A Modest Proposal for Solving the Older Worker Productivity Problem

Abstract

Following the satirist Jonathan Swift (https://www.gutenberg.org/files/1080/1080-h/1080-h.htm), I outline a “modest proposal” for managing the older worker productivity problem. Although meta-analyses by psychologists have consistently shown little or no relationship between age and worker productivity, older workers are clearly less productive from an economic perspective, given their average higher salary and, in the United States, their higher health care premiums. Building on Swift’s satire, I propose that to satisfy profit-hungry capitalist firms, older worker contracts should include an automatic salary deflator past age 50. This would improve older worker productivity from an economic perspective by reducing the denominator in their productivity index (output/input costs) and add to the stock of the common good.

In this commentary, I turn for inspiration to Jonathan Swift, the noted satirist, whose “modest proposal” to deal with the Irish poverty and famine problem was turning Irish children into food products in a 1729 essay entitled: “A Modest Proposal For preventing the Children of Poor People From being a Burthen to Their Parents or Country, and For making them Beneficial to the Publick.” His ironic solution was: “A young healthy child well nursed, is, at a year old, a most delicious nourishing and wholesome food, whether stewed, roasted, baked, or boiled; and I make no doubt that it will equally serve in a fricassee, or a ragout.” In this spirit, let me outline my “modest proposal” for dealing with the problems encountered by older workers who wish to stay employed longer than their employers want them.

First, I need to trace out how we arrived at the older worker retirement problem. It wasn’t that long ago that social insurance schemes were invented to allow older workers to retire with a guaranteed income (https://www.ssa.gov/history/briefhistory3.html). The wily Prussian Chancellor, Otto van Bismarck, set pensionable age for veterans at age 70, unreachable by most workers in 1889. However, for the first time in history, (a few) nonwealthy people could avoid working until they died by having a public means of support. The dawning of both the industrial revolution in the 19th century that moved workers from farms to factories, and the age of qualifying for retirement pensions in the 20th century, led to new ways to envision work and older workers.

Following the introduction of Social Security systems in the 20th century, the average age of retirement kept decreasing. For instance, U.S. workers started to take partial pensions at age 62 rather than waiting for full pension entitlement at age 65. This early retirement trend reversed over the past three decades (Munnell & Chen, 2015). One influence was likely the shift from defined benefit to defined contribution work pensions that moved risk from the firm to the worker. Baby boom cohorts have seen severe stock market crashes, such as that during the Great Recession, no doubt generating heightened fears of income insecurity in retirement. Coincidentally, today’s workers have expressed a desire to work longer (Gallup’s Economy and Personal Finance Survey, 2017).

Now, if workers wish to work longer, why don’t employers honor that desire? And why have policies such as the EU’s Lisbon Strategy aimed at retaining older workers, setting a goal of 50% employment of those aged 55–64 by 2010, been so unsuccessful (https://ec.europa.eu/social/BlobServlet?docId=15685&langId=en)? After all, decades of research have consistently shown that there is little or no decline in work productivity as a function of age (McEvoy & Cascio, 1989; Ng & Feldman, 2008; Waldman & Avolio, 1986). So why don’t more employers promote people working into later years?

I believe that the main problem is that the industrial-organizational psychology research community does not define productivity the same way that economists and business owners do. See, for instance, recent fretting over potential declines in total factor productivity because of the aging workforce (Aiyar et al., 2016; Maestas et al., 2016). Although age does not predict worker output or work quality ratings, age is a pretty good indicator of the size of the average industrial wage, (Charness, 2019), part of the denominator in an economic productivity index. Businesses tend to adopt an “output/input cost” definition of productivity. So if Joe at age 25 with an annual salary of 25,000 USD turns out the same number of widgets as Jim at age 65 with an annual income of 35,000 USD, it is pretty clear that Joe is more productive than Jim from an economic point of view, even though the two are equally productive for psychologists studying the relationship between output and age.

Indeed, there exists the interesting paradox that shows that firms are reluctant to let go of older workers during downturns in the economy compared with younger workers. However, when older workers are laid off, other employers are very reluctant to hire them. Hence, those workers suffer longer periods of unemployment before being rehired compared with younger workers and re-enter at much lower salaries than they commanded earlier (Johnson & Park, 2011).

One explanation offered for such contradictory valuation is the delayed payment contracts theory (Hutchens, 1986). It suggests that employers underpay younger workers and overpay older ones to prevent workers from shirking during lifetime employment. Workers, presumably, continue to work at peak productivity to maintain employment long enough to earn overpayments in later work life. Indeed, when job seniority is controlled for, older workers may be more likely to lose employment during downturns (Johnson & Mommaerts, 2011). Whether lifetime employment contract theory holds today in our fast-paced, internationally competitive workplace is an open question. Lifetime employment is slipping away from most workers. Tail-end baby-boom workers have changed jobs over a dozen times on average (News Release USDL-17–1158, Bureau of Labor Statistics, 2017) and hence likely are paid competitively at all stages of work life.

The other factor working against employment security for aging employees, at least in countries such as the United States, is their increasing cost of employee health plans. By the time U.S. workers reach 65 years of age, about 75% of them are likely to have multiple chronic medical conditions (https://www.cdc.gov/chronicdisease/about/multiple-chronic.htm). So even if Joe and Jim (above) turn out the same number of widgets and make the same salaries, younger worker Joe is more economically productive because he costs the company health plan a lower premium. Realization of the greater cost of older workers drives employers in the direction of replacing older workers with younger, healthier, cheaper laborers. Hence there is a need for legislation such as the U.S. age discrimination in employment act (ADEA: https://www.dol.gov/oasam/regs/statutes/age_act.htm) to try to prevent employers from acting on their profit-seeking motive. It apparently doesn’t always succeed (https://features.propublica.org/ibm/ibm-age-discrimination-american-workers/).

Now, as a basic economics course would teach us, the high seniority salary for older workers might be well-earned and represent a fair market price for the experienced older worker’s more efficient labor, at least in the case of perfect markets and all actors maximizing expected utility. Alas, as one of my mentors, Nobel Laureate Herbert Simon pointed out many years ago, perfect rationality is beyond individuals (and firms) because of their limited computational ability (Simon, 1955). As a result, people satisfice, or find good enough solutions, rather than optimize. People (and the firms they compose) have bounded rationality, and market inefficiencies are bound to develop. Subsequent Nobel prizes to Daniel Kahneman and Richard Thaler suggest that people behave every bit as irrationally as predicted.

With this backdrop, I return to the case for how to prolong worker longevity. It is really quite simple: pay them less as they age rather than more. Eliminate the “stickiness” of salaries through the ratcheting process that makes salaries go up and up, rather than up and down. Cross-sectionally speaking, salaries peak in a person’s 50s and start to decline (Charness, 2019), suggesting that adjustments are there to be made. Employers and employees could agree to build in an age-related salary deflator factor into their contracts. If aging workers disagree or change their mind on that deflator, they are of course free to go elsewhere. If employers believe that a worker is more valuable over time, they are free to offer additional salary to avoid losing the worker to a competitor. But, that deflator provides firms with a stronger incentive to retain even a worker who maintains steady state productivity as their increased health insurance cost will be balanced by their decreased salary cost. It also might induce a firm to retain an older worker who is showing declining productivity if the decline is balanced by a declining salary plus the avoided replacement cost (searching for, hiring, and training a replacement).

Now, this scheme of deflating salaries over time is predicated on the idea that lifetime employment with underpayment when young and overpayment when old is no longer an effective way to prevent shirking for those who remain with a firm for an extended employment period. An age-based deflator policy also is predicated on health insurance costs remaining a significant barrier to retaining older workers, which could be upended if universal health care coverage by the state replaces current work-related health insurance provision in the United States.

Such a policy also has a steep hill to climb legally because such contracts might violate the ADEA by using an age-based deflator factor. An age-based wage-deflator policy also chafes against the notion of equal pay for work of equal value, though that phrase belies the difficulty in finding perfect measures of work productivity, a problem plaguing the field of productivity evaluation for psychologists (e.g., Salthouse & Maurer, 1996) and economists alike. Finally, it is not immediately clear at what age to initiate a deflator: 40, 50, 60? The basic research on age and peak productivity suggests inflection points that vary by fields (Simonton, 1990) and there are clear tradeoffs between increasing knowledge and decreasing fluid ability resulting in inverted J-shaped productivity functions over career age. No doubt some very good research and consulting opportunities could be available to industrial-organizational psychologists to fit deflator functions to specific types of jobs and firms. Notwithstanding all these difficulties, this modest proposal to deflate older workers’ salaries should not be dismissed out of hand.

Returning to my muse, Jonathan Swift, I would argue that to the discerning palate this modest proposal may prove to be a tasty option. Much like those plumped up Irish toddlers, decreasingly expensive aging workers might be hard to resist for profit-hungry capitalist firms. To quote Swift: “I profess, in the sincerity of my heart, that I have not the least personal interest in endeavoring to promote this necessary work, having no other motive than the public good of my country, by giving some pleasure to the rich.” As a septuagenarian worker myself, I am clearly ready to see such a policy implemented in my work place to support the common good—or maybe not.