new institutional economics and economic history · new institutional economics 9 new institutional...

9New Institutional Economics

New Institutional Economics (NIE) has been celebratedas a path-breaking approach to the understanding ofcapitalism. This article advances a conceptual critiqueof NIE approaches to economic history. The authorsuggests that NIE cannot solve the underlying tension,that its economics remains ahistorical, and that whenhistory, social relations and realism are invoked, theeconomics disappears, being replaced by variouscultural and state-centred explanations. Therefore NIE

is not so much a research programme in progress,but rather an indication of the degeneration of thetools of neo-classical economics.

Introduction

‘I nstitutions’ has become a keyword within recent politicaland scientific discussion. It reflects the widespreadrealisation that a well-functioning market economy

presupposes an effective institutional framework to work itswonders. In the light of global turbulence, financial unrest,the developments in the former Eastern Bloc, continuedunderdevelopment in the ‘Third World’, etc., more and morebusinessmen, economists, and politicians (from neo-conservatives to the New Left) are converging on the ideathat economic development without ‘good institutions’ andan effective state is impossible.

New Institutional Economicsand economic historyDaniel Ankarloo

Capital & Class #7810

This focus on institutions raises interesting questions.First, it implies that a capitalist market economy cannot beleft to itself, but is a social system in need of design andsupport (e.g. law and order, state governance, fiscalregulation...). From the political right to the left theemphasis on the scope and aims of governance of coursediffers somewhat, but the underlying premise that marketsneed institutional support and an effective state is becomingcommon ground.1 Secondly, it has focused attention on thehistory of capitalist development.

However, there are also other issues that presentthemselves, which are seldom spelled out: For example, ifthe market is not ‘the invisible hand’ which integratesindividuals in society but rather something in need ofinstitutional integrative support from society, what good isthe market? If capitalist markets do not support society, i.e.form the basis on which society rests, but rather need supportfrom the institutions of society to work, why should theinstitutions of society support markets, rather than, say,community or provision based on need? Such analyses mayvery well lead us to question the very basics of capitalism.

In economics the defence for market capitalism isgrounded foremost in neoclassical theory, which maintainsthat markets equalise supply and demand ‘spontaneously’ in‘equilibrium’, traits of the market that provide for both co-ordination, freedom and ‘efficiency’ (in the sense of optimumresource allocation).2 Charges have, however, been madeagainst this theory for its lack of realism (regarding its basicassumptions), for its asocial character (in theorising ‘utilityfunctions’ and ‘factors of production’, rather than actual livingindividuals with social relations, cultures, values etc.) andfor its ahistorical approach to the economy.

In this light, it is of specific interest to focus on NewInstitutional Economics (). Recognising the limits inneoclassical theory, has gained substantial influence (mostspecifically in e.g. questions on policymaking and develop-ment economics) exactly because it attempts to makeneoclassical economic theory more realistic, more social,and historical in its approach.3 This is attempted by movingbeyond neoclassical economics with an institutional analysisyet at the same time sticking to its basics. Furthermore—due mainly to the works of Douglass C. North— hasbecome an established tradition in economic history inexplaining the rise and evolution of market capitalism.


can hence be defined as a ‘neoclassical’ institutionalism, whichpolitically comes close to the positions of classical liberalism.

The first section of this article describes the conceptualframework of , and how it differs from neoclassicaleconomics. The second section outlines a theoretical critiqueof . The initial focus is on the concept of ‘transactioncosts’, which I argue is deeply problematic. Thereafter, Ipoint to an underlying assumption of ‘why not markets?’ in, which I argue rests on ideological rather than historicalgrounds. I also argue that the emphasis on institutions,ideology, and the state within makes the economic spheredisappear in explanation or makes uncertain the place ofthat sphere in explanation.

The third section focuses on the economic history of

with emphasis on the work of Douglass North. Here I arguethat writers like North treat previous historical economicsystems as if some key institutions of market capitalism werealready present. Such a position is unwarranted if we aresupposed to explain the historical evolution to capitalism.

The fourth section criticises the possibility of conductingsuch ‘as-if economic history’ by critiquing the instrumentalistargument within economics, mostly associated with MiltonFriedman (who, of course, is not a theorist).

In the concluding remarks, the reasons for regarding

a failure are summarised. The underlying contradictionbetween the institutional analyses and the neoclassicaleconomic basis of indicates not so much the success orregeneration of the neoclassical research programme butrather the degeneration of neoclassical theoretical tools.

1. The basics of New Institutional Economics

NIE as a school of thought is influenced by rather diverseideas from liberal social thought. On the one handevolutionary arguments, ideas of selection, of ‘survival ofthe fittest’, of ‘the invisible hand’ etc. are in various waysdeveloped and elaborated upon by .4 On the other hand,the concepts of markets, supply and demand, marginalism,factors of production, etc., are inherited from neoclassicalorthodoxy. As two new institutionalists claim:

...[T]he exponents of modern institutional economicsapply the analytical apparatus of neoclassical theory (and


newer techniques) to explain the workings and evolutionof institutional arrangements and thus to expand the scopeand predictive power of microeconomics. (Furubotn andRichter, : )

Although this is true, the conceptual framework of doesdeviate from neoclassical theory in various ways. Let us turnto these concepts that are distinctive in .

Bounded rationality

Firstly, more and more ‘new institutionalists’ reject the‘unrealistic’ assumption of perfect information and hencethe concept of full rationality in favour of the concept of‘bounded rationality’. This concept—originating fromHerbert Simon—is a disputed one.5 In the hands of it istaken to mean that in the presence of uncertainty, informationproblems, and ‘transaction costs’, economic agents cannotgather all the information necessary to calculate, case-by-case, the optimal solution on the market. Individuals aretherefore ‘intendedly [sic!] rational but only limitedly so’(Williamson, : ). For writers, the maximisingequilibrium of neoclassical economics is as a rule not attainable.

To minimise the problem of uncertainty economic agentsdevise rules of thought and action to base their decisionson. They become ‘rule followers’ (Vanberg, ). Theserules are rather fixed and do not change with case-by-casecalculation. As one observer puts it: ‘A boundedly rationalindividual intends to maximise, but finds it costly to do so.’6

Transaction costs

The main theoretical achievement of is the concept oftransaction cost. It originates from Ronald Coase’s

article on the firm, where he poses the question in this vein:

But in view of the fact that it is usually argued that co-ordination will be done by the price mechanism, whyis...organisation necessary? Why are there these ‘islandsof conscious power’? ... Yet, having regard to the fact thatif production is regulated by price movements, productioncould be carried on without any organisation at all, wellmight we ask, why is there any organisation? (Coase, :)


His historic answer is the following:

The main reason why it is profitable to establish a firmwould seem to be that there is a cost of using the pricemechanism. (Coase, : )

This ‘cost of using the price mechanism’ is a transactioncost. The concept is far from easily understood, and it remainscontested since definitions vary (see Ankarloo, : -

). Douglass North (: ) defines transaction costs as‘the costs of defining, protecting and enforcing propertyrights’. Oliver Williamson (: -) quotes KennethArrow who talks of ‘the costs of running the system’, as ‘theanalogy to friction in physics’. Steven Cheung (: )writes ‘The determination of the piece-rate—a price—illustrates the costs of “discovering” prices’. Tharinn Eggert-sson (: ) claims that ‘transaction costs are opportunitycosts just like any other costs in economic theory...’ Dahlmansuggests that a ‘first approximation to a workable concept’is ‘search and information costs, bargaining and decisioncosts, policing and enforcement costs.’ (Dahlman, : -

)7

A theoretical linkage can now be seen. In the presence oftransaction costs, markets and prices are not sufficient tocreate neoclassical equilibrium (‘efficiency’). Furthermore,new institutionalists argue transaction costs are pervasive inall economies, in every economic system. The developmentof transaction costs is the key to the institutional structureof a society and economy. One cannot, therefore, over-stressthe importance of the concept in . As two proponentsexplain:

The concept of transaction costs is crucial to anyacceptable interpretation of how a capitalist marketeconomy actually functions. To see the truth of thisjudgement it is only necessary to consider a world wheretransaction costs are zero. In such a ‘frictionless’ worldeven basic institutions as e.g. money, the firm, publicregulation become irrelevant. (Furubotn and Richter,: )

The operation of the market economy is in based on athird condition, the distribution and protection of propertyrights.


Property rights

The focus on property rights in capitalism is, of course,very welcome. In the state of general equilibrium, propertyrights’ distribution is immaterial to the equilibrium solutionof the market, but in the light of bounded rationality andtransaction costs, the original property rights’ distributionis decisive for economic efficiency, in this neoclassical sense(Coase, ). With the property rights theory of , thingshave changed for neoclassical orthodoxy.8

Property rights are mainly defined as the right to use,derive an income from and sell an asset (Furubotn andRichter, : ). ’s definition of property rights is muchwider than the one of political economy. (For example, YoramBarzel (: ) contends that human rights ‘are simply apart of people’s property rights’. In such a definitioneverything from buying a tomato to committing suicide ordriving a car too fast is an exercise in private property rights.Obviously, this runs the risk of stretching the property rightsconcept to the point where it explains everything andtherefore nothing.)

The economic problem now becomes how and when areproperty rights exchanged in the economy. This is not asimple affair, since property rights over assets have to bespecified in a contract, and hence enforced by law. Questionsarise: What rights are bought and sold? Under whatconditions? What happens if some party breaks a contract?What constitutes a breach of contract?, etc. Depending uponthe answers to such questions and the exclusiveness of theproperty right (Is it only mine, or can someone else use ittoo?) the value of the asset for particular owners will vary.In order to ensure the exclusiveness of private property rights,they must be well delineated and enforced. The vehicle forthis enforcement is the state. If property rights are easilydelineated, well protected, and enforced by the state, thenthe transaction costs will be low, and the gains from tradeinherent in the neoclassical market argument will be realised.If not, exchange will not occur and the market will not beput to work. Here is a noteworthy twist to the argument.North summarises:

A theory of the state is essential because it is the statethat specifies the property right structure. Ultimately it isthe state that is responsible for the efficiency of the


property right structure, which causes growth or stagnationand economic decline. (North, : )

Like with classical liberal economics, trade in the marketremains the cause of growth in , but this is now madedependent upon the workings of the state, and the way inwhich property rights and institutions work to lowertransaction costs.

Institutions

Institutions are ‘the humanly devised constraints that shapehuman action’ (North, : ). With this concept, canarguably become more realistic in its approach. In the wordsof the founding father: ‘Modern institutional economicsshould study man as he is, acting within the constraintsimposed by real institutions’ (Coase, : ). Institutionsare both ‘informal constraints’ (custom, culture, incentives,taboos etc.) and ‘formal constraints’ (law, property rightsetc.). Institutions are the rules of the game in the economy, and‘organisations’ (‘the players of the game’) arise in response tothe institutional structure. North (: ) explains: ‘It is theinteraction between institutions and organisations whichshapes the institutional evolution of the economy.’

Acknowledging that neoclassical equilibrium does notcome about in the real world, and that the property rightstructure, the state and the institutional evolution of societydetermine economic performance, North comes to thefollowing conclusion:

History matters. It matters not just because we can learnfrom the past, but because the present and future areconnected to the past by the continuity of a society’sinstitutions. Today’s and tomorrow’s choices are shapedby the past. And the past can only be made intelligible asa story of institutional evolution. (North, : vii)

The distinctiveness of NIE

With its conceptual and theoretical innovations, attemptsto make neoclassical economics more realistic. The primeexample of this is the revision of the assumption of humanbehaviour into a conception of ‘bounded rationality’, whichis perceived to be more in line with actual human behaviour.


With the introduction of ‘transaction costs’, in combinationwith the focus on property rights, also takes economicsin a more social direction, allowing for and explaining bothhistorical evolution and different economic systems than themarket, as well as pointing to the necessity of institutionaland state support for the market to work. Having opened upthese avenues of theoretical elaboration, ‘institutions’ hasbecome the key concept in indicating the need for a morerealistic, social, historical approach to economics.

2. Theoretical problems in NIE

The overriding question is, however, Do the concepts of

contribute to economic history? What is the realisation that‘history matters’ and what is the focus on social and economicinstitutions really worth?9

Theoretical problems with ‘transaction costs’

Let me for clarity emphasise that the initial assumption ofthe Coaseian procedure is the idea that production andexchange at least in principle could be carried out solely bythe use of the price mechanism. If the distribution of resourcesin principle could be conducted through the price mechanismon the market, why does the ‘non-market’ of the firm ariseand allocate resources through internal planning and co-ordination by ‘entrepreneurs’? Underlying this line ofreasoning is the question, ‘Why are there not markets?’ Note,therefore, that transaction costs cannot explain the origin ofmarkets. Rather, in the theory, markets are assumed andfrom this assumption all other institutions are explained. Insome versions their explanation, in turn, is dependent uponwhich institutional arrangement is the more ‘efficient’.10

I want to point out two problems here: (i) the concept oftransaction cost is problematic in relation to the basics ofthe neoclassical theory, which it was supposed to supplement,and (ii) the concept is not a very helpful tool in explainingthe origins and evolution of the institutions of capitalism.

(i) Transaction costs and neoclassical economicsTransaction costs are the ‘cost’ or ‘price of using the pricemechanism’. Such a formulation of transaction costs runsinto an infinite regress of markets and prices. If there is a


price to knowing the price (and if all prices entail transactioncosts), then there has to be a price to knowing the price ofknowing the price as well, etc., etc. So every market musthave a meta-market in which the price of the transactioncost to the price at the lower market level is determined.There is a hierarchy of markets, each one nested inside theother, all of which have transaction costs.11

A possible solution to this problem would be to say thattransaction costs are costs, which are not reflected or seen asprices on the market. They are the ‘underlying costs ofexchange’ (North, : ). But then they are not bonafide prices. Hence the regress stops. This route, however,presents problems in relation to neoclassical theory, in whichindividuals adapt to price information. In institutionalanalyses, these important costs ultimately determine theevolution and structure of institutions and organisations.However, if these are not reflected in prices at all, the allegedlyintegrating capacities of the price mechanism and themarket—as argued by neo-liberal free marketeers—are putinto question. So, contrary to that argument, prices do notconvey all the information necessary for the individual tomake rational economic calculations.

There is a basic problem for any theory that poses thequestion of the efficiency of institutions and organisationbased on the neoclassical assumptions of choice andcalculation and which at the same time claims to haveexplained the origin of those institutions and organisations.How could the economic actors have interacted to create anefficient outcome, when they either did not have theinformation to do so (assuming that transaction costs werenot visible as prices), or they were acting in the sphere wherethis information was not to be had (assuming that transactioncosts were not on the market)?12 This question applies nomatter whether the theory uses the full or the boundedconception of rationality.

This realisation has some further consequences. It seemswe would have to admit that there are costs, which are notsolely subjective and dependent upon the choice of individualactors, but are ruled by some other underlying non-subjectivelogic. In short, this calls for a theory of ‘objective cost’,which, again, is neither decided on nor reflected on themarket.13 If we follow the definition of transaction costs,they must be conceptualised as being costs and underlyingmechanisms that are governed by other forces than the


calculations of individual economic actors. This leads ustowards the conclusion that—given transaction costs—institutions influence choice.

If, then, underlying costs and institutions play thisfundamental role it seems reasonable, indeed even necessary,to investigate the structure of these costs and institutionsinstead, rather than starting from consumer’s choice, andeconomic calculation or prices on the market.14 If so, anyconception of rationality and choice outside the giveninstitutional framework seems unwarranted. In aninstitutional analysis we cannot, then, start from the givenindividual of the neoclassical model in order to explaininstitutions, we must start from the individuals as they appearwithin the institutional framework itself. Accordingly, thewhole underpinning conceptual structure of ‘consumer’schoice’ and methodological individualism is in danger.

The limits of the price mechanism are in effect recognisedby the transaction cost theorists of . If we choose to goalong this line, we confront another major challenge to theorthodox theory of market and individual choice. The onlyway to escape the problem of the limits of the pricemechanism seems to be to argue that economic actors indeedcan make cost-benefit calculations without knowing theprices, yes, even in the absence of prices. This route, however,presents somewhat of a no-win-situation for the transactioncost theory notion of just widening or complementingorthodox price theory. N acknowledges that prices are nota sufficient condition for conveying all the information neededfor economic actors to make rational calculation (the caseof positive transaction costs). The only way to save the ideathat economic actors can, and do, make rational choices onthe cost and benefits of institutional arrangements andproperty rights structures is to depict these individuals asmaking these choices without the use of the pricemechanism.15 This, however, means that the price mechan-ism is not a necessary condition for rational economiccalculation. Well might we ask: If prices are neither a sufficientnor a necessary condition for economic calculation andinstitutional choice, what is left for them to do? And, what istheir explanatory value?16

We can sum up the problem in the following dilemma: Iftransaction costs are visible as prices on the market, thenthey too must have a transaction cost. If on the other theyare not visible as prices on the market, then economic actors


can gain no information of them and base their choices andresource allocation on them—at least, not the standard wayof calculation on price information. The existence of thisdilemma makes difficult the integration of transaction costswithin the conceptual boundaries of neoclassical economics.

(ii) Transaction costs and institutional historical changeN builds upon the realisation that the world is full oftransaction costs, information problems, non-equilibrium etc.Yet, the explanation of these phenomena starts from orthodoxequilibrium: a ‘zero transaction cost world’, where institutionsdo not exist (Coase, ). But how can transaction costs arisefrom an institution-free and frictionless framework? How,from a position where there are no institutions, do institutionsarise and change? And, how do real individuals act and createinstitutions when institutions are lacking? These are indeedrelevant questions for historical investigation.

We could relax these strict conditions and say that the‘zero transaction cost world’ indeed is a world whereinstitutions and organisations exist, but that it does not matterwhich ones as far as ‘efficiency’ is concerned. But, then thefollowing problem arises: If we conceptualise a world withouttransaction costs as one encompassing institutions andorganisations, then we have to admit that transaction costscannot in general be the cause of institutions and organisations.It may then, at best, be true that in a ‘zero transaction costworld’ with institutions, the question of which institutionsthat exist, is irrelevant from an efficiency point of view—theso-called Coase Theorem (Coase, ). But the structureof these institutions cannot be irrelevant if we want to explain—from this ‘zero transaction cost world’—the direction ofchange of these institutions in history. In general, this wasthe challenge set before itself to explain, the directionof institutional change, and its effect on economic growth,decline and efficiency.

Transaction costs were supposed to help explaininstitutions in history, but instead they are dependent uponthem. The concept seems, therefore, rather ill-equipped toexplain institutional changes in economic history.

Why not markets?

It seems that cannot explain the origins of the market.Rather, the market is assumed, and from the pre-existence


of markets the question is why other institutions andorganisations exist. This procedure is most clearly shown inOliver Williamson’s treatment of the institutions of capital-ism (Williamson, ; ), which simply starts from theassumption that ‘in the beginning there were markets.’ Theideological content of this assumption is apparent. Fromsuch an assertion, all other allocation systems are to be explained(away) as ‘inefficient’, as deviations or ‘constraints’ to the market,or at best only introduced where the market ‘fails’.

On the face of it, the property rights theorists of aredifferent. For example, Dahlman (), McCloskey (;

; ), Fenoaltea (; ) and North and Thomas() provide particular analyses, which argue thatcommunal property rights, the manor system, the open-fieldsystem etc., under given constraints, can be viewed as rationaland efficient in the neoclassical sense, e.g. in avoiding andspreading risk (McCloskey, ). The basis of the explan-ation is, again, that a prevailing property rights structureexists because it is more ‘efficient’ than its alternatives.

Although these arguments take some historical facts intoaccount, there are two problems.

The first is that the argument assumes cost-benefitcalculation on structures of property rights, and subsequent‘choice’. The underlying premise remains that the market isalways a possible structure of property rights to choose. Ifeconomic actors have not ‘chosen’ it, it is because given‘constraints’, transaction costs etc, make another propertyright structure more ‘efficient’. However, it is hard to seethat there was a historically independent choice available ofdifferent structures of property rights for economic agents,the way property rights theorists describe it. (In theory, wecould at best choose to buy different amounts of commoditieson the market, but not choose to ‘buy’ more or less marketorganisation itself.)17 It is not as though feudal lords andserfs initially made a calculation on whether to ‘choose’serfdom over a labour market and feudalism over capitalism.But it is exactly this analytical leap property rights theoristsmake when they explain why one institution exists ratherthan another and why institutions change. Take, for instance,this argument from Demsetz (which, in passing, managesto confuse slavery, wages and serfdom):

For example it might be thought that a firm which usesslave labour will not recognise all the costs of its activities,


since it can have its slave labour by paying subsistencewages only. This will not be true if negotiations arepermitted, for the slaves can offer to the firm a paymentfor their freedom based on the expected return to themof being free men. The cost of slavery can thus be inter-nalised in the calculations of the firm. The transition fromserf to free man in feudal Europe is an example of thisprocess (Demsetz, : ).

In short, in markets have to be described as present evenin their absence. (I return to this as a criticism of North’sinstitutionalism.)

The second problem is that, once we admit thatinstitutions and structures of property rights influence choice,then these institutions and the structure of property rightscannot be explained by those very same calculations andchoices. If the same rationality and efficiency argument isinvoked to explain both the persistence of an institution (e.g.serfdom) as well as its change (into e.g. wage labour) thenit is difficult to specify the explanatory capacities of theefficiency argument underlying the theory. It boils down tothis vacuous truism: ‘If the market is there, it is because it isefficient, and if it is not there it is because that is efficienttoo.’

For the existence of neoclassical economics remainsan ideological blockage. The underlying premise of ‘whynot markets’ effectively dismisses the very question economichistory puts on the agenda: ‘why markets?’ Markets are, afterall, results of historical evolution not its primordial condition.

‘The disappearance of the economic’

N is wide-ranging. North (; ), in particular, has,in recent years, tried to widen his institutionalism toincorporate everything from ‘ideology’, ‘mental models’ tolaw and the state. North has come to see economic growthor decline as determined by ‘informal constraints’ and thestate, and therefore not in and by the traditionallyconceptualised economic sphere of exchange and market.However, the further he goes along this road, the more thiseconomic sphere disappears in his model. The economiclogic of a system is no longer the cause of development andgrowth, development and growth become rather theconsequence of non-economic ‘constraints’.18 It is other


relations—‘ideology’, ‘mental models’ and the state—thatexplain different economic systems and their performance.

North, as well as in general, can only succeed in savingthe core of neoclassical economics by making it almostsuperfluous to the explanations in the model. The economicvirtually disappears. There is no mystery to this as North(: ) has conceded that neoclassical economics is notsuitable for explaining how economies develop. If neoclassi-cal economics is the only conception of the economic sphere has, then the economic sphere will ultimately have littlevalue in ’s explanations. Once it is admitted thatneoclassical economics is a static framework, the dynamicfor social change in history must come from somewhereelse. This is, it appears, the main theoretical function of theinstitution concept of , even what necessitates it (Ankarloo,: -). N does, therefore, not so much enhanceor move beyond neoclassical economics with its focus oninstitutions. Rather, the more realistic, social and historical tries to be, the more it also moves away from neoclassicaleconomics. In doing so, indicates not how fruitful its basiceconomic approach is, but rather the opposite.

This pressing dilemma is becoming clear to North inparticular. And, although he repeats throughout his workthat transaction cost concepts are only complements thatprovide explanatory variables alongside the traditionalconcepts of neoclassical economics, these traditionalvariables, prices, markets, technology and production, aremore and more pushed aside. In come instead ‘culture’,‘mental models’, ‘ideology’, law and the state, as the mainexplanatory variables. (This tension in North’s work isdiscussed at greater length in Ankarloo, : -.)

Consequently, in North’s version of , the problem ofunderstanding the economy and its institutions is reversedinto the one of giving these institutions a microeconomicexplanation. The economic conceptual apparatus, whichcalled for supplementation and revision, is now invoked toexplain the revision itself. N, therefore, tries to explainthe state and policy in economic terms. There is talk of thestate as ‘political markets’, of ‘political entrepreneurs’ ofvoting and ideology as ‘the cost of expressing your convic-tion’ (see North, : -; : -).

The hope to escape by bringing the economic back inagain in this fashion is illusory. Even if one could agree toconceptualise politics, state, culture and ideology as ‘markets’,


one would at least have to concede that these sorts of ‘markets’have a very different logic than economic ones. There is ingeneral no ‘voluntary contracting’, no exchange of propertyrights (not to mention the absence of commodity produc-tion), no economic calculation on price information etc. in‘political markets’. If there is any ‘exchange’ here, it is of avery different kind than market exchange. N suffers fromthe tendency to over-extend the idea of property rights.Similarly, by extending the concepts of market and exchangeto these other spheres, runs the risk of losing any particularexplanatory value that the concepts have. Indeed, it is hardto see how neoclassical theory, which on its own is incapableof explaining e.g. money, distribution, markets and efficiency,suddenly could explain ideology, culture and the state.‘Conceptual imperialism’, does not provide salvation for ;it is its dying breath. Even North, perhaps the must unorthodoxand original new institutionalist to date, cannot do awayfully with the barren burden of his original economicparadigm (North and Wallis, ; Ankarloo, : -

). N is caught in the middle: paying the double price ofboth abandoning and clinging on to concepts of neoclassicaleconomics at the same time.

3. NIE and economic history: The work of DouglassNorth

With its theoretical apparatus enters into the big debatesof economic history: transition debates, the origins ofcapitalist firm hierarchies etc. Yet examples abound of howhistorical investigation is not the driving force of theargument. Rather history is invoked to illustrate theoreticalpreconceptions. As an example, take explanations ofhierarchical capitalist firm relations.

Workers voluntarily undertake to be supervised... Theysubmit to being compelled to work harder than directincentives provide for, because the consequence is a higherexpected utility. (Stiglitz, : )

Or, in a similar vein, there is this ‘explanation’.

My favourite example is riverboat pulling before thecommunist regime, when a large group of workers


marched along the shore, towing a good sized woodenboat. The unique interest of this example is that thecollaborators agreed to the hiring of a monitor to whipthem. (Cheung, : )19

If this were all that had to offer, a real historical problemwould be to explain the influence of in economic history.So it is through the consistent application of the tools of

by Douglass North that has gained its influence in thisfield. His first path-breaking analysis of economic transfor-mation concerns the transition from feudalism to capitalism(co-authored with Robert Thomas). Here he contends thatthe manor system was ‘efficient’ and therefore chosen by theeconomic agents. He says:

…[T]he contractual arrangement of the manor can nowbe seen as an efficient arrangement for its day. Theobligation of the serf to provide labour services to hislord and protector, an input-sharing arrangement, waschosen because given the constraint of high transactioncosts involved in trading goods, it was the most efficient.The almost total absence of a market for goods, plus theexistence of a rudimentary market for labour, ensuredthat inputs could be shared with lower transaction coststhan would be involved in other contractualarrangements… The ‘quaint’ organisation of the classicmanor is therefore understandable as an appropriateresponse in the general absence of a market economy.(North and Thomas, : )20

The transformation of the manor system and feudalism isexplained by population growth and decline which changesthe market and the land-to-labour ratio in the economy.Thereby both prices and the transaction costs of serfdom vs.wage labour eventually changed (North and Thomas, ;

).

From handicraft to putting out system to the factorysystem spans more than three centuries: the key toexplaining the transformation is growth in the size of themarket and problems of quality control (that is,measurement of the characteristics of the good). In thecourse of the transformation in economic organisationwage labour developed… (North, : -)


North is sensitive to historical developments, as theformulations ‘absence of a market economy’ and ‘wage labourdeveloped’ indicate. The disturbing problem with North’seconomic history is all the same that the decisive institutionsand human characteristics that are specific to capitalism haveto be assumed to exist even when he admits capitalism isnot at hand. Otherwise the explanation does not work. But,if he acknowledges that markets and the institutions ofcapitalism originate from somewhere, and evolve, he cannotassume them to have existed from the start.

North does not explicitly do this, but in order for theexplanation to work the economic logic of the market has tobe described as present, even in the ‘absence of the market’.21

In the model, institutions evolve as rational responses torelative price changes, by, albeit ‘constrained’, capitalistmicro-rational economic men. But calculating what to buyonly makes sense when there already are commodities onthe market. These calculations cannot at the same time bethe cause of the market for these commodities.22 This isespecially true of the labour market. If wage labour ‘develops’,and there is ‘an absence of a market economy’ as Northconcedes, then the price of labour cannot be assumed fromthe start of the explanation. North fails to acknowledge thefact that the ‘free’ individual—the micro-calculatingconsumer—is the result of capitalist markets, rather thanthe other way around. In North’s theory this individual hasno history, and this problem of neoclassical theory is a criticallimitation on his programme of ‘history matters’.23

How about the land-to-labour ratio explanation in thedescription of the transition from feudalism to capitalism?We need not doubt that population change and a changingland-to-labour ratio will have effects on the economic system,and its underlying logic. The decisive counter-argument isthat these changes cannot be mediated as price changes andmicroeconomic choice, the way North and Thomas describeit, unless land and labour (or rather labour power) alreadyare commodities. They were not. This remains valid both asa historical remark, in relation to the historical developmentof proletarianisation and marketisation, and more impor-tantly as an analytical remark. Feudal work relations arenot wage labour relations. Indeed, North and Thomasacknowledge in their definition of feudalism that labour andland are not commodities. Though commodification didexist, it did so only to a small degree.24


We cannot have it both ways: either capitalism came fromsome different economic system, or else it is ubiquitous. Inthe first case, we have to describe the economic logic of thepreceding system in its own right—i.e. as non-capitalist. Inthe second case, this economic history does not explaincapitalism. Rather it assumes it from the start. If this secondcase is true, then economic history must be totally revisedfrom being the history that leads to capitalism, into a historyof capitalism.25

North and Thomas (; ) try to solve this dilemmaby using words like ‘implicit contracts’ implying, one mustsuppose, ‘implicit prices’. Unfortunately, North thenencounters the challenge found in the analysis of ‘transactioncosts’: How could the people of the feudal era, who suppo-sedly chose the most efficient property rights, make thesecalculations on implicit information? If there was no market,how could they visualise it, and economise on it? Consideringthe persistence of serfdom, even if individuals could calculateon a non-existent market, how could they act upon thisknowledge? Labour mobility was heavily restricted by theserf ’s bondage to his or her lord, which means that even ifmarket opportunities indeed were present, individuals couldnot take advantage of them.

The ahistorical assumptions and explanations constitutea consistent luminous thread in North’s work. North endsup, willy-nilly, in maintaining that all institutionalframeworks work ‘as if ’ capitalist micro-rational principlesare in place, even when he knows the principles are not(Mirowski, : -. cf. Brenner, ). He paintshimself into a corner because theoretically, in order to makehis explanations work, he has to accept, through thebackdoor, historical assumptions which he knows arefalsehoods.26 This is what I call a position of ‘as-if-economichistory’.

4. A critique of ‘as-if ’ methodology in economic history

Granted that the use of unrealistic assumptions, i.e. knownfalsehoods, is at the basis of , is there a possible defencefor them? Well, historically, within the subject of economics,yes. Because an escape from realism—at times conductedby economists, although not yet explicitly by exponents of—is to adopt an instrumentalist, ‘as-if ’ strategy. Here one


subscribes to the view, that theoretical concepts are merelyinstruments for the understanding and prediction ofoccurrences in reality. The concepts and theories themselveshowever are not realistic or non-realistic at all. Hence, theymake no ontological commitments, and have no truth-value.Criticism of them is futile. Within economics this line ofreasoning is principally associated with Friedman ().North himself describes the economist’s attitude:

Although I know of very few economists who really believethat the behavioural assumptions of economics accuratelyreflect human behaviour, they do (mostly) believe thatsuch assumptions are useful for building models of marketbehaviour in economics... (North, : )

The main point in Friedman’s methodology is the statementthat although the assumptions are decidedly false, thepredictions that follow from the theory work ‘as if ’ theassumptions were correct. His most famous example is thatleaves on trees of course do not calculate to maximise expo-sure to sunlight but if we investigate the position of leaveson trees it works as if they did. Another example is the skilledpool player, who—although he does not calculate using thelaws of Newtonian physics—plays ‘as if ’ he did so.

But could the problems of in relation to economichistory be solved by adherence to something like the Fried-man argument? Two decisive problems of the methodologysuggest a negative answer.

Firstly, following Friedman’s own argument, the ‘as-if ’method can at best show the irrelevance of assumptions,given the fact that the aim and test of science is to predict.27

The actual motives and actions of social actors are irrelevant(only the result matters). In order to uphold this method forhistory, we would have to maintain that the aim of historicalresearch is to describe what happened in history irrespectiveof people’s motives and irrespective of how they actually behavedand acted. This is surely absurd.

Secondly, in contrast, by definition the aim and test ofhistorical research cannot be prediction, even if we grantedthat it could be the aim for research in other sciences. Amain aspect of historical investigation is the aim of obtainingknowledge of the past, not prediction of the future. Ashistorians we do not want to predict what happened in humanhistory, by some invented behavioural assumptions, or even


known falsehoods. Rather we want to explain past historicaldevelopments, or at least describe them. In relation to ,which brings known falsehoods into its economic history,the following contradiction arises: If prediction is the onlyvalid test for science then, by definition, contrary to its promise,history does not matter. If on the other hand history does matter,then prediction cannot be the aim of the science, and it cannotvalidate the falsehoods of the theory.

If history does matter to the discipline of economics, andif institutions matter to our understanding of economies inhistory and the present, as above all North contends, thenany version of the ‘as-if ’ method—including the home-madeinstrumentalism of Friedman—is not viable. As Fleetwood(: ) has put it:

One only has to reflect upon this for a moment to see thisconclusion is self-evidently correct: if known falsehoodsare allowed into explanations imagine the bizarreexplanations that could be advanced.

5. Conclusion

There is an underlying contradiction in in its attempt tocreate a form of neoclassical economics which is morerealistic, more social and more historical in its approach.The concept of ‘institutions’ is invoked in in recognitionof the problems of neoclassical economics. What has emergedfrom this analysis is a picture of , in which, inasmuch as tries to retain the basic elements of neoclassical theory,the acknowledged non-realistic, asocial and ahistoricalfoundations re-emerge as major obstacles to the success ofthe project. When, on the other hand, , via its institutionalanalyses, does try to move in a realistic, social and historicaldirection, its neoclassical economic foundation is renderedincreasingly superfluous, or even contradicted. This dilemmais then ‘solved’ by bringing in, through the backdoor,historically known falsehoods and false historical startingpoints such as ‘in the beginning there were markets’ and byviewing history ‘as if ’ it were like the theory assumes, ratherthan developing theoretical formulations which incorporatehistorical research. History is understood in light of thepresent, rather than the present understood in the light ofhistory. To put it succinctly: where there is economics in


there is very little history, and where there is history there isvery little economics. Thus, the problems of economicorthodoxy are not solved but inherited. What is highlightedis the inadequacy of its explanation of how capitalist economieshave historically developed. N is not so much a thrivingresearch programme in progress, but rather a demonstrationof the degeneration of the tools of neoclassical economics.

In the end, were we to accept the premises of the ‘as-if ’approach to economic history underlying , we would riskmoving from a scientifically informed theory based onrational argument to theoretical formulations in which‘anything goes’, the basis of which would lie in trust andfaith, if not pure ideology. Any demarcation line betweenhistory and fiction would disappear. As for politics, the aboveconclusions suggest that inasmuch as neoclassical economicsfails to adequately provide theoretical grounds for the allegedblessings of the capitalist market economy, so does .Behind the theoretical niceties lurks the same capitalistapologetics that the ahistorical neoclassical approach toeconomics and economic history has fostered all along.

is not the solution to this problem but a testament to thefact that the problem remains.

* Daniel Ankarloo finished his PhD thesis on New Institutio-nal Economics on January at the Department ofEconomic History, Lund University, Sweden.

Notes

The author would like to thank Steve Fleetwood for his constructiveand incisive comments to a previous draft of the paper, as well as AlanFreeman for appreciative comments along the road to finishing thearticle.

1. The intellectual developments of former Hayekian John Gray andformer Marxist Geoffrey Hodgson are interesting examples of thisfact. All other disagreements aside, they both now converge on thisidea that institutions must support the market to work. See e.g.Gray () and Hodgson ()

2. All references to ‘efficiency’ below refer to this neoclassicalconception of efficiency as optimum recourse allocation on thebasis of given ‘constraints’.


3. On the different definitions and descriptions of , see Langlois(), Eggertsson (), Vromen (), Furubotn and Richter().

4. On this matter see Hodgson (a) and Vromen ().5. See Langlois () and Hodgson (b).6. Kreps, quoted from Furubotn and Richter (: ). For critical

expositions of the rationality assumptions of , see Hodgson (:chs. -), Mäki and Gustavsson and Knudsen () and Ankarloo(: -).

7. In the light of the widespread usage of transaction cost conceptstoday, it seems appropriate to stress that in Coase’s theory prices arethe cause of transaction costs. His original idea was that the onlyplausible economy without transaction costs would be a fully plannedeconomy, with no direction of the price mechanism. He says: ‘If Iwere asked to imagine an economic system in which transactioncosts did not exist, it would be a completely communist society.’(Coase : )

8. The property rights theory has its origin in in Coase () andhas subsequently been developed by Demsetz (; ), Furubotnand Pejovich (; ), Alchain and Demsetz (), De Alessi(), Barzel () and North and Thomas () to name a few.

9. What follows below is a summary of ideas more elaborately presentedin Ankarloo ().

10.See e.g. Alchian () and Williamsson (). North ()however does not subscribe to this view.

11.Mirowski (: ) in a different vein points to the problem of‘meta-markets’. He says: ‘...what structures organise this “meta-market”, to allow us to buy more or less market organisation...Who sets the price of the market? Who sets the price of the price ofthe market?’

12.The fact that transaction costs are not visible or measurable to theeconomic actors is mentioned in Magnusson and Ottosson (:

-) and Samuels and Medema (: -).13.‘Objective cost’ here is used only in contrast to the cost and price

formation of the market that neoclassical production functionanalysis and market analysis account for. Relating transaction coststo neoclassical economics does not entail that one should accept it.A disturbing critique against its subjective cost and price theory isthe argument that it is not possible for individuals to form andrevise their utility functions and relative demand functions withoutfirst knowing the relative prices of the goods and services which theycan choose from. Hence costs and prices are presupposed in theargument and not explained by utility functions, and cost-benefitcalculation. What Marx called ‘the illusion of competition’ (Marx,


: ch. ). We could of course say ‘value’ instead to solve theproblem.

14.This is indeed North’s definition of what economic history shoulddo, when describing the theories of Karl Marx and Joseph Schum-peter: ‘if economics is a theory of choice subject to specific constraints,a task of economic history was to theorise about those evolvingconditions.’ (North, : )

15.This idea that economic actors economise on transaction costs andrationally choose the most efficient among institutionalarrangements is pervasive in . The most prominent example isWilliamson’s theory of the choice of governance structures(Williamson, ; ).

16.A possible solution is to measure transaction costs in the economyin an indirect way. Attempts to measure the value of transactioncosts on various bases include North and Wallis (), Demsetz(), Williamson () and Furubotn and Richter (: -

). For all their merits and limitations, such measurementtechniques by researchers to determine costs which do not exist asprices, cannot answer what I argue to be the most fundamentalquestion facing : how can the actual economic actors determinethe level of transaction costs, without access to a price measure?

17.Of course individuals may choose to act in order to change andtransform given property right structures. But not by cost-benefitcalculation of having more or less of them, as would have it, butthrough (revolutionary) social praxis, in relation to already pre-given social structures which these individuals did not choose in thefirst place, as Marx would have it.

18.This analytical turn is clearly exposed in Rosenberg and Bridzell() which starts with an account of some of the major economicexplanations of economic growth presented in the history ofeconomic thought. Once these have been pushed aside as non-satisfactory by the authors, all that remains to discuss is politics, lawand culture.

19.Lazonick () effectively rebuts a famous argumentation of Alchainand Demsetz (). He states: ‘The key empirical point in theAlchian and Demsetz story is that with the rise of centralised powersources, “the measurement of marginal productivity...became moredifficult”. It will be of no help to us to examine the empiricalsources that Alchain and Demsetz used to support this contention—they did not cite any... Nor is it likely that they would have foundany support if they had bothered to look. The assertion concerningthe monitoring problems of a centralised power source is simplywrong – as are most of Alchain and Demsetz’s other statements...’(Lazonick, : -). As for Cheung’s example it forgets that


it is not the slaves who hire monitors to whip them, but the slaveowners, who do so!

20.North has later conceded that this view is over-simplified, because:‘…carrying over the modern-day notion of contract to the serf-lordrelationship is imposing a modern-day concept which is misleading.The serf was bound by his lord and his actions and movements wereseverely constrained by his status; no voluntary agreement wasinvolved. Nevertheless, it is crucial to re-emphasise a key point ofour analysis; namely, that it was the changing opportunity cost oflords and serfs at the margin which changed manorialism andeventually led to its demise.’ (North, : )

However, North’s correction is only half-hearted and misses theimportant point, In general, each individual lord only protected theserf from other lords or bandits etc. not from other serfs. So lordprotection is not a general explanation of the lord-serf relation assuch.

21.Phrases like ‘the absence of the market’ and the consistent view ofinstitutions as ‘constraints’ to the market, despite their sensitivityto history betray ideological residues in North’s thought. Why notjust describe ‘the presence of serfdom’ rather than ‘the absence of amarket’?

22.There are numerous critics of these conceptions in North’s work.See Mirowski (: -), Field () and Brenner (), whicheven today remains one of the most revealing critiques of the‘Smithian legacy’ to economics—and which, although not specificallyaddressed to North, applies to him as well.

23.There are examples of this analytical procedure in North’s writings,even to the prehistory of man. Here are some most telling quotes:‘Prehistoric man employed his labour in conjunction with naturalresources to produce his living. The natural resources whether animalsto be hunted or vegetation to be gathered, were initially held ascommon property. This type of property right implies free access byall to the resource. Economists are familiar with the propositionthat unconstrained access to a resource base will lead to its inefficientutilisation. This inefficiency as the demand for the resource increaseseventually leads to the depletion of the resource… This instance isan example of incentive failure caused by cultural or institutional(property rights) inadequacies.’ (North, : ) See also: ‘Let usexamine the situation where several bands compete for the samecommonly held migratory animals…The band then has the incentiveto exploit the resource to the point where the value of the lastanimal killed is equal to the private costs of killing it. The collectionwill continue until all of the income the scarce resource would haveearned under private property is dissipated. That is, in a competitive


situation no band has any incentive to conserve the resource, sincethe animals left to reproduce probably would be taken by the rivals.The stock of animals thus could be placed in danger of extinction.The crucial element causing this inefficiency is the lack of any barrierto the exploitation of the commonly owned resource base… Theresult is too many hunters.’ (North, : )

24.On the rate of proletarianisation see e.g. Tilly () who estimatesthat the proletariat between to grew eightfold, by

million people, almost equivalent to the total population growth ofEurope at the time.

25.On this matter see e.g. Wood ().26.It is for instance difficult to see why North clings on to transaction

cost theory in light of his own concession: ‘All the modernneoclassical literature discusses the firm as a substitute for market.[This] ignores a crucial fact of history: hierarchical organisationforms… predate the price-making market.’ (North, : )

27.The close relationship between Friedman’s instrumentalism and hispreoccupation for policy prediction is stressed by even his mostardent followers. See Frazer and Boland (: ): ‘Friedman ismore concerned with the immediate practical problems ofpolicymaking than with the philosophical problems that havetroubled those who have been searching for centuries for the onetrue method of finding the one true or acceptable theory.’

References

Alchain, A.A. () ‘Uncertainty, Evolution and Economic Theory’,Journal of Political Economy .

Alchian, A.A. and Demsetz H. () ‘Production, Information Costs,and Economic Organisation’, American Economic Review .

Ankarloo, D. () ‘“Institutions”’, What Is In a Word?’, Lund Studiesin Economic History . Lund.

Barzel, Y. () Economic Analysis of Property rights, CambridgeUniversity Press, Cambridge.

Brenner, R. () ‘The Origins of Capitalist Development: A Critiqueof Neo-Smithian Marxism’, New Left Review .

Cheung, S.N.S. () ‘The Contractual Nature of the Firm’, Journalof Law and Economics .

Coase, R.H. () ‘The Problem of Social Cost’, Journal of Law andEconomics .

_______ () ‘The New Institutional Economics’, Journal ofInstitutional and Theoretical Economics .


Coase, R.H. () ‘Comment on Cheung’, in Werin, L. and Wijkander,H. (eds.) Contract Economics, Blackwell, Oxford.

_______ [] () ‘The Theory of the Firm’, in Williamson andWinter (eds.) The Nature of the Firm: Origins, Evolution andDevelopment, Oxford University Press, Oxford.

Dahlman, C.J. () ‘The Problem of Externality’, Journal of Lawand Economics .

_______ () The Open Field System and Beyond: A Property RightsAnalysis of an Economic Institution, Cambridge University Press,Cambridge.

De Alessi, L. () ‘The Economics of Property Rights: A Review ofthe Evidence’, Research in Law and Economics .

Demsetz, H. () ‘The Exchange and Enforcement of PropertyRights’, Journal of Law and Economics .

_______ () ‘Towards a Theory of Property Rights’, AmericanEconomic Review, .

_______ () ‘The Cost of Transacting’, Quarterly Journal ofEconomics, .

Eggertsson, T. () Economic Behaviour and Institutions, CambridgeUniversity Press, Cambridge.

Fenoaltea, S. () ‘Risk, Transaction Costs and the Organisation ofMedieval Agriculture’, Explorations in Economic History .

_______ () ‘Whig History and the Common Fields’, inGustavsson, B. (ed.) Power and Economic Institutions, Edward Elgar,Aldershot.

Field, A.J. () ‘The Problem with Neoclassical InstitutionalEconomics, A Critique with Special Reference to the North andThomas Model of Pre- Europe’, Explorations in Economic History.

Fleetwood, S. () What Kind of Theory is Marx’s Labour Theory ofValue? A Critical Realist Inquiry, Capital & Class .

Frazer, Jr. W.J. and Boland, L. () ‘An Essay on the Foundations ofFriedman’s Methodology’, American Economic Review .

Friedman. M. () ‘The Methodology of Positive Economics’, inEssays in Positive Economics, Chicago University Press, Chicago.

Furubotn, E.G. and Pejovich, S. () ‘Property Rights and EconomicTheory: A Survey of the Recent Literature’, Journal of EconomicLiterature .

_______ () The Economics of Property Rights, Ballinger,Cambridge.

Furubotn, E.G. and Richter, R. () ‘The New InstitutionalEconomics: An Assessment’, in Furubotn and Richter (ed.) TheNew Institutional Economics, Texas A&M University Press, CollegeStation.


_______ () Institutions and Economic Theory. The Contribution ofthe New Institutional Economics, University of Michigan Press, AnnArbor.

Gray, J. () End Games, Polity Press, Cambridge.Hodgson, G.M. () Economics and Institutions—a Manifesto for a

Modern Institutional Economics, Polity Press, Cambridge._______ (a) Economics and Evolution: Bringing Life Back into

Economics, Polity Press, Cambridge._______ (b) ‘Transaction Costs and the Evolution of the Firm’,

in Pitelis, C. (ed.). Transaction costs, Markets and Hierarchies,Blackwell, Oxford.

_______ () Economics and Utopia – Why the learning Economy isnot the End of History, Routledge, London.

Langlois, R.N. () ‘Bounded Rationality and Behaviouralism: aClarification and Critique’, Journal of Institutional and TheoreticalEconomics .

Langlois, R.N. (ed.) () Economics as a Process: Essays in the NewInstitutional Economics, Cambridge University Press, Cambridge.

Lazonick, W. () Business Organisation and the Myth of the MarketEconomy, Cambridge University Press, Cambridge.

Magnusson, L. and Ottosson, J. () ‘Transaction Costs andInstitutional Change’, in Groenewegen (ed.) Transaction CostEconomics and Beyond, Kluwer Academic press, Boston.

Marx, K. [] () Capital vol. III, Penguin Classics, London.McCloskey, D.N. () ‘The Enclosure of Open Fields, Preface to a

Study of Its Impact on the Efficiency of English Agriculture in theEighteenth Century’. Journal of Economic History .

_______ () ‘The Persistence of English Common Fields’, inParker, W.N. and Jones, E.L. (eds.) European Peasants and theirMarkets: Essays in Agrarian Economic History, Princeton UniversityPress, Princeton.

_______ () ‘English Open Fields as Behaviour Towards Risk’,in Uselding, P. (ed.) Research in Economic History () Press,Greenwich.

Mirowski, P. () Against Mechanism, Rowman and Littlefield, ,.

Mäki, U. and Gustavsson, B. and Knudsen, C. (eds.) () Rationality,Institutions and Economic Methodology, Routledge, London.

North, D.C. () ‘Structure and Performance: The Task of EconomicHistory’, Journal of Economic Literature .

_______ () Structure and Change in Economic History, Norton,New York.

_______ () ‘The New Institutional Economics’, Journal ofInstitutional and Theoretical Economics .


North, D.C. () ‘A Transaction Cost Approach to the HistoricalDevelopment of Polities and Economies’, Journal of Institutionaland Theoretical Economics .

_______ () Institutions, Institutional Change and EconomicPerformance, Cambridge University Press, Cambridge.

_______ () Economic Performance through Time (The Nobel PrizeLecture).

North, D.C. and Thomas, R.P. () ‘An Economic Theory of theGrowth of the Western World’, Economic History Review .

_______ () The Rise of the Western World—a New EconomicHistory, Cambridge University Press, Cambridge.

North, D.C. and Wallis, J.J. () ‘Measuring the Size of the TransactionCost Sector in the American Economy, -’, in: Engerman,S. and Gallman, R. (eds.) Long Term Factors in American EconomicGrowth, Chicago University Press, Chicago.

_______ () ‘Integrating Institutional Change and TechnicalChange in Economic History’, Journal of Institutional and TheoreticalEconomics .

Rosenberg, N. and Bridzell, L.E. () How the West Grew Rich: theEconomic Transformation of the Industrial World, Basic books, NewYork.

Samuels, W.J. and Medema, S.D. () ‘Ronald Coase on EconomicPolicy Analysis: Framework and Implications’, in Medema, S.D.(ed.) Coasean Economics: Law and Economics and the New InstitutionalEconomics, Kluwer Academic Press, Boston.

Stiglitz, J.E. () ‘Incentives, Risk, and Information: Notes Towarda Theory of Hierarchy’, Bell Journal of Economics .

Tilly, C. () ‘The Demographic Origins of the European Proletariat’,in Levine, D. (ed.) Proletarianisation and Family History, AcademicPress, London.

Vanberg, V. () Rules and Choice in Economics, Routledge, London.Williamson, O.E. () Markets and Hierarchies—Analysis and Anti-

trust Implications, Free Press, New York._______ () Transaction Cost Economics: The Governance of

Contractual Relations, Journal of Law and Economics ._______ () The Economic Institutions of Capitalism, Free Press,

New York.Wood, E.M. () The Origins of Capitalism, Monthly Review Press,

New York.Vromen, J.J () Economics and Evolution, An Inquiry into the

Foundations of the ‘New Institutional Economics’ Routledge, London.

new institutional economics and economic history · new institutional economics 9 new institutional...

Documents