POPULATION/FAMILY/GENDER

GARY ''The Big Enchilada'' BECKER"
Treatise on the Family, Chapters 2 and 5

My oft-referenced friend Jeff says he finds thinking of Human Capital theory as an ideal type (Paula England apparently describes it this way somewhere) makes it much more palatable. I guess we're all entitled to our opinions. I think of HC theory more along the lines of an assessment of psychoanalytic theory I read the other day: it sounds neat and tidy, but there isn't much evidence for it.

Chapter 2, Division of Labor in Households and Families.

The various divisions of labor among family members are determined partly by biological differences and partly by different experiences and different investments in human capital. Specialization in the allocation of time and in the accumulation of human capital would be extensive in an efficient family even if all members were biologically identical.

Since married women have been specialized to childbearing and other domestic activities, they have demanded long-term contracts from their husbands to protect them against abandonment and other adversities. Shirking, pilfering and cheating are made easier by the extensive specialization and division of labor within families. These conflicts of interest can be reduced by monitoring behavior, including invasions of the privacy of members, expulsion and other punishments, and altruism.

Women not only have a heavy biological commitment to the production and feeding of children, but they also are biologically committed to the care of children in other, more subtle ways. Moreover, women have been willing to spend much time and energy caring for their children because they want their heavy biological investment in reproduction to be worthwhile. There is a complementarily between bearing and rearing the little darlings, since a mother can easily feed and care for older children while preggo with another (we gals are so darned efficient). An hour of household time or market time of women is not a perfect substitute for an hour of the time of men when they make the same human capital investment. Women have comparative advantage over men in the household when they make the same investments in human capital. An efficient household will allocate women's time mainly to the household, and men's time mainly to the market.

It gets better, though. Men and women are not only not perfect substitutes (because of biological differences in comparative advantage), but they are complementary. Because of complementarily, households with men and women are more efficient than households with only men or only women (My personal comment: this does not explain, though, why single sex households may have divisions of labor that follow traditional gender lines).

Since investment differences reinforce biological differences, bio comp adv cannot be easily disentangled from specialized investment. Specialized investments begin very early (while boys and girls are very young); they are made prior to the full knowledge of the biological orientation of children. It may be that there is some small proportion of ''deviant'' boys and girls (those biologically suited to household labor and market labor, respectively); however, since most boys and girls are not deviant, the optimal strategy in this condition of less than perfect foreknowledge of children's biological orientation, would be to make the usual gender-specialized investments. Probably a large fraction of deviants remain single, marry and then divorce, or remain in unsuccessful marriages.

The biological differences b/t men and women in the production and care of children, and the specialized investments in market and household skills that reinforce the biological differences, explain why the institution of marriage has been important in all societies.

People tend to have a preference for their own children. Women producing children can use their own milk as food and can more readily take care of young children while pregnant than while working in the marketplace. Additionally, most women have been reluctant to go to all the trouble of having children (investments) without having considerable control over rearing them. Also, own children are preferred because information is readily available about their intrinsic characteristics (they have half your genetic material), and their characteristics can be directly observed at birth and during infancy.

Chapter 5, The Demand for Children

Though some theories (eg, Darwin) might make one think that parents have an evolutionary ''push'' to have as many children as possible, there is an interaction between quality and quantity. A reduction in the number of children born to a couple can increase the representation of their children in the next generation if this enables the couple to invest sufficiently more in the education, training, and attractiveness of each child to increase markedly their probability of survival to reproductive ages and the reproduction of each survivor. Thus, the family maximizes a function including both quantity and quality of children.

Farm Children Farm families have been larger than urban families for hundreds of years. Part of the explanation is that food and housing have been cheaper on farms. In addition, the net cost of children is reduced if they contribute to family income by working at chores, working in the family business, or working in the marketplace. Then an increase in the earning potential of children would increase the demand for children. Farm families have had more children because children have been considerably more productive on farms than in cities. The contribution of farm children has decreased as agriculture has become more mechanized and complex. Both these factors have encouraged farm families to extend their children schooling. The cost and time of transport to school is larger for farm children, so the cost advantage of raising children on farms has been reduced, if not reversed, thus, farm families are having fewer children now.

AFDC. Programs providing aid to mothers with dependent children have reduced the cost of children; aid increases as the number of children increases, and the decline in the labor force participation of mothers induced by these programs reduces the opportunity cost of the time spent on children. Since mothers w/o mates have more readily qualified for aid, these programs have encouraged illegitimate births.

Married Women. The growth in the earning power of women during the last century in developing countries is a major cause of both the large increase in the labor force participation of married women and the large decline in fertility (the relative cost of children is affected by changes in the value of the time of married women).

Major changes in average fertility have not been caused by changes in sterility and knowledge of birth control. Improvements in birth control methods are mainly an induced response to other decreases in the demand for children, rather than an important cause of the decreased demand.

The demand for children is affected both by their cost and by real income. This is why in earlier times, wealthier western families had more children. However, sometime during the 19th century in the west, fertility and wealth became partially or wholly negatively related among urban families.

There is an interaction between the quality and quantity of children in a household. This interaction explains why the education of children, for instance, depends closely on the number of children, even though we have no reason to believe that education per child and number of children are close substitutes. The demand for children is highly responsive to price and perhaps to income, even when children have no close substitutes.

Jews. The high achievement and low fertility of Jewish families are explained by high marginal rates of return to investments in the health, education, and other human capital of their children that lower the price of quality relative to quantity.

Blacks. Blacks invest less in training because their rates of return on investments in health, education and other training have been lower than for whites. The quantity-quality interaction implies that blacks would respond to poorer investment opportunities with higher fertility. As opportunities for blacks have improved in recent years, they have invested more in their training and at the same time reduced their fertility relative to whites.

Educated Women. Since educated women have a lower demand for quantity of children, the interaction with quality implies that they would invest more in the education and other training of their children. The research showing positive relations between mother's education and child's education, this may not be evidence that the causation is from mother's education to child's education.

Economic development. Economic development effects fertility and the quality of children not only because incomes increase, but also because rates of return on investments in education and other human capital increase. Since even a 'pure' rise in income can reduced fertility through the interaction with quality, a rise in income combined with higher rates of return on quality could reduce fertility significantly. Thus, economic development can have significant effects on fertility.

JOHN BONGAARTS
''Framework for Analyzing the Proximate Determinants of Fertility''

Since studies aimed at examining the causes of fertility levels and their changes (especially as regards socioeconomic factors) are highly relevant as policy issues, it is important to construct a quality model of the factors involved. This is where Bongaarts (Bongo to his friends) comes in. He proposes a statistical model in this paper that attempts to define the relationship between intermediate fertility variables (IFV) and fertility. IFVs are biological and behavioral factors through which socioeconomic, cultural, and environmental variables affect fertility. The primary characteristic of an IFV is that it directly influences fertility (don't ask me why it is ''intermediate''). Anyhow...Differences among populations and trends in fertility over time can always be traced to variations in one or more of the IFVs. It works like this:

Indirect Determinants:	    Direct Determinants:
Socioeconomic, 		    Intermediate
cultural,     	----------+   fertility	        --------+Fertility
environmental		    variables
variables

''Natural fertility'' refers to situation/population where coupled do not practice deliberate fertility control dependent on the number of children they have. The maximum reproductive potential of even a naturally fertility population is never actually reached due to intervening factors (like IFV).

Here are the IFVs and what we need to know about them:

I. Exposure Factors

1. Proportion Married: intended to be a measure of the proportion of women of reproductive age that engage in sex regularly; the prevalent pattern of marriage in a population effects the character of women's reproductive lives; JB figures that: ,br> Total fertility rate (TFR) = index of proportion married (Cm) X total marital fertility (TM)
Empirical evidence points to the importance of proportions married in determining fertility variation, as countries with high fertility rates tend to have high proportions married (and vice versa).

II. Deliberate marital Fertility Control Factors

2. Contraception: responsible for the wide range in levels of fertility within marriage in contemporary populations. JB expresses marital fertility as the interaction of contraceptive practice (Cc :accounting for both prevalence of use and effectiveness) and natural fertility:
TM = Cc X total natural marriage fertility rate (TNM)
Empirical findings show that the highest natural fertility rates are found in developed countries, where higher use of contraceptives tends to offset this effect on overall fertility.
3. Induced Abortion: JB goes into some math modeling to address this factor, the main points being:
a) induced abortion always averts less than one birth (early return to ovulation)
b) number of births averted per induced abortion is largely independent of the age of the woman
c) number of births averted per induced abortion is strongly influenced by practice of contraception following the abortion.

III. Natural Marital Fertility Factors

4. Lactational Infecundability: lactation has an inhibitory effect on ovulation and thus increases the birth interval and reduces natural fertility. Birth interval has 4 components: a) infecundable interval immediately following birth {return to ovulation}; b) waiting time to conception; c) time added by spontaneous intrauterine mortality; d) nine-month gestation ending in live birth. Without lactation the birth interval is about 20 months. Lactation infecundability increases a) from above and in some cases can nearly double the birth interval.
5. Frequency of Intercourse: according to JB, coital frequency is not a very important determinant of fertility differences among populations ( ie: everyone's doin' it - at least as much as anyone else)
6. Sterility: *
7. Spontaneous Intrauterine Mortality: *
8. Duration of the Fertile Period: *
* These three are physiological factors assumed not to be under the control of individuals. Little is known about the genetic factors at work here, but two environmental factors (health and nutrition) are often considered significant determinants of fertility. There biological factors generally have little demographic importance - with three exceptions (only the first of which can show a particularly large effect):
a) high incidence of venereal disease in a population can affect fertility rates via sterility and high spontaneous intrauterine mortality
b) age at menarche is partly determined by nutrition
c) prenatal health can influence the stillborn rate.

That is the important stuff in a nutshell. JB also stresses the point that partial application of his model is possible. So some calculations are possible even if some categories of data are missing or lacking in quality. Finally Bongo states that analysis indicates that variations in four factors - marriage, contraception, lactation, and induced abortion - are the primary proximate causes of fertility differences among populations.

GEORGE BORJAS AND MARTA TIENDA
The Economic Consequences of Immigration

U.S. Immigration in Historical Perspective

Before the 1965 Amendment to the Immigration and Nationality Act (of 1952), the law held in place discriminatory national quotas which favored immigration from northern and western Europe. With the Amendment, discriminatory national quotas were abolished, and an emphasis was placed on family reunification (immigration for unification purposes is exempt from quotas). After 1965 the socioeconomic and regional composition of immigrants changed markedly. Immigrants were made up less and less of Europeans and more and more of people from Asia and the Americas. Refugee adjustments, also not subject to quotas, increased the number of Asian and Latin American immigrants (there just aren't too many refugees from Sweden).

Changed entry requirements, most notably the family reunification policy, have resulted in a bimodal skill distribution, with clusters in both white and blue collar occupations; this reflects the preponderance of Asians in white-collar occupations and Hispanics in blue-collar occupations.

Then Borjas and Tienda do various analyses to determine the impact of immigrants on the US economy, in particular addressing questions of whether immigrants affect the job opportunities and earnings possibilities of qualified Americans and whether they make disproportionate use of transfer programs. They conclude:
1) Although the volume of immigration has increased appreciably in recent decades, there is no basis for concluding that it has exceeded the growth rate or absorptive capacity of the US labor force.
2) There is a negative impact of immigrants on the earnings of native workers, but it is very small.
3) Immigrants who arrived during the 1970's are, on average, less skilled than those who arrived earlier, and their earnings do not rise as rapidly as previously thought 1, but this generalization varies by national origin. This finding is consistent with descriptive historical accounts about the changing socioeconomic and demographic composition of immigrants admitted since 1968. Two qualifications are in order: First, if the labor market experiences of recent immigrants, if judged to be different from those of earlier arrivals, may also reflect the sluggish character of the US economy during the 1970's, particularly the mid-period recession. Second, the observed cohort effects may reflect selective emigration or improved enumeration practices.
4) Hispanic and Asian immigrants as a group are more likely to receive transfer payments than are natives. This result reflects the greater pervasiveness of poverty among the foreign born rather than a higher propensity of foreign-born persons to participate in public assistance programs compared to (statistically) comparable natives.

LARRY L. BUMPASS
''What's Happening to the Family? Interactions between Demographic and Institutional Change''

In spite of the baby boom and other pronatalist forces encouraging reproduction, subreplacement fertility is still the norm in Western industrial societies; thus, the demographic transition is continuing in the West. Note: 1.) replacement fertility = the fertility level necessary for women to bear exactly enough daughters, given mortality, so that exactly 1 daughter per woman will reach reproductive age; 2.) the demographic transition = the shift from high levels of fertility and mortality to low levels of fertility and mortality.

Because fertility cannot be isolated theoretically from the institutional context in which it is embedded, theories about fertility decline are intrinsically also theories about changes in the family as an institution. It is important to regard family and fertility behavior as dynamic outcomes of profamilial and antifamilial factors. Changes in one family domain often contribute to further changes in that domain or in others.

Important Issues in the Study of Family and Fertility:

1.) Marital Disruption:

As most of us already know, the current level of marital disruption (mainly divorce) is very high, and this is true across all demographic and religious groups. The most important consequences of this are single-parent, especially single-mother, families. This instability challenges family lifestyles and norms, particularly those involving children. Divorce illustrates the force of secular individualism quite clearly. The interests of parents are becoming increasingly more legitimate as a reason for divorce, over the interests of the children. Insecurity about the viability of marriages may lead to reduced investments in some marriages which, in turn, lower the prospects that they will last (Becker, 1981). Hence, there may be feedback from demographic behavior to the institution of marriage itself.

2.) Cohabitation and Marriage:

The increase in cohabitation stems from the erosion of normative objections and the revolution in the sexual experience of unmarried persons. In addition, as the fragility of marriage is recognized, couples sense the need to ''try out'' marriage beforehand. Some implications of increasing frequency of cohabitation: 1.) a new meaning of being ''single''; 2.) sex, living arrangements, and parenting depend less on marriage; 3.) ''premarital divorces'' (i.e., breaking up before marriage) have helped to keep the divorce rate from getting higher; 4.) relationships of convenience; and 5.) both marital instability and cohabitation contribute to the decrease in the perceived necessity of marriage.

3.) Childbearing and Parenting:

The total fertility rate has been stable in the US over most of the two decades (70's and 80's) when marriage behavior has been changing so radically. One aspect of this trends the increase in nonmarital childbearing; which is evidence of , and contributes to, the erosion of norms against traditional behavior. If marriage assures neither a stable two parent family nor lifetime economic security for women, then the importance of marrying to ''legitimate'' a birth is not very compelling. Falling rates of marriage and remarriage and high rates of marital disruption have lengthened the unmarried period for risk for unplanned childbearing.

As we get more advanced birth control, there will be a greater delay in childbearing, which tends to lead to declines in intended fertility as well. Planned fertility will also decline because of: 1.) marital instability, which increases the costs of having children, and 2.) complex work schedules.

Bumpass reasserts the idea that there is an increase in emphasis on individuation as people tend to make decisions based on self-interest rather than child welfare (e.g., day care, working parents).

4.) Marital Relationships:

What men and women want out of marriage may be diverging, making it harder for bargains to be struck in the marketplace and contributing to lower marriage rates. Husbands and wives may have increasingly different expectations about the division of household labor and whether or not the wife should work. Most people agree that the household division of labor is unfair to working wives, but men in general still don't do half of the housework. Since women who think this division of labor is unfair are more likely than husbands to report trouble in the marriage, the extra load of work on women may be endangering marriages.

5.) Intergenerational Relationships:

Intergenerational relationships are a major source of social connectedness and support. However, these relationships are being altered by changes mentioned above. The full effect of marital instability on Intergenerational relationships will not be seen until the cohorts that have experienced the high instability of the last two decades approach retirement.

Conclusion:

Trends in cohabitation, marriage, fertility, and marital disruption are widely shared across Western industrial societies. Major causes include the individualizing tendency of participation in our economy and cultural values of individualization that both facilitate this participation and are reinforced by it. There is no reason to think that these processes are exhausted or are likely to reverse. Hence, the demographic transition in the West is likely to continue.

Note: Bumpass and Cherlin are both talking about similar trends in the family. However, Cherlin would question Bumpass' point of reference, arguing that these trends aren't all that unusual in the context of overall 20th century trends. In addition, Bumpass emphasizes changing values (a period effect) quite heavily. Cherlin looks at both period effects (such as values) and cohort effects (such as growing up during a specific economic period).

THOMAS BURCH AND BEVERLY MATTHEWS
''Household Formation in developed Societies''

Demographic changes in developed countries over the past 25 yrs:

1)decline in fertility to below-replacement levels
2)postponement/avoidance of marriage and increase in non-married cohab.
3)rise in divorce and instab. of nonmarital unions
4)rise in proportion of persons living in small households; decline in avg. household size

Household status is an instrumental good -- a means to various ends. Marriage decisions are increasingly subordinated to residence decisions. The emphasis is now on practical arrangements for everyday living, rather than kinship ties. There is a trend toward increasing simplicity of household ties.

Household Status as a Composite Good

Household status = number of persons with whom an individual shares a living unit. It is instrumental. Goods people seek when making decisions about household status:
shelter; care; independence; storage; companionship; economies of scale in food; household items; services; recreation; power

People may seek a combination of these, according to their economic status

Paying for household goods

Household formation involves a process of social exchange of household goods beyond those considered in micro-economics, such as companionship, labor, and subordination. The exchange of these goods has become problematic as people have become more entitled to income in their own right.
There is also the problem of the residential obligations of kinship -- sharing a household when one would rather not. Certain traditional kinship relations persist, even legally. Peoples' inclinations and social norms don't always point in the same direction (i.e., you hate your kid but you can't legally kick him out til he's 18). An adequate behavioral model must allow for situations of conflict between rational decisions, individual inclination, and social norms.

Some explanatory hypotheses
(these are not mutually exclusive; they are interactive and additive)

1)rising real income
As real income increases, so does the trend toward separate living. People can afford to forgo economies of scale from larger households. New institutions have increased the proportion of the population that can claim its own income (e.g., work for women, social security for the elderly)

2)availability of kin
Kin is still a constraint (legally) on residence arrangements. However, low fertility rates and high mobility rates (moving around and leaving relatives due to job opportunities, etc.) have reduced the chances that you'll live with large numbers of relatives.

3)changing preferences for privacy
The demand (not necessarily the preference) for privacy in the household may have increased. Possible contributing factors:
Protestantism/Enlightenment
postmaterialism (with inc. in ec security, indiv. becomes more introspective)
sexual revolution of 60's vs. parental expectations
opportunities for privacy outside home have declined (urbanization)

4) Role changes and household crowding
As men and women occupy more and more similar economic roles, competition in the household increases. Like Durkheim would argue, there is a decline of solidarity doe to a breakdown in the household DOL.

5)decline of household services entry of women into labor force makes co-residence less desirable.
decline in quant and qual of household services
increasing pressure on men to provide household services

6)technology, recreation, companionship
As technology develops and urbanization increases individuals are less dependent on one another for companionship (telephone, car, urban proximity)

Conclusion

The authors admit that their ideas are sketchy, and could be falsified quickly. However, there is a need for the clarification of a composite good that people seek in a household. Demographers need to find a middle ground between theory and the realism of family life.

GLEN CAIN
''Labor Force Concepts and Definitions in View of their Purposes'' (1980)

This reading centers around the debate concerning the adequacy of labor force statistics (which actually means the national unemployment rate).

Three major criticisms of the unemployment rate (UER) are central:
1) As a macroeconomic indicator, it is failing to measure conditions in the labor market regarding the demand for labor as accurately as it has in the past. The reasons cited for this effect are: a) high levels of unemployment since 1971 combined with inflation; b) the demographic shift to increased LFP by women and young people (who have higher ''natural'' unemployment rates than men); c) transfer payments increased as a source of income. In response, Cain believes that the unemployment rate remains adequate for its primary purpose, ie: a measure of cyclical changes in the real (as distinct from monetary) performance of the economy.

2) Unemployment rates no longer measure hardship as they did 10 to 20 years in the past because: a) changing demographic composition of the labor force means that a smaller proportion are primary wage earners; b) more families have more than one earner; c) increase of transfer payments. Cain says: ''Duh, '' first of all, as we saw above UER is not supposed to be a measure of hardship; and second, hardship applies to a household over a year's duration, while labor force calculations generally apply to individuals at a certain point in time.

3) Labor force stats are inadequate to measure labor market conditions for local areas, and therefore do not meet governmental needs for allocating federal aid to areas in various conditions of economic distress. Cain grants this criticism, but suggests (and later attempts to demonstrate) that the available alternative measures are inferior to unemployment rate (especially when the latter is modified).

In general, labor force stats are intended to serve three important purposes:
1) Measure the overall performance of the economy and provide signals to authorities responsible for monetary and fiscal policy.
2) Unemployment stats are directly related to transfer payments for those who have lost their jobs.
3) Used to allocate federal aid or expenditures to local areas
(We can see that some of the points of criticism above address these main purposes of labor force statistics.)

Building blocks of labor force statistics:
1) a population: those eligible to be in the labor force - non-institutionalized adults (age 16+)
2) an employed component: working for pay or profit during the reference week of survey (ie. CPS)
3) an unemployed component: not working but ''looking for work'' during specified 4-week period prior to and including the survey week

The 1961 Gordon Committee (assigned the task of investigating the adequacy of labor force stats) presented five criteria for the operational definitions of these labor force categories: a) objectively measured, b) obtainable at reasonable cost, c) readily understood, d) easily interpretable and sufficiently flexible to accommodate the needs of different users, e) market oriented, and f) keeping concepts constant over time to facilitate analytic work using time series data. (Cain tends to bring up these criteria when comparing other alternative stats to the UER that he feels better meets these requirements as a whole.)

One of the main shortcomings of UER is that it is highly subjective: relying on survey responses as to employment status, what constitutes a ''meaningful job'' or a ''sincere and authentic search'' for work.

As state previously, the main purpose of employment stats is to measure the overall performance of the economy and thereby guide macroeconomic policy - this enfolds three separate purposes : as a short-run forecast, an inflation indicator, and a measure of ''real'' economic performance. Cain states that only the last is served moderately well. In particular, the UER is useful as a measure of economic performance during the business cycle and the short run of several years (while a statistic such as LFP rate is better suited to measuring long run performance).

Although not intended to measure economic hardship, there are several reasons why unemployment statistics for local areas offer useful information for the allocation of federal aid:
1) UE stats based on unemployment insurance records may offer the single most valid and reliable indicator of economic distress for a local area, particularly for periods shorter than a year.
2) Unemployment generally connotes a) decline in one's income and well-being and b) an attachment to the labor force - both of which are relevant to many federal assistance programs.
3) Despite the limitations of UER as a measure of individual economic status: a) it can capture the - often high - personal cost of unemployment, as may effect skills and confidence; and b) the family unit is not permanent and the unemployed person may not always share fully in the family's income.

Cain concluded here that UE should not be relied upon heavily for measuring hardship and that the unemployment insurance program should be shifted more toward short-run support for active job-seekers.

Reviewing a variety of empirical evidence, Cain finds that - ceteris paribus - changes in demographic conditions have resulted in a higher unemployment rate in 1975/6 as compared to a point 20 years earlier. He also argues in favor of adjusting the UER in terms of age rather than other potential compositional factors (such as race, sex, etc.) since the former has a clear causal connection to unemployment, better relates to secondary issues of hardship, is generally accepted, and is less value-loaded than the alternatives.

Cain believes that on balance, transfer payment programs have probably increased the unemployment rate not just because they offer some incentive to be not employed, but more importantly because the registration required by these programs required those who may previously been not in the labor force (NLF) to become by definition ''unemployed'' - which artificially inflates the ranks of the unemployed. In sum, the recent increase in income transfer payment programs: 1) raises the ''frictional'' or ''natural'' level of unemployment, and 2) diminishes the overlap between unemployment and hardship.

In one of the last sections of this reading, Cain embarks on a critical evaluation of several alternatives to more traditional labor force statistics (the unemployment rate) - these include the employment/population ration, the Perry Index of Unemployment (something of a wage gap index), and two separate techniques for measuring labor-market related economic hardship. The discussion gets fairly complicated and detailed - and most importantly is not too relevant for our purposes. Suffice it so say that Cain concludes that UER is for various reasons still a superior (albeit not perfect) measure in comparison to the alternatives in assessing the performance of the labor market.

In conclusion, Cain makes a number of suggestions and recommendations - which although repetitious, I will recount because they emphasize some of the author's conclusions in sections of the paper not addressed in much detail here:
1) The main labor force statistics should be computed with a fixed age-distribution of weights and should be reported by year.
2) ''Discouraged workers'' should remain classified as NLF, preserving the ''active search'' criterion in defining unemployment.
3) The growth of the labor force attributed to the increased employment of students and women has led to a larger proportion of part-time workers in the labor force - a trend that may also become increasingly more common among older people.
4) The relations between family income, the sources of income, and transfer payments, on the one hand, and unemployment, on the other hand, require concurrent measures of these variables.
5) The relation of unemployment to nonmarket activities (school, training, child rearing, etc) needs to be studied to answer the question of how these activities affect the levels, trends and cycles of unemployment.
6) Continued analysis of search behavior of the unemployed is required to address the persistent criticism that the subjectivity involved in the responses defining unemployment makes them unreliable.

JOHN CALDWELL
Toward a Restatement of Demographic Transition Theory

Overview:

Demographic transition theory as espoused over the last 25 or so years, most notably by Frank Notestien, is wrong. Its emphasis on cultural props (traditional beliefs, religious superstitions, etc.) is misplaced. Both pre- and post-transition peoples behave rationally, even economically rationally, but ethnocentrism has blinded Western researchers to this fact.

Caldwell's Main Idea:

There are two types of fertility regime (with the exception of the situation at the time of transition): one where there is no economic gain to individuals from restricting fertility, and one where there is often or eventually economic gain to individuals from such restriction. The main determinant of this difference is the direction of net intergenerational wealth flows between parents and children. If net flows over the lifetime are in the direction of the parent, the highest fertility is rational. If net flows over the lifetime are in the direction of the children, zero fertility is rational. What prevents high fertility regime families from having the biological maximum number of children are social and psychological factors (such as, if you have 14 children running around they get on your nerves; or, the woman may declare she's done at 10 because she feels ''increasingly old and battered by reproduction''). What prevents low fertility regimes from having no children are, again, social and psychological factors.

Outline:

The mainstream arguments of Dem Tran theory, as outlined by Notestein, are that fertility is high in poor, traditional societies because of high mortality, lack of opportunity for individual advancement, and the economic value of children. In this body of theories, there is an implicit or explicit assumption of pre-transition irrationality (attitudes, beliefs, values, social institutions are believed necessary to keep fertility high, since high fertility is assumed to not be rational). Thus, when we find cases of high fertility persisting in new urban, industrial conditions, this theory would tell us that this is because of cultural lags.

What this theory misses, with its Western view of the world, is that the fundamental choices are social ones, and economic behavior is rational only insofar as it is rational within the framework established by social ends. We need to make the assumption that all societies are economically rational.

Further, when we consider past research bearing on the demographic transition, we need to recognize that is has often had the following flaws when applied to non-Western situations:
1) The magnitude and direction of wealth flows (money, goods, services, obligations, guarantees) has often been ignored.
2) The family of the fertility survey is often an artifact of the survey. No one considers the intricate system of decision-making and obligations that may far exceed the nuclear family or residential group and in which the nuclear family may not even be a recognizable subunit.
3) The nature of family formations and of related decisions in developing countries is frequently misunderstood. Family size decisions are usually out of the respondents' hands for a variety of reasons.
4) While fertility research is essentially a study of change, such investigations have been impeded by too much emphasis on modernization. Change can be understood only if emphasis is given to studying the fundamental nature of the society that is being subjected to new forces.

Then, using the Yoruba of Nigeria as an example, Caldwell talks about three kinds of society, Primitive, Traditional and Transitional:

1. Primitive society. Primitive society is where the largest organizational institution is the tribe, clan or village. The society or economy for the group is a single system in which the participants have time-honored roles and duties. There is usually communal land. Transactions and gifts are not very different from one another, especially since the latter are almost always the cause of two way flows of wealth; there is a security system of mutual obligations. Maximization of profit or other ends in good times is of small importance compared with the minimization of risks. Women will often forgo deserting or divorcing their husbands because of the prospect of a secure and relatively care-free old age of being cared for by their children. One lives with or very near to most of one's relatives; it is inconceivable that the nuclear family should crystalize out and should attempt to gain economic advantage over the larger unit.
It is the survival of the extended family system as economic change occurs that helps to sustain high fertility. This survival is rendered more likely by a system of mechanisms which retain the full vigor of the extended family system. Society is aware that conjugal sexual relations can intensify conjugal emotional relationships and that parent-child emotions can also become of overriding importance. Therefore, cultures successfully weaken both types of relationships; because, otherwise communal residence and occupational cooperation would be jeopardized. Networks of relatives are important in primitive and traditional societies both because they increase the size of the security system (risk management), and because they increase the number of close allies in political contests, where success is predicated on being able to tap more or better communal resources. They also increase the number of relatives who can attend celebrations and increase one's social importance and sheer consumption pleasure. Kids are also an important source of labor.

2. Traditional Society. These have greater overall organization (states,the Church, weakening of extended family, family no longer sole institution). This is a stage applicable in other parts of the world, but a ''distinction hardly worth making'' for the Yoruba.

In both Primitive and Traditional societies, net intergenerational wealth flows are from child to parent. Children do a great deal of work for their parents not only when young, but also during adulthood; they accept responsibility for parents in old age; they eventually bolster the family's political power and given it economic advantages; they ensure the survival of the lineage or family name and in many societies undertake the necessary religious services for ancestors.

In these relatively unchanging, non-introspective societies, no one recognizes these separate bonuses conferred by fertility; people just behave as they always have.2 Nevertheless, there are at least six economic advantages of kids in these societies:
1) Situational gain is of particular importance to patriarchal males (they're at the top of the pyramid of the family, control more people, control more wealth and power).
2) Children work providing goods and also services adults are loath to perform.
3) Adult children assist their parents with labor and gifts.
4) Adult children provide much of the family contribution to various ceremonies.
5) Adult children care for aged parents.
6) Parents can invest in kids' education to better their returns (ie, the educated kid can get a job in the city and send back bigger checks).

Economic rationality dictates maximum fertility. Social and emotional factors lead people to limit their fertility.

3. Transitinoal Society. The divide between high fertility and low fertility is not mechanistically determined by economic conditions. It is almost entirely a social phenomena, and can be reached only when the economy of the nuclear family has been largely isolated from that of the extended family and when a subsequent change of balance has occurred within the nuclear family. The necessity for economic nucleation arises in emotional nucleation, which is incompatible with the extended family economic system and its requirement of a parallel system of emotional obligations. The change of economic balance in the nuclear family means that the parents of the family are wholly in charge of their own family economy.

If we want to understand the transition from high to low fertility, we should study innovators (people who have lower fertility in higher fertility settings) and see how they differ. In Nigeria, they are more likely to be educated, more likely to be more highly educated, more likely to have husbands with non-manual jobs, and more likely to have had fathers with non-manual jobs. The innovators have been emotionally nucleated from their families, and have a greater concern with their children, and their children's future.

The cause of emotional nucleation IS NOT urbanization or industrialization, BUT IS Westernization, the importation of a different culture. Key is Christianization, as well as the mass import of European manners and models via mass media and mass education. The family, as taught by the school, is almost entirely the Western family (since virtually all the textbooks are from England). In the media, the great importance of sexual relations is touted; this cannot fail to affect the traditional system of family relations, by tending to strengthen the conjugal bond and so nucleate the family first emotionally, and then economically.

In historical Western Europe, the family was increasingly economically nucleated many, many years ago. This had the effect that Europe's population growth was lower than it would have otherwise been once mortality began to decline; and that the nuclear family acted as the basic unit of society.

Some comments on the export of the European social system as well as its economic system. This export has made both fertility and mortality declines possible in the Third World. However, the whole system of extended family obligations and the direction of net intergenerational wealth flow in favor of parents can be disrupted by political means (eg, China) with the same effect of reducing fertility.

An important thing to think about that goes in the face of some of standard demographic transition theory: maybe fertility and mortality levels are set independently.

JOHN CALDWELL
''Routes to Low Mortality in Poor Countries''

The focus of this paper is on how some modern, developing countries have achieved low mortality, and whether their routes could be followed by other poor populations. Countries may do this by political and social will, or, in the case of China, by political will alone. The route leads through exceptional inputs in the areas of education, health services and nutrition. Also important are ''breakthrough periods,'' periods of exceptional advancement.

Caldwell makes a list of third world countries, dividing them into poor and good health achievers. He characterizes them broadly, and then considers two countries and a state in more detail. The countries we will consider in detail, Costa Rica, and Sri Lanka, and the state of Kerala in India, have a mortality experience that is markedly at variance with what might seem to be dictated by the economic determinism exerted by per capita income levels. It is possible, then to break the economic shackles. They are all characterized by breakthrough periods since WWII.

In no country which performs well in health (low mortality, despite its poverty) is Islam culturally or politically dominant. Poor performing countries are overwhelmingly Muslim, or have large Muslim minorities. Islam is argued to have its effect through the position of some, low levels of female schooling, low levels of family planning, and limited access to employment outside the household (which means fewer women become nurses, a fact which has a direct impact on health services in the country). The enrollment of boys in schooling is also lower in Islamic countries than their economic rankings would suggest; there is Koranic schooling, but Caldwell argues that the effects of school have less to do with institutionalization than with the fact that modern schooling provides contact with the whole modern system.

-World: Some place mortality decline has been (unexpectedly) slowing down
-Ultimate constraints on mortality decline are economic
-examines Kerala (India), Sri Lanka, Costa Rica, and China
*How have these countries (state) achieved such low mortality when they are so poor?
*Can other countries (poor) in the ''third world'' follow their routes?

Questions Asked by Caldwell
1. How exceptional are their experiences?
2. How do these countries (with low mortality) contrast with countries with high mortality?
3. Was there a breakthrough period in achieving good health? How was it achieved?
4. Under what circumstances can political and social will be exercised?
5. For countries with different histories, are there policies that can be implemented to accelerate mortality reduction?

Variables Determining Health Achievement

1. Religion

-(9/11 of poor achievers Muslim, none of the high achievers are Muslim. Buddhism is common in high achievers)
-Differential education of women. (Muslim women are not educated)
-Limited autonomy of women
-Differential education of both sexes. (Muslim countries are less educated)
-Differential methods of schooling. (Muslims: at home, therefore they don't identify as much with institutions)

2. Health Services
-Density of doctors and nurses

3. Family Planning
-Fertility control [--] Reduction of infant mortality
-Greater care given when having fewer children

4. Colonization,br> -(Former British colonies have lower mortality)
-Education and health politics (language --] greater involvement in global health care

Parallels Between Sri Lanka, Kerala, and Costa Rica
-Female autonomy
-Dedicated to education
-Open political system
-Civilian society without rigid class structure
-Egalitarianism and radicalism and national consensus arising from political contest with marked elements of populism
-Dense settlement and cash crop farming in rural areas --] reduction in urban-rural mortality differential (which is key in reducing national mortality)

Position of Women and Children
(these are descriptive facts Caldwell notes about the position of women in these well-performing countries)
-Mothers are more likely to take action about sick children or about selves, will travel to health centers, e.g. be seen in public
-Girls more likely to remain in school (even after puberty)

Measures
-Scandal about girls assuming roles outside the house when they are unmarried but have gone through puberty.
-Scandal about older women appearing in public on their own initiative
*girls more likely to remain in school
*moms more likely to take action about sick children or themselves
*will travel to health centers
*will wait in queues of mixed sexes
*will argue with male physicians
-Woman's mortality and behavior (in the widest sense) are her own responsibility (rather than that of male relatives)
*broader responsibility in general
*deciding early and with certainty that children are sick and need rest/treatment
*will not worry about waiting to consult her husband or his family
-Women have autonomy
*treat daughters more like sons (equality of feeding and medical svcs)
*view sons and daughters in the same light
-Less concern about women's virginity
*marry later
*increased training/education
*increased work outside the home
*increased decision-making power about child health

Rural Health Service
-Substantial female autonomy
*women more likely to become nurses
*less likely to worry about daughters being in training schools
*less worry about working with male doctors or patients
*less worry about living away from home, without relatives
-Where rural girls are educated and parents/relatives are proud if they get professional jobs
*young women work cheaply as health auxiliaries or mid-wives
-Work in home areas (Sri Lanka and Kerala)
*effectiveness in house-to-house visits greater than for outsiders
(lack of local recruits mean no women health workers in some places)
-Status of women (and children) is strongly related to education
-Measures of a child's position is parental willingness to send him/her to school
-Economies unsuited to child labor (e.g. animal tending) such as coastal areas (fish more important) and communities that farm instead of herding
*children more likely to go to school
-More work for women
*children more likely to stay home (to help watch younger children)
-Government's view on/push for education

Conditions of Breakthrough
-Doesn't depend on level of technology
-Depends on ...
*density of service
*effeciency of service
*households visit (particularly for birth)
*immunization campaigns
*nutritional floors (maternal, school meals)

Addendum to Summary of Caldwell's ''Routes to Low Mortality in Poor Countries''

From the summary I started before I found out one had been written:

Caldwell makes a list of third world countries, dividing them into poor and good health achievers. He characterizes them broadly, and then considers two countries and a state in more detail. The countries we will consider in detail, Costa Rica and Sri Lanka, and the state of Kerala in India, have a mortality experience that is markedly at variance with what might seem to be dictated by the economic determinism exerted by per capita income levels. It is possible, then, to break the economic shackles. They are all characterized by breakthrough periods since WWII.

In no country which performs well in health (low mortality, despite its poverty) is Islam culturally or politically dominant. Poor performing countries are overwhelmingly Muslim, or have large Muslim minorities. Islam is argued to have its effect through the position of women, low levels of female schooling, low levels of family planning, and limited access to employment outside the household (which means fewer women become nurses, a fact which has a direct impact on health services in the country). The enrollment of boys in schooling is also lower in Islamic countries than their economic rankings would suggest; there is Koranic schooling, but Caldwell argues that the effects of school have less to do with institutionalization than with the fact that modern schooling provides contact with the whole modern system.

''On the Position of Women and Children''

Under the heading ''Measures,'' the list of characteristics is NOT a list of questions on some survey or something similar: it is a list of descriptive facts Caldwell notes about the position of women in these well-performing countries.

Another Important Point about Caldwell

Preston has argued elsewhere that the adoption of germ theory (the idea that there are little beasties that can get in our water or your ear or whatever and can make you sick -- even though you can't see them -- as opposed to evil spirits, the boogy man, etc.) was key in reductions in mortality. Caldwell thinks this is a load of crap, and cites some evidence (from Australia, for instance) of people who knew nothing about germ theory, but still had good sanitary practices. I mean, really, the point is that you don't drink the spoiled milk, rather than whether you think it's got Morganna, the Evil Demonness of Death living in it, or some teeny little micro-organism.

ANDREW CHERLIN
Marriage, Divorce, Remarriage (1981)

Chapter 1: The Trends

Though everyone loves to talk about the ''demise of the American family'' (because of the high divorce rates, single-parent families, etc.), Cherlin argues that this prediction is premature. The family has undergone many changes in recent decades, but some of the trends aren't' as unusual as people think.

When people talk of the ''demise,'' they tend to compare the contemporary family with that of the Cleavers (of Leave It To Beaver fame). Though we like to look at Beav's family as the stereotypical American family; in actuality, the 50's were an unusual time for family trends, especially in terms of the young age at marriage and the high rate of childbearing. These same trends for the 70's were more consistent with the typical patterns of the 20th century.

Looking at the young adults of the 70's, most people still marry, and those that divorce tend to remarry, usually within 3 years. (Note, however, that the divorce rate after remarriage is also high, probably due to complicated lifestyles and family structures.) While it is true that the divorce rate in the 70's is quite high, it has been high since the mid-19th century. The lifetime proportions ever divorced for people marrying in a given year have risen in a regular fashion for the past century. The apparent ballooning of divorce in the 70's only brought the numbers slightly higher than expected from his trend. In fact, the 50's were an unusual time because the rate was exceptionally low, compared to what the rising trend would have predicted.

Even given this high divorce rate, the annual rate of marital dissolution has not increased dramatically due to the low mortality rates and, hence, fewer marriages resulting in widowhood.

What is unique about he 70's is the high rate of cohabitation. Cohabitation is especially popular among urban young adult who are either better educated and cohabitated prior to marriage or who are less-well-educated but have had at least one previous marriage partner. For most, cohabitation is not a lifelong alternative to marriage, but a stage of intimacy prior to marriage or remarriage. Cohabitation is becoming institutionalizes as a first stage of marriage. Neither cohabitation or a later age at marriage changes patterns of marriage, divorce, and remarriage.

To conclude, when looking at trends such as these, it is important to chose your frame of reference carefully. It is important to keep in mind that the distinctive patterns of the young adults of the 50's reflect experiences of the depression and war years.

Chapter 2: The Explanations

There are two common explanations of the family trends of the 1950's:

1. Period Effect: Period explanations refer to the consequences of events that occur during the period studied. In this case, the argument is that patterns of marriage and childbearing changed as a result of a contemporaneous, society-wide shift in values. The 50's were a time of ''family values'' (a la Dan Quayle). Politicians and the media emphasized traditional family values and a retreat from public life to focus on personal life. The homemaker was extolled and the rise of suburbs created the space to raise the ideal family. But, Cherlin warns against emphasizing these values too much because this traditional rhetoric was accompanied by nontraditional social changes, such as more women entering the work force. Structural period effects include the postwar economic boom, which allowed people to marry young, have children, and still have a decent standard of living.

2. Cohort Effect: Cohort explanations refer to the consequences in later life of the early experiences or shared characteristics of particular birth cohorts. The young adults of the 50's grew up during the depression and war and were greatly affected by these experiences. For example, they were forced to assume adult roles at an early age and thus valued these roles. In addition, tough economic times produced lower standards of material things. Also, the relatively small cohort size meant better job opportunities and better economic standings on which to support families.

Cherlin notes, however, that period and cohort effects are often confounded and that both were operating and reinforcing each other to strengthen the trends of the 50's.

Explaining the 60's and 70's:
During this time, many family trends returned to ''normal.'' However, the recent rise in divorce is unprecedented in the speed in which the rate is increasing. This can be explained by the following factors:

Changes in attitudes toward divorce - it is not as stigmatized as it once was. (period effect) More married women work, which gives them the resources to leave an unhappy marriage. (period effect) The relative income hypothesis: The people divorcing in the 60's and 70's were raised in times of plenty, and the increase in women's employment and the decrease in overall income made a tight economic situation worse, leading to increased unhappiness. (cohort effect) Birth control may make it more practical to postpone marriage. (period effect) The marriage squeeze: Age at marriage may have risked because there are not as many men in the proper age range (2 to 3 years older) for women to marry. (cohort effect)

Once again, Cherlin acknowledges the importance of both period and cohort effects. He does speculate, however, that the increasing labor force participation of young married women ultimately will be seen as the most important stimulus to the initial rise in age at marriage and in divorce after 1960.

Overall, the trends in marriage, divorce, and childbearing since WWII are the result of one long term process (the development of advanced industrial society) and two specific historical events (the Great Depression and WWII).

Chapter 3: The Consequences

Divorce and remarriage increasingly have become a normal event in a person's life course. Family life continues to be centered around marriage. As marriage changes, the functions of marriage will be fulfilled in other ways. Of course, changes such as increasing divorce and remarriage, increasing single-parent households, and increasing families from remarriage profoundly affect the individuals involved, causing emotional strain, financial changes, changes in responsibilities and task management (such as child care), and changing conceptions of ''family.'' However, people learn to cope with these changes. An increase in institutional support for families and the development of social norms and standards of conduct will lessen the burden of these changes on individuals, particularly children.

Chapter 4: Black and White Differences

There is reason to believe that the postwar trends up and down have been parallel for most groups. However, there are some important differences in the typical family patterns of blacks and whites:
Blacks used to marry earlier than whites, but now they marry later
Black women have higher annual rates of out-of-wedlock childbearing
At any one time, there are proportionately more blacks than whites who are separated but not divorced
Once they are divorced, blacks tend to take longer to remarry

Some theorists point to the legacy of slavery to account for high levels of marital instability and out-of-wedlock childbearing. However, more recent research disproves such arguments by showing that two-parent families were more common under slavery and single-parent black families began to occur on a much larger scale within the last 50 years. Cherlin argues that he contemporary divergence of black and white family patterns goes back no further than the depression and has accelerated since 1960. The causes lie in the contemporary situation of urban, Northern blacks, rather than in a lingering heritage of slavery or a clash of traditional and modern cultures. Possible explanations lie in sharper social class distinctions, especially different educational and occupational opportunities (W.J. Wilson). Blacks have responded to such changes by relying less on marital ties and more on extended kin networks. However, marital instability is still high for blacks, especially lower-class blacks. This could be due to the negative effects of unemployment on marriage, urbanization, and the impact of social welfare policies.

1 Older cohorts of immigrants appeared to have assimilated well into the US labor market, in the sense that they were doing as well or better than native born workers twenty or so years after arrival. However, this finding may reflect the fact that unsuccessful immigrants are more likely to return to their country of origin (selective emigration), leaving successful immigrants to be counted by the census.
2. He seems to me to be saying that people behave economically rationally, but don't do so purposively. This begs the question of why they behave economically rationally, then. It seems functionalist to me.