Getting off the hedonic treadmill, one step at a time: The impact of regular religious practice and exercise on well-being

Abstract

Many studies have shown that few events in life have a lasting impact on subjective well-being because of people’s tendency to adapt quickly; worse, those events that do have a lasting impact tend to be negative. We suggest that while major events may not provide lasting increases in well-being, certain seemingly minor events – such as attending religious services or exercising – may do so by providing small but frequent boosts: if people engage in such behaviors with sufficient frequency, they may cumulatively experience enough boosts to attain higher well-being. In Study 1, we surveyed places of worship for 12 religions and found that people did receive positive boosts for attending service, and that these boosts appeared to be cumulative: the more they reported attending, the happier they were. In Study 2, we generalized these effects to other regular activities, demonstrating that people received boosts for exercise and yoga, and that these boosts too had a cumulative positive impact on well-being. We suggest that shifting focus from the impact of major life changes on well-being to the impact of seemingly minor repeated behaviors is crucial for understanding how best to improve well-being.

Introduction

Understanding the determinants of subjective well-being has important implications for economics. At the micro level, well-being has been shown to affect the behavior of individuals. For example, some studies have found that there is a positive relation between worker’s stable happiness and their work performance (Wright & Staw, 1999). People in positive moods have also been shown to be more creative problem solvers, more likely to attain a mutually favorable outcome while bargaining, and more willing to seek variety among positive choices (Isen, 2000).

At the macro level, policymakers must often decide among various programs that differ in the advantages they provide to the public. Understanding what factors truly improve well-being can guide such decisions, and help avoid choosing programs that bring expensive and non-lasting benefits. In contrast to models in which utility is inferred from people’s choices, recent economic conceptualizations of utility include people’s subjective feelings of utility – their reports of their subjective well-being – as an important input in determining overall utility. These reports of subjective well-being are increasingly a factor in determining the overall utility of some public policy decisions (e.g., Di Tella and MacCulloch, 2006, Kahneman et al., 2004b). As a result, the scientific study of subjective well-being has received increased attention from economists (see Frey and Stutzer, 2002, Kahneman et al., 2004a).

The literature on subjective well-being offers a paradox, however, that must be resolved before the construct can be fruitfully applied to improve people’s lives: most studies have shown that people’s overall level of happiness seems stubbornly impervious to change. While people accumulate experiences that most predict would affect their well-being, subjective well-being appears to be surprisingly stable. Indeed, if major life events such as winning a lottery fail to have a substantial lasting impact on well-being (Brickman, Coates, & Janoff-Bulman, 1978), it is hard to imagine that any single event could accomplish the feat. Therefore, it would seem that any policy geared towards maximizing subjective well-being would be doomed to fail from the outset. We believe, however, that it is possible to make lasting changes in subjective well-being if one focuses on the right types of behaviors. In this paper, we shift from a focus on the impact of single major life events toward a focus on the impact of seemingly minor behaviors such as exercising or attending religious services on well-being. We suggest that while single major events may be unlikely to have a lasting impact, smaller minor behaviors provide small boosts to well-being that can lead to real changes in overall well-being, especially if they are repeated with sufficient frequency over time: one cannot win the lottery every day, but one can exercise or attend religious services regularly, and these repeated behaviors may be enough to increase well-being over time.

Brickman and Campbell (1971) coined the term “hedonic treadmill” to describe the now widely accepted notion that though people continue to accrue experiences and objects that make them happy – or unhappy – their overall level of well-being tends to remain fairly static. The logic behind this argument stems from adaptation level theory (Helson, 1964), which argues that people perceive objects not in any absolute sense, but rather relative to a level established by previous experiences. Therefore, when people experience a positive event, two effects take place: in the short run, well-being increases; in the long run, however, people habituate to their new circumstances, which diminishes the positive effect of that event. In the most famous demonstration of this course of events, Brickman et al. (1978) interviewed a sample of lottery winners, as well as a sample of accident victims who had become paralyzed. The sample was chosen such that the major life event had happened to them within the previous year and a half, but at least a month before the interview, to allow for adaptation to occur. Their results showed that the lottery winners did not rate themselves as happier than the control group, and while the accident victims rated themselves as less happy than the other groups, they still rated themselves above the midpoint of the scale. Similarly, Suh, Diener, and Fujita (1996) showed that while major life events that had occurred within the previous 3 months predicted wellbeing, those occurring further back in time did not. Other studies have shown that there is no difference in well-being between people who had recently experienced a romantic break up versus those who had not, between assistant professors who had been denied tenure and those who had attained it, or between those whose preferred gubernatorial candidates had won or lost (Gilbert, Pinel, Wilson, Blumberg, & Wheatley, 1998). Taken together, these results suggest that most events have no lasting impact on our well-being (see Diener et al., 2006, Frederick and Loewenstein, 1999, Frey and Stutzer, 2002).

Worse still, of those investigations that have demonstrated a lasting impact of major events on well-being, most have been in negative domains. Dijkers (1997) conducted a meta-analysis of 22 studies that looked at the relationship between spinal cord injury and quality of life, and found that degree of disability had a negative effect on quality of life. Lucas, Clark, Georgellis, and Diener (2003) examined the effects of marital status over time with a large representative sample. Although marriage had a strong initial positive effect, this effect disappeared after just one year. People who had been widowed, on the other hand, never fully adapted and remained less happy than their baseline before the event occurred.

What then does account for people’s positive levels of well-being, if not major life events? Many studies have shown that there are strong effects of both genetic predisposition and stable personality traits on well-being. Lykken and Tellegen (1996) found that half of the variance associated with well-being was associated with genetic variation, and that this accounted for 80% of the stable component of the subjective well-being measure (see also Suh et al., 1996). Indeed, Headey and Wearing (1989) suggested that rather than life events causing changes in well-being, life events are to some degree endogenously caused by personality. Personality models of well-being thus suggest that though there may be short-lived effects of external shocks on well-being, people return to their baseline in the long run, a baseline which is determined primarily by their personalities and the events that those personalities cause them to pursue.

All of the previously cited research paints a rather discouraging picture about people’s ability to increase their own well-being. Although some major negative events seem to be able to create lasting changes (Dijkers, 1997, Lucas, 2005, Lucas et al., 2004), there seems to be little people can do to improve their well-being. At most, people may hope for temporary lifts from major life events (such as marriage, or winning the lottery) which quickly fade as they return to their usual baseline predetermined by genes and personality – requiring them to ever peruse the hedonic treadmill in the hope of finding some temporary increase in happiness. Given the frequency of winning the lottery or getting married (which for most people is a few times at most), the odds of improving well-being seem low. Of course, this is not as tragic as it sounds, since most people’s equilibrium state is somewhat happy (Diener & Diener, 1996). Nonetheless, it seems as though people generally should not even bother to pursue goals that make them happy (and not try as hard to avoid many of the activities that they expect to cause them unhappiness), as achieving them will not have any lasting impact. It also appears as though any economic policy aimed at improving people’s welfare is just a waste of time and money, since it will have no long-term effect. But is there truly nothing that people can do to improve their well-being?

We suggest that shifting from a focus on the impact of major life events to a focus on minor life events – the kinds of small activities people partake in every day – offers insight into how people might increase their well-being. Indeed, in contrast to the research reviewed above, some studies have shown that particular behaviors (such as religion and exercise) are related to higher levels of well-being. Importantly, these kinds of behaviors are repeated behaviors, rather than single-shot life events. We suggest that the cumulative impact of repeating minor but positive life events in the short-term – such as choosing to attend religious services each week or to work out several days a week – may be sufficient to increase well-being in the long-term.

We chose religious practice and exercise as our initial behaviors because both have been linked to well-being, and are precisely the kinds of minor repeated behaviors we propose may improve well-being. Indeed, religiosity and religious involvement have overwhelmingly been found to correlate with many measures of well-being. Myers (2000) reports data from a national sample showing that those who are most involved with their religion are almost twice as likely to report being “very happy” than those with the least involvement (see also Ferris, 2002). In a large cross-sectional national sample, Ellison (1991) found that religious variables accounted for 5–7% of the variance in life satisfaction (see also Witter, Stock, Okun, & Haring, 1985). Religious involvement has also been found to be positively related to more objective measures of well-being such as mental and physical health (e.g., Hackney and Sanders, 2003, Larson et al., 1992, Seybold and Hill, 2001). While not as widely studied, physical exercise is known to generate endorphins that improve mood (Thoren, Floras, Hoffmann, & Seals, 1990), regular engagement in exercise has been shown to have a positive impact on well-being (Biddle, 2000), and of course countless studies demonstrate the benefits of exercise for physical well-being (see Penedo and Dahn, 2005, Ross and Hayes, 1988).

How is it that religious involvement and exercise improve subjective well-being when most other factors – including events that seem to dwarf these in significance – seem to matter little? Why don’t people adapt to religion and exercise, behaviors that most people have been engaged in their entire life, as they do to most other things? We suggest that these behaviors have a causal effect on well-being because they give regular and reliable, albeit small, boosts to well-being each time a member participates in one of these activities. Though each boost is not large and fades over time, both religion and exercise encourage regular participation, which might cause these small boosts to aggregate over time, leading to increased well-being.

Some evidence for the impact of small repeated events comes from the finding that commuting, a regular daily activity, is rated as a highly negative experience (Kahneman et al., 2004a), and commuting time correlates negatively with subjective well-being (Stutzer & Frey, 2004). In addition, Lyubomirsky, Sheldon, and Schkade (2005) suggest that intentional activities, discrete actions in which people choose to engage, can lead to higher levels of well-being because they draw attention to positive events, thus preventing them from fading into the background. However none of these studies have documented whether these activities cause small boosts in well-being every time they are performed, as we propose.

In order to test whether people in fact do get small boosts from engaging in religious activity, we measured the subjective well-being of people as they entered and exited religious services (Study 1), and as they entered and exited the gym and yoga (Study 2). We expected to observe increases in well-being from before to after such behaviors. In both studies, we also assessed people’s reported frequency of such behaviors, to investigate whether increased frequency of engaging in these behaviors (and thus more frequent small boosts) was related to overall higher well-being.