I don’t often get time these days for a leisurely browse around the blogosphere. But I found a few spare minutes today, and happened to stumble across an unbelievably trivial but nasty spat between the lefties at Larva Rodeo and compulsive attention-seeking legal academic James McConvill. Don’t any of these people have better things to do with their time?
I’d like to say something positive about McConvill, if only because no-one else ever seems to do so and because he’s one of the few other Australian legal academics who blog (apart from his partner-in-crime Mirko “The Torturer” Bagaric and Kim Weatherall and Simon Evans from Melbourne Uni). Accordingly I was interested when McConvill mentioned that his doctoral thesis was about happiness research, a subject in which I’m also rather more idly interested. So I searched the legal bibliographic database Agisplus and discovered a recent article about happiness by McConvill and Bagaric in the Deakin Law Review. Sadly it seems to confirm Andrew Norton’s theory about McConvill’s intellectual output:
To get the most attention, the essay should be wrong. Logical essays are read and understood. But an illogical or wrong essay will prompt dozens of other authors to rise and respond, thus giving the author mounds of publicity.
McConvill and Bagaric’s happiness article is based predominantly on a 15 year old book by Christian psychologist David Myers. It cites a couple of other happiness studies by psychologists (mostly quite old ones), but makes almost no mention of the increasing number of recent economic analyses of happiness research.
As a 2003 article by Di Tella, MacCulloch, and Oswald observed:
Richard Easterlin (1974) began what remains a small literature, and recently updated his work in Easterlin (1995). Other contributions include Gruber and Mullainathan (2002), Von Praag and Frijters (1999), Ng (1996,1997), Blanchflower and Oswald (1999), Frank (1985), Inglehart (1990), Fox and Kahneman (1992), Frey and Stutzer (2000), Konow and Earley (1999), Oswald (1997), Winkelmann and Winkelmann (1998), Gardner and Oswald (2001), and Alesina, Di Tella, and MacCulloch (2001). Di Tella, MacCulloch, and Oswald (2001) study people’s preferences over inflation and unemployment. Di Tella and MacCulloch (1999) use happiness data to examine the properties of partisan versus opportunistic voting models. See Frey and Stutzer (2002) for a review.
And there’s been quite a lot written since then, including a recent paper by the blogosphere’s own Andrew Leigh and Justin Wolfers. McConvill and Bagaric mention only Inglehart and Frey, and then only in passing. They somehow reach the following conclusions:
Given that liberty is conducive to happiness, it follows that laws which curtail liberty are only justifiable if they will compel or guide people to engage in conduct that empirical evidence demonstrably suggests will promote happiness, or if the laws will compel people to desist from conduct which it is proven will harm them. At the most general level, this means that restrictive laws which do not directly harm others should be repealed unless there is evidence to suggest that the proscribed conduct is detrimental to one’s happiness. By way of example, planning laws restricting what one can do with his or her property and legal regulation of pornography should be relaxed. Moreover, legal regulation should be introduced discouraging people from engaging in passive pastimes. Thus, television coverage should be limited to, say, no more than four hours per day or, at the minimum, TV stations should be compelled to provide mental health warnings to their audience. …
In relation to laws levying taxation, we have noted that in terms of the effect of money on happiness (and hence the capacity to contribute to government revenue), there are three categories of people: those who do not have enough money, those who do have enough and those who have more than enough. There is obviously scope for considerable debate concerning exactly the rate at which tax should be levied for each of these categories. However, the following general observations can be made. People who do not earn enough money for the bare essentials are in a position where they are adversely affected by being required to divest any income through the taxation system. Thus, they should be taxed very lightly, if at all. People on roughly average incomes can afford to live with a little less, although every dollar less they have has a demonstrable connection with their level of happiness. Therefore they too should be taxed lightly. People on incomes above this can afford to pay a lot more tax and although they might not know it they will not experience any meaningful diminution in well-being if they have less money. They should be taxed heavily. We have suggested that this translates to the following rates of taxation. For income under approximately $US15,000 the rate should be 0 to 5 per cent. For income between $US15,000 to $US40,0000 ($US40,000 still being above the average level of income in most, if not all, OCED countries) the rate should be about 10 – 15 per cent. Incomes beyond this level should be taxed at approximately 75 per cent.
McConvill and Bagaric don’t appear to see any logical contradiction betwen advocating laws which promote liberty and urging a partial ban on TV and punitive rates of personal taxation. Fortunately they don’t appear to have noticed that happiness research also shows strong associations betwen happiness and marriage and religion. I hate to think what sorts of laws they’d recommend if they focused on those areas as well. I used to think that cross-disciplinary academic writing was a desirable antidote to excessive specialisation and rigid compartmentalisation of human knowledge. But the output of McConvill and Bagaric provides an excellent argument to the contrary.
Of course, there isn’t a simple straight line association between tax rates, growth and productivity. Countries with the lowest tax rates don’t necessarily enjoy higher productivity or faster economic growth, let alone greater happiness. But few would deny that a marginal tax rate of 75% on every dollar earned over $A53,000, which is what McConvill and Bagaric are advocating, would have a fairly serious effect on incentives towards work, productivity and growth. They would be on firmer ground if they argued for more modest increases in the total tax take without punitive rates on middle-higher income earners, and if they confined their policy prescriptions to utilisation of the additional revenue on enhanced health, education and training. Those services also correlate strongly with happiness and can be enhanced without any need to impose coercive laws.
Incidentally, one of the more succinct demolitions I’ve read of the nanny-stateist position on happiness research (exemplified in Australia by Clive Hamilton, Ross Gittins and Richard Eckersley and now in an extreme form by McConvill and Bagaric) is a CIS paper by Johan Norberg:
It is a simplification to say that growth does not contribute to happiness. In fact, one of the few things there is a consensus about in this very young field of science is that money does buy happiness. There is an extremely strong correlation between wealth and happiness. Low-income countries report low levels of happiness, middle-income countries report middle levels and high-income countries report high levels.
What the researchers say, though, is that this correlation levels off at a national income of about $10,000 a year. After $20.000, Layard says, ‘additional income is not associated with extra happiness’ (p. 33). The economist Richard Easterlin presented this surprising conclusion in a study about Japan. Since the 1950s, income in Japan increased almost 10-fold, but the Japanese are no happier today than they were then.
This is the most highlighted finding from the research. But in fact, it is yet to be proven. The fact that a higher income level does not translate into higher happiness does not mean that growth doesn’t. What we do know is that there is a big jump in reported wellbeing when countries move from about $5,000 to $15,000 a year. And this is consistent with a much more dynamic interpretation than Easterlin’s and Layard’s. …
And, after discussing an apparent large surge in recorded happiness levels in the immediate post-WWII period, Norberg observes:
That could help explain why happiness reached high levels in the Western world after the Second World War, when they were close to the Irish situation. With economies growing rapidly, people began to think that their children would enjoy a better life than they had. The fact that economic growth since then has not increased happiness much from those levels does not mean that it is useless¢â¬âit is the fact that growth has continued that makes it possible for us to continue to believe in the future, and to continue experiencing such high levels of happiness. The critics who think that they can conclude from the stability of happiness that zero-growth is preferable overlook that loss of income undermines happiness. And growth makes non-zero-sum games possible. Without it, whenever someone succeeds and gains, someone else has to fail.
Rising happiness
But there is a much more basic problem with Layard’s interpretations of the happiness research. In fact, happiness hasn’t stopped increasing in the West. On the contrary, in most Western countries where we have surveys since 1975, happiness has increased. Yes there are diminishing returns, but even at our standard of living, people do get happier when our societies grow richer, albeit at a slower rate.
Thus, advocacy of government intervention to impose happiness by force in place of materialism seems to owe more to the personal predilections of the advocates than anything else. McConvill and Bagaric unwittingly demonstrate this towards the end of their article. After advocating a legal system based on imposition of norms and behaviour shown by the psychologists’ research to be conducive to happiness (rather than on old-fashioned values like justice or human rights), they conclude:
This approach ought to be adopted unless it is established that either (i) there is a more important human goal than happiness; or (ii) human beings are so diverse that it is not feasible to make valid generalisations concerning the matters that make people happy.
But even if human beings aren’t so diverse as to render meaningless any generalisations about what makes people happy, why should we think that it’s a great idea to empower governments to impose happiness by coercive means? Mightn’t there be some rather obvious risks involved in surrendering that degree of personal autonomy to the state? McConvill and Bagaric’s writing is replete with logical non sequiturs like this. Not for nothing did Thomas Jefferson speak of the aim of government as promoting “life, liberty and the pursuit of happiness” rather than “life, liberty and the imposition of happiness whether you like it or not”. God save us from happiness Nazis and attention-seeking legal academics.
PS – In fact, you can make a strong case that even Jefferson took the happiness meme too far. As Frank Furedi argues (courtesy Patrick in the comment box):
Happiness may be a worthwhile objective, yet as the philosopher John Stuart Mill noted it cannot be the direct end of people’s activity. ‘Ask yourself whether you are happy, and you cease to be so’, observed Mill. He added that the ‘only chance is to treat, not happiness, but some end external to it, as the purpose of life’. As Richard Schoch recently put it, Mill understood that the ‘secret of happiness’ is ‘the paradox that you find it”¦only by searching for something else’. Happiness is the indirect outcome of engaging with others in the pursuit of civic virtues, and attempting to do good.
Hi Ken,
That’s a fair call. We chose this material for the article because we thought it gave a clear explanation of the issues for an audience that probably hadn’t been exposed to the literature before.
I’ve delved into more research happiness literature in my books “The False Promise of Pay for Performance” (2005) and “Shareholder Participation in the Corporation: A New Perspective” (2006).
I appreciate your analysis.
I should point out that the latter part of this post (starting with the extended quote from Johan Norberg) was added after James McConvill visited and commented. I hope he returns for a second bite, because it seems to me that there’s an interesting (and hopefully civil despite my gratuitous “happiness Nazi” slur) debate to be had here.
For links to stuff on happiness and economics
http://austrianeconomists.typepad.com/weblog/2005/08/happiness_and_e.html
No sense of humour today Ken?
I never read LP and share Andrew Norton’s view of McConvill. I still shudder remembering his article on judicial nominations. I am not sure it contained a single proposition that was completely and undeniably accurate, and it contained a good deal that were absolutely and incontrovertibly false. I haven’t been able to bring myself to read anything of his since.
Accordingly, I regret that I won’t be able to read this piece :)
Ken, I share your views on cross-disciplinary writing.
Regarding the use of happiness data to draw conclusions about optimal tax rates, one would of course also want to cite Andrew Oswald’s paper, “On the Common Claim that Happiness Equations Demonstrate Diminishing Marginal Utility of Income” (http://www2.warwick.ac.uk/fac/soc/economics/staff/faculty/oswald/concavity2005.pdf), as well as the huge, complex, and important literature on optimal taxation.
Thanks Andrew. The Oswald paper seems to be a rather long-winded statement of the obvious: that reported changes in happiness levels aren’t necessarily the same as actual changes in happiness levels. Nevertheless, it’s an obvious distinction that seems to have escaped Hamilton, Gittins, McConvill etc.
OTO the literature on optimal taxation is so complex and mathematical that I’m afraid I have Buckley’s chance of understanding it.
Ken
I note that you say that “you’d like to say something positive” about McConvill, then go on to say only negative things.
Unless I’m missing something.
Since you yourself recognise that he’s attention-seeking, why have a go at my post when you’re also feeding the McConvill commentery industry?
That should read “commentary”.
Kim
I wasn’t having a go at your post per se, but rather at the quality of the comments to both your post and McConvill’s. In both cases most were puerile in the extreme (although “Bill Posters” was obviously amused), and many of the comments were quite ill-tempered to boot. Nevertheless, I DID think that both your primary post and McConvill’s (which was essentially a whinge about Mark B’s allegedly obsessive focus on him) had somewhat trivial and self-referential subject matter.
You don’t think it’s astonishing and worthy of comment that someone should post their application to be a Professor on their blog and then handicap the “contenders”, Ken?
I don’t know that McConvill’s work warrants too much seriousness in evaluation in comments boxes. You’ve done a great job in your post of demonstrating how poor his scholarship is.
But don’t forget the nostrum from Andrew Norton you yourself quote!
Anyway, a lot of people had a good laugh over the comments on both posts, so perhaps the alleged goal of increasing the quantum of happiness is being advanced in an unintended fashion.
So why post on it?
BP
If you bothered to read my post, instead of making patently stupid debating points, you would discover that it wasn’t about either the subject matter of Kim’s post or McConvill’s. They were simply “hooks” to lead into a more serious discussion about happiness research.
That isn’t to suggest that blog posts should always be serious and worthy, or that trivial topics shouldn’t be covered. But tit-for-tat trivial, ad hominem posts each accompanied by a chorus of nasty comments, are not IMO to be admired.
That still begs the question of why you would post on McConvill’s “happiness research”, which appears to be rubbish.
If he wasn’t out there writing ridiculous op/eds and starting 5 online law reviews, you wouldn’t.
In fact I think a case could be made that McConvill’s publications constitute “trivial and self-referential subject matter”.
On the substantive issue, since I do like reading NG, KP et al, this piece seems modestly sensible:
I’ve always understood that the happiness research said that growth in incomes, not the level, was what made societies happy. As Hobbes said “felicity consisteth in prospering, not having prospered”.
Within societies, of course, it’s relative income. Or as a friend of mine says “happiness is earning a thousand dollars a year more than your brother-in-law”.
Inequality aversion is not a difficult concept, even if it does lead to mathematically murky waters. Imagine that God has just created you de novo. He tells you that:
– He’s going to get you to throw a dice to determine your whereabouts in the social scale (you have to do it rather than him because his omniscience would mess it up).
But before that you have to choose which society you prefer to live in – a rich and unequal one or a poorer but more equal one.
Inequality aversion is just a measure of which you’d choose.
DD
I wasn’t having difficulty understanding what inequality aversion actually was. It’s fairly self-evident. What I was interested in was how you would measure it in a given society. Andrew Leigh doesn’t know either, even though some of the optimal taxation literature factors the concept in as a variable. If optimal taxation formulae are to be meaningful, you would imagine that this concept needs to be capable of measurement.
Oh, now you’re asking far too much of the optimal tax literature. It’s precisely because lots of parameters in it can’t be easily measured that its (admittedly clever) mathematics has had little practical influence on policy.
But in fact inequality aversion is sorta measurable – mostly through observed behaviour in experiments, but also by the odd natural experiment. There’s also a strand of argument that we can derive it from the revealed preference of the electorate in the form of the actual tax and welfare systems they vote for, though I’m dubious of that approach myself – there’s far too much historical contingency, combined with status quo bias, in real life systems.
[…] issues have been aired before on CT: Don Arthur puts a version of objection E; Ken Parish here and here seems in effect to interpret point 5 as a rebuttal of point 8, although I may not have […]