ArticlePDF Available

Relationship between well-being and recycling rates: evidence from life satisfaction approach in Britain


Abstract and Figures

This study explores the relationship between self-reported well-being and recycling rates. The estimates are based on Britain using data from the British Household Panel Survey. The effects of recycling rates on individuals’ happiness are estimated. Two approaches are followed. The first approach refers to panel probit-ordinary least squares (OLS). The second approach is the latent class generalised ordered probit. The results support that a significant positive relationship between self-reported well-being and recycling is presented.
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
Download by: [] Date: 03 November 2015, At: 06:31
Journal of Environmental Economics and Policy
ISSN: 2160-6544 (Print) 2160-6552 (Online) Journal homepage:
Relationship between well-being and recycling
rates: evidence from life satisfaction approach in
Eleftherios Giovanis
To cite this article: Eleftherios Giovanis (2014) Relationship between well-being and recycling
rates: evidence from life satisfaction approach in Britain, Journal of Environmental Economics
and Policy, 3:2, 201-214, DOI: 10.1080/21606544.2014.883941
To link to this article:
Published online: 07 Feb 2014.
Submit your article to this journal
Article views: 208
View related articles
View Crossmark data
Relationship between well-being and recycling rates: evidence from
life satisfaction approach in Britain
Eleftherios Giovanis
Laboratory for the Analysis of Complex Economic Systems, IMT Institute for Advanced Studies,
Piazza S. Ponziano 6, 55100 Lucca, Italy;
Department of Economics, Royal Holloway
University of London, TW20 0EX, Egham, England
(Received 22 August 2013; accepted 13 January 2014)
This study explores the relationship between self-reported well-being and recycling
rates. The estimates are based on Britain using data from the British Household Panel
Survey. The effects of recycling rates on individuals’ happiness are estimated. Two
approaches are followed. The first approach refers to panel probit-ordinary least
squares (OLS). The second approach is the latent class generalised ordered probit. The
results support that a significant positive relationship between self-reported well-being
and recycling is presented.
Keywords: air pollution; happiness; life satisfaction approach; recycling; subjective
JEL codes: C23, C26, D60, H41, Q51
1. Introduction
Recycling has traditionally occurred because it has been economically viable. From the
1970s onwards, however, the perception in modern rich societies has been that we should
recycle even more, something that is expressed by existing or proposed solid waste legis-
lation. Recycling reduces the need for raw materials such as metals, forests and oil, and
so reduces the impact on the environment. Recycling saves energy, reduces raw material
extraction and combats climate change. The vast majority of studies have found that recy-
cling our rubbish is better for the environment rather than incinerating or landfilling it
(Waste and Resources Action Programme 2006; Department for Environment, Food and
Rural Affairs 2006). Virgin materials need to be refined and processed to create products,
requiring vast amounts of energy and the use of polluting chemicals further causing the
destruction of habitats. For example, making 1 tonne of aluminium needs 4 tonnes of
chemicals and 8 tonnes of bauxite – the mineral ore, and it takes 95% less energy
make a recycled aluminium can than it does to make one from virgin materials.
Solid waste facilities and landfill fires emit air pollutants, when waste is not recycled,
including carbon monoxide (CO), carbon dioxide (CO
), hydrocarbons (HC), particulate
matter (PM), nitrogen oxides (NO
) and sulphur dioxide (SO
). Recycling can potentially
cut down these emissions. Most of the UK’s waste is currently buried in landfill sites,
which releases climate-change gases and pollute the soil and water. Additionally, the pro-
cess of recycling and composting, from kerbside collection to the sorting and reprocessing
Ó2014 Journal of Environmental Economics and Policy Ltd
Journal of Environmental Economics and Policy, 2014
Vol. 3, No. 2, 201–214,
Downloaded by [] at 06:31 03 November 2015
of recyclables, creates more jobs than incineration and landfill (Renner 1991; Gray, Jones,
and Percy 2004).
More generally, economists have long worried about accounting for pollution (see
Leontief 1970 for an early example). To value the environment, two popular methods
exist: revealed preference and stated preference. The first method relies on hedonic price
analysis or the travel cost approach while the stated preference approach, based on contin-
gent valuation surveys, directly elucidate the environmental value from question. Both
methods have been widely used in practice (Carson et al. 2003).
Instead, this paper relies on life satisfaction approach (LSA). The approach offers sev-
eral advantages over other valuation techniques in the case where a direct question about
the public good is not available. For example, the approach does not rely on housing mar-
kets being in equilibrium – an assumption underpinning the hedonic property pricing
method – nor does it ask individuals to directly value the public good or bad in question,
as is the case in contingent valuation. Instead, individuals are asked to evaluate their gen-
eral life satisfaction. This is perceived to be less cognitively demanding, as specific
knowledge of the good is not required and respondents are not asked to perform the unfa-
miliar task of placing a monetary value on a public good. This approach entails the inclu-
sion of non-market goods as explanatory variables within micro-econometric functions of
life satisfaction along with income and other covariates. (Frey, Luechinger, and Stutzer
2010). Therefore, the LSA approach does not require awareness of causal relationships,
but simply assumes that recycling leads to change in life satisfaction. LSA is thus closely
related to hedonic pricing but relies on life satisfaction rather than house price to evaluate
how individuals value their environment. More precisely, LSA does not rely on the ability
of the respondents to account and consider all the relevant consequences of a change in
the provision of a public good. This paper proposes an econometric model to understand
and describe how the recycling rates are associated to well-being. Unfortunately, because
of the recycling prices data unavailability, only the recycling rates are included in the
The contribution of this paper is the examination of the relationship between self-
reported well-being and recycling rates using micro-level panel data controlling for vari-
ous factors, as demographic, regional and meteorological. Second, two methods are
applied: probit-ordinary least squares (OLS) with fixed effects and the latent class gener-
alised ordered probit (OP) model are employed. There are several key advantages of
using these estimates. First, it is possible to control for the local authority district-specific,
time-invariant characteristics. Second, estimating a latent class OP model we model also
for slope heterogeneity. The estimates account for the total sample of British Household
Panel Survey (BHPS), as well as for non-movers and movers within Great Britain.
2. Literature review
There are numerous studies on happiness economics. There is the general belief that
data on subjective well-being are valid and can be informative (Di Tella, MacCulloch,
and Oswald 2003; Pischke 2011). Research studies on happiness have identified vari-
ous personal, demographic and socio-economic factors of happiness that explain
observed happiness patterns. Some of the most important personal and demographic
characteristics which affect happiness are age, sex, marital status, the size of the house-
hold and the education level. Economic conditions like income and unemployment
have also a strong impact on people’s subjective well-being (Clark and Oswald 1994;
Easterlin 2001).
202 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
The most relevant study to our research is by Welsch and Kohling (2010). More spe-
cifically, the authors used a sample of 23,623 individuals in 27 countries in the time
period 1994–1999 using many of the variables used in our analysis, which are described
in the next section. They found a significant positive and linear relationship between recy-
cling and life satisfaction. However, in this study a much larger sample is examined using
only data for Great Britain, as well as we account for slope heterogeneity. Additionally,
in this study a panel data is used which allows us to identify the model from changes in
the pollution level using it as an instrument within individuals rather than between indi-
viduals. This reduces the possible endogeneity bias in the estimates since unobservable
characteristics of the neighbourhood that may be correlated with pollution, recycling rates
and life satisfaction are eliminated in a fixed-effect model.
Shen and Saijo (2007) examined the individual environmental concerns about recy-
cling and environmental quality in Shanghai based on a field survey conducted in
November 2006. They found that high-income and high-education classes are signifi-
cantly more concerned about recycling. Therefore, higher level of environmental quality
and recycling could be associated with higher levels of self-reported well-being. Also,
young people are more concerned with waste and recycling issues and they are willing to
sacrifice more life convenience for additional environmental quality including waste
management and recycling issues. Schubeler, Wehrle, and Christen (1996) present a con-
ceptual framework for waste management and recycling suggesting that the interaction
between waste handling procedures and public health conditions is influenced by climatic
conditions and characteristics of local, natural and ecological systems. Also, environment
health conditions may also be indirectly affected through the pollution of ground and sur-
face water by leachates from disposal sites. Air pollution is often caused by open burning
at dumps, and foul odours and wind-blown litter are common. As health status and condi-
tions are used as determinants of happiness, a relationship between recycling and well-
being also might be presented.
3. Theoretical framework
3.1. Theoretical model
There are two serious failures that arise in the management of solid waste. The first
relates to the negative externalities in the individual decision-making over waste genera-
tion and disposal. When individuals decide on how much to consume and what to con-
sume, they might not take into account how much waste they produce. Because the
external costs of waste generation, such as air pollution, are ignored by individuals,
more waste is produced and disposed of than is socially optimal. The second serious fail-
ure relates to the ways in which waste collection services are typically financed. Usually,
individuals pay for waste disposal in lump sums through general taxes or flat payments
to local governments or private collectors. Hence, waste disposal costs are not fully
reflected in the prices households face at the margin. In addition, individuals still face
zero prices for additional waste produced; thus they tend to produce and dispose of
more waste than if they were to pay for the additional garbage according to its social
marginal cost.
Addressing the issue of municipal solid waste is an important policy objective and one
which is becoming increasingly challenging to address. On the one hand, while the aware-
ness of the external effects of waste generation is increasing, there is resistance by society
to the development of new landfills and incineration facilities. On the other hand, solid
waste generation has grown significantly over the last decades as a result of higher
Journal of Environmental Economics and Policy 203
Downloaded by [] at 06:31 03 November 2015
incomes, more intensive use of packaging materials and disposable goods, and increased
purchases of durable material goods.
Next, we present the theoretical model. Assuming that some individuals may wish to
limit the amount of waste generated and sent to landfills or incinerators, the utility func-
tion is
Zindicates the commodity produced using inputs X,Gis the amount of garbage for
disposal, which is a function of inputs Xand time spent for separating the recyclables, S
and is a function of labour spent recycling some portion of the refuse generated by inputs
Xand lis the amount of leisure consumed. The marginal utilities are assumed to be U
>0 and U
0. The last term is an inequality because garbage generation will impact
the utility of some people negatively while it will not affect others. Next, the use of
inputs Xgenerates trash Tand it is a function, T(X), where T
>0. Trash may be sepa-
rated into garbage disposal or recycling and the production of recyclables Ris a function
of the total time spent separating recyclables Sand the amount of inputs Xavailable for
The amount of garbage is total trash less the recyclables and it is defined as
We assume that the budget constraint is constituted by household’s full income con-
sisted of wage and non-wage income and it is
wH þV¼px þfGðS;XÞð4Þ
where wis the wage, Vis the non-wage income, Hindicates the total hours worked, pis the
price for Xand fis the unit cost of garbage disposal. The household’s time constraint is
where Ais the total time available. Substituting Equations (2) and (3) into utility function
(1) and the budget constraint (4), the model is formulated in such a way that the variables
of interest are S,Xand l. The optimisation problem becomes
L¼U½ZðXÞ;GðS;XÞ;lþλ1ðwH þVpx f½TðXÞPðS;XÞ þ λ2ðDHlSÞ
204 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
The first-order conditions are
@X¼UZZXþUGðTXRXÞλ1½ðpþfðTXRXÞ  0ð7Þ
where λ
and λ
denote the shadow values of income and time, respectively. Furthermore,
Kuhn–Tucker conditions are required because some consumers do not recycle. Equation
(7) shows the optimum input level of Xwhich is affected by the utility of the input and
the potential disutility of the garbage produced, in the case that U
<0. Equation (8)
shows the optimum choice for Swhich is the time spent in recyclable preparation for
inputs X. Finally, Equation (9) shows the optimum choice for leisure. More specifically,
at an interior solution the marginal utility of leisure is equated with the shadow value of
3.2. Granger causality
In this section also the Granger causality methodology test is presented. The main interest
here is to examine if an inverse causality between well-being and recycling rates is pres-
ent, which might cause endogeneity bias. A time-stationary vector autoregression (VAR)
model adapted to a panel context as in Holtz-Eakin, Newey, and Rosen (1988) of the fol-
lowing form is estimated:
HPijt ¼aþX
bjk rec ratejtkþX
gijk HPijtkþmiþljþutþvijt ð10Þ
Relation (10) examines if recycling rates cause happiness. It is common in Granger-
causality studies to test whether causation runs in both directions. So although the main
focus of this paper is on testing whether recycling rates cause happiness and if so, with
which sign, also the following equation is estimated:
rec ratejt ¼aþX
bjk rec ratejtkþX
gijk HPijtkþmiþljþutþuijt ð11Þ
Based on relation (11), the causality from happiness to recycling rates is explored. In
order to test for Granger-causality between well-being and recycling rates, it is necessary
that the two time series are stationary. Based on Akaike (AIC) and Schwarz (SC) informa-
tion criteria, as well as based on the statistical significance of the coefficients, the opti-
mum lag length for Equations (10) and (11) chosen is 1. Equations (10) and (11) are
estimated using system generalized method of moments (GMM) proposed by Blundell
and Bond (1998). From Table 1 it becomes clear that recycling rates with one lag is statis-
tically significant and cause happiness. On the other hand, happiness does not cause
Journal of Environmental Economics and Policy 205
Downloaded by [] at 06:31 03 November 2015
recycling. Moreover, the Sargan test accepts the over-identifying restrictions in the GMM
estimations. In Table 1, the Granger causality test results are reported.
4. Econometric framework
4.1. Fixed-effect model
Happiness and life satisfaction can serve as an empirically valid and adequate approxima-
tion of individual welfare, in a way to evaluate directly the public goods. Additionally, by
measuring the marginal utility of public good or recycling rates in that case, the trade-off
ratio between income and the air pollution can be calculated. Therefore, the individual’s
reported happiness or life satisfaction levels can be treated as proxy utility data. However,
this seems to be a very strong assumption that is not supported. One way of limiting this
problem is to use panel data, so that the comparison is within individual over time, mak-
ing it more likely that it is meaningful. As such cross-sectional research is likely to be
biased, the following model of self-reported happiness for individual i, in area jat time t
is estimated:
HPijt ¼b0þb1rec ratejt þb2logðyitÞþb0zijt þgWjt þmiþljþutþljTþeijt ð12Þ
The dependent variable HP is the happiness response, subscript idenotes the individ-
ual, rec
is the recycling rate in linear in location jand in time t, respectively, log(y
denotes the logarithm of household income and zis a vector of household and demo-
graphic factors, as discussed in the next section. Wis a vector of meteorological variables,
as average, maximum and minimum temperature, wind speed and precipitation, in loca-
tion jand in time t. Wind direction could be useful; however, because of the data unavail-
ability it is not used in the study. Set m
denotes the individual fixed effect, l
is a location
(local authority) fixed effect, u
is a time-specific vector of indicators for the day and
month the interview took place and the survey wave, while l
Tis a set of area-specific
time trends. Finally, e
expresses the error term which we assume to be iid. Standard
errors are clustered at the local authority level. To limit endogeneity issue the population
of interest is limited to non-movers. Focusing on non-movers also allow us to capture
unobservable characteristics of the neighbourhood that may be correlated with pollution
and happiness that are fixed over time. Non-mover status is to be preferred, since this
indicates whether the individual has moved in comparison with its location at the last
wave (Taylor et al. 2010). In addition, by examining separately the non-movers the endo-
geneity issue is limited, since the decision to move may well be correlated to
Table 1. Granger causality test between well-being and recycling rates using GMM.
DV: happiness DV: recycling rates
Constant 0.8947 (0.0297)
0.9241 (0.1833)
Happiness with one lag 0.3768 (0.0058)
0.0744 (0.4333)
Recycling rates with one lag 0.0019 (0.0007)
0.6359 (0.0081)
Sargan test 2.145 (0.888) 2.841 (0.519)
Wald chi-square 11,570.92 [0.000] 18,347.26 [0.000]
No. obs 61,872 61,860
Note: Standard errors between brackets; p-values between square brackets;
denote significance at 1%
and 5% levels, respectively.
206 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
environmental quality including recycling. Furthermore, it is important to distinguish the
analysis into movers and non-movers because both groups may experience very different
dynamics regarding unemployment, wage earnings and quality of life, including school
among other factors.
In its current form, the model cannot be estimated by OP or logit using fixed effects.
Therefore, there are two options, either by estimating the model considering the depen-
dent variable as continuous or converting the dependent ordinal variable in continuous
variable assigning z-scores. This procedure was introduced by Van Praag and Ferrer-i-
Carbonell (2004). To compute probit-OLS, the categorical dependent variable is rescaled
by deriving Z-values of the standard normal distribution that correspond to cumulative
frequencies of the original categories. More specifically, the probit-OLS uses a transfor-
mation such that the new dependent variable takes the conditional mean – given the origi-
nal ordinal rating – of a standardised normally distributed continuous variable, calculated
based on the frequencies of the ordinal ratings in the sample (see Cornelissen 2006, for an
example). The advantages of this are that it is quicker to compute, as well as there is the
possibility of applying panel data methods, such as individual fixed effects. Although sat-
isfaction and happiness scores are collected on an ordinal scale, assuming cardinality of
satisfaction scores makes little difference to the results of regression analyses. Neverthe-
less, this study uses the probit-OLS to compare the results derived from OLS; however,
the results are not presented as the same. The reason why this framework is employed is
because it allows for fixed effects, while the OP model does not. In addition, these
estimates are used as robustness check to the traditional OP estimates. Van Praag and
Ferrer-i-Carbonell (2004,2006) show both heuristically and in several applications that
probit-OLS is virtually identical to the traditional OP analysis. Generally, both OLS and
probit-OLS have been compared with the ordered models and no difference has been
found among them (Van Praag and Ferrer-i-Carbonell 2006; Luechinger 2009,2010;
Stevenson and Wolfers 2008; Wunder and Schwarze 2010). The calculation of the depen-
dent ordinal variable can be stated as
HPi;j;t¼EðZjm1<Z<m2Þ¼½ðm1Þðm2Þ=½Fðm2ÞFðm1Þ ð13Þ
where Zis a standard normal random variable, wis the standard normal probability den-
sity function and Fis the standard normal cumulative density function (see Van Praag
and Ferrer-i-Carbonell 2004 for more details).
4.2. Latent class generalised ordered probit
Using the fixed- or random-effect models described in the previous sections, correct for
intercept heterogeneity, but they do not account for slope heterogeneity. One step further
is to model for slope heterogeneity. Therefore, this approach is asking not only how
much ‘money buys happiness’, but also ‘for whom it buys the most happiness’. The
model endogenously divides the observations –in a probabilistic sense – into separate
classes, which differ by the parameters – slope and intercept – of the relation between
income and happiness (Clark et al. 2005). This model assumes that an agent ievaluates
her health status at time t. Let b
denotes her answer, which belonging to a ordered set of
labels J¼fj1;j2;...;jJg, where Jdenotes the labels for j¼1, 2,...,J. The OP model is
usually justified on the basis of an underlying latent variable, HP, in our case, which is
Journal of Environmental Economics and Policy 207
Downloaded by [] at 06:31 03 November 2015
linear in unknown parameters, function of a vector of observed characteristics zand its
relationship to certain boundary parameters m. We can, therefore, write for simplicity the
So, model (1) is related to the observed outcome HP as
HP ¼
jif mj1<HPmj;for 1 <j<J
Jif mJ1HP)
with, under the assumption of normality, associated probabilities (Maddala 1983)of
PrðHP ¼0jzÞ¼Fðmj¼0z0gÞ
PrðHP ¼jjzÞ¼Fðmjz0gÞFðmj1z0gÞ;for <j<J
PrðHP ¼JjzÞ¼1FðmJ1z0gÞ)
Formally, a latent variable c
is defined, which determines latent class membership.
This is assumed to be a function of a vector of observed characteristics xwith unknown
weight band a random disturbance term eas
The overall probability of an outcome j¼1, 2, ...,Jis simply the sum of those
respective classes and has the form:
PrðHP ¼jjx;zÞ¼Prðc¼1jxÞPrðHP ¼jjz;c¼1ÞþþPrðc¼JjxÞPrðHP ¼jjz;c¼JÞ
So, for example, for those belonging to class 1, we have
PrðHP ¼0jz;c¼1Þ¼Fðx0bÞ½Fðz0g1Þ
PrðHP ¼jjz;c¼1Þ¼Fðx0bÞ½Fðm1;jz0g1ÞFðm1;j1z0g1Þ;for <j<J
PrðHP ¼Jjz;c¼11Þ¼Fðx0bÞ½1Fðm1;J1z0g1Þ )
The log–likelihood function, for a random sample of i¼1, ...,Nindividual, can be
written as
hijln½Prðyi¼jjxi;ziÞ ¼ X
208 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
where the indicator function h
hij ¼(1 if individual ichooses outcome j
0 otherwise )ði¼1; :::; N;j¼0;1; :::; JÞð21Þ
In this context, the estimated parameters of relation (4) are individual and potentially
time-varying. Therefore, in this general model heterogeneity is twofold; first, because the
‘marginal utility’ of income and the baseline-intercept level of self-reported happiness
are individual-specific, and second, because individuals may use different labels to
express the same level of happiness. The second heterogeneity may reflect variations in
attitudes towards pleasure, happiness, health and pain. Additionally, this model restricts
the marginal probability effects by design, whether the income and recycling effects dif-
fer based on the person’s well-being class.
5. Data
We use the BHPS, an annual survey of each adult member of a nationally representative
sample which started in 1991. Based on the data availability for the recycling rates, the
period examined in the current study covers the years 1999–2009. The BHPS takes place
during the whole year, except June and July. The variables included in vector Xare demo-
graphic and household variables as household income, age, family size or household size,
labour force status, house tenure, health status, marital status, education level, whether the
respondent lives in rural or urban area, and local authority districts. The income of the last
month is used as it is found to be significant. Also, the latter is measured in thousands of
pounds and has been converted to 2009 British pounds using the consumer price index (CPI).
The survey contains a question about their general happiness. General happiness is an
ordinal variable measured on a 4-point scale and the specific phrasing of the question is
the following: ‘Have you recently been feeling reasonably happy, all things considered’.
The meteorological variables are the average, minimum and maximum temperature,
wind speed and precipitation. The recycling rates have been derived from the UK
National Statistics, while the weather data have been derived from Met Office and the
National Climatic Data Center. The aggregation level of recycling rates is household and
is calculated based on the household waste which includes household collection rounds,
other household collections such as bulky waste collections, waste deposited by house-
holders at household waste recycling centres and recycling points/bring banks. In Table 2,
the summary statistics for recycling rate and income are reported.
6. Empirical results
In Table 3, the probit-OLS with fixed effects are reported.
It should be noticed that the
sum of non-movers and movers within Britain is not equal to the total sample. The reason
Table 2. Summary statistics.
Mean Standard deviation Minimum Maximum
Income 2694.672 2159.329 0 86,703.29
Recycling rates 23.293 11.659 1 62
Journal of Environmental Economics and Policy 209
Downloaded by [] at 06:31 03 November 2015
is that additional classes of moving status are included, as moving from abroad which
classes are not useful for the analysis, because the main interest of exploring the long-
term recycling effects on well-being is the respondents who move across Britain.
More specifically, the association between self-reported well-being and recycling rates
is positive and significant. This can be explained by the fact that it takes less energy to pro-
cess recycled materials than to process virgin materials. For example, it takes a lot less
energy to recycle paper than to create new paper from trees. The energy from transporting
virgin materials from the source is also saved. Saving energy also has its own benefits like
decreasing pollution. This creates less stress on own health and consequently increases hap-
piness. In addition, by saving energy in industrial production through recycling, the green-
house gas emissions from factories and industrial plants are lessened and the use of fuels
that emit harmful gases during production is also minimised. Furthermore, by recycling,
the waste materials that are placed into landfills are reduced, emitting less air pollutants.
Regarding the other coefficients, we observe that the coefficients of age and age
squared are negative and positive, respectively. Age is commonly found to have a U-
shaped relation to happiness, with those in middle age having lower happiness than the
young and the old (Blanchflower and Oswald 2004). Furthermore, a significant negative
association between poor health, unemployed and household size with well-being is
. Additionally, the respondents who own the house and who are married present
a positive and significant coefficient. All these findings are consistent with other studies
(Clark and Oswald 1996; MacKerron and Mourato 2009). On the other hand, respondents
who have the highest academic degree present a positive with happiness; however, the
Table 3. Probit-OLS happiness regressions.
Movers within
Great Britain
Recycling rate 0.0018 (0.0008)
0.0021 (0.0010)
0.0006 (0.089)
Household income 0.0293 (0.0123)
0.0278 (0.0122)
0.0248 (0.0375)
Age 0.0123 (0.0045)
0.0138 (0.0048)
0.0176 (0.0172)
Age square 0.00014 (0.00007)
0.00015 (0.00007)
0.00021 (0.0008)
Average temperature 0.0025 (0.0013)
0.0028 (0.0014)
0.0055 (0.0063)
Minimum temperature 0.0005 (0.0011) 0.0006 (0.0011) 0.0119 (0.0095)
Maximum temperature 0.0024 (0.0011)
0.0028 (0.0013)
0.0095 (0.0109)
Wind speed 0.0013 (0.0016) 0.0010 (0.0016) 0.0022 (0.0016)
Precipitation 0.0052 (0.0026)
0.0050 (0.0024)
0.0299 (0.0168)
Household size 0.0232 (0.0111)
0.0215 (0.0101)
0.0762 (0.0428)
Job status (unemployed) 0.202 (0.0421)
0.2373 (0.0437)
0.2794 (0.3231)
Marital status (married) 0.2411 (0.0934)
0.2440 (0.0966)
0.9220 (0.7252)
Tenure (house owned) 0.0612 (0.0310)
0.0740 (0.0322)
0.0252 (0.0277)
Highest degree (university or
0.0270 (0.128) 0.0716 (0.0150) 0.334 (0.617)
Health status (poor) 0.0192 (0.0079)
0.0181 (0.0075)
0.0492 (0.193)
Rural area 0.532 (0.243)
0.523 (0.121)
0.505 (0.576)
No obs. 135,710 112,638 8856
R square 0.4173 0.4327 0.8370
Omitted variables test 3.056 [0.0875] 2.677 [0.1023] 1.887 [0.1311]
Heteroskedasticity test 3.66 [0.0596] 3.27 [0.0612] 2.14 [0.0745]
Autocorrelation test 6.798 [0.0388] 5.255 [0.0514] 2.593 [0.1095]
Note: Standard errors between brackets; p-values between square brackets;
denote significance at
1%, 5% and 10% levels, respectively; clustered standard errors on local authority districts
210 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
coefficients are insignificant. Finally, the respondents who live in rural area present a
strong and positive association with happiness.
Regarding the meteorological data, maximum temperature and wind speed present the
expected negative and positive signs, respectively; however, wind speed is insignificant.
The precipitation, average and minimum temperature present positive signs; nevertheless,
minimum temperature is insignificant. Levinson (2012) finds no effect of precipitation
and a positive – though declining – effect of temperature on life satisfaction, while
Barrington-Leigh (2008) reports that life satisfaction varies significantly with the amount
of recent cloud cover. Finally, Lucas and Lawless (2013) find little evidence of a relation-
ship between any of a large number of weather variables and life satisfaction.
Finally, in Table 4, the latent class generalised OP estimates are reported. Using con-
ventional fixed or random effects corrects for intercept heterogeneity. However, latent
class models allow the parameters of the unobserved (latent) individual utility function to
differ across individuals, i.e. slope heterogeneity (Tinbergen 1991; Clark et al. 2005).
From Table 4, it becomes clear that recycling rates have significant stronger effects in
class 3 (same as usual), than in other classes, while the effects become less important con-
cerning classes 1 (much less happy) and 2 (less happy). Additionally, the income effects
become stronger in class 1, while are declined consecutively in classes 2 and 3. The mem-
bership of class 1 is 2.852%, while the memberships for classes 2 and 3 are 14.85% and
67.38%, respectively. The results can be explained by the fact that the individuals who
have self-reported as being less happy (class 1) might be more interested on basic needs,
job status and income, which the latter has the strongest effects among all classes. In addi-
tion, the effects of the rest variables, as personal, demographic and meteorological, are
similar to those in Table 4; however, the highest degree significant and positive effects on
subjective well-being for the individuals belonging to classes 1 and 2.
Table 4. Latent class generalised ordered probit regressions.
Class 1
(much less happy)
Class 2
(less happy)
Class 3
(same as usual)
Recycling rate 0.0011 (0.0020) 0.0015 (0.0007)
0.0023 (0.0009)
Household income 0.0473 (0.0235)
0.0409 (0.0157)
0.0171 (0.0078)
Age 0.0165 (0067)
0.0127 (0.0045)
0.0132 (0.0044)
Age square 2.2e0.4 (6.7e0.5)
1.7e0.4 (4.6e0.5)
1.3e0.4 (6.4e0.5)
Average temperature 0.0048 (0.0017)
0.0037 (0.0018)
0.0021 (0.0009)
Minimum temperature 0.0034 (0.0044) 0.00085 (0.0025) 0.0016 (0.0024)
Maximum temperature 0.0051 (0.0022)
0.0012 (0.0006)
0.0025 (0.0011)
Wind speed 0.0046 (0.0022) 0.0011 (0.0034) 0.0041 (0.0031)
Precipitation 0.0143 (0.0063)
0.0035 (0.0014)
0.0039 (0.0016)
Household size 0.0264 (0.0115)
0.0194 (0.0087)
0.0187 (0.0091)
Unemployed 0.3925 (0.1021)
0.1034 (0.0477)
0.1930 (0.0659)
Marital status (married) 0.2524 (0.1197)
0.0962 (0.0419)
0.2065 (0.1037)
Tenure (house owned) 0.1098 (0.0506)
0.0663 (0.0312)
0.0593 (0.0246)
Highest degree 0.3581 (0.1425)
0.2035 (0.1128)
0.1097 (0.1104)
Health status (poor) 0.0211 (0.0098)
0.0263 (0.0239) 0.0294 (0.0118)
Rural area 0.0448 (0.0675) 0.492 (0.234)
0.610 (0.257)
No obs. 135,710
LR chi-square 1640.82 [0.000]
Note: Standard errors between brackets; p-value between square brackets;
denote significance at 1%,
5% and 10% levels, respectively.
Journal of Environmental Economics and Policy 211
Downloaded by [] at 06:31 03 November 2015
Recycling can be the platform from which many people can be educated about their
environment and good citizenship. Councils should also promote and support waste mini-
misation schemes. These include the use of home composting, local bring banks and
household amenity sites, as well as opportunities to reduce waste and reuse items where
possible. For example, this could include preventing food waste and promoting furniture
reuse schemes, nappy washing services, local refillable schemes, and low packaging
shops and markets.
7. Conclusions
This study has used a set of panel micro-data on self-reported well-being from the British
Household Survey. LSA has been used to estimate the relationship between happiness
and air recycling rates.
LSA contains very useful information on individuals’ preferences. In addition, one
very strong point of life satisfaction is that it does not suffer from the contingent valuation
problem of large gaps between stated willingness to pay and willingness to accept. More-
over, the LSA can be very helpful in environmental and economic policy planning and
decisions. Future research suggests the study of alternative techniques, as dynamic panel
data regressions, as well as examination of recycling rates for specific materials, as paper,
aluminium and steel among others.
2. Based on Hausman and Breusch–Pagan Lagrange multiplier tests, fixed effects are preferred.
3. The results remain the same even when the health status is excluded from the regressions
accounting for the possibility of reverse causality. Therefore, based also on literature, we keep
this variable as it is useful to examine the effects of health status on happiness.
Barrington-Leigh, C.P. 2008. Weather as a Transient Influence on Survey Reported Satisfaction
with Life. Munich Personal RePEc Archive Working Paper 25736. Vancouver, BC: Department
of Economics, University of British Columbia.
Blanchflower, D.G., and A.J. Oswald. 2004. “Well-Being over Time in Britain and the USA.” Jour-
nal of Public Economics 88 (7–8): 1359–1386.
Blundell, R., and S. Bond. 1998. “Initial Conditions and Moment Restrictions in Dynamic Panel
Data Models.” Journal of Econometrics 87: 115–143.
Carson, R.T., R.C. Mitchell, W.M. Hanemann, R.J. Kopp, S. Presser, and P.A. Ruud. 2003.
“Contingent Valuation and Lost Passive Use: Damages from the Exxon Valdez Oil Spill.” Envi-
ronmental and Resource Economics 25: 257–286.
Clark, A.E., F. Etile, F. Postel-Viany, C. Senik, and K. Van der Straeten. 2005. “Heterogeneity in
Reported Well-Being: Evidence from Twelve European Countries.” The Economic Journal
115 (502): 118–132.
Clark, A.E., and A.J. Oswald. 1994. “Unhappiness and Unemployment.” The Economic Journal
104: 648–659.
Clark, A.E., and A.J. Oswald. 1996. “Satisfaction and Comparison Income.” Journal of Public
Economics 61: 359–381.
Cornelissen, T. 2006. Job Characteristics as Determinants of Job Satisfaction and Labour Mobility.
Discussion Paper No. 334. Hannover: Institute of Quantitative Economic Research, University
of Hannover.
212 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
Department for Environment, Food and Rural Affairs. 2006. Impact of Energy from Waste and
Recycling Policy on UK Greenhouse Gas Emissions, Final Report for Defra. Defra R&D
Project WRT 237. Oxford: Environmental Resources Management Limited.
Di Tella, R., R.J. MacCulloch, and A.J. Oswald. 2003. “The Macroeconomics of Happiness.” The
Review of Economics and Statistics 85: 809–827.
Easterlin, R.A. 2001. “Income and Happiness: Towards a Unified Theory.” The Economic Journal
111: 465–484.
Frey, B., S. Luechinger, and A. Stutzer. 2010. “The Life Satisfaction Approach to Environmental
Valuation.” Annual Review of Resource Economics 2: 139–160.
Gray, A., A. Jones, and A. Percy. 2004. Jobs from Recycling: Report on Stage II of the Research.
London: Local Economic Policy Unit (LEPU), South Bank University.
Holtz-Eakin, D., W. Newey, and H.S. Rosen. 1988. “Estimating Vector Autoregressions with Panel
Data.” Econometrica 56: 1371–1395.
Leontief, V. 1970. “Environmental Repercussions and the Economic Structure: An Input–Output
Approach.” Review of Economics and Statistics 52: 262–271.
Levinson, A. 2012. “Valuing Public Goods Using Happiness Data: The Case of Air Quality.”
Journal of Public Economics 96 (9–10): 869–880.
Lucas, R.E., and N. Lawless. 2013. “Does Life Seem Better on a Sunny Day? Examining the Asso-
ciation Between Daily Weather Conditions and Life Satisfaction Judgments.” Journal of
Personality and Social Psychology 104 (5): 872–884.
Luechinger, S. 2009. “Valuing Air Quality Using the Life Satisfaction Approach.” The Economic
Journal 119 (536): 482–515.
Luechinger, S. 2010. “Life Satisfaction and Transboundary Air Pollution.” Economics Letters
107 (1): 4–6.
MacKerron, G., and S. Mourato. 2009. “Life Satisfaction and Air Quality in London.” Ecological
Economics 68 (5): 1441–1453.
Maddala, G.S. 1983. Limited Dependent and Qualitative Variables in Econometrics. Cambridge:
Cambridge University Press.
Pischke, J.S. 2011. Money and Happiness: Evidence from the Industry Wage Structure. CEP
Discussion Paper No. 1051. London: Centre for Economic Performance, London School of
Economics and Political Science.
Renner, M. 1991. Jobs in a Sustainable Economy. Worldwatch Paper 104. Washington, DC: World-
watch Institute.
Schubeler, P., K. Wehrle, and J. Christen. 1996. Conceptual Framework for Municipal Solid Waste
Management in Low-Income Countries. Working Paper No. 9. UNDP/UNCHS (Habitat)/World
Bank/SDC Collaborative Programme on Municipal Solid Waste management in Low-Income
Countries. St Gallen: SKAT (Swiss Centre for Development Cooperation in Technology and
Shen, J., and T. Saijo. 2007. The Socioeconomic Determinants of Individual Environmental
Concern: Evidence from Shanghai Data. OSIPP Discussion Paper DP-2007-E-003. Osaka:
Osaka School of International Public Policy (OSIPP), Osaka University.
Stevenson, B., and J. Wolfers. 2008. “Economic Growth and Subjective Well-Being: Reassessing
the Easterlin Paradox.” Brookings Papers on Economic Activity 39 (1): 88–102.
Taylor, M.F., J. Brice, N. Buck, and E.P. Lane. 2010. British Household Panel Survey User
Manual Volume A: Introduction, Technical Report and Appendices. Colchester: University of
Tinbergen, J. 1991. “On the Measurement of Welfare.” Journal of Econometrics 50: 7–15.
Van Praag, B.M.B., and A. Ferrer-i-Carbonell. 2004. Happiness Quantified: A Satisfaction Calculus
Approach. Oxford: Oxford University Press.
Van Praag, B.M.B., and A. Ferrer-i-Carbonell. 2006. An Almost Integration-Free Approach to
Ordered Response Models. Tinbergen Institute Discussion Paper TI 2006-047/3. Amsterdam:
Tinbergen Institute.
Waste and Resources Action Programme. 2006. “Environmental Benefits of Recycling: An Interna-
tional Review of Life Cycle Comparisons for Key Materials in the UK Recycling Sector.”
Journal of Environmental Economics and Policy 213
Downloaded by [] at 06:31 03 November 2015
Welsch, H., and J. Kohling 2010. “Pro-environmental Behavior and Rational Consumer
Choice: Evidence from Surveys of Life Satisfaction.” Journal of Economic Psychology
(3): 405–420.
Wunder, C., and J. Schwarze. 2010. What (if Anything) Do Satisfaction Scores Tell us About the
Intertemporal Change in Living Conditions? SOEP Papers on Multidisciplinary Panel Data
Research. Berlin: DIW Berlin.
214 E. Giovanis
Downloaded by [] at 06:31 03 November 2015
... This evidence is closely related to the biophilia hypothesis understood as the innate emotional affiliation of human beings with other living organisms including the environment. Accordingly, humans having an innate affiliation with the ecosystem, are more likely to preserve it through actions such as recycling (Giovanis, 2014;Welsch and Kiihling, 2010;Hernandez et al., 2009;Yang et al., 2021), water conservation (Suarez-Varela et al., 2016;Welsch and Kuhling, 2011), energy conservation (Kaida and Kaida, 2016;Corral-Verdugo et al., 2011a;Muranko et al., 2019) or altruistic PEBs, such as volunteering in environmental care or preservation activities (Meier and Stutzer, 2008), which in turn increases individuals' welfare. Thus, actions that help the environment can lead to positive welfare effects similar to the pro-social altruistic behavior when we help others (Dunn et al., 2008(Dunn et al., , 2014Corral-Verdugo et al., 2011b) or even recover individuals' mental health after a stressful event (Hartig et al., 2001;Corral-Verdugo, 2012). ...
... In this model, in addition to showing that the signs and significance of the coefficients of interest are robust, it is also observed that the probability of reporting higher SWB increases monotonically with the intensity of participation in PEBs-except for the "always" category for water conservation. Overall, these results are consistent with previous findings related to recycling (Giovanis, 2014;Welsch and Kuhling, 2011), water conservation (Suarez-Varela et al., 2016;Welsch and Kuhling, 2011) and energy conservation (Kaida and Kaida, 2016;Corral-Verdugo et al., 2011a). Then, our empirical findings support our Hypothesis 1. Furthermore, they are also in line with the hypothesis that more frequently engaging in PEB is strongly correlated with life satisfaction for PEBs that have higher costs for the individuals in terms of money rather than efforts (Schmitt et al., 2018). ...
... In contrast, individuals in class 2 increasingly derive utility when the frequency of performing energy-saving activities increases, but with respect to the two remaining PEBs, the category always" is not significant. Therefore, the latter empirical finding supports our Hypothesis 2. In general, our results are in line with the findings of Giovanis (2014) and Muranko et al. (2019) for recycling and energy-saving PEBs and contrary to the results of Binder et al. (2020), who find negative relationships between these PEBs. ...
This article estimates the individuals’ non-pecuniary benefit of engaging in pro– environmental behaviors (PEB) using a large sample from Ecuador. As a novelty, we estimate a model that allows incorporating both unobserved heterogeneity in preferences and controlling for the potential endogeneity of income through instrumental variables. Although both problems have been addressed separately in the literature, we show that considering both sources of bias allows finding more accurate and credible monetary values. Our results show that subjective monetary evaluations regarding PEBs are generally overestimated (the income coefficient is underestimated), but not including unobserved heterogeneity hides important patterns for an important group of the sample with completely different preferences.
... The current work uses LCA to identify clusters of people within society that report low wellbeing. LCA is increasingly used to deal with the challenges of heterogeneity and endogeneity by allowing the latent (unobserved) characteristic to partition the data into clusters united by combinations of observable characteristics (see e.g., Anand et al. 2011;Brown et al. 2014;Clark et al. 2005;Fernandez-Blanco et al. 2009;Giovanis, 2014). LCA splits respondents into homogenous groups (latent classes) such that individuals in the same latent class will have similar response patterns to the independent variables while individuals across latent classes will have different response patterns to each other. ...
... For a more detailed comparison of alternative clustering approaches, see Appendix 6.2. LCA has been applied in research concerning a wide range of outcomes such as self-reported consumer taste (Fernandez-Blanco et al. 2009), financial satisfaction across life stages (Brown et al. 2014), and the relationship between self-reported wellbeing and recycling in Britain (Giovanis 2014). ...
Full-text available
Policymakers are generally most concerned about improving the lives of the worst-off members of society. Identifying these people can be challenging. We take various measures of subjective wellbeing (SWB) as indicators of the how well people are doing in life and employ Latent Class Analysis to identify those with greatest propensity to be among the worst-off in a nationally representative sample of over 215,000 people in the United Kingdom. Our results have important implications for how best to analyse data on SWB and who to target when looking to improve the lives of those with the lowest SWB (The authors owe a massive debt of gratitude to the Office for National Statistics for their support throughout this research. We are particularly grateful to Dawn Snape and Eleanor Rees for their valuable comments on earlier drafts of this paper, to Salah Mehad for the thorough review of methodology, and to Vahe Nafilyan for advice on clustering analysis. We also thank the anonymous reviewers for the very helpful comments. Thank you all very much.).
... Although there is a general agreement among researchers on this association, empirical studies have produced conflicting results. For instance, Frijters et al. (2004) and Giovanis (2014) found a significant positive association between happiness and income, while Clark (2003) found a negative relationship and Shields and Wailoo (2002) found insignificant relationships. Some researchers, however, found that income is a crucial determinant of happiness to a certain point, after which, there is a weaker correlation (Layard 2005;Graham 2005). ...
... The regression results of Model 1 indicate a significant positive relationship between HPI and per capita real income, suggesting that higher income leads to increase in the HPI. This result supports the findings of Frijters et al. (2004), Katsaiti (2012) and Giovanis (2014). The results further indicate that the coefficient of squared per capita real income is significantly negative, suggesting a non-linear relationship between per capita real income and HPI. ...
Full-text available
: The World Happiness Report 2018 ranks 156 countries by their happiness levels, and revealed a link between happiness and obesity. Despite the importance of this link, few studies have analyzed this relationship. Moreover, it may be the case that the relationship between happiness and obesity is non-linear. The relationship between happiness and income, however, has been studied by several researchers, particularly after the publication of Easterlin (1974). In his famous paradox, Easterlin found that after reaching a certain level, the further increase of material wealth no longer promotes happiness. Here, we investigate whether there is a quadratic relationship between happiness & obesity and happiness & income, for a panel of EU countries for the period 2005-2016, using the system GMM method. The empirical results suggest an inverse U-shaped relationship between happiness & obesity and happiness & income, implying that as obesity (income), represented by body mass index, increases, happiness first increases then stabilizes and finally decreases. Hence, the existence of an inverted U-shaped relationship between happiness and income supports the validity of the Kuznets curve hypothesis. Some control variables were also included in the regressions in order to solve omitted variable bias problems. The results indicate that income inequality and unemployment have a significantly negative impact on happiness.
... Welsch criticizes the limited application of happiness data to environmental issues while he asserts that the studies linking happiness (life satisfaction) to environmental factors are on the bottom rung of the ladder (Welsch, 2009). Giovanis estimated the effect of recycling rate on British individuals' happiness by using both the fixed-effect model and ordered probit model (Giovanis, 2014). Luechinger estimated the effect of $% & concentration on life satisfaction (Luechinger, 2009, s. 498). ...
... Giovanis estimated the effect of recycling rate on British individuals' happiness by using both the fixed-effect model and ordered probit model. His ordered probit model presents that the self-reported well-being of the total sample increases 0.0018% when the recycling rate reduces one percentage at the significance level 0.05, whereas the other environmental indicator 'precipitation' also increases the well-being and wind speed is not significant (Giovanis, 2014). ...
Full-text available
The municipalities in Turkey have a great extent statutory tools on coping with the environmental issues. Solid waste management, environmental health, and environmental management are among these responsibilities written merely down in the Municipal Law. Taking the Metropolitan Municipality Law into consideration, it is explicit that metropolitan municipalities own greater extent duties and powers than the other municipalities. In addition to these and according to the same Law mayors also have a responsibility, which is also a power, that is to take proper measures for tranquillity, well-being, health and happiness of the inhabitants. The municipalities necessarily, thus, have to provide the life satisfaction of their inhabitants through their several facilities. This study aims to estimate the effect of environmental factors, which are in the scope of municipalities' responsibilities, on life satisfaction. In this regard, we use both the indicator values of well-being index for provinces and the environmental statistics released by the Turkish Statistical Institute (TurkStat) by the year 2015 and 2018, respectively. Secondary, we also incorporated the environmental expenditures of the municipalities announced by the Ministry of Treasury and Finance into the model employed in the study. We employed the ordered logistic regression to estimate the effect of independent environmental variables on our dependent variable "the life satisfaction". The main results of our estimation indicate that some environmental factors, such as number of municipalities served by a sewerage system, the rate of municipal population served by drinking water treatment plants, the rate of the forest area per km 2 and environmental.
... The dependent variable of this study is the life satisfaction that is an ordered variable measured in a Likert scale from 0 (not satisfied at all) to 10 (completely satisfied). The relevant regressors are chosen based on the happiness literature (Clark and Oswald 1996;Brereton et al. 2008;Luechinger 2009;Levinson 2012;Chevalier and Giovanis 2012;Ferreira et al. 2013;Giovanis 2014;Giovanis and Ozdamar 2016b;Ozdamar and Giovanis 2017), including individual, demographic and household variables, such as the household income 2 , gender, age, household size, health status, job status, house tenure, marital status, education level. Also, we add variables representing the municipalities and community typology indicating whether the area is urban, suburban, and peripheral suburban, rural, agricultural and industrial among others. ...
... Education seems to be an important determinant similar to other studies, but the remarkable finding is that there is a negative relationship between life satisfaction and education up to a point, where there is no difference in life satisfaction reported, between those who have a postgraduate qualification and those with no qualification. While previous studies found a positive relationship (Easterlin 2001(Easterlin , 2006Bruni and Porta 2005;Ferreira et al. 2013;Giovanis 2014), our estimates are consistent with other studies which found a negative relationship, especially in the developed nations (see for example Veenhoven 1996;Dockery 2003Dockery , 2010. According to previous research studies, education is associated with the persons' likely future outcomes, including better employment opportunities and higher income which may result to further wealth and well-being improvement. ...
Full-text available
One important determinant of well-being is the environmental quality. Many countries apply environmental regulations, reforms and policies for its improvement. However, the question is how the people value the environment, including the air quality. This study examines the association between air pollution and life satisfaction using the Swiss Household Panel survey over the years 2000–2013. We follow a Bayesian network (BN) strategy to estimate the causal effect of the income and air pollution on life satisfaction. We look at five main air pollutants: the ground-level ozone (O3), sulphur dioxide (SO2), nitrogen dioxide (NO2), carbon monoxide (CO) and particulate matter of 10 micrometres (PM10). Then, we calculate the individuals’ marginal willingness to pay (MWTP) of reducing air pollution that aims to improve their life satisfaction. Beside the BN model, we take advantage of the panel structure of our data and we follow two approaches as robustness check. This includes the adapted probit fixed effects and the generalised methods of moments system. Our findings show that O3 and PM10 present the highest MWTP values ranging between $8000 and $12,000, followed by the remained air pollutants with MWTP extending between $2000 and $6500. Applying the BNs, we find that the causal effect of income on life satisfaction is substantially increased. We also show the causal effects of air pollutants remain almost the same, leading to lower values of willingness to pay.
... Furthermore, women and those living in large households report lower levels of life satisfaction. These findings are consistent with previous studies (Easterlin 2006;Dolan, Peasgood and White 2008;Giovanis 2014;Ngoo, Tey and Tan 2015). Age presents a quadratic relationship implying that up to age of 46 the life satisfaction is reducing and then starts to increase after that age. ...
For years, countries in Europe and around the globe have implemented various migration policies targeting the migrants’ economic and political integration and social inclusion. However, little is known about the impact of migration policies on migrants’ participation in socio-cultural activities and their link with well-being. The first aim of this study is to explore the effect of the Migration Act of 2000 in Germany on participation in socio-cultural activities of first-generation migrants. These activities are operationalized by attendance to cinema and jazz or pop concerts, theatre or opera, practising artistic activities, and participation in voluntary work. The second aim is to explore the impact of socio-cultural activities on subjective well-being (SWB), measured by life satisfaction, and how is compared between first-generation immigrants and natives. The Migration Act of 2000 was extended in 2005 to provide permanent residence permits to high-skilled migrants, accompanied by language and cultural orientation courses. We will implement a Difference-in-Differences (DiD) methodology comparing the relationship between socio-cultural participation and SWB of first-generation immigrants and natives. The results show that while first-generation immigrants participate less frequently in the socio-cultural activities explored, they experience an increase in participation after the implementation of the 2000 Migration Act. Furthermore, migrants report lower levels of SWB than natives, but their life satisfaction significantly improves with socio-cultural participation. The findings of this study have implications for policymakers and researchers, such as income, education, and employment promote migrant integration. Providing employment opportunities and a permanent residence permit, cultural participation, and thus, the integration of migrants can be successfully achieved. While there is a long debate about the effectiveness of migration integration policies, to the best of our knowledge, this is the first study exploring the impact of the Migration Act of 2000 on migrants’ cultural participation.
... However, since we follow the IV approach, we prefer the OLS to make the estimates comparable. Previous studies have applied OLS concluding that the estimates are similar (Clark and Oswald, 1996;Ahn and García, 2004;Giovanis, 2014). Another option is the "Probit OLS (POLS)" approach developed by Van Praag and Ferrer-i-Carbonnell (2006), which presents very similar estimates with those derived by OLS. ...
Full-text available
There is an increasing concern on the quality of jobs and productivity witnessed in the flexible employment arrangements. The aim of this study is to examine the relationship between various employment arrangements and the workplace performance. Home-based working-teleworking, flexible timing and compressed hours are the main employment types examined using the Workplace Employee Relations Survey (WERS) in years 2004 and 2011. The workplace performance is measured by two outcomes- the financial performance and labour productivity. First, the determinants of those flexible employment types are explored. Second, the ordinary least squares (OLS) method is followed. Third, an instrumental variable (IV) approach is applied to account for plausible endogeneity and to estimate the causal effects. The findings reveal a significant and positive relationship between these types of flexible employment arrangements and the workplace performance. Education, age, wage, quality of relations between managers-employees, years of experience, the area of the market the workplace is operated and the competition are significant factors and are positively associated with the propensity of the flexible employment arrangements implementation. This can have various profound policy implications for employees, employers and the society overall, including family-work balance, coping with family demands, improving the firm performance, reducing traffic congestion and stress among others. It is the first study that explores the relationship between flexible employment types and workplace performance using an IV approach. This allows us to estimate the causal effects of flexible employment types and the possible associated social implications.
... The recent interest in the LSA as a complementary tool to infer the monetary value of environmental goods relies on its alleged advantages over the CVM. First, LSA advocates sustain that respondents have little trouble answering life satisfaction questions as they are not requested to place a monetary value on environmental conditions, but to evaluate their life satisfaction (Giovanis, 2014). This is probably a cognitively less demanding task than answering a CVM questionnaire (Rojas, 2008;Frey et al., 2009;Welsch and Kühling, 2009). ...
In light of the growing interest and potential of the Life Satisfaction Approach (LSA) to value environmental attributes, this research paper aims to estimate the monetary value of water supply infrastructure improvements using this novel approach. A theoretical framework compares this method with the widely used Contingent Valuation method (CVM), highlighting the pros and cons of both approaches. Results show that the LSA can generate meaningful values of the non‐market benefits derived from improving water supply infrastructure, based on the impact that such benefits can have on individual’s life satisfaction. In addition, these values are not different from those obtained from the CVM. Thus the LSA arise as a useful complement to the more traditional methods while expanding the array of techniques available for the economic valuation of environmental goods in a cost benefit framework.
While aggregate recycling rates in developed countries have plateaued in recent years, the contamination rate of recycling streams due to consumers incorrectly recycling items that cannot be recycled has grown rapidly. We propose that this problem may be partially due to persuasive messages, such as pro‐environmental labeling on bins, that encourage recycling, but may lack guidance on how to do so accurately. For example, a number of public garbage receptacles across the U.S. are labeled “Landfill” instead of “Trash,” encouraging recycling by making the negative impact of garbage more salient. However, this labeling may also lead consumers to incorrectly “recycle” items that cannot be recycled (i.e., overinclusive recycling). Two field studies suggest that pro‐environmental receptacle labeling can lead to overinclusive recycling, and a controlled experiment provides preliminary process evidence involving anticipated emotion from trashing versus recycling. Research opportunities and public policy implications for pro‐environmental messaging are discussed.
Shreds of evidence from some developed countries have shown that green space has a significant positive effect on happiness, while few related studies have been done in developing countries. Based on panel data from Chinese Family Panel Studies (CFPS) and the Chinese National Bureau of Statistics, this paper provides an empirical analysis of the impact of green space on Chinese urban residents’ happiness levels by using the ordered probit model. The results show that per capita greenness has a significant negative effect on urban residents’ happiness in China and the effect weakens and turns into positive as residents’ income levels increase. Meanwhile, the relationships can vary among people of different educational levels and from different areas.
Full-text available
Weather conditions have been shown to affect a broad range of thoughts, feelings, and behaviors. The current study examines whether these effects extend to life satisfaction judgments. We examine the association between daily weather conditions and life satisfaction in a representative sample of over 1 million Americans from all 50 states who were assessed (in a cross-sectional design) over a 5-year period. Most daily weather conditions were unrelated to life satisfaction judgments, and those effects that were significant reflect very small effects that were only detectable because of the extremely high power of these analyses. These results show that weather does not reliably affect judgments of life satisfaction. (PsycINFO Database Record (c) 2013 APA, all rights reserved).