Dear Professor Krugman,
I am writing to you because three times over the last 14 months your authority has been invoked to me on behalf of the assertion that people who advocate shorter working time as a remedy for unemployment are guilty of a "lump-of-labor fallacy" assumption that there is only a fixed quantity of work in the world. As did John Maynard Keynes, I believe that working less is one of "three ingredients of a cure" for unemployment. I find it odd to learn that I (and presumably Keynes) am thereby assuming a palpable absurdity: that the amount of work to be done is invariant.
I have researched the history of the fallacy claim and published two scholarly articles on it and I have documented rather glaring discrepancies in the often-repeated claim. Because your authority on the alleged fallacy is so frequently cited, I would be extremely grateful if you would consider the evidence I outline below and respond to it. I believe the history is curious enough to be entertaining and thought provoking, whether or not you are persuaded by my presentation.
A column by you that has been held up to me as authoritative appeared in The New York Times on October 7, 2003. It was titled "Lumps of Labor." The first paragraph states as follows:
Economists call it the "lump of labor fallacy." It's the idea that there is a fixed amount of work to be done in the world, so any increase in the amount each worker can produce reduces the number of available jobs. (A famous example: those dire warnings in the 1950's that automation would lead to mass unemployment.) As the derisive name suggests, it's an idea economists view with contempt, yet the fallacy makes a comeback whenever the economy is sluggish.I would like to organize my presentation of the historical evidence by parsing your first two sentences into three parts: "Economists call it the 'lump of labor fallacy'"; "It's the idea that there is a fixed amount of work to be done in the world"; and "Any increase in the amount each worker can produce reduces the number of available jobs." I will conclude with a look at a insightful remark you made in 2002, "the 'productivity growth helps jobs' story, if that's what it is, is just the flip side of the lump-of-labor fallacy," in response to a Financial Times article by Glenn Hubbard.
"Economists call it the 'lump of labor fallacy'."
In 1890, Alfred Marshall used the expression, "fixed Work-Fund fallacy," in referring to what today would be called the lump of labor. Marshall expressed sympathy, on cultural and humanitarian grounds, for the objective of work time reduction while disputing its potential for expanding employment, except possibly in the case of shift work. Marshall's term was purposely evocative of the wages-fund doctrine of classical political economy, which John Stuart Mill had recanted twenty-one years earlier.
The facetious "Theory of the Lump of Labour" was coined by David Frederick Schloss, a lawyer and journalist, not an economist, in an 1891 article discussing worker's objections to piece-work. Schloss also professed sympathy for workers' demands for shorter working time and only objected to what he described as the systematic withholding of work effort – what Frederick Taylor was later to call "systematic soldiering."
John Rae, another journalist, attacked an unnamed fallacy in his 1894 book, Eight Hours for Work. For Rae, the alleged fallacy consisted of assuming that the relationship between hours and employment was "a simple sum in arithmetic." Rae was an advocate of the shorter working day, but not on the grounds that it would create jobs. He argued that an eight-hour day would improve productivity and concluded that those productivity gains would nullify any job creating effect.
It is illuminating to consider these three fallacy claimants from the 1890s collectively. They were expressing an idea and attitude that was clearly "in the air" but when their arguments are examined carefully they do not hold up well and, in crucial details, contradict one another. The American economist, Charles Beardsley, scrutinized John Rae's claims, concluding that they relied on the wages-fund doctrine and its underlying assumption that the relative shares of wages, profit and rent are immutable.
Marshall's "fixed Work-Fund fallacy" was described by A.C. Pigou, Marshall's pupil and successor as Cambridge Chair of Political Economy, as itself committing the fallacy of ignoratio elenchi. "If it were a good ground for rejecting an opinion that many persons entertain it for bad reasons," wrote Pigou, after restating the criticisms of the supposed belief in a fixed amount of work, "there would, alas, be few current beliefs left standing!" Pigou concluded that the employment potential of work time reduction depended ultimately on the effects of the increased leisure on workers' efficiency. Maurice Dobb offered an even more trenchant rebuttal to the fallacy claim in a 1928 Cambridge Economic Handbook, Wages:
…trade unionists in the nineteenth century were severely castigated by economists for adhering, it was alleged, to a vicious "Work Fund" fallacy, which held that there was a limited amount of work to go round and that workers could benefit themselves by restricting the amount of work they did. But the argument as it stands is incorrect. It is not aggregate earnings which are the measure of the benefit obtained by the worker, but his earnings in relation to the work he does — to his output of physical energy or his bodily wear and tear. Just as an employer is interested in his receipts compared with his outgoings, so the worker is presumably interested in what he gets compared with what he gives.Other economists -- Frank Carlton and Richard Lester, for example -- took exception to the "long run" Utopianism of the fallacy claim. They pointed out that what matters to workers are the immediate consequences of unemployment, not some hypothetically optimal long-run equilibrium.
"It's the idea that there is a fixed amount of work to be done in the world."
Although the Theory of the Lump of Labour and its cohorts emerged in the 1890s, "the idea that there is a fixed amount of work to be done" has a longer history. Specifically, a variant of the phrase can be traced back to commentary on the 1871 Engineers' strike for the nine-hour day in Newcastle, England.
A letter to the editor of The Times of London from "J.C.", published on September 21, 1871, criticized the views of an earlier writer, James Aytoun, as being "entirely based on the exploded fallacy that the amount of work to be done is a fixed quantity, unaffected by the wages paid or the number of labourers employed in producing it." Two weeks later, on October 6, the London correspondent for The New York Times, "F.H.J." filed a dispatch on the Newcastle strike, claiming that the strike leaders were secretly pursuing a sinister "ulterior design." "Their theory is that the amount of work to be done is a fixed quantity, and that in the interest of the operatives, it is necessary to spread it thin in order to make it go far."
The egregious error of "fixedness" would appear to owe something to William Thornton's critique of the wages-fund doctrine, John Stuart Mill's reply to Thornton and, perhaps, even to Karl Marx's 1865 polemic against John Weston in Wages, Price and Profit. In the latter, Marx wrote,
If our friend Weston's fixed idea of a fixed amount of wages, a fixed amount of production, a fixed degree of the productive power of labour, a fixed and permanent will of the capitalists, and all his other fixedness and finality were correct, Professor Senior's woeful forebodings would have been right, and Robert Owen, who already in 1816 proclaimed a general limitation of the working day the first preparatory step to the emancipation of the working class and actually in the teeth of the general prejudice inaugurated it on his own hook in his cotton factory at New Lanark, would have been wrong.Four years later, Thornton wrote, with regard to the wages-fund doctrine:
The believers in the wages fund, on the other hand, insist that whether labour be cheap or dear the whole body of employers always spend upon labour the utmost amount they can afford to spend. Very possibly they may be unconscious of insisting on this, but, consciously or unconsciously, this is demonstrably what they do insist upon. For they declare that the amount destined at any given time to the payment of wages, is a fixed amount – an amount so rigidly fixed that by applying to it the proper divisor you arrive infallibly at the average rate of wages. And they cannot but admit that this fixed amount may be the utmost that employers can afford; for, clearly, employers would pay the utmost they could afford for labour, rather than not get the quantity of labour they required. But if the fixed amount in question may be, it necessarily must be that utmost, for otherwise it would be not a fixed but a variable amount.The phrase about a fixed amount of work to be done cunningly commandeers the rhetoric of Marx's and Thornton's criticisms of the wages-fund doctrine, turns it around 180 degrees and uses it to argue against higher wages and shorter hours, exactly as the wages-fund doctrine had previously been used to argue against higher wages and shorter hours. Talk about snatching rhetorical victory from the jaws of doctrinal defeat!
In this strange doctrine it is, as Mr. Mill has pointed out, by implication affirmed that the demand for labour not only increases with the cheapness, but increases in exact proportion to it, the same aggregate sum being paid for labour whatever its price may be.' [compare also Thornton's phrase, "only a certain amount of work to be done," in Over-population and its Remedy]
"Any increase in the amount each worker can produce reduces the number of available jobs."
F.H.J.'s dispatch included a significant bridge to an earlier tradition of fallacy mongering. This is the theme of the "ulterior design" of unions. According to the Times correspondent, the Nine-Hour League was "only an offshoot of the Unions, and the great object of the Unions is to surround production with all manner of restraints and restrictions, so that it shall not be accomplished too fast or by a small number of workmen." That is, of course, the flip side of stating that "any increase in the amount each worker can produce reduces the number of available jobs."
F.H.J. was explicit about the conspiratorial nature of this object of restriction, as were John Wilson in "Economic Fallacies and Labour Utopias" (1871), James Ward in his egregiously plagiarized Workmen and Wages (1868), the anonymous author of a Quarterly Review article, "Trades Unions" (1867), Harriet Martineau in "The Secret Organization of Trades" (1859), Archibald Alison in "Trade Unions and Strikes" (1838) and Edward Carleton Tufnell in The Character, Object and Effects of Trades Unions (1834).
Tufnell's pamphlet contains the most complete prototype of what was to become the lump-of-labor fallacy claim. The credibility of that claim needs to be evaluated in the context of the author's disposition toward trade unions. "Were we asked to give a definition of a Trades' Union," the author stated at the book's conclusion, "we should say that it was a Society whose constitution is the worst of democracies — whose power is based on outrage — whose practice is tyranny — and whose end is self destruction."
Tufnell was an Assistant Poor Law Commissioner who served on the Whig government's Royal Commission aimed at deflecting agitation for the Ten-hour factory legislation. According to Sidney and Beatrice Webb's History of Trade Unionism, Tufnell's anonymously-published pamphlet was said to have been commissioned and paid for by the Whig government. The key text concerns the alleged motive of the Manchester Cotton Spinners' Union in supporting the Ten-hour Bill:
The Union calculated, that had the Ten-hour Bill passed, and all the present factories worked one-sixth less time, one-sixth more mills would have been built to supply the deficient production. The effect of this, as they fancied, would have been to cause a fresh demand for workmen; and hence, those out of employ would have been prevented from draining the pockets of those now in work, which would render their wages really as well as nominally high. Here we have the secret source of nine-tenths of the clamour for the Ten-hour Factory Bill, and we assert, with the most unlimited confidence in the accuracy of our statement, that the advocacy of that Bill amongst the workmen, was neither more nor less than a trick to raise wages—a trick, too, of the clumsiest description; since it is quite plain, that no legislative enactment, whether of ten or any other number of hours could possibly save it from signal failure.Walking back Tufnell's claim about the union's motive to the testimony before the Royal Commission, it is clear that Tufnell derived his conclusions from the supposition of a cotton manufacturer, Peter Ewart. Tufnell's question to Ewart was: "What do you suppose to be the chief motive for the operatives here advocating the Ten-Hour Bill?"
Besides being clearly labeled as supposition, Ewart's reply was more rambling, tentative and imprecise than Tufnell's sharply provocative allegations about the union's "secret motives" and "clumsy tricks." Ewart's testimony, in turn, can revealingly be interpreted as a rendition of the principles of popularized political economy – the wages-fund doctrine – such as was elaborated by Harriet Martineau in her 1832 story, "A Manchester Strike." In other words, what in Ewart's opinion made the supposed ideas of the union members fallacious was their non-compliance with the wages-fund doctrine. Moreover, the suspicion that they indeed held such views may have been suggested to Ewart not by their own words or actions but by a work of fiction.
Incidentally, the U.S. Commissioner of Labor investigated the ubiquitous claims of union restrictions on output and issued a 921-page special report in 1904, prepared and edited by pioneer labor economist John R. Commons, which found little substance to the claims, concluding, "the question of restriction of output… is not as simple as it has been supposed to be..." The report found that union regulations were aimed at ensuring that changes in work methods or organization were by mutual consent. To the extent such regulations were perceived by management as limiting output, it was in comparison to some hypothetical level of output that might presumably be attained if employers could impose their efficiency plans at will. Moreover, with respect to the reduction working time:
Considered solely with reference to speed or intensity of exertion, a moderate reduction in the number of hours of labor each day usually tends to increase the speed rather than to restrict it. From the standpoint of exertion, a reduction of hours is exactly the opposite from a restriction of output."The 'productivity growth helps jobs' story, if that's what it is, is just the flip side of the lump-of-labor fallacy."
One of the favorite unintended-consequences stories in economics is the idea that "technology creates more jobs than it destroys." This was a standard rebuke to Luddites in the early 19th century, who were portrayed as fearing that machines would create chronic unemployment. It closely resembles the case argued against the mercantilism of the early 18th century by Henri Martyn in Considerations on the East India Trade. The lump-of-labor fallacy appears as the negative version of this story. In fact, the fallacy is sometimes called the Mercantalist or Luddite fallacy.
There is a crucial difference between the two sides of the story, though. The technology creates jobs story is openly embraced by economists and triumphantly played as the trump card in debates over employment policy. The fixed-amount-of-work story, though, is only attributed by economists to Luddites, shorter work time advocates and other "naive populists" they wish to discredit. In both cases it is the economist (not infrequently, The Economist) speaking, telling the uninitiated to sit down and shut up.
In 1865, William Stanley Jevons introduced a paradoxical joker into the unintended consequences deck. In Chapter 7 of The Coal Question, "Of the Economy of Fuel," Jevons explicitly cited the argument that increased labor productivity expands employment as his model for analyzing the effects of fuel efficiency on fuel consumption: "…the same principles apply, with even greater force and distinctness, to the use of such a general agent as coal. It is the very economy of its use which leads to its extensive consumption."
The Jevons paradox presents economists with a dilemma that they have not squarely faced. If, as economists so often insist, technology creates more jobs than it destroys, fuel consumption operates on the same principle and the alternative job strategy of work time reduction is based on a fixed-amount-of-work fallacy; then job creation can only occur through increased fuel consumption. Or, conversely, fuel conservation can only occur at the expense of job destruction.
Unemployment exists and so does environmental crisis – so simply dismissing the three associated claims as fallacies is not enough. But one of these fallacies is not like the others. It applies to an attributed view, not a self-expressed one. On the contrary, advocates of work time reduction have often pointedly disavowed any assumption of a fixed amount of work, to no avail.
In The Rhetoric of Reaction: Perversity, Futility, Jeopardy, Albert O. Hirschman identified the three characteristic patterns named in the subtitle as recurring themes in reactionary rhetoric. The perversity claim is that a proposed reform will have results the opposite of what were intended. For example, reducing the hours of work will increase unemployment rather than decrease it. The futility claim argues that the "laws" of society (or in this case economics) prevent any effect whatsoever. It is sometimes argued that work-sharing doesn't reduce unemployment, it merely "spreads it around."
The jeopardy argument points to some cherished achievement that would be undermined by the reform. Legislated or collectively bargained work-time reduction would allegedly impinge on people's freedom to choose their own optimal combination of income and leisure. Various instances of the fallacy claim invoke each of Hirschman's three categories of perversity, futility and jeopardy. Some of the fallacy claims invoke more than one.
As Hirschman pointed out, often these arguments are advanced without substantiating evidence, purely on the strength of their seeming cleverness and irony. Hirschman described the perversity argument as a negative version of the "unintended consequences" story in which, for example, private vices are "led by an invisible hand" to produce public virtues. In the negative version, though, the unintended consequences are undesirable. In fact they are the opposite of what is desired.
Professor Krugman, I have presented only an outline and highlights that document the spurious nature of the fallacy claim. There are further twists and turns relating to the aggressive use of the fallacy claim by employers' organizations, the uncritical incorporation of the claim into textbooks and financial journalism, and the peculiar displacement of Sydney J. Chapman's once highly-regarded theory of the hours of labor from the theoretical canon of economics. I have discussed these issues elsewhere but have here restricted myself to material that responds directly to your own remarks.
As this is a question that bears on unemployment and the lack of an effective policy response to it, as well as on the protection of the environment, I hope that you will take the time to review the material I have brought to your attention and will either withdraw your support for the fallacy claim or, alternatively, explain why you continue to accept its validity (by "fallacy claim," let me reiterate that I am referring to economists' assertions that advocates of work time reduction believe in a fixed amount of work, not to the absurd proposition that there actually is a fixed amount of work). I would be more than happy to provide you with citations and links to all the articles I have mentioned in my letter.