The Successes and Failures of Professor Smith

Selected Papers No. 5 0

The Successes andFailures ofProfessor Smith

George J. Stigler

Graduate Schoolof Business

The Universityof Chicago

George J. Stigler is the Charles R. Walgreen Dis-tinguished Service Professor of American Institu-tions, and Director of the Walgreen Foundation,Graduate School of Business, at the Universityof Chicago.

Previously he served on the faculties of Colum-bia University, Brown University, the Universityof Minnesota and Iowa State College, and lec-tured at the London School of Economics.

A member of the Research Staff of the NationalBureau of Economic Research since 1941, he alsobelongs to the American Philosophical Society,the American Economic Association, the RoyalEconomic Society, the Econometric Society (Fel-low), and the American Statistical Association(Fellow), and he is a past president of theAmerican Economic Association.

In 1971 he was named Vice Chairman of theSecurities Investors Protective Commission, andin 1969-70 he was a member of the Blue RibbonDefense Panel. In 1960-61 he was Chairman ofthe Price Statistics Review Committee of theNational Bureau of Economic Research. In 1954-55 he was a member of the Attorney General’sCommittee for Study of Anti-Trust Laws.

He is the author of many articles in the field ofeconomics, as well as a number of books includ-ing which was first pub-lished in 1946; Capital and Rates of Return inManufacturing, 1963; The Intellectual and theMarket Place and Other Essays, 1963; Essays inthe History of Economics, 1965; The Organiza-tion of Industry, 1968; The Behavior of Indus-trial Prices (with J. K. Kindahl), 1970; andThe Citizen and the State, 1975.

“The Successes and Failures of Professor Smith”was presented at a meeting of the Mont PelerinSociety, St. Andrews, 1976. This selected paperis one of five for which Professor Stigler iswholly or jointly responsible. He and Paul A.Samuelson of the Massachusetts Institute ofTechnology were represented in Selected PaperNo. 7, “A Dialogue on the Proper EconomicRole of the State.” He is also the author ofSelected Paper No. 3, “The Intellectual andthe Market Place,” No. 13, “The Tactics of Eco-nomic Reform,” and No. 39, “Modern Man andHis Corporation.”

The Successes andFailures ofProfessor Smith

The Wealth of Nations appeared onMarch 9 of 1776, and perhaps sufficient timehas now passed to permit a fair estimateof Professor Smith’s triumphs and failures.It is a subject in which Smith himself wouldhave displayed a vivid and natural interest.John Rae recounts his final days:

When Smith felt his end to be approach-ing he evinced great anxiety to have allhis papers destroyed except the fewwhich he judged to be in a sufficientlyfinished state to deserve publication, andbeing apparently too feeble to undertakethe task himself, he repeatedly beggedhis friends Black and Hutton to destroythem for him . . . . Black and Huttonalways put off complying with Smith’sentreaties in the hope of his recoveringhis health or perhaps changing hismind; but at length, a week before hisdeath, he expressly sent for them, andasked them there and then to burnsixteen volumes of manuscript to whichhe directed them. This they did withoutknowing or asking what they contained.(Rae, 434.)

Only a man acutely sensitive to the opinionwhich posterity would hold of him wouldinsist upon such an act. Of course Smithwas wrong: there is no amount of mischiefand nonsense in sixteen volumes which wewould not have forgiven, especially since weknow he was given to neither mischiefnor nonsense.

If the time is ripe, I am less certain of thequalifications of the writer. There is a gameI sometimes play with children; I call it

2 Selected Papers No. 50

“Three Questions”. If all three questions areanswered correctly I promise a million dollars;no doubt the Securities and Exchange Com-mission will eventually prohibit the game,or the Federal Reserve System will make itviable. The first two questions present nodifficulty: perhaps the number of brothersand sisters the child has, and the city inwhich it lives. The third question is a dif-ferent matter. Once I asked “Who was AdamSmith’s best friend?” The reply from thischild was, “You are, Uncle George”. I hadsomeone like David Hume or James Huttonor Joseph Black in mind. Still, I have longbeen a good friend of Smith, though I haveno right to claim priority in his circle. I donot believe that my friendship will distortthe judgments I shall propose.

The task I set, in any event, is not theuninteresting one of praising or blamingSmith. The triumphs of any scholar arethose of his doctrines which he persuadeshis contemporaries and successors to heedcarefully. When Ricardo or John Stuart Millor Torrens adopted a theory of Smith’s, thatdid not necessarily mean that they acceptedit without qualifications but that their workand thoughts were directed by the formula-tion of Smith. Smith’s failures were, corre-spondingly, those theories which his successorseither ignored or rejected out of hand. WhenSmith was wrong, we would naturally expectable successors to ferret out the error, butwe shall also discover that some of Smith’sfinest theories suffered the fate of neglect.In any event, it is the judgment of the sciencethat is decisive in judging a scholar’s achieve-ments.

There is, I hope, an intrinsic interest inSmith’s triumphs and failures simply becausehe was as great an economist as has ever

George J . S t ig ler 3

lived. There is also a broader significanceto my query: can we determine the char-acteristics of theories that help or hurt theirreception ? 11. The Proper SuccessesA success or triumph is a proposition ineconomics that becomes a part of the workingsystem (the so-called paradigm) of con-temporary and subsequent economists. Theyaccept and u s e the proposition, with heavyemphasis upon the word use, or they rejectand the proposition, with heavyemphasis upon dispute. In either event, theirwork is influenced by the successful propo-sition and, indeed, measures the success. SoI repeat: a theoretical analysis is a successif it becomes a part of the living economicsof successors, and the success is ascribableto Smith if his formulation governs the lateruse of the theory, whether he invented it ornot. Hence I shall not attempt to determineSmith’s debts to his predecessors; suffice tosay they were large, but much smaller thanour debts to him. One can say of Smith whatNewton said of himself: “If I have seenfarther, it is by standing on the shouldersof giants”. That is appropriate to compari-sons with predecessors but not to comparisonswith contemporaries: they had the sameshoulders to stand on.

Smith had one overwhelmingly importanttriumph: he put into the centre of economicsthe systematic analysis of the behaviour ofindividuals pursuing their self-interest underconditions of competition. This theory wasthe crown jewel of The Wealth of Nationsand it became, and remains to this day, thefoundation of the theory of the allocationof resources. The proposition that resourcesseek their most profitable uses, so that inequilibrium the rates of return to a resource


in various uses will be equal, is still the mostimportant substantive proposition in all ofeconomics.

I do not know whether to list as a secondtriumph one enormously successful applica-tion of this theory of competitive prices,namely Smith’s theory of the differentialsin wage rates and profit among occupations.The famous list of cost factors which wouldgenerate apparent but not real differencesin rates of wages and profits - training,hardships, unemployment, and trust - wereaccepted, and in fact usually quoted verbatim,by Smith’s successors for a century.* Thisliterature is the direct ancestor of Marshall’sfamous chapters on wages (Marshall, Bk.VI,Chs.III-V), and of the modern theory ofhuman capital. So perhaps this special ap-plication of price theory deserves to be listedas his second success.

The third and final major success of Smithwas his attack on mercantilism. I measurea success by the impact of a scholar on otherscholars, not his impact upon public thinkingor public policy. Smith’s attack on protec-tionism in all its basic forms - tariffs,subsidies, compulsory use of domestic ship-pers, limitations on colonial enterprise, andthe like - rested squarely on his theory ofcompetitive prices. The crucial argumentfor unfettered individual choice in publicpolicy was the efficiency property of com-pet i t ion : the manufacturer or farmer orlabourer or shipper who was seeking tomaximise his own income would in the veryprocess be putting resources where they weremost productive to the nation.

Every individual is continually exertinghimself to find out the most advanta-geous employment for whatever capitalhe can command. It is his own ad-

George J . S t ig ler

vantage, indeed, and not that of societywhich he has in view. But the studyof his own advantage naturally, orrather necessarily leads him to preferthat employment which is most advan-tageous to the society. (Smith,I,454.)

This application of price theory was againa corollary of the main proposition, but itsdevelopment was so extensive and its successso great that it clearly deserves to be calledSmith’s third major triumph.

I have been most parsimonious in quotingSmith, and it is well to present a sampleof the power of his argument. Here is howhe describes one section of the English policyof mercantilism :

To found a great empire for the solepurpose of raising up a people of cus-tomers , may at first sight appear aproject fit only for a nation of shop-keepers. It is, however, a project al-together unfit for a nation of shopkeep-ers; but extremely fit for a nationwhose government is influenced byshopkeepers. Such statesmen, and suchstatesmen only, are capable of fancyingthat they will find some advantage inemploying the blood and treasure oftheir fellow-citizens, to found andmaintain such an empire. Say to ashopkeeper, “Buy me a good estate, andI shall always buy my clothes at yourshop, even though I should pay some-what dearer than what I can have themfor at other shops”; and you will notfind him very forward to embrace yourproposal. But should any other personbuy you such an estate, the shopkeeperwould be much obliged to your bene-factor if he would enjoin you to buyall of your clothes at his shop. England


purchased for some of her subjects, whofound themselves uneasy at home, agreat estate in a distant country. Theprice, indeed, was very small, andinstead of thirty years’ purchase, theordinary price of land in the presenttimes, it amounted to little more thanthe expense of the different equipmentswhich made the first discovery, recon-noitred the coast, and took a fictitiouspossession of the country. The landwas good and of great extent, and thecult ivators having plenty of goodground to work upon, and being forsome time at liberty to sell theirproduce where they pleased, became inthe course of little more than thirty orforty years (between 1620 and 1660)so numerous and thriving a people,that the shopkeepers and other tradersof England wished to secure to them-selves the monopoly of their custom.Without pretending, therefore, thatthey had paid any part, either of theoriginal purchase-money, or of thesubsequent expense of improvement,they petitioned the parliament that thecultivators of America might for thefuture be confined to their shop; first,for buying all the goods which theywanted from Europe; and, secondly,for selling all such parts of their ownproduce as those traders might find itconvenient to buy. (For they did notfind it convenient to buy every part ofit. Some parts of it imported intoEngland might have interfered withsome of the trades which they them-selves carried on at home. Those par-ticular parts of it, therefore, they werewilling that the colonists should sell

George J. Stigler 7

where they could; the farther off thebetter; and upon that account proposedthat their market should be confined tothe countries south of Cape Finisterre.)A clause in the famous Act of Naviga-tion established this truly shopkeeperproposal into a law. (Smith,II,613-14.)

From 1776 to today, the effect of this powerfulattack, reinforced by the theoretical advancesof Ricardo, Mill and others, established atradition of free international trade whicheven the most confirmed of economist-inter-ventionists seldom feel equal to attackingfrontally.

There is a fourth considerable success tobe credited to Smith: the formulation of thewages-fund theory. This theory explainedthe short-run level of average wages by theratio of funds for the payment of labour(the wages-fund) to the number of labourersemployed. It was saved from being a tautol-ogy by the implici t condit ion that overmoderate periods of time the wages-fundwas approximately constant in size. Puttingaside the question whether it was a usefultheory (I have argued that it was, Stigler1968), there is no doubt that it dominatedthe next hundred years of English economics.The uncertainty is how clearly Smith for-mulated the theory. He definitely assertedthe essence of the theory, as when he says:

The demand of those who l ive onw a g e s , it is evident, cannot increasebut in proportion to the increase of thefunds which are destined for the pay-ment of wages. These funds are of twokinds; first, the revenue which is overand above what is necessary for themaintenance; and, secondly, the stockwhich is over and above what is neces-


sary for the employment of theirmasters . (Smith,I,86 also I,110,453).

Smith’s theory of wages will be shown belowto rest on a wages-fund mechanism. Theonly real misgiving is that Smith did notexplicitly define the contents of the wages-fund.3

I am painting with a wide brush: insightsand arguments of lesser scope, which wouldbe a source of fierce pride to lesser economists,do not deserve inclusion here. The famousparadox of value concerning diamonds andwater, for example, which posed in inescap-able form the central problem for themarginal utility theory, would deserve atten-tion in any lesser man’s work. But the firstthree of these four successes I distinguishhave become a permanent part of economics.2. The Improper Successes

An improper success is an error or aninfertile and undevelopable subject or methodof analysis - but one that is influential withcontemporaries or successors. Most demon-strable errors, one hopes and believes, are soonferretted out, but the analysis that somehowfails to identify and organise and exploit auseful body of knowledge can only be dis-covered with time.

I would propose only one significant topicin Smith’s work that meets this description:his theory of productive and unproductivelabour.

There is one sort of labour which addsto the value of the subject upon whichit is bestowed: There is another whichhas no such effect. The former, as itproduces a value, may be called pro-ductive; the latter, unproductive labour.. . . (The) labour of the manufacturerfixes and realises itself in some partic-

George J. Stigler 9

ular subject or vendible commodity,which lasts for some time at least afterthat labour is past. It is, as it were,a certain quantity of labour stockedand stored up to be employed, if neces-sary, upon some other occasion. . . . Thelabour of the menial servant on thecontrary, does not fix or realise itselfin any particular subject or vendiblecommodity. His services generallyperish in the very instant of theirperformance . . . (Smith,I,330.)

The purpose of the distinction is clear: ifwe identify productive labour by the char-acteristic that its product can be accumulated,then capital formation can take place onlyout of the product of productive labour. Thedifficulties with the distinction are two. Evenif Smith is correct, the extensive employmentof productive labour merely theaccumulation of capital , and the actualformation of new capital requires a whollyindependent act of saving. Since mosttangible products are not accumulated ascapital but currently consumed, there couldbe the loosest of connections between theshare of labour that is productive and therate of capital growth.

There is a second difficulty : there areinvestment acts which are not the result ofproductive labour. Investments in what wenow call human capital do not becomeincorporated in a tangible saleable commodityas commonly understood. Yet Smith agreesthat one portion of the stock of a societyconsists of the acquired useful abilities ofits inhabitants to which he should haveadded the discovery of new knowledge:

The acquisition of (useful) talents, bythe maintenance of the acquierer during


his education, study, or apprenticeship,always costs a real expence, which isa capital fixed and realised, as it were,in his person. These talents, as theymake a part of his fortune, so they dolikewise of that of the society to whichhe belongs. (Smith,I,282)

Unless we include instruction and trainingas productive labour - and Smith lists “menof letters of all kinds” as unproductive labour- the existence of productive labour is noteven necessary to capital formation.4

The concept of productive labour nevermade a deep impression on Smith’s successors.Senior and McCulloch denied the distinctionand John Mill refined it almost out ofexistence.5 So it was a small impropersuccess.6

3. The Proper Failures

Smith’s failures to persuade economics were,like his successes, of two sorts: failures thatwere proper, and failures that should havebeen successes. We consider first the properfailures. A proper failure contains an analyt-ical error, or it presents an empiricallytrivial or mistaken view of the world.

The most conspicuous of Smith’s properfailures was the hierarchy of employmentsof capital, presented in Bk.II,Ch.V, “Of theDifferent Employments of Capital”.

A capital may be employed in fourdifferent ways: either, first in procuringthe rude produce annually required forthe use and consumption of society; orsecondly, in manufacturing and prepar-ing that rude produce for immediateuse and consumption; or thirdly, intransporting either the rude or manu-factured produce from the places wherethey abound to those where they are

George J . S t ig ler 1 1

wanted; or lastly, in dividing particularportions of either into such smallparcels as suit the occasional demandsof those who want them”. (Smith,1,361.)

Although all four activities are essential toone another or to “the general convenienceof the society”, capital is more productive -that is , sets more labour to work, andaugments more the annual produce of thesociety - if applied earlier in this sequenceof operations than if applied later. Theargument is simple: the capital of a retaileremploys only himself and possibly a clerk -the remainder of the capital goes to purchasethe goods he sells, and therefore to replacethe capitals of earlier stages. At the othere n d , “no equal capital puts into motion agreater quantity of productive labour thanthat of the farmer” (Smith, I,363), for all ofhis capital goes to support labour, and inaddition the fertility of nature is enlisted.

That Smith was in error is unequivocal.He allowed a system of financing to concealthe facts of economic life. If the consumer,instead of paying the retailer for the corn,had paid the farmer for raising it, the mill-wright for grinding it, the ship’s captain fortransporting it, and the retailer for stockingit, then everyone’s capital would have goneexclusively to the direct support of production,but nothing essential would have changed.

If Smith had really incorporated this errorinto his theoretical system, the effects wouldhave been disastrous: as one importantexample, the argument for private controlover investment would have been damagedbeyond repair. But it remained a localblemish (repeated, however, once, Smith, II,573), duly refuted by McCulloch and ignoredby Senior and J.S.Mill (McCulloch, 143ff).

1 2 Selected Papers No. 50

Only Malthus gives i t warm approval(Malthus, 30ff.)

This error is commonly, and no doubtcorrectly, attributed to the influence of thePhysiocrats: there is no such thing as a freetrip to Paris. But this is a history of theerror, not an explanation for Smith’s com-mission of it. (Does the explanation lie inhis anti-luxury viewpoint? - reference willbe made below to this attitude.)

A related error, and one to which Smithattached greater importance if measured bythe number of times it recurs in The Wealthof Nations, is the assignment of a hierarchyof social usefulness to domestic trade, foreigntrade, and the carrying trade to foreignnations (Smith, I,368ff, 454, 495-96; II,600-04,610-11,628-30). The internal trade, he argues,by the act of buying Scottish manufactures,carrying them to London, selling them andbuying English corn to return to Edinburgh,replaces two British capitals, whereas theforeign trade replaces only one British capitaland the carrying trade none. In addition thereturns of local trade are quicker than indistant trade. At this level of discourse, Smithis surely mistaken. If these various tradesare yielding equal annual rates of return oncapital, a shift from foreign to domestictrade would reduce aggregate national output(although the export of capital can of courseaffect wages). This error received no greaterapproval from Smith’s successors (thus,Ricardo,350-51).

A very different error, and possibly not anerror at all, is Smith’s measure of value -which came from the same source as thatwhich may have led him to over-value agri-culture. Smith was acutely sensitive to theinstability of monetary measures of value,and an appreciable fraction of The Wealth

George J. Stigler 13

of Nations is devoted to the chronicle ofcurrency debasement and inflation. Heproposes as the ultimate measure of valuethe disutility of an hour of ordinary labour:

Equal quantities of labour, at all timesand places, may be said to be of equalvalue to the labourer. In his ordinarystate of health, strength and spirits; inthe ordinary degree of his skill anddexterity, he must always lay down thesame portion of his ease, his liberty,and his happiness.

The price which he pays must alwaysbe the same, whatever may be thequantity of goods which he receivesin return for it. . . . Labour alone,therefore, never varying in its ownvalue, is alone the ultimate and realstandard by which the value of allcommodities can at all times and placesbe estimated and compared. (Smith,I,50-51.)

Smith’s error, if indeed it is an error, is toassume that the psychological cost of per-forming an hour of labour is more stable,in its significance to a person, than thepsychological pleasure from the consumptionof some bundle of goods. The instability oflabour disutility arises from at least threecircumstances :

1. It varies with the conditions of technology- for example, the lifting of heavy weightshas been almost eliminated in a modernsocie ty .

2. It varies with the degree of training ofthe worker: the disutility of acquiringlabour skills must be added to that ofperforming the work, and this additionwas already increasing secularly with theprogressive division of labour.


3. It varies with the hours of labour, andhence with income.

The corresponding view of a bundle ofconsumer goods yielding constant satisfactionas the unit of value is free of the seconddifficulty, possibly free of the first (dependinghow one views new commodities), but ofcourse not free of the third.

Smith’s rejection of consumption in fixingon a measure of value is attributable to hisbelief that luxuries are frivolous and yieldillusory pleasures that vanish in the act ofrealisation. This view is extensively arguedin his (Part IV, Ch.I) andreceives adequate expression in The Wealthof Nations.’ That Smith should attributeto almost all economic factors an illusionthat greater wealth yields greater satisfactions,an illusion that is perhaps never pierced, isone of his greatest idiosyncrasies.

Smith’s third error, and again perhaps weshould label it a misdirection, is his mone-tary theory, as presented in Bk.II, Ch.II, OnMoney. Smith believes that there is a fixeddemand for money in a society, in the specialsense that only a certain quantity of moneywill circulate and excessive sums will beexported (if the money is gold or silver)or be presented for redemption in gold (ifthe money is bank notes). The theory istenable as a first approximation if, as Smithassumes, the foreign exchanges are fixed, andthe paper currency is fully convertible: thetheory then is implicitly a simple purchasingpower parity theory.*

The complaint at Smith’s theory is notthat it is formally erroneous but that itrepresents a retrogression from the generalityand predictive power of the monetary theoryin Hume’s essays.

George J. Stigler 1 5

4. The Improper Failures

There remain the successes that Smith shouldhave achieved, but did not. It will appearparadoxical that Smith’s immense prestigeand vast powers of persuasion should havefailed to obtain acceptance of ideas thatwere correct, profound and fecund.

The first of these superior theories was arejection of the subsistence theory of wages.Smith, it will be recalled, gave four explicitreasons for believing wages were not gen-erally at subsistence level in Great Britain:

1. Summer wages exceed winter wages, butthe cost of subsistence varies inversely.

2. Subsistence varies substantially in costfrom year to year, but some wages changevery slowly.

3. Subsistence varies substantially from placeto place, but wages vary less by place.

4. Variations over time and place in the costof subsistence are often inverse to thoseof wages. (Smith,I,91-93.)

All of these proofs, particularly the first two,suffer from a concentration on short runcorrespondences of wages and the cost ofsubsistence, but they carry considerableweight. In addition, Smith offers the powerfullong run example of the differences in realwages between England and the Americancolonies (Smith,I,87-88)) an example whosepersistence made i t s tronger with eachpassing year.

Smith proposed an alternative theory, andone which was surely more valid than thesubsistence theory as a predictor of wagerates. He proposed that “the” wage rate of(say) unskilled labour was given by


Wage rate = Subsistence level +

A (Capital) , x > o

(Time)

that is, that population lagged changes incapital, so

. . . it is in the progressive state, whilethe society is advancing to the furtheracquisition, rather than when it hasacquired its full complement of riches,that the condition of the labouringpoor, of the great body of the people,seems to be the happiest and the mostcomfortable. It is hard in the stationary,and miserable in the declining state.(Smith,I,99.)

This wage theory, despite its great plausi-bility, was easily vanquished for a generationby Malthus’ simple theory (which set x= 0).

A second of Smith’s theories took slightlymore than a century to achieve currency -it was his theory of rent. He consistentlytreated the rent of land as it should betreated: any one use of land had to pay arent, which was a cost of production, to drawthe land away from other uses; whereas forall uses combined, rent was a residual. Thistheory is present in Bk.I, Ch.XI, The Rentof Land, with hardly any ambiguity butwith hardly any explicitness. Indeed, I usedto suspect my own reading of it until Idiscovered an early and wholly concordanttreatment by D.H. Buchanan. It is difficultin retrospect to see how the many recogni-tions of the alternative cost theory receivedso little attention, as when Smith says

As an acre of land, therefore, will pro-duce a much smaller quantity of theone species of food (meat) than of the


other (corn), the inferiority of thequantity must be compensated by thesuperiority of the price. If it is morethan compensated, more corn landwould be turned into pasture; and if itwas not compensated, part of what wasin pasture would be brought back intocorn. (Smith,I,l65, also 168, 175 etc.)

John Stuart Mill gave this theory timidrecognition, and Marshall refused to giveit full credit. But unlike the other theoriesof Smith under discussion, the correct theoryhere is only partly explicit and it was frag-mented in presentation, so he rather than hissuccessors deserves the larger blame for itsneglect .

The last of Smith’s regrettable failures isone for which he is overwhelmingly famous,the division of labour. How can it be thatthe famous opening chapters of his book,and the pin factory he gave immortality,can be considered a failure? Are they notcited as often as any passages in all eco-nomics? Indeed, over the generations theyare.9

The failure is different: almost no oneused or now uses the theory of division oflabour, for the excellent reason that thereis scarcely such a theory. The descriptionof division of labour was much enlarged inBabbage’s account of manufactures, and thephenomenon lies at the base of that part ofMarshall’s theory of external economics whichattends to localisation of industry. There aremore praises and even mild use of Smith’stheorem that the division of labour is limitedby the extent of the market in essays byAllyn Young and myself, and Ronald Coase’swork on the firm is clearly in the line ofdescent. (Young, Stigler,l951; Coase). Butthere is no standard, operable theory to


describe what Smith argued to be themainspring of economic progress.

Smith gave the division of labour an im-mensely convincing presentation: it seems tome as persuasive a case for the power ofspecialisation today as it appeared to Smith.Yet there is no evidence, so far as I know,of any serious advance in the theory of thesubject since his time, and specialisation isnot an integral part of the modern theoryof production - which may well be anexplanation for the fact that the moderntheory of economics of scale is little morethan a set of alternative possibilities.

5. The Recognition of Success and FailureIt is a general rule of scientific work that ascholar’s successes and failures are judgedby his contemporaries, and their judgmentis accepted by later scholars. Allowing forthe fact that Smith wrote when there werefew even part-time economists, so we mayperhaps treat the early nineteenth centuryas near-contemporary, his experience con-firms this rule. Certainly all of Smith’ssuccesses, proper and improper, were achievedwithin fifty years of initial publication. Allof his failures, proper and improper, weresimilarly achieved in this period. It isalmost (but not quite) tautological thatproper successes and proper failures bepromptly recognised as such, but it is notinevitable that an improper success soon bedespatched (recall that Smith’s, on productiveand unproductive labour, was a modestsuccess).

As for improper failures, the interestingpoint is that they do not influence the lateradoption of the neglected contribution. Whenthe theory of rent was finally set straight -a development which took place primarily


in the period from 1890-1910 - the correctformulation owed nothing to Smith’s pro-found but inarticulate insight. When theMalthusian wage theory was abandoned, asit was on an ever-widening scale from 1825on, it was not Smith’s highly plausible theoryof population lags that was adopted: in fact,for a long time, no alternative theory wasadopted and population receded from eco-nomic attention. It is perhaps rash tocomplete the trilogy, for these are threescientific tragedies, by asserting that whena theory of specialisation comes, it will owelittle to Smith, but on historical grounds evenrasher to assert the converse.

There are of course exceptions to the rulethat much later scientific descendants acceptthe judgment of a man’s contemporaries:perhaps Cantillon, von Thunen, Cournot,and Gossen are the leading exceptions in thehistory of economics.

These exceptions were for Schumpeter thesource of considerable lament at the closedminds and narrow visions of contemporaries(Schumpeter, 463ff. and elsewhere), but itseems unfruitful to expect of a science thatit immediately value all scientific work atits ultimate worth, never erring in deficitor excess.

The rule of the dominance of contempo-rary judgements, however, rests upon anotherbasis : science is a social pursuit of know-ledge, not a census of independent individuals.A scholar who does not influence his con-temporaries - who does not persuade themto work differently - is not an effectivemember of that science. Occasionally he mayindeed have been too farsighted - inventingdisc brakes before the internal combustionmachine was known - but this is so rarea cause of failure that there cannot be many


less efficient ways to discover good unex-exploited ideas than by reading earlier liter-ature. The overwhelming cause of failureof scholars is that their ideas were erroneousor infertile, or their development too prim-itive to provide useful guidance to theircontemporaries. That is the proper reasonfor a judgment of failure in a social enter-prise, and later scholars are quite sensible inaccepting the verdict in perpetuity.

6. The Sources of Success and FailureIn the long run nothing is more essentialto the theory than that it be right, but wecannot even pause for a new sentence beforeremarking that rightness means limitedwrongness. The theory must help in ex-plaining to the world what economics isattempting to understand, and a partialexplanation is better than none.

Logical error is sometimes enough todisqualify a theory - that is why Smith’sfour-layered hierarchy of employments ofcapital never had a prospect of scientificprosperity. Usually, however, it is possibleto retain the substance of a theory in alogical reformulation. Perhaps that is thereason that logical criticisms of Smith’shierarchical theory were not common -economists like Ricardo simply ignored thistheory, and it is difficult to doubt that theiraversion was at least partly due to a beliefthat the whole approach was infertile andcontrived.

Blemishes, however, will exist in everytheory: the logic may be reasonably rigourous- although standards of rigour are notunchanging - but the very formulation ofa problem will, in time, prove to be obtuse.

The acceptability of Smith’s theories, logicaside, was very little influenced by the

George J. Stigler 2 1

strength of the specific evidence he gave intheir support. His strong empirical argu-ments against a subsistence theory of wageswere ignored, whereas his support for thetheorem on equalisation of rates of returnunder competition was only casual andanecdotal. To say that a proof is non-specificand non-quantitative is not to say that itis unweighty : if the theorem was congruentwith widely observed phenomena - thegrowing number of members of prosperoustrades, the fall of prices of new goods overtime - then one was prepared to follow thetheorem into unobserved places. A centurylater, however, when Cliff-Leslie denied thetendency of rates of return to approachequality, there was precious little documentedevidence to refute him.

But the proof of the ubiquity of the divisionof labour certainly also met this test ofcongruence with common observation : infact, irresistibly so. Ask a modern economistto name an instance of non-specialisation oflabour and he will be lucky to rememberRobinson Crusoe. Yet, as we saw, there hasbeen scarcely any systematic or regular useof this concept in economic analysis.

We have already hinted at the differencebetween the fates of the theorems on divisionof labour and on rates of return. The latterwas a generalisation of enormous power; itcould be used immediately on the mostobtrusive and important questions : whysome occupations earned more than others;why mercantilism and similar state inter-ventions, as well as private monopoly, ledto misallocations of resources; and whowould bear various taxes. The theory ofdivision of labour is not devoid of conse-quences - that it is limited by the extentof the market makes it relevant to protec-

2 2 Selected Papers No. 50

tionism, for example - but the uses werefew and far between.

So Smith was successful where he deservedto be successful - above all in providing atheorem of almost unlimited power on thebehaviour of man. His construct of theself-interest-seeking individual in a com-petitive environment is Newtonian in itsuniversality. That we are today busilyextending this construct into areas of eco-nomic and social behaviour to which Smithhimself gave only unsystematic study istribute to both the grandeur and the dura-bility of his achievement.

Footnotes

1 tried to answer this question in Stigler, 1965, pp.66-155. The present approach is essentially independentalthough the answers overlap.

2 The role of trust was not analysed satisfactorily bySmith and its acceptance was much less complete. Fora modern interpretation of it, see Becker and Stigler.Smith’s fifth source of differences, the uncertainty ofsuccess, was not a cost-based differential, and it wasmuch disputed rather than generally accepted.

3 This is Taussig’s main reservation in the standardhistory of the wages-fund (Taussig, 1899,150-51). Butthe fund is not so easily defined as Smith’s successora n d T a u s s i g b e l i e v e d - i t w a s n o t s i m p l y , o r a l lof, “the consumable goods, in dealers’ hands, ready forpurchase by labourers” (Taussig,l48). The per iod ofadvance of the wages determines how “ready” the goodsneed be, as just one complication.

4 William Playfair pointed out , in h is edi t ion of TheWealth of Nations, that even the employment of menialservants might be a productive act:

“ . . . a cook for (example) is a menial servant, butin a tavern he enriches his master just as much asany other journeyman; and on the other hand, aservant that spins or sews for the use of her masteri n a p r i v a t e f a m i l y , i s o n l y a c t i n g a s a m e n i a lservant; she is but supplying his wants, and con-tributing to his comforts in the same manner aswhen she lights the fire, or washes the apartments,yet she is a productive labourer by this definition.”(Playfair,II,2n.)


Collected Works of John Stuart Mill, “On theWords Productive and Unproductive”, IV,280ff. A lessinfluential figure, David Buchanan, was a strong sup-porter of the distinction (Buchanan,13 l-36).

6 Hla Myint would wish to make the concept of pro-ductive labour and of economic growth more generallycentral to Smith’s work, and a wholly proper success(Myint, 20-24). I would assert the contrary: that growthis not the only important path to economic welfare inSmith’s system, and the concept of productive labouris not important to growth. But the modesty of thesuccess of Smith’s distinction is a matter of historicalrecord.

7 “For a pair of diamond buckles perhaps, or for some-thing as frivolous and useless, (the feudal lords) ex-changed the maintenance, or what is the same thing,the price of the maintenance of a thousand men fora year, and with it the whole weight and authoritywhich i t could give them.” (Smith,I,418-19.)

8 Smith assumes the full convertibility of paper money(Smith,I,329).

9 Babbage reports the dramatic story of how M. Pronywas enabled, by reading Smith on this subject, to planthe great mathematical tables of the French revolutionarygovernment (Babbage, 191-95).

References

Babbage, C., On the Economy of Machinery andManufactures, London, 1842.

Becker, G.S. and Stigler, G.J., “Law Enforcement,Malfeasance, and Compensation of Enforcers”,Journal of Legal Studies, January, 1974.

Buchanan, D., Observations on the Subjects Treatedof in Dr. Smith’s Inquiry into the Nature and theCause of the Wealth of Nations, London, 1817.

Buchanan, D.H., “The Historical Approach to Rentand Price Theory,” Economica, 1929, republished inA.E.A., Readings in the Theory of Income Distri-bution, Philadelphia, 1946.

Coase, R., “The Nature of the Firm,” Economica,1 9 3 7 .

McCulloch, J.R., The Principles of Political Econ-o m y , Edinburgh, 1825.

Malthus, T.R., Principles of Political Economy,London, 1820.

Marshall, A., Principles of Economics, Macmillan,1 9 6 1 .


Myint, Hla, “The Welfare Significance of Produc-tive Labour,” Review of Economic Studies, Winter,1943.

Playfair, W., ed., The Wealth of Nations by AdamSmith, London, 1805.

Rae, John, Life of Adam Smith, Macmillan, 1895.

Ricardo, D., Principles of Political Economy, Vol.1of Works and Correspondence, Cambridge, 1951.

Schumpeter, J.A., History of Economic Analysis,New York, 1954.

Smith , A. , The Theory of Moral Sentiments,Dublin, 1797.

Smith, A.; The Wealth of Nations, Clarendon Press,1 9 7 6 .

Stigler, G.J., “The Division of Labour is Limitedby the Extent of the Market,” Journal of PoliticalEconomy, 1951.

Stigler, G.J., Essays in the History of Economics,University of Chicago Press, 1965.

Stigler, G.J., “Mill on Economics and Society,”University of Toronto Quarterly, October, 1968.

Taussig, F.W., Wages and Capital, Appleton, 1899.Young, A.A., “Increasing Returns and EconomicProgress,” Economic Journal, 1928.

Documents

The Successes and Failures of Professor Smith