<<< speeches  

"Happiness" is not enough
Samuel Brittan: Templeton Lecture Inst. of Economic Affairs 22/11/01

The Title

The title of this talk is "Happiness" is not Enough. It is important to emphasise that the word happiness is in quotation marks. These are not an ideal device to show what I mean, but I cannot think of anything better. The reason for the quotation marks is that I have no wish to put forward some puritanical, self-denying or ascetic alternative.

On the contrary, a stress on the importance of personal happiness is part of the Enlightenment tradition. We are in the middle of a worldwide war - I am afraid not only of ideas, but physical as well - between Enlightenment ideas and an enemy, which for want of anything better, I must call religious fundamentalism -- not only Moslem fundamentalism, but Christian and Jewish fundamentalism as well.

I hope that I do not have to argue the Enlightenment case in front of this audience. What I wish to question are statistical measures of happiness which have recently been put forward as a guide for public policy. Thus it is, I hope, a friendly discussion within the Enlightenment camp about alternative ways of stating objectives.

My talk has been sparked off by an excellent study of survey research into peoples perceptions of their own happiness by two Swiss authors, Bruno S. Frey and Alois Stutzer (Happiness and Economics, Princeton University Press, December 2001; published in the UK by John Wiley & Sons). It is a good enough book to be worth disagreeing with its central contention, despite my being grateful both for its summary of existing research and for the stimulating ideas it contains.


I have an uphill task. The dominant outlook of English language political economy and quite a lot of political philosophy as well is utilitarianism. Its founding father, Jeremy Bentham, defined it as follows:

By the principle of utility is meant that principle which approves or disapproves of every action whatsoever, according to the tendency which it appears to have to augment or diminish the happiness of the party whose interest is in question... if that party be the community in general, then the happiness of the community: if a particular individual, then the happiness of that individual... The interest of the community then is, what? - the sum of the interest of the several members who compose it.

I hope you will notice the heavy emphasis on the individual and Bentham's disavowal of the idea of a collective consciousness. or of a state or nation being more important than the people who comprise it. I would count myself as qualified utilitarian: the nature of the qualifications will be touched on in the remarks that follow. .

After Bentham's time economists became more and more worried by the difficulties of assessing and measuring happiness. Instead they shifted towards interpreting utility in terms of the opportunity to satisfy desires - in technical terms being on the highest possible indifference curve. These desires are supposed to be revealed by observing peoples act or choices in the market place and elsewhere, e.g. in political voting.

This shift of emphasis was originally the work of economists looking for a way of making utilitarianism operational. I have in earlier work suggested that this interpretation is not merely a fall back due to measurement problems, but actually a preferable objective to which I have given the name Choice Utilitarianism (The Role and Limits of Government, 1983, chapter 2 and Capitalism with a Human Face, 1996, chapter 3,) It is preferable because it attaches a high value to individual choice and does not seek to peer into men's souls.

GDP Fallacies

It is a common fallacy to interpret either variety of utilitarianism in terms of maximising GDP or GDP per head. There are many components of utility which are excluded from conventional GDP measures. Well known examples include the value of work undertaken in the home, leisure and environmental harm such as pollution and - less discussed - environmental goods such as the value of a well kept front garden to those who just pass it by.

Attempts have been made to construct more comprehensive human development indices which give weight to factors such as literacy, access to clean water, life expectancy and so on. One problem is that the further you go in taking these things into account the more the index reflects the subjective values of those who draw it up. The placing of a country in a human development league can alter radically if it is done by an economist who is obsessed with inequality than if it is done done by one who is not. My own view is that adjustments to GDP should be confined to relatively straightforward matters, for instance working hours, and that other social indicators should be listed separately.

Happiness Research

But a much more fundamental challenge to choice utilitarianism has emerged. This is the claim that human happiness can be at least approximately measured, coupled with the value judgment that this should be the goal of public policy. In a sense it is Back to Bentham. This refurbished view is also associated with the use of questionnaires and opinion surveys, which have been common among empirical sociologists and students of politics, but which have late in the day been discovered by economists who have come upon them with all the thrill of the shock of the new.

These researchers are of course aware that individual happiness levels depend to a large extent on temperament, which in turn depends on both genetic and early environmental factors. They hope, reasonably enough, that these individual differences will cancel out if they take national or group averages.

It is of course possible to advocate basing policy on some direct measurement of happiness, without relying so much on questionnaire techniques. Who knows what direct physiological studies might one day be able to achieve? Even an untrained person can infer happiness among domestic cats by observing whether they are purring or not. But for the moment at least happiness advocacy is closely linked with these opinion studies.

Anything which extends the vision of empirical economics beyond those endless multiple regression analyses is to be welcomed. Too many of these regression analyses celebrate relatively small influences which are judged to be significant in the statistical sense. The variable being investigated is usually one of many; and an error in assessing any one of these other variables can throw out the whole study. I was first led to these criticisms by something quite important; namely studies by American econometricians which purported to show that the death penalty either was or was not a deterrent. Even the most enthusiastic of the death penalty advocates had to take into their equations the effects of many other variables such as unemployment, poverty rates and so on. An error in assessing any one of these throws out the whole value of the study and condemns people needlessly to the electric chair.

Survey Findings

But although I welcome the addition of survey studies to the all too limited toolbox of applied economics, I still have to question whether these happiness studies show what they are supposed to do. They are normally based on responses to questions such as Taking all things together, would you say that you are very happy, pretty happy or not too happy?

International comparisons of such studies produce some weird and wonderful results. Austria is rated as the least happy Western country with an index of 6.51 and Denmark the happiest with an index of 8.16. Indeed Nigeria comes out well above Austria. Countries such as Sweden and Finland with high suicide propensities also come towards the top of the European happiness league. Among the former Soviet republics the happiest league is Azerbaijan, with neighbouring Armenia coming at the bottom.

The most striking single result from these subjective happiness studies is that reported wellbeing does depend on income, but only up to a point. Despite the offbeat examples I have cited, on the whole people living in poor countries become happier with increasing average per capita income. Within a single developed country, better off people report themselves as being more satisfied with life than their poorer fellow citizens. Nevertheless the contribution of income flattens out at a level which Frey estimates as $10,000 per annum. Above that level, increases in average income per head contribute little to well being; even within a country such as the United States the reported happiness gains tend to vanish.

Indeed Frey has a chart showing that reported well being peaked in the USA in the late 1960s, despite a large increase in real income per head since then. He explains this as being as people adjusting their expectations to the rise in income. Some of us will recall that quite similar results were reported from the early studies of savings behaviour. The rich saved more than the poor. But when average national income increased the savings ratio did not rise at all. Indeed in the late 1990s US personal savings plunged dramatically with rising prosperity.

Misguided Policy Conclusions

Some economists have used these results as yet another debating point on an egalitarian agenda. If, beyond a certain level, people's well being reflects their relative position, why not redistribute income so that those who are in the middle or at the bottom no longer feel such intense relative deprivation. As far as I know, no one has demonstrated that the gains to the median citizen from soaking the rich would offset any loss that he or she might feel from the levelling up of the poor, which might lower his or her own self esteem.

But I am afraid that as a libertarian I would throw all these considerations out of the window. So called interdependent utilities have featured in economic writing well before people were asked questions about their happiness. Indeed it is the admission of jealousy and envy which weakens all forms of utilitarianism, whether traditional or new-fangled. One of the qualifications I would make to any form of utilitarianism is to rule out all the satisfactions deriving from the discomfiture of other people. In the words of the Nobel prize winning economist, John Harsanyi: All clearly antisocial preferences, such as sadism, envy resentment and malice should be excluded from the social utility function. (Contribution to A Sen and B Williams, ed., Utilitarianism and Beyond, 1982).

Other Results

It has long been known that unemployment makes many people unhappy; and Frey shows that this is not just due to loss of income but remains the case even if the responses are corrected for income effects. He also establishes that that actual inflation is disliked, and not merely unanticipated inflation as many economic writers suppose. My interpretation would be that most citizens cannot easily distinguish between a high and an unexpected inflation and find that any substantial rate of price increase contributes to the uncertainty and tension of life. I would suggest that some of their ignorance is shared even by economists and central bankers, as even they cannot distinguish between various kinds of inflation if they do not know what shocks -- such as oil price increases -- are likely to hit the economy or what changes in the policy regime governments might introduce.

Frey even attempts to estimate a subjective trade-off between inflation and unemployment, suggesting that it would take an 8.5 percentage points reduction in the rate of inflation to compensate for a 5 percentage point increase in the rate of unemployment. This estimate is worth reporting but only just, as it nourishes the fallacy that such choices are available, except very fleetingly.

A new result reported by Frey is that federalism and direct democracy contribute to personal happiness. The idea is that people enjoy being able to determine policies directly by referenda and by having small scale government relatively independent of decisions taken in the national capital. It is hardly surprising that Swiss authors should come to such conclusions; and their results are based on a study of the different Cantons which do have differences in the level both of local autonomy and the frequency and scope of referenda..

Indeed they report that the inhabitants of Basel (Landschaft) are 11 percentage points happier on average than inhabitants of Geneva where direct democracy is at its lowest. The authors discuss another interpretation: Living in a French speaking canton means significantly lower happiness, whereas living in the Italian speaking canton Tichino means significantly higher reported subjective well being than living in a German speaking canton.

More seriously, Frey refreshingly differs from many of the English and Eastern seaboard American economists in wanting to introduce corrective policies at the constitutional level rather than as immediate government measures. This might mean a focus, not on direct fiscal measures to reduce income disparities, but on a written constitution and voting rules other tan simple pluralities. This would entail, in my, view making it difficult both to dismantle the welfare state and to attempt to tax the rich until the pips squeak.

A Behaviourist Interpretation

There is a different way of looking at happiness which derives from a book by the English philosopher Gilbert Ryle, The Concept of Mind, published as long ago as 1949 (Penguin Classics edition with introduction by Daniel C. Dennet, 2000). The tendency of that book is to reinterpret mental phenomena, which are usually discussed as subjective feelings open only to introspection, in terms of observable behaviour. Ryle does not deal with happiness as such, but he does discuss enjoyment and pleasure, which come close to it.

He gives an example of person who is so absorbed in some activity, such as golf or argument, that he is reluctant to stop or even think of anything else. He is taking pleasure in or enjoying doing what he is doing, though he is in no degree convulsed or beside himself and though he is not, therefore, experiencing any particular feelings. Another of his examples is of a person who enjoys digging in his garden. This is not to say that he has been both digging and doing or experiencing something else, as a concomitant or effect of the digging; it is to say he dug with his whole heart in his task, i.e. that he has dug, is wanting to dig and not wanting to do anything else. His digging was his pleasure, and not a vehicle of his pleasure. There are not two activities going on side-by-side:, digging and an internal feeling of satisfaction.

Many philosophers do not believe that Ryle completely succeeded in banishing the subjective element. If one person who goes through all the motions of enjoying a game of golf but says he is in pain, then we can surely say he is less happy than another person who is not. And even if our subject is not in pain, he might be a grumpy individual who might believe that he would much rather have been a brain surgeon and is playing golf as a compensation.

Nevertheless the first clue to how happy people are is to look at how they behave and regard their answers to direct questions about their feelings as no more than useful supplement. The reported differences among happiness levels between countries may well reflect national cultures or even the exact meanings attached to happiness and related concepts when translated from one language into another.

Brave New World

If we take the happiness researchers seriously, it will not be enough to make a few mildly reformist suggestions, such as a Swiss-type constitution or fiscal transfers to reduce income disparities. We will have to go all the way towards Brave New World. This was, of course, a novel by Aldous Huxley, published as long ago as 1932. His New World differs from Orwell's 1984 because people are not made to conform by terror or fear of Big Brother. They are conditioned to do so by a selective breeding system applied when they are hatched in incubators. The population is divided into various categories ranging from the alphas, who give the directions, to the betas, gammas, deltas and epsilons - who do the worlds dirty work. And just to top up the effects of genetic engineering, there is a drug, soma, to be taken at any sign of waning happiness.

This brave new world is presented as a nightmare or dystopia. To understand why you will have to read or re-read the novel. No plot summary can be a substitute. Many of the initial readers were appalled by the description in the first few chapters of how children were conditioned to reject any books that might tell them of a historical past in which they would have had more control of their own destinies. Looked at from where we are now or even were in 1932, the picture is indeed a horrible one.

But we need to be careful. The inhabitants of Brave New World do not feel as we feel and are conditioned to enjoy their existence. The problem was previously aired by John Stuart Mill when he said it was better to be Socrates unhappy than a pig happy. I have always had a sneaking sympathy for the pig, so long as he can be bred to live as long as Socrates. But perhaps fortunately, we do not have that choice.

In all the 70 years that have passed since Huxley's novel appeared, genuine soma pills have not been invented. The hard drugs which some of the affluent young take to bring on a high seem in the end to produce squalor, suicides and other tragedies.

Huxley himself was ambivalent. In a new preface written in 1946, he criticised his own novel for offering only the choice between an insane life in Utopia and the life of a primitive in an [American] Indian village.. He rebuked himself for not having put forward a third possibility; simple human sanity. But he cannot keep up this optimistic vein. By the end of the preface he is talking in Orwellian terms of a choice between one supranational totalitarianism and a number of national militarist totalitarianisms.

He is also ambivalent about soma pills or their equivalent. Throughout his life he was looking for drugs which could bring on delightful feelings without an unfortunate kickback. He ended up by recommending mescaline for the purpose. But to the best of my knowledge there are still no happiness drugs devoid of unfortunate side or after-effects.

What was to me conclusive was that the conditioning in Brave New World was neither entirely painless nor completely successful. Young children are given electric shocks to induce distaste for flowers and books. And the action of the novel depends on there being people who are incompletely conditioned and who look back fondly to the world of Huxley's time and our own with all its imperfections.


What then is the alternative? It might come as an anti-climax; but I still think it is the choice utilitarian one of maximising the range of opportunities open to each individual. And I see the task of policy largely in negative terms; to remove obstacles to the exercise of individual choice rather than lots of fussy interventions on our behalf. But I must be frank and say that even such negative policies involve in my view a degree of income transfer towards the poor and less fortunate, which not everyone in this audience might welcome.

For about three quarters of the world's population a measure of success will still be real GDP per head, corrected for the worst absurdities, and supplemented by a few simple social indicators. But for the more affluent populations of North America and western Europe, economic growth in this sense is no longer a sensible objective of policy. It is much better that the growth rate should emerge from peoples own choices. So it would not be a disaster if after the recent traumatic events Americans adopted a quieter lifestyle with more emphasis on leisure and reflection, and working to live rather than living to work.

We should not throw out the baby with the bath water. GDP statistics will still remain useful for economic management and for looking at the way the national product is divided between different activities and different groups. Information on these matters should not take us along the road to serfdom. Indeed I have heard American friends observe that the social scientists who have been least tempted by collectivism are those who have the most detailed knowledge of the relevant facts and figures.

My conclusion is that the pursuit of happiness is and should remain a personal matter; and the people most likely to achieve this are not those who keep on asking themselves whether they are happy or unhappy, but who find worthwhile purposes and activities and concentrate on them. By all means make use of attitude surveys and similar devices; but let us do so first and foremost to satisfy our curiosity and not imagine that we have found the magic lodestar which has eluded thinkers of the past.

Samuel Brittan is a columnist at the Financial Times. His most recent books are Capitalism With A Human Face (Edward Elgar - 1995, Fontana - 1996) and Essays, Moral, Political and Economic (Edinburgh University Press 1998).

He is an Honorary Fellow of Jesus College, Cambridge; an Hon. Doctor of Letters (Heriot-Watt University, Edinburgh); an Hon. Doctor of the University of Essex. He has been visiting Professor at the Chicago Law School, a Visiting Fellow of Nuffield College, Oxford and an Honorary Professor of Politics at Warwick.

He has been awarded the George Orwell, Senior Harold Wincott and Ludwig Erhard prizes. He was a member of the Peacock Committee on the Finance of the BBC (1985-86).

He was knighted in 1993 for "services to economic journalism" and also became that year a Chevalier de la Legion d'Honneur.

  <<< speeches  
Site designed and managed by Andrew Heavens - aheavens@ftnetwork.com