The starting point: Work and working hours
It has been proposed to us recently that work and working (i.e., jobs) are inherently distasteful and punishing (Mumby, Reference Mumby2019), as well as meaningless, exploitative, and “inherently damaging to well-being and life satisfaction” (Alliger & McEachern, Reference Alliger and McEachern2024). The latter authors report on the so-called antiwork perspective—a “multifaceted, negative appraisal of work…in and of itself.” Yet on the other hand, Kaplan et al. (Reference Kaplan, Aitkin, Allan, Alliger, Ballard and Zacher2025) make the case for jobs being in many ways gratifying, fulfilling, self-actualizing, and virtuous. (One could forgive a young industrial-organizational (I-O) psychologist’s confusion regarding whether “work is wonderful!” or “work is the worst thing ever!”) Interestingly, a commentary on the Alliger and McEachern focal article proposed an empirically based middle ground—arguing against the ideological “anti-work perspective” by positing that much work and at least some jobs are gratifying and fulfilling (Lefkowitz, Reference Lefkowitz2024).
What are we to make of the Jekyll-and-Hyde characterization of work, working and organizations? If the original Stevenson (Reference Stevenson1886) allegory concerns the duality of human nature (good and evil), and we extend it to our domain of interest, it suggests that we should be wary of any extreme one-sided ideological characterization of jobs, work, and organizations (as well as “workers,” “management,” or “labor unions”).
Concomitantly, it also suggests that we should be focusing on particular features of work life that may have relevance: “As empirically grounded social scientists, I-O psychologists have devoted much of the 100 years or so of our existence investigating the positive and negative effects of particular features of jobs and organizations” (Lefkowitz, Reference Lefkowitz2024, p.33). That is not a criticism of the focal article; it points us in a number of “proposed research directions” (cf. Table 2). One might, however, quarrel with the overwhelming focus on only working hours—although one can imagine that the temptation to avail oneself of the intriguing “hook” provided by the Keynes prediction was irresistible. Indeed, the authors very creatively “use the prediction as a springboard to discuss other, related issues about the place of work in modern-day society” (ms., p. 7).
The predictive failure
Many work attributes and economic, political, and social factors are offered by Kaplan et al., as potential reasons why the prediction of J. M. Keynes failed—that is, that within 100 years (of his writing in 1930) the essential economic problem of humans to earn a living would be so diminished that we would all choose to be working just 15 hours/week.Footnote 1 As comprehensively reviewed by Kaplan et al., Keynes seems to have overvalued the importance of remuneration and material rewards, and greatly undervalued the intrinsic gratification that work can provide.Footnote 2
Economists have, of course, had a field day: “How could it be that a man of Keynes’s intelligence, with a deep understanding of economics and society, could be so right in predicting a future of economic growth and improving living standards and so wrong in understanding the future trends of labor and leisure, consumption and saving?” (Pecchi & Piga, Reference Pecchi, Piga, Pecchi and Piga2008, p. 3). Economists’ answers have tended to focus on macrolevel economic and demographic changes that Keynes did (could?) not have predicted concerning increases in life expectancy, much longer expected periods of retirement, needing to build up savings to pay for the expanded leisure time, and so on. For example, “enjoying a comfortable retirement is facilitated by accumulating assets early in life, and this has mitigated against larger reductions in annual hours of market work” (Crafts, Reference Crafts2022, p. 826).
But one critical factor of particular interest to I-O psychologists that has not received much, if any, attention by economists is an assumption that would seem to be implicit in the Keynes prediction that the growing profits resulting from enhanced technology, increasing efficiency and rising productivity would be distributed equitably, with workers receiving a fair share, hence not needing to work as hard or as much. But workers have not received their fair share. Up until around 1979 in the U.S., average worker compensation kept pace with rising productivity; however, from then through 2019, productivity rose 59.7% while compensation rose 13.7% (Mishel, Reference Mishel2021). During that same period, the percent change in real annual wages for the top 0.1% of the population rose more than 160% whereas that for the bottom 90% rose 26% (Mishel & Kandra, Reference Mishel and Kandra2020). From this perspective it is entirely understandable that the modest “decline in work hours was most pronounced in the earlier- and mid-20th century and has plateaued in recent decades, despite continuing, even accelerating, technological development” (Kaplan et al., p. 5). It is also worth noting that all this has been occurring despite the enactment in 1938 of the U.S. Fair Labor Standards Act, which introduced the minimum wage and mandatory time-and-a-half pay for overtime.
What other factor(s) might have contributed to little reduction in working hours, and to wage stagnation despite soaring productivity and profits over the past 50 years or so? Why has accelerating technological development and productivity led to commensurate increases in profitability and shareholder wealth but not to commensurate increases in wages? (Hint: it’s not an inevitable, “natural” phenomenon.) According to the U.S. Bureau of Labor Statistics (2024), union membership declined by half, from about 20% to 10%, from 1983–2023—from 17.7 million union workers to 14.4 million. The rate is only 6% in the private sector, where working conditions like hours are most susceptible to managerial control. The membership rate in the public sector is almost one-third (32.5%). Moreover, in 2023 as one would expect, the median weekly earnings of non-union workers were only 86% of that of union members ($1,090 vs. $1,263). Given the importance of hours and other working conditions to union–management relations, as well as the historical obliviousness and antipathy of I-O psychology to organized labor (Lefkowitz, Reference Lefkowitz2023; Stagner, Reference Stagner1981; Zickar, Reference Zickar2001, Reference Zickar2004), attention in this regard could be potentially very fruitful.
In addition to the decline in union membership, the situation has been exacerbated by U.S. government policies such as the 2017 regressive tax law and the very modest increases in the federal minimum wage to $7.25/h. in 2009, with no increases since then. In 1980, the FMW of $3.10 amounted to 45.3% of the average nonsupervisory worker wage; in 2020 it was only 29.4% of the average worker’s wage (Economic Policy Institute, 2019). The 2017 tax cuts overwhelmingly benefitted the very wealthiest people and provided virtually no tax relief for workers at or near the bottom in earnings.Footnote 3 Second, many are probably surprised to learn that The Fair Labor Standards Act does not limit the number of hours in a day or days in a week an employee may be required or scheduled to work, including overtime hours, if the employee is at least 16 years old….[nor does it] require vacation, holiday, severance, or sick pay; meal or rest periods, holidays off, or vacations; premium pay for weekend or holiday work; pay raises or fringe benefits; or a discharge notice, reason for discharge, or immediate payment of final wages to terminated employees. (emphasis added)Footnote 4
Conclusion
Kaplan et al., have used the Keynes prediction regarding work and leisure time as a means of posing a fundamental question: “Why do we continue to work as much as we do…[and what might that] tell us about the nature of work itself?” (p. 1). They have reflected on a considerable variety of potential answers from “a certain perspective…informed by I-O psychology (and related fields)” (p. 8). This commentary focuses rather narrowly (and incompletely) on one factor within that I-O domain that they did not consider explicitly: pay equity. One possible explanation for the oversight stems from having conceived Keynes’s prediction in terms of “gains in technology…allow[ing] individuals to choose to work fewer hours” (p. 6, emphasis added). Corporate power, lack of collective bargaining power, and regressive government wage policies have produced the pay equity described above—and worker choice arguably has had little to do with it.Footnote 5,Footnote 6
It would be a considerable understatement to say that the authors have provided I-O psychologists a great deal of food for thought. They have presented an enormous number of likely relevant answers to the question and emphasize the significance of the “tremendous variability across individuals and groups” (Kaplan et al., p. 31) regarding the determinants and effects of work hours; and that “trying to account for work hours is a complex affair, requiring consideration of several levels of analysis” (p. 33). I particularly applaud their effort to encourage I-O psychologists to look beyond our usual individual-level perspective and to pay greater attention to macrolevel, economic, and sociopolitical influences on all aspects of work behavior (Lefkowitz, Reference Lefkowitz2023; Schneider & Pulakos, Reference Schneider and Pulakos2022).
With respect to furthering the authors’ intention of “proposing specific ideas for organizational research” (Kaplan et al., p. 7), given the enormous number and variety of variables presented, it would have been particularly helpful to provide a conceptual structure (e.g., in reviewing the literature and in proposing future research) in which (a) the focal construct working hours is an antecedent variable, mediating or moderating variable, and/or outcome variable; (b) various operational definitions of the construct are illustrated, with their implications (e.g., which are influenced more/less by government policies or normative custom versus under individual control); and (c) complex issues of cross-level analyses are explicitly illustrated (e.g., national level government policies or cultural norms, with working hours measured at the group and/or individual level, nested within organizations, industry and geographic area).
