Thomas H. Davenport and Julia Kirby, the authors of Only Humans Need Apply, favor a job guarantee (JB) over a universal basic income (UBI). In this first part of a three-part article, I review their main argument, and assess one their central claims: the supposition that joblessness causes people to be less happy (irrespective of income).

Thomas H. Davenport and Julia Kirby, the authors of Only Humans Need Apply (2016, HarperCollins), believe that automation will radically change the structure of work. However, they further maintain that there is no need to fear a robot job apocalypse — in defiance of the picture painted by the popular CGP Grey video Humans Need Not Apply, to which the book’s title alludes. On their assessment, humans and machines will be able to work together in the new economy. Indeed, Only Humans Need Apply consists largely of strategies that workers can pursue to reduce their risk of losing their jobs to machines.

Davenport and Kirby are not proponents of universal basic income (UBI). Instead, as they mention briefly near the end of their book, they favor a job guarantee (JG) program. In this article, I will review their main argument against UBI, and present what I take to be its major shortcomings.

Something, though, should be said upfront: for the purposes of this article, I am happy to grant Davenport and Kirby’s premise that there’s no robot job apocalypse on the horizon. Indeed, I agree that it’s plausible that people will continue to create abundant opportunities for paid employment despite increased automation; after all, they have been doing so for decades already. However, as I’ve stressed elsewhere, there are plenty of reasons to support basic income that have nothing to do with automation; we don’t need to fear the rise of the robots. For one, I believe that a UBI should be demanded in part to free individuals from the need to sell their labor in the marketplace.

It’s on this last point that I markedly disagree with Davenport and Kirby, who hold that jobs are good. More specifically, Davenport and Kirby maintain that jobs are good for those who work at them — and for more than just income. It’s largely for this reason — which I’ll present in more detail below — that they eschew UBI in favor of JG.

1. The ‘Jobs Have Non-Monetary Value’ Argument

Davenport and Kirby devote only one chapter of Only Humans Need Apply to government policy, and only one short section of this chapter to universal basic income (pp. 241-243). (In fact, most of the relevant content was reprinted as an article in Fortune.)

Davenport and Kirby are quick to dismiss UBI — which they believe “misses the point” by neglecting the non-monetary value a jobs. It is worth noting what their complaints against UBI are not. For one, they do not worry that a UBI would be too expensive. Indeed, they assert that “the huge gains in productivity will mean we could afford, as a society, to go in either direction” of a UBI or JG (p. 243). Nor are they convinced that a guaranteed income would encourage laziness. They are clearly skeptical (as we’ll see later on), but at the same time they concede that this question is an empirical one — and go so far as to “applaud” cities like Utrecht for their willingness to experiment (p. 242).

For Davenport and Kirby, the deciding factor seems to be that (on their view) a JG provides benefits to individuals that exceed those provided by a UBI. What benefits? Well, jobs.

Jobs bring many benefits to people’s lives beyond the paycheck, among them the social community they provide through having coworkers, the satisfaction of setting and meeting challenging goals, even the predictable structure and rhythm they bring to the week. In 2005 Gallup began conducting a global opinion survey called World Poll. Analysis of the responses reveals that people with “good jobs” — which Gallup defines as those offering steady work averaging thirty or more hours per week and a paycheck from an employer — are more likely than others to provide positive responses about other aspects of their present and future lives.

Another World Poll question presents “aspects of life that some people say are important to them” and asks respondents to categorize each as to whether it is something essential they could not live without, very important, or useful but something they could live without. Gallup chair Jim Clifton says that by 2011, “having a good quality job” had reseach the top globally — putting it ahead of, for example, having a family, democracy and freedom, religion, or peace (pp. 7-8).

Work has value in itself as a way to find meaning in life. As we’ve noted, having a good job is the most desired thing in the world in global polls. Freud said that, “Love and work…work and love, that’s all there is.” Many studies have found that unemployed people are less happy, and that compensating them anyway doesn’t make them as happy as putting them back to work (p. 242).

To be sure, Davenport and Kirby are at times a bit flip. The Freud quotation, for instance, merits no more of a rebuttal than “The Beatles said, ‘All you need is love, love; love is all you need.'”

However, the basic worry — that work brings value to life that’s not provided by income alone — is indeed an important concern, and it merits a serious reply. Indeed, as I will concede later, it’s possible that, for some individuals, a JG would provide more benefit than a UBI — due precisely to the fact that jobs provide many people with rewards other than a mere paycheck. But I will argue that, all things considered, this is not a persuasive reason to favor a JG to UBI.

In particular, I want to highlight three main shortcomings of Davenport and Kirby’s argument:

1. It is inappropriate to extrapolate the results of studies like those in question (viz., surveys of unemployment and unhappiness) to a society with UBI. This is because a UBI itself might engender importantly different norms, values, and societal expectations.

2. Such generalizations ignore the fact that many individuals are discontent in their jobs, and that some would be happier (and more productive) if not confined to any traditional job. A UBI, but not a JG, would help such individuals immensely.

3. It’s important not to ignore that a UBI does not compel individuals to stop participating in paid employment; thus, it would not hurt those people who do have jobs and value them.

The decision between a UBI and JG must not construed as a choice between a society in which most individuals lack jobs and one in which they have them. After all, studies of guaranteed income, such as the Mincome experiment in Dauphin, Manitoba, have shown little negative effect on employment. It is a choice between a society in which those individuals choose jobs at their own will and one which the majority are forced to take jobs out of financial necessity.

In this article, I will focus on the first consideration above. Since UBI does not “cause” employment, it might seem that this issue is a red herring; nonetheless, it is illustrative to scrutinize some of the reasons that the extrapolation of results on “unhappiness” is unwarranted.

In a second article (to be published), I will turn to the second and third.

2. The Extrapolation Worry

Go back to one of Davenport and Kirby’s more provocative claims: “Many studies have found that unemployed people are less happy, and that compensating them anyway doesn’t make them as happy as putting them back to work” (p. 242). Let’s grant that this is true, and that there were no fatal flaws in the design of the studies. My claim is that, nonetheless, it’s inappropriate to extrapolate these results to a society in which a UBI has been enacted.

Suppose that, in general, unemployed people are less happy than employed people, and that this difference in happiness cannot be accounted for merely by the loss of income. Can we conclude that a job guarantee should be favored over a universal basic income?

I believe that the answer here is clearly no. This is because, in our actual society, there are other variables that are confounded with the presence or absence of paid employment. For one, to lack a job is to have a stigma (perhaps especially if one is receiving income without working). Secondly, jobs often play a large role providing individuals with a sense of meaning and personal identity. No doubt that bearing a stigmatized identity can contribute to unhappiness, as can the lack of a sense of identity and purpose. But what I want to stress here is that these correlations are present in our actual society — and there is good reason to believe that these connections would be attenuated by the institution of a UBI itself.

Put otherwise: The correlations between unemployment and stigma, and between employment and self-identity, are products of our society and culture. They would not necessarily hold in a society with a UBI. Thus, if it’s really the stigma and/or the loss of identity that causes many unemployed people to be unhappy, we can’t conclude that unemployment would have this same effect were a UBI to be put in place.

2.1 Unemployment and Stigma

Take the fact that unemployment is stigmatized. Some observations do suggest that stigma does play a role in contributing to the unhappiness associated with unemployment — such as the following three. First, retirees are not disproportionately unhappy. Second, in one study designed to test the stigma hypothesis, it was found that unemployed individuals enjoyed a boost in life satisfaction upon reaching retirement age, even though (curiously) employed individuals did not [1]. Third, if unemployment makes individuals unhappy, then the average happiness levels within societies should be expected to decrease when unemployment rates rises; however, this has been found not to happen [2]. Davenport and Kirby would argue that everyone wants a happy retirement, so whether being unemployed beforehand or not, wouldn’t or shouldn’t effect this. To be honest, retirement is closely linked to the age of senility. Therefore, this becomes the time for many to enjoy the last few happy years on the planet. That said, expecting a happy retirement is a normal human tendency. Many people expect a retirement devoid of the fear of writing a will or maybe getting in touch with a probate attorney in Denver (or nearby places) in advance to take care of the estate administration in case of death.

Anyway, these three observations can’t be explained merely by the hypothesis that people are unhappy when they lack a job. They can, however, be explained by the “stigma hypothesis”. Take the first two observations: although our society expects that able-bodied, non-elderly adults are employed full-time, this expectation does not hold for individuals past retirement age; “retirees” is a socially-acceptable, non-stigmatized category. Additionally, the third observation could be explained by noting that, when the unemployment rate is higher in a region, unemployment becomes more familiar, and thus ceases to bear as much of a stigma.

And we should add that there is also stigma associated with the receipt of “handouts”. Thus, if stigma is a large factor in the mechanism by which unhappiness contributes to unemployment, we should not expect that merely giving money to unemployed individuals would restore their happiness. If the money is perceived as a “handout” — a symbol of personal unfitness or inadequacy — then to accept it is to assume a stigmatized identity on top of a stigmatized identity.

That is our present world. Now, consider a world with UBI. In this world, a basic income is awarded to everyone. There is no need for anyone to prove their neediness in order to receive enough money to live, nor must anyone demonstrate that they are physically or mentally unable to work, or that they seeking employment but unable to find it. Because of the universality, there would be no stigma attached to individuals’ receipt of this form of cash assistance. Receiving the benefit could not in itself be construed as evidence of personal inadequacy.

Of course, it’s possible that individuals who did not work, living upon the basic income alone, would still be stigmatized in UBI-world. Perhaps they would still be branded as lazy, freeloaders, or incompetent to find work. However, in a society in which all individuals are guaranteed an income sufficient to meet their basic needs, some highly driven individuals might voluntarily opt out of the paid workforce in order to concentrate solely on their passions — artists and writers, independent researchers and open-source programmers, political activists and volunteers for humanitarian projects, and so on. If such individuals are numerous, successful, and productive, then unemployment might cease to be stigmatized, and could instead come to be regarded as a socially acceptable life-choice.

The final point relates directly to the second “confounder”: as a matter of fact, paid jobs provide a sense of meaning and identity to many people; however, this is not necessarily the case.

2.2 Employment versus Meaningful Work

Davenport and Kirby are surely correct that jobs function as an important “a way to find meaning in life”: for many people, having a job is a way to feel like one is providing some sort of important social contribution. And even those who find their jobs unfulfilling, perhaps even meaningless, might find in their job and career a source of self-identity. In American culture, a customary question to ask upon meaning a new acquaintance is “What do you do for a living?” Relatedly, a customary question to ask a child is “What do you do want to be when you grow up?”

In our current culture, job-centrism starts young — and persists. You might be interested in temporary work such as Interim Director Jobs. Maybe it is unsurprising that our culture should be this way: if jobs that consume most of our adult years are inevitable for us (given its financial necessity), then perhaps we might as well accept our jobs as core features — often the core features — of our personal identities.

But there seems to be nothing intrinsic about paid employment such that it should be more central to our self-identities than, say, unpaid work. Indeed, later in their section on UBI, Davenport and Kirby themselves state that volunteer service also “leads to greater happiness” (p. 243). Unemployed individuals seldom turn to volunteer service as a way to replace the lost non-monetary rewards of paid employment (I assume); however, this itself is plausibly an effect of society’s job-centrism. In our society, it is assumed that an unemployed, able-bodied person ought to devote as much effort as they can into searching for a new job; exclusive dedication to volunteer work, unless it is clearly a possible path to paid employment, is likely to be seen as imprudent and a waste of time.

Once again, however, a UBI might engender an entirely new culture — one which in more people, freed from the inevitability of full-time employment, turn to non-paid work to make their lives meaningful and valuable. This, I believe, is by no means an improbable effect of a UBI. After all, even in our present society, some individuals don’t turn to jobs for meaning and purpose — or would prefer not to. (I will return to this point in Part 2 as well.)

To take just one example, Zipcar CEO Robin Chase spoke of her research on “passion jobs” a recent White House roundtable discussion on automation an UBI. She has interviewed individuals from a cab driver who wrote music that made autistic children happy (but could not afford to pursue this passion full-time) to a computer programmer who slept on friends’ couches probably while being in the process to install git to write open-source software for 3D printers. In her informal research, she has encountered many people who are unable to pursue socially valuable and personally gratifying projects, simply because these projects are not financially lucrative; instead, these people are stuck in “crummy jobs”, detached from their passions. Chase herself supports a UBI as a way to allow individuals to pursue vocations that would give their lives much more meaning than the jobs to which they must resort for income.

If she is right, then a UBI might create an environment in which individuals routinely turn to voluntary work or other unpaid activities for meaning, fulfillment, and self-identity.

Similar points apply to other non-monetary benefits that many individuals derive from traditional jobs, such as those mentioned by Davenport and Kirby earlier in the book: “the social community [jobs] provide through having coworkers, the satisfaction of setting and meeting challenging goals, even the predictable structure and rhythm they bring to the week.” (I will return to similar points in Part II of this multi-part editorial.)

Presumably, most people are less happy when they lack engagement in projects, causes, and communities. In our present society, where jobs are (of necessity!) central to most people’s lives, lacking a job can mean lacking such a project, cause, or community. Again, however, this link is contingent — and could be severed through the institution of UBI itself. Thus, we can’t conclude that unemployment would have the same effect on unhappiness given UBI.

TL;DR –

Even though unemployment is correlated with unhappiness in modern developed societies (even when controlling for monetary factors), this result is plausibly a reflection of the job culture itself: it’s not that our culture values jobs because jobs intrinsically make us happy; it’s that being employed tends to make us happier because we are stuck in a culture that values jobs [3].

While the relationship between unemployment and unhappiness is no doubt highly complex — involving the interplay of more factors than we can reasonably discuss in a short response piece — the acknowledgement of the factors described above should at least lead us to question the appropriateness of invoking such studies in an argument for the superiority of a JG to UBI.


[1] Clemens Hetschko, Andreas Knabe, Ronnie Schöb (May 4, 2012) “Identity and wellbeing: How retiring makes the unemployed happier“, CEPR VOX.

[2] Cf. Petri Böckerman and Pekka Ilmakunnas (2006) “Elusive Effective of Unemployment on Happiness”, Social Indicators Research 79: 159-169.

[3] For more discussion of the non-naturalness of our modern notion of “work”, see this recent article: Ilana E. Strauss (Jun 8, 2016) “Would a Work-Free World Be So Bad?The Atlantic.

Davenport and Kirby: Full Bibliographical Entries

Thomas H. Davenport and Julia Kirby (2016) Only Humans Need Apply: Winners & Losers in the Age of Smart Machines, HarperCollins Publishers.

Tom Davenport and Julia Kirby (May 26, 2016) “What Governments Can Do When Robots Take Our Jobs“, Fortune Magazine (and reprinted in Yahoo Finance).


Reviewed by Tyler Prochazka

Photo (“Workers”) CC BY 2.0 Daily Sunny

Kate would like to thank her supporters on Patreon