Regardless of the motivation and intention behind it, the move to base government policy on measures of happiness is fraught with problems at every level, argues Mark D. White. Reorienting policy towards happiness suffers from problems of definition, measurement, and implementation.
The prospect of basing government policy on measures of happiness (or subjective well-beingas it is known in the academic parlance) is an appealing one for government leaders and citizens alike. At the most basic level, its rhetorical appeal cannot be denied: who can deny that happiness is a good thing? Ideally, focusing on happiness rather than narrower economic measures such as gross domestic product (GDP) would make policy more humane. It would avoid abstractions like income per capita and emphasize who gets that income and what do they do with it. Fundamentally, it would steer policy towards what truly matters to people rather than a rough, imprecise proxy for it.
Unfortunately, few of these aspirations will be realised by reorienting policy toward happiness, which suffers from three problems: issues of definition, measurement, and implementation. Each of these problems is fundamental in itself, and each successive one builds on the preceding.
First, “happiness,” like “justice” or “beauty,” is a notoriously vague concept. Philosophers have struggled for centuries to define it, and poets, novelists, and songwriters have tried to describe it, but no canonical statement has yet emerged. Most economists and psychologists who study happiness acknowledge this difficulty and either specify a meaning or leave it open for survey respondents to answer for themselves, both of which are fraught.
The problem is that there is no way to ensure that researchers and research subjects, or policymakers and citizens, understand “happiness” in the same way. Each of us has a sense of what it means, but given the vagueness of the concept, that understanding varies between different people as well as in different situations. If I ask my friend if she’s happy with her lunch or if she’s happy with her life, I’m asking about two very different things, never mind how she might interpret my question differently from what I meant (especially in the latter case). If survey questions leave the definition of happiness open, each respondent may be reporting on a different personal conception of happiness, and if researcher specify one conception of happiness, it will likely not correspond to what each respondent regards as happiness. Either way, researchers do not get an accurate and unified picture of what people regard as their happiness or what matters to them.
Second, measurement of happiness is difficult because of its essentially qualitative and subjective nature. Even if the definition of happiness were not an issue, accurate measurement would still be problematic because of how surveys are designed. Typically, respondents are asked to report some version of their happiness on a numerical scale with descriptive labels attached to each step. For instance, the scale may go from one to five, with one labeled “very unhappy,” two “somewhat unhappy,” up to five, “very happy.” Sometimes the endpoints are signified as “the most [or least] satisfied you can imagine,” in an attempt to normalise the measure to each person’s life circumstances.
There are numerous problems with this. For one, the descriptive labels themselves are vague; even if the definition of “happy” were clear, modifiers such as “somewhat” and “very” are not, and mean different things to different people. Also, the endpoints raise issues of their own: labels aside, does the lowest point signify the absence of happiness (which would seem like zero) or profound unhappiness (which would seem to merit a negative number)? Furthermore, people in different circumstances would have very different ideas about the best or worst life they can imagine, which turns an intended feature into a flaw. An heir to an industrial fortune may see the stars as the limit to his life, while a dishwasher at a greasy spoon may have more modest dreams. This raises the much larger issue of adaptation, the process by which people acclimate themselves to adverse circumstances and then report much higher levels of happiness than affluent researchers would expect.
Translating a qualitative feeling such as happiness into a numerical measure is also highly problematic. The use of a unit scale (1, 2, 3, and so forth) implies that the difference between each unit is the same, as are measures of space, time, or temperature. But this means that the difference between one, “very unhappy,” and two, “somewhat unhappy,” would have to be precisely the same as the difference between two, “somewhat happy,” and three, “neither happy nor unhappy.” This may seem like a technical point, but performing even basic arithmetic operations on the resulting scores, such as averaging them and identifying percentage changes, requires that each interval be identical. Some researchers go one step further and add scores from questions asking about different conceptions of happiness altogether, which implies that step on all scales are identical. Ultimately, happiness is a feeling to be described, not a quantity to be measured.
Third, even if we could solve issues with definition and measurement, there are still serious problems regarding how to implement policy based on measures of happiness. These problems are more political than conceptual, but demand a hard look at what the government would be trying to do with measures of happiness.
For example, what would the goals of a happiness-based policy be? Would the government enact policies to maximise gross domestic happiness, or would they choose a growth rate for it as they do for GDP? A growth rate would be problematic due to the hedonic treadmill, another version of adaptation by which we work hard to reach a higher level of happiness only to become accustomed to it, after which we work even harder in a futile attempt to achieve the same increase in happiness we thought we had! (This is related to setpoint theory, which implies that, to a large extent, each person’s baseline level of happiness is determined by genetics, which casts doubt on government efforts to increase it.) Also, neither goal would account for inequality in realised happiness, which is of no less concern than inequality of wealth. Adaptation is relevant here as well, because it is precisely those who need help the most who are most likely to report significant happiness despite meagre circumstances. Finally, how will sacrifices in current happiness to invest in future happiness, such as going to medical school or training to be a ballerina, be separated out of happiness measures that focus on the past or present but not the future?
Regardless of the motivation and intention behind it, the move to base government policy on measures of happiness is fraught with problems at every level. Furthermore, these problems are not simply a matter of developing better tools and techniques, but implications of the essential nature of happiness itself: vague, qualitative, and subjective. Happiness means something different to each person, in different circumstances, and at different points in his or her life. And what’s more, it is not necessarily the most important thing to each person, who may care about the well-being of loved ones, the situation in Gaza, or public reception of the new Doctor Who—aside from the impact of these things on his or her personal happiness. When all is said and done, in addition to the numerous problems with the concept, happiness may not be the right thing for government to focus on in the first place.
Note: This article gives the views of the author, and not the position of the British Politics and Policy blog, nor of the London School of Economics. Please read our comments policy before posting. Featured image credit: Caleb Roenigk CC BY 2.0
Mark D. White is chair of the Department of Philosophy at the College of Staten Island/CUNY in New York City. His latest book, on which this post is based, is The Illusion of Well-Being: Economic Policy Based on Respect and Responsiveness (Palgrave, 2014).
You could measure sense of coherence instead on a community or national level A strong SOC correalates to wellbeing and quality of live and one of its components is meaningfulness. Unfortunately it has often been seen as an individual capacity but it is the collective SOC which is interesting.