THE WORK OF MYTHSHARING

"It was the very dawning of day when the term 'Dignity of Labor' meant something" – George E. McNeill

In 1884, the Federation of Organized Trades and Labor Unions proclaimed that as of May 1, 1886, eight hours would become the length of the working day. In the months leading up to that deadline, thousands of workers across the country joined the struggle. On Mayday, hundreds of thousands went out on strike for shorter hours. In Chicago, “no smoke curled up from the tall chimneys of the factories and mills, and things had assumed a Sabbath-like appearance.” Then on May 4, reaction set in. Jeffory Clymer set the scene in "The 1886 Chicago Haymarket Bombing and the Rhetoric of Terrorism in America":

On 4 May 1886, about two thousand Chicagoans gathered at Haymarket Square to protest against the city's police, who had shot and killed at least two striking workers outside the McCormick reaper factory on the previous afternoon. The demonstration was peaceful, and only a few hundred people remained when, late in the evening, 170 Chicago policemen suddenly arrived and demanded that the protesters disperse. Nonplused by the anticlimactic arrival of the police at the close of a peaceful rally, Samuel Fielden, the evening's last speaker, pointed out the meeting's non-violent nature in response to the peremptory dispersal order. At this point in the exchange, someone tossed a dynamite bomb into the police ranks. The explosion immediately killed Officer Mathias Degan, wounded several others, and prompted a cacophony of gunfire, most of it from police pistols. In the chaos, the police shot several of their own officers as well as many of the fleeing civilians. While the number of dead among the police rose to seven over the next few days, the actual number of casualties among the protesters, like the bomb thrower's identity, was never determined.

In the aftermath, eight Anarchist labor leaders were put on trial as "accessories to murder":
August Spies
Albert Parsons
Samuel Fielden
Adolph Fischer
George Engel
Michael Schwab
Louis Lingg
Oscar Neebe
It was a kangaroo court with a rigged jury and predetermined outcome. Four of the convicted men were hanged. Louis Lingg escaped the noose by taking his own life. The remaining three were freed in 1893 by Illinois Governor Altgeld because they – and the dead men – were innocent of the crimes for which they had been convicted.

Eight hours for work, eight hours for rest, eight hours for what we will…

Ira Steward was a pioneering American labor leader and early proponent of the eight-hour day. His wife, Mary, is said to have written the couplet, which became popular among eight-hour day activists in the 19th century, "Whether you work by the piece or you work by the day, decreasing the hours increases the pay."

Frank Fetter was an economics professor in the early 20th century and an early adherent in the US to the Austrian school of economics. His textbook, Economics, published in 1916, contained one of the earliest textbook treatises on "the shorter day and the lump-of-labor notion." In his discussion, Fetter argued that the view that the shorter day would result in higher pay is "connected with the lump of labor notion" and is based on the assumption that "there is so much work to be done regardless of wages."

Although the dismissive phrase "fallacy of the lump of labour" had been coined by an economist, David Frederick Schloss, its use as a conclusive put-down of eight-hour day advocates was charted by the same yellow journalism pack of newspapermen who hanged Parsons, Engel, Spies and Fischer. To be fair to Fetter, his overall discussion of the shorter day did acknowledge the potential benefits of reduced working time. But his discussion was contorted into a frame that emphasized the supposedly fallacious thinking of eight-hour day advocates. No mention there of hanged labor leaders or lynch-mob newspapers.

O Kapteyn, My Kapteyn…

Fast forward to 2004. Professor Arie Kapteyn was lead author of an article titled "The Myth of Worksharing" published in Labour Economics. An earlier version of the article was issued as a discussion paper in 2000 by the Institute for the Study of Labor: "a place of communication between science, politics and business." One of the Institute’s current priorities is "mobility and flexibility of labor markets."

The framing of Kapteyn’s analysis follows a predictable lump-of-labor motif:

In public discussions the idea of worksharing often emerges as a potential instrument for reducing unemployment… However, economists as well as employers are mostly skeptical about the success of this policy prescription. The fallacy of this seemingly simple idea is made clear in the literature especially by its impact on wages, wage costs, and output.

A footnote on "fallacy" cites a 1997 Economist article, "One lump or two?" "It is depressing that supposedly responsible governments continue to pretend to be unaware of the old 'lump of labour' fallacy: the illusion that the output of an economy and hence the total amount of work available are fixed."

Aside from the unequivocal title of the article and the confidently formulaic framing of worksharing as being based on a fallacy, the paper’s conclusions are equivocal. The results from the authors’ literature survey are inconclusive. Their empirical analysis "does not provide any ground for the proposition that worksharing would reduce unemployment," which is a rather weaker claim than the "myth" and "fallacy" labels would suggest. And, hey, who knows? Reduction of working time might be a good thing ("welfare enhancing") after all.

But let’s return to Kapteyn’s empirical analysis for a moment:

Thus, the picture emerges that a reduction in working hours causes an increase in the real wage rate and, consequently, annihilates positive a direct effect of a reduction in working hours on the employment rate (Table 3) and turns it into a (insignificant) negative effect (Table 5).

Or, as a partial effect, "a one percent reduction in working hours results in a 0.38 percent increase in the employment rate and a 1.15 percent increase in the real wage rate." Substituting the wage equation into the employment equation, Kapteyn concludes that a "one percent reduction in working hours results, in the long run, in a 1.04 percent increase in the real hourly wage rate." Suggesting – as Mary Steward’s doggerel had it (and contra Frank Fetter) – that, "decreasing the hours increases the pay."

The bad news… wait for it… is that after substitution the employment rate goes from a 0.38 increase to a 0.27 decline that Kapteyn labels "insignificant". I’m guessing that means statistically insignificant. Which is to say, the model-building empirical mountain labored and gave birth to a mouse. Reading the fine print, there’s all sorts of stuff I could take issue with in the paper’s model specification. But to do so would be academic in the who-cares sense of the word. The point remains that the model-building, ‘empirical’ hocus pocus takes second billing to the myth/fallacy genuflection – a fallacy cut from the same evidential cloth as the Haymarket convictions in pursuit of the same end: suppression of the workers’ movement for an eight-hour day.

And on and on it goes...

September 2007, Marcello Estevão and Filipa Sá of MIT, the IMF and the IZA analyze the effects of the French 35-hour workweek in a paper prepared for the Panel Meeting of Economic Policy. They cite Kapteyn and present the following "rationale" for job creation through work sharing:

The idea of work sharing as an employment creation policy is simple: if the production of goods and services in an economy is fixed, then a reduction in hours can re-distribute the fixed amount of work across more people, increasing employment. In spite of its intuitive appeal, economists and policy makers are skeptical about the success of work sharing as it is rooted in the so-called “lump-of-labour fallacy”: the false premise that the amount of output in the economy is fixed.

They then go on to "analyze" the effects of the French policy using a model that presupposes precisely the optimizing equilibrium relationships between hours, wages and employment that the real theory of the hours of labor – Chapman’s – puts into question. The problem is, if you understand the implications of Chapman’s theory, you can’t do the kind of model building these folks do to “find” the conclusions they do. Why? Because the whole model building enterprise is explicitly predicated on a suspension of Chapman’s theory. It would make as much sense to conclude that since Parsons, Engel, Spies and Fischer were hanged, they must have been guilty.

The great "myth of worksharing" is that these economists who purport to analyze it empirically are doing any such thing. The touchstone is the ritual incantation of the bogus lump-of-labor fallacy claim.

Put up or shut up

Is there no economist with the integrity to stand up and defend the lump of labor fallacy claim that they reiterate with such abandon? Or are they like the jurors at the trial of the Haymarket martyrs?

I am offering a $10,000 prize, in Canadian funds, to be awarded to the author who directly and conclusively refutes the argument in "Why Economists Dislike a Lump of Labor," (Review of Social Economics, September 2007). My argument is that the authenticity of the lump-of-labor fallacy claim, with regard to unemployment and the hours of work, is questionable; that various explanations of it are inconsistent and contradictory and that Sydney J. Chapman’s neglected theory of the hours of labor presents a more coherent analysis of the reduction of working time than the often-cited fallacy claim.

The article must be accepted for publication in one of the 30 top-ranked economics journals. Only anonymously peer-reviewed articles are eligible, not book reviews, commentary or other journal front or back matter. Journal rankings will be taken to be those specified in Kalaitzidakis, Mamuneas and Stengos "Rankings of Academic Journals and Institutions in Economics." Journal of the European Economic Association 1 (December 2003).

(And, no, it won't count if the article simply re-asserts the fallacy claim and follows that by grinding out yet another "empirical analysis". The article must directly confront the argument of my article. The question of whether the article successfully refutes my argument I will leave to be resolved by publishability.)

Enter by notifying me, Tom Walker, by mail or email that your article has been accepted for publication in one of the qualified journals. Mail address is 1204 Lakewood Drive, Vancouver, BC, V5L 4M4. Email: lumpoflabor [at] gmail [dot] com.

Deadline for entry is 11:59 p.m. on January 31, 2010. In the event an article is under review by a qualified journal on the deadline date, an extension may be granted provided the article was submitted to the journal on or before December 31, 2009. All requests for extension of the contest entry deadline must attach a copy of the submitted draft.

If no contest entry meets all of the above criteria, a consolation prize of $1000 Canadian may be awarded to an anonymously peer-reviewed article meeting the core argument criteria, published in a journal ranked 31-159 by Kalaitzidakis et al. Articles published in a non-ranked journal and articles submitted to but rejected by an economics journal may be given consideration for the consolation prize at the sole discretion of the contest organizer.

Prize money will be awarded only to the author or authors of the article specified in the winning entry. In the event of group authorship, prize money will be divided equally between the authors unless specified otherwise and agreed in advance by all authors.