«The Growth of Protestantism in Brazil and its Impact on Income, 1970-2000 Joseph E. Potter Ernesto F. L. Amaral Robert D. Woodberry Introduction ...»
The Growth of Protestantism in Brazil and its Impact
on Income, 1970-2000
Joseph E. Potter
Ernesto F. L. Amaral
Robert D. Woodberry
Protestantism has expanded rapidly in Brazil in recent decades resulting in a substantial religious
transformation in a country in which, not long ago, nearly all citizens were at least nominally Catholic.
The churches that have grown the fastest have Pentecostal or neo-Pentecostal roots, such as the
Assemblies of God or the Universal Church of the Kingdom of God. When the vibrancy and impact of these Churches are discussed in the popular press, reference is often made to former drunks who, after converting to Pentecostalism, have transformed their lives for the better. A typical description is found in
a story filed by Reuters reporter, Todd Benson, shortly before the Pope’s visit to Brazil in 2007:
“For years, Ronaldo da Silva's daily routine consisted of drinking himself into a stupor until he passed out on a sidewalk. Now he spends his days praying and singing with hundreds of fellow Christians at the Universal Church of the Kingdom of God in Carapicuiba, a sprawling shantytown on the outskirts of Sao Paulo where Pentecostal congregations are found on just about every block.
‘I'd probably be dead or in jail if it weren't for this church,’ said da Silva, a 38-yearold former Catholic who claims God cured him of epilepsy and helped him straighten out his life when he converted to Pentecostalism a decade ago.
Conversions like da Silva's are increasingly common all over Brazil, where a boom in evangelical Protestantism is steadily chipping away at the supremacy of the Roman Catholic Church.” In this paper, we seek to assess whether the increase of Protestantism in Brazil has led to an increase in male income. If the type of experience described above is indeed widespread, we might expect to find an impact of Protestantism among the least fortunate segments of the population most afflicted by alcohol dependency, binge drinking, and drugs. But if conversion to Protestantism is especially likely to occur among those in the worst circumstances and who have a history of problematic behavior, it might well be that even after conversion Protestants might be no better off than other members of the population even though their personal lot had improved considerably. We tackle this problem of selectivity using microdata from the Brazilian censuses of 1970, 1980, 1991, and 2000, and and income analyzing the association between Protestantism/ the group rather than the individual level.
Protestantism in Brazil:
Protestant missionary activity began in Brazil in the mid-nineteenth century.Missionaries founded schools, hospitals, radio stations, printing presses, and clinics, yet made few converts until the second half of the 20th century (Ferriera 1959; Belloti 2000; Chestnut 1997). The number of Protestants also increased through immigration, particularly German immigration to southern Brazil. Yet, even today these mission/mainline Protestant denominations (Lutheran, Presbyterian, Anglican, and Baptist)1 make up only about 4% of the population (Instituto Barsileiro de Geografia e Estatística 2002).
The majority of Protestants in Brazil are Pentecostal. Pentecostals first came to Brazil in the early 20th century, but in the second half of the 20th century spread rapidly – particularly among the poor and in urban areas. Most Pentecostals are part of Brazilian initiated and controlled denominations and churches and have less connection with foreign missionaries. Leadership in Pentecostal churches typically requires less formal education and leaders are often recruited from the local community. Moreover, and highly Both the “mainline” and “mission” label can be misleading. “Mainline” can be misleading because the word comes from the United States and often has the connotation to being theologically liberal, where as many of the “mainline” Protestants in Brazil are theologically conservative. “Mission” can be misleading since a significant portion of these Protestants are immigrants (particularly German Lutherans), and all these denominations are now firmly under indigenous leadership.
participatory services focusing on spiritual gifts are easily accessible and relevant to the felt needs of ordinary people (Vingren 1987; Chestnut 1997; Mariano 2004). According to the 2000 Brazilian census, 11.7% of the population now consider themselves Pentecostal.
Both mainline/mission and Pentecostal denominations are Protestant and both have traditionally emphasized abstaining from alcohol and drugs (or at most moderate use of alcohol), avoidance of sex outside of marriage, commitment to family, modesty in dress, avoidance of corruption, etc., (e.g., Chestnut 1997). Many Pentecostal groups are even more restrictive on the use of alcohol than other Protestants.
However, there are also significant differences between Pentecostal and non-Pentecostal Protestants.
Non-Pentecostal Protestants in Brazil typically base religious authority primarily on the Bible and have a formally educated clergy (formal education is required for ordination). Thus, non-Pentecostal Protestants have traditionally put a strong emphasis in both literacy and formal education and are disproportionately educated for their class backgrounds (Chestnut 1997).
On the other hand, Pentecostals base religious authority both on the Bible and on the gifts of the Holy Spirit. Pastoral leadership is based more on perceived spiritual gifting and less on formal education or academic degrees than for non-Pentecostals. This enables Pentecostals to reproduce leaders quickly and keeps leadership closer to the conditions and understanding of marginalized congregants, but it may minimize the impact of Pentecostalism on educational outcomes relative to non-Pentecostal Protestantism. While Biblical literacy is important to Pentecostals, formal education may not be emphasized as much (Garrard-Burnett and Stoll 1993; Noll 2002). In recent years Pentecostals have founded schools all around Brazil (particularly elementary schools), however it is still not clear whether Pentecostals will have as important an influence on education as non-Pentecostal Protestants have.
Within Pentecostalism there are several distinct families of denominations/churches and a bewildering array of denominations. However, for our purposes the most important distinction is between “traditional” Pentecostals and “neo-Pentecostals” (Cox 1995).2 Neo-Pentecostals are also often called “Word of Faith” Pentecostals (by insiders) or “Health and Wealth” Pentecostals (by outsiders). NeoPentecostal theology emphasizes that God materially blesses those who have faith and serve him well.
One common teaching is that if people want to become rich, they should give “seed money” to their local church. God will see this seed and bless contributors with a far greater financial harvest. In this context public displays of wealth are acceptable, even by pastors, since it reflects God’s material blessing not a lack of spirituality. One of the fastest growing (and most notorious) neo-Pentecostal denominations in the world is the Universal Church of the Kingdom of God (IURD) established in Rio de Janeiro in 1977. It now owns a TV station, has churches across Brazil, and has established branches in many other Latin American countries and Portuguese speaking regions of Africa (Mariano 2004; Freston 2001).
While both Protestant and Catholic clergy decry violence, premarital pregnancy, and abuse of drugs and alcohol, ethnographic evidence suggest that Pentecostals are more successful than the Catholic Church at recruiting young people and convincing them to make a more radical break from dominant cultural patterns with respect to alcohol use and sexual behavior (e.g., Burdick 1996).3 Unfortunately, there is less ethnographic evidence about the impact of neo-Pentecostal churches on alcohol consumption and other economically relevant behaviors. However, the impression of many observers is that neoPentecostals are less demanding than many traditional Pentecostals (e.g., Mariano 2004).
Protestants’ Impact on the Economy:
There are a number of mechanisms through which Protestantism may influence economic outcomes in Brazil. Some mechanisms are the result of distinct emphases within Protestantism relative to Catholicism or Afro-Brazilian religious traditions. Other mechanisms are the result of religiosity in general, and would apply equally to Protestants and Catholics. However, in the Brazilian context Other distinctions are between Unitarian and Trinitarian Pentecostals, whether speaking in tongues is an essential sign of being baptized by the Holy Spirit or merely a common sign of it, etc.
This evidence is consistent with ethnography and statistical research from other countries in Latin America (e.g., Brusco 1995;
Protestants are typically far more religiously active than Catholics and thus may accrue more of the benefits and costs.
First, Protestantism and religiosity may influence health which in turn influences economic outcomes. Internationally, there is substantial evidence that for Christian groups, religiosity is associated with greater health and longer life expectancies (e.g., Hummer et al. 1999; Lehrer 2004; Woodberry 2008a; 2008b). Research on religion and health within Brazil is limited. However, there is evidence that infant mortality is about 10 percent lower among Protestants than Catholics (holding all else equal) – although the reduction is greater for non-Pentecostal than for Pentecostals (Wood, Williams, and Chijiwa 2007). To the extent that health increases people’s work efficiency and lower mortality increases their time horizons, this may influence economic outcomes, particularly for poor people.
Second, both religiosity and theologically conservative Protestantism may influence the investment fathers make in their families. Internationally, there is substantial statistical evidence for this (e.g. Waite and Lehrer 2003; Wolfinger and Wilcox 2008; Woodberry 2008a; 2008b). In Latin America, ethnographic evidence suggests a similar pattern (Burdick 1993; Brusco 1995; Sexton and Woods 1977;
Sherman 1997; Robbins 2004). While this may influence the economic outcomes for children (particularly those from poor families), it is less likely to influence economic outcomes within the short time frame covered by this paper.
Third, there is substantial international evidence that both Protestantism and religiosity are associated with greater aversion to corruption and greater rule following behavior (Guiso, Sapienza and Zingales 2003; Stark 2001; Stack and Kposowa 2006; Woodberry 2008a; 2008b). Ethnographic evidence from Latin American suggests a similar pattern (e.g., Robbins 2004: 136). Protestants have their share of corruption, including some high profile corruption scandals by neo-Pentecostal leaders in Brazil (Freston 2001; Woodberry and Shah 2004). However, to the extent that the spread of Protestantism and religiosity minimizes corrupt behavior among the laity, it may foster development in communities with more Protestants and greater religiosity. Still, this is likely to help community-wide economic development, not individual economic attainment. Aversion to corruption may especially diminish incomes of those in the middle and upper classes, at least in a society with high levels of overall corruption. However, it may make Protestants and religiously active people more attractive employees, and thus enhance their economic prospects – particularly from poor communities.
Fourth, in most societies, Protestantism has been central to the expansion of both the supply of and demand for education (Woodberry 2004; 2008). In societies with short histories of Protestant religious competitions, Protestants typically are disproportionately educated and have higher educational expectations for their children (Woodberry 2004; Garner 2000; Blunch 2008; Heaton, James, and Oheneba-Saky 2008; Zhai 2006; Roemer 2008). Ethnographic evidence from Latin America also suggests that Protestants are more likely to be literate and invest more in educating their children relative to others from their economic background (Tax and Hinshaw 1970; Early 1973; Sexton 1978; Annis 1987; Brusco 1995; Sherman 1997).
While Catholics have often founded and run some of the best schools (as they do in Brazil), prior to Vatican II (1965) Catholics did not invest heavily in education for non-elites except when competing with Protestants (Woodberry 2004). To the extent that Protestantism and religiosity increase the supply of private schools, increase the quality of schools available to the poor through educational competition (Gallego and Woodberry 2008a; 2008b), and increase the incentives of poor people to acquire literacy and formal education, they may increase long-term economic development for both communities with more Protestants and for the children of Protestant parents.
Fifth, international evidence suggests that both Protestants and highly religious respondents are more likely to volunteer both formally and informally. This is true both for religious and non-religious organizations (Woodberry 2004; 2008; Verba Schlozman and Brady 1995; Smidt 2003). To the extent that this voluntary activity provides economically relevant services, it may foster long-term economic development in communities where more Protestants and religiously active adherents are present. This may be especially relevant in poor communities where for-profit suppliers are less prevalent and less affordable to poor people.
Finally, since the 19th century conversionary forms of Protestantism have been consistently associated with lower-levels of alcohol consumption. This is economically relevant in communities where alcohol abuse is widespread. In the 19th century, revivalist Protestants helped spur the temperance movement in reaction to a rapid increase in the supply and consumption of distilled alcohol. Historians suggest that temperance was associated with economic uplift in both frontier and urban communities (Johnson 1978; Blocker 1989) and Protestant activists helped spread temperance worldwide (Woodberry 1996).
Contemporary statistical evidence from the US suggests that highly religious people are less likely to abuse alcohol (e.g., Bazargan et al. 2004; Ford and Kadushin 2002) ethnographic evidence from Latin American suggests that Protestantism diminishes alcohol and drug consumption as well (Burdick 1993;