Herewith my op ed from the Herald and Age today.
What is the good life and are we living it?
Assessing and measuring wellbeing has vexed us since ancient times. But a funny thing happened on the modern world’s way to the answer. The metric that economists used to dampen down the business cycle – Gross Domestic Product (GDP) – received such prominence that it ‘went viral’ as we say these days. It became the default measure of national progress.
But there’s lots wrong with GDP as a measure of economic wellbeing let alone more general wellbeing. Measuring gross activity, it ignores the growth and depreciation of assets – such as buildings, equipment, natural resources like farmland and mineral deposits, biodiversity and clean air. And that’s not to mention the greatest asset of all – our knowhow.
Moreover GDP is measured by money changing hands. So converting bread, mince and salad into a hamburger increases GDP in McDonalds but not at home. More starkly, an evening of passion and pleasure only adds to wellbeing as measured by GDP if it happens in a bordello! More broadly still, GDP takes no account of the distribution of income or of our physical or social wellbeing.
But considering how different all these phenomena are, how can we possibly measure their sum impact on national wellbeing in a single number? Because it would ‘dumb down’ complex issues, economics Nobel Laureate Amatya Sen initially refused to participate in the construction of a single index of human development to help guide development in poorer countries. But he relented because he appreciated that, however unsatisfactory a single wellbeing index might be, it was better than the alternative. Given the thirst for simple answers, the alternative is even more dumbing down as would occur if GDP yet again filled the vacuum.
The Herald/Age Lateral Economics (HALE) Index of Wellbeing takes in social, environmental, educational and health outcomes. But aggregating them into a single measure requires the wisdom of Solomon. How would you weight the environment against education for instance?
Many existing indices of wellbeing give equal weighting to the different dimensions that they consider. Thus for instance the Kingdom of Bhutan’s famous index of Gross National Happiness takes nine ‘dimensions’ of wellbeing and weights them equally. The architects of this measure claim that equal weighting avoids bias, but this is smoke and mirrors as an unexamined choice is essentially arbitrary.
Thus in Bhutan’s measure, Dimension Two of wellbeing is ‘time use’. Should it really receive equal weight to Dimensions Six and Eight – education and living standards? Wouldn’t extreme poverty or illiteracy be worse than serious time imbalances in one’s time-use? And how would you measure such an imbalance given wide ranging preferences between different people and trade-offs between the labour-leisure choice and other values like health and education?
The HALE Index makes these choices by beginning with GDP – which in fact correlates surprisingly well, though not perfectly, with many of the most important aspects of wellbeing such as educational, environmental, social and political cohesion and health. We then adjust GDP for factors – both economic and non-economic.
Beginning with Net National Income (NNI), which is essentially GDP adjusted for the depreciation of physical capital, we adjust our index to reflect the depletion of non-renewable resources by mining and the discovery of new mineral resources. We allow for any degradation of agricultural land. We also adjust the index by estimating a cost to represent the risk we are taking today of substantial climate change in the future.
However, the biggest news in our index is that we bring Australians’ knowhow into the picture. For this stock of ‘human capital’ is many times more valuable than all other (tangible) assets combined. In our index, human capital is measured by the risk of early childhood deprivation, school results when compared with other countries and post-secondary education.
But what about more value-laden questions? How much weight should health and employment outcomes carry and how should we treat changes in the distribution of income?
Agreeing with former UK Chancellor Denis Healey that policy’s task is “eroding by inches the conditions which produce avoidable suffering”, we went in search of something which one might call ‘Gross National Suffering”.
Here we used evidence from surveys that ask people to rate their own happiness or life satisfaction. Indices built directly from this data rarely move much. Most people rate themselves at around 7.5 out of 10 with the private ups and downs of different lives cancelling each other out when aggregated.
However, the survey data enable us to detect specific associations between lower self-reported wellbeing and specific phenomena. Of course many problems produce suffering in individual lives, but not amongst sufficiently many Australians to move a national index. But some things make a difference even in the scheme of the whole population’s wellbeing.
Underemployment (including unemployment), overwork, mental illness and obesity are all relatively widespread and strongly influence reported wellbeing. And the direct wellbeing surveys also give us a way of escaping from the arbitrariness of our own subjective values when considering how changes in the distribution of income should be considered when measuring total wellbeing.
The data suggest that on average for those Australians on very low incomes, a $6,000 increase in income is associated with increased self-reported wellbeing by one percentage point. By contrast the same increment in happiness requires over $100,000 for a household already earning over $100,000 a year.
These relativities enable us to adjust aggregate income growth in the economy for its efficacy in improving people’s reported wellbeing. When national income flows to lower-income households it funds more improvement in wellbeing than when the same amount goes to higher income families.
Does the sum of these manoeuvres solve the age old dilemma of measuring our wellbeing? Far from it. Having sought the wisdom of Solomon, what wisdom we’ve gained has been that of Socrates. We know only too well how little we know.
All that acknowledged, are we proud of what we’ve come up with – and with the praise our index has already received – as a reasonable compromise between heart and mind, rigour and common sense? Definitely.
Good to see this up after reading Ross Gittins piece at Fairfax. Clearly neolib economics has been selective in what criteria are employed to measure well-being eg GDP, ever since it was decided that wars on inflation are prioritised against the unemployment scourge.
I know people love single numbers, but perhaps you could try and teach people that many things have multiple dimensions that arn’t necessarily comparable due to being qualitatively different.
What comes next?
Sounds good. Given that the HPI and other such measures publish their methodology (ie, weights and variables), are you going to or have you already publish(ed) the methodology?
What’d be interesting is using a crowd-sourcing approach to the methodology and looking at the correlation in rates of change (and or levels depending on the circumstance) among the different methodologies.
The reason I suggest that is, as you say, GDP or GNP or Real Net Disposable Income is highly correlated with other measures of well-being, probably due to their correlation with taxation revenue, which permits ammeliorating market failures such as infant mortality.
So it’d be interesting to see how lots of different indices perform stacked up against each other. Also, as you say, choosing weights for the variables is a horrible task requiring the wisdom of Solomon. So crowd-sourcing (via publishing the methodology and a suitable data set to save the lazy having to re-create it) would allow number crunchers to do something other than read blogs during their lunchtime.
Yes, the methodology is documented – and in a report to Fairfax and my understanding is that they intend to publish it, but the decision as to when is up to them.
Don, tomorrow’s papers have a description of the broad results for this quarter and how they differ from GDP and Saturday gets the full report.
:-) funny. I was just in Tokyo this week talking about GNH around the world. Many of the delegates were from official statistical agencies and they were increasingly talking about how private companies are stepping into the area they used to own. Guess the HALE is competition for the ABS! Good, they can use a bit of competition.
The big querie is whether you, unlike the ABS, are willing to have the underlying micro-data available to researchers and policy makers?
Hmm. After the HALE, I suggest the Herald/Age Reality Testimonial of International Excellence. The HeARTIE.
Here’s the LE report as made available by Fairfax, for those who were interested.
I second Paul’s comment.
Without releasing the data and weights, what hope do you have for promoting the index as reliable, particularly when it conflicts with popular perceptions? Then again it’s your baby, not mine.
Paul Walter,
I think your comments are confused. ‘Neoclassical economics’ doesn’t teach that GDP is good for measuring wellbeing. It teaches all the things that I pointed out above about its shortcomings. And as for there being some mechanical tradeoff between inflation and unemployment, other things being equal, higher inflation (at least above low levels like 2-3%) is likely to reduce long run growth.
There are exceptions to this – if I were in Europe or the US I’d be in favour of higher inflation as the lesser of the evils to choose from (we don’t have the same problems) but simply assuming some simple tradeoff between low inflation and higher unemployment isn’t very realistic.
Nicholas Gruen, It’s not ME that’s assuming some simple “trade off between low inflation and high employment”, or not, but if that’s the case, if I am, its only becuase that’s what the Murdoch press and media have been telling me since the ‘seventies high unemployment recession and its “austerity”. EG, the need for a fairer way to measure well being than GDP, mediated through/by vested interests.
That’s how the Gittins argument struck me. That they use GDP morphing from a snapshop of economic activity and its trends into something our media interlocutors employ as definitive in defining for the public when they’re well off or not, maybe depending on how the proprietors are feeling that morning after the early gloomy briefing with their economics advisers.
But you are right- economics is your field and I have no difficulty whatsover in playing the role of Socrates on this occasion- like the rest of the “99%” faced with lack of rquisite wit or education and not acceptably cogniscent of the nuances you bring into play. Will sleep on it- meh!
“More starkly, an evening of passion and pleasure only adds to wellbeing as measured by GDP if it happens in a bordello!”
I looked forward to reading how you went about measuring the value of evenings of ‘passion and pleasure’ Nick. But after through reading the report I was unable to find any mention of the topic and how it contributes to the wellbeing index. Very disappointing as I was looking forward to presenting the other half with some evidence based suggestions on how we could improve the nations wellbeing.
Mea Culpa Carjig75.
Nick,
I am guessing from your silence you will not be making the underlying data available? Oh irony of ironies ….
Hi Paul,
Apologies for not responding sooner.
My silence has been because of my busyness, and the fact that our work is commercial and so our client needs to be consulted. That hasn’t happened yet. There would also be some minor logistical issues to deal with.
But I’m well disposed to releasing our underlying data.
In any event, all the underlying data we use is already public.
I have posted some comments on HALE on my blog that might be of interest to someone:
http://wintonbates.blogspot.com/2011/12/what-factors-should-be-taken-into.htm
and
http://wintonbates.blogspot.com/2011/12/have-important-factors-been-omitted.html
[…] is the key driver of improvements in human capital. Nicholas Gruen has written about the Index here. Dennis and Nicholas have made me think about the value we add to a person’s entry profile […]
[…] ‘wellbeing period’ if I might call it that, Lateral Economics built the Herald/Age Lateral Economics (HALE) Index of Wellbeing. In fact, someone from Treasurer Joe Hockey’s office did ring about it and responded to my […]