Low fertility in Latin America: an analysis using cohort size - - PDF document

Low fertility in Latin America: an analysis using cohort size variation and age- specific growth rates Vanessa di Lego Ana Paula Verona Cedeplar/UFMG/Brazil Abstract Some important studies have impinged upon the size of generations as playing a major role on shaping the future population age structure in Latin America and the Caribbean (LAC) countries (Wong 2009). To discuss this specific matter, the former work relied mainly upon UN projections variants to evaluate the probabilistic impact of future generation of newborns on the age structure and its age-dependency ratios. However, it could not take advantage of the yet to be launched census and surveys results for these set of countries after year 2010 and the 2015 UN Revision, when important changes took place. In this present work we indicate when exactly in time from 1950 to 2100 the cohort of females entering reproductive ages is smaller than the previous generation for all LAC countries, thus presenting the moment when the contribution of cohort size relatively starts to potentially affect future fertility trends and population size. We also compute age-specific growth rates from 1950-2100 for all females in reproductive ages to track how fast are these cohorts changing in size, and discuss how aspects of the fertility transition as applied to low European fertility contexts are not suitable for low LAC fertility contexts. INTRODUCTION Fertility has declined steeply and to surprisingly low levels in the developing world, challenging demographers’ former prospects and many UN projections (Bongaarts and Bulatao, 1999; United Nations Revision 2009). No variant hypothesis considered that

fertility would continue on such a dramatic downward trend, often due to these countries’ poor socioeconomic development and other common vulnerable indicators related to precluding fertility transition, to which Latin America and the Caribbean (LAC) is no exception (Wong et al 2000). The focus on the developing world as regards fertility has usually been on how to deal with its high levels and its negative consequences. The typical framework used to explain fertility transition and low fertility levels in developed countries was not only unable to foresee such a fast-paced and dramatic decline in the developing world, but is also feeble to explain how fertility determinants operate at such challenging contexts as Latin America and the Caribbean. This scenario urges demographers and specialists to revisit the framework that has been used to explain the determinants and pathways to lower fertility, since much has been done to explain those levels at developed countries, but not at the developing ones. Research show that for 13 countries that represent 79.2% of Latin American population fertility has continued to decline substantially, and not only in those countries where levels were above 4 or 5 children per women, but also for those with TFRs below 3 (Cavenaghi and Alves 2009). As with Mediterranean countries such as Italy, Spain, and Portgual, fertility in Latin American countries did not stall at around two children per women and clearly indicate a trend moving towards extremely low levels. Despite this common fact with the former that the fertility transition did not stall at around the 2.1 replacement level, LAC countries, some with remarkably different socioeconomic backgrounds, are reaching below or replacement fertility levels at a much faster pace than developed countries, inevitably leading to a faster aging process, while still having vulnerable labor markets, scanty or inexistent welfare systems and high inequality. In this work we analyze the possible relationship between cohort size variation at reproductive ages and fertility for a selected LAC countries, and discuss some of the particular aspects of below replacement fertility levels at those countries. BACKGROUND Total Fertility Rates and the Dynamics of Decline Figure 1 shows total fertility rate estimates from 1950 to 2015, for (panel a) selected regions in the world and (panel b) some selected European and Latin American

• countries. As it is well known, the fertility rates in Europe are well below replacement

level since 1975, as indicated by the solid horizontal red line in the Figures. On the contrary, by 2015 Latin America and the Caribbean as a whole was reaching replacement level, while the World is still behind. When looking more closely into some countries of both the LA and Caribbean region and Europe, however, we see that many countries have not only already reached below replacement levels in the former, but also a couple have done so reaching in 1980 the same level of France for the same year, as is the case of Barbados and Cuba. Nonetheless, what is most striking when comparing the profiles of fertility decline in both sets of countries is the rapid pace of decline of LA countries and in most cases the high level from which those countries start their fertility transition. Figure 1. Total fertility rate (TFR), selected Latin American and Caribbean Countries, and selected regions in the world, 1950-2015

a) Selected Regions in the World b) Selected Countries

2 3 4 5 6 1950-19551955-1960 1960-19651965-1970 1970-19751975-1980 1980-19851985-19901990-1995 1995-20002000-2005 2005-20102010-2015

Year TFT(UN estimates)

Region

EUROPE LATIN AMERICA AND THE CARIBBEAN WORLD

2 4 6 1950-1955 1955-19601960-1965 1965-19701970-1975 1975-19801980-19851985-1990 1990-19951995-2000 2000-20052005-2010 2010-2015

Year TFT(UN estimates)

Region

Aruba Barbados Brazil Cuba Dominican Republic France Germany Sw ed

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

Figure 2 below shows the profile of some LA and Caribbean countries more closely, considering also the projections until year 2100. Again, the speed of decline is striking.

As discussed in more detail ahead, particularly striking is the case of Brazil, where the TFR dropped from a 6.15 children per women in 1950 to a level of 1.75 children in

• 2015. Colombia also reaches below replacement level in 2005, while Chile reached that

level between years 1995-2000, Puerto Rico between 1990-1995, and Cuba between 1975-1980. Figure 2. Total fertility rate (TFR), selected Latin American and Caribbean Countries, 1950-2100

2 4 6 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055 2060 2065 2070 2075 2080 2085 2090 2095

Year TFT(UN estimates

Country Brazil Chile Colombia Cuba Dominican Republic Latin America and the Caribbean Mexico Paraguay Peru Puerto Rico

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

Within the European context, as shown in Figure 3, even when looking at the most dramatic case of Slovakia, where the TFR declines from 3.5 children per women between 1950-1955 to approximately 1.3 between 2010-2015, it is only a twofold decline during 60 years, while Brazil decreased almost six times in 65 years. No country in Europe underwent such a substantial decline, nor did they decline from such high levels of fertility.

Figure 3. Total fertility rate (TFR), selected European Countries, 1950-2015

1.5 2.0 2.5 3.0 3.5 1950-1955 1955-1960 1960-1965 1965-1970 1970-1975 1975-1980 1980-1985 1985-1990 1990-1995 1995-2000 2000-2005 2005-2010 2010-2015

Year TFR(UN estimates)

Region

Austria France Germany Slovakia Spain Sw eden Ukraine

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

The case of Brazil As previously mentioned, Brazil is the most radical case of fertility decline among the countries of South America and the whole of Latin America and the Caribbean. As indicated on the Figure below, the total fertility rate plummeted from a level of 6.15 children per woman in 1950 to a level of 1.75 in 2015. In only 65 years the total fertility rate in Brazil declined six times. No other LAC country underwent such a dramatic

• decline. As we can see on Figure 4, even countries that reached below replacement

levels much earlier than Brazil, and also at a very fast pace, as is the case of Cuba and

Barbados, did not start their transition at such high levels of fertility. Brazil started its fertility decline at a level of fertility higher than all Latin American and Caribbean countries taken together. But what still troubles specialists are the determinants of this

• decline. It happened with no directed effort or public policies with the explicit aim of

lowering fertility. Also, as it will be discussed in more detail for all LAC countries ahead, Brazil had a long history of widespread female sterilizations as a means of contraception. Although sterilization emerged initially among the upper class in Brazil, it spread throughout all classes, generations, racial groups and other segments of society, leading to what some researchers denominated a “culture of sterilization” (Berquó ). The prevalence of pairs of mothers and daughters or sisters who have been sterilized increases in number with educational level, although it is present as well among women with no education” (1999a, p. 214)”. In fact, a recent study confirms that the implementation of the sterilization law is far from desirable and leads Berquó and Cavenaghi (2002) to conclude that “the law has changed little the usual practice of sterilization and yet not satisfying individual reproductive rights”. For a more extensive review on the subject and

• n the influence of the Brazilian government and laws on fertility behavior see Goldani

(2001), Berquó, Elza, and Suzana M. Cavenaghi (2002), and Caetano, André Junqueira (2001).

Figure 4. Total fertility rate (TFR), selected Latin American and Caribbean Countries, 1950-2015

2 3 4 5 6 195019551960196519701975198019851990199520002005201020152020202520302035204020452050205520602065207020752080208520902095

Year TFR UN estimates

Country Barbados Brazil Cuba Latin America and the Caribbean

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

Determinants of Fertility Decline: revisiting the Fertility transition framework with some of the Latin American and Caribbean challenges It is out of scope of this work to extensively debate all determinants of fertility decline and how they have been dealt with in the literature. However, some specific aspects that are typically mentioned in the European contexts as having an important impact on fertility may have a different meaning in the LAC countries, and that is why we will mention some of them in this section.

Fertility and Development The relationship between fertility and level of development is a complex one and often approached to explain fertility differentials. Recently, the well-established linear and negative association between fertility and development (Bryant 2007) was challenged by evidence indicating that the relationship is not linear, but J-shaped (Myrskyla et al 2009). While low and average levels Human Development Index (HDI) are associated with low fertility, high levels of HDI are associated to an increased fertility, inverting the relationship to a positive one. However, other work for the US and Brazilian context did not find this association (Furuoka 2010; Junior et al 2013). In LAC countries this relationship is not so simple and some countries that experience low fertility rates, such as Cuba, have a worse HDI rank score than other countries in the same region with higher fertility. And the opposite is also true. So other factors play an important role to distinguish why some countries have higher or lower fertility levels than others in the LAC region. Fertility and Contraception Contraception is regarded as one of the most important proximate determinants of fertility decline (Bongaarts 1982). The main difference when acknowledging fertility in advanced societies, relative to developing ones, is that the former are societies in which birth control is the default option (Balbo et al 2013). This fact alone points out the importance of rethinking the role of contraception in the fertility transition framework when analyzing developing countries, and especially LAC ones. Still, a key factor in the dramatic decrease in fertility rates has been the increased use of contraception in LAC countries (UN 2015). Contraceptive prevalence almost doubled in the world between 1970 and 2015, from 36% in 1970 to 64% in 2015 (UN 2015). LAC was the region with an especially rapid growth in contraceptive use, passing from about 60% in 2000 to 72.7% in 2015, among married women of 15-49 years of age. The most recent estimates indicate that Nicaragua was the country with the highest contraceptive rate

(79.5%) in the LAC region, followed by Brazil (79%). However, out of those 72.7% total contraceptive use in Nicaragua, 25.7% are due to female sterilization, a long-term, irreversible method. The most extreme case is of Dominican Republic, (71.8% of contraceptive use and 41% are of female sterilization), followed by Puerto Rico (78.3%

• f contraceptive use and 36.6% are of female sterilization), Mexico (72.6% of

contraceptive use and 36.3% female sterilization), Nicaragua (79.5% of contraceptive use and 29.3% female sterilization), and Colombia (78.2% contraceptive use and 34.4%

• f female sterilization), as shown in Figure 5 below. Brazil has increased other methods

such as the pill (24.1%), but female sterilization is still a high 28.4%, when we compare with the 4.5% of female sterilization in Europe. The contribution of this relatively high percentage of women using irreversible methods must be taken into account when considering the future of fertility in LAC countries. In the framework used to study low fertility, not only is birth control considered accessible to all women, but also the methods available are different, allowing more control not

• nly over their parity, but also the timing of births. This bears important implications on

the quantum and tempo discussion. The high proportion of women in LAC countries using irreversible contraceptive methods means that considerable cohorts of women will never catch up their fertility, and as a consequence future fertility rates prospects may be even more dramatically low. This urges researchers to also focus on the impact of female sterilization on fertility, in the sense that a considerable proportion of females are simply not capable of postponing births, and also on the meaning of tempo effects when applied to LAC countries. Fertility and Timing of Motherhood Two of the most important components of fertility are tempo and quantum. Tempo refers to the timing of births and quantum to the total amount births. With the mean age at childbearing increasing since the 70s in OECD countries, time has gained attention from researchers.

Figure 5. Contraceptive prevalence among married or in-union women aged 15 to 49 by method and region, 2015

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). Trends in Contraceptive Use Worldwide 2015 (ST/ESA/SER.A/349).

However, the focus was on the “postponement transition”, as ages at first birth increased considerably (Kohler, Billari & Ortega 2002a).Time, however, is an aspect of fertility that can be considered both in terms of aging and youth, meaning that births can be postponed and anticipated. The focus has been on postponement, since the phenomenon rose in developed countries and followed their fertility decline, aging the fertility structure. In some developing countries, the fertility transition is, on the contrary, followed by a “rejuvenating” process of their fertility structures, with the decline first happening to women of older reproductive ages. LAC countries, again, are no exception and present an additional feature since before the onset of the fertility transition: the consistently and persistent high adolescent fertility. These countries experience persistently high fertility rates at young ages, even when completing the demographic transition and reaching below replacement fertility levels. This trend has drawn attention from many researchers (Berquó and Cavenaghi 2005; Alves and Cavenaghi 2009; Caterline and Mendoza 2008; Rodriguez-Vignoli 2008; 2011). In Brazil, by 2010, the

total fertility rate had already reached below replacement level at 1.9 children per women (Alves 2012) and the age-specific fertility rates for younger women, which experienced an increase between years 1991 and 2000, also started to fall, but still remained high. In addition, the fertility of the first two age groups of Brazilian females contributed with 45% of the total fertility rate in 2010, when fertility rate was below replacement level. Figure 6 indicates for the whole LAC region how adolescent fertility rate was high in 1960 and remained at high levels in 2015, despite decrease. The figures for LAC countries are more than twofold those of OECD countries, with the curve being out of the confidence intervals from a linear regression. This challenges the conception that low fertility is necessarily related to older reproductive age structures. Figure 6. Adolescent fertility rate (births per 1,000 women ages 15-19), Selected Regions, 1960-2015

Source: The World Bank, 2016

Fertility and First Unions Many researchers have debated on how the unions in Latin America are characterized by an early entry into first union and an increasing proportion of informal ones, despite great social transformations in Latin America in recent decades (e.g., the significant increase in female schooling), (Rosero-Bixby 1996; Vignoli 2005; Esteve, Lesthaeghe, and Lopez-Gay 2012). This aspect already challenges the very beginning of the transition in Europe, which was characterized by late and non-universal marriage and relatively high fertility within marriage. Brazil has a later age at union formation, compared to the other LAC countries, but research has shown that other factors, such as religion, can influence the interaction between age at first union and the formality of the union (Verona et al 2015). The fact is that nuptiality indicators have always been more important to explain the patterns of fertility decline in the European context, but do not seem no important in the LAC context. This is mainly due to this pattern of early union (marriage or consensual union) that has shown no significant changes in that over the last 40 years (Cavenaghi and Alves 2009). However, some research has shown that proportion of mothers below age 30 has dropped substantially in most Latin America countries, indicating a weakening early motherhood (Rosero-Bixby, Castro-Martín, and Martín-García 2009). These are the main aspects that are discussed as determinants of fertility and that have a different take when we analyze them in LAC countries. The aim is to urge researchers to revisit those aspects when studying LAC fertility, especially in a this new coming scenario of below replacement fertility for the region as a whole. MATERIAL AND METHODS Data We used population estimates and total fertility rates and projections from the World Population Prospects: The 2015 Revision, United Nations, Department of Economic and

Social Affairs, Population Division, for both European and Latin American and Caribbean countries. The estimates range from 1950-2015 and the projections from 2015-2100. The female population counts were from mid-year estimates of the time

• period. We also used data from Eurostat European population estimates from 1960-

2010 and CELADE (Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision). Method We estimated the age-specific growth rates between 1950 and 2015 of females at reproductive ages. The rationale of this method is that any population age structure in a given moment of time is the product of its past fertility, migration and mortality

• schedules. The measure that relates this set of historical vital rates to its current rates

are the set of age-specific growth rates, since in it is contained all the demographic history of a population (Preston and Coale 1982; Horiuchi and Preston,1988). By computing the age-specific growth rates it is possible to investigate if the cohorts of women reaching reproductive age are growing at a negative or positive rate and if the absolute cohort sizes are relatively changing. Also, in this approach, besides fertility, migration and mortality are already taken into account, since the proportion of persons in any age group is the product of those components. Therefore, in order to study cohort variation and size throughout time we do not need to explicitly account for those demographic components in this work, except if we wanted to decompose the relative contribution of each of the demographic phenomena. In that manner, we computed (1) for each reproductive age group, starting from 15-19 until 45-49, for all Latin American and Caribbean countries, as well as the European ones, from 1950 to 2100, estimating the growth rates at each 5-year interval. (1) [ ln ( , )] ( , ) a N x t r x t at  The aim is to analyze at what rate are the cohorts of reproductive age females changing every 5 years from 1950 until 2100. After computing the age-specific growth rates, we also estimated the proportion of females at each reproductive age group relative to the

total female population, in order to assess if their contribution in terms of percentage of the total population is also changing. RESULTS Cohort size variation

• This Figure above shows how the decrease in cohort size of the first reproductive age

group starts more consistently in year 1980 in Latin America and the Caribbean and about 10 years earlier for Europe. From year 2000 onwards, the decrease in the proportion of females entering their reproductive ages falls more substantially both in Europe and in Latin America and the Caribbean. In the former region the percentage of females in the first reproductive age group is significantly smaller than the latter throughout the whole period until 2015. However, the slope of decline is approximately the same, with the percentage of females of 15-19 years in Europe declining from 6.5% in 2000 to approximately 5% in 2015, while in Latin America and the Caribbean the decline is from 10% to 8.5%. Hence, in both regions the decline was of about 1.5%. However, when we consider the projections, it is clear how for Latin America and the Caribbean the pace of decline is much faster than Europe, with the latter maintaining a relatively constant percentage of females on all reproductive ages, relative to the total female population. The low and medium variants scenario of projections indicate a fast and dramatic decrease of almost twofold the proportion of females entering their reproductive ages in Latin America. This means that the proportion of females between 15-19 years old in 2015 is almost 2 times larger than in 2060. The cohort of females is hence smaller by double its initial size.

Figure 5. Proportion of females in each age group (% of total female population), by cohort and year , Selected Regions

15-19 20-24 25-29 30-34 35-39 40-44 45-49

0.04 0.06 0.08 0.10 1960 1980 2000 1960 1980 2000 1960 1980 2000 1960 1980 2000 1960 1980 2000 1960 1980 2000 1960 1980 2000

Year %females to total population

Region EUROPE LATIN AMERICA AND THE CARIBBEAN WORLD

Source: UN(2015)

Figures 6 and 7 above show a dramatic decrease in the proportion of females at the beginning of reproductive ages, starting from 2000, indicating important cohort size

• change. Cuba is the country with the most substantial decline in cohort size, with

females between 15-19 representing 10% of the total female population in 1955 and almost halving by 2015, representing 5.9% of the population. However, after 2050 the projections indicate a small increase in the amount of females in this age group. By 2050, the proportion of females entering their reproductive ages represents 5-6% of total population on all theses selected countries, approximately the same percentage of Europe between years 2005-2010. Figure 8 concentrates on the proportion of females entering reproductive ages in countries where below replacement fertility is already a reality or will inevitably be in the projected years. The blue line is a linear regression smoother showing the strong downward trend of the 15-19 female cohort sizes along Latin American and Caribbean. This Figure also suggests a linear relationship between below replacement fertility and the proportion reaching reproductive age that deserves further investigation. The red lines are pointing the position of Latin America and Caribbean as a whole on year 2020, relative to the other countries.

Figure 6. Proportion of females in each age group (% of total female population), by cohort and year, Selected Regions, UN Projections

15-19 20-24 25-29 30-34 35-39 40-44 45-49

0.03 0.04 0.05 0.06 0.07 0.08 0.03 0.04 0.05 0.06 0.07 0.08 0.03 0.04 0.05 0.06 0.07 0.08

High variant Low variant Medium variant

2020 2040 2060 2080 2100 2020 2040 2060 2080 2100 2020 2040 2060 2080 2100 2020 2040 2060 2080 2100 2020 2040 2060 2080 2100 2020 2040 2060 2080 2100 2020 2040 2060 2080 2100

Year %Females to total Population

Region EUROPE LATIN AMERICA AND THE CARIBBEAN WORLD

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition

Figure 7. Proportion of females in each age group (% of total female population), by cohort and year selected countries.

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition

15-19 20-24 25-29 30-34 35-39 40-44 45-49

0.050 0.075 0.100 1950 2000 2050 21001950 2000 2050 21001950 2000 2050 21001950 2000 2050 21001950 2000 2050 21001950 2000 2050 21001950 2000 2050 2100

Year CohortProp

Country Brazil Chile Cuba Haiti Latin America and the Caribbean Peru

Figure 8. Proportion of females in the first reproductive age group (% 15-19 of total female population), by cohort and year, Latin American and Caribbean Countries at below replacement fertility level

Argentina Aruba Bahamas Barbados Belize Boliv ia Brazil Chile Colombia Costa Rica Cuba Curaçao Ecuador El Salv ador Grenada Guadeloupe Guatemala Guy ana Haiti Honduras Virgin Islands Jamaica Martinique Mexico Nicaragua Panama Paraguay Peru Puerto Rico Dominican Republic Saint Vincent and the Grenadines Saint Lucia Suriname Trinidad and Tobago Uruguay Venezuela (Boliv arian Republic of ) Latin America and the Caribbean 0.06 0.08 0.10 0.12 2000 2025 2050

Year Proportion of 15-19 females to total population

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

Age-specific growth rates We also computed the age-specific growth rates to investigate further the rate of change of cohort size among reproductive ages. Figure 9 shows the overall trend for all countries of Latin American and the Caribbean, between years 1950 and 2100. The red lines indicate the rate and year at which some countries already present growth rates below zero. The blue line is a linear regression smoother showing the direction of the growth rates along the years. It is clear that the trend is downward, indicating that the growth rates becoming negative at all age groups. However, the most important is the first age group, 15-19 years old, where by 2050 virtually all countries in Latin America and the Caribbean will present a negative growth rate. However, it is important to note that this does not imply a smaller cohort in terms of absolute size, but already indicates that the negative growth will inevitably lead to smaller cohorts in the future. Figure 10 shows a selected group of countries and their growth rates. Again, Cuba presents a distinguished result, having an extremely negative growth rate, coherent with its already long-term fertility decline. However, it is still a remarkable change from 1975 and 2000. Another country that deserves attention is Brazil. In year 2000 the age group 15-19 presents, for the first time, a negative growth rate.

Figure 9. Age-Specific growth rates (r), female reproductive ages, all countries of Latin America and the Caribbean, 1950-2100

15-19 20-24 25-29 30-34 35-39 40-44 45-49

• 0.1

0.0 0.1 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 Year r

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

Figure 10. Age-Specific growth rates (r), female reproductive ages, Selected countries of Latin America and the Caribbean, 1950-2100

15-19 20-24 25-29 30-34 35-39 40-44 45-49

• 0.05

0.00 0.05 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100 1950 2000 2050 2100

Year r

Country Brazil Chile Colombia Cuba Latin America and the Caribbean Peru

Source: United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition/ Latin American and Caribbean Demographic Centre, ECLAC Population Division - Population database, the 2015 revision

DISCUSSION The resulting rates are the same: below replacement fertility. But do the principles governing the fertility transition and leading to low fertility in the developed world are also the ones in the developing world? Are the consequences the same? How do policy makers and researchers have to adapt the existing framework to such distinguishing international contexts? These were some of the questions that were the backdrop of this

• work. We argue that the forces driving fertility to low levels in LAC countries are not only

very different from the developed world, but are also accompanied by very different profiles of contraceptive use, first unions and timing of motherhood. Taken together, these aspects indicate the need to study fertility transition and the phenomenon of below replacement fertility in LAC countries in a different manner, urging us to question if the same framework used to study below replacement fertility in advanced societies are suitable for the developing world. In other words, Cuba and Western Europe, both with the same TFR of 1.63 in 2010, are not the same phenomenon. Further, some authors argue that unless structural changes in education and the labor market, as well as full access to reproductive health and effective contraception take place, the fertility schedule in LAC countries will hardly resemble those of developed countries, where women aged 20-24 still have low fertility compared to the 25 – 29 age group (Alves and Cavenaghi 2009, Goldani 2002). However, we argue that structural changes and access to reproductive and effective contraception are fundamental to guarantee the young population a better quality of life, qualified labor, as well as their reproductive rights, in conformance to those affirmed by international organizations since ICPD Cairo. Nonetheless, despite those important changes, our results show that fertility will inevitably decline in the next few years, due simply to smaller cohorts that are now entering reproductive ages. If important structural changes, public policies and unmet demand for contraception take place, those factors will only accelerate the process, but will not be the main drivers. The demographic force at work is powerful and alone will push fertility rates even further. Coupled with cohorts of women that

underwent sterilization and will not contribute catching up their fertility, we can expect substantial fertility decline in LAC countries in the next decades. In addition, these results provide support for discussing the low-fertility trap hypothesis, developed by Lutz et al (2006) for European contexts, in developing country scenarios. This hypothesis has three components: a demographic one based on the negative population growth momentum; a sociological one based on the assumption that ideal family size for the younger cohorts is declining as a consequence of the lower actual fertility they see in previous cohorts; and an economic one based on the first part of Easterlin’s relative income hypothesis. Lutz et al (2006; 2007) argue that all those three factors would work towards a downward spiral in births in the future, being practically an irreversible process. In 2010, Brazil already experiences a below replacement level fertility, a smaller cohort of women entering reproductive ages and, as Casterline and Mendoza (2008) have shown, the perception of ideal family size is decreasing in all LAC

• countries. In Brazil, 28% of women aged 15-24 reporting 0 or 1 as the ideal number of

children in year 2000. In 84% of all LAC countries the wanted TFR is less than replacement-level. Indeed, in four countries the wanted TFR is less than 1.5 births per woman: Brazil, Bolivia, Colombia, and Peru. Countries that still have high fertility rates also present low levels of wanted fertility, such as Bolivia and Peru (Casterline and Mendoza 2008). Some important limitations of this work are noteworthy. Fertility decline is not a steady

• phenomenon. Not only is it different among countries, but also between social groups

within countries. Important variation has been reported by subpopulations and by subregions, including between urban and rural areas, according to level of income and education, as well as by ethnic and indigenous group (Chackiel and Schkolnik 1992; Schkolnik and Chackiel 1998; ECLAC 2005; Dias Júnior et al 2008). Due to LAC’s

• verarching socioeconomic inequality and fast paced demographic transition,

considering those differences are even more important. This fact was not explicitly taken into account in this work. Relevant regional differences and socioeconomic gradients were not appreciated. In addition, since this work skirted around the necessity

• f revisiting some concepts of the typical framework of fertility transition and its

suitability to analyze the phenomenon LAC countries, this work was far from an extensive debate on each determinant of fertility. It lacks deepening on gender issues, nuptiality, the role of culture, and the important role of abortion in some of the LAC countries in shaping their reproductive outcomes. However, since our main objective was to show the overall changes in reproductive age cohort size throughout time and its importance in discussing the future of overall fertility, it did not affect much our

• conclusions. But future research must consider each of these elements in return.

ACKNOWLEDGEMENTS I thank the CNPq Brazilian institution for funding my research and providing me with the

• pportunity to develop my work. I also thank all members of Cedeplar, my home

institution, for all the intelectual and friendly support.

REFERENCES Balbo, N., Billari, F. C., & Mills, M. (2013). Fertility in Advanced Societies: A Review of

• Research. [journal article]. European Journal of Population / Revue européenne de

Démographie, 29(1), 1-38. doi: 10.1007/s10680-012-9277-y Berquó, Elza (1999). Ainda a Questão da Esterilização Feminina no Brasil. In Giffin, Karen and Sarah H. Costa, Org. Questões da Saúde Reprodutiva. Editora FIOCRUZ. Rio de Janeiro. Berquó, Elza (1999a). Sterilization and Race in São Paulo. In Reichmann Rebecca,

• Edit. Race in Contemporary Brazil: From Indifference to Equality. The Pennsylvania

State University Press. University Park, Pennsylvania. Berquó, Elza, and Suzana M. Cavenaghi (2002). The Impacts of the Voluntary Sterilization law on Reproductive Rights of Men and Women. Paper presented at the Annual Meeting of the Population Association of America, 9-11 May, Atlanta, Georgia. Bongaarts, J., and R. A. Bulatao. (1999) ―Completing the Demographic Transition.‖ Policy Research Division, Working Paper, No. 125. New York: Population Council. Caetano, André Junqueira (2001). Fertility Transition and the Diffusion of Female Sterilization in Northeastern Brazil. The Roles of Medicine and Politics. In the CD-ROM

• f the XXIV General Population Conference of International Union for Scientific Studies
• f Population, IUSSP. Salvador , 18-4 August, 2001.Bahia.
• Cavenaghi. Suzana, and José Eustáquio Alves (2009). Fertility and contraception in

Latin America: historical trends, recent patterns. In: Demographic transformations and inequalities in Latin America / Organization of Suzana Cavenaghi. – Rio deJaneiro: ALAP Chackiel, J., and S. Schkolnik. (1992) ―La transición de la fecundidad en América Latina.‖ Notas de población (55). Santiago, Chile: Latin American Demographic Centre. Dias Júnior, Cláudio Santiago, Verona, Ana Paula de Andrade, Pena, João Luiz, & Machado-Coelho, George Luiz Lins. (2008). Fecundidade das mulheres autodeclaradas indígenas residentes em Minas Gerais, Brasil: uma análise a partir do Censo Demográfico

• 2000. Cadernos

de Saúde Pública, 24(11), 2477-

• 2486. https://dx.doi.org/10.1590/S0102-311X2008001100003

ECLAC (Economic Commission for Latin America and the Caribbean). (2005) Social Panorama of Latin America, 2005 (LC/L.2288-P/E). Santiago, Chile. United Nations publication, Sales No. E.05.II.G.161. FURUOKA, F. (2010).The fertility-development relationship in the United States: new evidence from threshold regression analysis. Economics bulletin, v. 30, n. 3, p. 1.808- 1.822.

Goldani, Ana Maria (2001). Rethinking Brazilian Fertility Decline. In the CD-ROM of Brazilian Demography Sessions, ABEP/IUSSP. XXIV General Population Conference of International Union for Scientific Studies of Population (IUSSP), Salvador-, Brazil, 18-24 August 2001. Junior, A.FA.; Salvato, M. A.; Lanza Queiroz, B. (2013). Desenvolvimento e Fecundidade no Brasil? Revers’ao da Fecundidade para Municípios mais Desenvolvidos ? Planejamento e políticas públicas. n. 41/jul./dez. Livi-Bacci, M. (2012) Low Fertility in Historical Perspective. Population and Development review 38 (Supplement): 72–82 Lutz W. et al. (2006)The Low Fertility Trap Hypothesis: Forces that may lead to further postponement and fewer births in Europe. Pages 167-192 in D. Philipov, A.C. Liefbroer, and F.C. Billari (Guest eds.), Vienna 14 Yearbook of Population Research. Lutz, W. (2007). The future of human reproduction: Will birth rates recover or continue to fall? Ageing Horizons 7: 15-21. Oxford Institute of Ageing. Kohler, H. P., Billari, F. C., & Ortega, J. A. (2002a). The emergence of lowest-low fertility in Europe during the 1990s. Population and Development Review, 28, 641–681. Rosero-Bixby, L., Castro-Martín, T., & Martín-García, T. (2009). Is Latin America starting to retreat from early and universal childbearing? Demographic Research, 20(9), 169-194. Schkolnik, S., and J. Chackiel. (1998) ―América Latina: la transición demográfica en sectores rezagados.‖ Notas de población (67/68) (LC/G.2048/E). Santiago, Chile: Economic Commission for Latin America and the Caribbean (ECLAC). United Nations, Department of Economic and Social Affairs, Population Division (2015). World Population Prospects: The 2015 Revision, DVD Edition. United Nations, Department of Economic and Social Affairs, Population Division (2015). Trends in Contraceptive Use Worldwide 2015 (ST/ESA/SER.A/349). Verona, A. P., Dias Jr, C., Fazito, D., & Miranda-Ribeiro, P. (2015). First conjugal union and religion: Signs contrary to the Second Demographic Transition in Brazil? Demographic Research, 33(34), 985-1014. Wong, L. R., J. A. M. Carvalho, and A. Aguirre. (2000). Duración de la transición demográfica en América Latina y su relación con el desarrollo humano.‖ Estudios Demográficos y Urbanos (043) :185–207. México D.F.: El Colegio de México.