According to his numbers (or, should I say, the numbers spreading through the media), a marriage for a man results in a $32,000 increase in happiness. That's the lump sum Frijters says the man would need to receive out of the blue to make him as happy as his marriage will over his lifetime. But his wife would need much less, only about $16,000. Similarly, when it comes to divorce, men are so devastated, it feels like they lost $110,000, while their hardened ex-wives only feel a $9000 loss. These data are just anectdotal, however - they have yet to be published in any peer-reviewed paper.
What, if anything, does this really mean? Does it mean that men value their spouses more? No, it doesn't. It may mean any number of things, really. It helps to start by understanding where the numbers come from.
The dollar values that Frijters assigns to life events come from mining a unique set of data that has tracked the happiness and major life events of about 10,000 Australians once a year since 2001 Participants were asked to describe how satisfied they are with their lives on a scale of 0 to 10. That number often changed before, during, or after major life events and sudden changes in income. From these changes, Frijters put cash values on the effects of happiness of major events such as marriage, divorce and illnesses.
But this isn't just any 1:1 correlation between happiness and finances; the math behind the dollar amounts Frijters gives is staggeringly complex. For example, the upcoming paper that explains his analysis in general (though not between men and women) includes this equation for the valuing of life events:
He goes onto explain that this equation can be solved through at least two approaches, and that each gives different numbers for answers, and he compares and contrasts the different values obtained from each. First off, he compares value changes in happiness due to life events to known changes in happiness due to those who received a financial windfall; ie winning the lottery. This is the approach he recommends, and is new to the field. The other method compares changes in happiness using a "classic" approach, by valuing things at the level of stated preferences by individuals. The second method, he explains, tends to lead to higher compensation values and less accuracy.
While on the one hand the monetary numbers seem straightforward, further analysis reveals they are anything but. For example, using Frijters analysis, anyone with a higher income will see life changes as a much more costly life event. As the paper states, "The absolute value of the income change will depend on the individualís initial income level, with wealthier individuals receiving higher absolute compensation." This, of course, doesn't mean that it's harder or worse for someone who was originally at a high income to recover from such an event than someone who started at a lower income; indeed, in real life, it's often the opposite.
Given this level of complexity, it's hard to explain exactly what it means for one person to value a life event higher than another, especially when it's unclear exactly which method was used to get the sex differences.
It's at this point that I question the separation between men and women. As far as I can tell, no corrections were made for income level between the two groups. Considering the pay gap is fairly high in Australia (higher than even some third world countries like Sri Lanka), we would expect that men would automatically have higher compensations. Without accounting for this, it seems difficult to determine exactly how much difference there is between men and women. Even still, it's simply possible that men and women value money differently. For example, if for a woman, a gift of $100,000 only increases her happiness by 0.1 on a scale of 1 to 10, whereas for a man it increases his happiness by 0.5, we would expect that the compensations for life events would therefore be very different, too, though the actual happiness gained may be the same. In the end, what was the difference in happiness gained or lost between the two groups?
In part, I would just really like to see the data that led to all these headlines. It seems shoddy at best that this information is being circulated as truth without any means for someone to take a look at the hard numbers. And trust me, I looked. No one can cite one reference to the "study" this sex-separated data is in, except for one site which said the data was merely presented to some people at some university. Call me a scientist, but I want to see the results myself, published in a peer-reviewed journal.
That's not to say that what Frijters is doing is meaningless or irrelevant. Frijters work is very contemporary and absolutely important to the field of economics. Understanding these kinds of events in terms of monetary gains and losses is crucial to insurance and legal groups which have to do exactly that every day. Some of the more interesting finds from this data were that, on average, it takes about 2 years to adjust to a life change, with negative changes felt much more strongly and lasting much longer than positive ones. The death of a loved one, for example, is devastating across the board, and most never regain the same level of happiness they had before. However, translating high-level, complex economic analysis down to a catchy headline misses the point of his work.
References
Frijters P, Johnston DW&Shields MA. 2010.Happiness Dynamics with Quarterly Life Event Data. IZA Discussion Paper No. 3604. Available at SSRN: http://ssrn.com/abstract=1184013 or
doi:10.1111/j.0042-7092.2007.00700.x
. ...i'll post later...to me the gender-brain differences are very solid but largely ignored as non-politically correct by yank press......