This is a response to the objections raised against the Stevenson & Wolfers paper “Subjective Well‐Being and Income: Is There Any Evidence of Satiation?” in the “Analysis” part of the post “Well-being, income and economics: why money does not buy happiness” that was published in Pompeunomics. The objections will be criticized in the same order they are presented in the original Easterlin paradox post. All works mentioned in the text that are not cited, can be found in the post published at Pompeuonomics just mentioned. Quotes, unless explicitly mentioned are from the Pompeuonomics post.
The use of the logarithmic function
The use of the logarithmic function: “I am very convinced it does not hold in practice.” Well, conviction must be well grounded. Reasons to hold this conviction must be showed and they have been not, at least explicitly showed in the post. What the authors suggest seems plausible at first sight, since the evidence they provide does show that, for proportional increases in income, happiness also increases and it does so steadily. We do not need to assume anything, the linear-log relationship (increases in income increase happiness but extra income “buys” a smaller amount of additional happiness) between well being and income corresponds to the data, so it’s a finding of their research, not an assumption [see also Stevenson & Wolfers (2008)]. After all, what Stevenson & Wolfers show, is that money seems to exhibit diminishing returns, nothing unexpected by the average economist.
Also, the use of econometrics should not be an objection unless it’s shown that they have been not properly used and the fact that econometric analysis has its limitations does NOT imply that its use in the Stevenson & Wolfers paper is inappropriate since it has not been shown that such limitations apply to the Stevenson & Wolfers paper. Since this has not been, at least, explicitly showed in the Pompeuonomics post, we should give the benefit of the doubt to Stevenon & Wolfers until further research shows up with more robust estimations or shows methodological issues with their research [like Easterlin (2010) (which by the way uses an OLS regression) critique of Stevenson & Wolfers (2008) because of the use of short term instead of long term analysis, this critique does not apply Stevenson & Wolfers paper under examination since in a mimeo they find very similar results that hold over the long run, Sacks, Stevenson and Wolfers (2013), see also Veenhoven, Ruut, and Floris Vergunst (2013)].
Do Stevenson & Wolf misinterpret Easterlin?
The idea that Stevenson & Wolf misinterpret Easterlin is also difficult to cope with. Easterlin obviously tries to push the idea that increasing aggregate income does not lead to increasing happiness. See Figure 1 of Easterlin’s paper which implies such vision since it’s a cross-country analysis.
This collides head to head with the Stevenson & Wolfs findings of their cross-country analysis and just in the first paragraph of the Stevenson & Wolfers paper, they note that Easterlin did not make a satiation hypothesis (at least in his 1974 paper). They also note that Easterlin’s “original” claim is contradictory with other studies by Stevenson & Wolfers and others that point to a robust positive relationship between income and well being across countries and over time, so some researchers have pushed forward what they call “modified-Easterlin hypothesis”, that the Easterlin paradox applies, but only after certain threshold is surpassed. More clarification on the matter can be found in the footnotes of the second page in the Stevenson & Wolfers paper:
We should note that the term “modified-Easterlin hypothesis” is something of a misnomer, as Easterlin himself is not among those claiming a satiation point. Instead, Easterlin and Sawangfa (2009) make the even stronger claim rising aggregate income is not associated with rising subjective well-being at any level of income. While incorrect, it is not uncommon, however, to attribute the “modified Eaterlin hypothesis” to Easterlin, and indeed, his citation for the IZA Prize says that: “Societies with higher material wealth are on average more satisfied than poorer ones, but once the participation in the workforce ensures a certain level of material wealth, guaranteeing basic needs, individual as well as societal well-being as a whole are no longer increasing with a growth of economic wealth.”
The other claim Easterlin makes is that happiness rises with relative income. Easterlin pushes such view in the within-country analysis, showing that the richer showed reported higher happiness than the poorer, so the main thesis of Easterlin is that what matters for happiness is relative income, in fact this is how the Pompeuonomics post interprets Easterlin: “He makes a very important point, that after all what might matter is relative income, that is, not so much what you earn but how does it compare with your family, circle of friends and society”. For the relative income -> happiness hypothesis to hold, we should see that happiness does not increase over time (as income grows) in a given society, we should see more or less a horizontal line (like the previous graph but for each country, or at least above a certain level, as the modified Easterlin hypothesis suggests) for Figure 2 of the Stevenson & Wolf paper, which we do not see.
This might be explained by the relative income -> happiness hypothesis by pointing to the increase in inequality, but the countries analyzed all exhibit substantially different evolutions of inequality [also, global inequality has fallen, see Xavier Sala-i-Martín (2006)] and if only relative income matters, we should still see that up to a point or always, that lines are horizontal, because the relative position implies a zero sum view at a country aggregate level, the more reported happiness by the richest because they perceive themselves to have a higher income than others would be balanced by a lower happiness reported from lower income people who would now see themselves in a lower relative position, or at least, we should see a significant decrease in the slope’s since the happiness effect due to an increase in absolute (not relative) income would disappear once the threshold has been surpassed distorting the relation found by Stevenson & Wolfers. This is not what the graph shows, therefore the relation between happiness and income holds, no threshold is found.
Another interesting fact brought up in the post, is the fact that spending money on others, can cause one’s happiness to increase more than if one spends it on oneself. At first sight, this fact seems to contradict the relative income -> happiness hypothesis since giving one’s money to others decreases one’s relative position. The fact that we care about others does not seem to fit well with the relative income à happiness hypothesis. Thus, Stevenson & Wolfers do not malinterpret Easterlin and they do not find evidence for both Easterlin’s claims and the modified Easterlin hypothesis.
Social Manipulation & Methodological Individualism
“ Is happiness something that a person decides by herself, bringing it to an oversimplistic reductionist view, or have aspirations been influenced by social manipulation by means of the business system, as explained on the article? Omitting answering to this question is very consistent with the principle of methodological individualism which is the framework in which neoclassical economists have been operating by negating a macroeconomics holistic”.
Omitting the answer to this question does not show anything about Stevenson & Wolfers operating under certain methodological approaches. It shows nothing since that question is not what they intended to answer with their research; they only intended to look out for empirical evidence of the Easterlin paradox. To answer such question one must pursue further research. Since this was not what Stevenson & Wolf intended to research, it can NOT be an objection to their work.
Irrelevance due to the limits to growth
This last objection can’t be an objection to the Stevenson & Wolfers paper, only if one wants to derive policy from Stevenson & Wolfers results (such as establishing the objective of increasing income or redistributive policies due to the diminishing returns of income, although for this last policy, a cardinal unit of happiness would be needed. This would face the same problems and is likely to be impossible, just as the construction of a cardinal utility function. One also needs to assume that the government’s aim should be to increase happiness and it has the right to pursue the actions necessary to achieve that objective, those are controversial moral and political premises). Their paper, does not explicitly propose any policy, their work remains strictly in the realm of positive research, so to attach a certain ideology to Stevenson & Wolfers is at the very least misleading, and even if they did so, this would not invalidate the findings that compromise the Easterlin paradox, only that certain policies might not follow from those findings.
I find the criticisms to the Stevenson & Wolfers paper and also the claims of unscientificness of the economics profession (difficult to see how one can prove such claim with the analysis of only one paper, does one observation make the whole sample then?) void of substance. Maybe it’s time to bury the Easterlin paradox until further research shows up. This is not to say that relative income may not matter, but is not the only thing that matters, absolute income is also part of the equation and a threshold for such effect is not found (see conclusions of the Stevenson & Wolfers paper for clarification and a mention of Kahneman & Deaton). The only thing I think we should be skeptic is about the whole “happyonomics” foundations, specially the possibility of aggregating happiness between individuals as McCloseky (2012) notes.
Easterlin, Richard A. “Does economic growth improve the human lot? Some empirical evidence.” In Nations and Households in Economic Growth: Essays in Honor of Moses Abramowitz, by Paul A David and Melvin W. Reder. New York: Academic Press, Inc., 1974. http://huwdixon.org/teaching/cei/Easterlin1974.pdf
Easterlin, Richard A., et al. “The happiness–income paradox revisited.”Proceedings of the National Academy of Sciences 107.52 (2010): 22463-22468. http://www.pnas.org/content/107/52/22463.full.pdf+html
McCloskey, Neirdre. “Happyism: The creepy new economics of pleasure.” The New Republic. June 8, 2012 http://www.newrepublic.com/article/politics/magazine/103952/happyism-deirdre-mccloskey-economics-happiness
Sacks, Daniel W., Betsey Stevenson, and Justin Wolfers. Growth in income and subjective well-being over time. mimeo, 2013. http://126.96.36.199/mfile/files/Jurnal/Jurnal%202012-2013/Growth%20in%20Income%20and%20subjective%20well-being%20over%20time.pdf
Sala-i-Martin, Xavier. “The world distribution of income: falling poverty and… convergence, period.” The Quarterly Journal of Economics (2006): 351-397. http://www.columbia.edu/~xs23/papers/pdfs/qjec.2006.121.2.pdf
Stevenson, Betsey, and Justin Wolfers. Economic growth and subjective well-being: Reassessing the Easterlin paradox. No. w14282. National Bureau of Economic Research, 2008. http://www.nber.org/papers/w14282.pdf
Stevenson, B. and Wolfers, J. (2013). “Subjective well-being and income: Is there any satiation?,” American Economic Review: Papers & Proceedings, 103(3): 598-604 http://users.nber.org/~jwolfers/papers/Satiation(AER).pdf
Veenhoven, Ruut, and Floris Vergunst. “The Easterlin illusion: economic growth does go with greater happiness.” (2013). http://mpra.ub.uni-muenchen.de/43983/1/MPRA_paper_43983.pdf
Author: Gordon Geko