Contrary to the deterioration hypothesis, we find that market-oriented societies have a greater aversion to unethical behavior, higher levels of trust, & are not significantly associated with lower levels of morality under any model specification

The moral costs of markets: Testing the deterioration hypothesis. Justin Callais, Colin Harris, Ben Borchard. Journal of Economic Behavior & Organization, Volume 204, December 2022, Pages 200-220. https://doi.org/10.1016/j.jebo.2022.10.007

Abstract: The expansion of markets has generated significant material benefits. Yet some worry that this increase in wealth has come at a significant moral cost. Markets may crowd out or even corrupt existing moral values, causing moral deterioration. We test this hypothesis using both fixed effects and matching methods to estimate the impact of market institutions on a society's moral values. Contrary to the deterioration hypothesis, we find that market-oriented societies have a greater aversion to unethical behavior, higher levels of trust, and are not significantly associated with lower levels of morality under any model specification. Furthermore, we find that becoming more market oriented does not cause a significant reduction in a society's moral values. Together, our results suggest that being or becoming more market oriented does not cause moral deterioration.

Introduction

The world is becoming more interconnected due in part to the expansion of markets. And while most people accept that markets generate significant material benefits, there remains debate “concerning whether the wealth that societies gain by embracing markets comes at too high a moral cost” (Storr and Choi 2019: 11). On one side of the debate is the position that markets crowd out or even corrupt existing moral values, resulting in moral deterioration (Rousseau [1754] 1984; Marx [1844] 2000; Radin 1987, 1989; Anderson 1993; Sandel 2012; Bowles 2016). On the other side is the claim that markets actually promote moral virtues, or at worst act as neutral spaces to be filled in by the existing values of the market participants (Mandeville [1714, 1732] 1988; Montesquieu [1748] 1989; Smith [1759] 1982, [1776] 1977; Friedman [1962] 2002; McCloskey 2006; Storr 2009; Storr and Choi 2019). Even though each position presents a testable hypothesis, the debate has largely remained one of philosophical conjecture. This is unsatisfying as the question of the moral costs of markets is “at root an empirical, rather than a philosophical, claim” (Storr and Choi 2019: 12). Do markets cause moral deterioration?

The most thorough attempt at addressing this question empirically comes from Storr and Choi (2019). Their aim is to assess the moral character of markets and the impact that market institutions have on morality, and in doing so provide a plausible retort to both the market's most stringent critics and its most tepid supporters.1 A major contribution of their work is in identifying empirical measures which either directly or indirectly measure morality.2 Indirect measures include variables like income, life expectancy, and infant mortality. These variables cannot directly address the empirical claim that markets cause moral deterioration, yet nonetheless have moral significance. If markets made people worse off on a variety of important margins, markets may be morally suspect regardless of their effect on moral values.

The direct measures capture moral values related to trust, tolerance, materialism, fairness, altruism, and the acceptance of unethical behavior using nationally representative surveys or behavioral variables like how much (and often) people give to charity. These measures capture at least part of the concern critics have over the deterioration of moral values, particularly as it relates to cosmopolitanism and integrity. Sandel (2012: 6) for example, a prominent moral critic of markets, suggests that the “appropriate way” to treat and value human beings is as “persons worthy of dignity and respect, rather than as instruments of gain and objects of use.” Being trustworthy and treating others as equals is a part of it means to treat people appropriately. If market institutions cause moral deterioration, we should expect lower levels of these measured moral values in market-oriented societies.

Storr and Choi (2019: 243) find the opposite: “Rather than being incompatible with morality, markets are not only consistent with morality but seem to promote morality.” Still, their empirical approach is limited, about which they are very forthright.3 We build on the foundation of their work using panel data. We first test the markets-cause-moral-deterioration hypothesis using both fixed effects and matching methods to estimate the impact of market institutions on a society's moral values. Contrary to the deterioration hypothesis, we find that market-oriented societies have a greater aversion to unethical behavior, higher levels of trust, and are not significantly associated with lower levels of morality, regardless of which measure is used for moral values or market orientation.

The results of these tests (section 5.1 and 5.2) are comparable to Storr and Choi's (2019) in that we find that being a market-oriented society is not associated with lower levels of moral values (and in some instances it is associated with higher levels). However, fixed effects and matching methods alone cannot address the problem of endogeneity. Market-oriented societies may have higher levels of moral values and yet the expansion of markets may still deteriorate morals. Thankfully, while our move to panel data reduces the number of countries in the sample compared to Storr and Choi's (2019) cross-section, the added time component allows us to investigate how changes in market orientation impacts moral values. By looking at instances of pro-market reform, we are able to employ differencing and matching to remove time-invariant factors and balance covariates, bringing our estimates closer to causal (An and Winship 2017). With this method, we find that becoming a more market-oriented society does not cause a significant reduction in a society's moral values. Together, our results suggest that being or becoming a more market-oriented society does not cause moral deterioration.

The rest of the paper proceeds as follows. Section two outlines the central claims of the deterioration hypothesis, including the mechanisms by which deterioration is believed to occur and the main moral values likely to be affected. Section three explains our data for market orientation and moral values. Section four outlines our empirical strategy for testing the deterioration hypothesis and section five presents our results. Section six concludes.

Men, but not women, in medium-status occupations report lower life satisfaction than those above or below when stuck in the middle

Occupational status and life satisfaction in the UK: The miserable middle? Yannis Georgellis et al. Journal of Economic Behavior & Organization, Volume 204, December 2022, Pages 509-527. https://doi.org/10.1016/j.jebo.2022.10.045

Abstract: We use British panel data to explore the link between occupational status and life satisfaction. We find puzzling evidence for men of a U-shaped relationship in cross-section data: employees in medium-status occupations report lower life satisfaction scores than those of employees in either low- or high-status occupations. This puzzle disappears in panel data: the satisfaction of any man rises as he moves up the status ladder. The culprit seems to be immobility: the miserable middle is caused by men who have always been in medium-status occupations. There is overall little evidence of a link between occupational status and life satisfaction for women, although this relationship for higher-educated women does look more like that for men.

Keywords: Occupational statusLife satisfactionOccupational mobility

JEL I31J24Z13

5. Discussion and conclusion

Our results have supplied one answer to the question of whether labour-market success paves the path to a happier life. This success can be defined in many ways, and we have here focused on occupational status (with and without controlling for labour income), as measured by the CAMSIS scale.

Occupational status is not the same thing as income, as there is (at least for men) a significant relationship between status and life satisfaction conditional on labour income;22 this status relationship is far weaker for women. Contrary to almost all of the existing literature, we allow for this occupational-status relationship to be non-monotonic, considering the correlation between life satisfaction and low-, medium- and high occupational-status jobs. In the cross-section (for men) this relationship does indeed turn out to be non-monotonic: medium-status men are less satisfied than are either those with low- or high-status jobs.

This finding might be thought to be in line with aspirations and frustration, as in the Silver-Medal effect (Medvec et al., 1995), where the dominant counterfactual for Silver Medallists is the Gold Medal, whereas that of the Bronze Medallist is no medal at all. In our context, individuals who move from low- to middle-status jobs may be frustrated not to have made the move up to high-status jobs. On the contrary, men who have always been low status do not experience this kind of frustration.

The data that we analyse is panel, and a number of individuals change from jobs with one occupational status to another over time. This allows us to address the Silver-Medal effect: Are men who rise from low to middle status less satisfied (as in the cross-section)? The panel estimation results tell a different story: following the same individual over time, those who move up the occupational-status ladder report higher levels of life satisfaction. Rather than frustration, moving up the status ladder produces satisfaction.

The panel and cross-section results are therefore contradictory. One interpretation of the perhaps surprising cross-section finding is that middle-status jobs have more unobserved unattractive features than do low-status or high-status jobs, and so are genuinely less attractive. But our panel findings of a positive relationship between life satisfaction and status run contrary to this argument. Another possibility is that there is something about individuals in medium-status jobs, rather than the characteristics of the jobs themselves: the men who end up in middle-status jobs may well be “unhappy types”. We check this by splitting the sample up into individuals who never change occupational status group and those who do change, and re-estimating cross-section regressions. The results are clear-cut: medium-status men who are (status-) mobile are more satisfied than low-status men; immobile medium-status men are less satisfied than low-status men.

The miserable middle does not then reflect the experience of men who move through medium-status jobs, but rather the experience of men who never leave them. Immobility in the middle is worse than immobility in low- or high-status jobs. This might be thought of as a dynamic version of the Silver-Medal hypothesis: these men have the Silver occupational medal, and even over time will never have any other type.

Following on from the heterogeneity analysis in Section 4.3, it is of interest to reflect on why some workers are observed to be immobile. Two possible barriers to job movements are family commitments and labour-market conditions. We have investigated the role of these barriers by first estimating separate regressions for male parents and male non-parents, finding only a very slight difference between the two, and no difference in the male cross-section status coefficients according to home-ownership. On the contrary, there does seem to be a role for labour-market health, with the miserable middle for men only being found in high-unemployment regions (see Table B10 in Online Appendix B).23 As such, the immobility that seems to lie behind our main results suggest that labour-market health may play a role in producing this specific type of immobility.

It is worth underlining the differences we find in the labour market between men and women. There is first a positive relationship between earnings and life satisfaction for both sexes, but one that is larger in size for men than for women. At the same time, there is little association between occupational status and life satisfaction for women, while that for men is U-shaped in the cross-section and positive in panel data. In the panel results, men gain more from both status and labour income than do women. This is consistent with men being more comparison-sensitive than women in the labour market, and with the notion of labour-market success being affected by social norms. Fortin (2005) notes that, across most OECD countries, these norms reinforce women's role as homemakers and men's role as breadwinners, corresponding to the observed patterns in labour-force participation. In this context, the stronger correlation between occupational status and life satisfaction for men may be unsurprising. There is however heterogeneity in this sex difference, and higher-educated women look more like men in terms of the relationship between occupational status and life satisfaction. With the sharp rise in women's education (which now exceeds that of men),24 occupational status may be poised to play an increasingly important role in the subjective well-being of all workers.