Bill Totten's Weblog

Sunday, March 20, 2005

Population and Sustainable Development (2 of 2)

by Herman E Daly

from his Beyond Growth (Beacon Press, 1996)

Chapter 9

Marx and Malthus in Northeast Brazil: A Note on the World's Largest Class Difference in Fertility and Its Recent Trends

During the early 1970s I published several articles dealing with differential fertility and distribution of income per head in Northeast Brazil. <1> The central thesis of the study was that even though aggregate gross national product per head for Northeast Brazil (the largest poor region in the Western Hemisphere) had grown very rapidly over the previous decade, this growth represented "swelling" rather than development, because the income per head of the poorer 80% of the population was at best constant, while that of the richer 20% was growing very rapidly. One very important reason for this disparity, and for the absence of any net "trickle down" effect, was the more rapid population growth of the lower class, due to much higher fertility which more than compensated for the higher mortality of the poor. To the Marxist notion of exploitation based on class monopoly of the means of production there must be added a notion of Malthusian or "Roman" exploitation based on class monopoly of the means of limiting reproduction. In ancient Rome (as in Northeast Brazil) the role of the proletariat was to procreate a plentiful supply of laborers and servants for the republic (that is, the patricians).

In policy discussions of the 1960s the demographic problem was commonly dismissed by pointing to the fact that total GNP was growing at over 6% while population was growing at around 3%. The consequent growth in GNP per head of around 3% was considered more than adequate. Add to that statistics showing the sparse density of population per square kilometer in Amazonia, and the demographic problem, it was thought, could be dismissed as an exaggeration at best, or neo-Malthusian apologetics at worst. This dismissal, of course, was as statistically inept as the famous recipe for "fifty percent rabbit stew" (one rabbit, one horse). As will be seen later, this error is still prevalent in high circles.

The estimate of total fertility that I made for the late 1960s was eight surviving children for the poorer eighty to ninety percent of the population and four surviving children for the richer ten to twenty percent. This class difference in net reproduction ratio of the order of 100% strongly affects the distribution of income per head. The effects may be summarized under three headings.

1. The denominator effect. A given family income divided among more people results in a lower family income per head. Since families with lower incomes have about twice as many children as those with higher incomes, the effect of differential fertility on the distribution of income per head is very large and very regressive.

2. The numerator or wage effect. The rapid reproduction of the lower class results in an unlimited supply of cheap labor. Lower wages (smaller numerator for the lower class) mean higher profits, more reinvestment, and faster growth in aggregate income (higher numerator for the upper class). But growth was never fast enough to offset the effect of differential reproduction, and did not, therefore, exert an upward pressure on wages leading to the expected "trickle down". Also, the abundant supply of cheap labor allows the educated wives of members of the upper classes to employ domestic servants, thus freeing themselves for relatively high-paying jobs, thereby reinforcing class inequality in the numerator (total family income). In Northeast Brazil, poorer families have fewer breadwinners than richer families, even though they have more members. In 1970, 65.1% of the families with income per head less than half the minimum wage had only a single breadwinner, while the corresponding figure for those with incomes more than twice the minimum wage was only 54-5%. <2>

3. The age of structure effect. More rapid population growth in the lower class results in a younger average age than in the upper class and in a higher dependency ratio. This makes it more difficult for members of the lower class to save, and also tends to keep wives out of the labor force. The opposite is the case for the upper class. More speculatively, the lower average age of the lower class enhances the dominance of the upper for the simple reason that, up to a limit not yet reached, older people find their greater knowledge and experience of life to be an advantage in the political and economic domination of a younger group. In 1970, for Brazil as a whole, the percentage of the population younger than fifteen years in the lowest income class (less than half minimum wage) was 47.8%. For the next lowest category (between 50 and 100% of minimum wage) the figure was 28.7%, while for the highest category (more than two minimum wages) it was only 21.1%. <3> Since we would expect people under fifteen to have lower incomes on average than those over fifteen, it stands to reason that in a class with 48% of its members less than fifteen years old, other things equal, income per head would be lower than in a class with 21% of its members under fifteen. Since differential fertility directly determines age structure it also indirectly determines distribution of income per head to a significant degree.

My basic conclusion in 1970 was that a family planning program aimed at the democratization of birth control was a necessary (but obviously not sufficient) condition for economic development, understood as improving the lot of the bottom eighty percent of the population of Northeast Brazil. Such a thesis did not, and still does not, enjoy official acceptance, even though it is widely debated. <4> The opposition to family planning during the 1960s was quite intense and came from the conservative wing of the Catholic church, the nationalists, the military, the leftists, and the oligarchy. With such an alliance of enemies it was not to be expected that population control would become official policy soon, and it has not. The common opposition of the leftists and the oligarchy to population control is especially interesting since they hold opposite expectations about the political consequences of rapid population growth in the lower class. The leftists think (or thought?) that it would hasten the revolution by building up the pressure of misery, while the oligarchy apparently believes that it will increase stability by absorbing any surplus above subsistence that the proletariat might use on its own behalf. The nationalists and the military want an abundant proletariat to serve in the army and colonize the Amazon to secure it against foreign penetration. The leftists want a growing proletariat to fight for the revolution, and the oligarchy wants a growing proletariat to work in its factories, farms, and households at low wages. Everyone wants someone else to do their dirty work, and the larger the exploited population the better for those who want to keep their hands clean. Both leftists and oligarchs seemed to believe that "foxes should not advocate birth control for rabbits!"

What are things like now, some fifteen years later? What new information has become available? To what extent does it support or contradict the picture presented in my article of 1970 and summarized above? What changes have occurred in official attitudes and policy?

Two new sources of information are relevant. First, the census of 1970 gives new information on fertility and mortality differentials by income in 1970. As will be seen below, this information strongly supports the basic thesis. Secondly, in the PNAD-1977 <5> comparable information is given for that year, indicating a rather significant lowering of total fertility, along with a narrowing of the extreme class differences. Information from both sources was elaborated in a joint project by UNICEF and IBGE published in 1982. <6> Table 1 is reproduced from that study. {Sorry, unable to include the Table in this post.}

The most striking feature of Table 1 is the enormous drop in total fertility <7> between the adjacent income categories of "below half minimum wage" and "half to 1 minimum wage". For the Northeast (1970) the fall is from 9.26 to 2.38 - a reduction of nearly seven births between the lowest class (comprising 84.6% of the population of women between the ages of fifteen and forty-nine), and the adjacent "lower middle class" (containing 5.6% of the same population). For the two highest categories, together accounting for about 3% of the population, total fertility is a little higher at 2.83. Using the latter figure as a basis of comparison gives a class differential of 6.43. We can say with some confidence that the class differential in total fertility lies between 6.4 and 7.0.

My estimate in 1970 of eight and four for the lower and upper classes, respectively, referred to surviving children rather than live births. Given the infant mortality rates in the Northeast of around fifteen to twenty percent <8>, a total fertility of 9.26 is roughly consistent with an average of eight surviving children, so in retrospect that estimate still looks good. However, for the upper class I evidently overestimated fertility. Apparently it was around 2.38 rather than 4 (correction for infant mortality in the upper class would be minor, but would lower the 2.38 figure, thus making my estimate of four surviving children an even greater overestimate). My implicit estimate of the class differential was, therefore, about 5 (= 9 - 4), while the real differential was apparently close to 7 (= 9.26 - 2.38). Consequently the effect of differential fertility on income distribution was actually underestimated in my paper of 1970 because the differential itself was underestimated. The new data strengthen the main argument of that paper, for 1970. However, Merrick and Berquo in their very thorough and recent study report a total fertility for the lower class of 8.55, and for the upper class of 3.95 for the Northeast in 1970. <9> These figures, cited from a study by Carvalho and Paiva, reflect different statistical definitions of social class, based on total household income in absolute cruzeiros rather than family income per head in multiples of a minimum wage. A significant difference in total fertility is apparent regardless of the exact definition of lower and upper class.

There are, of course, limitations in the use of either total family income or family income per head. As Wood and Carvalho remark, "Larger households tend to have larger total incomes because of the greater number of earners, but smaller incomes per household member". <10> When households are grouped by income per head, differential fertility and income inequality are greater than when households are grouped according to total household income. Neither category is "biased". They are simply different concepts, each useful and reasonable as long as we remember the definitions and avoid mixing them in comparisons.

Between 1970 and 1977, however, there was a substantial drop in total fertility for all classes together in all regions of the country. In the Northeast it fell by about 2 (from 8.36 to 6.22). Moreover, the gap between the groups "below half minimum wage" and "half to one minimum wage" narrowed from around 7 to around 4.4 (7.73 - 3.37). Part of this narrowing was due to a very surprising increase in the fertility of the "half to one minimum wage" category (from 2.38 to 3.37), an increase of over forty percent in seven years. Why this strong movement against the trend for this particular class which was previously the least fertile? There are good reasons to believe that this reflects a statistical reclassification rather than a real change in behavior. Since monetary correction for inflation lagged behind the rate of inflation between 1970 and 1977, the real purchasing power of the minimum wage was eroded, and although some people from the lowest category nominally passed into the next income category there was no change in either their real incomes or their reproductive behavior. Consequently, these people raised the fertility of the nominal income class into which inflation moved them. Clearly the real class differentialin fertility in 1977 was greater than 4.4, although still significantly lower than the difference of 7 recorded in 1970 - probably around 5.4 on the reasonable assumption that the entire increase in fertility in the "half to one minimum wage" class was spurious.

Between 1970 and 1977 there has been a clear narrowing of extreme class differences in fertility, and a consequent weakening of the effect on distribution of income per head compared to 1970. However, lower-class total fertility was still over twice that of the upper class, so the effect remains highly significant. Moreover, since total income growth has now slowed drastically, the pressure of high fertility on income per head in the lower class may now be even greater in absolute terms.

The differential of nearly seven recorded in 1970 was probably the greatest class difference in fertility in the world, and perhaps in all history. A cursory review of the literature yields a few benchmark comparisons which do not prove that Northeast Brazil has had the highest class difference in fertility in the world but, nevertheless, do render that conjecture very plausible.

1. For Peru in 1960, Stycos <11> defined five socioeconomic classes and found the difference in total fertility between the lowest (7.6) and the highest class (3.8) to be 3.8.

2. For Mexico City in 1970, Zambrano Lupi <12> found that total fertility of mothers with less than three years of education was 7.85 and that of mothers with university education 4.53, a difference of 3.32.

3. In Calcutta, Pakrasi and Haider <13> found that total fertility in the class with the lowest monthly expenditure was 7.95 against 4.52 in the class with the highest monthly expenditure, giving a class differential of 3.43.

4. For the United States in 1910, Petersen <14> reports a difference between the total fertility of "rural workers and farmers" (5.6) and "urban professionals" (1.8) of 3.8.

5. For Brazil as a whole in 1970, Wood and Carvalho <15> discovered that total fertility for the lowest of four household income classes was 7.6 while that of the highest class was 3.3, giving a difference of 4.3.

In all of these cases the comparisons are between a top class and a bottom class, with a substantial excluded middle, while in Northeast Brazil there is no excluded middle since our bottom category alone contains eighty percent of the population and we are comparing it with the adjacent category. Even so, the differences cited above fall far short of the figure of nearly seven for Northeast Brazil in 1970, and are also well below the 1977 figure of 5.4. I doubt whether any other society in the world has such a large fertility difference between classes of comparable inclusivity.

Furthermore, the difference is not between indigenous peoples and colonizers, nor between religious or linguistic subcultures. The fertility difference in Northeast Brazil is one of economic class. Both classes speak the same language, belong to the same church, are of the same racial background, watch the same TV novelas, go to the same futbol games, and vote (or are forbidden to vote) in the same elections. A class difference in total fertility of almost seven children seems so extreme that it should have attracted a great deal of attention. It did not. Nor is much attention now paid to the current difference of about 5.4, which must also be among the very highest in the world. Why have these striking facts been so little noticed?

One possibility, of course, is that the figures are wrong. The majority (but by no means unanimous) reaction of my economist colleagues at the Federal University of Ceara, when presented with Table 1, was one of disbelief. The fertility differences, they said, were too large and occurred too abruptly between adjacent classes, which were not as different as all that on the social scale. They may be right. On the other hand, the source (IBGE) is the best available, and two independent data sets were used, both giving the same general picture. Furthermore, the two adjacent classes between which the big change occurred are less close than might appear because the first category (below half minimum wage) is open-ended downward and contains around 80% of the total number of families and 92% of the total number of individuals in 1970. The family incomes per head of most people in this populous category are actually less than one-quarter of the minimum wage. In 1970, 67% of the families and 76% of the individuals were in the group "below one-quarter minimum wage. Thus, the median for the open-ended category, "below half minimum wage" is going to be well below one-quarter of the minimum wage. The median of the next category (half to one minimum wage) can be approximated by the mid-point of 75% of minimum wage. Viewing it in this way, we see that the median income per head of the second category is three times that of the first, even though they are adjacent.

The fact that these figures provoke disbelief as a first reaction among many nordestinos may be more an indication of customary blindness to the social reality they reflect, than a reason to doubt the numbers. That at least was the interpretation of some of the people at IBGE in Rio de Janeiro who carried out the study, when informed of the disbelief prevalent among my colleagues in Ceara. In any event, the subject deserves more study by believers and doubters alike.

A broader question that also deserves more study was raised years ago by Joao Lyra Madeira. <17> Why has the whole subject of fertility been tainted with a virtual taboo in Brazilian demography and economics? Nowadays "taboo" is perhaps too strong a word - yet there remains a certain "disinclination" to study fertility in its socioeconomic context. When a subject is taboo it usually reveals an injustice too blatant to defend openly, but too important to the interests of the status quo to challenge openly. I am reminded of the segregated Southern United States of my childhood. It was obvious to an unexceptional child that, by and large, the blacks did the dirty work and were poor, while the whites did the more interesting work and were better paid. And segregation laws were intended to keep it that way. It was acceptable to take explicit notice of this fact as long as one did not dwell on it, or pursue it too deeply. Very few whites in the South of the 1940s really hated blacks (or "colored people" as was then the polite term). In fact, most whites liked blacks on a personal level and "in their place", as the cliche went. Simiilarly, I suggest that upper-class Brazilians in the Northeast have a genuine affection for their lower-class countrymen on a personal level and "in their place", which means working in the factories, farms, and households of the upper class. Northeast Brazil is much less racist than the United States, but economic class divisions are stronger.

In a class society it is natural that there should be class differences in everything, including fertility. The fact that differential fertility reinforces the basic class structure of the society is what makes it inconvenient to study the issue in depth. It is inconvenient to the oligarchy because it exposes another dimension of class exploitation. One might think, therefore, that Marxists would be eager to expose the facts. However, they are unwilling to admit any cause for poverty other than monopoly ownership of the means of production. Neo-Malthusian policies are especially distasteful to Marxists because they hold out the possibility of improvement by individual action, thereby weakening class solidarity and taking some wind out of the sails of their single cure for all ills, the class revolution. Consequently, the subject is neglected.

Because of this neglect the exceptionally large fertility differentials in Northeast Brazil are still awaiting adequate "explanation". As I have indicated, I believe the basic explanation lies in the class nature of the society. As I argued in my studies of 1970 and 1971, the literal role of a proletariat is to proliferate. <18> The literal meaning of the word "proletariat" in ancient Rome was those with many children, the poorest class of society whose members were exempt from taxes and whose service to the republic was mainly in the procreation of children. Implicit in this literal meaning and explicitly developed in Malthusian and neo-Malthusian thought is the association of poverty with rapid proliferation. By Marx's time the word had largely lost its Latin meaning, and Marx severed any remaining etymological connection with proliferation by redefining the word to mean nonowners of the means of production who must sell their labor to the capitalists in order to survive. A theory of poverty is also implicit in Marx's definition, specifically that poverty is a consequence of the monopoly ownership of the means of production by a few and the consequent nonownership of the means of production by many. Ideologically, in the sense of providing a single cause and a single cure for poverty, the two traditions conflict. But logically the Marxian and Malthusian theories do not conflict. It is quite possible for one class to have a monopoly of both the means of production and the means of limiting reproduction. In fact, Marx himself tells us that mere possession of land and capital is not sufficient to make a man a landlord or a capitalist if there be lacking the requisite social correlative, the proletarian with no alternative but to sell his labor to the capitalist. <19> Rapid reproduction of "correlatives" supports and reinforces the unequal distribution of productive wealth that is the dominant feature of social life in Northeast Brazil.

Further demographic evidence of the class nature of the society, and partial explanation of the extreme fertility differences, is provided by the statistics on class differences in mortality found in Table 2. {Sorry, unable to include the Table in this post.}

For the Northeast as a whole in 1970 there was a difference in life expectancy at birth of 17.5 years between the lowest category (below half minimum wage) containing over eighty percent of the population, and the highest category (more than twice the minimum wage) containing about one percent of the population. In the urban Northeast the difference was nearly twenty years! A species, or class, facing high death rates (low life expectancy) must have high birthrates to survive. Although highly significant, the class differences in mortality are less dramatic than those in fertility. That is, higher fertility in the lower class much more than compensates for higher mortality.

By 1977 the corresponding class difference in life expectancy in the Northeast had fallen to about fifteen years. The biggest change, almost eleven years, occurred between the categories "half to one minimum wage" and "one to two minimum wages", whereas the big change in fertility over the same period was between the previous two categories of"below half minimum wage" and "half to one minimum wage". The fall in the birthrate thus occurred before that of the death rate as we move up the scale of family income per head - a fact which seems to weaken the hypothesis that birthrates fall as a lagged response to falling death rates. In sum, class differences in mortality are not as extreme as the fertility differences, nor is the dividing line so dramatic, nor at the same income level. Nevertheless, mortality differences add significantly to the total picture of the real dimensions of a class society. The demographic dimensions of class in Northeast Brazil may be summarized as follows.

1. In 1970 there was a class difference in total fertility of almost seven births and a difference in life expectancy at birth of around eighteen years.

2. In 1977 there was a class difference in total fertility of around 5.4 and a difference in life expectancy at birth of around fifteen years.

There has been a dramatic improvement over the seven years, even though there is still a long way to go. The change in mortality was an objective of public policy. The change in fertility was not sought directly by policy, but was a consequence of other factors. What other factors?

One cannot appeal to the usual "demographic transition thesis" because the economic condition of the lower eighty percent did not improve, yet that is where the bulk of the reduction in fertility occurred. Furthermore, as we have noted, the big fall in fertility occurred at lower levels of income per head than did the big fall in mortality. In fact it could be argued for the Northeast, as Merrick and Berquo have done for Brazil as a whole <20>, that the deteriorating real income of the lower class led them to reduce their fertility in order to maintain their rising consumer expectations. Alternatively, and perhaps more relevant to the Northeast, they suggest that the increase in the misery of the lower class may have lowered fecundity and the will to reproduce. The fall in fertility was not a consequence of improved economic conditions for the masses. Rather, it seems to have been the result of a certain democratization of birth control in terms both of attitudes and access to devices. This incipient spread of habits from the upper to the lower class was not a policy objective of the government - far from it. It seems to have happened in response to a latent desire on the part of the proletariat to control reproduction, once the practice received a certain legitimacy.

This increased legitimacy came from several sources. First, urbanization helped spread information. Secondly, the liberal wing of the Catholic church gave more emphasis to "responsible parenthood" than to the relative acceptability of alternative birth control methods. Thirdly, the practice of birth control by the upper class has a natural diffusion from donas de casa to empregadas, and from doctors to patients, because caring human relationships develop across class lines. Fourthly, pills and condoms are now generally available, the latter openly displayed in supermarkets, which was certainly not the case during the 1960s. Fifthly, there has been a general sexual revolution in the sense that sex moved from a mildly taboo subject ( Brazilians were never puritans) to an object of intense commercial exploitation via television (novelas and ads), cinema (the now famous Brazilian pornochanchadas), and magazines ranging from sexy to pornographic sold at every newspaper stand. It is interesting that in matters related to sex, censorship was practically abandoned about the same time that political censorship became intense. Some people argue that this was a ploy to give the illusion of a free press. Free pornography was an easy substitute for free speech. The unintended consequence may have been to hasten the spread of birth control by putting an intense commercial spotlight on sex and rendering legitimate the open discussion of everything related to sex, including birth control. <21>

The de facto attitude of the government is laissez-faire - private birth control initiatives are tolerated, but the official attitude is still pro-natalist. Some believe that the government would like to do a U-turn, as happened in Mexico, but that they are restrained from this by the fact that they do not want any more conflicts with the Catholic church, since they are already getting considerable criticism from the church on other matters of human rights and social justice.

In contrast to the notion that the government really wants birth control but is restrained by the political context, is the view that the government is genuinely pro-natalist and is not at all likely to change. Historically, Brazil has always followed a cheap labor policy. <22> In the early nineteenth century, slavery represented a policy of providing cheap involuntary labor from Africa. When slavery was abolished in the late nineteenth century, subsidized immigration of poor Southern Europeans provided a cheap source of labor well into the twentieth century, until such time as the natural rate of increase of the Brazilian working class was itself sufficient to guarantee cheap labor. The singular lack of enthusiasm shown by recent Brazilian governments toward a policy of spreading birth control among the working class (the upper class already practices it) may be seen as a continuation of the historical cheap labor policy, although in a passive rather than active mode.

Further support for the second view can be found in various official public pronouncements. In 1982, a study group of the Escola Superior de Guerra published a short book entitled The Brazilian Demographic Problem <23>, which reflects the official view, illustrated by the following quotations:-

"A country of continental dimensions such as Brazil, with fabulous natural resources, abundant wealth, and without prejudice in matters of race, colour or religion needs a population sufficient to occupy and defend its territory from international greed.

"The demographic policy implicit in anti-natalist campaigns judged by some to be absolutely necessary for our development, would result in the stagnation or regression of the growth of our population - we who already exist in such small numbers in a country so large and with inexhaustible resources. A policy of stimulating births thereby assuring more economic development will permit us to have more Brazilian workers, technicians and scientists. In addition to producing consumer and producer goods, with Brazil industrialized, our technicians and scientists will be able to transform our heavy industry partially into an industry capable of producing military goods." [page 75]

"From the military point of view, population is power, and in Brazil, in spite of Malthusian campaigns, population growth has historically served the country in that we have more rapid economic growth than that of countries with low natality, such as Argentina, Chile, and Uruguay." [page 72]

The authors of this booklet approvingly quote Minister Joao Paulo dos Reis Veloso's statement that "a country like Brazil has no right to create birth control programmes, because while our population grows at 2.5 per cent per year, our economic development grows at 9 per cent". [page 76]

Note that the last statement is precisely the misleading "rabbit stew" aggregation considered earlier, which I was at pains to refute in my paper in 1970 - and which also demonstrates that this "second look" is not uncalled for. Of course, the economic growth of 9% proved short-lived in any case.

To summarize, the basic conclusions of this "second look, fifteen years later" are listed below.

1. Northeast Brazil has probably the highest class difference in fertility of any society. The size of class differences in fertility was understated in my study in 1970, and consequently the arguments based on that difference now have stronger empirical support than was available at that time.

2. Between 1970 and 1977, fertility fell significantly in all classes. Class differences were reduced but remained very high by international standards. The effect of differential fertility on income distribution remains very important in spite of the real progress made in lowering fertility.

3. A Marxian-Malthusian definition of social class, in terms of control versus non-control of both production and reproduction, fits the Northeast, and offers a possibility for integrating the valid insights of both traditions. This is important because with the current rebirth of Marxist economics in Brazilian universities, Malthusian insights are in danger of being lost or discarded along with the brittle analytical bones of value-desiccated neoclassical models. The democratization of control over reproduction is no less (and no more) important than the democratization of land ownership in the Northeast. Everyone talks about land reform, but so far few talk about reproduction reform. Paradoxically, reproduction reform seems to be actually occurring faster than land reform.


I am grateful to the Fulbright Commission for a lectureship that allowed me to spend three months in Northeast Brazil in 1983. Also, for discussions and suggestions I am indebted to my colleagues at the Universidade Federal do Ceara, the Fundacao Joaquim Nabuco, and the Institute Brasileiro de Geografia e Estati'stica. Responsibility for all points of view and any errors, of course, rests with me.

1. "The Population Question in Northeast Brazil: Its Economic and Ideological Dimensions", Economic Development and Cultural Change, July 1970; "A Marxian-Malthusian View of Poverty and Development", Population Studies, May 1971.

2. Perfil Estatistico de Crinacas e Maes no Brasil: Caracteristicas Socio-demogrdficas, 1970-1977 (Rio de Janeiro, IBGE), 99. This study will henceforth be cited as "Perfil'."

3. Perfil, 89.

4. See comment by Yony Sampaio, with my reply, in Economic Development and Cultural Change, January 1976.

5. PNAD is an acronym for Pesquisas Nacionais por Amostra de Domicilio (National research based on household sample).

6. See no 3, above.

7. Total fertility is the number of live births per woman if she survived to menopause and were subject to the age-specific fertility rates currently prevailing in the population in question.

8. For 1970 the child mortality below two years of age was 192.3 per thousand, and in 1977 154.7 per thousand (Perfil, 55).

9. Thomas W Merrick and Elsa Berquo, The Determinants of Brazil's Recent Rapid Decline in Fertility (Washington, DC: National Academy Press, 1983), 24.

10. Charles H Wood and Jose Alberto Carvalho, "Population Growth and the Distribution of Household Income: The Case of Brazil", The Sociological Quarterly 23 (Winter 1982): 53.

11. J Mayone Stycos, "Social Class and Differential Fertility in Peru", Proceedings of the International Population Conference, New York, 1961, vol 2 (London, 1963), 123-38.

12. Jorge H Zambrano Lupi, "Fertility and Educational Status in Mexico City" (in Spanish), Demografia y Economia 13 (4), no 40 (1979): 442.

13. Kanti Pakrasi and Ajit Haider, "Fertility in Contemporary Calcutta: A Biosocial Profile", Genus 37 (3-4). (July-December 1981): 201-19.

14. William Petersen, Population (New York: Macmillan Co, 1975), 527.

15. Wood and Carvalho, "Population Growth", n 11, see their Table 1, 54.

16. Perfil, 157.

17. Joao Lyra Madeira, "Migracoes Internas no Planejamento Economico", in Migracoes Internas no Brasil, edited by Manoel A Costa (Rio de Janeiro: Institute de Planejamento Economico e Social, 1971), 42.

18. See note 1 above.

19. See Karl Marx, Capital, chapter 33, "The modern theory of colonization", pages 379-83 in Great Books edition (University of Chicago, 1952).

20. Merrick and Berquo, Determinants, 82, 83.

21. For an interesting study of changing sexual attitudes in different social classes, see Rose Marie Muraro, Sexualidade de Mulher Brasileira (Editora Vozes, Petropolis, R J, Brazil, 1983).

22. Nathaniel H Leff, Underdevelopment and Development in Brazil, 2 vols ( London: George Allen and Unwin, 1982). See especially chapter 4, volume 1.

23. O Problema Demografeo Brasileiro (Associacao dos Diplomandos da Escola Superior da Guerra, Grupo 05, Belo Horizonte, MG, 1982). (My translations.)

Bill Totten


Post a Comment

<< Home