Should happiness-maximization be the goal of government?

Grant Duncan, PhD

Massey University, Albany

Auckland, New Zealand

Email: L.G.Duncan@massey.ac.nz

This is the author’s manuscript of the published article:

Duncan, G. (2010) ‘Should happiness-maximization be the goal of government?’ Journal

of Happiness Studies, Volume 11, Number 2, 163-178.
DOI: 10.1007/s10902-008-9129-y
Should happiness-maximization be the goal of government?

Abstract

Recent social surveys of happiness (subjective well-being) have given a new

stimulus to utilitarian political theory by providing a statistically reliable measure
of the ‘happiness’ of individuals that can then be correlated with other variables.
One general finding is that greater happiness does not correlate strongly with
increased wealth, beyond modest levels, and this has led to calls for governments
to shift priorities away from economic growth and towards other social values.
This paper notes how the conclusions of this research help to address some of the
traditional objections to utilitarianism. The question of how happiness research
findings can be used to set happiness-maximization goals for public policy needs
careful examination, as the translation from research to policy is not always
straightforward. Some empirical and ethical objections to this ‘new utilitarianism’
are raised. The complicating factors of public expectations of, and trust in,
governments pose obstacles to any proposal that happiness research may lead to
changes in public policy and hence to ‘happier’ populations.

Introduction

Recent research on happiness and subjective well-being has prompted a re-examination

of the traditional utilitarian principle that the maximization of happiness should be
adopted by governments as an aim of law and public policy. This is an old idea that
reappears in various guises, for example:

Whatever the form or Constitution of Government may be, it ought to have no

other object than the general happiness.
Thomas Paine, The Rights of Man [1790] (1996, 164).

A measure of government … may be said to be conformable to or dictated by the

principle of utility, when … the tendency which it has to augment the happiness
of the community is greater than any which it has to diminish it.
Jeremy Bentham, An Introduction to the Principles of Morals and Legislation
[1789] (Warnock, 1962, 33).

The US Declaration of Independence of 1776 specifies ‘the pursuit of happiness’ as one

of the principal inalienable rights of all citizens. The 1942 Beveridge Report in the UK,
which laid the groundwork for Britain’s welfare state, referred to ‘the happiness of the
common man’ as the basic objective (Beveridge, 1942, 171). And, according to an
influential, though Whiggish, history of New Zealand:

2
 Social Security is … clearly an investment in the future personnel of industry as
well as in the happiness of the citizenry (Sinclair, 1991, 271).

For present purposes, I am mainly interested in that version of the happiness-

maximization principle that has re-emerged from research on happiness in the fields of
psychology, sociology and economics over the last three decades. To reiterate the
question, then, this article asks whether the maximization of happiness should be adopted
by governments as an aim of law and public policy. I emphasize the word should,
moreover, because the theory that I am considering here involves an ethical case for a re-
evaluation of social-political aims. Let me illustrate this with a few quotes from the recent
happiness literature.

People can and do experience lasting changes in their well-being as a result of life
events. Appropriate public policies can increase the average level of subjective
well-being, and it is conceivable that individuals, with greater knowledge of the
social mechanisms governing their lives, might themselves deliberately choose
courses of action that would permanently improve their happiness (Easterlin
2003a, italics added).

. . . [P]ublic policy should be about enhancing happiness or the welfare of people,

now and in the future (Ng and Ho 2006, 1).

. . . [T]he greatest happiness principle deserves a more prominent place in policy

making (Veenhoven 2004, 676).

Veenhoven (2004) gives one of the most carefully considered examples of this case for
the application of happiness research to public policy, based on what he calls a ‘new
utilitarianism’. He points out that a number of social factors over which governments
have some authority and control (such as respect for the rule of law and civil rights,
economic freedom, and tolerance of minorities) are positively correlated with popular
happiness levels. He concludes that the happiness of a society could therefore be raised
through the application of appropriate public policies, just as public health promotion
requires appropriate policy actions.

Diener and Seligman (2004) put forward a well-researched case in favour of nations
adopting systems of well-being indicators, partly to supplement existing economic
indicators. They do not argue that such social indicators should serve a prescriptive
function in the field of public policy. They see them as serving an evaluative role in
assessing the impact or success of national policies, and hence their conclusion is more
modest than Veenhoven’s. In each case, though, the conclusion is based on retrospective
data and on statistical correlations. We do not yet have prospective research that indicates
a causal relationship between the implementation of a new policy and an effect on
happiness.

The present discussion takes its impetus, then, from philosophical and social-research
literature. Apart from the well-known examples of the Kingdom of Bhutan’s policy of

3
Gross National Happiness and the reference in the US Declaration of Independence to
‘the pursuit of happiness’, governments around the world have so far not yet firmly
established happiness as an over-arching goal or evaluative criterion of public policy in
practical terms. So, the question posed here is a largely hypothetical one about the
possible political applications of happiness research. It considers some of the pragmatics
of public policy, and some of the ethical and empirical questions that have yet to be
addressed by any proposal to use happiness research to directly inform policy-making. As
such, this article does not seek to take a fixed stand one way or the other, but it considers
the ‘pros and cons’, poses some critical questions, and seeks to stimulate future theory
and research on this question.

So, what are some of the key conclusions of happiness research that are said to have some
relevance for public policy? Diener and Seligman (2004) provide a useful literature
review. Well-being surveys across nations have suggested that there are diminishing
returns to incomes above US$10,000 per annum. Further, once other health and social
factors that also correlate with both affluence and subjective well-being are statistically
controlled, “the effect of income on national well-being becomes nonsignificant” (Diener
and Seligman 2004, 5). In fact, aggregate happiness-survey scores remained static in the
USA and Japan, even after decades of strong economic growth; and the self-reported
happiness of individuals does not increase into middle-age, even when incomes and
wealth do. The communities with the highest subjective well-being or happiness are not,
therefore, those who are the wealthiest, but rather those who enjoy good health, effective
social and political institutions, high trust and social cohesion, and low corruption. More
pertinent to matters of government and policy, Diener and Seligman (2004) review
research reports that have found correlations between subjective well-being and
countries’ respect for human rights and freedoms, democratic institutions and political
stability.

Below, I introduce some reasons for interpreting such results with caution. But, at face
value, they appear to carry a message for policy-makers whose goals may have been
focused narrowly on economic growth as an objective and less concerned about health
and social issues and about good governance. But, does it also mean that happiness
surveys could become a tool for planning and evaluating public policy, and that
happiness itself should be a goal of public policy and/or a means of assessing its success?
To address this, we need a way of defining public policy, and of understanding its aims
and methods and how these are chosen and enacted. We also need to consider the bridge
between social research and policy actions, as this may not be as simple and unobstructed
as it may at first seem. Does research simply quantify pre-existing social problems and
test hypotheses about their causes for policy-makers then to digest the results and take
appropriate actions; or is the process of research somehow constructive of social
problems and thus actively shaping political concerns? But, before we do that, let us
consider how happiness researchers define happiness itself.

4
How do researchers define happiness?

Statistical data about the happiness of populations come from various surveys. This is
usually done in the form of questions about how happy or how satisfied with life the
respondent feels. It is a question about subjective well-being, but the results tend to be
fairly consistent and to correlate well with other measures of well-being factors. One
might say that ‘happiness’ is whatever the survey respondents think it is when asked, but
researchers often see a need to define the term. Authors in this field have provided
definitions that are quite diverse. Layard defines happiness as ‘feeling good – enjoying
life and wanting the feeling to be maintained’ (2005, 12); and Myers calls it ‘a high ratio
of positive to negative feelings’ (2004, 522). These definitions of happiness are the
simplest, based on positive feelings, and are the closest to the traditional Benthamite
meaning. Others take a slightly broader view, using the term ‘subjective well-being’ to
encompass both an affective evaluation of oneself at present, as well as a rational
evaluation of satisfaction with one’s life a whole. Frey and Stutzer point out that surveys
can measure such subjective well-being, and that these measures ‘serve as proxies for
“utility”’ (Frey and Stutzer 2002, 405).

Happiness, according to some psychologists, needs to be broken down into three strongly
correlated, but independent, factors: subjective well-being, life satisfaction, and absence
of depression or anxiety. Subjective well-being is the moods or feelings that people have
of joy or elation. Life satisfaction refers to qualities or circumstances of life, such as
personal wealth, family relationships, community participation, employment, goal
achievement etc., that may cause satisfaction or dissatisfaction. And, finally, the absence
of depression, anxiety, insecurity, etc. does not in itself constitute happiness or well-
being, but is nonetheless an important prerequisite (Argyle 2001).

There is another definition of happiness that deserves attention here, because it has been
developed in the context of a discussion that addresses the same question as the present
paper: the relevance of happiness to the aims of government. Veenhoven, who advocates
a ‘new utilitarianism’ and argues that happiness can be promoted by public policy,
defines happiness as ‘the overall enjoyment of your life as a whole’ (2004, 664). This
sounds more like Aristotelian eudaimonia than Benthamite utility. In fact, the quote
comes from a chapter in a volume on ‘positive psychology’ which explicitly seeks its
historical and philosophical roots in Aristotelian ideas about character and virtue, and
hence advocates a eudaimonic approach to ‘the good life’ – in which happiness (as good
feelings or satisfaction with life) would be only one dimension alongside ‘authenticity’,
‘continuous development’ and ‘the meaningful life’ (Jorgensen and Nafstad 2004).1

So, while each of the definitions of happiness may be sound enough in its own right, the
happiness literature in economics and psychology proposes diverse definitions deriving
from well-established, but rival, philosophical positions. Are we talking about subjective
feelings of pleasure, as balanced against pain – or are we using the term in reference to a
broader moral evaluation of ‘a good life’? Clearly, the happiness research field has
diverging definitions of its key term, but this may be seen as a healthy diversity of
opinion, rather than a fundamental flaw. And both the eudaimonism of the Aristotelian

5
tradition and the utilitarianism of Bentham and Mill have one significant thing in
common: they both hold up the potential for the realization of happiness/eudaimonia as
an ultimate goal, achievable in this worldly life. It is neither an abstract ideal, nor
something to be hoped for in an after-life. And both traditions have proposed that
happiness is relevant to the ethics and goals of the individual and of the state.

The matter would become even more complicated if one were to include non-western
concepts of happiness. Bhutan, which has adopted the policy of Gross National
Happiness, is officially a Buddhist state with a Buddhist concept of happiness (Mancall
2004). Buddhism advocates a detachment from worldliness as a means to overcome
suffering and to achieve enlightenment; and it is not really about individual happiness as
an immediate or transient experience. By contrast, western utilitarianism and
eudaimonism tend to be more oriented towards achieving life-satisfaction or pleasure in
the world as we know it empirically.

What is public policy?

So, while both eudaimonism and utilitarianism see happiness as an ultimate goal for the
individual, they both also extend this goal, by aggregation, to the political community as
a whole. In his Politics, Aristotle argues that the state exists to ensure the survival of its
members, but, more than that, it also exists to achieve ‘the good life’ (eudaimonia). In a
similar vein, Bentham says that the actions of governments can be evaluated in terms of
the aggregate effects they have on people’s happiness. Both assume that the goals of the
state can be inferred from the goals of its members, by aggregation.

It should be noted at this point that some other prominent thinkers have rejected
happiness as an ethical-political goal. Kant did not dismiss the importance of happiness,
but he did not see it as a suitable guide to moral reasoning, as it is too self-regarding. He
preferred instead a deontological approach as expressed in his categorical imperative.
More recently, Amartya Sen (1985) has extensively critiqued happiness and utilitarian
ethics in relation to development policy, and he advances a theory of capabilities instead.

Nonetheless, contemporary happiness researchers promote happiness as a political goal

based on correlations between social and institutional factors, over which governments
have some control or influence, and subjective well-being. The latter, moreover, has been
associated with other well-being factors, such as good health and longevity. Although
real-world governments do not have ‘Ministries of Happiness’, it would appear logical to
argue that, in as much as public policy can influence people’s happiness, governments
have (at least) an interest in maximizing it – if not a moral obligation to do so. By this
view, a low level of subjective well-being among some populations is as much a
‘problem’ for policy-makers as, say, poor public hygiene would be, even if it may require
less urgent attention. And thus hypothetically a suitable happiness objective needs to be
set, and policy programmes put in place to achieve it. The UK Prime Minister’s Strategy
Unit, for example, considered a paper on life-satisfaction and happiness research in 2002.
This discussion paper concluded that there is ‘a case for state intervention to boost life

6
satisfaction due mainly to evidence of direct impacts on life satisfaction of government
activities, together with strong evidence of the dependence of individuals’ wellbeing on
the actions of others’ (Donovan and Halpern 2002, 4).

Such a case relies upon a rational-instrumental model of public policy. It maintains that
scientific evidence of social conditions and problems is objective and that it leads
logically to proposals for remedies. The job of the policy analyst is to examine alternative
courses of action and to advise political decision-makers on the most economical or cost-
effective option. This assumes that implementation of such a remedy is a technical
process that has reasonably predictable outcomes, and that governments have the
authority and the means to carry out the desired programmes. When it comes to
something as inoffensive and universally desirable as happiness, moreover, how could
anyone rationally object? Social researchers and policy technocrats could rule, provided
we know it is making people happier.

Even if one cannot object to happiness per se, the instrumental model of public policy
that would see happiness research as translatable unproblematically into policy
programmes, and hence into social outcomes, is not the only way in which to consider the
role of policy. A second model of public policy could be called ‘interactionist’. This
considers the complex set of institutions and actors that form and perform policy as a
system, with subsystems and feedback processes. By this model, policy involves more
than just routinely-acting and stable hierarchies, but depends also on the complex and less
predictable effects of changeable networks and complex governance structures, some of
which occur outside the scope of state authority. For example, Rhodes (1996) talks about
‘governance as self-organizing networks’ operating across organizations in private
enterprise, civil society and central and local government. These networks challenge
rational, hierarchical models of government because they ‘become autonomous and resist
central guidance’ (p. 667). Policy objectives from the centre may be modified,
renegotiated, or even resisted, on such terrain. Such a model would place a notion like
happiness onto complex and contestable territory. So any liberal-democratic polity would
then need to consider how the ideal of happiness may be expressed within diverse
communities, be they secular or fundamentalist, mainstream or minority-group, etc.
Individual and cultural differences in values and policy priorities regarding how people
achieve happiness or ‘a good life’ would thus come to be relevant, and contestable,
matters in such a policy-making process.

A third approach to public policy could be called the ‘constructivist’ model. By this
model, the political process of ‘problem-formation’ is itself treated critically as
problematic (e.g. Dean and Hindess 1998). Hence, social conditions are not simply
objective states waiting passively to be measured by social researchers for the
information of policy-makers. Rather, the surveying of communities, especially
concerning something as subjective as happiness, actively constructs the awareness of the
phenomenon, and hence pre-forms the framework in which it may be treated as a
‘problem’ that legitimates political attention and governmental actions. The constructivist
approach is then also concerned with the ways in which policy instruments give effect to
programmes that normalize certain predictable or calculable patterns of behaviour in the

7
interests of, say, safety, financial prudence, healthy eating, or indeed happiness and well-
being. Such a perspective takes a critical view of the power relations and strategies that
enact programmes of problem-formation and normalization, treating them as interested
and contingent, rather than as neutral and necessary. By this approach, then, one may ask
how happiness research, once released into the domain of political actions, is actively
reconstructing our understanding of our own well-being as a ‘problem’ that requires
collective efforts and powers to address it; and also how the resultant debates and
practices may serve to redirect values and behaviours into normative patterns. For
example, a presumed governmental duty to maximize happiness could become implicitly
an individual duty to be happy – and hence this may imply a moral failure in those who
fall short of that, due to not following socially prescribed norms, and hence elicit a range
of new constraints and imperatives around the lives of citizens who are construed as
‘autonomous’, but whose ‘well-being’ is taken to be a collective concern.

My impression is that happiness researchers – like many other researchers in social

sciences and public health – tend to assume the rational-instrumental model of public
policy. The above discussion of alternative models does not invalidate the instrumental
model in practice, though, and I am not privileging any one of them. After all, one would
still expect well-formulated policies to have some rational grounding in evidence, to have
clear instrumental goals with authoritative backing, and to be performed by regulated
systems that seek to achieve those goals. But the alternative models do call us to exercise,
at the very least, some caution and modesty about understanding the complexity and
subtlety of policy processes and their origins and outcomes. They help us to understand
how sometimes well-intentioned political programmes can come to grief through public
misunderstanding and opposition, or how some critical reflexivity may be required in
order to fully appreciate the complex and political nature of apparently technical
processes of public policy.

To illustrate the complexity of the relationship between research findings and their
application to policy, consider the case of obesity as a social problem. Obesity is a major
concern for all affluent nations, it has well-known health risks, and it has worrying cost
implications for governments and health-care systems. It would appear to be a simple
matter to translate the mass of scientific evidence into effective public policy. After all,
unfettered food production and consumption in affluent societies is only resulting in
higher rates of obesity; so governments should take action where markets have failed. All
they have to do, it would seem, is to encourage people to change their diets and to get
more exercise. But the reality is much more complex than that. Policy-makers are
confronted with competing theories, complex and cross-cutting issues, huge vested
interests, ideological disagreements, a multitude of possible interventions, and frequent
resistance to perceived government ‘interference’. Despite the growing sophistication of
publicly available knowledge about obesity and its prevention, the rates of obesity seem
not to decline (Lang and Rayner 2007).

The skeptical note that this introduces is not meant to imply that either obesity-reduction
or happiness cannot or should not be a goal of public policy. It is only to suggest that the
translation from social research into effective political action is often fraught with

8
complexity and frustration. The next sections of this paper take up the theme mainly from
the point of view of utilitarian prescriptions for governments, rather than the Aristotelian
perspective – partly to simplify and to shorten the discussion, but also to recognize that
the utilitarian approach appears to be the more common in the happiness research
literature, especially among economists. The points arising from this discussion will show
that the case for happiness as a goal of good government does run into complications, in
the light of ethical and empirical issues.

The ensuing discussion will largely assume that public policy takes place in the
environment of a developed nation with a good standard of governance, although the
situation of developing or less stable countries will be taken into account when necessary.

Traditional objections to utilitarianism

The findings of happiness research represent an interesting development in utilitarian

thought. Now, utilitarianism has become a diverse set of doctrines, and has had to adapt
to numerous problems. One of the basic problems was that utility, being subjective and
having no basis for interpersonal comparison, could not be measured. So, the Benthamite
principle of utility (based on ‘the greatest happiness’) was, for the purposes of economics,
stripped of its subjectivism by Samuelson’s principle of ‘revealed preference’. This
meant that preferences are not regarded in subjective terms, but defined purely by the
behavioral choices that consumers make in a market. This led some policy-makers to the
idea that measures of aggregate economic output or income could be used as proxies for
the welfare of the people, as if greater consumption implicitly means greater happiness.
(That is, if we get more of what we prefer, we must be better off.) It appears that this
simple equation is no longer taken literally. United Nations guidelines on national
accounts, for example, point out that GDP is an economic indicator, should not be used to
stand for ‘well-being’, and is not supposed to represent directly the ‘utility’ that people
may (or may not) derive from the consumption of goods (United Nations Statistics
Division 2007). So, let us assume that utilitarianism, in the field of public policy, is not
necessarily about the reduction of public values to economic statistics – and hence that
happiness research may now have something useful to add to political deliberation and
policy decision-making.

Furthermore, while Bentham had a theory about calculating and maximizing happiness,
but lacked any interpersonally comparable data, contemporary happiness researchers
might claim that social surveys now supply data that is reasonably reliable and valid, thus
filling that gap. Veenhoven (2004), for example, argues that happiness surveys are valid
and reliable measures, and that the construct of happiness that the surveys are using is
interpersonally and cross-culturally comparable. Hence, this research literature may be
seen as giving new stimulus to utilitarianism. Although some, especially in the American
‘positive psychology’ movement, as pointed out above, appear to take a more
eudaimonistic approach to happiness, it is assumed here that contemporary happiness
research is predominantly utilitarian in its philosophical foundations. From a political
point of view, it leads us to the conclusion that the results of happiness research, based on

9
social surveys and multi-variate analyses, provide evidence of the personal, social and
economic conditions that are most likely to maximize happiness. This seems to result in a
form of rule-utilitarianism which says that, if governments adopt certain policies and
institutional practices, then people will be happier.

The precepts and prescriptions of contemporary happiness research have succeeded in

overcoming some of the traditional objections to the political applications of
utilitarianism. Relevant authors argue that happiness, as a mental state, is now measurable
and can be enhanced through appropriate actions (Layard 2005; Veenhoven 2004), and so
happiness surveys, results from which are now quite widespread and available globally,
provide the ability to observe and aggregate – and hence potentially to predict and even to
enhance – people’s happiness. Below, I place in italics some common objections to
utilitarianism, and immediately following is the kind of response that happiness
researchers can plausibly give in reply.

Objection 1: So long as the person is ‘happy’ – even if it requires the administration of

opiates – it does not matter if she is illiterate and malnourished.2

Happiness surveys find that greater happiness is correlated with good health, longevity
and higher levels of educational achievement, and so it supports the pursuit of a wider
range of well-being and development factors (see for example Easterlin 2003b), and the
spectre of a starving, yet ‘happy’, person is nothing but a ‘straw man’ in this argument.
Further, it is sensible to suppose that policies with a happiness goal may be quite different
in a country where illiteracy and malnourishment are widespread compared with a
country where unmet needs are not as ‘basic’ and which is instead concerned more with
social inclusion in the context of material prosperity. It is thus unfair to attribute to
happiness research literature a narrow subjectivism, devoid of interest in objective and
variable social conditions.

Objection 2: ‘The greatest happiness’ of the majority may be sought by means that
produce misery for a minority, or that at least involve some trade-off between people, in a
way that violates our understandings of human rights or moral principles.3

Happiness research findings may suggest that the greatest gains in aggregate happiness
can be made by attending to the well-being of society’s worst-off, especially those who
are poor (by progressive taxation and redistribution, and by employment security) and
those with mental illness (by improving mental health services) (Layard 2005).
Furthermore, democratic societies that respect universal human rights tend to have
happier people (Diener and Seligman 2004).4

Objection 3: ‘Happiness’ as an ultimate goal lacks adequate normative content, so a

person may justify, say, gambling, taking drugs or spending his savings on the grounds
that it makes him as happy as, or happier than, he would be if saving for retirement and
adopting a healthy lifestyle.5

10
Happiness research (for example Easterlin 2003b) does inquire into which happiness-
altering circumstances or behaviours avoid the effects of ‘adaptation’ (a short-term boost
in happiness, after which there is no long-term benefit), and hence can reach conclusions
that reflect upon more ‘sustainable’ gains in well-being.

But, while the conclusions about social and political values that may be derived from
happiness research can possibly withstand some of the standard objections to
utilitarianism, there are other obstacles in the way of drawing the conclusion that these
research results (regardless of how rigorous they may be) should be applied to public
policy. To make this case, we can explore some logical, philosophical problems, as well
as some potential empirical grounds that may lead us to ask whether public policy can
achieve any further gains in happiness (as reflected in surveys).

The proposition that happiness should become an aim of public policy relies on two main
types of assumption: an ethical assumption about political obligations, and an empirical
assumption about effectiveness. The ethical assumption is basically that, because we have
statistical analyses that show the conditions under which people are most likely to say
that they are happy, then we ought to use collective political means to maximize those
conditions. The empirical assumption is that, by having governments make the policy
reforms that appear to be needed, happiness-survey responses will improve. I will
examine each in turn.

Some ethical issues

The ethical claims for happiness as a political goal are often premised on the notion that
happiness is a universally understood and desired goal. For Richard Rorty (1999), for
example, the goal of giving all people ‘an equal chance of happiness’ is of prime
importance and has an a priori status for which there are no rational, empirical or
supernatural grounds. It is simply ‘a goal worth dying for’, about which there can be no
further argument. While it may be true that all languages contain words like the basic
idea of happiness as ‘good feelings’ (Wierzbicka 1999), however, the supposedly ‘self-
evident’ idea that happiness is something all humans universally want may only be self-
evident in the sense of being tautological. Happiness is defined as a good or desirable
state and hence, purely by definition, it is good, or all desire it. Hence, the statement
‘happiness is a universal good’ is quite circular, and no more informative than saying
‘pain always hurts’.

We may be able to get a little further, though, by stating the conditions which appear to
lead to a higher frequency of reports of high levels of happiness – just as it helps to know
what is most likely to cause pain, so that we can avoid such things. Now, a common
problem with any simplistic version of utilitarianism is the desire to leap from the
possession of such facts to the claim that we ought to do something. So, while it may be
perfectly ‘natural’ – an observable fact of human behaviour – that we take steps to make
our lives happier and to avoid pain, it should not automatically be assumed that this can
be used as a fundamental principle of ethics. Certainly, those happiness researchers who

11
use their empirical findings as premises for an ethical and political case may be guilty of
overlooking the extra logical steps needed. They generally do not stop to argue why the
achievement of higher aggregate scores on happiness surveys is good (given that it would
be circular simply to say that this result would be ‘good’ because it would signify that
people are happier). The fact that we want to be happy does not ineluctably lead to the
conclusion that happiness-maximization should be our ethical guide. Indeed, there are
respectable branches of moral philosophy that argue that it should not be, and instead
principles such as freedom, human rights, duties, virtue, capabilities or fairness may be
more relevant.

I will not try to settle this particular philosophical debate, but simply present it as a
problem worth considering. The grounds for happiness get marshier, however, when we
examine another common assumption in the happiness research literature. This is the
‘leap’ involved in proceeding from matters based upon personal subjective feelings to
matters of collective political decision-making. Happiness is subjective, and refers to
feelings of pleasure and satisfaction with life that can only be directly experienced by a
person privately. Hence, we have to be cautious about the scope of reference appropriate
to the word happiness. I can comment as the expert on my own happiness, and can share
empathetically in the happiness expressed by someone else who is close to me. So, it may
even be meaningful to assert ‘we are happy’, if the ‘we’ is inclusive of people close
enough to share each other’s feelings. But does the statistical aggregation of many
anonymous individuals’ estimation of their own private feelings of happiness amount to
something that is simply the sum of its parts? Or, is the well-being of the whole
something more than the net-sum ‘well-beings’ of its members?

It is clearly feasible to ask individuals to rate their own happiness numerically, and we
can aggregate and analyze those responses statistically. But, a large collection of people
is not the subject of happiness. Happiness, as an experience, or as something expressed
verbally or non-verbally, does not apply to more than one person at a time – or, at best, to
no more than a close group of persons. Any expression of personal happiness, moreover,
is mediated by factors unique to each context, such as a survey respondent’s
interpretation of the questioner’s motives for asking about happiness, or willingness to
disclose feelings to strangers. Aggregated scores from national surveys of happiness,
therefore, may not be the best way to estimate the well-being of Belgian or Japanese
society – because whole societies are not the grounds upon which the feeling of happiness
has content or practical meaning.6

Coming back to the ethical issues, then, a similar problem applies in moving from matters
of individual choice to matters of social or political choice. Even if I accept that
happiness can form a guide to my own ethical reasoning and decision-making, it is quite
another matter to generalize this to a category to which happiness may not properly apply
(that is, a whole society). As John Rawls put it, ‘[utilitarianism] improperly extends the
principle of choice for one person to choices facing society’ (Rawls 1999, 122), and the
same critique may be applicable to happiness survey findings if applied to collective
political decision-making. Even if we assumed that happiness is the ultimate personal
goal, happiness does not automatically become the ultimate political goal.

12
Communities do not feel happiness, only individuals feel it. So, while it may behove one
person to act in a certain way to see to his or her own happiness, the ‘happiness’ of a
whole collective or community of persons may be empty of content or reference, and
there may be no obvious pathway to improving happiness at that level of action. Two
different people may find satisfaction in customs or policies that are diametrically
opposed to one another. (Think about the legality of prostitution in one community versus
the ‘freedom’ to have child brides in another.) So aggregations of happiness scores may
elide underlying social and political conflicts, and hence happiness surveys may not
always be useful in guiding real-world political decisions about how the law should treat
various behaviours.

There is a further logical argument that advocates of happiness-maximization need to

make which bridges this gap between the individual level, at which happiness (as an
experience and as an ethical precept) may make sense, and the collective level at which
public policy actions occur, the level of whole communities and populations. The case
would need to be made that the goals of a nation can be viewed simply as the net-sum of
the happiness-maximization goals of the millions of individuals who inhabit it. In other
words, the whole is simply the sum of its parts, and the ethics of state action are to be
determined by the aggregated desires of individuals.

Even if we were persuaded that happiness surveys are relevant information for public
policy (which they may yet be), they would not overcome the basic ethical-political
problems of the policy-making process. No matter how much information is produced to
show under what social conditions people report greater happiness, these conclusions, on
their own, can have no morally compelling basis for the actions of governments. Indeed,
some would argue that happiness, while of paramount importance, is rightly the concern
of the individual (including loved ones), but not of government. Happiness is a private
value, not a public one; and many of the key factors associated with happiness (such as
social connectedness and health) may rely more on private choices than on public policy.
The neo-liberal suspicion of Big Government, or Nanny State, is likely to be scornful of
any well-intentioned ‘therapeutic’ effort to lend the powers of the state to the ‘problem’
of helping us to be happier (Furedi 2006). The obligations of government, it is often
argued, are merely to preserve the liberties of the individual. (This is a libertarian
standpoint and I will not presently defend it, however.) Further, though, there are other
principles for social decision-making, such as Rawls’s justice as fairness, or Sen’s theory
of capabilities, against which happiness-maximization has to justify itself as an equal, if
not superior, guide. So, even if we can claim that people in well-governed societies report
higher levels of happiness, we have not yet established that happiness is the aim of good
government.

In fact, happiness has been used by political ideologies ranging from utopian socialist to
social-democratic to libertarian, each with its own programme for political, social and
economic reforms, and with policy objectives that conflict with one another. Claiming
that happiness is the goal of good government and that we have scientific evidence about
what makes people happier, therefore, does not actually help us to solve the problem of

13
divergent social values and policy objectives. Happiness research can be used
persuasively and ideologically for diametrically opposed political aims and does not
provide us with a set of recommendations that over-ride political antagonisms and public
debate. One may claim that happiness is the goal of good government, as Tom Paine did,
but that leaves us back at square one when it comes to arguing about how best to
maximize happiness in terms of many practical policy choices. In short, evidence about
happiness does not trump political opinion-formation and public debates about core social
values (Duncan 2007). But it may at least inform our public debates with empirical data,
as Diener and Seligman (2004) suggest.

Some empirical questions

Having considered some ethical questions so far, what then of the empirical issues that
bear upon the idea that governments should act to maximize happiness?

The ‘Easterlin paradox’ is based on the empirical finding that growing markets have not
been associated with growth in aggregate happiness-survey results. This normally is
explained by ‘rising expectations’, and leads to calls for better public policy to restore the
pursuit of happiness (Easterlin 1974, 2001). The effects of the acquisition of new material
goods or enhanced income tend to be ‘self-limiting’, and do not lead to sustained gains in
subjective well-being. Good health and social connectedness, on the other hand, appear to
contribute more sustainably to people’s happiness than the pursuit of wealth and material
goods (Easterlin 2003b), so governments should do more about the former things, for the
sake of our happiness.

While this leads happiness researchers to claim that government should not focus too
heavily on economic growth, they have not asked about the potential ‘unhappiness’ that
might result from letting economies stagnate. Happiness research also tells us that high
unemployment and job insecurity cause lower levels of happiness (Frey and Stutzer
2002), so the consequences of not pursuing economic growth may also be undesirable, if
we accept the findings of the research.

Further, there may also be a ‘rising expectations’ effect around public services and
policies – such that ‘better’ or ‘more’ public services would also lead to static happiness-
survey findings, or even paradoxically to greater discontent. Indeed, it must not be
forgotten that, as developed countries got wealthier, their public and regulatory services
also improved – with massive benefits for health, education and public safety. In most
advanced industrial nations, the post-War decades saw rising incomes as well as rising
social well-being, in part brought about by improved public policy and technical
improvements in public services, such as public health systems – paid for by growing
economic output. This resulted in measurable well-being improvements in longevity,
infant mortality, educational participation and achievement, public hygiene and safety
regulations, etc. And yet, happiness surveys remained static over that period in many
countries, including the USA, for which we have data going back to 1946 (Veenhoven
2007). In other words, while Easterlin warned us that rising wealth did not correlate with

14
rising happiness, one must note that rising social well-being and expanding public
services (over the same decades) also did not correlate with rising happiness. Perhaps the
problem is that, as public services in health, education, etc. improved, public expectations
of what governments can and should deliver also rose accordingly, leading people to feel
no more satisfied or happy than they did before. And so the citizen fails to see him- or
herself as a participant in genuine social progress.

Richard Sennett noted this effect from his experience working with the New Labour
government in the UK. While he described that government as genuinely progressive, he
also observed that it had a tendency to take a short-term ‘consumerist’ approach to policy
– ever striving to produce new policies for the consumer-citizen and ‘abandoning them as
though they have no value once they exist’. The beneficiaries of the policies failed to
credit the politicians with the achievement of progress; the politicians blamed the public
for being ‘ungrateful’; and the Opposition accused the Government of being ‘out of
touch’ (Sennett 2006, 176). The danger may lie in regarding politics and public policy as
if we were consumers of ‘products’ or ‘brands’. Viewing the democratic policy-making
process in terms of whether a specific programme of action may ‘make me happier’ is
only going to make that consumerist mentality stronger.

There are some – but only a few – developed countries whose surveys have shown long-
term trends towards higher levels of happiness (if we can take the survey results literally
for the moment). Notable for this are Denmark and Italy. Meanwhile, Belgium had
gradually declining happiness scores, and Great Britain’s were static (Veenhoven 2007).
But, it is not yet clear what the governments of Denmark and Italy may have been doing,
and the governments of Belgium and Great Britain not doing, to have contributed to these
trends. In the case of Belgium, internal disunity caused by differences between its
constituent communities could well be a factor, and it is conceivable that ‘better policy’
could reduce that. National disunity also exists between northern and southern Italy (not
to mention corruption and political instability in Rome) and yet happiness ratings seem to
be on the rise there. Moreover, the Thatcher revolution made Britons neither more nor
less happy. In light of these paradoxes, it may be more parsimonious to hypothesise that
public policies have little at all to do with the trends in national survey results. It may be
the case that better public policy does not result in greater happiness – at least in
countries that already enjoy high levels of social well-being – as each improvement in the
public services or in the social environment is met with rising public expectations,
leaving people’s sense of satisfaction or well-being more or less static.

The best evidence supporting the notion that governance and public policy have an effect
on happiness comes from the negative cases, such as Russia, where political turmoil,
economic failure and corruption since the collapse of communism appear to be correlated
with low and declining surveyed happiness levels (Saris and Andreenkova 2001). In
general, moreover, there are lower happiness results in those developing countries that
suffer from economic hardship and political instability or human-rights violations
compared with developing countries that are the most prosperous and that are democratic
and stable. One would expect the individual happiness of Zimbabweans to rise after
political and economic stability is restored there. But one does not need happiness

15
surveys to justify democratic elections in that country, nor in any other. So, the most one
can say is that the failure effectively to govern a country may cause unhappiness. But
from this one cannot claim either that better public policies in already well-governed,
affluent countries will lead to greater happiness, or that ‘the greater happiness’ is a
necessary or sufficient justification for good governance.

The case in favour of making happiness an aim of government needs also to take account
of empirical trends in public attitudes towards government and political engagement in
democracies. Surveys of trust in government consistently declined over recent decades in
developed nations, though there is some uncertainty about how such public perceptions
are related to the actual performance of government agencies (Van de Walle and
Bouckaert 2003), and these results bear no correlation with well-being indicators
(Killerby 2005). Furthermore, a common trend in western democracies is towards lower
participation rates in elections, lower political-party membership rates, and less voter
loyalty (or more voter volatility) (Mair 2006). The affluent democracies of the west are
showing less trust in and commitment to political institutions and processes. This may be
an unfortunate trend, but it is empirically grounded, and it lends no support to the idea
that publics in the west will put their trust into governments and public officials in
matters of maximizing the happiness of all citizens, even though there is still a reasonable
expectation that relevant social and health services ought to be provided by government.

So, when thinking about the empirical effectiveness of any public programme aiming to
maximize happiness, we need to ask whether governments are in a position to use their
legitimate authority to this effect and to gain public consensus. Further, though, we would
need to be assured that governmental actions can positively influence happiness, as
measured by social surveys, and the evidence in favour of this proposition has yet to be
gathered. We do have evidence that some of the institutional and social or economic
factors over which governments do have some influence correlate with measured
happiness; but this does not yet prove that reforms to public policy, beyond what has
already been achieved in developed nations, will necessarily raise happiness levels. It
may be that the relationship between happiness and public perception of good
government is as paradoxical as that between economic growth and happiness. While
happiness research may supply us with many useful results, one should not leap to the
conclusion, as does Layard (2005), that there is now a ‘science’ of happiness that is
capable of showing us how to reform our personal and political choices. The application
of such findings to political aims and to public policy needs to take account of some of
the complexities of the relationship between what governments do, in terms of public
services, and how people feel about, or derive satisfaction from, their lives.

Conclusion

Happiness research, based principally on social surveys of subjective well-being, has

given a new impetus to utilitarian prescriptions for good government. One of its most
significant findings has been that, beyond poverty, the relationship between wealth and
happiness is, at best, weak – and further that aggregate self-reported happiness does not

16
increase, in developed countries, as economies grow and as people get wealthier. Instead,
a range of other variables correlate with changes in subjective well-being, including
physical and mental health and social belonging. These findings have led to calls for
happiness to be adopted as an aim (if not the aim) of government, and hence for public
policy to be guided by happiness research.

It may, however, be wrong to conclude that happiness research findings can be translated
directly into authoritative actions by governments, in the interests of the well-being of all
members of society, other than to provide feedback on the social programmes that
governments already deliver. Future debate about the politics of happiness would need to
take into account the complexity of policy networks and processes, and the diversity of
communities and their values, and to treat reflexively the questions of problematization
and normalization that may underlie the apparently benign and innocuous prescription
that happiness be the goal of good government.

While happiness research does overcome some of the traditional objections to utilitarian
political theory, the case for its direct application to public policy is challenged when
closely examined on ethical and empirical grounds. The utilitarian case for making
happiness a guiding collective value is not as ‘self-evident’ as it may at first seem; and
countervailing trends in rising public expectations of, and reducing trust in, governments
and public services may be adding complexity to the scene in which any programme for
maximizing happiness would have to take part.

Happiness research appears to support what Aristotle said: that you don’t have to be rich
to be happy and to live a satisfying life – though poverty is a major drawback. Growing
economic productivity and personal material wealth in developed countries appear not to
result in higher scores on happiness surveys. But, if this means that the growth of
economies in the developed world has not caused much progress towards a post-
Enlightenment objective of happiness-maximization, would governmental action succeed
instead? I have raised some doubts about that, but conclusive evidence has yet to be
produced. Public policy does not lend itself to controlled trials, but future evaluative
research on the impact of policy reforms on happiness-survey results may assist us.
Perhaps such prospective research on governmental effectiveness needs to be performed.
In the meantime, I conclude that happiness-maximization, as a social goal, is not an
obligation of government.

17
References

Argyle, M. (2001). The Psychology of happiness, 2nd edn. Hove: Routledge.

Beveridge, W. (1942). Social insurance and allied services. London: His Majesty’s
Stationery Office.

Dean, M. & Hindess, B. (Eds) (1998). Governing Australia: Studies in contemporary

rationalities of government. Cambridge: Cambridge University Press.

Donovan, N. & Halpern, D. (2002). Life-satisfaction: The state of knowledge and the
implications for governments. London: Strategy Unit.

Duncan, G. (2007). After happiness. Journal of Political Ideologies, 12(1), 85–108.

Diener, E. & Seligman, M.E.P. (2004). Beyond money: toward an economy of well-
being. Psychological Science in the Public Interest, 5(1), 1–31.

Easterlin, R.A. (1974). Does economic growth improve the human lot? Some empirical
evidence. In I.P.A. David & M.W. Reder (Eds), Nations and households in economic
growth: Essays in honor of Moses Abramovitz (pp. 89–125). New York: Academic Press.

Easterlin, R.A. (2001). Income and happiness: Towards a unified theory. The Economic
Journal, 111(473), 465–484.

Easterlin, R.A. (2003a). Towards a better theory of happiness. Conference abstract, The
Paradoxes of Happiness in Economics, 21–23 March 2003, University of Milano-
Bicocca.

Easterlin, R.A. (2003b). Explaining happiness. Proceedings of the National Academy of

Sciences of the United States of America, 100(19), 11176-11183.

Frey, B.S. & Stutzer, A. (2002). What can economists learn from happiness research?
Journal of Economic Literature, 40(2), 402–435.

Furedi, F. (2006). Why the ‘politics of happiness’ makes me mad. Spiked-Online, 23

May 2006. Available at
http://www.spiked-online.com/index.php?/site/article/311/ accessed 30 January 2007.

Jorgensen, I.S. & Nafstad, H.E. (2004). Positive psychology: historical, philosophical and
epistemological perspectives. In P.A. Linley & S. Joseph (Eds), Positive psychology in
practice (pp 15–34). Hoboken: John Wiley.

Killerby, P. (2005). ‘Trust me, I’m from the government’: the complex relationship
between trust in government and quality of governance. Social Policy Journal of New
Zealand, 25, 1–15.

18
Lang, T. & Rayner, G. (2007). Overcoming policy cacophony on obesity: an ecological
public health framework for policymakers. Obesity Reviews, 8 (suppl. 1), 165–181.

Layard, R. (2005). Happiness: Lessons from a new science. New York: Penguin.

Mair, P. (2006). Ruling the void? The hollowing of western democracy. New Left
Review, 42, 25–51.

Mancall, M. (2004). Gross national happiness and development: An essay. In K. Ura &
K. Galay (Eds), Gross National Happiness and development (pp. 1 – 54). Thimpu: Centre
for Bhutan Studies.

Myers, D.G. (2004). Psychology. New York: Worth.

Ng, Y.-K. & Ho, L.S. (2006). Introduction: happiness as the only ultimate objective of
public policy. In Y.-K. Ng & L.S. Ho (Eds), Happiness and public policy: Theory, case
studies and implications (pp. 1–16). Houndmills: Palgrave Macmillan.

Paine, T. (1996). Rights of man. Ware: Wordsworth.

Rawls, J. (1999). A theory of justice (revised edn). Cambridge MA: Belknap.

Rhodes, R. A. W. (1996) The new governance: Governing without government. Political

Studies, XLIV, 652—667.

Rorty, R. (1999). Philosophy and social hope. London: Penguin.

Saris, W.E. & Andreenkova, A. (2001). Following changes in living conditions and
happiness in post communist Russia: the Russet Panel. Journal of Happiness Studies,
2(2), 95–109.

Sen, A. (1985). Well-being, agency and freedom: the Dewey Lectures 1984. The Journal
of Philosophy, 82(4), 169–221.

Sennett, R. (2006). The culture of the new capitalism. New Haven: Yale University Press.

Sinclair, K. (1991). A history of New Zealand (4th Ed.) Auckland: Penguin.

United Nations Statistics Division (2007). 1993 System of National Accounts,

Introduction: The aggregates of the system as indicators of economic activity and
welfare. http://unstats.un.org/unsd/sna1993/tocLev8.asp?L1=1&L2=10. Accessed 7
September 2007.

19
Van de Walle, S. & Bouckaert, G. (2003). Public service performance and trust in
government: the problem of causality. International Journal of Public Administration,
26(8&9), 891–913.

Veenhoven, R. (2004). Happiness as a public policy aim: the greatest happiness principle.
In P.A. Linley & S. Joseph (Eds), Positive Psychology in Practice (pp. 658–678).
Hoboken: John Wiley.

Veenhoven, R. (2007). Trend average happiness in nations 1946–2006: how much people
like the life they live. World Database of Happiness, Trend Report 2007-1.
http://www.worlddatabaseofhappiness.eur.nl/hap_nat/nat_fp.htm.

Warnock, M. (Ed.) (1962). Utilitarianism. Glasgow: Collins.

Wierzbicka, A. (1999). Emotions across languages and cultures: Diversity and universals.
Cambridge: Cambridge University Press.

Acknowledgements

I am grateful to Nikil Mukerji for comments on an early draft, and to Professor Ruut
Veenhoven for supplying data.

1
The eudaimonic approach to happiness research is given more thorough discussion in a
recent issue of the Journal of Happiness Studies (issue 1, 2008).
2
‘If a starving wreck, ravished by famine, buffeted by disease, is made happy through
some mental conditioning (say, via the “opium” of religion), the person will be seen as
doing well on this mental-state perspective, but that would be quite scandalous’ (Sen
1985, 188).
3
A typical hypothetical example used here would be that of a surgeon who has five
patients whose lives depend on organ transplants. Should he dissect, and hence kill, one
person with healthy organs to save five others? To do so would appear to maximize
utility, but is also fundamentally wrong. A real-world example comes from the Kingdom
of Bhutan, which espouses a policy of Gross National Happiness, but in 1991 rescinded
citizenship from, and then expelled, its Nepalese-Hindi minority – about 100,000 people
– who continue to languish in refugee camps. At the time of writing, a group of these
refugees was resettling in my own community. Can the forced expulsion of an unwanted
minority be justified by the happiness of the majority?
4
This is perhaps being too generous to Diener and Seligman, however, as their review
article does contain the assertion that ‘… market democracies have much more well-
being than totalitarian dictatorships, so military expenditures that protect and extend

20
democracy will increase global well-being’ (2004, 24). Published not long after the
invasion of Iraq, this statement seems to be guilty of the kind of moral problem often
associated with utilitarianism.
5
This is a version of one of oldest objections to utilitarianism. It may be put somewhat
like this: Suppose playing tic-tac-toe gives me more pleasure than listening to Bach; then
we would have to suppose that the former is of higher moral worth, in my case. There is
something inherently wrong with this, and Mill struggled to get around the problem by
arguing that, in the estimation of anyone with sufficient experience of both forms of
pleasure, Bach would be the clear favourite. So, similarly, we might argue that anyone
with sufficient experience and knowledge of both will see that saving one’s money for
retirement is morally superior, because it will bring greater lasting happiness, to spending
one’s money in a casino.
6
J.S. Mill commits all of the fallacies described above. He argues that, because people
desire happiness, happiness is desirable. Since all this says is, ‘happiness is desirable
because people desire it’, the so-called ‘proof’ is tautological. Further, the only reason he
can give as to why people desire the ‘general happiness’ is that people desire their own
happiness. Proceeding thus from a tautology, which he assumes to be a ‘fact’, he surmises
that ‘happiness is a good’; and because it is a good to each individual person, it must
therefore be ‘a good to the aggregate of all persons’ (Mill, in Warnock 1962, 288–9). He
does not stop to ask whether ‘happiness’ has any meaning beyond the subjective
experience of one person at a time. However, he then proceeds to argue that ultimately
we desire only happiness, because all other desirable things are desired only as a means to
greater happiness. But the very premises of his case have to be dismissed. Furthermore,
happiness suits his purposes as a super-ordinate goal partly because it is free of content
and can thus be linked to any other desirable goals.

21