Christopher Barrington-Leigh

I am grateful for all the deep thinkers and wise people who have taken time to write. Below, I have tried to respond to each of the major themes raised.

First, because the formal, quantitative research on life satisfaction is what underlies almost all I wrote, I must start by suggesting that I did not succeed in conveying to all readers how empirical this field is. Of course, using subjective well-being (SWB) should attract all the normal criticisms of reductionism and utilitarianism. However, this field is not philosophy, nor the introspection of a group of academics. My article is certainly normative, but writing that way is a new venture for me, a quantitative empirical researcher.

Some, like Emily Huddart Kennedy, wanted more of the empirical evidence, but my piece was not a technical summary of my work nor a literature review. The empirical literature in economics and psychology on the determinants of life satisfaction is now truly vast, and appears to be consistent and robust enough to be ready for policymakers in many contexts.¹

Many concerns can likely be alleviated by well-established facts from the enormous body of research into life satisfaction (and other measures of SWB). For instance, the degree to which happiness and even cognitive evaluations like life satisfaction are genetically determined is not a mystery; indeed, it is quite limited. If it were the bulk of the story, we would not find the enormous differences in life satisfaction across individuals (or communities or countries) with different (objective) life circumstances.

Lucie Middlemiss suggests that my claim that we are “social beings” overlooks that we may have different interpretations of what “social” means. In the data I use, no one is asked what they value in life, what they think “social” means, or whether they want to have a more social life. They are merely asked (to paraphrase the life satisfaction question) how good life feels overall, as a single (numeric) answer, as well as other more objective things, including “social capital” measures like how often they see family and friends, and so on. All the conclusions and projections about which I write are based on statistical analysis using this single subjective assessment as the measure of well-being, which leaves absolutely everything to each individual’s priorities and values. When I say “humans appear to be remarkably social beings,” I am referring to specific statistical findings, now repeated countless times by different researchers with different datasets, now comprising millions of respondents and more than 150 countries. To the extent that a particular “social” experience (believing local strangers are likely to return a lost wallet, frequency of contact with family, etc.) comes out of the statistics as important for life satisfaction, that is because it is a commonality across people. And some of these findings, as I’ve recounted, are overwhelming.

Individual differences are the very foundation of the life satisfaction approach, because it is up to the individual to evaluate her life based on whatever she deems important (the more cognitive factors likely to influence her answer), combined with whatever drives the frequency with which she feels positive emotions (the more affective factors likely to influence her answer).

Middlemiss says, “It could be anything from greeting your neighbor on the street, to using fashion to express your sense of identity and belonging, to being a community activist.” Agreed! To the extent that we also happen to have measures of each of those in the same dataset (and let’s, over time, include anything we suspect is important), we can test how important each is for explaining differences in life satisfaction—on average, or for particular subsets of the population (income ranges, geographic locations, gender, race, etc.). In fact, questions about neighbors, about identity and belonging, and about community involvement are all common, as are more conventional ones about income, labor, etc. These are not questions evaluating the importance of those things; they simply ask about experiences (or feelings, in the case of belonging) as objectively as possible, and later on, the statistical analysis is tasked with evaluating their “importance” for contributing to overall life evaluation.

“Consume less and you will be happy” is not my message, and never have I seen empirical evidence for such a finding, despite some of what makes it to the press. I advocate simply (a) that we should build good lives, subject to some (many) material constraints, (b) that we can get from here to there without living worse lives along the way, and (c) that this should be our story to get everyone on board.

Unlike rhetoric from the past (“women are happy being in caring roles,” “civilizing the natives will make them happy”), the SWB literature is empirically accountable. However, this does not mean that it cannot be abused or, even more likely, that there are some findings which would put policymakers in awkward positions of unwanted paternalism when pro-well-being policies conflict with people’s intuition. Luckily, we can simply ignore those policies. There is plenty to do.

Deric Gruen is skeptical that “happiness can provide much insight without complementary quantitative metrics.” I agree wholeheartedly about the complementarity. It would be meaningless, policy-wise, to pursue SWB without understanding the objective, policy-amenable levers which appear, empirically, to shift it.

For the record, though, Gruen also calls happiness qualitative. In fact, the measures of SWB I refer to, with life satisfaction being of most interest, are quantitative. Anamaria Aristizabal also raises this question of subjectivity (a special thanks to her for her uplifting and beautiful contribution).

At the individual level, life satisfaction is quantitative yet subjective (strange—i.e., possibly unfamiliar—but true). However, once collected, these data at the aggregate level also provide perfectly objective (repeatable, verifiable) and quantitative facts. That the average (or distribution of) life satisfaction that people will report in your community is measurably, repeatably, and statistically-significantly different from that of the community on the other side of the tracks is an objective fact.

However, as Aristizabal’s response highlights, the language itself is powerful. In many places where SWB data are being taken very seriously by people who like statistics (e.g., UK at the national level), the discourse surrounding re-envisioning metrics for social progress tends to open up a space for people to more freely reflect and report what they think their values are.² The data derived from analysis of SWB may have something different to say, because there is no reason people should have perfect knowledge about how happy they would be in other potential realities, but in many places, this process has legitimized new priorities, even without deriving them directly from analysis of the SWB data. The more time we spend thinking about what well-being means to us, the better, and results from the science of it can help to inform our beliefs (and further displace any non-empirical preconceptions we may have absorbed from the existing cultural paradigm).

Reading Aristizabal’s eloquent account of her work, it occurred to me that what we need/what I am advocating is just more “life coaching” for society at large (as well as individually, starting in school). Life satisfaction research can provide evidence-based advice to individuals, to take or leave. And it can provide evidence-based advice to policymakers, who must figure out what outcomes they are ultimately accountable to.

Sandra Waddock provides examples of three well-being-oriented indicators. I have recently done a survey of these and more than seventy others which may be useful to readers interested in the relationship between objective and subjective measures, how they can be used together, and what trends are evident in the use of these indicators to inform policy.³

Middlemiss muses that if we are fixated on happiness (i.e., SWB), then unhappy people become “problems.” She says that we should instead focus on the things that we think might be an intermediate objective for aiming at happiness (inequality, education, health, environmental sustainability, etc.). I agree, absolutely. SWB measurements provide a way to derive that list of objectives, assign relative importance to each item, and monitor overall success, but ultimately, the items in that list are usually measured more directly and objectively in order to guide policy in detail.

As I stressed in my essay, “pursuing well-being doesn’t ensure environmental sustainability.” Conflating them degrades both objectives. If readers came away thinking I advocate moving “to new measures of well-being as a sustainability strategy,” as Deric Gruen wrote, then my exposition certainly failed.

My perhaps-simple view of sustainability is that we should impose a large set of legally enforced constraints on material flows (1970s Herman Daly), while pursuing, within those constraints, things which are empirically shown to support good lives (life satisfaction). My argument, then, is about whether a shift towards that scenario of collective constraints can be carried out with steady improvements of SWB. This can be contrasted with the question of whether individuals can find their own trajectory of Tim Kasser’s pro-ecological behaviors (“PEB”s) while benefiting as individuals.

In fact, I would never equate “living an ecological lifestyle,” defined by personal consumption choices, with sustainability. In my conception, the ecological constraints must go far beyond such mechanisms and are ultimately things which must be collectively imposed (“Mutual Coercion, Mutually Agreed Upon”). Individual behavior has limitations in scalability, equity, and accountability. It is, above all, characterized by the basic collective action problem known as the Tragedy of the Commons.

By contrast, collectively embraced constraints do not require individual initiative of the sorts described by Kasser. Of course, those PEB actions are associated with political change required for new constraints to be imposed, and in my view, participating in PEBs even changes one’s own values (rather than just the other way around), paving the way for making future collective rules.

A not-uncommon intuition in response to the literature on life satisfaction is that it must be about, or come from, what Deric Gruen calls “white, wealthy environmental circles.” I don’t think I should speak for other groups, nor impose identities on colleagues who might not fit that preconception. However, I believe the attitude of wanting to change a given (unsightly?) material situation (“Those poor people need a roof, first!”) is in fact one judgmental and colonial attitude which SWB research can free us from. (Basic needs for the poor is, of course, the usual justification for outsiders to extract resources.). Indeed, research and practical experience both say, in countless contexts and interventions, that any change one is trying to make in any community or society can be done better when attention is paid to the ultimate outcomes that matter, like making better lives, rather than narrowly focusing on preconceived material (income, particular physical health measures, etc.) goals. Example after example of programs aiming to improve employment/income/health/education/incarceration/development shows that you can do better by thinking first about human outcomes experienced by the people involved. This informs the process as well as the ultimate attention and resource allocation of policy.

Besides (to Tadhg OʼMahony and Sandra Waddock’s points), in terms of life satisfaction, the data show clearly that “poor” and “unhealthy” people and societies are not happy, so who says we are at risk of trading off “grinding poverty” for something else?

Emily Huddart Kennedy (thanks for her careful reading, insightful summary, and clear articulation of the common reaction about class bias and inequality) introduces us to Jim and Charles as a way to express a similar reaction to my exposition. I am advocating for finding a path that involves ecological limits yet is not frightening to people. Thus, when I write that “a reduction in potential income will not necessarily reduce well-being (particularly for those in upper-income brackets),” it can be taken to mean that there is an opportunity to redistribute more while improving lives at all strata, however you slice them.

Thinking about issues of race, class, and gender is always important and valuable in any context. However, imposing ideals like “local food” onto the happiness literature is unfair unless we see rigorous studies in which life satisfaction has been shown to vary with “local food” access, independently from other standard (including demographic) variables. If not, this may be a straw man. And, lest this be misread again, the literature does not say that material status or health is not important also. It simply says that there are relatively material-cheap ways to improve lives.

Life satisfaction data can continue to inform us of which life conditions are most important in practice (universally or for particular groups), so as not to rely on some set of specific ideals of wealthy white males. Ultimately, lessons from life satisfaction will be useful if they are relevant to the bulk of the population. To the extent that I have not, in this very condensed format, managed to suggest why that is likely or possible, more work must be done. As Emily Huddart Kennedy suggests, such work involves both research and normative efforts.

On the related issue of inequality, Anders Hayden expressed concern that a well-being-derived argument would focus more on mental health services while doing less to address material poverty. I would say yes, let’s follow the science and outcomes on this, without immutable preconceptions. The next dollar should be spent on whatever will do the most good. If that is addressing mental health before addressing “material poverty,” then so be it. Now, this sounds like a straw man to me, since housing, income, and mental health are intricately linked. Moreover, no matter which you are trying to address with a given policy, the well-being literature has plenty to say on how to go about it in order to have extra (free) co-benefits for well-being.

While everything I wrote about SWB was based on quantitative research, the part about “zero marginal cost” was speculation and extrapolation. Sylvia Lorek and Deric Gruen objected especially to my example of 3D printers as a decentralizing force in production. Putting aside technical quibbles (we can nowadays put more things than imaginable as “ink” in 3D printers, including living cells to build tissue/organs), I believe we need not be fixated on one possible bad outcome when envisioning the option space. This is the difference between backcasting, which the link Lorek shares advocates, and forecasting. My essay was never intended to make predictions. I did not state that a technology will reduce resource consumption. I pointed out that technological shifts can become drivers of big social change (one way or the other), and that to the extent current trends might reduce centralization of production, they may allow us to reflect more independently on what we want, and they may naturally increase the role for collectively made decisions. I certainly did not advocate having everything made by cookie cutters for us. In fact, I claimed that creativity may become less fettered by capitalist motives. Maybe the “Global Village Construction Set” is a better illustration of the democratization that some are thinking about in the context of 3D printing?

I have enormous apprehension about the equity impacts of genetic engineering and artificial intelligence. But societies that are following the social objectives I advocate will be the ones best able to deal with these impacts, to form appropriate policies, to have the heart for appropriate redistribution, etc. Dignity and trust are important for getting anything done (democratically), and I am only pointing out that (a) opportunities await us and (b) positive future stories that may help enable the right kind of engagement and investment (collective, rather than defensive).

Technology will not save us; it responds to constraints. For example, I am currently advising my university to implement a policy for curtailing our academic air travel until such time as air travel is electric or sustainably biofuel-based.

On the topic of optimism, I share Anders Hayden’s concerns about the reality of my hope that we may be able to improve life quality monotonically while capping and reducing material impacts. Hayden gives the example of air travel. However, I think the result if we don’t embrace positive human outcomes will be vastly less material restraint than if we do.

Indeed, given that succeeding in implementing “needed” environmental limits is by no means assured, it may be that it is not politically possible to cut consumption more than can be compensated by better policies for whatever society values as well-being. However, I don’t believe in that absolute, either. Possibilities are many, and we must strive for the best meaningful outcome that appears possible.

1. Not all experts agree, however. See “Understanding the Effect of Policy on National Wellbeing,” Centre for Economic Performance, London School of Economics and Political Science, March 2017, http://cep.lse.ac.uk/wwp/past_survey.asp?num=4.
2. Christopher Barrington-Leigh, “The Role of Subjective Well-Being as an Organizing Concept for Community Indicators,” in Community Quality-of-Life Indicators: Best Cases VII, edited by Meg Holden, Rhonda Phillips, and Chantal Stephens (New York: Springer, 2017), 19-34, http://wellbeing.ihsp.mcgill.ca/publications/Barrington-Leigh-CIC2016-SWB-community-indicators.pdf; Christopher Barrington-Leigh and Alice Escande, “Measuring Progress and Well-Being: A Comparative Review of Indicators,” Social Indicators Research (December 2016): 1–33. 
3. Ibid.