William J. BaumolNew York University and Princeton University
The basic hypothesis is that, while the total supply of entrepreneursvaries among societies, the productive contribution of the societ\ 'sentrepreneurial acti\ ities varies much more because of their alloca-tion between producti\e activities such as innovation and largeKunproductive activities such as rent seeking or organized crime. Thisallocation is heavih influenced by the relati\e payoffs society offersto such activities. This implies that policv can influence the allocationof entrepreneurship more effectively than it can influence its supply.Historical evidence from ancient Rome, early China, and the MiddleAges and Renaissance in Europe is used to investigate the hy-potheses.
It is often assumed that an economy of private enter-prise has an automatic bias towards innovation, but thisis not so. It has a bias onh towards profit. [HOBSBAWM
1969, p. 40]
When conjectures are offered to explain historic slowdowns or greatleaps in economic growth, there is the group of usual suspects that is
I am very grateful for the generous support of the research underlying this paperfrom the Di\ision of Information Science and Technolog\ of the National ScienceFoundation, the Price Institute for Entrepreneurial Studies, the Center for Entrepre-neurial Studies of the Graduate School of Business Administration, New \ork Univer-sity, and the C. V. Starr Center for Applied Economics. I am also very much indebtedto Vacharee Devakula for her assistance in the research. I owe much to Joel Mok\ r,Stefano Fenoaltea, Lawrence Stone, Constance Berman, and Claudia Goldin for helpwith the substance of the paper and to William Jordan and Theodore Rabb for guid-ance on references.
regularly rounded up—prominent among them, the entrepreneur.Where growth has slowed, it is implied that a decline in entrepre-neurship was partly to blame (perhaps because the culture's "need forachievement" has atrophied). At another time and place, it is said, theflowering of entrepreneurship accounts for unprecedented expan-sion.
This paper proposes a rather different set of hypotheses, holdingthat entrepreneurs are always with us and always play some substantialrole. But there are a variety of roles among which the entrepreneur'sefforts can be reallocated, and some of those roles do not follow theconstructive and innovative script that is conventionally attributed tothat person. Indeed, at times the entrepreneur may even lead aparasitical existence that is actually damaging to the economy. Howthe entrepreneur acts at a given time and place depends heavily onthe rules of the game—the reward structure in the economy—thathappen to prevail. Thus the central hypothesis here is that it is the setof rules and not the supply of entrepreneurs or the nature of theirobjectives that undergoes significant changes from one period to an-other and helps to dictate the ultimate effect on the economy via theallocation of entrepreneurial resources. Changes in the rules andother attendant circumstances can, of course, modify the compositionof the class of entrepreneurs and can also alter its size. Without deny-ing this or claiming that it has no significance, in this paper I shall seekto focus attention on the allocation of the changing class of entrepre-neurs rather than its magnitude and makeup. (For an excellent analy-sis of the basic hypothesis, independently derived, see Murphy,Shleifer, and Vishny [1990].)
The basic proposition, if sustained by the evidence, has an impor-tant implication for growth policy. The notion that our productivityproblems reside in "the spirit of entrepreneurship" that waxes andwanes for unexplained reasons is a counsel of despair, for it gives noguidance on how to reawaken that spirit once it has lagged. If that isthe task assigned to policymakers, they are destitute: they have nomeans of knowing how to carry it out. But if what is required is theadjustment of rules of the game to induce a more felicitous allocationof entrepreneurial resources, then the policymaker's task is less for-midable, and it is certainly not hopeless. The prevailing rules thataffect the allocation of entrepreneurial activity can be observed, de-scribed, and, with luck, modified and improved, as will be illustratedhere.
Here, extensive historical illustrations will be cited to impart plausi-bility to the contentions that have just been described. Then a shortdiscussion of some current issues involving the allocation of entrepre-neurship between productive and unproductive activities will be of-
ENTREPRENEURSHIP 895
fered. Finally, I shall consider very briefly the means that can be usedto change the rules of the game, and to do so in a manner thatstimulates the productive contribution of the entrepreneur.
I. On the Historical Character of the Evidence
Given the inescapable problems for empirical as well as theoreticalstudy of entrepreneurship, what sort of evidence can one hope toprovide? Since the rules of the game usually change very slowly, a casestudy approach to investigation of my hypotheses drives me una\ oid-ably to examples spanning considerable periods of historv and en-compassing widely different cultures and geographic locations. HereI shall proceed on the basis of historical illustrations encompassingall the main economic periods and places (ancient Rome, medievalChina, Dark Age Europe, the Later Middle Ages, etc.) that the eco-nomic historians almost universallv single out for the light thev shedon the process of innovation and its diffusion. These will be used toshow that the relative rewards to different tvpes of entrepreneurialactivity have in fact varied dramatically from one time and place toanother and that this seems to have had profound effects on patternsof entrepreneurial beha\ior. Finallv, evidence will be offered suggest-ing that such reallocations can have a considerable influence on theprosperity and growth of an economy, though other variables un-doubtedly also play substantial roles.
None of this can, of course, be considered conclusive. Vet, it issurely a standard tenet of scientific method that tentative confirma-tion of a hypothesis is provided by observation of phenomena that thehypothesis helps to explain and that could not easily be accounted forif that hypothesis were invalid. It is on this sort of reasoning that Ihope to rest m\ case. Historians have long been puzzled, for example,by the failure of the societv of ancient Rome to disseminate and putinto widespread practical use some of the sophisticated technologicaldevelopments that we know to have been in its possession, while in the"High Middle Ages," a period in which progress and change werehardly popular notions, inventions that languished in Rome seem tohave spread like wildfire. It will be argued that the hypothesis aboutthe allocability of entrepreneurial effort between productive and un-productive activity helps considerably to account for this phenome-non, though it certainly w ill ?wt be claimed that this is all there was tothe matter.
Before I get to the substance of the discussion, it is important toemphasize that nothing that follows in this article makes any pretenseof constituting a contribution to economic history. Certainlv it is notintended here to try to explain any particular historical event. More-
896 JOURNAL OF POLITICAL ECONOMY
over, the analysis relies entirely on secondary sources, and all thehistorical developments described are well known to historians, as thecitations will indicate. Whatever the contribution that may be offeredby the following pages, then, it is confined to enhanced understand-ing and extension of the (nonmathematical) theory of entrepre-neurship in general, and not to an improved analysis of the historicalevents that are cited.
II. The Schumpeterian Model Extended:Allocation of Entrepreneurship
The analysis of this paper rests on what seems to be the one theoreti-cal model that effectively encompasses the role of the entrepreneurand that really "works," in the sense that it constitutes the basis for anumber of substantive inferences.^ This is, of course, the well-knownSchumpeterian analysis, whose main shortcoming, for our purposes,is the paucity of insights on policy that emerge from it. It will besuggested here that only a minor extension of that model to encom-pass the allocation of entrepreneurship is required to enhance itspower substantially in this direction.
Schumpeter tells us that innovations (he calls them "the carryingout of new combinations") take various forms besides mere improve-ments in technology:
This concept covers the following five cases: (1) the in-troduction of a new good—that is one with which consumersare not yet familiar—or of a new quality of a good. (2) Theintroduction of a new method of production, that is one notyet tested by experience in the branch of manufacture con-cerned, which need by no means be founded upon a discov-ery scientifically new, and can also exist in a new way ofhandling a commodity commercially. (3) The opening of anew market, that is a market into which the particularbranch of manufacture of the country in question has notpreviously entered, whether or not this market has existedbefore. (4) The conquest of a new source of supply of rawmaterials or half-manufactured goods, again irrespective ofwhether this source already exists or whether it has first to be
' There has, however, recently been an outburst of illuminating writings on thetheory of the innovation process, analyzing it in such terms as races for patents in whichthe winner takes everything, with no consolation prize for a close second, or treatingthe process, alternatively, as a "waiting game," in w hich the patient second entrant mayoutperform and even survive the first one in the innovative arena, who incurs the bulkof the risk. For an overview of these discussions as well as some substantial addedinsights, see Dasgupta (1988).
ENTREPRENEURSHIP 897
created. (5) The carrying out of the new organization of anyindustr\, like the creation of a monopolv position (for ex-ample through trustification) or the breaking up of a mo-nopoly position. [(1912) 1934, p. 66]
The obvious fact that entrepreneurs undertake such a varietv oftasks all at once suggests that theory can usefully undertake to con-sider what determines the allocation of entrepreneurial inputs amongthose tasks. Just as the literature traditionally studies the allocation ofother inputs, for example, capital resources, among the various in-dustries that compete for them, it seems natural to ask what m-fluences the flow of entrepreneurial talent among the various activi-ties in Schumpeter's list.
Presumablv the reason no such line of inquirv was pursued bySchumpeter or his successors is that any analvsis of the allocation ofentrepreneurial resources among the five items in the preceding list(with the exception of the last—the creation or destruction of a mo-nopoh) does not promise to yield any profound conclusions. There isno obvious reason to make much of a shift of entrepreneurial activityaway from, say, impro\ement in the production process and towardthe introduction of new products. The general implications, if any,for the public welfare, for producti\ it\ growth, and for other relatedmatters are hardlv obvious.
To deri\e more substanti\e results from an anahsis of the alloca-tion of entrepreneurial resources, it is necessar\ to expand Schumpe-ter's list, whose main deficiencv seems to be that it does not go farenough. For example, it does not explicith encompass innoxative actsof technolog\ transfer that take ad\antage of opportunities to in-troduce already-available technology (usually with some modificationto adapt it to local conditions) to geographic locales whose suitabilityfor the purpose had previouslv gone unrecognized or at least unused.
Most important for the discussion here, Schumpeter's list of entre-preneurial acti\ ities can usefulh be expanded to include such items asinnox ations in rent-seeking procedures, for example, discovery of apre\iously unused legal gambit that is effecti\e in di\erting rents tothose who are first in exploiting it. It may seem strange at first blush topropose inclusion of acti\ ities of such questionable \alue to societv (Ishall call them acts of "unproducti\ e entrepreneurship') in the list ofSchumpeterian innovations (though the creation of a monopoly,which Schumpeter does include as an innovation, is surely as ques-tionable), but, as will soon be seen, this is a crucial step for the analysisthat follows. If entrepreneurs are defined, simplv, to be persons whoare ingenious and creatixe in finding ways that add to their ownwealth, power, and prestige, then it is to be expected that not all of
JOURNAL OF POLITICAL ECONOMY
them will be overly concerned with whether an activity that achievesthese goals adds much or little to the social product or, for that mat-ter, even whether it is an actual impediment to production (this no-tion goes back, at least, to Veblen [1904]). Suppose that it turns out, inaddition, that at any time and place the magnitude of the benefit theeconomy derives from its entrepreneurial talents depends substan-tially, among other variables, on the allocation of this resource be-tween productive and unproductive entrepreneurial activities of thesorts just described. Then the reasons for including acts of the lattertype in the list of entrepreneurial activities become clear.
Here no exhaustive analysis of the process of allocation of entrepre-neurial activity among the set of available options will be attempted.Rather, it will be argued only that at least one of the prime determi-nants of entrepreneurial behavior at any particular time and place isthe prevailing rules of the game that govern the payoff of one entre-preneurial activity relative to another. If the rules are such as toimpede the earning of much wealth via activity A, or are such as toimpose social disgrace on those who engage in it, then, other thingsbeing equal, entrepreneurs' efforts will tend to be channeled to otheractivities, call them B. But if B contributes less to production or wel-fare than A, the consequences for society may be considerable.^
As a last preliminary note, it should be emphasized that the set ofactive entrepreneurs may be subject to change. Thus if the rules ofthe game begin to favor B over A, it may not be just the same individ-uals who switch their activities from entrepreneurship of type A tothat of type B. Rather, some persons with talents suited for A maysimply drop out of the picture, and individuals with abilities adaptedto B may for the first time become entrepreneurs. Thus the allocationof entrepreneurs among activities is perhaps best described in the wayJoan Robinson (following Shove's suggestion) analyzed the allocationof heterogeneous land resources (1933, chap. 8): as the solution of ajigsaw puzzle in which the pieces are each fitted into the places se-lected for them by the concatenation of pertinent circumstances.
III. Entrepreneurship, Productive andUnproductive: The Rules Do Change
Let us now turn to the central hypothesis of this paper: that theexercise of entrepreneurship can sometimes be unproductive or even
^ There is a substantial literature, following the work of Jacob Schmookler, providingstrong empirical evidence for the proposition that even the allocation of inventiveeffort, i.e., the directions pursued by inventive activities, is itself heavilv influenced byrelative payoff prospects. However, it is now agreed that some of these authors go toofar when they appear to imply that almost nothing but the demand for the product ofinvention influences to any great extent which inventions will occur. For a good sum-mary and references, see Abramovitz (1989. p. 33).
ENTREPRENEURSHIP 899
destructive, and that whether it takes one of these directions or onethat is more benign depends heavily on the structure of payoffs in theeconomy—the rules of the game. The rather dramatic illustrationsprovided by world history seem to confirm quite emphatically thefollowing proposition.
PROPOSITION 1. The rules of the game that determine the relativepayoffs to different entrepreneurial activities do change dramaticallyfrom one time and place to another.
These examples also suggest strongly (but hardly "prove") the fol-lowing proposition.
PROPOSITION 2. Entrepreneurial behavior changes direction fromone economy to another in a manner that corresponds to the varia-tions in the rules of the game.
A. Ancient Rome
The avenues open to those Romans who sought power, prestige, andwealth are instructive. First, it may be noted that they had no reserva-tions about the desirability of wealth or about its pursuit (e.g., Finley1985, pp. 53—57). As long as it did iwt involve participation in industry orcommerce, there was nothing degrading about the wealth acquisitionprocess. Persons of honorable status had three primary and accept-able sources of income: landholding (not infrequently as absenteelandlords), "usury," and what may be described as "political pay-ments":
The opportunity for "political moneymaking" can hardlybe over-estimated. Money poured in from booty, indem-nities, provincial taxes, loans and miscellaneous extractionsin quantities without precedent in Graeco-Roman history,and at an accelerating rate. The public treasury benefited,but probably more remained in private hands, among thenobles in the first instance; then, in appropriately decreasingproportions, among the equites, the soldiers and even theplebs of the city of Rome. . . . Nevertheless, the whole phe-nomenon is misunderstood when it is classified under theheadings of "corruption" and "malpractice", as historiansstill persist in doing. Cicero was an honest governor of Ciliciain 51 and 50 B.C., so that at the end of his term he hadearned only the legitimate profits of office. They amountedto 2,200,000 sesterces, more than treble the figure of600,000 he himself once mentioned {Stoic Paradoxes 49) toillustrate an annual income that could permit a life of luxury.We are faced with something structural in the society. [Finley1985, p. 55]
JOURNAL OF POLITICAL ECONOMY
Who, then, operated commerce and industry? According to Veyne(1961), it was an occupation heavily undertaken by freedmen—former slaves who, incidentally, bore a social stigma for life. Indeed,according to this writer, slavery may have represented the one avenuefor advancement for someone from the lower classes. A clever (andhandsome) member of the lower orders might deliberately arrange tobe sold into slavery to a wealthy and powerful master.^ Then, withluck, skill, and drive, he would grow close to his owner, perhapsmanaging his financial affairs (and sometimes engaging in somehomosexual activity with him). The master then gained cachet, after asuitable period, by granting freedom to the slave, setting him up witha fortune of his own. The freedmen, apparently not atypically, in-vested their financial stakes in commerce, hoping to multiply themsufficiently to enable them to retire in style to the countryside, there-after investing primarily in land and loans in imitation of the upperclasses.
Finally, regarding the Romans' attitude to the promotion of tech-nology and productivity, Finley makes much of the "clear, almosttotal, divorce between science and practice" (1965, p. 32). He goes onto cite Vitruvius's monumental work on architecture and technology,in whose 10 books he finds only a single and trivial reference to meansof saving effort and increasing productivity. Finley then reports thefollowing story:
There is a story, repeated by a number of Roman writers,that a man—characteristically unnamed—invented un-breakable glass and demonstrated it to Tiberius in anticipa-tion of a great reward. The emperor asked the inventorwhether anyone shared his secret and was assured that therewas no one else; whereupon his head was promptly re-moved, lest, said Tiberius, gold be reduced to the value ofmud. I have no opinion about the truth of this story, and it isonly a story. But is it not interesting that neither the elderPliny nor Petronius nor the historian Dio Cassius was trou-bled by the point that the inventor turned to the emperor fora reward, instead of turning to an investor for capital withwhich to put his invention into production?^ . . . We must
^ Stefano Fenoakea comments that he knows no documented cases in which thisoccurred and that it was undoubtedly more common to seek advancement throughadoption into an upper-class family.
^ To be fair to Finley, note that he concludes that it is not really interesting. North andThomas (1973, p. 3) make a similar point about Harrison's invention of the ship'schronometer in the eighteenth century (as an instrument indispensable for the deter-mination of longitude). They point out that the incentive for this invention was a largegovernmental prize rather than the prospect of commercial profit, presumably becauseof the absence of effective patent protection.
ENTREPRENEURSHIP 9OI
remind ourselves time and again that the European experi-ence since the late Middle Ages in technologv, in the econ-omy, and in the \alue s\stems that accompanied them, wasunique in human histor\ until the recent export trend com-menced. Technical progress, economic growth, productivity,even efficiencv have not been significant goals since the be-ginning of time. So long as an acceptable life-st\ le could bemaintained, however that was defined, other values held thestage. [1985, p. 147]
The bottom line, for our purposes, is that the Roman reward svs-tem, although it offered wealth to those who engaged in commerceand industrv, offset this gain through the attendant loss in prestige.Economic effort "was neither the wav to wealth nor its purpose. Cato'sgods showed him a number of wavs to get more: but the\ were allpolitical and parasitical, the wavs of conquest and booty and usury;labour w as not one of them, not even the labour of the entrepreneur"(Fmlev 1965. p. 39).
B. MedirL'al China
In China, as in man\ kingdoms of Europe before the guarantees ofthe Magna Carta and the re\ i\ al of tow ns and their acquisition ofprivileges, the monarch commonh claimed possession of all propertyin his territories. As a result, particularly in China, when the sover-eign w as in financial straits, confiscation of the propertv of wealthysubjects was entireh in order. It has been claimed that this led thosewho had resources to avoid investing them in any sort of visible capitalstocks, and that this, in turn, was a substantial impediment to eco-nomic expansion (see Balazs 1964, p. 53; Landes 1969, pp. 46—47;Rosenberg and Birdzell 1986, pp. 119-20; Jones 1987, chap. 5).
In addition, imperial China reser\ ed its most substantial rewards inwealth and prestige for those who climbed the ladder of imperialexaminations, which were heavilv devoted to subjects such as Confu-cian philosophy and calligraphv. Successful candidates were oftenawarded high rank in the bureaucracv, high social standing denied toanyone engaged in commerce or industry, even to those who gainedgreat wealth in the process (and who often used their resources toprepare their descendants to contend \ ia the examinations for a posi-tion in the scholar bureaucracv). In other words, the rules of the gameseem to have been heavilv biased against the acquisition of wealth ayidposition through Schumpeterian behavior. The avenue to success layelsewhere.
Because of the difficulty of the examinations, the mandarins(scholar-officials) rarely succeeded in keeping such positions in their
JOURNAL OF POLITICAL ECONOMY
own families for more than two or three generations (see Marsh 1961,p. 159; Ho 1962, chap. 4 and appendix). The scholar families devotedenormous effort and considerable resources to preparing their chil-dren through years of laborious study for the imperial examinations,which, during the Sung dynasty, were held every 3 years, and onlyseveral hundred persons in all of China succeeded in passing themeach time (E. A. Kracke, Jr. in Liu and Golas [1969, p. 14]). Yet,regularly, some persons not from mandarin families also attainedsuccess through this avenue (see, e.g.. Marsh [1961] and Ho [1962]for evidence on social mobility in imperial China).
Wealth was in prospect for those who passed the examination andwho were subsequently appointed to government positions. But thesources of their earnings had something in common with those of theRomans:
Corruption, which is widespread in all impoverished andbackward countries (or, more exactly, throughout the pre-industrial world), was endemic in a country where the ser-vants of the state often had nothing to live on but their verymeager salaries. The required attitude of obedience tosuperiors made it impossible for officials to demand highersalaries, and in the absence of any control over their activitiesfrom below it was inevitable that they should purloin fromsociety what the state failed to provide. According to theusual pattern, a Chinese official entered upon his duties onlyafter spending long years in study and passing many exami-nations; he then established relations with protectors, in-curred debts to get himself appointed, and then proceededto extract the amount he had spent on preparing himself forhis career from the people he administered—and extractedboth principal and interest. The degree of his rapacity wouldbe dictated not only by the length of time he had had to waitfor his appointment and the number of relations he had tosupport and of kin to satisfy or repay, but also by the pre-cariousness of his position. [Balazs 1964, p. 10]
Enterprise, on the other hand, was not only frowned on, but mayhave been subjected to impediments deliberately imposed by theofficials, at least after the fourteenth century A.D.; and some histo-rians claim that it was true much earlier. Balazs tells us of
the state's tendency to clamp dowti immediately on any formof private enterprise (and this in the long run kills not onlyinitiative but even the slightest attempts at innovation), or, ifit did not succeed in putting a stop to it in time, to take over
ENTREPRENEURSHIP 903
and nationalize it. Did it not frequently happen during thecourse of Chinese history that the scholar-officials, althoughhostile to all inventions, nevertheless gathered in the fruits ofother people's ingenuity? I need mention only three exam-ples of inventions that met this fate: paper, invented by aeunuch; printing, used by the Buddhists as a medium forreligious propaganda; and the bill of exchange, an expedientof private businessmen. [P. 18]
As a result of recurrent intervention by the state to curtail the libertyand take over any accumulated advantages the merchant class hadmanaged to gain for itself, "the merchant's ambition turned to becom-ing a scholar-official and investing his profits in land" (p. 32).
C. The Earlier Middle Ages
Before the rise of the cities and before monarchs were able to subduethe bellicose acti\ ities of the nobilit), wealth and power were pursuedprimarilv through military acti\itv. Since land and castles were themedieval forms of wealth most highlv valued and most avidly soughtafter, it seems reasonable to interpret the warring of the barons ingood part as the pursuit of an economic objecti\e. For example, dur-ing the reign of William the Conqueror (see, e.g., Douglas 1964),there were frequent attempts by the barons in Normandy and neigh-boring portions of France to take over each other's lands and castles.A prime incentive for William's supporters in his conquest of Englandwas their obvious aspiration for lands."^ More than that, violent meansalso served to provide more liquid forms of income (captured trea-sure), which the nobilitv used to support both private consumptionand in\estment in military plant and equipment, where such itemscould not easily be produced on their own lands and therefore had tobe purchased from others. In England, with its institution of primo-geniture (the exclusive right of the eldest son to inherit his father'sestate), younger sons who chose not to enter the clergv* often had nosocially acceptable choice other than warfare as a means to make theirfortunes, and in some cases they succeeded spectacularly. Thus notethe case of William Marshal, fourth son of a minor noble, who rose
^ The conquest has at least two noteworthy entrepreneurial sides. First, it involved aninnovation, the use of the stirrup bv the Normans at Hastings that enabled William'swarriors to use the same spear to impale a series of \ ictims with the force of the horse'scharge, rather than just tossing the spear at the enem\, much as an infantrvman could.Second, the invasion was an impressi\e act of organization, with William having toconvince his untrustworthy allies that they had more to gain by joining him in Englandthan by staying behind to profit from his absence by trying to grab awav his lands asthey had tried to do many times before.
JOURNAL OF POLITICAL ECONOMY
through his military accomplishments to be one of the most powerfuland trusted officials under Henry II and Richard I, and became oneof the wealthiest men in England (see Painter 1933).
Of course, the medieval nobles were not purely economic men.Many of the turbulent barons undoubtedly enjoyed fighting for itsown sake, and success in combat was an important avenue to prestigein their society. But no modern capitalist is a purely economic maneither. What I am saying here is that warfare, which was of coursepursued for a variety of reasons, was also undertaken as a primarysource of economic gain. This is clearly all the more true of themercenary armies that were the scourge of fourteenth-centuryFrance and Italy.
Such violent economic activity, moreover, inspired frequent andprofound innovation. The introduction of the stirrup was a requisitefor effective cavalry tactics. Castle building evolved from wooden tostone structures and from rectangular to round towers (which couldnot be made to collapse by undermining their corners). Armor andweaponry became much more sophisticated with the introduction ofthe crossbow, the longbow, and, ultimately, artillery based on gun-powder. Military tactics and strategy also grew in sophistication.These innovations can be interpreted as contributions of military en-trepreneurs undertaken at least partly in pursuit of private economicgains.
This type of entrepreneurial undertaking obviously differs vastlyfrom the introduction of a cost-saving industrial process or a valuablenew consumer product. An individual who pursues wealth throughthe forcible appropriation of the possessions of others surely does notadd to the national product. Its net effect may be not merely a trans-fer but a net reduction in social income and wealth.^
*^ In saying all this, I must not be interpreted as taking the conventional view thatwarfare is an unmitigated source of impoverishment of any economy that unquestion-ably never contributes to its prosperity. Careful recent studies have indicated thatmatters are more complicated (see, e.g., Milward 1970; Olson 1982). Certainly theunprecedented prosperity enjoyed afterward by the countries on the losing side of theSecond World War suggests that warfare need not always preclude economic expan-sion, and it is easy to provide earlier examples. The three great economic leaders of theWestern world preceding the United States—Italy in the thirteenth-sixteenth cen-turies, the Dutch Republic in the seventeenth and eighteenth, and Great Britain in theeighteenth and nineteenth—each attained the height of their prosperity after periodsof enormously costly and sometimes destructive warfare. Nevertheless, the wealthgained by a medieval baron from the adoption of a novel bellicose technique can hardlyhave contributed to economic growth in the way that resulted from adoption of a newsteelmaking process in the nineteenth century or the introduction of a product such asthe motor vehicle in the twentieth.
ENTREPRENEURSHIP 9O5
D. The Later Middle Ages
By the end of the eleventh centurv the rules of the game had changedfrom those of the Dark Ages. The revival of the towns was well underway. They had acquired a number of privileges, among them protec-tion from arbitrary taxation and confiscation and the creation of alabor force by granting freedom to runaway serfs after a relativelybrief residence (a year and a day) in the towns. The free-enterpriseturbulence of the barons had at least been impeded by the church'spacification efforts: the peace and the (later) truce of God in France,Spain, and elsewhere; similar changes were taking place in England(see, e.g., Cowdrev [1970]; but Jones [1987, p. 94] suggests that somefree-enterprise military activity bv the barons continued in Englandthrough the reigns of the earlier Tudors in the sixteenth centurv). Allthis subsequently "gave way to more developed efforts to enforcepeace b\ the more organized governments of the twelfth century"(Brooke 1964, p. 350; also p. 127). A number of activities that wereneither agricultural nor militarv began to yield handsome returns.For example, the small group of architect-engineers who were incharge of the building of cathedrals, palaces, bridges, and fortressescould live in great luxury in the service of their kings.
But, apparently, a far more common source of earnings was thewater-driven mills that were strikingly common in France and south-ern England by the eleventh century, a technological innovationabout which more will be said presenth. An incentive for such techni-cal advances mav have been the monopoly thev conferred on theirowners rather than any resulting improvement in efficiency. Suchmonopoly rights were alike sought and enforced by private parties(Bloch 1935, pp. 554—57; Brooke 1964, p. 84) and by religious or-ganizations (see below).
The economic role of the monks in this is somewhat puzzling—theleast clear-cut part of our story.' The Cistercian abbeys are generallyassigned a critical role in the promotion of such technological ad-yances. In some cases they simph took o\er mills that had been con-structed by others (Berman 1986, p. 89). But the Cistercians im-proyed them, built many others, and yastly expanded their use; at
' Bloch (1935) notes that the monasteries had both the capital and the large numberof consumers of flour necessar) to make the mills profitable. In addition, they were lesslikely than lav communities to undergo military siege, which, Bloch notes, was (besidesdrought and freezing of the waterways) one of the main impediments to adoption ofthe water mill, since blocking of the waterway that droye the mill could threaten thebesieged population with starvation (pp. 550-53).
JOURNAL OE POLITICAL ECONOMY
least some writers (e.g., Gimpel 1976, pp. 3-6) seem to suggest thatthe Cistercians were the spearhead of technological advance.
Historians tell us that they have no ready explanation for the entre-preneurial propensities of this monastic order. (See, e.g., Brooke[1964, p. 69] and also a personal communication to me from Con-stance Berman. Ovitt [1987, esp. pp. 142-47] suggests that this mayall have been part of the twelfth-century monastic drive to reduce oreliminate manual labor in order to maximize the time available for theless onerous religious labors—a conclusion with which Bloch [1935, p.553] concurs.) But the evidence suggests strongly that avid entrepre-neurs they were. They accumulated vast tracts of land; the sizes oftheir domesticated animal flocks were enormous by the standards ofthe time; their investment rates were remarkable; they sought to exer-cise monopoly power, being known, after the erection of a water mill,to seek legal intervention to prevent nearby residents from continu-ing to use their animal-powered facilities (Cimpel 1976, pp. 15-16);they were fierce in their rivalrous behavior and drive for expansion,in the process not sparing other religious bodies—not even otherCistercian houses. There is a "record of pastoral expansionism andmonopolies over access established by the wealthiest Cistercian houses. . . at the expense of smaller abbeys and convents . . . effectivelypushing out all other religious houses as competitors" (Berman 1986,p. 112).
As with early capitalists, the asceticism of the monks, by keepingdown the proportion of the monastery's output that was consumed,helped to provide the resources for levels of investment extraordi-nary for the period (pp. 40, 83). The rules of the game appear to haveoffered substantial economic rewards to exercise of Cistercian entre-preneurship. The order obtained relatively few large gifts, but in-stead frequently received support from the laity and from the churchestablishment in the form of exemptions from road and river tollsand from payment of the tithe. This obviously increased the marginalyield of investment, innovation, and expenditure of effort, and theevidence suggests the diligence of the order in pursuing the resultingopportunities. Their mills, their extensive lands, and their large flocksare reported to have brought scale economies and extraordinaryfinancial returns (chap. 4), Puritanical, at least in earlier years, in theirself-proclaimed adherence to simplicity in personal lifestyle while en-gaged in dedicated pursuit of wealth, they may perhaps represent anearly manifestation of elements of "the Protestant ethic." But what-ever their motive, the reported Cistercian record of promotion oftechnological progress is in diametric contrast to that of the Romanempire.
ENTREPRENEURSHIP 907
E. Fourteenth Centuiy
The fourteenth century brought with it a considerable increase inmilitary activity, notably the Hundred Years' War between Franceand England. Pa\offs, surely, mtist ha\e tilted to favor more thanbefore inventions designed for militarv purposes. Cannons appearedas siege devices and armor was made heavier. More imaginative wardevices were proposed: a windmill-propelled war wagon, a multibar-reled machine gun, and a di\ ing suit to permit underwater attacks onships. A pervasive business enterprise of this unhappy century of warwas the companv of mercenary troops—the condottiere—w ho roamedEurope, supported the side that could offer the most attractive terms,and in lulls between fighting, when unemployment threatened, wan-dered about thinking up military enterprises of their own, at theexpense of the general public (Gimpel 1976, chap. 9; see also McNeill1969, pp. 33-39). Clearly, the rules of the game—the system of entre-preneurial rew ards—had changed, to the disadvantage of productiveentrepreneurship.
F. Early Rent Seeking
Unproductive entrepreneurship can also take less violent forms, usu-ally involving various types of rent seeking, the tvpe of (possibly)unproductive entrepreneurship that seems most relevant today. En-terprising use of the legal system for rent-seeking purposes has a longhistory. There are, for example, records of the use of litigation in thetw elfth century in w hich the proprietor of a w ater-driven mill soughtand won a prohibition of use in the vicinity of mills driven bv animalor human power (Gimpel 1976, pp. 25—26). In another case, theoperators of tw o dams, one upstream of the other, sued one anotherrepeatedly at least from the second half of the thirteenth century untilthe beginning of the fifteenth, when the downstream dam finallysucceeded in driving the other out of business as the latter ran out ofmoney to pay the court fees (pp. 17-20).
In the upper strata of society, rent seeking also gradually replacedmilitary activity as a prime source of wealth and power. This transi-tion can perhaps be ascribed to the triumph of the monarchies andthe consequent imposition of law and order. Rent-seeking entrepre-neurship then took a variety of forms, notably the quest for grants ofland and patents of monopoly from the monarch. Such activities can,of course, sometimes prove to contribute to production, as when therecipient of land given by the monarch uses it more efficiently thanthe previous owner did. But there seems to have been nothing in the
JOURNAL OF POLITICAL ECONOMY
Structure of the land-granting process that ensured even a tendencytoward transfer to more productive proprietors, nor was the individ-ual who sought such grants likely to use as an argument in favor of hissuit the claim that he was likely to be the more productive user (interms of, say, the expected net value of its agricultural output).
Military forms of entrepreneurship may have experienced a renais-sance in England in the seventeenth century with the revolt againstCharles I. How that may have changed the structure of rewards toentrepreneurial activity is suggested by Hobsbawm (1969), who claimsthat at the end of the seventeenth century the most affluent mer-chants earned perhaps three times as much as the richest "mastermanufacturers."^ But, he reports, the wealthiest noble families proba-bly had incomes more than 10 times as large as those of the richmerchants. The point in this is that those noble families, according toHobsbawm, were no holdovers from an ancient feudal aristocracy;they were, rather, the heirs of the Roundheads (the supporters of theparliamentary, or puritan, party) in the then-recent Civil War (pp.30—32). On this view, once again, military activity would seem to havebecome the entrepreneur's most promising recourse.
But other historians take a rather different view of the matter.Studies reported in Thirsk (1954) indicate that ultimately there waslittle redistribution of property as the result of the Civil War and therestoration. Rather it is noted that in this period the "patrician elitesdepended for their political power and economic prosperity on royalcharters and monopolies rather than on talent and entrepreneurialinitiative" (Stone 1985, p. 45). In this interpretation of the matter, itwas rent seeking, not military activity, that remained the prime sourceof wealth under the restoration.
By the time the eighteenth-century industrial revolution ("the" in-dustrial revolution) arrived, matters had changed once again. Ac-cording to Ashton (1948, pp. 9—10), grants of monopoly were in goodpart "swept away" by the Monopolies Act of 1624, and, we are told byAdam Smith (1776), by the end of the eighteenth century they wererarer in England than in any other country. Though industrial activ-ity continued to be considered somewhat degrading in places in whichindustry flourished, notably in England during the industrial revolu-tion there was probably a difference in degree. Thus Lefebvre (1947,p. 14) reports that "at its upper level the [French] nobility . . . wereenvious of the English lords who enriched themselves in bourgeois
® The evidence indicates that the wealth of affluent families in Great Britain con-tinues to be derived preponderantly from commerce rather than from industry. Thiscontrasts with the record for the United States, where the reverse appears to be true(see Rubinstein 1980, pp. 22-23. 59-60).
ENTREPREXEURSHIP 9O9
wavs," w hile in France "the noble 'derogated' or fell into the commonmass if [like Mirabeau] he followed a business or profession" (p. 11).(See, however, Schama [1989], who tells us that 'even a cursorv exam-ination of the eighteenth-centurv French economv . . . reveals thenobilitv deeply involved in finance, business and industrv—certainlvas much as their British counterparts. . . . In 1765 a royal edictofficially removed the last formal obstacles to their participation intrade and industr)" [p. 118].) In England, primogeniture, by forcingyounger sons of noble families to resort to commerce and industry,apparentlv was imparting respectabilitv to these activities to a degreethat, while rather limited, mav have rarelv been paralleled before.
The central point of all the preceding discussion seems clear—perhaps, in retrospect, self-evident. If entrepreneurship is the im-aginative pursuit of position, with limited concern about the meansused to achieve the purpose, then we can expect changes in the struc-ture of rewards to modify the nature of the entrepreneur's acti\ ities,sometimes drastically. The rules of the game can then be a criticalinfluence helping to determine whether entrepreneurship will be al-located predominantlv to activities that are productive or unproduc-tive and even destructive.
IV. Does the Allocation between Productiveand Unproductive EntrepreneurshipMatter Much?
We come now to the third proposition of this article.PROPOSITION 3. The allocation of entrepreneurship between pro-
ductive and unproductive activities, though bv no means the onlvpertinent influence, can have a profound effect on the innovativenessof the economy and the degree of dissemination of its technologicaldiscoveries.
It is hard to believe that a svstem of pavoffs that moves entrepre-neurship in unproductive directions is not a substantial impedimentto industrial innovation and growth in productivitv. Still, historv per-mits no test of this proposition through a set of anvthing resemblingcontrolled experiments, since other influences did, undoubtedly, alsoplay important roles, as the proposition recognizes. One can onlv notewhat appears to be a remarkable correlation between the degree towhich an economv rewarded productive entrepreneurship and thevigor shown in that economy's innovation record.
Historians tell us of several industrial "near revolutions" that oc-curred before the industrial revolution of the eighteenth century thatare highly suggestive for our purposes (Braudel [1986, 3:542-56];for a more skeptical view, see Coleman [1956]). We are told that two
91O JOURNAL OF POLITICAL ECONOMY
of the incipient revolutions never went anywhere, while two of themwere rather successful in their fashion. I shall report conclusions ofsome leading historians on these episodes, but it should be recognizedby the reader that many of the views summarized here have beendisputed in the historical literature, at least to some degree.
A. Rome and Hellenistic Egypt
My earlier discussion cited ancient Rome and its empire as a case inwhich the rules did not favor productive entrepreneurship. Let uscompare this with the evidence on the vigor of innovative activity inthat society. The museum at Alexandria was the center of technologi-cal innovation in the Roman empire. By the first century B.C., that cityknew of virtually every form of machine gearing that is used today,including a working steam engine. But these seem to have been usedonly to make w hat amounted to elaborate toys. The steam engine wasused only to open and close the doors of a temple.
The Romans also had the water mill. This may well have been themost critical pre-eighteenth-century industrial invention because(outside the use of sails in transportation by water) it provided thefirst significant source of power other than human and animal labor:"it was able to produce an amount of concentrated energy beyond anyother resource of antiquity" (Forbes 1955, 2:90). As steam did inmore recent centuries, it offered the prospect of providing the basisfor a leap in productivity in the Roman economy, as apparently itactually did during the eleventh, twelfth, and thirteenth centuries inEurope. Yet Finley (1965, pp. 35-36), citing White (1962), reportsthat "though it was invented in the first century B.C., is was not untilthe third century A.D. that we find evidence of much use, and not untilthe fifth and sixth of general use. It is also a fact that we have noevidence at all of its application to other industries [i.e., other thangrinding of grain] until the very end of the fourth century, and thenno more than one solitary and possibly suspect reference . . . to amarble-slicing machine near Trier."
Unfortunately, evidence of Roman technical stagnation is onlyspotty, and, further, some historians suggest that the historical re-ports give inadequate weight to the Roma« preoccupation with ag-ricultural improvement relative to improvement in commerce ormanufacture. Still, the following quotation seems to summarize theweight of opinion: "Historians have long been puzzled as to why thelandlords of the Middle Ages proved so much more enterprising thanthe landlords of the Roman Empire, although the latter, by and large,were much better educated, had much better opportunities for mak-ing technical and scientific discoveries if they had wished to do so"
ENTREPRENEURSHIP 9 I I
(Brooke 1964, p. 88). It seems at least plausible that some part of theexplanation is to be found in the ancient world's rules of the game,which encouraged the pursuit of wealth but severely discouraged itspursuit through the exercise of productive entrepreneurship.^
B. Medieval China
The spate of inventions that occurred in ancient China (before it wasconquered by the barbarian Yuan dynasty in 1280) constituted one ofthe earliest potential revolutions in industry. Among the many Chi-nese technological contributions, one can list paper, (perhaps) thecompass, waterwheels, sophisticated water clocks, and, of course, gun-powder. Yet despite the apparent prosperitv of the Sung period(960-1270) (see, e.g., Eiu and Golas 1969), at least some historianssuggest that none of this spate of inventions led to a flowering ofindustry^^^ as distinguished from commerce and some degree of gen-eral prosperity. And in China too, as we have seen, the rules did notfavor productive entrepreneurship. Balazs (1964, p. 53) concludesthat
what was chieflv lacking in China for the further develop-ment of capitalism was not mechanical skill or scientific ap-titude, nor a sufficient accumulation of wealth, but scope forindividual enterprise. There was no individual freedom andno security for private enterprise, no legal foundation forrights other than those of the state, no alternative investmentother than landed property, no guarantee against beingpenalized by arbitrary exactions from officials or against in-tervention by the state. But perhaps the supreme inhibiting
^ It has been suggested by historians (see, e.g., Bloch 1935, p. 547) that an abundanceof slaves played a key role in Roman failure to use the water mill widely. Howeyer, thismust imply that the Romans were not efficient wealth seekers. As the cliometric litera-ture has made clear, the cost of maintaining a sla\ e is not low and certainly is not zero,and slayes are apt not to be efficient and dedicated workers. Thus if it had beenefficient to replace human or animal power by the inanimate power of the waterways,failure to do so would haye cut into the wealth of the slayeholder. in effect saddling himwith the feeding of unproductiye persons or keeping the slayes who turned the millsfrom other, more lucratiye, occupations. Perhaps Roman landowners were fairly un-sophisticated in the management of their estates, as Finlev (1985, pp. 108-16) suggests,and, if so, there may be some substance to the h\pothesis that slayers goes far toaccount for the failure of water mills to spread in the Roman economy.
'° Also, as in Rome, none of this was associated with the emergence of a systematicbody of science inyoKing coherent theoretical structure and the systematic testingof hypotheses on the basis of experiment or empirical obser\ation. Here, too, thethirteenth-century work of Bishop Grosseteste, William of Henley, and Roger Baconwas an early step toward that unique historical phenomenon—the emergence of asystematic body of science in the West in, say, the sixteenth century (see Needham1956).
9 ^ 2 JOURNAL OE POLITICAL ECONOMY
factcjr was the overwhelming prestige of the state bureau-cracy, which maimed from the start any attempt of thebourgeoisie to be different, to become aware of themselvesas a class and fight for an autonomous position in society.Free enterprise, ready and proud to take risks, is thereforequite exceptional and abnormal in Chinese economic his-tory.
C. Slow Growth in the ''Dark Ages"
An era noted for its slow growth occurred between the death ofCharlemagne (814) and the end of the tenth century. Even this periodwas not without its economic advances, which developed slowly, in-cluding the beginnings of the agricultural improvements that at-tended the introduction of the horseshoe, harness, and stirrup, theheavy plow, and the substitution of horsepow^er for oxen, w hich mayhave played a role in enabling peasants to move to more populousvillages further from their fields (see White 1962, p. 39 ff.). But, still,it was probably a period of significantly slower growth than the indus-trial revolution of the eleventh—thirteenth centuries (Gimpel 1976),about which more will be said presently. We have already seen thatthis was a period in which military violence was a prime outlet forentrepreneurial activity. While this can hardly pretend to be the expla-nation of the relative stagnation of the era, it is hard to believe that itwas totally unimportant.
D. The ''High Middle Ages"
A good deal has already been said about the successful industrialrevolution (and the accompanying commercial revolution sparked byinventions such as double-entry bookkeeping and bills of exchange[de Roover 1953]) of the late Middle Ages, whose two-century dura-tion makes it as long-lived as our own (see Carus-Wilson 1941; White1962; Cimpel 1976).
Perhaps the hallmark of this industrial revolution was that remark-able source of productive power, the water mills, that covered thecountryside in the south of England and crowded the banks of theSeine in Paris (see, e.g., Gimpel 1976, pp. 3-6; Berman 1986, pp. 8 1 -89). The mills were not only simple grain-grinding devices but accom-plished an astonishing variety of tasks and involved an impressivevariety of mechanical devices and sophisticated gear arrangements.They crushed olives, ground mash for beer production, crushed clothfor papermaking, sawed lumber, hammered metal and woolens (aspart of the "fulling" process—the cleansing, scouring, and pressing of
ENTREPRENEURSHIP 913
woven woolen goods to make them stronger and to bring the threadscloser together), milled coins, polished armor, and operated the bel-lows of blast furnaces. Their mechanisms entailed man\ forms ofingenuitv. Gears were used to translate the vertical circular motion ofthe efficient form of the waterwheel into the horizontal circular mo-tion of the millstone. The cam (a piece attached, sav, to the axle of thewaterwheel, protruding from the axle at right angles to its axis ofrotation) served to lift a hammer and to drop it repeatedlv and auto-matically (it was apparentlv known in antiquitv, but mav not ha\ebeen used with waterwheels). A crank handle extending from the endof the axle transformed the circular motion of the wheel into the backand forth (reciprocating) motion required for sawing or the operationof bellows. The most sophisticated product of all this mechanical skilland knowledge was the mechanical clock, which appeared toward theend of the thirteenth centurv. As White (1962, p. 129) sums up thematter, 'the four centuries follo\s ing Leonardo, that is. until electricalenerg\ demanded a supplementarv set of devices, were less techno-logicalh engaged in discovering basic principles than in elaboratingand refining those established during the four centuries beforeLeonardo." ^
In a period in which agriculture probablv occupied some 90 per-cent of the population, the expansion of industrv in the twelfth andthirteenth centuries could not b\ itself ha\e created a major upheavalin Ii\ing standards.^" Moreover, it has been deduced from what littlewe know of European gross domestic product per capita at the begin-ning of the eighteenth centurv that its average growth in the preced-ing six or seven centuries must have been verv modest, since if thepoverty of that later time had represented substantial growth from
' ' As was alreadv noted, science and scientific method also began to make an appear-ance with contributions such as those of Bishop Grosseteste and Roger Bacon. U'alterof Henle\ championed controlled experiments and obser\ ation over recourse to theopinions of ancient authorities and made a clear distinction between economic andengineering efficiencN in discussing the advisabilitv of substituting horses for oxen.Bacon displa\ed remarkable foresight when he wrote, circa 1260, that "machines maybe made b\ which the largest ships, with onlv one man steering them, will be movedfaster than if the\ were filled with rowers; wagons ma\ be built which will mo\e withincredible speed and without the aid of beasts; fl\ing machines can be constructed inwhich a man . . . ma\ beat the air with wings like a bird . . . machines will make itpossible to go to the bottom of seas and rivers" (as quoted in White [1962, p. 134]).
'" But then, much the same was true of the first half centun of "our" industrialrevolution, which, until the coming of the railwavs, was centered on the production ofcotton that perhaps constituted onl\ some 7—8 percent of national output (Hobsbawm1969, p. 68). Initiallv, the eighteenth-century industrial revolution was a \er} minoraffair, at least in terms of investment levels and contributions to output and to grow thin productivity (perhaps 0.3 percent per vear) (see Landes 1969, pp. 64-65; Feinstein1978, pp. 40-41; Williamson 1984).
9^4 JOURNAL OF POLITICAL ECONOMY
eleventh-century living standards, much of the earlier populationwould surely have been condemned to starvation.
Still, the industrial activity of the twelfth and thirteenth centurieswas very substantial. By the beginning of the fourteenth century,according to Gimpel (1976), 68 mills were in operation on less thanone mile of the banks of the Seine in Paris, and these were supple-mented by floating mills anchored to the Grand Pont. The activity inmetallurgy was also considerable—sufficient to denude much ofEurope of its forests and to produce a rise in the price of wood thatforced recourse to coal (Nef [1934]; other historians assert that thisdid not occur to any substantial degree until the fifteenth or sixteenthcentury, with some question even about those dates; see, e.g.. Cole-man [1975, pp. 42-43]). In sum, the industrial revolution of thetwelfth and thirteenth centuries was a surprisingly robust affair, andit is surely plausible that improved rewards to industrial activity hadsomething to do with its vigor.
E. The Fourteenth-Century Retreat
The end of all this period of buoyant activity in the fourteenth cen-tury (see the classic revisionist piece by Lopez [1969] as well as Gimpel[ 1976, chap. 9]) has a variety of explanations, many of them having noconnection with entrepreneurship. For one thing, it has been de-duced by study of the glaciers that average temperatures dropped,possibly reducing the yield of crops (though recent studies indicatethat the historical relation between climatic changes and crop yields isat best ambiguous) and creating other hardships. The plague re-turned and decimated much of the population. In addition to thesedisasters of nature, there were at least two pertinent developments ofhuman origin. First, the church clamped down on new ideas andother manifestations of freedom. Roger Bacon himself was put underconstraint. ^ The period during which new ways of thinking broughtrewards and status was apparently ended. Second, the fourteenthcentury included the first half of the devastating Hundred Years'War. It is implausible that the associated renewal of rewards to mili-tary enterprise played no part in the economic slowdown.
F. Remark on "Our'' Industnal Revolution
It need hardly be added, in conclusion, that the industrial revolutionthat began in the eighteenth century and continues today has brought
'^ The restraints imposed by the church had another curious effect: they apparentlymade bathing unfashionable for centuries. Before then, bathhouses had been popularas centers for social and, perhaps, sexual activity; but by requiring separation of thesexes and otherwise limiting the pleasures of cleanliness, the church undermined theinducements for such sanitary activities (see Gimpel 1976, pp. 87-92).
ENTREPRENEURSHIP 915
to the industrialist and the businessperson generally a degree ofwealth and a respect probably unprecedented in human historv. Thefact that this period yielded an explosion of output at least equallyunprecedented is undoubtedly attributable to a mxriad of causes thatcan probably never be discovered fullv and whose roles can ne\ er bedisentangled. Yet the continued association of output growth withhigh financial and respectabilit\ rewards to productixe entrepre-neurship is surelv suggesti\e, e\en if it can hardh be taken to beconclusive evidence for proposition 3, which asserts that the allocationof entrepreneurship does really matter for the vigor and innovative-ness of an economy.
V. On Unproductive Avenues for Today'sEntrepreneur: A Delicate Balance
Todav, unproductive entrepreneurship takes manv forms. Rent seek-ing, often via activities such as litigation and takeovers, and tax eva-sion and avoidance efforts seem now to constitute the prime threat toproductive entrepreneurship. The spectacular fortunes amassed bvthe "arbitrageurs'" revealed bv the scandals of the mid-1980s weresometimes, surely, the reward of unproductive, occasionalh illegal butentrepreneurial acts. Corporate executives devote much of their timeand energ\ to legal suit and countersuit, and litigation is used to bluntor prevent excessi\e vigor in competition bv rivals. Huge awards bythe courts, sometimes amounting to billions of dollars, can bring pros-perity to the victor and threaten the loser with insohencv. When thishappens, it must become tempting for the entrepreneur to select hisclosest advisers from the law vers rather than the engineers. It inducesthe entrepreneur to spend literalh hundreds of millions of dollars fora single legal battle. It tempts that entrepreneur to be the first tosue others before those others can sue him. (For an illuminatingquantification of some of the social costs of one widelv publicized legalbattle between two firms, see Summers and Cutler [1988].)
Similarly, taxes can ser\e to redirect entrepreneurial effort. AsLindbeck (1987, p. 15) has observed, "the problem with high-tax soci-eties is not that it is impossible to become rich there, but that it isdifficult to do so by w ay of productive effort in the ordinary produc-tion system." He cites as examples of the resulting reallocation ofentrepreneurship " 'smart' speculative financial transactions withoutmuch (if any) contribution to the productive capacity of the economy"(p. 15) as well as "illegal 'business areas' such as drug dealing" (p. 25).
In citing such activities, I do not mean to imply either that rent-seeking activity has been expanding in recent decades or thattakeover bids or private antitrust suits are ahvays or even preponder-antly unproductive. Rather, I am only suggesting where current rent-
9l6 JOURNAL OF POLITICAL ECONOMY
seeking activities are likely to be found, that is, where policy designersshould look if they intend to divert entrepreneurial talents into moreproductive channels.
The main point here is to note that threats of takeovers are some-times used as a means to extract "greenmail" and that recourse to thecourts as a means to seek to preserve rents through legally imposedimpediments to competition does indeed occur, and to suggest that itis no rare phenomenon. This does, then, become an attraction forentrepreneurial talent whose efforts are thereby channeled into un-productive directions. Yet, to the extent that takeovers disciplineinefficient managements and that antitrust intervention sometimes islegitimate and sometimes contributes to productivity, it would seemthat it will not be easy to change the rules in a way that discouragesallocation of entrepreneurial effort into such activities, without at thesame time undermining the legitimate role of these institutions. Somepromising proposals have been offered, but this is not a suitable placefor their systematic examination. However, a few examples will bereported in the following section.
VI. Changes in the Rules and Changes inEntrepreneurial Coals
A central point in this discussion is the contention that if reallocationof entrepreneurial effort is adopted as an objective of society, it is farmore easily achieved through changes in the rules that determinerelative rewards than via modification of the goals of the entrepre-neurs and prospective entrepreneurs themselves. I have even gone sofar as to use the same terms to characterize those goals in the verydifferent eras and cultures referred to in the discussion. But it wouldbe ridiculous to imply that the attitudes of a wealth-seeking senator inRome, a Sung dynasty mandarin, and an American industrialist ofthe late nineteenth century were all virtually identical. Still, the evi-dence suggests that they had more in common than might have beenexpected by the casual observer. However, even if it were to transpirethat they really diverged very substantially, that w ould be of little useto the designer of policy who does not have centuries at his or herdisposal and who is notoriously ineffective in engendering profoundchanges in cultural influences or in the structure of preferences. It isfor this reason that I have chosen to take entrepreneurial goals asgiven and to emphasize modification in the structure of the rewardsto different activities as the more promising line of investigation.
This suggests that it is necessary to consider the process by whichthose rules are modified in practice, but I believe that answers to eventhis more restricted question are largely beyond the powers of the
ENTREPRENEURSHIP 9 1 7
historians, the sociologists, and the anthropologists into whose do-mains it falls. One need only review the disputatious literature on theinfluences that led to the revival of trade toward the end of the earlyMiddle Ages to see how far we still are from anything resembling firmanswers. Exogenous influences such as foreign in\asions or unex-pected climatic changes can clearly play a part, as can deyelopmentswithin the economy. But the more interesting observation for ourpurposes is the fact that it is easy to think of measures that can changethese rules quicklv and profoundly.^"*
For example, the restrictions on royal grants of monopolies im-posed by Parliament in the Statute of Monopolies are said to havereduced substantiallv the opportunities for rent seeking in seven-teenth- and eighteenth-centurv England and mav have moved reluc-tant entrepreneurs to redirect their efforts toward agricultural im-provement and industry. Even if it did not succeed to anv substantialextent in reallocation of the efforts of an unchanged bodv of entre-preneurs from one of those types of activity to the other, if it in-creased failure rates among the rent seekers while not impedingothers who happened to prefer producti\ e pursuits, the result mighthave been the same. Similarly, tax rules can be used to rechannelentrepreneurial effort. It has, for instance, been proposed thattakeoyer actiyity would be reoriented substantially in directions thatcontribute to productiyity rather than impeding it bv a "revenue-neutral" modification in capital gains taxes that increases rates sharplyon assets held for short periods and decreases them considerabh forassets held, sav, for 2 years or more. A change in the rules thatrequires a plaintiff firm in a private antitrust suit to bear both parties'legal costs if the defendants are found not to be guilty (as is done inother countries) promises to reduce the frequency with which suchlawsuits are used in an attempt to hamper effective competition.
As has already been said, this is hardly the place for an extensivediscussion of the design of rational policy in the arena under consid-eration. The objective of the preceding brief discussion, rather, hasbeen to suggest that there are identifiable means by which the rules ofthe game can be changed effectiveh and to illustrate these meansconcretelv, though hardh attempting to offer any generalizationsabout their character. Certainly, the few illustrations that have justbeen offered should serve to confirm that there exist (in principle)
'"* Of course, that still leaves open the critical metaquestion. How does one go aboutchanging the society's value system so that it will want to change the rules? But that isnot the issue with which I am grappling here, since I see no basis on which the econo-mist can argue that society ought to change its values. Rather, I am positing a societywhose values lead it to favor productivity growth and am examining which instrumentspromise to be most effective in helping it to pursue this goal.
9l8 JOURNAL OF POLITICAL ECONOMY
testable means that promise to induce entrepreneurs to shift theirattentions in productive directions, without any major change in theirultimate goals. The testability of such hypotheses indicates that thediscussion is no tissue of tautologies, and the absence of references tothe allocability of entrepreneurship turned up in extensive search ofthe literature on the entrepreneur suggests that it was not entirelyself-evident.
VII. Concluding Comment
There is obviously a good deal more to be said about the subject;however, enough material has been presented to indicate that a minorexpansion of Schumpeter's theoretical model to encompass the deter-minants of the allocation of entrepreneurship among its competinguses can enrich the model considerably and that the hypotheses thathave been associated with the model's extension here are not withoutsubstance, even if none of the material approaches anything thatconstitutes a formal test of a hypothesis, much less a rigorous "proof."It is also easy to confirm that each of the hypotheses that have beendiscussed clearly yields some policy implications.
Thus clear guidance for policy is provided by the main hypothesis(propositions 1—3) that the rules of the game that specify the relativepayoffs to different entrepreneurial activities play a key role in deter-mining whether entrepreneurship will be allocated in productive orunproductive directions and that this can significanth affect the vigorof the economy's productivity growth. After all, the prevailing lawsand legal procedures of an economy are prime determinants of theprofitability of activities such as rent seeking via the litigative process.Steps such as deregulation of the airlines or more rational antitrustrules can do a good deal here.
A last example can, perhaps, nail down the point. The fact thatJapan has far fewer lawyers relative to population and far fewer law-suits on economic issues is often cited as a distinct advantage to theJapanese economy, since it reduces at least in part the quantity ofresources devoted to rent seeking. I he difference is often ascribedto national character that is said to have a cultural aversion tolitigiousness. This may all be very true. But closer inspection revealsthat there are also other influences. While in the United States legalinstitutions such as trebled damages provide a rich incentive for onefirm to sue another on the claim that the latter violated the antitrustlaws, in Japan the arrangements are very different. In that countryany firm undertaking to sue another on antitrust grounds must flrstapply for permission from the Japan Fair Trade Commission. But
ENTREPRENEURSHIP 919
such permission is rareh given, and, once denied, there is no legalavenue for appeal.
The overall moral, then, is that we do not have to wait patientlv forslow cultural change in order to find measures to redirect the flow ofentrepreneurial activitv toward more productive goals. As in the illus-tration of the Japanese just cited, it mav be possible to change therules in wavs that help to offset undesired institutional influences orthat supplement other influences that are taken to work in beneficialdirections.
References
Abramo\itz. Moses. Thinking about Growth, and Other Essays of Economic Growthand Welfare. Xew York: Cambridge Uni\. Press, 1989.
Ashton, Thomas S. The Industrial Rrcolution, 1760-1830. London: OxfordUniv. Press, 1948.
Balazs, Etienne. Ghmese Civilization ayidBureau^racy: Variations on a Theme. XewHa\en. Conn.: Yale Univ. Press. 1964.
Berman, Constance H. "Medieval Agriculture, the Southern French Coun-tryside, and the Earlv Cistercians: A Stud\ of Fortv-three Monasteries."Trans. American Philosophical Soc. 76, pt. 5 (1986).
Bloch, Marc. "Avenement et conquetes du moulin a eau." Annales d^HistoireEconojnique et Sociale 7 (November 1935): 538—63.
BraudeL Fernand. Civilization and Capitalism. 15th—18th Century. Vols. 2, 3.Xew York: Harper and Row, 1986.
Brooke. Christopher X. L. Europe in the Central Middle Ages, 962—1154. Lon-don: Longman, 1964.
Carus-W ilson, Eleanora M. An Industrial Revolution of the Thirteenth Cen-tun."Eron. Hist. Rrc. 11. no. 1 (1941): 39-60.
Coleman, Donald C. "Industrial Growth and Industrial Revolutions." £ro«o/«-/ra 23 (Februarv 1956): 1-22.
. Industry in Tudor and Stuait England. London: Macmillan (for Econ.Hist. Soc), 1975.
Cowdrev. H. E. J. T h e Peace and the Truce of God in the Eleventh Cen-turs . ' Past arid Present, no. 46 (Februarv 1970), pp. 42-67.
Dasgupta. Partha. "Patents. Prioritv and Imitation or. the Economics of Racesand Waiting Games." Econ.J. 98 (March 1988): 66-8(3.
de Roover, Ravmond. "The Commercial Revolution of the 13th Centurv." InEnterprise and Secular Change: Readings in Economic History, edited bv Fred-eric C. Lane and Jelle G. Riemersma. London: Allen and Unwin, 1953.
Douglas. Da\id G. William the Conqueror: The Xorrnan Impact upon England.Berkelev: Univ. Galifornia Press, 1964.
Feinstein, G. H. "Gapital Formation in Great Britain." In The Cambridge Eco-nomic History of Europe, vol. 8, pt. 1, edited bv Peter Mathias and M. M.Postan. Gambridge: Gambridge L ni\. Press, 1978.
Finlev, Moses I. "Technical Innovation and Economic Progress in the AncientWorld." Econ. Hist. Rei'. 18 (August 1965): 29-45.
. The Ancient Economy. 2d ed. London: Hogarth, 1985.Forbes. Robert J. Studies in Ancient Technology. Leiden: Brill, 1955.
92O JOURNAL OF POLITICAL ECONOMY
Gimpel, Jean. The Medieval Machine: The Indmtrial Revolution of the MiddleAges. New York: Holt, Reinhart and Winston, 1976.
Ho, Ping-Ti. The Ladder of Success in Imperial China, 1368-1911. New York:Columbia Univ. Press, 1962.
Hobsbawm, Eric J. Industry and Empire from 1750 to (he Preseyil Day. Har-mondsworth: Penguin, 1969.
Jones, Eric L. The European Miracle: Environments, Economies, and Geopolitics inthe History of Europe and Asia. Cambridge: Cambridge Univ. Press, 1987.
Landes, David S. The Unbound Prometheus: Technological Change and IndustrialDevelopment in Western Europe from 1750 to the Present. New York: CambridgeUniv. Press, 1969.
Lefebvre, Georges. The Coming of the French Revolution, 1789. Princeton, NJ.:Princeton Univ. Press, 1947.
Lindbeck, Assar. 'The Advanced Welfare State." Manuscript. Stockholm:Univ. Stockholm, 1987.
Liu, James T. C, and Golas, Peter J., eds. C^hange in Sung China: Innovation orRenovation? Lexington, Mass.: Heath, 1969.
Lopez, Robert S. "Hard Times and Investment in Culture." In The Renais-sance: A Symposium. New York: Oxford Univ. Press (for MetropolitanMuseum of Art), 1969.
McNeill, William H. History of Western Civilization. Rev. ed. Chicago: Univ.Chicago Press, 1969.
Marsh, Robert M. The Mandarins: The Circulation of Elites in China, 1600-1900.Glencoe, III.: Free Press, 1961.
Milward, Alan S. The Economic Effects of the Two World Wars on Britain. Lon-don: Macmillan (for Econ. Hist. Soc), 1970.
Murphy, Kevin M.; Shleifer, Andrei; and \ ishny, Robert. 'The Allocation ofTalent: Implications for Growth." Manuscript. Chicago: Univ. Chicago,1990.
Needham, Joseph. "Mathematics and Science in Ghina and the West." Scienceand Society 20 (Fall 1956): 320-43.
Nef, John U. 'The Progress of Technology and the Growth of Large-ScaleIndustry in Great Britain, 1540-1640." Econ. Hist. Rev. 5 (October 1934):3-24.
North, Douglass C, and Thomas, Robert Paul. The Rise of the Western World: ANew Economic History. Cambridge: Cambridge L'niv. Press, 1973.
Olson, Mancur. The Rise and Decline of Nations: Economic Growth, Stagflation,and Social Rigidities. New Haven, Gonn.: Yale Uni\. Press, 1982.
Ovitt, George, Jr. The Restoration of Perjection: Labor and Technology in MedievalCulture. New Brunswick, NJ.: Rutgers Univ. Press, 1987.
Painter, Sidney. William Marshal: Knight-Errant, Baron, and Regent of England.Baltimore: Johns Hopkins Press, 1933.
Robinson, Joan. The Economics of Imperfect Competition. London: Macmillan,1933.
Rosenberg, Nathan, and Birdzell, L. E., Jr. How the West Grew Rich: The Eco-nomic Transformation of the Industrial World. New York: Basic Books, 1986.
Rubinstein, W. D., ed. Wealth and the Wealthy in the Modern World. London:Groom Helm, 1980.
Schama, Simon. Citizens: A Chronicle of the Erench Revolution. New York:Knopf, 1989.
Schumpeter, Joseph A. The Theoiy of Economic Development. Leipzig: Dunckerand Humblot, 1912. English ed. Gambridge, Mass.: Harvard Univ. Press,1934.
ENTREPRENEURSHIP 92 1
Smith, Adam. An Inquiry into the Xature and Causes of the Wealth of Nations.1776. Reprint. Xew York: Random House (Modern Library), 1937.
Stone, Lawrence. "The Bourgeois Revolution of Seventeenth-Genturv En-gland Revisited." Past and Present, no. 109 (November 1985). pp. 44-54.
Summers, Lawrence, and Gutler. David. 'Texaco and Pennzoil Both LostBig." A'fu' York Times (Februarv 14, 1988).
Thirsk. Joan. ' T h e Restoration Land Settlement."y. Modern Hist. 26 (Decem-ber 1954): 315-28.
Veblen, Thorstein. The Theory of Business Enterprise. New York: Scribner,1904.
