(March), 1-17. Reprinted in Friedman (1987), 387-403.

Friedman, Milton (1969), The Optimum Quantity of Money and Other Essays, Chicago,Aldine.

Friedman, Milton (1970), "A Theoretical Framework for Monetary Analysis", 78 Journalof Political Economy, (Mar./April). Reprinted in Gordon, Robert J. (ed.), MiltonFriedman's Monetary Framework, Chicago, University of Chicago Press.

Friedman, Milton (Leube, Kurt R. ed.) (1987), The Essence of Friedman, Stanford, HooverInstitution Stanford University.

Friedman, Milton and Schwartz, Anna J. (1963), A Monetary History of the United States,1867-1960, Princeton, Princeton University Press.

Friedman, Milton and Schwartz, Anna J. (1982), Monetary Trends in the United States andthe United Kingdom, Chicago, University of Chicago Press.

Fudenberg, Drew, and Tirole, Jean (1991), Game Theory, Cambridge (MA), The MIT Press.

Goff, Brian (1996), Regulation and Macroeconomic Performance, Boston, KluwerAcademic Publishers.

Gordon, Richard L. (1994), Regulation and Economic Analysis, A Critique over TwoCenturies, Boston and Dordrecht, Kluwer Academic Publishing.

Gordon, Robert Aaron, and Klein, L. R. (eds.) (1965), Readings in Business Cycles,Homewood, Richard D. Irwin.

Gordon, Robert J. (1990), "What is New-Keynesian Economics?", 28 Journal of EconomicLiterature, (September) 1125-1171.

Gordon, Robert J. (1993), Macroeconomics, 6th ed., Glenview, Scott-Foresman.

Greenwald, Bruce and Stiglitz, Joseph (1993), "New and Old Keynesians", 7 Journal ofEconomic Perspectives, (Winter), 23-44.

Haberler, Gottfried (ed.) (1944), Readings in Business Cycle Theory, Homewood, RichardD. Irwin.

Hahn, Frank and Solow, Robert M. (1995), A Critical Essay in Modern MacroeconomicTheory, Cambridge (MA), The MIT Press.

Haberler, Gottfried (1958), Prosperity and Depression: A Theoretical Analysis of CyclicalMovements, new and revised (fourth) edition, Cambridge (MA), Harvard UniversityPress. (First three editions 1937, 1939, and 1941 Geneva, League of Nations)

Hayek, Friedrich A. (1945). "The Uses of Knowledge in Society", 35 American EconomicReview, (September), 519-530. Reprinted in Hayek (1948), 77-91 and Hayek (1984),211-224.

Hayek, Friedrich A. (1948), Individualism and Economic Order, Chicago, Ill., Universityof Chicago Press.

Hayek, Friedrich A. (Nishiyama, Chiaki and Leube, Kurt R. eds.) (1984). The Essence ofHayek, Stanford, Hoover Institution Stanford University.

Hicks, John R. (1937), "Mr. Keynes and the Classics", 5 Econometrica, (April), 147-159.Reprinted in Fellner and Haley (1951), 461-476 and Hicks, J. R., (1982), Money Interestand Wages, Collected Essays on Economic Theory, v. II, Cambridge (MA). HarvardUniversity Press and Oxford, Basil Blackwell, 100-115.

Hicks, J. R. (1950), A Contribution to the Theory of the Trade Cycle, Oxford, OxfordUniversity Press.

Houthakker, H. S. (1957), "Can Speculators Forecast Prices?", 39 The Review of Economicsand Statistics, (May), 143-151.

Hume, David (1985), Essays: Moral, Political, and Literary, Indianapolis, Liberty Classics.

Kelman, Mark (1993), "Could Lawyers Stop Recessions? Speculations on Law andMacroeconomics", 45 Stanford Law Review, (May), 1215-1310.

Keynes, John Maynard (1936), The General Theory of Employment, Interest and Money,New York, Harcourt Brace and London, Macmillan and Company. (Available in theRoyal Economics Society collected works series.)

Kindleberger, Charles P. (1986), The World in Depression, 1929-1939, Berkeley, Universityof California Press.

King, Robert G. (1993), "Will the New Keynesian Macroeconomics Resurrect the IS-LMModel?", 7 Journal of Economic Perspectives, (Winter), 67-82.

Klevorick, Alvin K. (1991), "Directions and Trends in Industrial Economics: A ReviewEssay on the Handbook of Industrial Organization", in Brookings Papers on EconomicActivity: Microeconomics 1991, 241-280. (Comments , 265-279).

Kreps, David M. (1990), A Course in Microeconomic Theory, Princeton, PrincetonUniversity Press.

Krouse, Clement G. (1990), Theory of Industrial Competition, Cambridge (MA), BasilBlackwell.

Kuhn, Thomas (1970), The Structure of Scientific Revolutions, 2d ed., enlarged, Chicago,University of Chicago Press.

Layard, Richard, Nickell, Stephen, and Jackman, R. (1991), Unemployment:Macroeconomic Performance and the Labor Market, Oxford, Oxford University Press.

Leijonhufvud, Axel (1968), On Keynesian Economics and the Economics of Keynes, A Studyin Monetary Theory, New York, Oxford University Press.

Lindbeck, Assar (1993), Unemployment and Macroeconomics, Cambridge (MA), The MITPress

Lindbeck, Assar and Snower, Dennis (1988), The Insider-Outsider Theory of Employmentand Unemployment. Cambridge (MA), The MIT Press.

Lipsey, Richard G. (1960), "The Relationship between Unemployment and the Rate ofChange of Money Wages in the United Kingdom 1862-1957: A Further Analysis", 27Economica, (February), 1-31. Reprinted in Gordon and Klein (1965), 457-487.

Lucas, Robert, E., Jr. (1972). "Econometric Testing of the Natural Rate Hypothesis", inEckstein, Otto (ed.), The Econometrics of Price Determination Conference, Washington(DC), U.S. Federal Reserve System Board of Governors, 50-59. Reprinted in Lucas(1981), 90-103.

Lucas, Robert, E., Jr. (1972), "Expectations and the Neutrality of Money", 4:2 Journal ofEconomic Theory, (April), 103-124. Reprinted in Lucas (1981), 66-89.

Lucas, Robert E., Jr. (1973), "Some International Evidence on Output-InflationTradeoffs,"63 American Economic Review, (June), 326-334. Reprinted in Lucas (1981),131-145.

Lucas, Robert E., Jr. (1975), "An Equilibrium Model of the Business Cycle", 83 Journal ofPolitical Economy, (December), 1113-1144. Reprinted in Lucas (1981), 179-214.

Lucas, Robert E., Jr. (1976), "Econometric Policy Evaluation: A Critique", in Brunner, Karland Meltzer, Allan (eds.), The Phillips Curve and Labor Markets, Carnegie-RochesterConference Series on Public Policy, v. 1, 19-46. Reprinted in Lucas (1981), 104-130.

Lucas, Robert E., Jr. (1977), "Understanding Business Cycles", in Brunner, Karl and Meltzer,Allan (eds.), Stabilization of the Domestic and International Economy,Carnegie-Rochester Conference Series on Public Policy, 7-29. Reprinted in Lucas,(1981), 215-239.

Lucas, Robert E., Jr. (1981), Studies in Business Cycle Theory, Cambridge (MA), The MITPress.

Lucas, Robert E., Jr. (1981), "Tobin and Monetarism: A Review Article", 19 Journal ofEconomic Literature, (June), 558-567.

Lucas, Robert E., Jr. (1987), Models of Business Cycles, Oxford, Basil Blackwell.

Lucas, Robert E., Jr. and Sargent, Thomas J. (eds.) (1981), Rational Expectations andPractice, Minneapolis, Minn., University of Minnesota Press.

Lutz, Friedrich A. and Mints, Lloyd W. (eds.) (1951), Readings in Monetary Theory,Philadelphia, Blakiston (subsequently Homewood, Richard D. Irwin).

McChesney, Fred S. and Shughard, William F., II (eds.) (1995), The Causes andConsequences of Antitrust, Chicago, University of Chicago Press.

Mankiw, N. Gregory (1985), "Small Menu Costs and Large Business Cycles: AMacroeconomic Model of Monopoly", 100 Quarterly Journal of Economics, (May),529-537. Reprinted in Mankiw and Romer (1991), v. 1, 29-42.

Mankiw, N. Gregory (1989), "Real Business Cycles: A New Keynesian Perspective", 3Journal of Economic Perspectives, (Summer), 79-90.

Mankiw, N. Gregory (1990), "A Quick Refresher Course in Macroeconomics", 28 Journalof Economic Literature, (December), 1645-1660.

Mankiw, N. Gregory and Romer, David (eds.) (1991), New Keynesian Economics, 2 v.,Cambridge (MA), The MIT Press.

Margo, Robert A. (1993), "Employment and Unemployment in the 1930s", 7 Journal ofEconomic Perspectives, (Spring), 41-59.

Mas-Colell, Andreu, Whinston, Michael D., and Green Jerry R. (1995), MicroeconomicTheory, New York, Oxford University Press.

Mason, Edward S. (1957), Economic Concentration and the Monopoly Problem. Cambridge(MA), Harvard University Press.

Mason, Edward S. (1982), "The Harvard Department of Economics from the Beginning toWorld War II", 97 Quarterly Journal of Economics, (August), 383-433.

Means, Gardiner (1939), The Structure of the American Economy, Washington, NationalResources Council.

Means, Gardiner (1972), "The Administered-Price Thesis Reconfirmed", 62 AmericanEconomic Review, (June), 292-306.

Miller, J. Preston (ed.) (1994), The Rational Expectations Revolution: Readings from theFront Line, Cambridge (MA), The MIT Press.

Mises, Ludwig von (1966), Human Action: A Treatise on Economics. 3d rev. ed., Chicago,Henry Regnery Company. (Now published by Contemporary Books Inc., Chicago)

Modigliani, Franco (1944), "Liquidity Preference and the Theory of Interest and Money",12 Econometrica, (January), 45-88. Reprinted in Lutz and Mints (1951), 186-239.

Okun, Arthur (1975), "Inflation: Its Mechanics and Welfare Costs", Brookings Papers onEconomic Activity #2, 351-401. Reprinted in Mankiw and Romer (1991), v. 2, 327-375.

Patinkin, Don (1965), Money, Interest and Prices, 2d ed., New York, Harper and Row.(Available minus appendixes from MIT Press)

Peltzman, Sam (1991), "The Handbook of Industrial Organization: A Review Article", 99Journal of Political Economy, 201-217.

Phillips, A. W. (1950), "Mechanical Models in Economic Dynamics", 17 Economica N.S.,(#67 August), 283-305.

Phillips, A.W. (1954), "Stabilization Policy in a Closed Economy"n 64 Economic Journal,(June), 290-323.

Phillips, A. W. (1956), "Some Notes on the Estimation of the Time-Forms of Reactions inInterdependent Dynamic Systems", 23 Economica NS, (#90 May), 99-113.

Phillips, A. W., (1957), "Stabilization Policy and the Time Form of Lagged Responses", 47Economic Journal, (June), 265-277. Reprinted in Gordon and Klein (1965), 66-79.

Phillips, A. W. (1958), "The Relation Between Unemployment and the Rate of Change ofMoney Wage Rates in the United Kingdom, 1861-1957", 25 Economica NS, (Issue #100 November), 283-299.

Plosser, Charles L. (1989), "Understanding Real Business Cycles ", 3 Journal of EconomicPerspectives, (Summer), 51-77.

Posner, Richard A. (1976). Antitrust Law, An Economic Perspective, Chicago, Universityof Chicago Press.

Romer, Christina D. (1993), "The Nation in Depression", 7 Journal of EconomicPerspectives, (Spring), 19-39.

Romer, David (1993), "The New Keynesian Synthesis", 7 Journal of EconomicPerspectives, (Winter), 5-22.

Samuelson, Paul A. (1939), "Interactions between the Multiplier Analysis and the Principleof Acceleration", 21 Review of Economic Statistics, (May), 75-78. Reprinted in Haberler(1944), 261-269 and Samuelson (1966), v. 2, 1107-1110.

Samuelson, Paul A. (Stiglitz, Joseph E. ed.) (1966), The Collected Scientific Papers of PaulA. Samuelson, , 2 v., Cambridge (MA), The MIT Press.

Samuelson, Paul A. (Merton, Robert C. ed.) (1972), The Collected Scientific Papers of PaulA. Samuelson, , vol. 3, Cambridge (MA), The MIT Press.

Samuelson, Paul A. (Nagatani, Hiroaki and Crowley, Kate eds.) (1977), The CollectedScientific Papers of Paul A. Samuelson , vol. 4, Cambridge (MA), The MIT Press.

Samuelson, Paul A. (Crowley, Kate ed.) (1986), The Collected Scientific Papers of Paul A.Samuelson, , v. 5, Cambridge (MA), The MIT Press.

Samuelson, Paul A. and Solow, Robert M. (1960), "Analytic Aspects of AntiinflationPolicy", 50 American Economic Review papers and proceedings, (May), 177-194.Reprinted in Samuelson (1966), v. 2, 1336-1353.

Sargent, Thomas J. (1973), "Rational Expectations, the Real Rate of Interest and theNatural Rate of Unemployment", Brookings Papers on Economic Activity, 429-480.Reprinted in Lucas and Sargent (1981), 159-198.

Sargent, Thomas J. (1976), "A Classical Macroeconomic Model for the United States", 84Journal of Political Economy, 207-237. Reprinted in Lucas and Sargent (1981),521-551.

Sargent, Thomas J. (1976), "The Observational Equivalence of Natural and Unnatural RateTheories of Macroeconomics", 84 Journal of Political Economy, 631-640. Reprintedin Lucas and Sargent (1981), 553-562.

Sargent, Thomas and Wallace, Neil (1975), "'Rational Expectations', The OptimalMonetary Instrument and the Optimal Money Supply Rule", 83 Journal of PoliticalEconomy, (April), 241-254. Reprinted in Lucas and Sargent (1981), 215-228.

Sargent, Thomas and Wallace, Neil (1976), "Rational Expectations and the Theory ofEconomic Policy", 2 Journal of Monetary Economics. (previously Federal Reserve Bankof Minneapolis Studies in Monetary Economics 2) Reprinted in Lucas and Sargent(1981), 199-213.

Sawyer, Malcolm C. (1991), "Post-Keynesian Macroeconomics", in Greenaway, David,Blarney, Michael, and Stewart, Ian (eds.), Companion to Contemporary Economics,London, Routledge, 184-206.

Scherer, F. M., and Ross, David (1990), Industrial Market Structure and EconomicPerformance, 3d ed. Boston, Houghton-Mifflin Co.

Schmalensee, Richard, and Willig, Robert D. (eds.) (1989), Handbook of IndustrialOrganization, 2v., Amsterdam, North-Holland (Elsevier).

Slutsky, Eugen (1937), "The Summation of Random Causes as the Source of CyclicProcesses", 5 Econometrica, (April), 105-146.

Smithies, Arthur and Butters, J. Keith (eds.) (1955), Readings in Fiscal Policy, Homewood,Richard D. Irwin.

Snowdon, Brian, Vane, Howard, and Wynarczyk, Peter (1994), A Modern Guide toMacroeconomics: An Introduction to Competing Schools of Thought, Aldershot, EdwardElgar.

Spulber, Daniel F. (1989), Regulation and Markets. Cambridge (MA), The MIT Press.

Stadler, George W. (1994), "Real Business Cycles", 32 Journal of Economic Literature,(December), 1750-1783.

Stigler, George J. (1988), Memoirs of an Unregulated Economist, New York, Basic Books.

Stigler, George J. (ed.) (1988), Chicago Studies in Political Economy, Chicago, Universityof Chicago Press.

Temin, Peter (1989), Lessons from the Great Depression: The Lionel Robbins Lectures for1989, Cambridge (MA), The MIT Press.

Temin, Peter (1993), "Transmission of the Great Depression", 7 Journal of EconomicPerspectives, 87-102.

Tirole, Jean (1988), The Theory of Industrial Organization, Cambridge (MA), The MITPress.

Tobin, James (1993), "Price Flexibility and Output Stability: An Old Keynesian View", 7Journal of Economic Perspectives, 45-65.

Viscusi, W. Kip, Vernon, John M., and Harrington, Joseph E., Jr. (1995), Economics ofRegulation and Antitrust, Cambridge (MA), The MIT Press.

Woglom, Geoffrey (1988), Modern Macroeconomics, Glenview, Scott, Foresman.