attached below
jojom3
happiness.pdf
This article was downloaded by: [Library Services, University of the West of England] On: 12 December 2014, At: 13:42 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Regional Studies Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/cres20
The Happiness of Cities Richard Florida a , Charlotta Mellander b & Peter J. Rentfrow c a Martin Prosperity Institute, Rotman School of Management , University of Toronto , 101 College Street, Suite 420, Toronto , ON , Canada , M5G 1L7 E-mail: b Prosperity Institute of Scandinavia , Jönköping International Business School , Box 1026, S-551 11 , Jönköping , Sweden c Department of Social and Developmental Psychology , University of Cambridge , Free School Lane, Cambridge , CB2 3RQ , UK E-mail: Published online: 25 Aug 2011.
To cite this article: Richard Florida , Charlotta Mellander & Peter J. Rentfrow (2013) The Happiness of Cities, Regional Studies, 47:4, 613-627, DOI: 10.1080/00343404.2011.589830
To link to this article: http://dx.doi.org/10.1080/00343404.2011.589830
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
The Happiness of Cities
RICHARD FLORIDA*, CHARLOTTA MELLANDER† and PETER J. RENTFROW‡ *Martin Prosperity Institute, Rotman School of Management, University of Toronto, 101 College Street, Suite 420, Toronto, ON,
Canada M5G 1L7. Email: [email protected] †Prosperity Institute of Scandinavia, Jönköping International Business School, Box 1026, S-551 11 Jönköping, Sweden.
Email: [email protected] ‡Department of Social and Developmental Psychology, University of Cambridge, Free School Lane, Cambridge CB2 3RQ, UK.
Email: [email protected]
(Received July 2010: in revised form May 2011)
FLORIDA R., MELLANDER C. and RENTFROW P. J. The happiness of cities, Regional Studies. This research examines the factors that shape the happiness of cities, whereas much of the existent literature has focused on the happiness of nations. It is argued that in addition to income, which has been found to shape national-level happiness, human capital levels will play an important role in the happiness of cities. Metropolitan-level data from the 2009 Gallup–Healthways Survey are used to examine the effects of human capital on city happiness, controlling for other factors. The findings suggest that human capital plays the central role in the happiness of cities, outperforming income and every other variable.
Happiness Well-being Human capital Income Cities
FLORIDA R., MELLANDER C. and RENTFROW P. J. 城市幸福感，区域研究。多数研究集中于讨论与国家幸福感相关的问 题，本研究则关注构成城市幸福感的因素。研究认为，就城市层面而言，除了收入这一同样包含在国家层面的幸福 感中的要素外，人力资本水平也将成为城市幸福感的重要影响因素。研究采用 2009 年 Gallup-Healthways 调研中大都 市层面的数据，在控制其他影响变量的基础上考察了人力资本对城市幸福感的影响。研究表明，人力资本对于城市 幸福感的形成发挥着中心作用，超过了收入以及其他变量的影响力。
幸福感 福利 人力资本 收入 城市
FLORIDA R., MELLANDER C. et RENTFROW P. J. Le bonheur des grandes villes, Regional Studies. Cette recherche examine les facteurs qui influencent le bonheur des grandes villes, tandis que beaucoup de la documentation actuelle porte sur le bonheur de la nation. On affirme que le capital humain joue un rôle important dans le bonheur des grandes villes en plus du revenu, qui s’avère influencer le bonheur au niveau national. On emploie des données auprès des zones métropolitaines et qui proviennent de l’enquête Gallup–Healthways de 2009 pour examiner les effets du capital humain sur le bonheur des grandes villes, tout en tenant compte d’autres facteurs. Les résultats laissent supposer que le capital humain joue le rôle central quant au bonheur des grandes villes, dépassant celui du revenu et de tout autre variable.
Bonheur Bien-être Capital humain Revenu Grandes villes
FLORIDA R., MELLANDER C. und RENTFROW P. J. Das Glück von Städten, Regional Studies. In diesem Beitrag untersuchen wir die Faktoren, die sich auf das Glück von Städten auswirken, während sich ein Großteil der vorhandenen Literatur auf das Glück von Nationen konzentriert hat. Wir argumentieren, dass zusätzlich zum Einkommen, das den Ergebnissen zufolge das Glück auf nationaler Ebene prägt, für das Glück von Städten das Niveau von Humankapital eine wichtige Rolle spielt. Mit Hilfe von Daten der Gallup–Healthways-Erhebung von 2009 auf Metropolitan-Ebene untersuchen wir die Auswirkungen von Humankapital auf das Glück von Städten unter Berücksichtigung von anderen Faktoren. Die Ergebnisse lassen darauf schließen, dass das Humanka- pital für das Glück von Städten eine entscheidende Rolle spielt und eine wichtigere Bedeutung als das Einkommen und jede andere Variable besitzt.
Glück Wohlbefinden Humankapital Einkommen Städte
FLORIDA R., MELLANDER C. y RENTFROW P. J. La felicidad de las ciudades, Regional Studies. En este estudio examinamos qué factores repercuten en la felicidad de las ciudades, mientras que en la mayoría de la bibliografía existente se ha prestado más atención a la felicidad de las naciones. Sostenemos que además de los ingresos, que se ha observado que determinan la felicidad a nivel nacio- nal, los niveles de capital humano desempeñan un importante papel en la felicidad de las ciudades. Utilizamos datos recabados en una encuesta realizada por Gallup–Healthways en 2009 en un marco metropolitano para examinar los efectos del capital humano
Regional Studies, 2013
Vol. 47, No. 4, 613–627, http://dx.doi.org/10.1080/00343404.2011.589830
© 2013 Regional Studies Association http://www.regionalstudies.org
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
en la felicidad de las ciudades, considerando también otros factores. Los resultados indican que el capital humano desempeña una función central en la felicidad de las ciudades, superando a los ingresos y todas las otras variables.
Felicidad Bienestar Capital humano Ingresos Ciudades
JEL classifications: I0, J24, R0
INTRODUCTION
Much of the debate over happiness or subjective well- being – defined as people’s subjective cognitive and affective evaluations of their quality of life – has centred on the role of income. While EASTERLIN’s (1974, 1995) cross-national work found that the relationship between income and happiness holds only within and not across countries, more recent work by DEATON (2008) and STEVENSON and WOLFERS (2008) has chal- lenged this view, finding a strong relationship between income and happiness across nations. GRAHAM (2009) offers the paradox of the ‘happy peasant and the miserable millionaire’ as an important factor to consider here, suggesting that people’s expectations often adapt to their income level and financial stability.
The present analysis seeks to shed additional light on the ongoing debate over happiness and well-being by focusing on the metropolitan level. Considering happiness at the metropolitan level is important and interesting; individuals tend actively to select their place of residence in light of the job opportunities, public goods and services they provide (SJAASTAD, 1962; TIEBOUT, 1956), identify closely with and derive both satisfaction with their community (FLORIDA, 2009; FLORIDA and MELLANDER, 2010) and emotional attachment from the city in which they live (FLORIDA and MELLANDER, 2010; FLORIDA et al., 2011). While national-level studies have stressed the connection between happiness and income, drawing from studies of metropolitan economic per- formance it is argued here that human capital is likely to play a considerable role in metropolitan happiness. This hypothesis is tested through statistical analyses of the effects of human capital on happiness, alongside income, wages and economic output as well as income inequality, unemployment, housing cost and afford- ability, density, age, commuting time, and climate. The measure employed to assess metropolitan happiness comes from newly available data from a Gallup– Healthways Survey. As far as the authors are aware, it is the first comprehensive dataset that tracks happiness and well-being at the metropolitan level, providing data from a large-scale survey of individuals across 184 metro regions. Data-matching reduces the size of the sample to 170 metros – roughly half of all US regions.
The findings support the hypothesis. Using corre- lation and regression analysis, this investigation finds that capital is a key determinant of city or metropolitan well-being; and that housing values and unemployment also play a role in metropolitan well-being.
THEORY AND CONCEPTS
The happiness–income connection
Most of the research on place and well-being has focused on national differences (DIENER and DIENER, 1995; DIENER et al., 1995, 2003; DIENER and LUCAS, 1999; INGLEHART and KLINGEMANN, 2003; LYNN and STEEL, 2006; STEEL and ONES, 2002; VEENHO- VEN, 1993), and the results have revealed robust national differences in life satisfaction (DIENER, 2000; DIENER et al., 1995; VEENHOVEN, 1993). Explanations for these differences have frequently focused on the relationship between well-being and income. Early research found that the relationship between income and happiness holds within countries, but not between them (EASTERLIN, 1974, 1995). However, more recent research (DEATON, 2008; STEVENSON and WOLFERS, 2008) found a closer connection between happiness and income, suggesting that people with high incomes are happier than those with lower incomes, both in absolute and in relative terms. Cer- tainly one reason why income might be important to life satisfaction is that individuals must meet their basic needs for food, water, and shelter in order to survive. Another reason may be that wealth affords people opportunities and experiences to enrich their lives.
While income levels matter for happiness, work by GRAHAM (2008) shows the relationship between the two is relative. Graham contends that although people can be happy at lower levels of income, they are far less happy when there is greater uncertainty over their future wealth. This suggests that the income–happiness link is based not only on individual perceptions, but also on social and economic context.
The role of human capital
Both individual-level and cross-national studies have found that human capital plays, at best, a small role in happiness and well-being. A meta-analysis by WITTER et al. (1984) found that education is a small but positive contributor to subjective well-being in adults, account- ing for 1–3% of the variance. LAYARD (2005) found that education has a small (but significant) direct effect on happiness and that it also affects happiness indirectly by raising personal income. Curiously, there is some evi- dence that better-educated individuals report greater dissatisfaction than their less-educated counterparts, so the relationship cannot simply be seen as higher edu- cation equals more happiness. DIENER et al. (1999)
614 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
suggest the diminishing returns of education to well- being may be the result of education raising people’s goals and aspirations to a point that is very challenging to achieve, and therefore distressing.
While the links between human capital and well- being do not operate strongly at the individual level, there is evidence for a stronger relationship at aggregate levels of analysis. Using a measure of well-being derived from the Gallup–Heathways survey, RENTFROW et al. (2009) found a close correlation between human capital and happiness at the state level. More recently, LAWLESS and LUCAS (2010) examined the effects of human capital as well as other variables on subjective well-being across US counties. Their measure of subjective well-being was based on state surveys of life satisfaction collected by the Centers for Disease Control and the Prevention Behavioral Risk Factor Surveillance System. They found a strong association between human capital and life satisfaction at the county level. While the findings of these two studies are encouraging, they are based on bivariate analysis only, and did not control for the effects of income and other factors in a multivariate context. The present investigation goes beyond these previous studies by explicitly testing for the effects of human capital on happiness through a series of regression models that control for the effects of income and other key economic factors.
Other factors that contribute to happiness and well-being
This research considers a number of factors that the literature identifies as influencing happiness and well- being at the individual and/or national levels. Building on theoretical and empirical work in the urban and regional literature, this investigation includes variables that have been identified as potential predictors of metropolitan-level economic performance and/or community satisfaction.
Income inequality. GLAESER et al. (2009) found that people in unequal communities are more likely to report lower levels of happiness compared with residents in communities with lower levels of inequality. CLARK and OSWALD (1996) also found that workers are sensi- tive to their relative (or comparison) wage level. STUTZER (2004) posited that current income compared with aspired income accounts for some of the variance in happiness. These findings suggest that relative income might be a better predictor of happiness than absolute income.
Unemployment. Several studies in economics have probed the relationship between unemployment and happiness (CLARK and OSWALD, 1994; CLARK et al., 2001; WINKELMANN and WINKELMANN, 1998). BLANCHFLOWER and OSWALD (2004), for instance, found a negative relationship between unemployment
and happiness at the individual level. LAWLESS and LUCAS (2010) found a sizeable negative correlation between unemployment and happiness at the county level.
Commuting. Commuting time has been found to have a significant negative effect on happiness and well- being. KRUEGER et al. (2008) found the second most unpleasant activity of the day to be a long commute to work. STUTZER and FREY (2008) also found that people with longer daily commutes reported systemati- cally lower subjective well-being. LAWLESS and LUCAS (2010) found commuting time to be negatively associ- ated with happiness at the county level, although the correlation was relatively weak compared with other factors in their analysis.
Housing. Housing is the single biggest cost factor for most individuals and households. It might be expected that happiness is higher in places where housing is more available, less expensive and more affordable. However, RENTFROW et al. (2009) found happiness to be associated with higher median housing values at the state level. LAWLESS and LUCAS (2010) found mixed results for housing variables. Like RENTFROW et al. (2009), they found housing costs to be significantly related to happiness, although at a weaker level than for human capital or income. However, they found nega- tive correlations for indicators that measure housing prices as a percentage of income, such as the percentage of people whose mortgage or rent exceeds 35% of income. They also found negative correlations for housing cost variables – including housing value, median mortgage and media rent – when they used partial correlations for median household income.
The urban economics and regional science literatures (ALBOUY, 2008, 2009; FLORIDA and MELLANDER, 2010; GLAESER et al., 2001; GYOURKO et al., 2006) found that housing costs are somewhat of a proxy for higher levels of amenities and a generally higher quality of life. Thus, housing costs may accompany other attributes that positively affect happiness and well-being. Campbell and colleagues (CAMPBELL and CONVERSE, 1972; CAMPBELL et al., 1976) suggest that the objective character of housing is less important than the degree to which housing conditions meet aspirations. Furthermore, the comparison between where one currently lives and the best place one lived previously appears to be more important than the differ- ence between one’s current housing compared with that of others.
Several studies suggest other factors affecting the relationship between housing and life satisfaction (CARP, 1975; CATTANEO et al., 2007; KOZMA and STONES, 1983; MICHALOS, 1982). The link between housing satisfaction and well-being appears to vary in urban and rural contexts. For example, KOZMA and STONES (1983) found that housing satisfaction was a
The Happiness of Cities 615
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
significant predictor of happiness for urbanites, but not for rural dwellers. MITCHELL’s (1971) research on high-density housing in Hong Kong, China, found that the link between housing and happiness is not driven by housing per se, but by the factors that tend to accompany high-density housing in the neighbour- hood, such as dirty and crowded street environments that are socially unhealthy for communities.
Density. According to various studies, the effects of density on happiness are mixed. BELL (1992) points out that around the world there is a fairly consistent belief that less dense communities provide a higher quality of life than urban living. LAWLESS and LUCAS’s (2010) county-level study finds that life satisfaction is higher in counties with smaller and less dense popu- lations. PUTNAM (2000) found higher levels of social capital in rural and suburban areas compared with larger, more diverse urban areas. DAVIS and FINE- DAVIS (1991) found that rural dwellers in Ireland were more satisfied with life than were those in urban areas. Life satisfaction was related to a more relaxed, less pressured way of life – although this finding could also be a function of lower expectations. RICHMOND et al. (2000) found that respondents with 1 acre or more of land around their home reported greater satis- faction with their area and community compared with residents with less land. On the other hand, CRIDER et al.’s (1991) research in rural Pennsylvania, USA, found that rural location was only marginally associated with community satisfaction and there was no strong relationship between place of residence and happiness.
Age. Recent research on subjective well-being has identified a ‘U’-shaped relationship between well- being and age, with well-being being high among young and older adults, and lower among middle- aged adults (BLANCHFLOWER and OSWALD, 2008; DEATON, 2008). Most studies of national differences in well-being have not reported relationships between the average age of citizens and well-being, so it is not clear whether the age of citizens relates in any way to happiness. On the one hand, metropolitan areas with younger residents may be higher in well-being because younger people experience fewer health problems compared with older adults. On the other hand, it is conceivable that areas with more older adults are wealthier and therefore healthier and happier (WILKINSON, 1992) compared with areas with fewer young people.
Climate. Conventional wisdom suggests that happiness is related to climate and weather – specifically, that people are happier in warmer and sunnier places. The findings of the research literature are more mixed. In an investigation of temperature and happiness in sixty- seven countries, REHDANZ and MADDISON (2005) found that high average temperatures in the coldest
month were positively associated with happiness, while high mean temperatures during the hottest month were negatively associated with happiness. These findings are consistent with work of KELLER et al. (2005), which found that pleasant weather in spring was associated with a positive effect, but warm weather in summertime was negatively associated with positive effect. One hypothesis for these findings is that during the cold months and early spring, people have been deprived of nice weather and spent much of their time indoors, so seeing the sun and being able to spend time outside comfortably in spring is more appreciated.
SCHKADE and KAHNEMAN (1998) found that people believed the weather to be better, on average, in California than in the Midwest, and thus expected California residents to be happier than people living in the Midwest. However, while Californians were indeed more satisfied with their climate than were Midwesterners, self-reported overall life satisfaction was the same in both California and the Midwest, suggesting that weather may not have a consistent effect on well-being. MUESER and GRAVES (1995) posited that as societies become richer, people are willing to sacrifice their personal income to live in more pleasant, sunnier climes.
Even though the academic literature provides only limited evidence of the view that climate is associated with happiness, the popular belief persists that people who live in warmer, sunnier climates tend to lead happier lives.
Summary
National and individual-level research has found a close association between income and happiness, but only a small role for human capital. Studies of metropolitan economic performance have found that human capital is a strong predictor of regional economic outcomes including incomes. Moreover, recent state- and county-level studies have found that human capital is associated with happiness at those geographic scales. The present investigation aims to inform one’s under- standing of geographical differences in happiness by determining the role of human capital when controlling for income and other key economic factors that have been found to be important to life satisfaction across US metropolitan areas, which comprise economically integrated and meaningful geographic units.
Variables, data and methods
Next, a series of statistical techniques are used to examine the relationships between human capital and happiness across metropolitan areas, while controlling for the effects of income and other key economic factors. This section outlines the major variables, data sources and methods used in these analyses.
616 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
Dependent variable
Well-being index. The dependent variable is a multifa- ceted variable for well-being. It is based on the Gallup– Healthways Well-being Index. It ranges from a low of zero to a high of 100 and consists of six sub-indices for life evaluation, emotional health, work environ- ment, physical health, healthy behaviours, and access to basic needs. The telephone survey was conducted by the Gallup Organization and tracked US residents between 2 January and 29 December 2009. It includes 353000 individuals; and covers 184 US metropolitan regions. The maximum expected error varies based on metro size, ranging from less than 1% in larger cities with more than 1 million population to 3.1% for the smallest metros. Data are only available for the overall well-being index and not for the sub-indices on which it is based.1
Independent variables
The analysis employs a range of independent variables:
Human capital. The central hypothesis is that there is a strong association between human capital and happiness at the metropolitan scale. Human capital has been found to be strongly associated with metropolitan level econ- omic performance. The variable used for human capital is the standard one: the share of the labour force with a bachelor’s degree or above. The data are from the 2006 US Census.
Income measures. Three variables were used to capture the effects of income on material well-being:
. Income is the sum of the amounts reported separately for wage or salary income including net self-employ- ment income; interest, dividends, or net rental or royalty income or income from estates and trusts; social security or railroad retirement income; Sup- plemental Security Income (SSI); public assistance or welfare payments; retirement, survivor, or disability pensions; and all other income. It is measured on a per capita basis and is from the 2005 US Census.
. Wage income is a narrower measure and covers total money earnings received for work performed as an employee in the region. This measure includes wages, salary, armed forces pay, commissions, tips, piece-rate payments, and cash bonuses earned before deductions were made for taxes, bonds, pensions, union dues, etc. It is measured on a per worker basis and is gathered from the from the Bureau of Labor Statistics for 2006.
. Gross regional product is a measure of the value of regional production and is a reflection of regional productivity levels, but also a regional standard of living. The variable used here is calculated on a per capita basis using data from the 2005 US Census.
Income inequality. The measure of income inequality used was the standard Gini coefficient of inequality in household incomes. The data were from the Census’ American Community Survey for 2006–2008.
Unemployment. Two measures of unemployment were employed:
. Unemployment rate, that is, the share of the labour force without employment. The data are for December 2009.
. Change in unemployment measures the share of the labour force that was unemployed between Decem- ber 2008 and December 2009. Both are from the Bureau of Labor Statistics.
Housing. Three measures were used for housing:
. Housing cost measures the median values of owner- occupied housing units for 2009. The data are from the American Community Survey.
. Housing affordability is a ratio of housing costs (median housing value) to wages. The wage data were from the Bureau of Labor Statistics for 2006; the housing cost data were from the American Community Survey.
. Homeownership share is the share of the owner-occu- pied housing units. The data source is the American Community Survey for 2006–2008.
Density. The measure of density is population per square kilometre. The data are from the 2006 US Census.
Age structure. This is the population median age taken from the American Community Survey, US Census Bureau, for 2006–2008.
Average commute time. This was the average commuting time to work; the source was the American Community Survey from the US Census Bureau for 2006–2008.
Climate. Three climate variables were employed: average temperature in January, in July, and the differ- ence between the two (to capture temperature stability around the year). The data were from the US Geologi- cal Survey.
Table 1 presents the descriptive statistics for all variables.
Appendix A provides descriptive statistics for the US metropolitan regions that are not included in this analy- sis due to missing values, mainly associated with the cov- erage of the Gallup–Healthways Well-being Index. The Gallup–Healthways metros have higher mean values for income, wage income and gross regional product per capita, higher human capital levels, greater inequality, lower rates of unemployment, higher housing values, and higher densities than the metros with missing
The Happiness of Cities 617
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
values. The two sets of metros differ little in temperature or climate.
Methods
A variety of statistical methods were used. First, bivariate correlation analyses were conducted to identity relation- ships between well-being and key economic and social factors. Second, partial correlations were run, control- ling for the effects of the income measures. Finally, multivariate ordinary least-squares (OLS) regression analysis was used to examine the effects of human capital on happiness, controlling for income measures and other factors.
Findings
To orient the analysis, Fig. 1 provides a map of the vari- ation in happiness across US metro areas.
Overall, the difference between the highest and lowest well-being scores revealed no dramatic disparities between American cities. The lowest score is 59.5 for Fort Smith, Arkansas–Oklahoma, while the highest is 72.5 for Boulder, Colorado. Nonetheless, the map illus- trates that there is significant variation between regions. In general, the happiest (black) regions tend to be part of the large mega-regions near the coasts, including Seattle and Portland (Cascadia), parts of the Bay Area (Nor-Cal) and Greater Los Angeles (So-Cal), and Boston and Washington (Bos-Wash). Cities in the Midwest and Texas typically scored near the middle of the well- being scale, while Florida and some parts of the South displayed considerably lighter shades, suggesting rela- tively low scores. In general, dramatic variation
between adjacent regions is rare, suggesting that proxi- mity matters somewhat to the happiness of regions.
Correlation analysis
Table 2 summarizes the results of the bivariate corre- lations among the key measures. Since previous research (STEVENSON and WOLFERS, 2008) found a significant relationship between income and well-being at the national level, partial correlations were run, controlling for the effects of income on happiness using the best- performing measure of income or material well-being: wage income.
Human capital. The correlation analysis provides partial confirmation of the core hypothesis. It finds a strong, positive and significant correlation with metropolitan well-being. In fact, the correlation between the two is the highest of any variable in the analysis (0.681). The correlation between human capital and happiness remains the strongest among all variables when partial correlations controlled for the effects of wage levels (0.582) were run. Fig. 2 is a scatter-graph of the relation- ship between human capital and well-being. The line is steep and the fit is good.
Income. The correlations for the main income measures – average income, wages and output per capita – are significant, but weaker than for human capital. The strongest correlation is for wage income (0.450) followed by income (0.403), and then gross regional product or economic output per capita (0.372). Wage income is also the control variable in the partial correlations above. Fig. 3 illustrates the
Table 1. Descriptive statistics
n Minimum Maximum Mean Standard deviation
Well-being 184 59.50 72.50 65.897 2.123 Human capital 170 0.09 0.32 0.170 0.040 Gross regional product per capita 170 19828 86069 48502 12369 Income 170 20837 69118 41531 7422 Wagesa 170 32639 72277 44841 5639 Income inequality 184 0.395 0.543 0.447 0.023 Median housing valueb 184 76000 638300 203470 102866 Housing-to-wage ratio 170 1.93 14.19 4.526 2.487 Homeownership share 171 0.519 0.817 0.678 0.052 Unemployment 184 4.10 17.50 9.663 2.655 Unemployment change 184 0.40 5.00 2.50 0.979 Population density 170 5.16 1020.34 143.29 146.68 Age 171 24.4 46.8 36.58 3.31 Commute time 171 19.61 37.88 25.20 3.46 January temperature 181 9.40 66.50 36.89 12.18 July temperature 181 61.80 91.15 75.29 5.58 Temperature stability 181 11.40 59.64 38.40 10.14 Valid n (list-wise) 170
Notes: aExpressed in thousands of US dollars in the regression analysis. bExpressed as 100000 of US dollars in the regression analysis.
618 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
bivariate relationship between wages and well-being. The fitted line is not as steep as the one for human capital above.
Income inequality. Metropolitan-level happiness is not associated with income inequality. Correlations of –0.040 in the bivariate correlation and –0.012 in the partial correlation were found.
Unemployment. Unemployment is significantly related to metropolitan-level happiness. The correlations to the two unemployment variables are negative and significant: –0.344 for the unemployment rate and –0.300 for the annual change in unemployment. These correlations remain significant when partial correlations controlling for wage levels, at –0.305 and –0.248, were run. Fig. 4 shows the relationships between well-being and unem- ployment and changes in unemployment.
Housing. Housing values are also significantly corre- lated with metropolitan well-being (0.494), and stay sig- nificant also when wages are partialled out (0.301). The measure for housing affordability – the housing value- to-wage ratio – is also significantly associated with metropolitan happiness (0.483), and remains significant in the partial correlation (0.388). Homeownership is negatively and significantly associated with metropolitan happiness, but becomes insignificant when controlling for regional wages. Fig. 5 plots the relationship between well-being and median housing values, as well as between well-being and housing-to-wage ratio.
Fig. 1. Well-being for 184 US metropolitan regions
Table 2. Correlation analysis findings for metropolitan well- being
Bivariate correlation
Partial correlation controlling for average
wage
Gross regional product per capita
0.372*** n.a.
Income 0.403*** n.a. Wages 0.450*** n.a. Human capital 0.681*** 0.562*** Income inequality –0.040 –0.012 Median housing value 0.494*** 0.301*** Housing-to-wage ratio 0.483*** 0.388*** Homeownership –0.187** –0.053 Unemployment –0.344*** –0.305*** Change unemployment –0.300*** –0.248*** Density 0.088 –0.195** Age –0.290*** –0.298*** Commute time 0.083 –0.161** January temperature –0.018 0.081 July temperature –0.217*** –0.024 Temperature stability –0.098 –0.108
Note: *Statistically significant at the 0.01 level; **statistically signifi- cant at the 0.05 level; and ***statistically significant at the 0.01 level.
The Happiness of Cities 619
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
Density. Density is not correlated with metropolitan- level happiness, but the correlation is significant and negative (–0.195) when controlled for wages. In other words, more dense places are related to lower levels of happiness when wages are controlled for.
Age. The median age of the population is significant and negatively related to happiness (–0.290). In other words, younger places are happier places. This relation even becomes slightly stronger when one controls for wage (–0.298).2
Commute time. Commute time is not associated with metropolitan happiness, but when controlled for wages the correlation is negative and significant (–0.161).
Climate. The results for the climate variables suggest that climate does not play a role in metropolitan happi- ness. The correlations are insignificant for two climate variables: January temperature and temperature stability. The third variable, July temperature, is negatively and
significantly related to metropolitan happiness (–0.217), though this relationship becomes insignificant when wages are controlled for.
Regression findings
Next are the findings of the regression analysis. Only those variables that were statistically significant in the correlation analysis are included, while those that were insignificant are removed. Given the nature of the vari- ables, one would expect to find some collinearity. Thus, several regressions were run, substituting variables that correlate strongly with one another. Variance inflation factor (VIF) value tests were also run to control for possible multicollinearity effects.
The first model examines the relationship between metro-level happiness and income, since earlier research suggests a close connection between the two in cross- national and individual levels studies. It includes the best performing measure of regional ‘income’ – wage income – from the previous correlation analysis. Other key economic measures that were significant in the correlation analysis are also included. These include two measures for unemployment – unemployment rate and change in the unemployment rate – and two measures for housing: median housing value and the housing value-to-wage ratio. Table 3 summarizes the results for these regressions.
The first version of the model (equation 1.1) exam- ines how much of the variation is explained by wage income alone. R2 is 0.20, suggesting that wage income on its own explains roughly 20% of the variation in well-being. The second version of the model (equation 1.2) adds the unemployment variables. R2
for this model increases slightly to 0.292. The relatively high VIF values show that there is strong collinearity between the two unemployment variables shown by the relatively high VIF values. Still, the unemployment rate variable – that is, the variable for the absolute level of unemployment – shows a relatively stronger relation- ship to metropolitan happiness than change in the unemployment rate. The third version of the model (equation 1.3) includes the two housing variables along- side wage income. R2 here is 0.328, just slightly better than equation (1.2) for just wage income in combi- nation with unemployment, and approximately 0.13 more than if wage income alone is used. Again, note the strong collinearity between these housing variables which generate VIF values between 54 and 78. That said, the housing affordability measure (housing cost to wages) is the stronger measure of the two. The fourth version of the model includes wage income alongside the best-performing unemployment and housing vari- ables. This model (equation 1.4) generates an R2 of 0.398, which is the best performing of all the models, explaining roughly 40% of the variation.
The first model above sets up a second model which also includes human capital alongside wage income.
Fig. 2. Human capital and well-being
Fig. 3. Wage income and well-being
620 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
Also included are the best-performing housing and unemployment variables, as well as commuting time and density (which were positive and significant in the partial correlation analysis) as control variables. Table 4 summarizes the results of these models.
The first version of the model (equation 2.1) exam- ines how much of the variation is explained by human capital alone. The regression generates an R2 of 0.464, implying that human capital alone explains almost half of metropolitan happiness. This is more than double that of the wage-income model above (equation 1.1). The second model (equation 2.2) adds the variables for wage income, unemployment level, housing cost to wages and also age. The model generates an even higher R2 of 0.578, and human capital remains the strongest variable. Human capital is excluded from the next model (equation 2.3) and the regression is rerun. This regression generates an R2 of 0.482, smaller than for equation (2.2) and just slightly more than equation (2.1). The fourth model (equation 2.4) puts human capital back in, in place of the wage income variable. This version of the model generates an R2 of 0.578. Human capital thus adds significantly more explanatory
power to well-being than the best-performing measure of income – wage levels income. Note, however, that this model (equation 2.4) adds approximately only 0.11 in additional explanatory power to the R2 value beyond the initial model (equation 2.1), which was for human capital alone. Clearly, human capital is the strongest-performing variable in this series of models explaining metropolitan well-being and happiness.
Taken together, the multivariate regression results show the significant impact of human capital on metro- politan happiness. The best-performing measure of income – wages – is also significant, but it does not perform as well as human capital. Furthermore, a wide body of research in urban economics (GLAESER et al., 1992; GLAESER and MARÉ, 2001; RAUCH, 1993) finds that human capital levels are the key predictor of income to begin with.
Unemployment is also a significant factor: regions with lower levels of unemployment have significantly higher levels of happiness. Housing costs are also signifi- cantly associated with happiness,3 but interestingly hap- piness is higher in metropolitan areas where housing is less affordable. While this may seem counterintuitive
Fig. 4. Unemployment, change in unemployment and well-being
Fig. 5. Housing values, housing-to-wage ratio and well-being
The Happiness of Cities 621
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
at first, research in urban economics (ALBOUY, 2008, 2009; FLORIDA and MELLANDER, 2010; GLAESER et al., 2001; GYOURKO et al., 2006) finds that metropo- litan housing prices reflect a combination of productivity and quality-of-life amenities. Accordingly, one would expect higher housing prices and higher housing price- to-wage ratios in places that are more economically pro- ductive and offer higher levels of amenities and quality of life, two factors that are likely to affect happiness and sub- jective well-being. In these regions, individuals are willing and able to pay more for these housing-related attributes, which in turn affect happiness. A significant and negative relationship is also found between age struc- ture and happiness. Younger cities are happier cities. This investigation also finds that a number of factors thought to affect metropolitan, city or community-level happiness – density and commute time – do not appear to play a statistically significant role in metropolitan-level happi- ness. It is worth noting, however, that some of the vari- ables (change in unemployment and housing values) excluded from the regression analyses due to collinearity problems are associated with housing affordability and levels of unemployment, both of which were significant in the multivariate regression analysis.
CONCLUSION
The present research examined the factors that influence happiness at the metropolitan level. The metropolitan
level is important since individuals actively seek out locations, identity with their place of residence, and derive considerable satisfaction as well as emotional attachment from them. While previous cross-national research (DEATON, 2008; DIENER, 1984; DIENER and DIENER, 1995; EASTERLIN, 1974, 1995; STEVENSON and WOLFERS, 2008) focused on the effects of income and observed that human capital accumulation plays only a limited role in national-level happiness, the present authors hypothesized that human capital would play a significant role in happiness at the metro- politan level, given its strong role in shaping metropoli- tan economic and social outcomes. The hypothesis was tested using new data from a large-scale Gallup–Health- ways survey of happiness and well-being across 170 metropolitan areas. Correlation analyses were con- ducted, including partial correlations to control for wage levels, and regression analysis to probe for the effects of human capital on metropolitan-level happi- ness, alongside income and related measures of econ- omic performance, unemployment, income inequality, housing cost and affordability, density, age structure, commuting time, and climate, which are thought to effect happiness.
Overall, the findings confirmed the human capital and happiness hypothesis. Specifically, the analysis shows that human capital plays a key role in metropoli- tan-level happiness and well-being, more so than income wages or other variables. This is in line with pre- vious studies at the state (RENTFROW et al., 2009) and
Table 3. Regression findings for wage income, unemployment, and housing variables
Equation (1.1) Equation (1.2) VIF Equation (1.3) VIF Equation (1.4) VIF
Constant 58.148*** (1.191) 60.839** (1.277) 50.991*** (3.044) 62.294 (1.294) Wage income 0.172*** (0.000) 0.163*** (0.000) 1.006 0.284*** (0.067) 7.531 0.092*** (0.027) 1.332 Unemployment rate –0.201* (0.081) 2.392 –0.261** (–0.049) 1.013 Change in unemployment –0.136 (0.222) 2.386 Housing value –0.035*** (0.012) 78.213 Housing-to-wage ratio 2.084*** (0.577) 54.490 0.463*** (0.085) 1.326 R2 0.202 0.292 0.328 0.398 R2 adjusted 0.197 0.279 0.315 0.387 n 169 169 169 169
Note: VIF, variance inflation factor.
Table 4. Human capital regression results
Equation (2.1) Equation (2.2) VIF Equation (2.3) VIF Equation (2.4) VIF
Constant 59.677*** (0.527) 66.017*** (1.887) 67.681*** (2.065) 66.012*** (1.794) Average wage level 0.000 (0.033) 2.906 0.128*** (0.029) 1.740 Unemployment –0.073 (0.049) 1.422 –0.227*** (0.048) 1.043 –0.074** (0.047) 1.351 Housing-to-wage ratio 0.394*** (0.075) 1.423 0.491*** (0.084) 1.359 0.394*** (0.074) 1.396 Human capital 36.292*** (3.009) 27.727*** (4.546) 2.799 27.705*** (3.507) 1.676 Population density 0.001 (0.001) 1.823 –0.002 (–0.002) 1.814 –0.001 (0.001) 1.692 Age –0.119*** (0.035) 1.048 –0.146*** (0.038) 1.032 –0.119*** (0.035) 1.048 Commute time –0.053 (0.042) 1.782 –0.076 (–0.047) 1.767 –0.053 (0.041) 1.695 R2 0.464 0.578 0.482 0.578 R2 adjusted 0.461 0.560 0.463 0.563 n 169 169 169 169
Note: VIF, variance inflation factor.
622 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
the county levels (LAWLESS and LUCAS, 2010). Further- more, a wide body of research in urban economics (GLAESER et al., 1992; GLAESER and MARÉ, 2001; RAUCH, 1993) finds that human capital levels are the key predictor of income. Unemployment is also a sig- nificant factor: regions with lower levels of unemploy- ment have significantly higher levels of happiness. Housing costs are significantly associated with happiness, but interestingly happiness is higher in metropolitan areas where housing is less affordable. While this may seem counterintuitive at first, research in urban economics (ALBOUY, 2008, 2009; FLORIDA and MELLANDER, 2010; GLAESER et al., 2001; GYOURKO et al., 2006) finds that metropolitan housing prices reflect a combination of productivity and quality-of-life amenities. Accordingly, one would expect higher housing prices and higher housing price-to-wage ratios in places that are more economically productive and also offer higher levels of amenities and quality of life, two factors that are likely to affect happiness and subjec- tive well-being. In these regions, individuals are willing and able to pay more for housing-related attributes, which in turn affect happiness. The inverse relationship between metropolitan-level well-being and age is likely because areas with more young people are generally heal- thier compared with places with more older adults.
This statistical analysis also finds that a number of factors that have been thought to affect metropolitan, city or community-level happiness, such as density and commute time, do not appear to play a statistically sig- nificant role in metropolitan-level happiness. Density is not correlated with metropolitan-level happiness, and the correlation is negative and significant (–0.195) when controlling for wages. That is, more dense places appear to generate lower levels of happiness when wages are held constant. While numerous studies have found commute time to be among the factors that contribute to low levels of happiness, commute time is not significant in the regressions here, and it is not significantly associated with happiness in the correlation analysis. It is, however, negatively (though weakly) associated with happiness when partial correlations are run that control for wage levels. While some continue to believe people are happier in warmer and sunnier places, the present results do not support that claim. The correlation results for the three climate variables suggest that climate does not play a role in metropolitan happiness. The correlations are statistically non-significant for two climate variables: January temperature and temperature stability. The third variable, July temperature, is negatively and signifi- cantly related to metropolitan happiness, though this relationship is eliminated when wages are partialled out.
More broadly, the results from this investigation suggest that the effect of income on happiness at the metropolitan level is driven more by human capital than by income. Not only does human capital signifi- cantly affect income, but also it affects a number of
variables associated with well-being, such as a sense of control over life, stable and supportive relationships, and occupational resilience. Human capital thus influ- ences several important life domains and its effects can be observed at the metropolitan level.
Higher levels of human capital are associated with better employment opportunities and greater employ- ment options. JUDGE et al. (2001) found that individuals with a college degree not only have more occupational opportunities, but also engage in more challenging, sti- mulating and satisfying work. Education gives people access to non-alienating paid work and economic resources that, along with schooling itself, increase the sense of control over life and explains much of edu- cation’s positive effects on psychological well-being. Human capital is also closely associated with unemploy- ment. College-educated adults in the United States face levels of unemployment that are markedly lower than those for high-school graduates and even more so for individuals without high-school degrees. Compared with people without a college degree, college-educated adults also tend to have higher-level and more flexible skills that enable them to switch jobs more readily if laid off, or to find more fulfilling employment if their job becomes less interesting.
Human capital also contributes to people’s ability to respond to challenges effectively and to maintain a sense of mastery over the environment. MIROWSKY and ROSS (1989) and ROSS and MIROWSKY (1992) observed a positive relationship between education and sense of control, such that well-educated individuals had a greater sense of personal control than the less-edu- cated, even after adjusting for employment, job auton- omy, earning, minority status, age and marital status (BIRD and ROSS, 1993; ROSS and MIROWSKY, 1992). High personal control facilitates the development of effective and flexible coping strategies (MIROWSKY and ROSS, 1989; TURNER and NOH, 1983; WHEATON, 1983).
Furthermore, human capital is positively related to other factors that affect happiness, such as stable social relationships and social support (ROSS and WILLIGEN, 1997). Human capital is associated with more stable marriages and family ties – factors that are closely corre- lated with subjective well-being. Research has found college-educated adults in the United States (GLENN and SUPANCIC, 1984) and Norway (LYNGSTAD, 2004) have lower levels of divorce and more stable mar- riages compared with adults with less education. More educated individuals postpone marriage and have more opportunities over time to select more suitable partners (DIXON, 1978; GOLDSTEIN and KENNEY, 2001). In these ways, human capital affords people opportunities and resources to enrich their lives and thus has both direct as well as indirect effects on happiness.
However, it is important to acknowledge that research at the national level has not found a strong
The Happiness of Cities 623
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
relationship between human capital and well-being. Why might human capital or education level have a stronger relationship with happiness at the metropolitan level than at the national level? It is clear that income itself turns on human capital. Previous research has documented the strong effects of human capital on metropolitan-level income, so it could be said that human capital in effect captures or accounts for much of the effect of income on happiness as well. It is simply a better underlying measure of the key factor that affects happiness.
Previous research (BERRY and GLAESER, 2005) has also shown growing variation or ‘divergence’ in human capital across metropolitan regions. Metropoli- tan regions are natural economic units – by definition comprising a labour market and commuting shed – so they are more coherent economic units than nations or even states. The metropolitan level is the scale where key economic drivers such as human capital truly matter, and where their effects are less muddied by political or jurisdictional boundaries. Higher levels of human capital are not only associated with higher income, but also higher human capital metros have many other attributes that have been found to affect
well-being. Higher levels of human capital at the metro level are associated with better health outcomes, lower levels of smoking and obesity, lower levels of crime, better schools, better quality housing, more natural amenities, higher levels of consumer amenities such as restaurants and cultural amenities, higher levels of openness and diversity, and a higher quality of life more generally – all factors that can and do affect happi- ness. While this research has documented the connec- tion between human capital and happiness, further investigation is needed to gauge more precisely the interaction between human capital and these and other metropolitan-level factors that may directly or indirectly shape happiness and well-being.
Whether by providing greater occupational opportu- nities or an enhanced sense of control over one’s environment, education offers important advantages to individuals. The present work shows that those advan- tages are far reaching and can actually influence the sub- jective well-being of one’s friends and neighbours. If the psychological and physical health of individuals is to improve, it is crucial that a clearer understanding of pre- cisely how human capital impacts the well-being of cities and metropolitan regions must be gained.
APPENDIX A
Table A1. Descriptive statistics for excluded regions due to a lack of well-being data
n Minimum Maximum Mean Standard deviation
Gross domestic product per capita 152 19727 77581 41175.71 9854 Average income 153 22213 64687 36070 5553 Average wage 152 31566 52169 40287 3596 Income inequality 183 0.381 0.515 0.440 0.027 Median housing value 175 76400 456100 147590 56650 Housing-to-wage ratio 152 2.07 8.82 3.583 1.176 Homeownership 175 0.493 0.807 0.68 0.057 Unemployment 183 4.00 27.70 9.43 3.46 Change in unemployment 183 –1.20 5.90 2.42 1.17 Human capital 153 0.07 0.27 0.142 0.042 Population density 152 2.59 222.86 58.19 36.50 Age 175 25.30 50.90 35.08 4.11 Commute time 175 15.79 31.36 22.23 .2.09 January temperature 182 3.95 64.00 33.91 12.00 July temperature 182 61.90 93.70 75.95 5.43 Temperature stability 182 11.90 65.35 42.04 8.78 Valid n (list-wise) 149
624 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
NOTES
1. While there are no publicly available data for the sub- indices at the metro level, RENTFROW et al. (2009) exam- ined the relations between the well-being index and the sub-indices at the state level: life evaluation (current and future prospects) correlates with 0.61; emotional health (daily positive and negative experience) correlates with 0.58; physical health (for example, body mass index (BMI) and days absent from work) correlates with 0.58; healthy behaviour (for example, smoking, eating habits
and exercise) correlates with 0.58; work environment (for example, job satisfaction and relations with supervi- sors) correlates with 0.46; and basic access (clean water, medicine, healthy food and healthcare access) correlates with 0.24, but is insignificant.
2. Age squared was also employed to capture non-linearities. The results remained negative and significant.
3. While this study employ 2009 Census data for housing values, the regressions were also run with 2006 housing value data (before the financial crisis) with similar results.
REFERENCES
ALBOUY D. (2008) Are Big Cities Bad Places to Live? Estimating Quality of Life across Metropolitan Areas. Working Paper Series Number 14472. National Bureau of Economic Research (NBER), Cambridge, MA (available at: http://www.nber.org/papers/ w14472).
ALBOUY D. (2009) What are Cities Worth? Land Rents, Local Productivity, and the Capitalization of Amenity Values. Research Working Paper Series Number 14981. National Bureau of Economic Research (NBER), Cambridge, MA (available at: http://www. nber.org/papers/w14981).
BELL D. H. (1992) The law clinic as a regional center: looking for solutions to rural southern housing problems, Washington University Journal of Urban and Contemporary Law 4, 101–126.
BERRY C. R. and GLAESER E. L. (2005) The divergence of human capital levels across cities, Papers in Regional Science 84(3), 407–444.
BIRD C. E. and ROSS C. E. (1993) Houseworkers and paid workers: qualities of the work and effects on personal control, Journal of Marriage and Family 55(4), 913–925.
BLANCHFLOWER D. G. and OSWALD A. J. (2004) Money, sex and happiness: an empirical study, Scandinavian Journal of Economics 106(3), 393–415.
BLANCHFLOWER D. G. and OSWALD A. J. (2008) Is well-being U-shaped over the life cycle?, Social Science and Medicine 66(8), 1733–1749.
CAMPBELL A. and CONVERSE P. E. (1972) The Human Meaning of Social Change. Russell Sage Foundation, New York, NY. CAMPBELL A., CONVERSE P. E. and RODGERS W. L. (1976) The Quality of American Life: Perceptions, Evaluations, and Satisfactions.
Russell Sage Foundation, New York, NY. CARP F. M. (1975) Impact of improved housing on morale and life satisfaction, Gerontologist 15(6), 511–515. CATTANEO M. D., GALIANI S., GERTLER P. J., MARTINEZ S. and TITIUNIK R. (2007) Housing, health, and happiness, SSRN eLibrary
(available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=981822). CLARK A., GEORGELLIS Y. and SANFEY P. (2001) Scarring: the psychological impact of past unemployment, Economica 68(270),
221–241. CLARK A. E. and OSWALD A. J. (1994) Unhappiness and unemployment, Economic Journal 104(424), 648–659. CLARK A. E. and OSWALD A. J. (1996) Satisfaction and comparison income, Journal of Public Economics 61(3), 359–381. CRIDER D. M., WILLITS F. K. and KANAGY C. L. (1991) Rurality and well-being during the middle years of life, Social Indicators
Research 24(3), 253–268. DAVIS E. E. and FINE-DAVIS M. (1991) Social indicators of living conditions in Ireland with European comparisons, Social Indicators
Research 25(2/4), i–365. DEATON A. (2008) Income, health, and well-being around the world: evidence from the Gallup World Poll, Journal of Economic
Perspectives 22(2), 53–72. DIENER E. (1984) Subjective well-being, Psychological Bulletin 95, 542–575. DIENER ED. (2000) Subjective well-being: the science of happiness and a proposal for a national index, American Psychologist 55(1),
34–43. DIENER ED. and DIENER M. (1995) Cross-cultural correlates of life satisfaction and self-esteem, Journal of Personality and Social
Psychology 68(4), 653–663. DIENER ED., DIENER M. and DIENER C. (1995) Factors predicting the subjective well-being of nations, Journal of Personality and
Social Psychology 69(5), 851–864. DIENER ED. and LUCAS R. E. (1999) Explaining differences in societal levels of happiness: relative standards, need fulfillment,
culture, and evaluation theory, Journal of Happiness Studies 1(1), 41–78. DIENER ED., OISHI S. and LUCAS R. E. (2003) Personality, culture, and subjective well-being: emotional and cognitive evaluations
of life, Annual Review of Psychology 54(1), 403–425. DIENER ED., SUH E. M., LUCAS R. E. and SMITH H. L. (1999) Subjective well-being: three decades of progress, Psychological Bulletin
125(2), 276–302.
The Happiness of Cities 625
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
DIXON R. B. (1978) Late marriage and non-marriage as demographic responses: are they similar?, Population Studies 32(3), 449–466.
EASTERLIN R. (1974) Does economic growth improve the human lot? Some empirical evidence, Nations and Households in Economic Growth 89–125.
EASTERLIN R. A. (1995) Will raising the incomes of all increase the happiness of all?, Journal of Economic Behavior and Organization 27(1), 35–47.
FLORIDA R. (2009) Who’s Your City?: How the Creative Economy is Making Where to Live the Most Important Decision of Your Life. Random House of Canada, Toronto, ON.
FLORIDA R. and MELLANDER C. (2010) There goes the metro: how and why bohemians, artists and gays affect regional housing values, Journal of Economic Geography 10(2), 167–188.
FLORIDA R., MELLANDER C. and STOLARICK K. (2011) Beautiful places: the role of perceived aesthetic beauty in community satisfaction, Regional Studies 45(1), 33–48.
GLAESER E. L., KALLAL H. D., SCHEINKMAN J. A. and SHLEIFER A. (1992) Growth in cities, Journal of Political Economy 100(6), 1126–1152.
GLAESER E. L., KOLKO J. and SAIZ A. (2001) Consumer city, Journal of Economic Geography 1(1), 27–50. GLAESER E. L. and MARÉ D. C. (2001) Cities and skills, Journal of Labor Economics 19(2) (available at: http://papers.ssrn.com/sol3/
papers.cfm?abstract_id=920975). GLAESER E. L., RESSEGER M. and TOBIO K. (2009) Inequality in cities, Journal of Regional Science 49(4), 617–646. GLENN N. D. and SUPANCIC M. (1984) The social and demographic correlates of divorce and separation in the United States: an
update and reconsideration, Journal of Marriage and Family 46(3), 563–575. GOLDSTEIN J. R. and KENNEY C. T. (2001) Marriage delayed or marriage forgone? New cohort forecasts of first marriage for U.S.
women, American Sociological Review 66(4), 506–519. GRAHAM C. (2008) The New Palgrave Dictionary of Economics. The Economics of Happiness, 2nd Edn. Palgrave Macmillan, Basingstoke. GRAHAM C. (2009) Happiness Around the World: The Paradox of Happy Peasants and Miserable Millionaires. Oxford University Press,
New York, NY. GYOURKO J., MAYER C. and SINAI T. (2006) Superstar Cities. Working Paper Series Number 12355. National Bureau of Economic
Research (NBER), Cambridge, MA (available at: http://www.nber.org/papers/w12355). INGLEHART R. and KLINGEMANN H. D. (2003) Genes, culture, democracy, and happiness, in SUH E. M. (Ed.) Culture and Subjective
Well-being, pp. 165–183. MIT Press, Cambridge, MA. JUDGE T. A., THORESEN C. J., BONO J. E. and PATTON G. K. (2001) The job satisfaction–job performance relationship: a qualitative
and quantitative review, Psychological Bulletin 127(3), 376–407. KELLER M. C., FREDRICKSON B. L., YBARRA O., CÔTÉ S., JOHNSON K., MIKELS J., CONWAY A. and WAGER T. (2005) A warm heart
and a clear head, Psychological Science 16(9), 724–731. KOZMA A. and STONES M. J. (1983) Predictors of happiness, Journal of Gerontology 38(5), 626–628. KRUEGER A. B., KAHNEMAN D., FISCHLER C., SCHKADE D., SCHWARZ N. and STONE A. A. (2008) Time use and subjective well-
being in France and the U.S, Social Indicators Research 93(1), 7–18. LAWLESS N. M. and LUCAS R. E. (2010) Predictors of regional well-being: a county level analysis, Social Indicators Research 101(3),
341–357. LAYARD R. (2005) Happiness: Lessons form a New Science. Penguin, London. LYNGSTAD T. (2004) The impact of parent’s and spouses’ education on divorce rates in Norway, Demographic Research 10(5),
121–142. LYNN M. and STEEL P. (2006) National differences in subjective well-being: the interactive effects of extraversion and neuroticism,
Journal of Happiness Studies 7(2), 155–165. MICHALOS A. C. (1982) The satisfaction and happiness of some senior citizens in rural Ontario, Social Indicators Research 11(1), 1–30. MIROWSKY J. and ROSS C. E. (1989) Social Causes of Psychological Distress. de Gruyter, New York, NY. MITCHELL R. E. (1971) Some social implications of high density housing, American Sociological Review 36(1), 18–29. MUESER P. R. and GRAVES P. E. (1995) Examining the role of economic opportunity and amenities in explaining population
redistribution, Journal of Urban Economics 37(2), 176–200. PUTNAM R. D. (2000) Bowling Alone: The Collapse and the Revival of American Democracy. Simon & Schuster, New York, NY. RAUCH J. (1993) Productivity gains from geographic concentration of human capital: evidence from the cities, Journal of Urban
Economics 34, 380–400. REHDANZ K. and MADDISON D. (2005) Climate and happiness, Ecological Economics 52(1), 111–125. RENTFROW P. J., MELLANDER C. and FLORIDA R. (2009) Happy States of America: a state-level analysis of psychological, economic,
and social well-being, Journal of Research in Personality 43(6), 1073–1082. RICHMOND L., FILSON G. C., PAINE C., PFEIFFER W. C. and TAYLOR J. R. (2000) Non-farm rural Ontario residents’ perceived
quality of life, Social Indicators Research 50, 159–186. ROSS C. E. and MIROWSKY J. (1992) Households, employment, and the sense of control, Social Psychology Quarterly 55(3), 217–235. ROSS C. E. and WILLIGEN M. V. (1997) Education and the subjective quality of life, Journal of Health and Social Behavior 38(3),
275–297. SCHKADE D. A. and KAHNEMAN D. (1998) Does living in California make people happy? A focusing illusion in judgments of life
satisfaction, Psychological Science 9(5), 340–346. SJAASTAD L. A. (1962) The costs and returns of human migration, Journal of Political Economy 70(5), 80–93.
626 Richard Florida et al.
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
STEEL P. and ONES D. S. (2002) Personality and happiness: a national-level analysis, Journal of Personality and Social Psychology 83(3), 767–781.
STEVENSON B. and WOLFERS J. (2008) Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox. Working Paper Series Number 14282. National Bureau of Economic Research (NBER), Cambridge, MA (available at: http://www.nber.org/ papers/w14282).
STUTZER A. (2004) The role of income aspirations in individual happiness, Journal of Economic Behavior and Organization 54(1), 89–109.
STUTZER A. and FREY B. S. (2008) Stress that doesn’t pay: the commuting paradox, Scandinavian Journal of Economics 110(2), 339–366.
TIEBOUT C. M. (1956) A pure theory of local expenditures, Journal of Political Economy 64(5), 416–424. TURNER R. J. and NOH S. (1983) Class and psychological vulnerability among women: the significance of social support and
personal control, Journal of Health and Social Behavior 24(1), 2–15. VEENHOVEN R. (1993) Happiness in Nations: Subjective Appreciation of Life in 56 Nations 1946–1992 Erasmus University Press,
Rotterdam. WHEATON B. (1983) Stress, personal coping resources, and psychiatric symptoms: an investigation of interactive models, Journal of
Health and Social Behavior 24(3), 208–229. WILKINSON R. G. (1992) Income distribution and life expectancy, British Medical Journal 304(6820), 165–168. WINKELMANN L. and WINKELMANN R. (1998) Why are the unemployed so unhappy? Evidence from panel data, Economica 65(257),
1–15. WITTER R. A., OKUN M. A., STOCK W. A. and HARING M. J. (1984) Education and subjective well-being: a meta-analysis,
Educational Evaluation and Policy Analysis 6(2), 165–173.
The Happiness of Cities 627
D ow
nl oa
de d
by [
L ib
ra ry
S er
vi ce
s, U
ni ve
rs it
y of
t he
W es
t of
E ng
la nd
] at
1 3:
42 1
2 D
ec em
be r
20 14
