Chapter 1 of TRANSFORMING NEW ZEALAND. This is a draft. Comments welcome.
Keywords: Growth & Innovation;
The Purpose of Policy
Asking about the purpose of public policy is only a little less frustrating than asking about the purpose of life. But if we dont ask either question we may lock ourselves into paths which are patently wrong. There is a danger of New Zealand doing this, if public policy focusses exclusively on maximizing material output, as measured by GDP or one of its related indicators.
As Appendix I reports, GDP has a number of fundamental deficiencies. There has been a number of attempts to modify GDP for these weaknesses, but they miss the point. The connection between wellbeing, or any other plausible purpose of public policy, is tenuous. Basically material goods and services are a means to an end, not the end in itself. If we focus too much on them, we may miss the whole purpose of life. However, once this is recognised, it makes sense to increase material goods and services – perhaps indicated by GDP, but subject to caveats – in pursuit of the higher goal.
Throughout this chapter I am going to refer to NNP (net national product). This is GDP less depreciation (that is consumed capital), which converts the ‘gross’ to the ‘net’, and less the (net) income of foreigners in the country, which shifts the measure from a ‘domestic’ to a ‘national’ basis. The use of NNP (a.k.a. National Income) is to remind the debate not to get too obsessed with GDP.
Focusing on NNP (or GDP) sets the economic perspective of transforming New Zealand. Much of this book is about how to increase material output. Of all the intermediate targets of the good life, it is the one scientists (albeit economic scientists) know most about. But it is always in the context of the wider aims, of using the additional economic activity to transform New Zealand for a wider purpose.
Perhaps that wider purpose might be described as ‘wellbeing’. To begin with the little we know about
wellbeing and material output.
GDP and Wellbeing
The assumption which underlies much of the economic policy debate is that New Zealand should aim for higher real incomes and production, measured by GDP. The difficulties with GDP as a measure of material output are analysed in appendix I of this book. Here we accept the indicator at face value, as does much of economic policy, and ask the more fundamental question, what is the evidence that material output contributes to some higher objective of wellbeing?
Nineteenth century economists tended to focus on material output, assessing how well off someone was by the amount they could consume. Given the much higher levels of hardship of their day, that notion was perhaps justifiable. But it still dominates today’s economics. Pushed, an economist might say it is better to have more material goods and services than less, and if all the other things which make up human happiness are assumed as given, higher material consumption is better.
Economists – or at least the good ones – have been aware of the importance of the assumption, but until recently they were not able to evaluate it in any scientific way. Now that we can, we find that ‘more means better’ proves to be only marginally correct, and that it is not nearly as important – directly anyway – as economic policy assumes.
A major source of evidence comes from an official US survey which each year asks whether each of the 1500 odd respondents were ‘happy’. (The question is motivated by the US Declaration of Independence that among ‘certain unalienable rights’ is ‘the pursuit of happiness’.) Since the questions have been asked over a number of years (together with a whole range of other personal variables), it is possible to study the trends and associates of happiness. Of course, happiness is not quite the same as personal wellbeing (consider a person on a psychedelic high). But the survey raises serious doubts about the importance of material consumption (and hence NNP) as a good indicator of wellbeing.
For instance, US citizens are no happier today than they were in 1972, when the survey began (or even back to 1946 according to older surveys), despite major increases in material consumption and NNP per capita over the period. There are differences between groups is even more puzzling. Men report themselves less happy than women, but over the period they have been getting slightly more happier and women slightly less happy (so the gender difference is converging). This is an astonishing finding, given the social changes over the last three decades are generally thought to favour women. No one is sure why. A cynical possibility is that ‘women’s liberation’ is making women as unhappy as men. Another non-obvious outcome is that happiness changes over a lifetime, initially decreasing as one gets older, hitting the bottom in the thirties, and rising thereafter.
Even more surprising is that while real incomes have risen over the quarter of a century, average happiness has not. If we look at any point in time (‘cross-sectionally’, as the jargon says, rather than ‘longitudinally’) we find that there is a very slight improvement in happiness for those with higher incomes in the community. The effect is small, as is evident by a comparison of the advantage of extra income to the advantage of being married, for respondents report being married is a much happier state than being widowed, separated, divorced or never married. (That is an average, of course. As Jane Austin reminds us, ‘happiness in marriage is entirely a matter of chance.’) The economists calculate being average married generates the same additional happiness as an additional income of $US100,000 a year. These figures apply for the US, but some less comprehensive European data generally supports the broad conclusions. (The annual sum capitalises to at least a $1,000,000. Look at the (average) spouse and think ‘you’re a million dollar baby.’)
Economic variables over which the government has some influence, and which give a much better increase of happiness than income, is the more years of education the happier. It is also happier to have a job (for the same income). This last result is intriguing, for it suggests that work is valuable in itself, and that job creation may generate greater happiness, even if that reduces average incomes. I would not jump to the conclusion that make-work schemes or low paid jobs are necessarily a good thing. Other studies suggest that work has to be seen as socially valuable (and adequate remuneration is a signal that it is).
More recently the World Values Study asked across many different countries ‘are you happy’ and ‘are you satisfied with life’. (There is a translation problem. Is the word ‘satisfied’ in English equivalent to ‘satisfait’ (French), ‘sodisfatto’ (Italian), or ‘zufrieden’ (German)?. However the Swiss response is substantially higher than their counterparts speaking the same languages in France, Germany and Italy, which suggests the study is measuring behavioural, and not language, differences.)
In 1998 (not a year of outstanding economic performance) 95 percent of New Zealanders said they were ‘happy’, which put them second equal in the world with Switzerland and Sweden, just behind Iceland. Some 84 percent said they were satisfied with their life. That is 14th in the world –– in the middle of the OECD. Apparently there are a lot of New Zealanders who are happy, but striving for better.
That New Zealand ranks higher on these measures than it does on per capita NNP is no surprise given the deficiencies of aggregate material product as a measure of wellbeing. There is not a very good correlation among the rich OECD countries: The responses of New Zealand and the US are much the same even though US per capita NNP is about 40 percent higher than New Zealand’s. There is a relationship between material production and happiness for those countries whose per capita NNP is less than 70 percent of New Zealand’s (perhaps justifying the nineteenth century focus on material prosperity as a measure of wellbeing). In poor countries economic growth increases happiness and life satisfaction. As a country gets richer, growth does not.
Among the factors which depress the responses are being in a Communist nation, although (sadly) happiness and satisfaction deteriorated when they threw off their Communist shackles. While there does not seem to be any connection among high income countries between wellbeing and measures of civil liberties and political rights, poorer countries with democratic institutions seem to be happier. Regrettably the studies have not yet investigated the extent to which high employment and low unemployment affect happiness and life satisfaction. An eyeball over the data suggests there may be a statistically significant correlation, although the effect of superior labour market conditions may not be strong.
In summary then NNP, or other measures of material production and consumption, do not seem to be particularly relevant to the promotion of social welfare as indicated by happiness and satisfaction – directly anyway. This chapter will argue that NNP contributes indirectly to some of the other factors which make up wellbeing.
Alternatives to Material Output
The economist with the high standing who has written most about alternatives to material output is Amartya Sen who starts with the notion of ‘functionings’ which summarise the life a person might lead. Some functionings are elementary: being well nourished and disease free. Some are more complex: having self respect, preserving human dignity, taking part in the life of the community.
(The list has echoes of Abraham Maslow’s hierarchy of needs:
High level needs
– need for cognitive understanding;
– need for self actualisation;
– esteem needs;
– needs for belongingness and love;
– safety needs;
– physiological needs.
Low level needs.
Observe that measures of material output such as NNP only directly address the lowest – and possibly the second to lowest – needs on the hierarchy, again warning us that we should not get obsessed with NNP.)
Sen then introduces the key notion of ‘capability’ which refers to the alternative functionings (‘life choices’) a person might have. His notion of well being is not what you consume (‘opulence’ as Sen calls it) but the choices (or ‘capabilities’) the individual has. These are:
(1) Political freedoms: ‘the opportunities that people have to determine who should govern, and on what principles, and also include the possibility to scrutinize and criticize authorities, to have freedom of political expression and an uncensored press, to enjoy the freedom to choose between different political parties, and so on.’
(2) Economic facilities: ‘the opportunities that individuals enjoy to utilize economic resources for the purpose of consumption or production or exchange.’
(3) Social opportunities: ‘the arrangements that society makes for education, health care and so on.’
(4) Transparency guarantees: ‘the need for openness that people can expect: the freedom to deal with one another under guarantees of disclosure and lucidity.’
(5) Protective security: ‘needed to provide a social safety net for preventing the population from being reduced to abject misery, and in some cases even starvation and death. Its domain includes fixed institutional arrangements such as unemployment benefits and statutory income supplements to the indigent as well as ad hoc arrangements such as famine relief or emergency public employment to generate income for destitute.’
So material consumption is only a part of the totality of capabilities.
Like the New Right, Sen talks of ‘freedom’ and ‘choice’. But unlike the New Right, Sen does not assume that if everybody has ‘freedom’ (in the New Right sense) in a ‘free’ market they will all be better off. This is partly because Sen is an expert on inequality, and he knows ‘free market’ outcomes will have some people worse off (as happened in New Zealand). But the capability approach also diverges markedly from the New Right when it concludes that government spending (on education, on health care, on other things such as the environment, culture and recreation facilities) can enable individuals to do and be so much more.
The logic of Sen’s approach led to the World Bank’s ‘Human Development Index’ (HDI), which is intended to replace NNP per capita as an indicator of the state of progress of a nation. It combines per capita material output with measures of educational and health achievement. Thus a rise in life expectation increases the HDI even if NNP per capita goes down. Literacy is there, because the capability markedly affects our ability to flourish and enjoy life. The index has to be simple, because most countries do not have complicated statistical bases. Had it been practical, Sen would have wanted a measure of the differences between men and women, which can be grotesque.
Sen’s concerns are illustrated by Kerala, one of the poorest states of India, nonetheless has a life expectancy of over 73 years, not too different from the New Zealand expectancy of 76.9 years. In contrast, Afro-Americans, living in some of the richest cities in the world, have a shorter life expectancy than those in Kerala. Despite its material limitations, Kerala has organized itself – including its education and health systems – to give its residents substantial freedoms, including that of longevity.
New Zealand is 19th in the world on current HDI standings, three above its NNP per capita ranking. (The biggest divergence in the country rankings is for South Africa, whose HDI ranking is a whopping 44 places below its NNP per capita ranking, a terrible indictment on apartheid’s record on health and education, that will take generations to repair.)
Constructing a Human Development Index for Well-off Nations
However, the World Bank’s interests are poor countries, and the index is not be very good at discriminating between rich countries. Their literacy and longevity are much the same. So should we go back to NNP per capita which shows bigger variation?
Not necessarily. The big gap between New Zealand and the US in per capita NNP terms does not reflect relative happiness, and probably does not reflect the difference in average wellbeing either. Instead, let us try to elaborate the HDI approach for rich countries, with an index which has five (rather than three) primary components: material consumption; health; education; diversity for choice; and leisure. However, with Sen, we assume that there are some basic conditions which are already attained:
– effective civil rights;
– personal security;
– national security;
– environmental sustainability.
We shall see that they affect the other measures.
Material Consumption: It would be easy to measure material consumption by per capita NNP. But we need to recognise that some elements of NNP contribute to civil rights, personal security national security, and environmental sustainability. Since there are considered prior to the measure, they should be deducted from NNP to get a better measure of material consumption. Moreover spending on health, educational attainment and creativity should also be deducted, because they appear directly or implicitly in the other three components of the index.
Health: The HDI uses longevity to measure health. Countries with better statistical systems are beginning to be able to adjust each of year of life for its quality. (The competing measures are QALYs – quality adjusted life years – and DALYs – disability adjusted life years. )
Educational AttainmentThe HDI gave a weighting to literacy and years of schooling. Richer countries might still want to use literacy measures in their index (although the more sophisticated ones of the ….), and also the proportion of the adult population with university degrees, and the hours they experience through a life time of post-secondary education including tertiary, vocational and continuing. Various institutions such as libraries, galleries and museums are also a part of this.
Notice the measure does not include numbers currently in educational institutions nor the attainments of the current crop of graduates (except as a part of the whole of adulthood educational attainment). That is because they are additions to the stock, not the stock itself. It would be as inappropriate to measure the stock of investment by the plant and equipment being installed.
Diversity for choice Choice is key to Sen’s account of human wellbeing. Health, education and leisure all enhance choice, but we need a measure which directly indicates the possibilities. We are interested in real choice – a problematic notion since who would think a plethora of cheeses was a real choice (unless they were French)? Sen is more concerned about life paths, and that might be better measured by indicators of diversity and creativity. Heterogeneity is important.
There is no diversity component in the HDI, because it is difficult to measure even in rich countries. Perhaps relevant are such factors as ethnic and religious diversity. Opportunities for women and the disabled are good tests too.
Richard Florida proposes a ‘bohemian’ index based on the ‘number of writers, designers, musicians, actors and directors, painters and sculptors, photographers and dancers’ – he means people who describe their occupation in these terms. Perhaps we should be less precious and include sportsmen, in part because we have a different agenda to Florida. (He is trying to find measures of economic performance: we are trying to measure the choices which promote a quality of life.) It is a hard one t measure this one, but we would miss the point of human development were it left out.
American workers work 1867 hours a year, Dutch ones work 1347 hours, some 28 percent less. In fact hourly labour productivity in the Netherlands is higher than the United States (by almost 17 percent) even though NNP per capita is lower by 23 percent). The Dutch have chosen to take out their higher productivity with more worker leisure, rather than more material consumption. It is a valid choice, and needs to be included in any index of human development. Ideally rather the hours worked should include travelling time to and from work. And probably some adjustment needs to be included from forced leisure from involuntary unemployment.
(We will find that New Zealanders work slightly longer than the OECD average, so unless involuntary unemployment is low – probably not – and travelling time to work is low – possibly – New Zealand’s leisure is not especially high.)
It is not proposed here to create a human development index. That is technically a very big job, especially if it is done for a number of countries. The point of this section is to emphasise that NNP per capita is not the sole criteria human development, nor the sole objective of a nation.
The point of NNP
Now it may seem that by this stage the significance of NNP has been reduced to almost nothing, or at least to the point with which it is hardly worth bothering about. In fact the conclusion is exactly the opposite. Material output is needed to attain the other ends. It is necessary to run the institutions which give us civil rights, personal security, national security, and environmental sustainability. It not only gives the wherewithal for material consumption, we need it in the pursuit of health of education of diversity and of leisure (allowing that high productivity makes leisure more feasible).
So even if NNP does not contribute directly to national welfare, or only to a small extent directly, it contributes indirectly in very important ways. NNP is lower in the hierarchy of national goals, but it is a means to higher ones, and to some extent we know a bit about enhancing NNP, getting it to grow, so it is worthwhile pursuing it, but not by compromising higher goals. Recall that Ceausescu’s Rumanian abrogated civil rights, screwed back consumption, discouraged leisure and so on to maximise NNP growth. It was not successful, thank the lord. Otherwise there would be more others arguing for such policies. When we pursue NNP (or whatever measure of material output) we must always check that we are not compromising what we really care about.
The great thing is we know a little about how to enhance NNP, of getting it to grow. This is what this much of this book is about.