Richard A. Easterlin, “Happiness and Economic Growth: The Evidence.” USC Dornsife Institute for New Economic Thinking, Working Paper No. 14-03, November 6, 2014.
• This paper covers much of the same ground as Easterlin (2013) – and hence this outline bears a close resemblance to the outline for the previous Easterlin article, too.
• According to the time-series data, rate of improvement in Subjective Well-Being (SWB) does not seem to depend on GDP growth, in: (1) developed countries; (2) developing countries; or (3) transition countries. Doubling the rate of GDP growth does not seem to change the growth of life satisfaction.
• In cross-country studies at a point in time, SWB growth is positively correlated with GDP growth. Combined with the lack of connection in the time-series evidence, we have the Easterlin paradox.
• For Latin American countries, even increases in financial satisfaction do not seem to rise with economic growth.
• In recent decades, China has seen unparalleled economic growth – but there is no evidence that Chinese people have gotten happier; perhaps the opposite, in fact.
• In the short run, there is a positive connection between SWB and economic growth – but there is no long-run connection. Economic growth does not buy happiness.
• Sacks, Stevenson, and Wolfers (2012) choose their time periods in ways that drive their results: they capture the short-term connection between SWB growth and GDP growth, but miss the long-term independence. Using the maximum time periods for which data are available restores the Easterlin effect.
Monday, March 28, 2016
Thursday, March 10, 2016
Florack and Sheffrin (2013) on Psychology and Taxes
Nicole E. Florack and Steven M. Sheffrin, “Psychological Non-Equivalence of Tax Bases: An Empirical Investigation.” Proceedings of the 106th National Tax Association, November 2013; available at SSRN: http://ssrn.com/abstract=2376990.
• Identical tax schedules can be framed either as wage taxes or as consumption taxes. Florack and Sheffrin ask if the framing will influence decision making, as well as preferences for one frame as opposed to the other.
• They use an internet survey to ask about: (1) the willingness to take a second job and (2) the willingness to alter prospective retirement age and current consumption in the face of higher future taxes.
• The authors find that the framing matters quite a bit. Wage taxes undermine the willingness to take on a second job to a greater extent than do equivalent consumption taxes. This asymmetry remains (actually, is amplified) even after the equivalence of the two tax schemes carefully is explained to the survey respondents.
• Retirement ages are more likely to be raised when wage taxes increase as opposed to when (equivalent) consumption taxes increase; the two tax frames also differentially affect the time path of pre- and post-retirement consumption.
• Though the two frames produce identical effects on budget sets, the longer respondents considered the issue (with respect to the second job scenario), the less likely they were to favor the wage tax framing. In the early going, the wage tax framing was more popular than the consumption tax framing. After the explanation of equivalence, the spending frame is more popular, but oddly, only about one-third of the respondents view the equivalent tax frames as, well, equivalent. The spending frame is more popular with respect to retirement decisions, too.
• The authors suggest that tax salience is driving some of the results. In particular, when deciding whether to take a second job, a wage tax seems more salient, more relevant, somehow, than a consumption tax – and hence the wage tax deters more people from taking a second job.
• The (eventual) preference for the consumption tax might be related to a sense of control: perhaps respondents think that they can avoid a consumption tax through increased saving. This view, within the framework of the model, is illusory, but perhaps not so illusory for similar taxes in the real world. The notion of self-administration of taxes (via spending decisions) is a related control-like feature.
• Identical tax schedules can be framed either as wage taxes or as consumption taxes. Florack and Sheffrin ask if the framing will influence decision making, as well as preferences for one frame as opposed to the other.
• They use an internet survey to ask about: (1) the willingness to take a second job and (2) the willingness to alter prospective retirement age and current consumption in the face of higher future taxes.
• The authors find that the framing matters quite a bit. Wage taxes undermine the willingness to take on a second job to a greater extent than do equivalent consumption taxes. This asymmetry remains (actually, is amplified) even after the equivalence of the two tax schemes carefully is explained to the survey respondents.
• Retirement ages are more likely to be raised when wage taxes increase as opposed to when (equivalent) consumption taxes increase; the two tax frames also differentially affect the time path of pre- and post-retirement consumption.
• Though the two frames produce identical effects on budget sets, the longer respondents considered the issue (with respect to the second job scenario), the less likely they were to favor the wage tax framing. In the early going, the wage tax framing was more popular than the consumption tax framing. After the explanation of equivalence, the spending frame is more popular, but oddly, only about one-third of the respondents view the equivalent tax frames as, well, equivalent. The spending frame is more popular with respect to retirement decisions, too.
• The authors suggest that tax salience is driving some of the results. In particular, when deciding whether to take a second job, a wage tax seems more salient, more relevant, somehow, than a consumption tax – and hence the wage tax deters more people from taking a second job.
• The (eventual) preference for the consumption tax might be related to a sense of control: perhaps respondents think that they can avoid a consumption tax through increased saving. This view, within the framework of the model, is illusory, but perhaps not so illusory for similar taxes in the real world. The notion of self-administration of taxes (via spending decisions) is a related control-like feature.
Busse et al. (2013) on Salience in Car Markets
Meghan R. Busse, Nicola Lacetera, Devin G. Pope, Jorge Silva-Risso, and Justin R. Sydnor. “Estimating the Effect of Salience in Wholesale and Retail Car Markets.” American Economic Review 103(3): 575-79, 2013 [pdf].
• People have limited attention, and hence, even the provision of full information might not lead to “optimal” decision making.
• Retail prices for used cars show a significant discontinuity when the mileage on those cars passes from 9,999 to 10,000 miles; similar discontinuities exist at higher multiples of 10,000, too. Apparently buyers focus on the left-most digit of the mileage, and pay less than full heed to the other digits, so they overpay for a car with 69,950 miles on it, for instance. The overpayment is significant, on the order of $300.
• Car owners respond to this bias by trading in their cars at a higher rate before they reach milestones such as 50,000 miles. That is, left-digit bias affects not only prices of used cars, but also the composition of used cars offered for sale.
• Wholesale prices for used cars show similar price discontinuities at major milestones, but these seem largely to be reflecting the retail effects of left-digit inattention.
• People have limited attention, and hence, even the provision of full information might not lead to “optimal” decision making.
• Retail prices for used cars show a significant discontinuity when the mileage on those cars passes from 9,999 to 10,000 miles; similar discontinuities exist at higher multiples of 10,000, too. Apparently buyers focus on the left-most digit of the mileage, and pay less than full heed to the other digits, so they overpay for a car with 69,950 miles on it, for instance. The overpayment is significant, on the order of $300.
• Car owners respond to this bias by trading in their cars at a higher rate before they reach milestones such as 50,000 miles. That is, left-digit bias affects not only prices of used cars, but also the composition of used cars offered for sale.
• Wholesale prices for used cars show similar price discontinuities at major milestones, but these seem largely to be reflecting the retail effects of left-digit inattention.
Levinson (2013) on Happiness and Policy
Arik Levinson, “Happiness, Behavioral Economics, and Public Policy.”
NBER Working Paper No. 19329, August 2013.
• It is hard to know the costs imposed by an increase in noise pollution or by unemployment. Happiness data offer a relatively new way to judge those costs.
• The notion is that you can examine average happiness in an area with high noise pollution, say, and, average happiness in a quiet area. If the quiet area is happier, we can ask, by how much would average income have to increase in the noisy area to bring about the same happiness level as in the quiet area (controlling for everything else, of course)? This methodology produces an estimate for the willingness-to-pay for a reduced noise environment.
• The happiness approach to valuing a non-market public good like “quiet” has many pitfalls, but perhaps adaptation (or habituation, the hedonic treadmill) is foremost among them. People adapt pretty quickly to permanent situations, so that most bad events or conditions, in the long run, do not lower subjective well-being. Does this mean that these bad events impose no costs in the long run? (The analogous issue arises for good events or conditions.)
• Temporary events, like particularly bad noise one day, should not influence evaluative measures of subjective well-being, but do affect momentary happiness, and hence should affect hedonic happiness measures. It turns out that, through projection bias, evaluative measures of long-term satisfaction actually are affected by minor, short-term events like a bad noise or weather day, or by finding a dime.
• Analysts can be careful, but in the end, we still seem to have an insoluble problem, that policy makers should care most about long-term conditions, but because of adaptation, happiness only varies with short-term conditions.
• It is hard to know the costs imposed by an increase in noise pollution or by unemployment. Happiness data offer a relatively new way to judge those costs.
• The notion is that you can examine average happiness in an area with high noise pollution, say, and, average happiness in a quiet area. If the quiet area is happier, we can ask, by how much would average income have to increase in the noisy area to bring about the same happiness level as in the quiet area (controlling for everything else, of course)? This methodology produces an estimate for the willingness-to-pay for a reduced noise environment.
• The happiness approach to valuing a non-market public good like “quiet” has many pitfalls, but perhaps adaptation (or habituation, the hedonic treadmill) is foremost among them. People adapt pretty quickly to permanent situations, so that most bad events or conditions, in the long run, do not lower subjective well-being. Does this mean that these bad events impose no costs in the long run? (The analogous issue arises for good events or conditions.)
• Temporary events, like particularly bad noise one day, should not influence evaluative measures of subjective well-being, but do affect momentary happiness, and hence should affect hedonic happiness measures. It turns out that, through projection bias, evaluative measures of long-term satisfaction actually are affected by minor, short-term events like a bad noise or weather day, or by finding a dime.
• Analysts can be careful, but in the end, we still seem to have an insoluble problem, that policy makers should care most about long-term conditions, but because of adaptation, happiness only varies with short-term conditions.
Deaton and Stone (2013) Identify “Two Happiness Puzzles”
Angus Deaton and Arthur A. Stone, “Two Happiness Puzzles.” American Economic Review 103(3): 591-97, 2013 [pdf].
• If the Easterlin Effect is correct, and if it arises from the fact that people care about relative, not absolute income, then a rise in someone else’s income hurts me. That is, individual income increases in these circumstances involve a negative externality – and perhaps such increases should be dissuaded by a Pigovian tax, as with other negative externalities.
• Deaton and Stone can find little or no evidence for the Easterlin Effect either within the US or across countries, however – this is their first happiness puzzle. Their non-result is especially strong when using an evaluative measure of happiness – the rung (0 to 10) of a ladder representing quality of life – as opposed to a hedonic measure, based on reported happiness (reported today concerning yesterday). Indeed, there is some evidence that living in the same zip code as rich people raises average happiness. For Deaton and Stone, universal increases in income will increase happiness markedly.
• Religious individuals tend to report higher evaluative happiness – but more religious countries or states do not appear to be happier, even (in cross country data, though not in the US) after controlling for average income. This aggregation inconsistency is the second happiness puzzle, and it applies only to evaluative happiness, not hedonic happiness. (More religious countries report higher average hedonic happiness, but not higher evaluative happiness.) In the US, religious states also seem to have worse social problems, like crime and ill health. One speculative explanation is that social problems tend to cause people to be religious, and as the problems are ameliorated, religiosity declines.
• If the Easterlin Effect is correct, and if it arises from the fact that people care about relative, not absolute income, then a rise in someone else’s income hurts me. That is, individual income increases in these circumstances involve a negative externality – and perhaps such increases should be dissuaded by a Pigovian tax, as with other negative externalities.
• Deaton and Stone can find little or no evidence for the Easterlin Effect either within the US or across countries, however – this is their first happiness puzzle. Their non-result is especially strong when using an evaluative measure of happiness – the rung (0 to 10) of a ladder representing quality of life – as opposed to a hedonic measure, based on reported happiness (reported today concerning yesterday). Indeed, there is some evidence that living in the same zip code as rich people raises average happiness. For Deaton and Stone, universal increases in income will increase happiness markedly.
• Religious individuals tend to report higher evaluative happiness – but more religious countries or states do not appear to be happier, even (in cross country data, though not in the US) after controlling for average income. This aggregation inconsistency is the second happiness puzzle, and it applies only to evaluative happiness, not hedonic happiness. (More religious countries report higher average hedonic happiness, but not higher evaluative happiness.) In the US, religious states also seem to have worse social problems, like crime and ill health. One speculative explanation is that social problems tend to cause people to be religious, and as the problems are ameliorated, religiosity declines.
Easterlin Continues to be Paradoxical (2013)
Richard A. Easterlin, “Happiness, Growth, and Public Policy.” Economic
Inquiry 51(1): 1–15, January 2013 [pdf].
• Sacks, Stevenson, and Wolfers (2012) choose their time periods in ways that drive their results; using the maximum time periods available restores the Easterlin effect.
• Rate of improvement in subjective well-being (SWB) does not seem to depend on GDP growth, in: (1) developed countries; (2) developing countries; and (3) transition countries.
• For developing countries, even financial satisfaction does not seem to rise with economic growth.
• In recent decades, China has seen unparalleled economic growth – but there is no evidence that Chinese people have gotten happier; perhaps the opposite, in fact.
• In the short run, there is a positive connection between SWB and economic growth – but there is no long-run connection. Economic growth does not buy happiness.
• Full employment and the provision of a social safety net raise SWB.
• Germany pioneered safety nets in the 1880s; its experience suggests that today’s developing countries, for the most part, can afford social safety nets, too.
• Sacks, Stevenson, and Wolfers (2012) choose their time periods in ways that drive their results; using the maximum time periods available restores the Easterlin effect.
• Rate of improvement in subjective well-being (SWB) does not seem to depend on GDP growth, in: (1) developed countries; (2) developing countries; and (3) transition countries.
• For developing countries, even financial satisfaction does not seem to rise with economic growth.
• In recent decades, China has seen unparalleled economic growth – but there is no evidence that Chinese people have gotten happier; perhaps the opposite, in fact.
• In the short run, there is a positive connection between SWB and economic growth – but there is no long-run connection. Economic growth does not buy happiness.
• Full employment and the provision of a social safety net raise SWB.
• Germany pioneered safety nets in the 1880s; its experience suggests that today’s developing countries, for the most part, can afford social safety nets, too.
Wednesday, March 9, 2016
Datta and Mullainathan (2014) on “Behavioral Design”
Saugato Datta and Sendhil Mullainathan, “Behavioral Design: A New Approach to Development Policy.” Review of Income and Wealth 60(1): 7–35, March 2014.
• Behavioral economics can help us design development policies that will work. There are no guarantees, of course – much of what we have learned comes from small-scale field experiments, and their external validity and ability to scale are not yet proven.
• The poor are like the rest of us, though their poverty adds to their psychological burdens.
• A farmer’s failure to use fertilizer might not arise from an underlying preference not to use it; indeed, it can happen even where there is a desire to use fertilizer. People procrastinate, and they do so repeatedly; people also lack self-control to resist temptations, including temptations to use resources for things other than fertilizer. As a result, the timely provision of fertilizer at a small subsidy can have significant impacts on usage; likewise, a commitment savings plan can spur fertilizer use.
• Four types of mental resources are scarce: (1) self-control; (2) attention; (3) cognitive capacity; and (4) understanding. More information or even financial subsidies may not resolve the problems created by these scarce resources.
• Datta and Mullainathan argue that development problems might be fairly situation-specific, and that the binding constraint or constraints might implicate behavioral issues. (Compare with Dani Rodrik's work (with co-authors) on "growth diagnostics".) Datta and Mullainathan suggest a behavioral mapping of problems that reveals where these constraints (bottlenecks) might bind. Commitment savings accounts, steady income flows, appropriate defaults, reward lotteries, and reminders: these are some strategies that can help to overcome behavioral bottlenecks.
• Behavioral economics can help us design development policies that will work. There are no guarantees, of course – much of what we have learned comes from small-scale field experiments, and their external validity and ability to scale are not yet proven.
• The poor are like the rest of us, though their poverty adds to their psychological burdens.
• A farmer’s failure to use fertilizer might not arise from an underlying preference not to use it; indeed, it can happen even where there is a desire to use fertilizer. People procrastinate, and they do so repeatedly; people also lack self-control to resist temptations, including temptations to use resources for things other than fertilizer. As a result, the timely provision of fertilizer at a small subsidy can have significant impacts on usage; likewise, a commitment savings plan can spur fertilizer use.
• Four types of mental resources are scarce: (1) self-control; (2) attention; (3) cognitive capacity; and (4) understanding. More information or even financial subsidies may not resolve the problems created by these scarce resources.
• Datta and Mullainathan argue that development problems might be fairly situation-specific, and that the binding constraint or constraints might implicate behavioral issues. (Compare with Dani Rodrik's work (with co-authors) on "growth diagnostics".) Datta and Mullainathan suggest a behavioral mapping of problems that reveals where these constraints (bottlenecks) might bind. Commitment savings accounts, steady income flows, appropriate defaults, reward lotteries, and reminders: these are some strategies that can help to overcome behavioral bottlenecks.
Thursday, March 3, 2016
Sandel (2013) on Markets and Morals
Michael J. Sandel, “Market Reasoning as Moral Reasoning: Why Economists Should Re-engage with Political Philosophy.” Journal of Economic Perspectives 27(4): 121-140, Fall 2013.
• Putting goods into the market domain can change their nature; therefore, economics cannot avoid ethics.
• Does an increase in efficiency make a society better off? The answer depends on one’s view of what constitutes the social good. Economists tend to implicitly adopt a utilitarian approach. Selling slots in universities, or votes, or babies, might promote efficiency, but might nonetheless be morally objectionable.
• Shouldn’t we be wary of the “voluntary trade is mutually beneficial” claim? Unequal starting points might imply coercion. Are there societal conditions that are not consistent with the provision of meaningful consent to a transaction?
• Markets might corrupt goods or crowd out other types of incentives. People might do things voluntarily that they would be unwilling to do for payment.
• Selling spots to attend Congressional hearings would demean the institution. It would be better to distribute tickets by an online lottery, and make the tickets non-transferable.
• Fines indicate social disapproval; they are not equivalent to fees, and people do not like it when intentional offenders treat fines as fees.
• The Intuits in Canada receive a quota for killing walruses – an exemption that recognizes the longstanding role of walrus hunting in the Intuit civilization. The sale by the Intuits of some of that quota to “sport” hunters corrupts the meaning of the exemption. A further issue is whether society need respect the perverse preference of hunters to shoot defenseless walruses.
• Are love and benevolence non-renewable, depletable resources, or are they augmented with use?
• Putting goods into the market domain can change their nature; therefore, economics cannot avoid ethics.
• Does an increase in efficiency make a society better off? The answer depends on one’s view of what constitutes the social good. Economists tend to implicitly adopt a utilitarian approach. Selling slots in universities, or votes, or babies, might promote efficiency, but might nonetheless be morally objectionable.
• Shouldn’t we be wary of the “voluntary trade is mutually beneficial” claim? Unequal starting points might imply coercion. Are there societal conditions that are not consistent with the provision of meaningful consent to a transaction?
• Markets might corrupt goods or crowd out other types of incentives. People might do things voluntarily that they would be unwilling to do for payment.
• Selling spots to attend Congressional hearings would demean the institution. It would be better to distribute tickets by an online lottery, and make the tickets non-transferable.
• Fines indicate social disapproval; they are not equivalent to fees, and people do not like it when intentional offenders treat fines as fees.
• The Intuits in Canada receive a quota for killing walruses – an exemption that recognizes the longstanding role of walrus hunting in the Intuit civilization. The sale by the Intuits of some of that quota to “sport” hunters corrupts the meaning of the exemption. A further issue is whether society need respect the perverse preference of hunters to shoot defenseless walruses.
• Are love and benevolence non-renewable, depletable resources, or are they augmented with use?
Roth (2007) on Repugnance
Alvin E. Roth, “Repugnance as a Constraint on Markets.” Journal of Economic Perspectives 21(3): 37-58, Summer 2007.
• Repugnance often presents a real limit on what types of markets can be implemented; in this sense, repugnance is like resource and incentive constraints.
• Arguments focusing on the gains to trade have little traction against repugnance of kidney sales.
• Sometimes repugnance is combined with concerns about negative externalities, addiction, or hidden coercion.
• Closely related activities can differ substantially on the repugnance scale: dwarf-tossing (often repugnant) v. wife-carrying (often acceptable). The set of transactions that are viewed as repugnant changes over time. [Maybe those kidney sales are becoming less repugnant, too.]
• Sometimes the transactions themselves, as gifts, are not repugnant, but adding money makes them so; many people who object to kidney sales find kidney donation to be admirable.
• Repugnance often presents a real limit on what types of markets can be implemented; in this sense, repugnance is like resource and incentive constraints.
• Arguments focusing on the gains to trade have little traction against repugnance of kidney sales.
• Sometimes repugnance is combined with concerns about negative externalities, addiction, or hidden coercion.
• Closely related activities can differ substantially on the repugnance scale: dwarf-tossing (often repugnant) v. wife-carrying (often acceptable). The set of transactions that are viewed as repugnant changes over time. [Maybe those kidney sales are becoming less repugnant, too.]
• Sometimes the transactions themselves, as gifts, are not repugnant, but adding money makes them so; many people who object to kidney sales find kidney donation to be admirable.
Subscribe to:
Posts (Atom)