Comparing Work Ethic and Success between Protestants and Catholics, Slides of Religion

Download Comparing Work Ethic and Success between Protestants and Catholics and more Slides Religion in PDF only on Docsity!

Barcelona Economics Working Paper Series

Working Paper nº 497

Protestants and Catholics: Similar Work

Ethic, Different Social Ethic

Benito Arruñada

September 2010

Benito Arruñada∗

Protestants and Catholics:

Similar Work Ethic, Different Social Ethic**

The Economic Journal , 2010, 120 (547), 890-918.

Abstract

This article develops two hypotheses about economically-relevant values of Christian believers, according to which Protestants should work more and more effectively, as in the “work ethic” argument of Max Weber, or display a stronger “social ethic” that would lead them to monitor each other’s conduct, support political and legal institutions and hold more homogeneous values. Tests using current survey data confirm substantial partial correlations and possible different “effects” in mutual social control, institutional performance and homogeneity of values but no difference in work ethics. Protestantism therefore seems conducive to capitalist economic development, not by the direct psychological route of the Weberian work ethic but rather by promoting an alternative social ethic that facilitates impersonal trade.

JEL Classification: D23, E0, N4, O39, Z Keywords: Religion, values, Weber, institutions, enforcement.

∗ (^) Pompeu Fabra University and Barcelona GSE. Trias Fargas, 25. 08005-Barcelona (Spain). E-

mail: benito.arrunada@upf.edu. ** (^) Preliminary versions of this paper were circulated under the title “The Economic Effects of

Christian Moralities”. The author acknowledges the comments received from Doug Allen, Manuel Bagüés, Robert Barro, Lee Benham, Albert Carreras, Nuno Garoupa, Dean Lueck, Phillip E. Keefer, Rachel McCleary, Debin Ma, Eduardo Melero, Lewis S. Mudge, Andrew Scott, Paul Zak, Luigi Zingales, three referees and seminar participants at Berkeley, FAES, Harvard, Instituto de Empresa, ISNIE, UAM, UPF and Washington University, as well as the research assistance of Petar Balachev. The usual disclaimers apply. The author acknowledges financial support from the Barcelona GSE Research Network and the Government of Catalonia; the Spanish Ministry of Science and Innovation, through grant ECO2008-01116; and the European Commission, through the Integrated Project “Reflexive Governance in the Public Interest”. The data utilized were documented and made available by the Zentralarchiv für Empirische Sozialforschung, University of Cologne, and were collected by independent institutions in each country. Neither the original collectors nor the Zentralarchiv bear any responsibility for the analyses or interpretation presented here.

social control, more supportive of institutions, less bound to close circles of family and friends and to hold more homogeneous values.

These results are in line with Glaeser and Glendon (1998) who, after modelling the incentives in Calvinist predestination, confirm empirically that Catholicism and Protestantism associate with different social interactions; and with Guiso, Sapienza and Zingales (2003), who also find that Catholicism and Protestantism offer pros and cons with respect to economic attitudes. The argument is partly similar to that of Becker and Woessmann (2009), who argue that Protestant regions grew faster because Protestant emphasis on reading the Bible led, as a side effect, to greater investment in literacy and human capital. My argument is also one of side effects but related to the development of a social ethic that favoured market transactions and market- enhancing institutions.^1

The present work differs from, and complements, previous works in several important ways. First, some previous works supply heterogeneous empirical correlations between religion and economic performance or attitudes. As a consequence, their findings are hard to evaluate whereas the empirical tests in this article are developed from an analytical framework that allows a set of testable hypotheses to be drawn up, painting a more systematic and theoretical picture of the links between alleged causes and effects, which comes closer to examining cultural innovation instead of taking culture as given. Second, the effect of religion is estimated not only on values but also on personal outcomes such as working hours, education and personal success. Third, using data from the 1998 International Social Survey Programme (ISSP) surveys conducted in 1998 and 1999 in 32 countries allows estimation of the effects of religion when strong believers allegedly played a greater role in society.

The rest of the article is structured as follows. Section 1 develops a framework to analyze the economic impact of religions, based on a typology of enforcement systems, applies it to the changes introduced by the Protestant Reformation and defines on this basis two testable hypotheses according to which Protestantism favours a more productive work ethic or a stronger social ethic that leads Protestants to exert greater social control, support the rule of law more and hold more homogeneous values. Section 2 describes the ISSP data, the specific variables employed and the statistical regressions used to test the hypotheses. Section 3 presents and discusses the results, according to which Protestantism promotes values favourable to capitalist relations based on impersonal trade, with no perceptible impact on the work ethic. Section 4 concludes.

(^1) Ekelund et al. (2006: 189-231) discuss other possible reasons, such as simpler churches and

liturgies, and fewer holidays and pilgrimages, as to why the Protestant Reformation may have affected economic growth.

1. Analytical Framework

1.1. The Effect of Religion on Enforcement and Growth

In all societies, individual behaviour is constrained by norms and rules that humans define and enforce by different means. Religion is one of these means. To analyze the effects of Christian religions, I will distinguish three types of structures according to which party is responsible for enforcing the more or less implicit terms of exchange in a given interaction. Under “first party” enforcement, the obliged individuals evaluate their own conduct in relation to their own reading of a moral code, a code that includes many economically-relevant preferences, towards effort, thriftiness, and so on. In accordance with this evaluation, individuals sanction themselves with some psychological compensation, which in Christianity is related to the idea of “salvation” and eternal life in heaven. “Second party” enforcement is based on verification and sanction by the party suffering the consequences of breach. In addition to partners in a standard economic exchange, peers in groups are also second parties to the extent that they exert pressure on noncompliant members through diverse means, from shaming to ostracizing or even killing them. Lastly, under “third party” enforcement, more or less specialized agents, such as political rulers, judges and police forces, verify the behaviour of group members and punish those who do not follow the rules.

Religion obviously influences first party enforcement. For example, relative to the more individualistic pagan views prevalent at the time of the Roman empire, Christianity greatly reinforced first party enforcement by imposing on believers the then novel moral duty of helping their neighbours (Stark, 1996). In this Christian spirit, natural feelings of compassion, which Roman patricians were educated to suppress, were the cornerstone of Adam Smith’s moral conception in his Theory of Moral Sentiments (1759). Less obviously, religion also affects the functioning of second and third party enforcement. In primitive societies, there is even little separation between religious and civil law. In more developed theocratic regimes, religious authorities also take over or dominate political rulers, and there is little or no separation between religious and civil law. Christianity followed a long and tortuous path along these lines. It started as a minority cult whose views conflicted with those that were politically correct at the time, but became the State religion in the last centuries of the Roman empire. During the late Middle Ages, and especially on the eve of the Reformation, the Church was not only the monopoly supplier of religious services but also the main provider of educational, legal, bureaucratic and welfare services; it commanded substantial military forces; the papal state constituted a political power in itself; and churchmen were the main political officials all across Europe (Cameron, 1991).

1.2. Comparative Analysis of Christian Moral Enforcements

In this context, the Protestant Reformation radically modified both the contents of moral rules and the enforcement mechanisms of moral and civil rules alike. The Reformation directly affected the three enforcement systems:

neighbours’ opinions on them. Mutual social control thus avoided the obvious risk that self- examination might result in more lenient standards of conduct. These consequences were more explicit in stricter communities. For example, the Geneva of Calvin adopted many intrusive social controls, such as “family visitations,” by which two elders regularly visited each home to discuss the spiritual health of each family. Another classic account of these practices in mutual control was also given by Weber in his description of American sects (1920).

Third, in contrast to the more independent Catholic Church, reformers were more supportive of political and legal institutions, often because they needed political support in their fight against Catholicism. The Medieval Church had been an international power that held very substantial wealth and limited the power of political rulers, frequently opposing them. Where the Reformation succeeded, most Church property was soon seized by political rulers, and the previous Church privileges were removed. Furthermore, lay rulers rapidly asserted their domination of religious affairs. Consequently, many Protestant churches rapidly became appendices of local or national rulers. Moreover, in contrast to the ambivalent support provided by the Catholic Church to political rulers and the nuanced advice the Church gave to the laity for dealing with rulers, reformed churches more flatly affirmed that believers had to obey their rulers, as Luther exhorted very early on (1523). Such concentration of political power may have and often did result in tyranny. However, in a similar fashion to self-control, it also tended to be restrained by more active mutual social control.

In addition, the changes introduced by the Reformation also made values more homogeneous among believers. This should lower the transaction costs of impersonal trade, the sort of trade between strangers that has been considered crucial in capitalist economic development at least since North and Thomas (1973), and by authors in different disciplines, such as Granovetter (1985) and Seabright (2004). The two religions provide markedly different setups in this regard:

Primarily, Catholic practice favours more diverse moral standards because of both the contents and its moral code and, mainly, its enforcement mechanisms. With respect to the code, its prioritization of the family encourages selective charity and reinforces in believers their natural tendency to favour their relatives and possibly friends over strangers. It therefore enhances the use of double standards. With respect to enforcement, both the theology of salvation by works and the practice of private confession of sins to a priest support heterogeneous standards. First, salvation by works involves an element of individual fine-tuning because works cannot be evaluated without considering the possibilities of each individual and moral standards are adjusted to specific circumstances. In fact, medieval theologists had minutely devised prescriptions for each case, developing the body of “casuistry” literature. Private confession also adjusted moral standards to each individual: priests were trained to adapt the moral code to the strength of the penitent, even negotiating penance with them. In addition, the theology of Purgatory made it possible for merits to be traded amongst believers and with the Church, which reinforced inequality (Arruñada, 2009). The sale of indulgences also caused greater moral disparity among believers (in addition to considerable rent-seeking, as emphasized by Ekelund et al. (1992, 1996, 2002, 2006).

Conversely, greater homogeneity in the moral standards of Protestantism derives from its emphasis on universal charity and its greater reliance on “external” sources for enforcement, both second parties and legal institutions. First, the Protestant ideal tends to place obligations to strangers on a par with those to family members (McCleary, 2007). Second, compared to the secret judgments of the confessional, reliance on external enforcement is likely to produce more

equal treatment, as modelled by Glaeser and Glendon (1998). Examples abound, starting with John Calvin’s insistence on treating all believers equally, or the community responsibility system practiced by the American sects described by Weber (1920), which is somehow similar to the late medieval system analyzed by Greif (2002). In this extreme case, as individuals are liable for the debts of their colleagues, they will insist that they meet the standards of the group, both on admission and later on. Third, more generally, legal enforcement applies the same principle on a larger scale, by providing impartial enforcement of obligations without paying attention to who the parties are and, in particular, regardless of whether they are locals or outsiders.

Summing up, the analysis supports two distinctive hypotheses. The work ethic hypothesis predicts that Protestant believers work more and more effectively than Catholics. The social ethic hypothesis predicts that Protestants show greater concern for social interactions, in terms of at least social control, rule of law and homogeneity of values.

2. Data and Tests

2.1. Data

The tests will rely on several econometric models built with cross-section data from the 1998 religion module of the International Social Survey Programme (ISSP) Survey. The ISSP is a continuing program of cross-national collaboration on surveys covering topics of importance for research in the social sciences. The surveys were conducted in 1998 and 1999 (8 countries) in a total of 32 countries, most of them developed countries, with sample sizes between 804 and 2,488, and a total of 39,034 observations. After dropping observations with missing values in the independent variables, 19,246 observations on Protestants and Catholics remain, which are distributed across countries as summarized in Table 1. In each of these national surveys, the questionnaires of the religion module included at least 72 questions about the respondents’ feelings (for instance, their happiness), values (tolerance of homosexuality, confidence on parliaments, trust in strangers and so on), religious beliefs (in heaven, in hell, etc.) and practice (church attendance), social habits (different volunteer activities), and opinions (government responsibilities), etc.; as well as a full set of demographic variables (sex, marital status, education, earnings, etc.). Detailed explanation of all variables is given in Table 6 in the Annex.

These surveys have been used in many other studies.^3 Using ISSP data instead of the World Values Survey reduces sample size and the number of countries and variables. It provides, however, a more detailed measurement of the intensity of beliefs, which allows me to estimate its impact and to distinguish the fixed effect of “belonging” to a religion from the variable effect of “believing” its doctrine more or less strongly.

(^3) Full information on the ISSP surveys is available at http://www.issp.org (accessed January 13,

2009), including data and codebooks for the religion 1998 module and a list of the thousands of studies using the ISSP data.

Church ). The hypothesis predicts that, whatever the effect of education, it will be smaller or even negative for Catholics, because most reformed churches relinquished some of their functions in moral enforcement, correspondingly empowering individuals and encouraging them to learn. Therefore, substitution between education and church enforcement was accomplished by Protestants centuries ago, while the Catholic Church has retained a more active role in moral enforcement. As a consequence, education should be a complement of religion for Protestants but a substitute for Catholics.

Lastly, by similarly testing if greater levels of education affect Religious practice differently for Catholics and Protestants, where Religious practice is an index built with four variables measuring prayer frequency, participation in church activities, self-description as a religious person and frequency of attendance at religious services. Considering the results in the literature (Iannaccone, 1998: 1470), greater education is expected to increase religious participation. However, the prediction concerns only differences between religions—in particular, it predicts that this effect will be significantly smaller for Catholics than for Protestants. The rationale is the same as for confidence in the Church.

2.3.2. Rule of Law

In terms of the rule of law, the social ethic hypothesis predicts that Protestants support political and legal institutions more than Catholics. It will be tested by examining how Protestants and Catholics compare with respect to three variables: First, their Tolerance of tax fraud. The fact that taxes are necessary for operating political and legal institutions upholds the notion that less tolerance of tax fraud is related to greater support of such institutions. In addition, the Catholic Church has been relatively lenient with respect to tax fraud, not considering it as a serious sin, perhaps as a remnant of the medieval times when it competed for tax money with civil rulers. Given the greater support historically given by Protestant churches to political rulers, it is predicted that within each country and also controlling for a full set of demographic variables, Catholics will be more tolerant of tax fraud.

Second, their willingness to Cover up for friends , measured through respondents’ reactions to the hypothetical case in which they are riding in a car driven too fast by a close friend who hits a pedestrian, and this friend asks the respondent to tell the police that he was obeying the speed limit. The variable therefore measures individuals’ unwillingness to voluntarily cooperate with the legal system and its representatives when this cooperation conflicts with friendship ties. It is assumed to indicate how much support respondents provide to third party enforcement. Also in this case, it is predicted that Catholics will be more willing to cover up for their friends.

Lastly, I will examine respondents’ confidence in political and legal institutions, measured with the index Trust institutions , which combines confidence in the parliament, and the court and legal system. The logic behind these proxies is straightforward, as Parliament establishes the legal rules and courts enforce them. Our prediction is that Catholics will be less confident of their political and legal institutions.

2.3.3. Homogeneous Values

Lastly, the social ethic hypothesis predicts that Protestants hold more homogeneous values than Catholics. It will also be examined in three ways: First, the homogeneity of moral standards

within the Protestant and Catholic religion will be directly tested by comparing the residuals of the regressions on values between the two groups, expecting a higher variance among Catholics.

Second, homogeneity of values will be tested indirectly by examining how Protestants and Catholics compare with respect to a measure of interpersonal generalized trust, given by responses to the question, “Generally speaking, would you say that people can be trusted or that you can’t be too careful in dealing with people?” ( Trust strangers ). This variable is often related to “social capital” (Putnam, 1993) and is thought to convey trust in strangers (Knack and Keefer, 1997). Its connection to homogeneity is clear if people trust strangers more in homogeneous societies, as argued by Alesina and LaFerrara (2000, 2002), an idea grounded both on social (Tajfel and Turner, 1979) and evolutionary psychology (Trivers, 1971). The prediction is that Catholics should trust strangers less than Protestants.

Complementarily, I will finally check the importance individuals give to their family, to examine the argument put forward by Putnam (1993), Fukuyama (1995) and others that there is substitution between the strength of family ties and formal institutions. Lacking direct measures, I will use as a proxy the difference in individuals’ stated tolerance of heterosexual relations before and after marriage. The prediction is that this index of Importance of family , measured as the gap between tolerance of premarital sex and adultery, will be greater for Catholics.

2.4. Models

2.4.1. Main Model

To perform most of these tests, the following equations will be estimated:

Yi = α 0 + α 0c Catholic

+ β 1 Faith + β 1c Catholic × Faith

+ β 2 Religious upbringing + β 2c CatholicR × Religious upbringing

+ β 3 Education + β 3c Catholic × Education

+ Σ t ( β t Control m variables )

+ Σ r ( β r Country n dummies ) (1)

where each dependent variable, Yi , represents a value or action, as stated by respondents to the survey, from their weekly working hours to their trust in strangers. Half of the dependent variables are expressed in terms of categories with a natural order. Ordered probit models are estimated in these cases, and ordinary least squares are used for the rest. The regressions of Working hours and Positive working hours are tobit and probit models. All equations were estimated using weight adjustments with robust (Huber-White) standard errors. Neither the choice of model nor the use of survey estimation materially affect the results.

Independent variables are as follows:

• Catholic is a binary variable that takes value one for respondents who state that they belong to the Catholic religion, zero otherwise. The default category is Protestant.

country effects but not for the intensity of their faith, upbringing and education. Estimates of these average effects are given in the first row of Panel B in Table 4.

3. Results

Table 3 presents summary statistics of the data used for the analysis. Most differences in means and standard deviations for Protestants and Catholics are consistent with the hypotheses. Table 4 presents the main results, obtained by estimating the effects of the two religions on a variety of values, conducts and facts. Catholicism is represented by a dummy binary variable, with Protestantism as the omitted category. Results for constants, control variables and country dummies are omitted but their use is supported by their statistically significant correlations with the dependent variables and the fact that most control variables show the expected signs. The Table contains two panels. Panel A shows the results for models in equation (1), with variable effects, while Panel B presents the average effects obtained with the simplified model.

3.1. The Work Ethic Hypothesis

Results hardly provide any support for the work ethics hypothesis. Despite the fact that on average Protestants work 8.5% more hours and show greater personal success, the significance of these differences disappears after controlling for demographic and country effects according to the non-significant coefficients of fixed effects in Table 4, both for the standard person in the sample (panel A) and for the standard member of each religion (panel B).

Results on working hours, a variable that measures working time at all jobs, but not time worked at home, are slightly ambiguous with respect to the differential effects of the intensity of religious belief and upbringing between Catholics and Protestants. As shown in panel A, for Catholics, stronger faith and religious upbringing are associated with fewer working hours (equation [1]). However, this result seems to be driven by the fact that the only working hours considered are those worked in jobs. Catholics are indeed less likely to report positive working hours (equation [2]). However, for those Catholics with positive working hours, belief intensity is associated with a greater number of working hours (equation [3]), a result driven fully by Catholic women. Overall, the results seem to reflect a difference between the two religions as to occupational patterns, which cannot be fully clarified without knowing how many hours the two types of respondents work outside their jobs. With the data available, the observed difference— which refers only to the effect of faith and upbringing—cannot be ascribed to a differential willingness to exert effort. It seems more likely to respond or at least be heavily influenced by different priorities in allocating time between the family and the outside world, an interpretation that fits in well with the greater importance that Catholics grant to the family (section 3.2.3).

Results for personal success are even clearer, with no significant differences being observed when controlling for demographic and country variables, neither in the fixed effect of belonging to a religion nor in the variable effects of religious belief intensity or upbringing.

3.2. The Social Ethic Hypothesis

3.2.1. Social Control

Results confirm the three social control predictions. First, average Catholics not only volunteer work significantly less than Protestants (on average, about .159 standard deviations) but volunteering increases with faith and upbringing less than half for Catholics than what it does for Protestants. Catholic values are thus more weakly linked to volunteering and therefore are allegedly less conducive to mutual control, to the extent that mutual control relies on volunteer enforcement work and many organizations based on volunteer work also act as enforcement devices by, for example, screening access to social networks.

Second, the data show an acute contrast in the relationship between education and confidence in churches and religious organizations: better-educated Protestants trust them more whereas better-educated Catholics trust them less (from a rate of .074 standard deviations of trust for each deviation in education for Protestants to -.090 deviations for Catholics). Catholics also show more confidence in their Church, and their confidence increases with their faith and religious upbringing. However, greater confidence can be interpreted in different ways.

Third, education has a similarly contrasting effect on religious practice because, even if in this case education has for Catholics a slightly positive instead of a negative effect, this attenuation probably is at least partly driven by the likely presence of a social element in the demand for religious practice, as argued by Sacerdote and Glaeser (2008).

These contrasting effects of education support the idea that for Protestants education complements religion whereas for Catholics education substitutes for religion. It therefore seems that the Catholic Church is less in tune with its more educated laity, possibly due to the greater role the Catholic Church plays as an enforcer, which conflicts more with educated laity.

3.2.2. Rule of Law

Results clearly confirm two of the three predictions according to which Protestants support political and legal institutions more than Catholics.

First, Catholics are significantly more tolerant of tax fraud than Protestants, in line with Guiso, Sapienza and Zingales (2003). Furthermore, even if strong Catholic beliefs are associated with less tolerance of tax fraud, this association is about a third weaker than among Protestants. The result therefore confirms that, as predicted, Catholic values are less supportive of those political and legal institutions that are financed with taxation.

Second, Catholics are also more willing to cover up for their delinquent friends in dealings with the police, and strong Catholic beliefs have a similarly positive but weaker effect. This result confirms that Catholic values elicit less cooperation from citizens in the functioning of legal institutions when they conflict with smaller social circles, such as the ones defined by friendship ties.

Third, no significant differences are observed between Protestants and Catholics with respect to their confidence in political and legal institutions, measured through the index of Trust institutions. Strong Catholic believers even show more confidence than strong Protestant believers, with Catholic faith showing almost twice the effect of Protestant faith, in line with

that are crucial in making cooperative strategies viable in repeated games, for instance when implementing a “tit for tat” strategy, and the correlation established between trust and economic growth (Knack and Keefer, 1997; Zak and Knack, 2001).

Third, Catholics not only give more importance than Protestants to family ties, but Catholic beliefs and upbringing are also positively related to the proxy of family importance, which is in line with the argument of Putnam (1993). This greater importance of the family tallies with the propensity shown by Catholics to take occupational choices that favour “production” within the family and might hinder the functioning of institutions, for example through nepotism. When adding the greater proclivity of Catholics to cover up for their friends and their lesser trust in strangers, the “social ethic” of Catholicism seems to favour personal exchange to the detriment of impersonal exchange.

3.3. Discussion

These results suffer several limitations. First of all, finding significant correlations does not necessarily imply causation. It is possible that hidden variables may be affecting both the religiosity and other values. The analysis attenuates this problem by relying on within-country variation and by estimating fixed and variable effects, as it can be assumed that the fixed effects are relatively more exogenous. In addition, as argued by Guiso, Sapienza and Zingales (2003, 2006), the significance of the religious upbringing variable supports the notion that causality runs from religion to values.

Second, the data are built from statements on values instead of observations on actual behaviour. Given the nature of some of the questions (for example, whether one tolerates tax fraud), one should expect some bias caused by a certain tendency to lie because of “political correctness”. However, given that the analysis here focuses on differences between Catholics and Protestants, this bias should matter only to the extent that the proclivity to lying varies systematically between both religions.

Third, many of these tests use values to test hypotheses about actions. Some support for this approximation is given by empirical tests showing that, when responding on values, people tend to convey information on their own predispositions. For instance, self-reported trust has also been shown to be a good proxy of trustworthiness actions in an experimental setting (Glaeser et al. , 2000).

Fourth, given that these results have been obtained with current data, it could be claimed that the two theologies of salvation in Catholicism and Protestantism did provide different work ethics effects in the past but these differential effects have now disappeared. Such dilution of differences seems unlikely, however, when considering that social norms change slowly; therefore, current values may inform us on the effects of past values. As Glaeser and Glendon put it, “current social norms may still be the legacy of prior religious beliefs” (1998: 431). Moreover, if we take the differential effects of intense beliefs as a proxy for the differential effects that both religions may have had in a more religious past, the estimates lend more support to the social ethic than to the work ethic hypothesis. Even the differences observed in working hours seem to reflect occupational choices that are more consistent with the social ethic hypothesis.

Lastly, large sample size tends to cause statistical significance even with slight substantive effects. However, in this case, when evaluating substantive significance, it must be remembered that the effects of religions are underestimated because only within-country variation is considered. Much of the influence that religion has exerted throughout history has been through changes in the fabric of different societies. This effect is embodied here in the country dummy variables. This ensures that observed differences are not wrongly attributed to religion, but it also reduces the apparent explanatory power of religion. The overall consistency of results, both internal—between the battery of variables used in the paper—and external—with respect to those in the literature—, also rules out the possibility that they might be driven by mere chance, in the spirit of Meehl (1978).

4. Concluding Remarks

Overall, the article finds little support in survey data on currently held values for Weber’s “work ethic” hypothesis in “The Protestant Ethic,” by which Protestants would tend to work more and more efficiently than Catholics. It finds substantial support, however, for an alternative “social ethic” argument, as Protestant values are shown to shape a type of individual who exerts greater effort in mutual social control, supports institutions more and more critically, is less bound to close circles of family and friends and also holds more homogeneous values. In Weberian terms, the data are therefore more supportive of Weber’s view in “The Protestant Sects,” with its emphasis on mutual social control.

In line with these results, the economic contribution of the Protestant Reformation would have been connected not to the psychology of individuals regarding economic activity but to their empowerment as citizens vis-à-vis other citizens, the community and the State, affecting the relative effectiveness of alternative enforcement systems. The consequences for economic growth and the development of Capitalism would be related, first, to the greater effort that individuals are willing to exert in informal social enforcement; second, to the contribution that having more independent individuals makes to the design and functioning of political and legal institutions; and, lastly, to the greater homogeneity of values among individuals. All these features work in favour of anonymous markets, as they facilitate legal enforcement and reduce the cost of impersonal exchange.

Evans, T. D., Cullen, F. T., Dunaway, R. G., and Burton, Jr., V. S. (1995). ‘Religion and crime reexamined: The impact of religion, secular controls, and social ecology on adult criminality’, Criminology , vol. 33(2), pp. 195-224.

Freeman, R. B. (1986). ‘Who escapes? The relation of churchgoing and other background factors to the socioeconomic performance of black male youths from inner-city tracts’, in (R. B. Freeman and H. J. Holzer, eds.), The Black Youth Employment Crisis , pp. 353-76, Chicago and London: University of Chicago Press.

Fukuyama, F. (1995). Trust: The Social Virtues and the Creation of Prosperity , New York: Free Press.

Glaeser, E. L., and Glendon, S. (1998). ‘Incentives, predestination and free will’, Economic Inquiry , vol. 36(3), pp. 429-43.

Glaeser, E. L., Laibson, D., Scheinkman, J. A., and Soutter, C. L. (2000). ‘Measuring trust’, Quarterly Journal of Economics , vol. 115(3), pp. 811-46.

Granovetter, M. (1985). ‘Economic action and social structure: The problem of embeddedness’, American Journal of Sociology , vol. 91(3), pp. 481-510.

Greif, A. (2002). ‘Institutions and impersonal exchange: From communal to individual responsibility’, Journal of Institutional and Theoretical Economics , vol. 158(1), pp. 168-204.

Guiso, L., Sapienza, P., and Zingales, L. (2003). ‘People’s opium? Religion and economic attitudes’, Journal of Monetary Economics , vol. 50(1), pp. 225-82.

Guiso, L., Sapienza, P., and Zingales, L. (2006). ‘Does culture affect economic outcomes?’, Journal of Economic Perspectives , vol. 20(2), pp. 23-48.

Iannaccone, L. R. (1998). ‘Introduction to the economics of religion’, Journal of Economic Literature , vol. 36(3), pp. 1465-95.

Inglehart, R. (1999). ‘Trust, well-being and democracy’, in (M. Warren, ed.), Democracy and Trust , pp. 88-120, New York and Cambridge UK: Cambridge University Press.

Inglehart, R., and Norris, P. (2004). Sacred & Secular: Religion and Politics Worldwide , Cambridge UK: Cambridge University Press.

Knack, S., and Keefer, P. (1997). ‘Does social capital have an economic payoff? A cross-country investigation’, Quarterly Journal of Economics , vol. 112(4), pp. 1252-88.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R. W. (1997). ‘Trust in large organizations’, American Economic Review , vol. 87(2), pp. 333-38.

Levene, H. (1960). ‘Robust tests for equality of variances’, in (I. Olkin, ed.), Contributions to Probability and Statistics , pp. 278-92. Stanford, CA: Stanford University Press.

Luther, M. (1523). Secular Authority: To What Extent it Should Be Obeyed , January, in Martin Luther’s Basic Theological Writings (T. F. Lull, ed.), 1989. Minneapolis: Fortress Press.

McCleary, R. M. (2007) ‘Salvation, damnation and economic incentives’, Journal of Contemporary Religion , vol. 22(1), pp. 49-74.

Meehl, P. E. (1978). ‘Theoretical risks and tabular asterisks: Sir Karl, Sir Ronald, and the slow progress of soft psychology’, Journal of Consulting and Clinical Psychology , vol. 46(4), pp. 806-34.

North, D. C., and Thomas, R. P. (1973). The Rise of the Western World: A New Economic History , Cambridge UK: Cambridge University Press.

Putnam, R. (1993). Making Democracy Work: Civic Traditions in Modern Italy , Princeton, NJ: Princeton University Press.

Rodinson, M. (1974). Islam and Capitalism , London: Allen Lane. (English translation by Brian Pearce).

Sacerdote, B. I., and Glaeser, E. L. (2008). ‘Education and religion’, Journal of Human Capital , vol. 2(2), pp. 188-215.

Seabright, P. (2004). The Company of Strangers: A Naturals History of Economic Life , Princeton and Oxford: Princeton University Press.

Smith, A. (1759). The Theory of Moral Sentiments , Indianapolis: Liberty Fund (1982).

Stark, R. (1996). The Rise of Christianity: A Sociologist Reconsiders History , Princeton: Princeton University Press.

Stulz, R. M., and Williamson, R. (2003). ‘Culture, openness and finance’, Journal of Finance Economics , vol. 70(3), pp. 313-49.

Tajfel, H., and Turner, J. (1979). ‘An integrative theory of intergroup conflict’, in (W. G. Austin and S. Worchel, eds.), The Social Psychology of Intergroup Relations , pp. 33-47. Monterrey, CA: Brooks/Cole.

Trivers, R. L. (1971). ‘The evolution of reciprocal altruism’, Quarterly Review of Biology , vol. 46(1), pp. 35-57.

Weber, M. (1904-1905). The Protestant Ethic and the Spirit of Capitalism , Routledge, London and New York, 1992. Translated by Parsons, T., 1930, from the revised version originally published in the series Gesammelte Aufsätze zur Religionssoziologie , 1920. Tubingen: J. C. B. Mohr (Paul Siebeck).

Weber, M. (1920). ‘The Protestant sects and the spirit of capitalism’, http://www.ne.jp/asahi/moriyuki/abukuma/weber/world/sect/sect_frame.html, accessed January 19, 2009.

Zak, P. and Knack, S. (2001). ‘Trust and growth’, Economic Journal , vol. 111(470), pp. 295-