525786
research-article2014
PSSXXX10.1177/0956797614525786Tay et al.Living Among the Affluent: Boon or Bane?
Research Report Psychological Science 2014, Vol. 25(6) 1235–1241 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0956797614525786 pss.sagepub.com
Living Among the Affluent: Boon or Bane? Louis Tay1, Mike Morrison2, and Ed Diener3,4 1
Department of Psychological Sciences, Purdue University; 2Department of Psychology, University of Western Ontario; Department of Psychology, University of Illinois; and 4The Gallup Organization, Omaha, Nebraska
3
Abstract This study examined whether national income can have effects on happiness, or subjective well-being (SWB), over and above those of personal income. To assess the incremental effects of national income on SWB, we conducted crosssectional multilevel analysis on data from 838,151 individuals in 158 nations. Although greater personal income was consistently related to higher SWB, we found that national income was a boon to life satisfaction but a bane to daily feelings of well-being; individuals in richer nations experienced more worry and anger on average. We also found moderating effects: The income-SWB relationship was stronger at higher levels of national income. This result might be explained by cultural norms, as money is valued more in richer nations. The SWB of more residentially mobile individuals was less affected by national income. Overall, our results suggest that the wealth of the nation one resides in has consequences for one’s happiness. Keywords national income, personal income, happiness, subjective well-being, spillover effects, multilevel analysis Received 5/24/13; Revision accepted 2/3/14
Does national income have incremental effects on individuals’ subjective well-being (SWB) beyond the effects of their personal income? Whether SWB improves or deteriorates with greater national income is controversial because of differing expectations that stem from contrasting value judgments about money and the pursuit of it. Differing opinions notwithstanding, bringing scientific evidence to bear on this issue is critical, as nations are increasingly seeking economic progress. Unraveling whether raising the average income of citizens has positive or negative effects on SWB will potentially have important implications for economic and public policy (Oswald, 1997). Although many studies have shown that personal income robustly predicts SWB (e.g., Stevenson & Wolfers, 2008), it is not clear whether national income incrementally predicts SWB. Previous research examining the effects of regional wealth on SWB has focused primarily on social comparison effects. One approach has been to examine how income norms, operationalized as average incomes of people demographically matched to respondents, predict SWB over and above the effects of
respondents’ individual income (e.g., Ferrer-i-Carbonell, 2005). An indirect approach has been to examine whether ranking of individual income is more predictive of SWB than absolute individual income is (e.g., Boyce, Brown, & Moore, 2010). By contrast, few studies have directly examined effects of regional wealth on SWB. In one such study, Luttmer (2005) examined neighborhood wealth and found that it had negative effects beyond those of individual income. Yet, to our knowledge, no studies have examined incremental effects of national income on SWB, which we term spillover effects. It has been proposed that regional wealth can generate negative spillover effects for a variety of reasons. Wealth increases frequently co-occur with industrialization and environmental degradation, which in turn can lower SWB (Di Tella & MacCulloch, 2008). Further, Ng, Diener, Aurora, and Harter (2009) proposed that as a society grows richer, its members are likely to experience Corresponding Author: Louis Tay, Purdue University, 703 Third St., West Lafayette, IN 47907 E-mail: [email protected]
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Tay et al.
1236
Fig. 1. Illustrative examples of positive and negative spillover effects of national wealth. In each graph, the solid lines show individual-level subjective well-being as a function of annual household income, separately for nations with average annual household incomes of $5,000 and $25,000; the dotted lines show nation-level subjective well-being as a function of national household income. As illustrated, positive spillover occurs when individuals at the same level of absolute income have higher SWB if they are living in a richer nation, and negative spillover occurs when individuals at the same level of absolute income have lower SWB if they are living in a richer nation.
greater levels of worry and anxiety because of a faster pace of life (Levine & Norenzayan, 1999) and an overabundance of choices (Schwartz, 2004). Striving to make the best choice under time pressure can be detrimental to SWB. Negative spillover effects can also arise from social processes such as upward income comparisons, which can lead individuals who live in a richer nation to experience lower SWB than those who live in a poorer one even though they have the same absolute income level. National wealth can also produce benefits, however. Greater national income has been shown to be associated with higher levels of material welfare (e.g., having basic needs for food and shelter met) for citizens (Diener, Ng, Harter, & Arora, 2010). Further, greater national income could lead to better infrastructure, such as transportation, communication, sanitation, and health-care systems, thereby improving the standard of living for individuals, and enhancing SWB, regardless of personal income (e.g., Aschauer, 1989). Empirically, positive or negative spillover effects would be evident if the magnitude of the relation between income and SWB varies between individual and national income, a pattern indicating that aggregate income has effects on SWB that are over and above those of individual income. Positive spillover occurs when the nationlevel slope of this relation is steeper than the
individual-level slope; negative spillover occurs when the individual-level slope is steeper than the nation-level slope. As illustrated in Figure 1, positive spillover occurs when individuals at the same level of absolute income have higher SWB if they are living in a richer nation. Conversely, negative spillover occurs when individuals at the same level of absolute income have lower SWB if they are living in a richer nation. There are good reasons to expect that national wealth may exert either positive or negative spillover effects on happiness, but the direction of the effect may depend on the type of SWB measured. SWB comprises life evaluations, positive feelings, and negative feelings (Diener, 1984). Research suggests that income is more strongly associated with life evaluations than with daily emotional experiences (Kahneman & Deaton, 2010). It has been proposed that life evaluations correspond more to external living conditions than do daily emotional experiences, which are more sensitive to daily life events (Tay, Chan, & Diener, 2013). Hence, higher levels of national income may increase life evaluations, but have less bearing on feelings, which are more tied to individual lifestyles and choices. Aside from its spillover effects, national wealth may moderate the extent to which individual income is linked to SWB. According to the culture-norm hypothesis,
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Living Among the Affluent: Boon or Bane? 1237 societies tend to prize those attributes with which they are endowed (Fulmer et al., 2010). Thus, money is valued more in richer nations (Myers, 2000), and as a result, national wealth may amplify the effect of income on SWB. If this is true, the link between income and SWB would be stronger in richer nations than in poorer ones. Income has decreasing marginal utility for SWB (e.g., Inglehart, Foa, Peterson, & Welzel, 2008); that is, increases in income buy greater increases in SWB at lower levels of income. However, according to the culture-norm hypothesis, although decreasing marginal utility holds within countries, its effect could be weakened in richer nations because the income-SWB link is stronger in richer nations: The same proportional increase in income might yield a greater increase in SWB in a richer nation, so that the decline in utility is reduced. We expected that national wealth would moderate the relation between income and SWB such that greater national income would enhance the effects of income on happiness. Finally, we expected that effects of national income on SWB would be more likely to be felt by individuals who identified more strongly with their environments. Indeed, the socioethological perspective of self and social behaviors posits that identity and well-being are more context dependent for less residentially mobile individuals (Oishi, 2010; Oishi, Lun, & Sherman, 2007). Thus, we hypothesized that national income would be more strongly related to the SWB of individuals who were less mobile.
Method We used data from the Gallup World Poll, which surveyed income and SWB across 158 nations from 2005 to 2011. The sample comprised 838,151 individuals and averaged 5,305 (SD = 3,020) individuals per country.
Individual income—specifically, annual household income (in international dollars)—was calculated from World Bank purchasing power parities (PPP). These values were log-transformed and then averaged at the national level to create the measure of national income. Because most variability in national income was attributable to between-nation differences (96.8% of the variance) rather than within-nation changes over time (3.2% of the variance), we averaged national annual household income across the years, treating yearly observations as repeated measures. Averaging across years yielded a high reliability (.99) for national income. Residential mobility was measured by whether or not respondents planned to move to another country in the next 12 months.
Analyses To disentangle the effects of individual-level income and average income within nations, we used cross-sectional random-coefficients multilevel modeling (Raudenbush & Bryk, 2002). Random coefficients that were significant at the .05 level were included in the model. Individual-level income was group-mean-centered, and national-level income was uncentered. We tested for an incremental effect of national income by examining whether the effect of national income was significantly different from the effect of individual income (see Enders & Tofighi, 2007). In all our analyses, we included age, gender, and education level (all grand-mean centered) as control variables at the individual level. The multilevel model was specified as follows:
Level 1: SWBij = b0j + b1j age + b2j gender + b3j education
Measures
We examined effects of individual household income and national-level (average) household income on both current life evaluations and affective SWB. Our measure of current life evaluations was a single item from Cantril’s (1965) Self-Anchoring Ladder; respondents indicated their assessment of their life on a scale from 0 (worst possible life) to 10 (best possible life). Affective well-being was assessed by asking respondents whether they had experienced various feelings “a lot” in the previous day; the response scale was dichotomous (1 = yes, 0 = no). An index for positive feelings was created by aggregating responses to two items, “smile/laugh” and “enjoyment” (αindividual = .61; αcountry = .83); an index for negative feelings was created by aggregating responses to three items, “worry,” “sadness,” and “anger” (αindividual = .61; αcountry = .78).
+ b4j [log(income) – log(ıncome)] + rij
Level 2: b0j = b00 + b01 log(ıncome) + u0
b1j = b10 + u1
b1j = b20 + u2
b3j = b30 + u3
b4j = b40 + b41 log(ıncome) + u4,
where i and j index individual and nation, respectively. Coefficient β4j represents the effect of pooled withinnation individual-level income; this within-nation individual-level effect was obtained by within-nation centering, subtracting the average logged individual income of a nation, log(ıncome), from logged individual
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Tay et al.
1238 income. Coefficient β01 represents the effect of nationallevel income (i.e., the average of logged individual income for a nation). For more information on the model and additional analyses not reported here, see the Supplemental Material available online.
Results Testing for the presence of context effects We calculated intraclass correlations (ICCs) representing the proportion of SWB variance that could be accounted for by national conditions, such as national income. The ICCs for the three measures of SWB were moderate to large. About 23% of the variability in life evaluations was attributable to the country respondents lived in (ICC = .23), and about 5% of the variability in affective SWB was attributable to the country respondents lived in (ICCs = .06 for positive feelings and .04 for negative feelings). These findings showed that national context has effects on SWB and provided an empirical basis for examining whether national income has incremental effects on SWB. They also show that national context matters more for life evaluations than for feelings.
Incremental effects of national income We found that national income had significant incremental effects on SWB over and above the effects of individual income (ps < .001). The effects of national income on life evaluations (β = 1.98) were stronger than the effects of individual-level income (β = 1.20). Thus, national wealth enhances life evaluations over and above individual income. Although national income was positively linked to positive feelings (β = 0.04), individual income had a stronger link with positive feelings (β = 0.12). These findings indicate that national wealth had negative spillover effects on positive feelings. For negative feelings, negative spillover was large. Greater national income was associated with higher negative feelings (β = 0.02), whereas greater individual income was linked to lower negative feelings (β = −0.08). Post hoc analyses showed that national income was linked to worry and anger (ps < .05), but not sadness (n.s.), which is consistent with the idea that economic development results in a faster pace of living that is more stressful (Levine & Norenzayan, 1999; Ng et al., 2009). Figure 2 shows that, overall, greater national income is a boon to life evaluations but a bane to affective SWB.
Moderating effects Analyses confirmed the culture-norm hypothesis. Significant interaction terms indicated that national wealth
enhanced the effect of individual income on life evaluations (β = 0.26), positive feelings (β = 0.02), and negative feelings (β = –0.04), ps < .05. As illustrated in Figure 2, regression lines for higher national incomes had steeper slopes, particularly in the case of life evaluations and negative feelings; this effect was less pronounced for positive feelings.
Mobility Residential mobility moderated the effect of national income on life evaluations (β = –0.022, p < .05), but not affective SWB (ps > .05). As shown in Figure 3, the effects of national income were stronger for individuals who were not planning to move than for those who were.
Discussion To our knowledge, this is the first study to disentangle the effects of national and personal income to show that national income has spillover effects on happiness. We found evidence that SWB could be partially accounted for by where respondents lived, and we estimated that 23% of life-evaluation scores could be accounted for by the nation respondents resided in. Thus, national context has strong effects on SWB, and especially life evaluations. Pursuing SWB is not merely an individual affair, as national context and one’s views of one’s country matter for personal SWB (Morrison, Tay, & Diener, 2011). Whether one lives in a rich nation, such as The Netherlands, or in a poor one, such as Zimbabwe, dramatically affects one’s sense of well-being. Researchers have proposed that national wealth can generate both positive and negative effects on SWB, so we tested for both positive and negative spillover. There was support for both, despite our finding that greater individual-level income was consistently associated with higher SWB. National income had a positive spillover effect on life evaluations but negative spillover effects on affective SWB. Table 1 presents predicted values for life evaluations, positive feelings, and negative feelings of individuals at two different income levels in nations of varying wealth. As shown, these predicted values differ greatly between nations at the 90th and 10th percentiles of income: The difference is about 1 point for life evaluations (i.e., about 1/10 of the 11-point scale) and up to about .20 point for affective SWB (i.e., about 1/5 of the 1-point scale). Our results present a nuanced picture of the effect of national income on happiness. Spillover effects are positive for life evaluations, but negative for daily affective SWB. Thus, both positive and negative contextual effects appear to be at work. Compared with neighborhood
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Living Among the Affluent: Boon or Bane? 1239
Fig. 2. Results of the multilevel analysis: subjective well-being (from top to bottom: life evaluations, positive feelings, and negative feelings) as a function of logged annual household income. In each graph, the four solid lines are regression lines for the effect of individual-level income on the indicated component of subjective well-being at four levels of national income (average household income at the national level); each line spans the range from 2 standard deviations below the average household income to 2 standard deviations above the average household income for the given band of national income. The dotted lines are regression lines for the relation between national-level income and subjective well-being.
wealth (Luttmer, 2005), national wealth more likely translates into better infrastructure, which in turn benefits not only the wealthy but also the poor. Further, better infrastructure and conveniences are evaluable and thus translate into the evaluative component of SWB, life evaluations (see Hsee & Zhang, 2010). Nevertheless, in richer nations, environmental degradation, a faster pace of life, and upward comparisons may be detrimental to everyday experiences and feelings of happiness. In short, national wealth generation spills over to create, on average, higher levels of life evaluations, but not necessarily higher daily affective SWB.
Moderating effects of national income on the incomehappiness relationship were also clearly evident. Although personal income has decreasing marginal utility for SWB, the same increase in personal income yields a greater increase in SWB in richer nations than in poorer nations. These results are in line with the culture-norm hypothesis, which predicts that individuals more similar to the majority are likely to enjoy greater happiness (Fulmer et al., 2010); in this case, richer individuals gain more in SWB in richer societies than in poorer societies. When judging their SWB, less residentially mobile individuals weigh national wealth and possibly other
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Tay et al.
1240 Move to another country in the next 12 months Yes No 8
Life Evaluations
7
6
5
000 $50, 000 $60, 000
$40,
000 $30,
000 $20,
000 $10,
3
$5,0
00
4
log(National Income) Fig. 3. Life evaluations as a function of logged national income for respondents who did and did not intend to move to another country in the next 12 months.
aspects of national quality of life as more important than do individuals who are more mobile. Specifically, in rich nations like the United States (average annual household income ≈ $60,000), life evaluations were 0.18 points lower for more mobile individuals (those who were planning to move in the next 12 months) than for less
mobile individuals (those who were not planning to move in the next 12 month), but in poor nations like Rwanda (average annual household income ≈ $2,000), life evaluations were 0.14 points higher for more mobile than for less mobile individuals. The range of this effect was about 0.32 points on the 10-point scale, or a difference of about 3% of the scale’s range. Although the effect appears to be small relative to the scale, it can translate to large practical effects at the population level. In contrast, there was less evidence that residential mobility influenced daily affective SWB, which is consistent with our other findings. Our results suggest that evaluative SWB is more sensitive to external macroenvironments, whereas daily affective SWB is less so, perhaps because it may be more closely tied to daily life events (e.g., Kahneman, Krueger, Schkade, Schwarz, & Stone, 2004). In conclusion, although greater personal income is associated with greater happiness, national wealth can influence happiness above and beyond the effects of personal income. Living among the affluent is both a boon and a bane: It can promote more positive life evaluations, but the opportunity cost is lower feelings of happiness. Also, we found that money is more strongly linked to SWB in wealthier countries, likely because a higher value is placed on money and materialistic goods in such countries (Kasser, 2002). However, having plans to leave one’s country can reduce the effects of national income. The relationship between income and SWB has been studied for some time, but past work has examined the influence of personal and national income independently. The present study illustrates the importance of examining them together and using multiple measures of SWB. Personal income and national wealth both matter for SWB, but the relationships observed depend on how rich
Table 1. Predicted Happiness for Individuals at Different Income Levels, in Poor and Rich Nations National income Income level and measure of happiness Median worldwide income ($6,847)a Life evaluations (0–10) Positive feelings (0–1) Negative feelings (0–1) Median U.S income ($52,762)a Life evaluations (0–10) Positive feelings (0–1) Negative feelings (0–1)
Very poor (10th percentile)
Poor (25th Average (50th Rich (75th percentile) percentile) percentile)
Differential between the Very rich (90th 90th and 10th percentile) percentiles
4.80 .75 .20
4.99 .72 .22
5.28 .67 .26
5.50 .62 .30
5.69 .56 .36
5.59 .92 .15
5.82 .89 .16
6.19 .85 .19
6.47 .80 .22
6.74 .75 .27
0.88 –.19 .16 1.15 −.17 .12
Note: The predicted values are estimates based on the multilevel model. The following are examples of countries at the different levels of national income—very poor: Ethiopia, Mali, and Indonesia; poor: India, Yemen, and Nigeria; average: Macedonia and Ecuador; rich: Greece, Cyprus, and Portugal; very rich: Finland, Japan, United Kingdom, Germany, and Hong Kong. a These income values are estimated from the Gallup data.
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Living Among the Affluent: Boon or Bane? 1241 respondents are, how tied they are to their country and the measure of SWB one is looking at. Author Contributions The data used in this study were obtained from The Gallup Organization via E. Diener. L. Tay developed the study concept and analyzed and interpreted the data. L. Tay drafted the manuscript, and M. Morrison and E. Diener provided critical revisions. All authors approved the final version of the manuscript for submission.
Declaration of Conflicting Interests The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.
Supplemental Material Additional supporting information may be found at http:// pss.sagepub.com/content/by/supplemental-data
References Aschauer, D. A. (1989). Is public expenditure productive? Journal of Monetary Economics, 23, 177–200. Boyce, C. J., Brown, G. D. A., & Moore, S. C. (2010). Money and happiness: Rank of income, not income, affects life satisfaction. Psychological Science, 21, 471–475. Cantril, H. (1965). The pattern of human concerns. New Brunswick, NJ: Rutgers University Press. Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542–575. Diener, E., Ng, W., Harter, J. K., & Arora, R. (2010). Wealth and happiness across the world: Material prosperity predicts life evaluation, whereas psychosocial prosperity predicts positive feelings. Journal of Personality and Social Psychology, 99, 52–61. Di Tella, R., & MacCulloch, R. J. (2008). Gross national happiness as an answer to the Easterlin Paradox? Journal of Development Economics, 86, 22–42. Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old issue. Psychological Methods, 12, 121–138. Ferrer-i-Carbonell, A. (2005). Income and well-being: An empirical analysis of the comparison income effect. Journal of Public Economics, 89, 997–1019. Fulmer, C. A., Gelfand, M. J., Kruglanski, A. W., Kim-Prieto, C., Diener, E., Pierro, A., & Higgins, E. T. (2010). On “feeling right” in cultural contexts: How person-culture match affects self-esteem and subjective well-being. Psychological Science, 21, 1563–1569.
Hsee, C. K., & Zhang, J. (2010). General evaluability theory. Perspectives on Psychological Science, 5, 343–355. Inglehart, R., Foa, R., Peterson, C., & Welzel, C. (2008). Development, freedom, and rising happiness: A global perspective (1981–2007). Perspectives on Psychological Science, 3, 264–285. Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional well-being. Proceedings of the National Academy of Sciences, USA, 107, 16489– 16493. Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone, A. A. (2004). A survey method for characterizing daily life experience: The Day Reconstruction Method. Science, 306, 1776–1780. Kasser, T. (2002). The high price of materialism. Cambridge, MA: MIT Press. Levine, R. V., & Norenzayan, A. (1999). The pace of life in 31 countries. Journal of Cross-Cultural Psychology, 30, 178– 205. Luttmer, E. F. P. (2005). Neighbors as negatives: Relative earnings and well-being. The Quarterly Journal of Economics, 120, 963–1002. Morrison, M., Tay, L., & Diener, E. (2011). Subjective well-being and national satisfaction. Psychological Science, 22, 166– 171. doi:10.1177/0956797610396224 Myers, D. G. (2000). The funds, friends, and faith of happy people. American Psychologist, 55, 56–67. Ng, W., Diener, E., Aurora, R., & Harter, J. K. (2009). Affluence, feelings of stress, and well-being. Social Indicators Research, 94, 257–271. Oishi, S. (2010). The psychology of residential mobility: Implications for the self, social relationships, and wellbeing. Perspectives on Psychological Science, 5, 5–21. Oishi, S., Lun, J., & Sherman, G. D. (2007). Residential mobility, self-concept, and positive affect in social interactions. Journal of Personality and Social Psychology, 93, 131–141. Oswald, A. J. (1997). Happiness and economic performance. The Economic Journal, 107, 1815–1831. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage. Schwartz, B. (2004). The paradox of choice: Why more is less. New York, NY: HarperCollins. Stevenson, B., & Wolfers, J. (2008). Economic growth and happiness: Reassessing the Easterlin paradox (Brookings Papers on Economic Activity). Washington, DC: The Brookings Institution. Tay, L., Chan, D., & Diener, E. (2013). The metrics of societal happiness. Social Indicators Research. Advance online publication. doi:10.1007/s11205-013-0356-1
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
research-article2014
PSSXXX10.1177/0956797614525786Tay et al.Living Among the Affluent: Boon or Bane?
Research Report Psychological Science 2014, Vol. 25(6) 1235–1241 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0956797614525786 pss.sagepub.com
Living Among the Affluent: Boon or Bane? Louis Tay1, Mike Morrison2, and Ed Diener3,4 1
Department of Psychological Sciences, Purdue University; 2Department of Psychology, University of Western Ontario; Department of Psychology, University of Illinois; and 4The Gallup Organization, Omaha, Nebraska
3
Abstract This study examined whether national income can have effects on happiness, or subjective well-being (SWB), over and above those of personal income. To assess the incremental effects of national income on SWB, we conducted crosssectional multilevel analysis on data from 838,151 individuals in 158 nations. Although greater personal income was consistently related to higher SWB, we found that national income was a boon to life satisfaction but a bane to daily feelings of well-being; individuals in richer nations experienced more worry and anger on average. We also found moderating effects: The income-SWB relationship was stronger at higher levels of national income. This result might be explained by cultural norms, as money is valued more in richer nations. The SWB of more residentially mobile individuals was less affected by national income. Overall, our results suggest that the wealth of the nation one resides in has consequences for one’s happiness. Keywords national income, personal income, happiness, subjective well-being, spillover effects, multilevel analysis Received 5/24/13; Revision accepted 2/3/14
Does national income have incremental effects on individuals’ subjective well-being (SWB) beyond the effects of their personal income? Whether SWB improves or deteriorates with greater national income is controversial because of differing expectations that stem from contrasting value judgments about money and the pursuit of it. Differing opinions notwithstanding, bringing scientific evidence to bear on this issue is critical, as nations are increasingly seeking economic progress. Unraveling whether raising the average income of citizens has positive or negative effects on SWB will potentially have important implications for economic and public policy (Oswald, 1997). Although many studies have shown that personal income robustly predicts SWB (e.g., Stevenson & Wolfers, 2008), it is not clear whether national income incrementally predicts SWB. Previous research examining the effects of regional wealth on SWB has focused primarily on social comparison effects. One approach has been to examine how income norms, operationalized as average incomes of people demographically matched to respondents, predict SWB over and above the effects of
respondents’ individual income (e.g., Ferrer-i-Carbonell, 2005). An indirect approach has been to examine whether ranking of individual income is more predictive of SWB than absolute individual income is (e.g., Boyce, Brown, & Moore, 2010). By contrast, few studies have directly examined effects of regional wealth on SWB. In one such study, Luttmer (2005) examined neighborhood wealth and found that it had negative effects beyond those of individual income. Yet, to our knowledge, no studies have examined incremental effects of national income on SWB, which we term spillover effects. It has been proposed that regional wealth can generate negative spillover effects for a variety of reasons. Wealth increases frequently co-occur with industrialization and environmental degradation, which in turn can lower SWB (Di Tella & MacCulloch, 2008). Further, Ng, Diener, Aurora, and Harter (2009) proposed that as a society grows richer, its members are likely to experience Corresponding Author: Louis Tay, Purdue University, 703 Third St., West Lafayette, IN 47907 E-mail: [email protected]
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Tay et al.
1236
Fig. 1. Illustrative examples of positive and negative spillover effects of national wealth. In each graph, the solid lines show individual-level subjective well-being as a function of annual household income, separately for nations with average annual household incomes of $5,000 and $25,000; the dotted lines show nation-level subjective well-being as a function of national household income. As illustrated, positive spillover occurs when individuals at the same level of absolute income have higher SWB if they are living in a richer nation, and negative spillover occurs when individuals at the same level of absolute income have lower SWB if they are living in a richer nation.
greater levels of worry and anxiety because of a faster pace of life (Levine & Norenzayan, 1999) and an overabundance of choices (Schwartz, 2004). Striving to make the best choice under time pressure can be detrimental to SWB. Negative spillover effects can also arise from social processes such as upward income comparisons, which can lead individuals who live in a richer nation to experience lower SWB than those who live in a poorer one even though they have the same absolute income level. National wealth can also produce benefits, however. Greater national income has been shown to be associated with higher levels of material welfare (e.g., having basic needs for food and shelter met) for citizens (Diener, Ng, Harter, & Arora, 2010). Further, greater national income could lead to better infrastructure, such as transportation, communication, sanitation, and health-care systems, thereby improving the standard of living for individuals, and enhancing SWB, regardless of personal income (e.g., Aschauer, 1989). Empirically, positive or negative spillover effects would be evident if the magnitude of the relation between income and SWB varies between individual and national income, a pattern indicating that aggregate income has effects on SWB that are over and above those of individual income. Positive spillover occurs when the nationlevel slope of this relation is steeper than the
individual-level slope; negative spillover occurs when the individual-level slope is steeper than the nation-level slope. As illustrated in Figure 1, positive spillover occurs when individuals at the same level of absolute income have higher SWB if they are living in a richer nation. Conversely, negative spillover occurs when individuals at the same level of absolute income have lower SWB if they are living in a richer nation. There are good reasons to expect that national wealth may exert either positive or negative spillover effects on happiness, but the direction of the effect may depend on the type of SWB measured. SWB comprises life evaluations, positive feelings, and negative feelings (Diener, 1984). Research suggests that income is more strongly associated with life evaluations than with daily emotional experiences (Kahneman & Deaton, 2010). It has been proposed that life evaluations correspond more to external living conditions than do daily emotional experiences, which are more sensitive to daily life events (Tay, Chan, & Diener, 2013). Hence, higher levels of national income may increase life evaluations, but have less bearing on feelings, which are more tied to individual lifestyles and choices. Aside from its spillover effects, national wealth may moderate the extent to which individual income is linked to SWB. According to the culture-norm hypothesis,
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Living Among the Affluent: Boon or Bane? 1237 societies tend to prize those attributes with which they are endowed (Fulmer et al., 2010). Thus, money is valued more in richer nations (Myers, 2000), and as a result, national wealth may amplify the effect of income on SWB. If this is true, the link between income and SWB would be stronger in richer nations than in poorer ones. Income has decreasing marginal utility for SWB (e.g., Inglehart, Foa, Peterson, & Welzel, 2008); that is, increases in income buy greater increases in SWB at lower levels of income. However, according to the culture-norm hypothesis, although decreasing marginal utility holds within countries, its effect could be weakened in richer nations because the income-SWB link is stronger in richer nations: The same proportional increase in income might yield a greater increase in SWB in a richer nation, so that the decline in utility is reduced. We expected that national wealth would moderate the relation between income and SWB such that greater national income would enhance the effects of income on happiness. Finally, we expected that effects of national income on SWB would be more likely to be felt by individuals who identified more strongly with their environments. Indeed, the socioethological perspective of self and social behaviors posits that identity and well-being are more context dependent for less residentially mobile individuals (Oishi, 2010; Oishi, Lun, & Sherman, 2007). Thus, we hypothesized that national income would be more strongly related to the SWB of individuals who were less mobile.
Method We used data from the Gallup World Poll, which surveyed income and SWB across 158 nations from 2005 to 2011. The sample comprised 838,151 individuals and averaged 5,305 (SD = 3,020) individuals per country.
Individual income—specifically, annual household income (in international dollars)—was calculated from World Bank purchasing power parities (PPP). These values were log-transformed and then averaged at the national level to create the measure of national income. Because most variability in national income was attributable to between-nation differences (96.8% of the variance) rather than within-nation changes over time (3.2% of the variance), we averaged national annual household income across the years, treating yearly observations as repeated measures. Averaging across years yielded a high reliability (.99) for national income. Residential mobility was measured by whether or not respondents planned to move to another country in the next 12 months.
Analyses To disentangle the effects of individual-level income and average income within nations, we used cross-sectional random-coefficients multilevel modeling (Raudenbush & Bryk, 2002). Random coefficients that were significant at the .05 level were included in the model. Individual-level income was group-mean-centered, and national-level income was uncentered. We tested for an incremental effect of national income by examining whether the effect of national income was significantly different from the effect of individual income (see Enders & Tofighi, 2007). In all our analyses, we included age, gender, and education level (all grand-mean centered) as control variables at the individual level. The multilevel model was specified as follows:
Level 1: SWBij = b0j + b1j age + b2j gender + b3j education
Measures
We examined effects of individual household income and national-level (average) household income on both current life evaluations and affective SWB. Our measure of current life evaluations was a single item from Cantril’s (1965) Self-Anchoring Ladder; respondents indicated their assessment of their life on a scale from 0 (worst possible life) to 10 (best possible life). Affective well-being was assessed by asking respondents whether they had experienced various feelings “a lot” in the previous day; the response scale was dichotomous (1 = yes, 0 = no). An index for positive feelings was created by aggregating responses to two items, “smile/laugh” and “enjoyment” (αindividual = .61; αcountry = .83); an index for negative feelings was created by aggregating responses to three items, “worry,” “sadness,” and “anger” (αindividual = .61; αcountry = .78).
+ b4j [log(income) – log(ıncome)] + rij
Level 2: b0j = b00 + b01 log(ıncome) + u0
b1j = b10 + u1
b1j = b20 + u2
b3j = b30 + u3
b4j = b40 + b41 log(ıncome) + u4,
where i and j index individual and nation, respectively. Coefficient β4j represents the effect of pooled withinnation individual-level income; this within-nation individual-level effect was obtained by within-nation centering, subtracting the average logged individual income of a nation, log(ıncome), from logged individual
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Tay et al.
1238 income. Coefficient β01 represents the effect of nationallevel income (i.e., the average of logged individual income for a nation). For more information on the model and additional analyses not reported here, see the Supplemental Material available online.
Results Testing for the presence of context effects We calculated intraclass correlations (ICCs) representing the proportion of SWB variance that could be accounted for by national conditions, such as national income. The ICCs for the three measures of SWB were moderate to large. About 23% of the variability in life evaluations was attributable to the country respondents lived in (ICC = .23), and about 5% of the variability in affective SWB was attributable to the country respondents lived in (ICCs = .06 for positive feelings and .04 for negative feelings). These findings showed that national context has effects on SWB and provided an empirical basis for examining whether national income has incremental effects on SWB. They also show that national context matters more for life evaluations than for feelings.
Incremental effects of national income We found that national income had significant incremental effects on SWB over and above the effects of individual income (ps < .001). The effects of national income on life evaluations (β = 1.98) were stronger than the effects of individual-level income (β = 1.20). Thus, national wealth enhances life evaluations over and above individual income. Although national income was positively linked to positive feelings (β = 0.04), individual income had a stronger link with positive feelings (β = 0.12). These findings indicate that national wealth had negative spillover effects on positive feelings. For negative feelings, negative spillover was large. Greater national income was associated with higher negative feelings (β = 0.02), whereas greater individual income was linked to lower negative feelings (β = −0.08). Post hoc analyses showed that national income was linked to worry and anger (ps < .05), but not sadness (n.s.), which is consistent with the idea that economic development results in a faster pace of living that is more stressful (Levine & Norenzayan, 1999; Ng et al., 2009). Figure 2 shows that, overall, greater national income is a boon to life evaluations but a bane to affective SWB.
Moderating effects Analyses confirmed the culture-norm hypothesis. Significant interaction terms indicated that national wealth
enhanced the effect of individual income on life evaluations (β = 0.26), positive feelings (β = 0.02), and negative feelings (β = –0.04), ps < .05. As illustrated in Figure 2, regression lines for higher national incomes had steeper slopes, particularly in the case of life evaluations and negative feelings; this effect was less pronounced for positive feelings.
Mobility Residential mobility moderated the effect of national income on life evaluations (β = –0.022, p < .05), but not affective SWB (ps > .05). As shown in Figure 3, the effects of national income were stronger for individuals who were not planning to move than for those who were.
Discussion To our knowledge, this is the first study to disentangle the effects of national and personal income to show that national income has spillover effects on happiness. We found evidence that SWB could be partially accounted for by where respondents lived, and we estimated that 23% of life-evaluation scores could be accounted for by the nation respondents resided in. Thus, national context has strong effects on SWB, and especially life evaluations. Pursuing SWB is not merely an individual affair, as national context and one’s views of one’s country matter for personal SWB (Morrison, Tay, & Diener, 2011). Whether one lives in a rich nation, such as The Netherlands, or in a poor one, such as Zimbabwe, dramatically affects one’s sense of well-being. Researchers have proposed that national wealth can generate both positive and negative effects on SWB, so we tested for both positive and negative spillover. There was support for both, despite our finding that greater individual-level income was consistently associated with higher SWB. National income had a positive spillover effect on life evaluations but negative spillover effects on affective SWB. Table 1 presents predicted values for life evaluations, positive feelings, and negative feelings of individuals at two different income levels in nations of varying wealth. As shown, these predicted values differ greatly between nations at the 90th and 10th percentiles of income: The difference is about 1 point for life evaluations (i.e., about 1/10 of the 11-point scale) and up to about .20 point for affective SWB (i.e., about 1/5 of the 1-point scale). Our results present a nuanced picture of the effect of national income on happiness. Spillover effects are positive for life evaluations, but negative for daily affective SWB. Thus, both positive and negative contextual effects appear to be at work. Compared with neighborhood
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Living Among the Affluent: Boon or Bane? 1239
Fig. 2. Results of the multilevel analysis: subjective well-being (from top to bottom: life evaluations, positive feelings, and negative feelings) as a function of logged annual household income. In each graph, the four solid lines are regression lines for the effect of individual-level income on the indicated component of subjective well-being at four levels of national income (average household income at the national level); each line spans the range from 2 standard deviations below the average household income to 2 standard deviations above the average household income for the given band of national income. The dotted lines are regression lines for the relation between national-level income and subjective well-being.
wealth (Luttmer, 2005), national wealth more likely translates into better infrastructure, which in turn benefits not only the wealthy but also the poor. Further, better infrastructure and conveniences are evaluable and thus translate into the evaluative component of SWB, life evaluations (see Hsee & Zhang, 2010). Nevertheless, in richer nations, environmental degradation, a faster pace of life, and upward comparisons may be detrimental to everyday experiences and feelings of happiness. In short, national wealth generation spills over to create, on average, higher levels of life evaluations, but not necessarily higher daily affective SWB.
Moderating effects of national income on the incomehappiness relationship were also clearly evident. Although personal income has decreasing marginal utility for SWB, the same increase in personal income yields a greater increase in SWB in richer nations than in poorer nations. These results are in line with the culture-norm hypothesis, which predicts that individuals more similar to the majority are likely to enjoy greater happiness (Fulmer et al., 2010); in this case, richer individuals gain more in SWB in richer societies than in poorer societies. When judging their SWB, less residentially mobile individuals weigh national wealth and possibly other
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Tay et al.
1240 Move to another country in the next 12 months Yes No 8
Life Evaluations
7
6
5
000 $50, 000 $60, 000
$40,
000 $30,
000 $20,
000 $10,
3
$5,0
00
4
log(National Income) Fig. 3. Life evaluations as a function of logged national income for respondents who did and did not intend to move to another country in the next 12 months.
aspects of national quality of life as more important than do individuals who are more mobile. Specifically, in rich nations like the United States (average annual household income ≈ $60,000), life evaluations were 0.18 points lower for more mobile individuals (those who were planning to move in the next 12 months) than for less
mobile individuals (those who were not planning to move in the next 12 month), but in poor nations like Rwanda (average annual household income ≈ $2,000), life evaluations were 0.14 points higher for more mobile than for less mobile individuals. The range of this effect was about 0.32 points on the 10-point scale, or a difference of about 3% of the scale’s range. Although the effect appears to be small relative to the scale, it can translate to large practical effects at the population level. In contrast, there was less evidence that residential mobility influenced daily affective SWB, which is consistent with our other findings. Our results suggest that evaluative SWB is more sensitive to external macroenvironments, whereas daily affective SWB is less so, perhaps because it may be more closely tied to daily life events (e.g., Kahneman, Krueger, Schkade, Schwarz, & Stone, 2004). In conclusion, although greater personal income is associated with greater happiness, national wealth can influence happiness above and beyond the effects of personal income. Living among the affluent is both a boon and a bane: It can promote more positive life evaluations, but the opportunity cost is lower feelings of happiness. Also, we found that money is more strongly linked to SWB in wealthier countries, likely because a higher value is placed on money and materialistic goods in such countries (Kasser, 2002). However, having plans to leave one’s country can reduce the effects of national income. The relationship between income and SWB has been studied for some time, but past work has examined the influence of personal and national income independently. The present study illustrates the importance of examining them together and using multiple measures of SWB. Personal income and national wealth both matter for SWB, but the relationships observed depend on how rich
Table 1. Predicted Happiness for Individuals at Different Income Levels, in Poor and Rich Nations National income Income level and measure of happiness Median worldwide income ($6,847)a Life evaluations (0–10) Positive feelings (0–1) Negative feelings (0–1) Median U.S income ($52,762)a Life evaluations (0–10) Positive feelings (0–1) Negative feelings (0–1)
Very poor (10th percentile)
Poor (25th Average (50th Rich (75th percentile) percentile) percentile)
Differential between the Very rich (90th 90th and 10th percentile) percentiles
4.80 .75 .20
4.99 .72 .22
5.28 .67 .26
5.50 .62 .30
5.69 .56 .36
5.59 .92 .15
5.82 .89 .16
6.19 .85 .19
6.47 .80 .22
6.74 .75 .27
0.88 –.19 .16 1.15 −.17 .12
Note: The predicted values are estimates based on the multilevel model. The following are examples of countries at the different levels of national income—very poor: Ethiopia, Mali, and Indonesia; poor: India, Yemen, and Nigeria; average: Macedonia and Ecuador; rich: Greece, Cyprus, and Portugal; very rich: Finland, Japan, United Kingdom, Germany, and Hong Kong. a These income values are estimated from the Gallup data.
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
Living Among the Affluent: Boon or Bane? 1241 respondents are, how tied they are to their country and the measure of SWB one is looking at. Author Contributions The data used in this study were obtained from The Gallup Organization via E. Diener. L. Tay developed the study concept and analyzed and interpreted the data. L. Tay drafted the manuscript, and M. Morrison and E. Diener provided critical revisions. All authors approved the final version of the manuscript for submission.
Declaration of Conflicting Interests The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.
Supplemental Material Additional supporting information may be found at http:// pss.sagepub.com/content/by/supplemental-data
References Aschauer, D. A. (1989). Is public expenditure productive? Journal of Monetary Economics, 23, 177–200. Boyce, C. J., Brown, G. D. A., & Moore, S. C. (2010). Money and happiness: Rank of income, not income, affects life satisfaction. Psychological Science, 21, 471–475. Cantril, H. (1965). The pattern of human concerns. New Brunswick, NJ: Rutgers University Press. Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542–575. Diener, E., Ng, W., Harter, J. K., & Arora, R. (2010). Wealth and happiness across the world: Material prosperity predicts life evaluation, whereas psychosocial prosperity predicts positive feelings. Journal of Personality and Social Psychology, 99, 52–61. Di Tella, R., & MacCulloch, R. J. (2008). Gross national happiness as an answer to the Easterlin Paradox? Journal of Development Economics, 86, 22–42. Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old issue. Psychological Methods, 12, 121–138. Ferrer-i-Carbonell, A. (2005). Income and well-being: An empirical analysis of the comparison income effect. Journal of Public Economics, 89, 997–1019. Fulmer, C. A., Gelfand, M. J., Kruglanski, A. W., Kim-Prieto, C., Diener, E., Pierro, A., & Higgins, E. T. (2010). On “feeling right” in cultural contexts: How person-culture match affects self-esteem and subjective well-being. Psychological Science, 21, 1563–1569.
Hsee, C. K., & Zhang, J. (2010). General evaluability theory. Perspectives on Psychological Science, 5, 343–355. Inglehart, R., Foa, R., Peterson, C., & Welzel, C. (2008). Development, freedom, and rising happiness: A global perspective (1981–2007). Perspectives on Psychological Science, 3, 264–285. Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional well-being. Proceedings of the National Academy of Sciences, USA, 107, 16489– 16493. Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone, A. A. (2004). A survey method for characterizing daily life experience: The Day Reconstruction Method. Science, 306, 1776–1780. Kasser, T. (2002). The high price of materialism. Cambridge, MA: MIT Press. Levine, R. V., & Norenzayan, A. (1999). The pace of life in 31 countries. Journal of Cross-Cultural Psychology, 30, 178– 205. Luttmer, E. F. P. (2005). Neighbors as negatives: Relative earnings and well-being. The Quarterly Journal of Economics, 120, 963–1002. Morrison, M., Tay, L., & Diener, E. (2011). Subjective well-being and national satisfaction. Psychological Science, 22, 166– 171. doi:10.1177/0956797610396224 Myers, D. G. (2000). The funds, friends, and faith of happy people. American Psychologist, 55, 56–67. Ng, W., Diener, E., Aurora, R., & Harter, J. K. (2009). Affluence, feelings of stress, and well-being. Social Indicators Research, 94, 257–271. Oishi, S. (2010). The psychology of residential mobility: Implications for the self, social relationships, and wellbeing. Perspectives on Psychological Science, 5, 5–21. Oishi, S., Lun, J., & Sherman, G. D. (2007). Residential mobility, self-concept, and positive affect in social interactions. Journal of Personality and Social Psychology, 93, 131–141. Oswald, A. J. (1997). Happiness and economic performance. The Economic Journal, 107, 1815–1831. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage. Schwartz, B. (2004). The paradox of choice: Why more is less. New York, NY: HarperCollins. Stevenson, B., & Wolfers, J. (2008). Economic growth and happiness: Reassessing the Easterlin paradox (Brookings Papers on Economic Activity). Washington, DC: The Brookings Institution. Tay, L., Chan, D., & Diener, E. (2013). The metrics of societal happiness. Social Indicators Research. Advance online publication. doi:10.1007/s11205-013-0356-1
Downloaded from pss.sagepub.com at University Library Utrecht on March 16, 2015
