The policy implications of findings in happiness studies partly depend on our understanding of the reasons for these findings. Most economists focus on the important role of relative standing. Thus, in his seminal paper, Easterlin (1974) uses it to explain the failure of happiness to increase. Once over the subsistence level, happiness depends more on relative than absolute levels of incomes, consumption, or other objective variables.Footnote 1 In fact, Knight et al. (2013) show that relative income was at least twice as important for individual happiness as absolute income, even in rural China where people were barely above the subsistence level. Studies in the West show less dramatic results but still have relative incomes at least half as important as absolute incomes. Frank (1999), Ireland (1998), and others correctly draw the conclusion that this implies very high corrective taxes on incomes (Cf. Luo et al. 2018). In fact, all of the income taxes of most countries could be justified as corrective taxes on the relative competition effects alone. If we add the very substantial environmental costs of most production and consumption, general taxation is likely to be below its optimally efficient level. Rather than imposing excess burdens or distortionary costs, taxes are corrective and this efficiency gain could be increased by increasing tax rates! This would stand traditional public economics on this aspect on its head! (Yan et al. 2021.)

Another important factor accounting for the findings in happiness that appear to be inconsistent with traditional presumption in economics is the importance of adaptation effects (people get used to their standard of living) and the underestimation of these effects by most if not all people.Footnote 2 There are also a host of other results indicating that individuals are far from being perfectly rational. Moreover, this imperfection does not just cause random biases on either side. Rather, partly due to the nature of accumulation instinct and partly to nurture (peer pressure and commercial advertising), there is a consistent bias towards excessive materialism (Ng 2003). A distinction between utility (representing preference) and welfare (happiness) is thus necessary (Chap. 2). Traditional welfare economics and cost–benefit analysis are based on individual preferences or willingness to pay. When individual preference and welfare systematically diverge, adjustments may be necessary even in the absence of external effects.

A cost–benefit analysis aiming to maximize happiness (or welfare; the two terms are used interchangeably) may be quite different from one aiming at net-income or even preference maximization (including Pareto optimality in the sense of preference). For example, if certain protection measure is shown to cost the economy more than the total wages of the protected workers, most economists regard this as more than conclusive proof that the measure is inefficient. However, while this is inefficient in terms of income maximization, it needs not be inefficient in terms of happiness maximization. The protective measure may well still be inefficient in happiness terms if unemployment will only increase very temporarily without the measure. If displaced workers could get alternative jobs quickly, no protection is usually the best choice. However, if prolonged unemployment will be involved, it may be worth spending more than the total wages to protect the jobs. This is so because of two results in happiness studies: 1. Unemployment causes a lot of unhappiness which is ways beyond the mere losses in incomes (Winkelmann and Winkelmann 1998; see also Sect. 10.3); 2. At least for rich countries, more incomes no longer contribute to happiness at the social level in the long run (Chap. 7). It may thus make sense to spend a lot of money at negligible marginal happiness to protect jobs that are important for happiness. Economists’ case against protection has to rest more with the working of the market to make unemployment temporary. This also implies the higher importance of job search or transition assistance.

Consider issues like accidents/risk/security, health care, and the value of life. Modern economics has replaced incomes-based analysis with willingness-to-pay or preference-based analysis. If we ignore possible concern for others, ignorance and irrational preferences, individual preference and welfare coincide and analysis based on preference and that based on welfare are equivalent at the individual level. At the social level, we then only have to take into account external effects and issues of equality that economists are already familiar with. However, recent results in happiness studies and behavioural economics suggest that individual choices often involve imperfect information and/or imperfect rationality and this is so especially for choices involving the future and changes in small probabilities. In combination with the basis of excessive materialism/consumerism mentioned above, this may make people engage in excessive competition/consumption (to the detriment of health and family life), under-save for old age (making compulsory superannuation possibly sensible and actually practised in many countries), and also a host of other welfare-reducing activities not warranted even at the individual level (before counting the additional external costs through the environmental and relative-income effects). Thus, asking people’s willingness to pay for a marginal increase in safety or a slight reduction in the risk of a fatal accident may not give reliable values of life in the sense of true expected welfare maximization.

On the one hand, such figures may be under-reported due to people’s pressure for present consumption and hence under-rating the willingness to pay for safety. On the other hand, these figures may be over-reported due to the innate irrational fear of death (which has clear selection value but may be inconsistent with welfare maximization). Thus, it is desirable to supplement the willingness-to-pay studies with happiness studies. For expected welfare maximization,Footnote 3 the value of life should equal the total happiness of remaining life divided by the true happiness value of a marginal dollar. Since this latter value is likely to be very low (if still positive) even just taking into account the adaptation effect alone, the correct value of life may be very large even at the individual level. At the social level, this is even more so, as the true happiness value of a marginal dollar has to be further discounted by environmental effects and the mutually offsetting relative-income effects. This may partly explain the very large sums of money some decision makers at the social level are willing to spend to improve safety that most economists regard as many times beyond the efficient levels. While there may be some efficiency problems at the public decision-making level, it may also be the case that economists should revise their analysis to be more consistent with welfare maximization.

A particular area where very substantial adjustments are necessary is public spending. Since government spending on public goods has to be financed from taxation and government spending may involve some unavoidable inefficiency, most economists emphasize the excess burden of taxation and are in favour of lowering government spending. Their favorite slogan is: Big society, small government. This position ignores the probably greater inefficiency of private production and consumption due to unaccounted-for environmental costs and the mutually cancelling effects (at the social level) of competition for relative standing. In addition, if additional private consumption no longer contributes to happiness at the social level, even if the monetary costs of public spending are very high, but the ultimate happiness costs may be zero. Thus, provided that the relevant items of public spending do contribute to happiness ultimately, they may still be worth the high costs and some inefficiency in public spending. In addition, there are also arguments (Kaplow 1996; Ng 2000) that, even if we ignore these factors, the spending side tends to produce offsetting effects to the excess burden on the taxation side, as discussed in the next chapter.

If individual choices are subject to high degrees of informational imperfection and irrationality, does that mean that central planning is better than the market economy? Our experience gives a clear and resounding ‘No!’. Soviet Union, Eastern European countries, and China all adopted central planning without success. Central planners are also subject to imperfections in knowledge, rationality, and motivation. Much government interventions especially nationalization in Western countries also do not give an encouraging experience. We should not go back to central planning and use inappropriate government intervention. These paternalistic measures not only usually less efficient than the imperfect private market, they also temper with individual freedom excessively. Nevertheless, while the government should not be too paternalistic, some milder degree of so-called ‘soft paternalism’ or libertarian paternalism may work to revise the mistakes of imperfect private decisions. This is advocated by, among others, Richard Thaler (Nobel laureate in economics in 2017) and his collaborator Cass Sustein, including in their joint book called Nudge (Thaler and Sustein 2009; updated edition). See also: Camerer et al. (2003), Lambert (2017); for criticism/evaluation of soft paternalism, see: Qizilbash (2012), Rebonato (2014), Fumagalli (2016), Epstein (2018); for welfare-improving measures beyond soft paternalism, see Bhargave and Loewenstein (2015).

One very successful soft-paternalistic policy is nudging people to have more savings. Under saving for retirement while in prime working ages is an important problem in many Western countries. In the U.S. Congress, the rightist conservatives and the leftist liberals united to push through the policy of ‘Save more tomorrow’. If you ask people to save more right now, resistance is very high. Instead, people are requested to agree now that, when their salaries increase next, they will save more a proportion of the increase. This does not require people to cut down consumption and hence wins much higher rates of acceptance. Apart from the U.S., other countries including the U.K. and South Korea also accept some soft-paternalistic policies.

Another mild policy of restriction is for countries/states where legal gambling is allowed, including Singapore and some states in the U.S., to allow people to apply for restricting themselves or family members from entering casinos. In the State of Michigan in the U.S., the person who first applied for such a restriction on himself was also the first to be arrested for violating that restriction. After signing to restrict himself, he could not control himself and sneaked into a casino to play blackjack. After being discovered, he was not only fined but his winning of more than a thousand U.S. dollars was also confiscated.

Why do people restrict themselves? Studies show that emotion and reasoning are handled in different parts in our brain. While one is calmly reasoning, one may realize that gambling is no good for oneself. However, when one is emotionally attracted, reasoning may not be able to control your emotion.

If people have demand for restricting themselves, why could such restriction not be supplied by the market? This is so because, in the absence of the power of enforcement by the government, the emotional self will cancel the restriction by the rational self. For example, in the eighteenth century, the romantic poet Samuel Taylor Coleridge of the U.K. employed workers to prevent him to go into opium-smoking houses. However, when he wanted to get in, he would warn the workers that, if they prevented him, he would call the police to arrest them.

Due to the widespread existence of laziness, procrastination, vulnerability to temptations, and many other reasons of imperfect rationality, problems like obesity, alcohol poisoning, inadequate savings, etc. are prevalent. Thus, if excessively illiberal policies are avoided, some soft-paternalistic measures may be considered to help people to become healthier and happier. One simple policy is to impose much higher taxes on those demerit goods like cigarettes, soft drinks and candies to reduce their consumption. Apart from this consumption reduction effects, the extra revenues generated could be used for poverty reduction and environmental protection.

In the light of recent results in happiness studies, Layard (2005) has convincingly argued for the need to rethink public policy with respect to many areas including work-life balance, family life, helping the poor, eliminating high unemployment, mental illness, and community life. Here, I wish to mention a couple of areas of public spending that are more likely to increase happiness than private competitive spending.

First, in the light of the threatening effects of climate change and other environmental disruption, public actions in controlling pollution both in the form of taxing external costs and in abatement spending will likely be necessary to protect our future. Economists are familiar with the desirability in principle of taxing external effects in accordance to the damages inflicted. However, for environmental disruption that affects the long future, it is difficult if not impractical to estimate. Even if this is so, for most cases where some abatement is desirable (true for most serious environmental problems including climate change), it is desirable to tax disruption at least at the marginal cost of abatement (which is easier to estimate than the marginal damage of disruption). Such a tax will normally yield total revenue in excess of the optimal amount of abatement spending (Ng 2004). However, due to the global nature of some if not a large part of environmental disruption, international cooperation will likely be necessary. (On ways to foster international cooperation and compliance, see Ng and Liu 2003.)

If we proceed along an environmentally responsible path of growth, our great grandchildren in a century will have a real per capita income 5 to 6 times higher than our level now. Is it worth the risk of environmental disaster to disregard environmental protection now to try to grow a little faster? If this faster growth could be sustained, our great grandchildren would enjoy a real per capita income 7 to 8 times (instead of 5–6 times) higher than our level now. However, they may live in an environmentally horrible world or may well not have a chance to be born at all! The correct choice is obvious.

Easterly (1999) shows that, with economic growth, while some quality-of-life indicators increase, others decrease. Rather, it is the advance in knowledge at the world level that is more positively associated with higher values of quality-of-life indicators. In addition, as knowledge is largely a global public good with long-term benefits, it is likely to be well under-supplied. Thus, public investment in education and research to promote the advance of knowledge will most certainly yield very high benefits. I also support Layard’s (2005, Chap.13) endorsement of the appropriate use of drugs to promote happiness.Footnote 4 However, brain stimulation and genetic engineering provide far more potential, as discussed in Chap. 12. Especially for those with some understanding of economics, the case for higher public spending is pursued at a deeper level in the next chapter, taking account of the argument of Kaplow, a well-accomplished economist of the Harvard Law School.