This is a version of a paper issued in February 1993 as a
          University of Bradford. Department of Social and Economic
          Studies. Research Paper.


          J.M.BUCHANAN'S 'DEFENSE OF ORGANIZED CRIME': FORM,CONTENT AND
          INTENT
          1.INTRODUCTION.

                The literature of economics has not traditionally
          contained in depth analyses of the style of presentation of
          contemporary journal articles. The convention has been to
          present an original paper which opens with a brief nod of
          recognition to precursors. This form of exegesis reflects
          economists' desire to be seen as scientific. In the scientific
          milieu a new paper is worth publishing if it constitutes an
          advance on an old paper. The content of older papers is regarded
          as correct or incorrect on grounds of valid application of
          mathematical and/or statistical methods. Mulling over the style
          and form of economic writing has traditionally been restricted to
          historians of economic thought. Even here the surest means of
          publication has been evaluation of historical work in terms of
          its contribution to a scientific flow towards greater
          correctness. This is now changing, to some extent,  due to the
          influence of Donald McCloskey (1985) as evidenced by the papers
          collected in Samuels (1990).
                McCloskey's pioneering book contains a content analysis of
          Muth's presentation of his original article on rational
          expectations. The aim is to discover why economists found this
          persuasive. McCloskey's position is that, far from being
          scientific extracters of truth, economists are engaged in the act
          of persuading readers to their point of view. Persuasion is

          achieved through such strategies as appeals to authority,
          mathematical formulation and proof, statistical techniques and
          rhetorical use of language.  Further, mathematical and
          statistical techniques are often used in ways contrary to their
          methodological underpinnings, for example hypothesis test results
          are often deployed in a manner unacceptable in classical
          statistics. This tension arises because economists wish to use
          them to persuade rather than, according to received wisdom, as
          divining rods in the pursuit of truth.

                In this paper I provide a scrutiny of the form,content and
          intent of James M. Buchanan's contribution to the economics of
          organized crime. This is of interest for a number of reasons. It
          allows us to see how a Nobel laureate engages in the act of
          persuasion. It affords the opportunity to examine the congruity
          between the author's avowed methodological position and his
          analysis of specific policy problems.  In recent times, 'empire
          building' by economists has been robustly defended as
          unproblematic. The subject of organized crime brings to light the
          potential for contradictions when economists move their modus
          operandi into areas of social and political controversy. James
          Buchanan has gained distinction from winning a Nobel Prize for
          contributions to economics which emphasize the relationship
          between politics and economics [Sandmo(1990)]. He is also noted
          for going against the mainstream of economics, to a degree, in
          applying contractarian views to public finance [Congleton(1988)].
          He makes many claims for his oeuvre which distance him from the

          content and methodology of mainstream economics. The following
          claims are of the greatest relevance for this paper:

          1. A subjectivist approach is taken to preferences and costs in
          contrast to the neoclassical assumption of well behaved,
          exogenously given preference structures and objectively given
          production functions. Following from 1 are points 2 and 3:
          2. Buchanan has never carried out any empirical work and hardly
          ever refers to that of others [ibid.p.62].
          3. "He objects to the use of a social welfare function" .....
          given that "it is wholly inappropriate to attribute goals to
          society as such". [ibid.p.57]
          4. In place of a social welfare function he stresses the
          importance of efficiency where his "ideas about efficiency are
          not the same as the ones that are current in neoclassical
          economics." [ibid.p.57]. His notion of efficiency is basically
          that embodied in the Coase theorem i.e. any contract arrived at
          by voluntary 1 cooperation is necessarily efficiency enhancing.
          Sandmo points out that this approach to efficiency does not
          appear to take us to a fruitfully distinct alternative to the
          neoclassical.

                Determining the substance of these claims requires a
          detailed consideration of Buchanan's treatment of concrete
          problems. Sandmo(1990) does this for many areas although not for
          that considered here. So, at one level this paper is complementary
          to his. At another, it is significantly different as Sandmo's
          paper is strictly in the 'economics as a positive science' mode.

          2. THE CONTEXT OF THE PAPER.
                In 1968 Gary Becker initiated crime and punishment as a new
          sub-field of economics. His analysis was driven firmly by
          conventional welfare considerations. The object was to derive
          optimality conditions for social responses to crime [Stigler
          (1970)]. The explanation of criminal behaviour operated on the
          individual level and was a straightforward application of
          mainstream utility-maximizing rational choice economics in a
          risky environment. Becker's work was developed by others for the
          most part empirically. The main thrust of this work was to
          establish a deterrent effect of punishment in opposition to the
          perceived received wisdom of earlier sociological work [see
          Cameron(1988)].  Theoretical and empirical derivations of the
          deterrent effect were premised on the notion of a rationa
          individual making a decision, at the margin, on entering into
          illegitimate activity.
                Until Buchanan's paper appeared, the issue of criminal
          firms had not been considered. Criminal acts had been treated as
          individual choices analogous to the traditional problem of
          determining optimal hours of paid labour. Given the themes
          identified we can suggest that Buchanan's research agenda was to
          determine whether the deterrence (punishment) approach to
          organized crime provides an optimal policy response. A
          supplementary question would be the precise amount of deterrence.
          This does not come to the surface because of the (mainly)
          negative answer to the first question. Given that Buchanan has an
          allegedly different approach from Chicago methodology [as

          expounded in Becker and Stigler(1977)] this should be revealed in
          his handling of this topic.

          3.CONTENTS.
                Buchanan begins with the following quotation from Samuel
          Butler:
          "... we should try to make the self interest of cads a little
          more coincident with that of decent people"
                This is merely a figurehead to the paper as the discussion
          starts without further reference to it. What is the purpose of
          this quote ? Firstly it implies erudition on the part of the
          author. Also it functions as as an appeal to authority by harking
          back to the well known remark by Adam Smith on the role of self
          interest of producers (butchers,bakers) in ensuring that we are
          all fed, clothed etc. This theme is only picked up in section V
          of Buchanan's paper by which time we will all, he hopes, be
          convinced by the argument. There we find a potted history of the
          idea (Mandeville,Hume,Smith) that individual selfishness can work
          to the social good complete with a reference to the above cited
          quote. Following this Buchanan links his own analysis to the
          eighteenth century individualist position. He ends the paper by
          reworking Smith's butcher's self-interest quote as follows: "It
          is not from the public-spiritedness of the leaders of the Cosa
          Nostra that we should expect to get a reduction in the crime rate
          but from their regard for their own self-interests" (p.407). This
          neatly rounds up the suggestion in the opening quote from Butler.

          To the workaday economist all of this is just extra cream on the
          cake as it is irrelevant to the 'scientific' content of the
          paper. To Buchanan it is part of his agenda to establish himself
          as a profound thinker, and to be seen as part of the great
          tradition of free market thought.  Indeed he refers to Smith's
          butcher quote as containing the "philosophical foundations of
          competitive economic organization" (p.406).
                The paper proper commences with a basic exposition of the
          neoclassical theory of monopoly. The simple idea presented is
          that if this market structure restricts the output of goods then
          it must also restrict the output of bads. There follows the
          question of whether organized crime is a good thing through its
          being a monopolization of 'bad' production. To concretize his
          exposition the case of prostitution is seized upon as an area in
          which to show the application of his arguments. This is used
          early in the paper but later sections do not use any examples
          whatsoever. Prostitution is discussed without reference to any
          empirical evidence. Buchanan refers to what 	he believes other
          people believe	 under the vague and sweeping label 'journalistic
          discussion'. This term is not defined and there are no explicit
          quotations to illustrate what it might be. It is typically used
          in a way which suggests an attempt to denigrate the beliefs of
          opponents.
                The case of prostitution raises the problem of choosing a
          social welfare function but this is not faced head on. This is
          hardly surprising given point 3 above. Nevertheless an
          implicit social welfare function can be gleaned from statements

          that prostitution is a 'bad' in some 'social sense' (p.395)
          because of 'almost universal legal prohibitions' but it must be a
          good to those who pay for it.  If we stand the argument on its
          head we would have to say that a universally allowed activity
          (say some noxious pollutant) is a good. There seems to be
          a confusion here between the notion of 	marginal	 externalities
          and the 	absolutist	 notion of goods and bads.  For some reason,
          Buchanan seems reluctant to treat explicitly prostitution/crime
          as just another externality. This is quite unlike the mainstream
          literature [Hahn(1972),Usher(1986,1987)].  It is also hard to
          see how the application of contractarianism could clear up this
          anomaly.

                How do we get from individual goods to social bads? This
          is not taken up; we might assume a simple majority rule or sum
          of utilities function whereby any activity is a social bad if
          aggregate social losses exceed aggregate social gains. Such a
          function is rejected outright by Buchanan(1964). It implies the
          inclusion of welfare losses from meddlesome preferences i.e.
          those who object to prostitution 	per se	 rather than
          externalities which might arise from it. It should be noted that
          Buchanan fails to indicate 	any	 of the sources of externality in
          the case of prostitution. The statement that an activity is a bad
          because of almost universal legal restriction seems to imply that
          the law is a straightforward manifestation of the social welfare
          function (as revealed through voluntary exchanges) rather than
          the preferences of dominant rent-seeking groups. This is in

          apparent conflict with much of Buchanan's other writings which
          not only reject the notion of a a social welfare function but
          also emphasize the role of rent-seeking in public policy.
                There are also contradictions between the method of
          arriving at the notion of prostitution as a social bad and
          Buchanan's views on efficiency. I.e. if the meddlesome offended
          are not offended enough to engage in a voluntary contract with
          the perpetrators then can we not conclude that the activity is
          not a bad after all? Perhaps there is some impediment to
          voluntary exchange in which case an optimal policy could
          concentrate on its removal rather than prohibition of the
          activity which is a bad to some people but not to others.

                One thing which would help make an activity an absolute
          social bad would be a fixed externality, in the case of
          prostitution an objecter would suffer a large and equal loss of
          utility whether there were one or a million prostitutes with
          one or a million customers. Even in such a case the proportion of
          objecters and others would determine whether the activity was a
          marginal bad. If we take a more extreme case (such as religious
          objecters), where no amount of monetary compensation can nullify
          a perceived evil and the bad is still output invariant then
          reductions of output would bring only losses [to those who enjoy
          the good] and no gains.
                Section II of Buchanan's paper presents the 'simple
          geometry' of the analysis. Analogy with work in the fields of

          international trade,duopoly,voting and public goods (p.396) is
          used to promote the idea that the "reaction curve construction"
          is both sound and useful. In fact, the empirical work on the
          economics of crime which appeared contemporaneously with his
          paper would see his diagram as a straightforward supply and
          demand analogue [Cameron(1989)].  It may be that Buchanan
          presents his geometry as a reaction curve construction rather
          than a standard market supply and demand model because he wishes
          to avoid being pushed into a distinctly non-subjectivist corner
          of assuming given costs and preferences. Criminal activity is
          shown as a negative function of the amount of resources devoted
          to law enforcement. This presupposes a deterrent effect which
          Buchanan does not even deign to discuss 2. The supply of law
          enforcement resources is seen as a positive function of the
          amount of crime. In supply and demand terms the former function
          is the supply of crime and the latter is the demand for
          protection against crime.  Diminishing returns to enforcement
          effort are assumed merely on the grounds that "it is reasonable
          to hypothesize" (p.397) their presence. This is incorporated in
          the diagram but plays no part in the central argument. Its
          importance is as a persuasive device which reassures the
          professional economist that s/he is on familiar ground whatever
          discomfort the topic may engender.
                The problem of aggregating over individual criminals to
          obtain (what is in effect) a supply curve is not discussed
          although subsequent discussion makes it clear that perfect
          competition is being assumed.  An obscure discussion (p.404)

          almost says as much about the demand for law enforcement:
          "The community enforcement-response function necessarily
          describes outcomes generated by the interaction of many
          behavioral components; in many respects such response are more
          closely analogous to competitive, than to monopoly, behavior."
          There is also no explicit statement that individual criminals are
          utility maximizers. Presumably this is because an economist would
          be expected to take this for granted. Nonetheless given
          Buchanan's attachment to subjectivism there should be some scope
          for disagreement with the Chicagoan penchant for applying
          subjective expected utility [	pace	 Cameron(1989)].
                The geometrical construction is given further credibility
          by some 'intermediate micro' type informal stability analysis.
          Again this is not strictly necessary to the central argument;
          presumably it serves as further reassurance of familiar territory
          to persuade us to go along with the author.
                Section III of Buchanan's paper, titled "The Predicted
          Effects of Criminal Monopoly" contains the heart of the paper. It
          begins by distinguishing two types of crime.  The first is
          "socially bad" but constitutes a "good" to customers of
          prohibited products; prostitution is again raised as a "typical
          case" (p.396). Here we find an attack on "journalistic
          discussion" for labelling this type of crime as 'victimless'.  No
          rationale for this attack is given. Perhaps this would have
          exposed the pointlessness of distinguishing activities along the
          socially bad/privately good dichotomy. All criminal activities

          are presumably bad in some sense given that they are illegal (as
          Buchanan points out; see above) but they must also be seen as
          goods by some individuals,  otherwise they would not rationally
          have engaged in them. This applies equally well to Buchanan's
          second type of crime which is a forced transfer in the absence of
          mutual agreement.  No additional light is shed in this part of
          the discussion on the problem of the social welfare function.
                The equilibrium conditions for factor inputs in perfectly
          competitive and monopolistic industries are then stated.  There
          then follows the extraordinary claim that rational monopolists
          are profit-maximizers (p.399).  One could make the case that
          criminal monopolists may be more likely to depart from profit
          maximization given that they can not have shareholders exerting
          pressure on them. This may lead to a greater or lesser output
          than Buchanan expects depending on the precise form of the
          objective function of the managers of the criminal cartel.
          It is interesting to note on this point a lacuna in Buchanan's
          treatment of public expenditure whereby [Sandmo(1990 p.55)]
          he fails to advance an explicit characterization of the maximand
          of supplying bureaux. The reluctance to explicitly analyze the
          goals of organizations may be a reflection in miniature of his
          stance on the social welfare function [try substituting the word
          organization for society in the quote at point 3. in our
          introduction].

                 With reference to his simple diagram Buchanan gets us
          safely home to a standard claim that monopoly, for bads, is
          Pareto superior in a partial equilibrium setting:
          "Resources involved in enforcement may be freed for the
          production of alternative goods and services that are positively
          valued; the taxpayer has additional funds that he may spend on
          alternative publicly provided or privately marketed goods and
          services." (p.402).
                This is a potentially uncomfortable position. Buchanan has
          exercised persuasion to convince us of the correctness of his
          analytical proposition. But, he is somewhat wary of the ethics of
          being seen to endorse government supprt of organized crime. So,
          section IV. is entitled "Possible Objections to Criminal
          Monopolies". First, he sets up some very soft targets
          (distribution, joint supply), purely in the domain of 'positive'
          economics, that can be knocked down with some simple theory. He
          sees the third case as more serious. It emerges from the
          possibility that monopoly may be "more efficient" (p.403) than
          competition. A precise definition of the intended usage of
          efficiency is not given. Nor is there any empirical justification
          for the argument beyond the claim that "It seems plausible (..)
          at least under some circumstances" (p.403). Once again Buchanan's
          resistance to empirical work goes further than mere generalizing
          from the work of others (or any kind of 'casual' empiricism) to
          pure back of an envelope quantification not dissimilar to that of
          the bygone era of Keynes,Harrod,Hicks et al. From the following

          statement (p.403) "To the extent that there are significant
          economies of large scale in crime, monopoly organization will
          tend to be relatively more efficient" it appears that the
          argument is simply one of differently shaped cost curves for
          monopoly and competitive crime organizations.
                With the central proposition firmly established on
          theoretical grounds Buchanan is quick to distance himself from
          its real world implications. He says that some will worry that
          the government is being asked to "deal with the syndicate" but
          "I should emphasize that there is nothing of the sort implied in
          the analysis to this point. In its strictest interpretation the
          analysis carries no policy implications at all".(p.403-4)
          How can this be?  After all, the pivotal feature of the paper is
          the claim that we should have a different policy stance towards
          organized crime from that normally advocated.  The next sentence
          says "It merely suggests that that there may be social benefits
          from the monopoly organization of crime". Surely if there are
          benefits they ought to be seized?; this is the gospel of Pareto
          optimality.  The back pedalling might be due to doubt about the
          form and membership of the social welfare function e.g. criminals
          may or may not be included [see Cameron(1989a). It seems not,as
          Buchanan steadfastly continues to confront this directly.
                Despite the baldness of the above quote about the total
          absence of policy implications, Buchanan sticks to his view that
          criminal monopoly may be socially beneficial but is keen to
          cleave a distinction between passive acceptance of criminal
          monopolies by law enforcement agencies and pro-active

          promulgation of such entities. On the face of it, it is hard to
          see much of a distinction on pure economic theory grounds; so
          some persuasion is needed. Buchanan adopts the twin-pronged
          strategy of making technical sounding noises and 'telling a
          story'.  The technical argument is that the "trading" solution
          "lies off the community's enforcement response"(p.404). This is
          just a restatement of the traditional monopoly result that
          equilibrium is on the demand schedule and off what would be the
          supply schedule 	if there were competition	. In the crime case what
          this boils down to is a smaller outlay of resources by the
          enforcement bureau than that  voted to it by the community. The
          story given is that this would lead to "pressure on politicians
          to break the agreement"(p.404). Buchanan slips this in without
          ever giving any hint of an explicit model of the relationship
          between citizen-voters, politicians and law enforcement managers.
          It is not obvious that the Buchanan story will come to pass.
          Given the asymmetric information which is characteristic of the
          bureaucratic scenario there is surely scope for the shortfall of
          resource use to be hidden at some level of organizational
          expenditures [Cameron & Golby(1992)]. Buchanan (p.404) throws in
          an additional difficulty that pro-active promotion of criminal
          monopoly may be difficult because it is hard to identify the
          criminal syndicate. Having said this he simply repeats his
          original position: "Passive acquiescence in the syndication of
          crime is a wholly different policy stance from active
          negotiations with identified leaders." (p.404/405). There still
          seems to be no justification for the proposed asymmetry.

          Deliberate passivity, in the pursuit of an optimum, surely
          requires the same level of knowledge as active intervention.

          Section IV of Buchanan's paper concludes with a very banal
          paragraph about law enforcement techniques. Its presence is due
          to the fact that Buchanan is extremely worried that readers may
          be drawing anarchistic conclusions from his tacit endorsement of
          organized crime. So worried is he, that he even feels the need to
          remind us that the enforcement agencies will still be making
          efforts to deter crime: "The analysis does nothing toward
          suggesting that enforcement agencies should not take maximum
          advantage of all technological developments in crime
          prevention,detection and control" (p.405).
          If the enforcement agencies are extracting rents then they would
          	not	 be taking maximum advantage (as far as the ultimate consumer
          is concerned) from such shifts in the production function. As we
          have indicated above, Buchanan thinks this would only be a
          problem with an active stance of encouraging criminal monopolies.
          So he would hope to have headed off this line of criticism in
          part from his vague endorsement of the assumption that the voting
          of resources to law enforcement follows some analogue of a
          competitive process.
                In Section V, Buchanan returns to the agenda of the Butler
          quote. He re-states that "there is no argument for monopolistic
          restriction .. whether this be done via governmental agencies ...
          or by profit-seeking private entrepreneurs."(p.406) in an economy

          of pure exchange of goods. No clarification is given on why it is
          deemed fit that the state should have a monopoly in the
          administration of law enforcement activities. Thorough-going free
          market thinkers [cp. Rothbard(1973,1977)] would dispute this. The
          elementary welfare proposition of the paper is reiterated in
          slightly different words from earlier viz. that monopolization of
          bads may shift us in the direction of social optimality. The
          elementary welfare corollary that there is no guarantee that it
          will get us to the optimum is then stated.  Buchanan points out
          that unambiguous conclusions cannot be reached "for the cases
          here 'goods' and 'bads' are mixed" (p.406).
                We then jump back to the solid ground of activities that
          are "unambiguously 'bads' in the social or public sense" (p.407).
          It will be apparent from our earlier discussion that Buchanan
          does not make a very good job of conceptualising social/public
          bads. He then simply assumes that pure bads exist so that the
          'competition in goods is to the public good argument' "advanced by
          Mandeville and Smith becomes applicable in reverse" (p.407).

          4. SUMMARY AND CONCLUSION.
                The most striking feature to emerge from the above
          discussion of the content of Buchanan's paper is the presence of
          contradictions. At the most basic level Buchanan seems at many
          points to go against, or at least fails to follow up, the essence
          of his avowed methodology. This is in accordance with McCloskey's
          (1985) observation that neo-classical economists swear adherence

          to a pseudo-Popperian modernism which they constantly violate in
          the process of their day-to-day work.
                Surprisingly, for someone whose contributions are notably
          in the area of positive public sector analysis, Buchanan gets
          into considerable confusion about the conceptualisation of crime
          as an externality.  As we have seen, he makes use of the notion
          of a pure bad. It is hard to imagine the production of a pure
          bad; why would anybody want to generate such an entity? The
          activities regarded, by Buchanan, as pure bads are acts of
          redistribution (albeit involuntary) rather than production. It
          becomes impossible to make judgements about the optimal policy
          responses toward such acts without passing judgement on
          alternative distributions. A higher level of crime is better for
          some people: criminals themselves, sellers of crime prevention
          equipment and insurance recipients of stolen goods who receive
          them at discounted prices. Who is to say that the gains do not
          outweigh the losses ? The distribution of income is not a pure
          public good although there have been highly problematic attempts
          to characterize it as such [see Thurow(1971, p.328-329) and
          Ng(1983,207-211) for a critique].  Buchanan's rejection of an
          explicit social welfare function leaves a vacuum, as far as
          distributional judgements are concerned, which he does not fill
          with a meaningful alternative.

                This paper makes it clear that the nub of the argument
          could be presented, without doing it any great violence, in a
          very short space of time. Indeed, there are two paragraphs (one
          near the beginning and one near the end) which do so. The rest of
          the material might be expected to flesh out the argument. This is
          not the case; after all there is (quite deliberately) no
          empirical evidence.  So we might feel that the additional
          material is irrelevant. It is not irrelevant once the primary
          objective of persuasion is laid bare. The paper reflects an
          intention to sustain the literary form in economics despite the
          drive within the profession to alleged scientific status. The
          essence of literary form is the use of rhetoric to attempt
          persuasion.

          NOTES
          * I am grateful to Professor Murray Milgate for comments on an
          earlier version of this paper.
          1. A whole can of worms would be opened by debating the
          definition of voluntary. A good example is to be found in the
          debate on the use of the market to allocate bodily organs where
          there is great disquiet about the poor selling crucial organs
          whilst still alive [see Hilton(1990)].Buchanan desists.
          2. The evidence on this is far from conclusive; see
          Cameron(1988).



          REFERENCES.
          Becker,G.S.1968. Crime and Punishment: An Economic Approach.
          	Journal of Political Economy	 76(2) 169-217
          Becker,G.S. & Stigler,G.J.1977. De Gustibus Non Est Disputandum.
          	American Economic Review	 67(1) 76-90
          Buchanan,J.M.1964. What Should Economists Do?.	Southern Economic
          Journal	 30 213-222
          Buchanan,J.M.1973. A Defense of Organized Crime. in Rottenberg,S.
          (ed.) 	The Economics of Crime and Punishment	 Washington American
          Enterprise Institute. Reprinted in Andreano,R. & Siegfried,J.J.
          1980. 	The Economics of Crime.	 Wiley,New York.
          Cameron,S.1988. The Economics of Crime Deterrence: A survey of
          Theory and Evidence. 	Kyklos	 34 301-323
          Cameron,S.1989. A Subjectivist Perspective on the Economics of
          Crime.	 The Review of Austrian Economics	 3  31-44
          Cameron,S.1989a. On the Welfare Economics of Capital Punishment.
          	Australian Economic Papers	 28(53) 253-266
          Cameron,S. & Golby,D.1992. Public Goods and Policing.	Economics	
          28 Pt.3. No.119 104-108  August
          Congleton,R.D. 1988. An overview of the contractarian Public
          Finance of James Buchanan. 	Pulbic Finance Quarterly	 16(2) 131-157
          Hahn,R.G.1972. Crime and the Costs of Crime: An Economic Approach
          	Journal of Research in Crime and Delinquency	 9
          Hylton,K.N.1990. The Law and Economics of Organ Procurement.	Law
          and Policy	. 12(3) 197-224
          McCloskey,D.1985.	The Rhetoric of Economics	 University of
          Wisconisn Press.
          Ng,Y-K.1983. 	Welfare Economics	. Revised Edition. MacMillan.

          Rothbard,M.1973.	For a New Liberty	. MacMillan. New York.
          ----------.1977. Punishment and Proportionality. Ch.11. in
          Barnett,R.E. & Hagel,J. 1977.	Assessing the Criminal: Restitution,
          Retribution and the Legal Process.	Ballinger. Cambridge,Mass.
          Samuels,W.J.1990.(ed.).	 Economics as Discourse. An Analysis of
          the Language of Economists.	 Boston. Kluwer.
          Sandmo,A.1990. Buchanan on Political Economy. A Review Article.
          	Journal of Economic Literature	 28(1) 50-65
          Stigler,G.J.1970. The Optimum Enforcement of Laws.	Journal of
          Political Economy.	 78
          Thurow,L.C.1971. The Income Distribution as a Pure Public Good.
          	Quarterly Journal of Economics	 85 327-336
          Usher,D.1986. Police,punishment and public goods.	Public Finance	
          41(1)  96-114
          Usher,D. 1987. Theft as a paradigm for departures from
          efficiency.	Oxford Economic Papers.	 39(2) 235-252