The President's Council on Bioethics click here to skip navigation


Thursday, October 17, 2002

Session 2: Choosing the Sex of Children: Demographics

Nicholas Eberstadt, Henry Wendt Chair in Political Economy, American Enterprise Institute

CHAIRMAN KASS:  All right.  In this second session on choosing sex of children, we turn to certain demographic implications of the use of this capacity, and we're very fortunate to have Nick Eberstadt, who is the holder of the Harry Wendt Chair in Political Economy at the American Enterprise Institute, to make a presentation to us this morning.

It was actually a talk that I heard Nick give.  It must be four years ago on this topic that opened my eyes to how things quietly initiated here in the United States for  one reason wind up having very powerful effects across the globe to which we should pay attention.

And when this topic came up, I thought it would be worthwhile for the Council to hear about the uses of sex determination practices.

And, Nick, thanks very much for joining us.  We look forward to your talk.


DR. EBERSTADT:  Leon, thank you very much.

Ladies and gents, it's an honor to be with you here this morning, and I'm going to start off more or less exactly where Dr. Haney left off in his excellent presentation.

What I'm going to try to do very quickly is provide you with some information about the state of conditions in the world today with respect to the secondary sex ratio, usually called the sex ratio at birth, and also to offer some speculations about some of the implications of trends that are now developing.

Demography is a study about three centuries old, maybe a little bit more than three centuries old at this point.  As soon as students of demography began to look at patterns in human numbers, one of the first things that they recognized was that there was quite a regular pattern with respect to births, slightly more boys born than girls born.

And this was noted and commented upon at the beginning of the study of demography.  I'll read you something that Johann Sussmilch, who was an early demographer, wrote in 1741.  He said, referring to other early demographers, that "Grant, Durham, and others have suggested the Creator has reasons for insuring four to five percent more boys than girls lie in the fact that it compensates for the higher male losses due to the recklessness of boys, to exhaustion, to dangerous occupations, to war, to seafaring and immigration, thus maintaining the balance between the two sexes so that everyone can find a spouse at the appropriate time for marriage."

Well, as you will appreciate, demographers are rather more loath today to talk about divine intent, but they do recognize that there is over time and over space a consistent pattern of somewhat excess male mortality over female mortality.  So that the early surfeit of boys, if you will, more or less evens out by marriageable ages so that cohorts are more equal in numbers.

Now, demographers have noted variations in the secondary sex ratio or the sex ratio at birth associated with a variety of factors.  They have noted variations with respect to ethnicity.   They've noted variations with respect to parity, which is to day birth order, and with the age of parents.

They've noted some variations with respect to the nutritional status of parents.  Nutritional status may have some slight influence upon sex ratio.  Also various sorts of diseases and disabilities.

And there is an ongoing question/discussion, call it a debate, in certain areas of demography about whether there may be an adaptive response with respect to sex ratio at birth, that is to say, it's called the operational sex ratio or the sex ratio of the adult population may have some influence upon the rising cohort's sex ratios.

I have no opinion about this work.  I consider it an unsettled discussion which continues, and I wouldn't suggest that there's any conclusive findings that have been found in this area, but for noting all of these influences upon sex ratios at birth, I think the overwhelming biological fact about sex ratios at birth in large and regularly constituted populations is regularity and the stability of the sorts of numbers one sees produced.

These ratios are almost constant in large populations over time and over space.  Just to give you by way of background some perspective on this, these are the sex ratios at birth by ethnicity for the United States from the early 1980s, and you will see that for the country as a whole, it was about 105 live baby boys born for every 100 live baby girls.  There were differences by ethnicity, but these were not dramatic, let's say.

By the same token, you can see differences in sex ratio at birth with respect to birth order, the parity.  In general the sex ratio at birth is somewhat lower at higher parity or higher birth orders than at lower parity or lower birth orders.

And this result for the United States in 1984 is hardly uncommon.  We could have used data from many other countries, many other times to replicate that.

This was 1984, and we live in 2002 today, and some things have started to change even in the United States itself.  And one of the things we are beginning to see in the U.S. are sex ratios for at least certain groups within the population which would be very hard to explain on the basis of purely random occurrence since we're dealing with large numbers of people.

The odds against some of these sex ratios at birth look very forbidding, very imposing.  Let me show you here.

These numbers compare sex ratios at birth in the United states in 1984 and in the year 2000, which is the latest year for which we have complete birth data, and you'll see that there's not terribly much difference in the sex ratio at birth for the total population or for the so-called white population or the so-called black population.

But when one gets down to Asian Americans, to the Chinese and Japanese ethnicities in the United States, we're seeing some very substantial increases in sex ratio at birth, and those would be very hard to explain on a purely biological basis.

Over the last decade or so, we find, indeed, that there are many places around the world where these sex ratio at birth has started unaccountably or seemingly biologically unaccountably to rise.

Let me show you some data from other countries.  These data were gathered from the United Nations' statistical office which publishes an annual demographic year book.  Of course, it's never up to date, and of course, when it is compiled, the data are from earlier years.  These are the most recent data that the U.N. demographic year book has pulled together on live births in countries and areas where vital registration is nearly complete.

And what I have listed here are simply countries reporting now a sex ratio at birth of 107 baby boys per 100 baby girls or higher.  In ordinarily constituted populations ratios of 103, 104, 105, even slightly over 105 are not things that would look unusual.

One, oh, six starts to need some explanation, and 107 just usually doesn't happen, and that's why I chose 107 as the cutoff there.

And you will see here a number of different regions of the world represented with these unusually high sex ratios at birth:  some Latin American countries, Salvador, Venezuela, most dramatically Cuba, where we have a phenomenal 118 baby boys per 100 baby girls, some of the Maghreb  countries, the North Africa, Tunisia, and  Egypt. 

They represented also some of the post Communist states from the former Soviet Bloc, Belarus, Bulgaria.  I think we have excluded some additional post Communist countries whose birth numbers were rather lower, but whose sex ratios were very high.  I think Moldova, Estonia, Lithuania could also be added into this grouping.

And then, of course, we have a grouping of East Asian countries and regions:  Hong Kong, the Republic of Korea being most prominent there.

One of the problems in tracking changes in sex ratios at birth around the world is that most of the world's population is not found in countries that have complete or nearly complete vital statistical systems, vital registration systems.

By the U.N. Population Division's estimate, in fact, less than one person in 12 in the low income world lives in a region that would be described as a country with complete or near complete vital registration.  So that's a big problem in trying to track trends in sex ratio at birth.

However, we can draw some inferences about changing patterns of sex ratio at birth by looking at the numbers of children at young ages reported in national censuses and other sorts of demographic survey data.  And I will show you this imperfect, but perhaps indicative proxy for a number of other countries.

These data were compiled by the U.S. Census Bureau's International Programs Center.  These are their estimates for 1998 of sex ratios for children under the age of five and for particular countries where the ratio was 107 or higher.  And you'll see, again, a representation of an Islamic society in Tunisia.  You'll find some post Communist representation in the form of Serbia and Macedonia.  For the most part what one's seeing there are East Asian countries:  Singapore, Taiwan, China, Hong Kong, and South Korea.

And the sex ratio in some of those settings for children seem to be very, very unusually high indeed.

There is another smaller area of the world whose imbalances I've just become aware of thanks to a colleague at the Census Bureau. Dennis Donahue kindly supplied me with this table.  But in the Caucasus area of the former Soviet Union, the ratio of children under one year of age — these are not live births, but tabulated infants and through census materials and demographic counts — the ratio of children under one year of age has risen very rapidly and dramatically in Armenia, Azerbaijan, and Georgia.

These Caucasus societies represent a diversity of religions.  Azerbaijan is primarily Islamic.  Armenia and Georgia are not, but as of the year 2001, the sex ratio for infants under one year of age was approaching 120 in each of these three countries.

Let me turn to East Asia which Dr. Haney discussed a little earlier.  This slide, which is put together by Daniel Goodkind at the U.S. Census Bureau shows reported sex ratios at birth for a number of East Asian countries.  You'll see that the sex ratio at birth for Japan, this lower line, falls within the range that would ordinarily be expected of a large human population.

For all the other countries and regions, for Singapore Chinese, for Hong Kong, for Taiwan, for South Korea, in particular for South Korea, some of these ratios have risen really quite extraordinarily.  In the past several years South Korea's sex ratio at birth had declined, but it has declined only to a degree.  It's still in the range of 110 baby boys per 100 baby girls, something like that.

The big enchilada is China, which is the most populous country in the world, and China does not have complete birth registration statistics.  So determining the actual sex ratio at birth requires quite a bit of inference.

As you'll see, there is a discrepancy in China between hospital records, which are certainly not complete with respect to annual births, servicing only a limited fraction of the population, predominantly the urban fraction, and records from vital statistics, limited though they may be.

There is a discrepancy there, and this is a discrepancy which demographers have puzzled over.  It has suggested to some that the imbalance in China may not be as great as some observers have feared.

I think, however, the weight of evidence from demographic records leans towards the more pessimistic rather than, if you will, the more optimistic assessment, and that's a judgment reinforced by the initial data from the November 2000 Chinese census. 

If one attempts to do reconstructions from that and from previous censuses, one sees a sex ratio at birth that has risen perhaps from around 108 20 years ago to something like 117 today, and if one looks at the sex ratio of children under the age of five from this and previous censuses, one gets even more extreme indications of increasing imbalance between young boys and young girls.

As Dr. Haney's presentation indicated previously, the imbalance in these societies is concentrated disproportionately in higher parity births, although as I think I already showed you, in ordinary biologically constituted population, sex ratios at higher birth orders tend actually to decline slightly.

But what we find in many East Asian societies today is an increasing imbalance, increasing sex ratio at birth with higher birth order parities.  This slide presents data collected by two doctors in Hong Kong at St. Margaret's Hospital.  Hong Kong does not ordinarily offer comprehensive data on sex ratios by birth order, although it does have complete vital registration.

St. Margaret's Hospital, I believe, handles about a sixth of the births in Hong Kong, SAR.  So it is by no means comprehensive, but it is indicative, and as you'll see there, Wong and Ho show that for births that were second births, there was quite a significant distinction between sex ratios for parents, for mothers, whose first birth had been a boy and whose first birth had been a girl, and the phenomenon is even more extreme for third order births.  In the cases where the two previous births to a mother had been girls, the sex ratio was 137.

Even with small numbers of births, as in this sort of situation, the odds against this being a natural occurrence become quite astronomical.

Dr. Haney already showed a slide indicating changing sex ratios by birth order.  These are data for Taiwan in 1990 and for China in '89, and as you see, when one gets up to birth order four, fourth births in Taiwan, by 1990 we are talking about sex ratios of 160 boys per 100 girls.

But it gets better than that.  Let me show you South Korea.  In 1992, by the time one is talking about fourth order births, we are above 200 boys for every 100 girls born, and you can see there that blue line is South Korea in 1992.  The red line is South Korea only 12 years earlier, in 1980, for the wide dissemination of technology, making available sex determination and, therefore, sex selective abortion.

In all of these cases, the inference that one would draw, I think, is that sex determination has led to sex selective abortion as a main driver of these biologically unnatural results.  Female infanticide may have played a more pronounced role in China, especially in the late 1970s and early 1980s, but increasingly, I think we can draw the reasonable inference that sex determination and sex selective abortion has been the primary instrument at work here.

Dr. Kass asked me to gather for you some information on gender preference on the parts of parents, and I had assumed this was going to be a very easy task for me, very easy part of my presentation, and was quite surprised to find that it was not.

The reason for the opacity or the difficulty here is that the ordinary or the regular surveys that are circulated through much of the world to determine demographic health fertility patterns, the so-called DHS survey, as a rule now simply do not contain any information to present to respondents with respect to gender preference of children.  Such information was collected in earlier rounds of world fertility surveys, but as it happens, for the most part these are not available now for most low income countries.

It's unfortunate.  As you can see, it's an increasingly unfortunate oversight. 

However, one of the countries that does collect information on gender preference among married women is India, and it's not an inconsiderable portion of humanity.  Roughly one out of six people in the world today live in India.  It's a population of over a billion.

And this shows you results from the national family health survey, second round done about three or four years ago.  Almost everyone, almost all mothers asked to have no children, and most of those who had one child desired more children, and as you'll see below, when asked whether they preferred a boy or a girl, some said no preference.  Some said leave it to God, but of those who expressed a preference, there's an overwhelming preponderance of preference for boys, four to one for the sample overall, and in some cases even higher ratios.

Let me show you data for a particular state in India, the State of Punjab.  Punjab is a smaller sample, and we have asterisks there because in some cases there weren't enough respondents to arrive at statistically significant results, but for the Punjab as a whole, out of the almost 3,000 married women questioned, the preference for an additional child being a boy as opposed to a girl was on the order of ten to one.

Now, let me show you some results from the latest Indian census.  India is another country that has no – that does not yet have complete vital registration of births and deaths for the country as a whole.  So one has to work with census materials and other sorts of large counts to do reconstructions or to draw certain sorts of inferences.

The March 2001 Indian census did account by age for the country as a whole and its various provinces, and what I'll show you now is the sex ratio for children under age seven or children zero to six for different provinces in India.

You will see that there are a number of states and territories in which the sex ratio for surviving children, surviving up to the age of seven, falls within the range that we might call biologically expected, maybe up to about there, but there are a  number of states for the most part in northern India where the ratio is outside of historical biological norms.  Mostly these are states in northern India.

But I'll point out in particular to you the results for Punjab.  In Punjab today for every four girls under the age of seven there are over five boys, and it's probably also worth noting that these newly found differences do not comport with patterns of tradition or with patterns of traditionally construed under development.

These heightened abnormal sex ratios at birth are instead very closely associated with modernization or with some particular variant of modernization.  Punjab, for example, is one of the most prosperous provinces in all of India.  I think we see Delhi.  Delhi at 116 is also one of the most — is an urbanized area with one of the highest income levels in India and one of the higher literacy levels in India.

Thus also Chandigarh, thus also Haryana, and if we go back and review some of those other data that I showed you, I think we'd say that this imbalance in sex ratios at birth has coincided with improved levels of income with higher educational attainment and with heightened interaction with other cultures and other economies, with what we would call globalization.

So far from being traditionally driven, these trends have certainly coincided and comported with some form of modernization.

I will get to that in a moment.

I think it also mentioned that these heightened abnormal sex ratios at birth have been associated with fertility decline, and they're most obviously in low or sub-replacement fertility populations.

As you saw from the East Asian data, all of the East Asian countries that we were discussing are currently at or below replacement levels of overall childbearing, which is to say that if current childbearing patterns continued indefinitely and there were no immigration, no net immigration, population would stabilize and ultimately decline indefinitely.

The phenomenon of choosing higher parity, the sex of higher parity children seems to be more of a pronounced phenomenon in the context of declining or sub-replacement fertility, and it is worth noting that the Punjab area of India has been an area of rapid fertility decline.  It is not the lowest fertility area in India.  There are some sub-replacement areas of southern India that do not exhibit this extreme imbalance between male and female children, but Punjab is an area of rapid fertility decline and is now close to the replacement level.

Dr. Kass asked me if I would draw together some data on sex preference and sex selection in the Arab-Islamic world.  I think I showed you earlier some indications that Tunisia, which is one of the few areas in the Arab-Islamic expanse to have hit replacement or sub-replacement fertility, is now exhibiting unusually high sex ratios at birth or excuse me.  I think it's ratios of children under the age of five, sex ratios of children under the age of five.

And there is precious little information on gender preference from survey and demographic data for this great expanse of humanity, but one of the few surveys that I could find is actually for the Palestinian authority, and it isn't quite as neat as I would like something to present to you, but I think it is indicative nonetheless of what we have here from a demographic survey conducted in the late 1990s, is the preference expressed by I think it was a total of 3,000 married women for an additional child or for an additional son or an additional daughter if one already has a son or daughter, if one already has four-plus sons or four-plus daughters.

And I hope this presentation isn't too confusing, but I think you'll see from these numbers a very strong and pronounced disposition towards son  preference across the board here.

Interestingly enough, the sex ratio at birth for babies within the Palestinian authority area and for Muslims in Israel proper is no different.  It's not appreciably different from that of Israeli Jews.  Both sex ratios for now are in the vicinity of 105, slightly below 105 baby boys per 100 baby girls.

But what we have not yet seen in Palestinian areas or in the rest of the Arab-Islamic expanse is the  dramatic dissemination of relatively inexpensive techniques of gender determination prenatally, and so this is a phenomenon which we may yet experience, although we have not thus far.

What are the consequences of gender imbalances?  One of the most obvious possible consequences of the sorts of gender imbalances we have seen developing around different parts of the world is a later, potentially inevitable, inexorable imbalance in the marriage market.  If there are too many boys to marry off to a given number of young ladies, the market can't clear.

In the past that has not been too much of a problem even in areas where boys in any given age group have tended to outnumber girls in any given age group because for the most part, over the last century certainly world population has been rising, and that has meant that each year slightly more girls were being born in any given birth cohort.  So that matching up or pairing simply would entail an average difference in age at marriage.

You will appreciate, to make a very crude example, if population were growing, if birth numbers were growing by about two percent a year and there was a ten percent imbalance between young men and young women, it would take roughly five years in difference in average age at marriage to make everything square.

But the arithmetic becomes very much less forgiving when one is dealing with sub-replacement populations rather than populations that are at or above replacement level.  And as you will have seen already, the gender imbalances that we have seen developing are most pronounced in societies that are precisely sub-replacement or below replacement fertility contexts.

What I wanted to show you here is some speculative projections for china, and I have to emphasize that these are speculative.  The data on the year 2000 are not really terribly speculative.  We've got those data, but then the question is:  what will China look like, say, 25 years from now?

What I have put together for you here is a Chinese population structure and age-sex structure based on the presumption or the assumption that the 117 to 100 imbalance, implied imbalance, in sex ratio at birth that we've seen from the recent Chinese census is, in fact, real rather than a statistical artifact, and that that imbalance continues from 2000 to 2025.

I think even by eyeballing this, you can see that there are at younger ages an awful lot more greens than blues in this figure.  What does that, what would that hypothetically mean?

Well, let's look at this.  The U.S. Census Bureau takes the cautious and, I think, respectable posture that until there is overwhelmingly persuasive evidence to the contrary, they will be projecting China's future population on the basis of an assumption that the true imbalance, the true sex ratio at birth is 109 to 100 rather than 117 as recently reported.

If one accepts that, by the year 2025, the sex ratio for young men and young women ages 20 to 35 would have risen from about 106 reported roughly today to about 109.  There'd be a deficit; there'd be a shortage in this particular cohort of about 13 million women as opposed to men.

Now, China has been a society, a cultural setting, where near universal marriage has been the expectation.  One takes a look at previous censuses or demographic data on China.  Ninety-six, 97, 98 percent of women report eventually — report having been ever married by the time and heads towards older ages.

But that means that three, four percent of Chinese women do not get married, do not take a husband, and if one deals very crudely here, that would be suggesting that something like ten, 11, 12 percent of the Chinese men in the Census Bureau's projections in this year would have to find wives if they were to find wives from outside of this cohort.

If one takes an assumed sex ratio of 117 and projects that forward, the numbers are even more dramatic.  We would end up with a shortage within these given cohorts of about 16 million potential brides or 16 million excess husbands.

And if one stuck with this same back-of-the-envelope sort of calculation that two, three, four percent of women would end up never marrying, we'd be talking about 13, 14, 15 percent of this cohort having to find wives or partners from outside of this grouping or never marry.

Now, what are the social and economic and political implications of having a large group of men for whom the expectation of never marrying is fairly plausible?  My impression, my very unscientific impression from demographic history is that it very, very much depends.

One can see all sorts of ominous arguments about unsocialized or unsocializable young men causing social strains and perhaps even political problems, and you know, prima facie, I think that argument is inherently plausible.

But it is also true that there have been large regions of the world in which the expectation of never marrying has been quite real for large proportions of the population and where  social fabrics have dealt with this in sort of a regular and non-catastrophic manner.

Pre-industrial Europe, for example, is a setting that comes to mind, although in Europe west of the Danube today a very large proportion of women end up eventually marrying, end up being ever married.  A hundred years ago or so that was not at all the case, and at the end of the 19th Century and earlier, it was not at all uncommon to find birth cohorts of women and of men in which 15 or 20 or 25 percent total never married.

Now, in the western European context, western European culture provided for mechanisms to deal with this phenomenon.  There were mercenary armies.  There's the Catholic church.  There's respectable spinsterhood and bachelorhood, and all of these were social conventions which helped to deal with this demographic phenomenon.

As best I can tell, in East Asian and in Confucian societies, there is no parallel set of mitigating social mechanisms, and rather to the contrary, the expectation seems to be quite strongly to encourage marriage, if possible, in part to continue the family line and to respect ancestors and all of the rest.

Social conventions, I think, would have to change very dramatically and very quickly in large portions of East Asia and perhaps elsewhere to deal with this impending gender imbalance, and it seems to me a very reasonable question to ask whether one can expect this to occur in such a very short period of time.

It's all speculative, but it seems a question worth asking, and I think I'll stop there.

CHAIRMAN KASS:  Thank you very much, Nick, for a very interesting and provocative presentation.

If we could get the lights, we could start our discussion.

Frank Fukuyama.

PROF. FUKUYAMA: Nick, thanks.  I've been reading your stuff for many years, and it has all been extremely useful.  I've got a series of small factual questions and a larger one at the end, but maybe you could just answer them serially.

You know, Amartya Sen wrote this famous book about 100 million missing women, but from your data, it looks like that's actually a very low figure.  Do you  know how he got that or what period corresponds to?

Because it looks like, you know, just China alone over the next decade is going to  —

CHAIRMAN KASS:  Would you turn the mic on?

DR. EBERSTADT:  From the 1980s.  It was based on U.N. Population Division estimates of total world population.

PROF. FUKUYAMA: Okay, and on Korea, the drop that occurred from the early '90s to the late '90s, from about kind of the high teens to, you know, like 110, my understanding is that sex selection was actually illegal there in that the reason that that drop happened is that they actually started enforcing their existing laws.

DR. EBERSTADT:  Yes, it had been illegal.  New legislation had come on the books in the 1980s with the widespread advent of amniocentesis and ultra sound.  The laws were, as you can tell, completely ignored.

Then in the mid to late 1990s, the government started to pay more attention to these practices.  Civil society was also important there.  There were festivals in  South Korea held, "love your daughter" festivals and things like this, and that started to make some sort of more general impact.

PROF. FUKUYAMA: I understand the ratios also varied by province quite substantially.

DR. EBERSTADT:  Yes, yes.  That's right.

PROF. FUKUYAMA: Do we have any historical data on rates of female infanticide in Asia just as a point of comparison for the current sex ratios?

DR. EBERSTADT:  We can only draw inferences about infanticide from, as you will, the missing girls in earlier censuses and very limited registries.

In China, in particular, there's a longstanding imbalance between males and females at almost every given age.  Life expectancy for men at younger ages quite surprisingly is higher than life expectancy for women, and that suggested a whole plethora of discriminatory practices, not only infanticide, but discriminatory practices after those young ages.

PROF. FUKUYAMA: Okay, and then I guess the longer question, you know, there was a book published by Marcia Gutentag and Paul Secord.  I guess it was over a decade ago called Too Many Women, talking about some of the social consequences of, you know, these unbalanced sex ratios, and I guess while the conventional wisdom is having too many men is a problem, they make the point that there may be some compensating advantages, which is that if you have a sex ratio that's tilted towards men, it actually puts women in the driver's seat in marriage markets, and that one of the arguments they made was with the Baby Boom one of the reasons that the sexual revolution happened when it did was that because of this phenomenon of men marrying younger women, they had more choice, and in effect that skewed the sex ratio towards women, which led to, you know, a breakdown of family life and so forth.


PROF. FUKUYAMA: And that you could expect the opposite to happen if you had a sex ratio skewed towards men.

Is that a respectable argument?  I mean, what do you make of that?

DR. EBERSTADT:  That's a very reasonable, Economics 101 sort of argument, I think, but not all parts of the world proceed under the sorts of premises that we presume to be in place in an Economics 101 setting.  In China already, the increased value of women, increased scarcity value of women has led anecdotally to distinctly increased reports of woman stealing, of trafficking, of virtual enslavement, and that is, I suppose, a less attractive, but still quite real manifestation of an improved value of women.

DR. FOSTER:   Just a brief comment.  If I remember it correctly, about the last week The New York Times, you know, had an article about single women in the United States, and if I remember the figure correctly, that above the age of 15, about 48 percent of all women are single.

And I don't know where it is in other places, but if a wife of a mature man dies in Dallas, instantly he is assaulted by women who want to have a date or go to dinner or something of that sort.  I mean, there does seem to be in our country a very large group of the excess women, and I don't know what to make of that.  I'm just commenting on it.

PROF. GEORGE:  It depends on what you count as our country, Daniel.

DR. FOSTER:  Well, I believe — I don't know, Robby, whether I'm following you or not, but I believe that Texas did join the Union some time ago.



PROF. DRESSER:  These are questions for you to comment on, but as for the group as well to think about.

I guess one issue this raises for me is how should we as a Council look at patterns and practices in other nations.  Certainly it would have an impact on U.S. medicine when people from other cultural backgrounds come here and want particular interventions that are more popular in other nations, but should we go beyond that?  And how should we think about that?

And then the second is provoked, I guess, a little bit by Leon's point about orthodox Jewish preferences for sons.  I think we're looking at sex selection as part of our, quote, enhancement project, I think.  Is that right?


PROF. DRESSER:  Maybe it's related to some other project I don't know about, but anyway, if we're to think about sex selection as being done for non-health related purposes, putting aside defining what falls under health related purposes, but if we're just to say, okay, for other reasons, is it possible to make arguments that would say certain reasons deserve more weight than others?

So a religious tradition, a cultural tradition, family balancing; is it possible to really discriminate or differentiate between preferences, saying some are unjustified gender or sex discrimination and others are not, or should they all be treated equivalently, whether that is to say they're all okay or they're all not okay?

CHAIRMAN KASS:  Nick, do you want a word on this or not?

Let me say something about the first question.  I mean one could argue that what goes on in other countries with the use of these technologies is a curiosity and we should pay attention, but finally not of concern to this body, and that might be closer to the truth than not.

On the other hand, as has been pointed out in previous discussions, what goes on in other countries has an impact on what happens here, either where precedents set elsewhere come to be argued for here.  I mean if the British are doing embryo research, why shouldn't we, or conversely, they would say, "Look.  If the United States isn't doing sex selection, you know, maybe we shouldn't either."

But there's also these — there are sort of social and political ramifications of these technologies, and Nick, in his characteristic understatement, just alluded I think.  The numbers are by themselves interesting, but what they actually mean socially and what they mean internationally in an age of globalization is, it seems to me, of importance and, at the very least, it might be worth our while to call attention to what goes on with the use of these technologies and perhaps even recommend nothing so radical, nothing more radical than the need to monitor and pay attention to what is happening and where in ways which we are not at the moment doing.

That would be at least the minimum suggestion as to why that kind of conversation — I mean, the material that Nick has raised is, I think, of importance to us.

And if I might simply respond just to open up the other question, yeah, there is a question here as to whether or not — I mean, whether sex selection is an enhancement or not could be discussed.  That it is a non-medical or nontherapeutic use of medical technique is part of what makes this of interest.

And I think one of the reasons it's part of this conversation is that it's one of a whole series of developments produced for medically related purposes, but which will yield individuals, unless there are regulations, but will yield the desire, will support the desire for individuals to use new technical power to achieve their desires either for self-enhancement or for the control of their offspring or for the control of behavior of others.

The question is whether if those things are worrisome, what, if anything, can and should be done about it.  I think that's the context, and what are the reasonable reasons for using this thing, except for the prevention of disease is, it seems to me, interesting.

And does one sit in judgment of the people in Punjab or is this one of those occasions where who are we to judge the cultural preferences of other people or, for example, the cultural preferences of American subcultures, the first slide that Nick showed about beginning to see changes in the United States in the East Asian populations where the sex ratio begins to approach 110.

DR. EBERSTADT:  I think the ones I showed were 107.  The Philippine Americans, it's almost 110.



DR. ROWLEY:  Well, I'd like to go back.  Just some of the techniques that are used to achieve these changing ratios, and one of them obviously is sex selection before implantation, and the other is selective abortion, and it's sort of surprising to me, say, in countries that I assumed had  restricted availability of some of these technologies, that there are changes.

For instance, you had the statistic about American Indians going from 101 in 1984 to 103.5 in 2000.  I'm not sure whether that's within the range of variation because one thinks of American Indians as not having access necessarily to fairly sophisticated technologies.

And then you also quoted countries like El Salvador with 107, the Philippines, 108.7, Egypt, 108.7, and Pakistan, 110.9.  It would seem to me it's a small portion of the population within those countries that would have access to more sophisticated technologies.

But the implication of all of this is that it's really some kind of sophisticated technology and access to it that is causing these changes because these are birth ratios.  These aren't zero to four where you can think of infanticide.

So what is going on actually in these countries that may be associated with these changes?

DR. EBERSTADT:  I would not myself become at least yet too alarmed by the data reported there for the American Indian population even though the reported ratio, sex ratio at birth in the years that I chose had increased.  103 is still well within the range of ordinarily reported biological variability.

I think one starts to ask questions to be arbitrary at about 106, and one starts to have alarm bells go off at about 107.

With respect to Pakistan, the data there for Pakistan were from their vital registration system, which I don't think is complete.  I would guess in the case of Pakistan — I would have to go back and double check, but I would guess that those are hospital deliveries, and hospital deliveries are going to have the enhanced techniques that you asked about.

With the case of El Salvador or Egypt, medical services are presumably much more limited than for native American population in the United States.  I'm only speculating, and I have no basis in fact for this, but just to speculate about this, if a relatively small proportion of parents felt rather strongly about some preference issues, it would be possible to alter to some degree the national sex ratio at birth without having comprehensive medical services and availability, but I don't know.  That's a speculation.

DR. ROWLEY:  Well, if I can just follow on with this, and again, toward the end you also tied this in with more sophisticated family planning, if you will.  So if you only want to have two or, say, three children and you already have one or two children of a certain sex and you're going to have just one more child, then may well pay much more attention, and your data has suggested if you already had two females and you were going to have only one more child, it was very important to you that you have a boy.

I happen to have four sons and only four children.  So —

DR. EBERSTADT:  And I have three daughters.

DR. ROWLEY:  So, you know, years ago I'm not sure if really effective sex selection had been available whether I might have taken advantage of it for family balance.  It wasn't available.  It's not clear that it is even yet available unless you to go really extraordinary lengths.

But I think that these are intellectually interesting issues.  Again, the question whether it's a matter for us to consider and pursue, I think, is the kind of discussion that Michael was suggesting that we have.

CHAIRMAN KASS:  Michael, and then Bill May, and then Gil.

PROF. SANDEL:  Well, these were two such fascinating presentations, and I'd like to go back to the ethical questions that they raise.

First, a comment that occurred to me at the end of our first session, or thought rather.  There are lots of reasons to worry about sex selection, but I think that our greatest contribution can be to resist the reflect or the tendency to translate what troubles us into overly familiar terms.

And that comes out if we say, well, it must be the means.  And so people have views one way or the other about certainly infanticide, about abortion, and for that matter, about embryos.  So there's a tendency to translate it into that aspect or, on the other hand — and in this report, Leon, that you referred to we have in our binder, to translate it to different kind of familiar terms.

Well, it's going to lead to gender discrimination.  We're familiar in our society arguing about those two things, about the status of embryos, on the one hand, and about gender discrimination, on the other.

But what's distinctive about this question of sex selection — and this connects to the broader issue of enhancement — is that there are reasons to worry that go beyond those two familiar reasons, which isn't to diminish those familiar ones, but this question is philosophically interesting and challenging because not everything that's objectionable about it can be translated either into worries about embryos, on the one hand, or about discrimination, on the other.

And that takes us to the goals or the goods or the ends to reflect on, and looking at the data, the demographic data, we all shudder at the numbers, but trying to make sense of the shuddering is what's interesting.

There are worries obviously about the social consequences, and here we enter into the kind of speculation in the exchange between Frank and Nick about, you know, will all of those extra men in  China create instability, make China more warlike.  Will it make the situation with women better?  Will it make it worse?  There's that set of worries.

But beyond the worries, there's something also chilling about this, quite apart from speculating about the social consequences for China or for the world or for war or for instability or for all of these extra men rattling around there.  And that's what we should try to get at.

Is it chilling?  Well, one thing is, oh, it's chilling because we know lying behind there, you know, is some infanticide or abortion  or the killing of embryos.

But there's another dimension to what makes it chilling, and that has to do with — well, I'm not sure, but here's a speculation or a question just to invite us to direct some of our energy anyhow in our reflection to this other set of issues.

Is it that we see in those ratios, rising ratios, is it that we see here's what happens when we have the technology to actually implement the things that for various reasons we might desire?

And is the danger — is what's chilling the power that the technology give us?  And may shrink before the consequence of that technology when we see it here, or is what's chilling that the technology being available just reveals to us vividly desires, things that people want, that themselves — we think people shouldn't want that sort of thing.

And suddenly the technology lays bare desires that  we haven't been able to act upon in other times.  So here's one way of testing that independent of the technology.

Is it wrong  or not wrong, but is there something morally troubling or questionable about praying for a boy rather than a girl?  And is the problem with the technology that it answers the prayer that's independently objectionable or it enables us to answer the prayer, to perhaps obviate the need for the prayer?  But is that the underlying desire; is it the prayer?  Would that be objectionable?

And if it's not, then we have to look elsewhere, but if it is, then that leads us to some difficult moral terrain, but at least it's the kind of terrain that may kind of get at this issue of enhancement.

CHAIRMAN KASS:  Could I as Chair ask members in the queue if they would waive the queue and address the question?  Because I think it's a nice — if we could at least response to Michael's challenge, if people would like to address it rather than just run away from it?

Gil, do you want to speak to it?  Both you and Bill were in there.

PROF. MEILAENDER:  I do want to speak to it.


PROF. MEILAENDER:  Maybe he does.

DR. MAY: But go ahead.

PROF. MEILAENDER:  Well, yeah, I want to pick up on that actually relate to where it was Rebecca who sort of started us down this road to some degree by thinking about kind of how this fit with our larger concerns.

And let's assume an unobjectionable means for the time being, you know, whatever that means.  It probably means different things to different people, but let's assume an unobjectionable means so that we're talking about just the goal of getting a child of the desired sex.

I don't myself think there's anything wrong with the desire, and if somebody said, "I'm praying to have a boy," that's all right.

If I may just kind of switch issues for a moment, St. Paul prays to die and be with Christ.  He doesn't act on that, however.

So that, I mean, I don't think there's necessarily anything wrong with the desire.  So that's not — if we're looking for kind of a deeper concern, I don't think that's it, that it reveals to us some desire that's suspect.  I don't think there's anything wrong with wanting a boy or wanting a girl.  I don't think there's anything wrong with not caring either, but it doesn't matter in that sense.

I think the issue is when desire turns to action, it's control, and I think to me at least the issue is or one of the issues anyway is the kind of control of offspring that is set up there in the sense that means a less than unconditional affirmation of somebody who turns out to be other than one desired.

That to me would be the problem, not just having a desire. 

I think there's a second kind of issue that probably moves beyond —

PROF. SANDEL:  Could I just add quickly?  But you don't think the desire could ever condition the affirmation?

PROF. MEILAENDER:  I think it could, but I think it's much more likely that the availability of the technology is what brings the conditions rather than just having the desires.  Because technology turns out not just to free us and give us new options, but to shape us and constrain us in various ways.

So that, in fact, is what I think generally happens.  Now, if I may make one more point that kind of in a way relates a little more to where Rebecca had started us, but I think an additional question here — I mean I think you're right, Michael that it's important to ask, you know.  Get out the means and everything.  Is there any reason to just object to choosing the sex and can we articulate what it is?

But relates to the demography stuff is if there are undesirable larger consequences, one of the issues that we're raising then is if my desire is innocuous, if there's nothing wrong with it, whether I have to suffer for the larger good, whether my desire should be thwarted, whether we should regulate it.

And that is a question I think at least in a society like ours that also needs to be faced.

CHAIRMAN KASS:  Bill, Bill May.

DR. MAY: Partly a response to yours, but also to return to the issue you were raising.  As I listened to you, I couldn't help but think about how studying comparative information of this kind accentuates for us the difficulties of anticipating consequences.

As I think about India, just a passing word.  The fact that towards the third and fourth child, one even more intensely wanted a boy does relate to economic circumstances there.  The dowry that has to accompany the girl into a marriage, and I know farmers there who have gone into major debt and wanted a boy.

Terrific pressure on boys later on.  If they don't happen to hold to that system of a dowry and coming to this country and working on computer work to be able to send money back at home to help out the father to escape from the economic bag or trap he was in.

So there are lots of pressures of that kind that haven't surfaced in this discussion, but at the same time it was very interesting you mentioned that at Punjab modernization tended to move us in this direction.  I wondered to what degree that related to the fact that the boys in those provinces or regions, modernization meant to move to the city, and there is more mobility in the boys and girls, and so there might be more reason, again.

But all of this traffics in the question of micro controlling as opposed to macro controlling, and you say the problem of micro controlling is acute only if it's going to produce macro problems, and that may be the wrong way of looking at it, and that I think is what you, Michael, are forcing us to consider.

The deeper problem as it relates to parenting for me is the whole question of how parenting relates us to the question of the unbidden and the unelected in life.

I mean, we elect a mate in our kind of culture, but we pick up with that a lot of things that we haven't elected:  the in-laws, and the genetic load, and so forth.  There's a whole unelected, and the testing out of a relationship very much depends on our capacity to rise to what has been unbidding and unelected.

And that's certainly also true of children.  In that very delicate issue is it okay to desire something else because is that going to hamper your ability to be open to what is unelected is the exchange I see between the two of you.

You happened to mention prayer, and it's a very interesting issue, intercessory prayer in the setting of the Christian tradition.  Is intercession the way in which we, absent of technology, look to God to provide us with the deus ex machina response?

Within the machine we don't have the technology.  So we need this cosmic bellhop who will do certain things to fulfill our wants and our desires.

And there's another way of looking at intercessory prayer, is that circumstance under which we now temper our desires.  We express them, but we temper them.  "Let this cup pass from me, but not my will, but thine be done," and suggests a way in which one hands over into the hands of another, and that other is not defined as the cosmic bellhop.

But the way in which I am sustained and supported to be open to the unbidden in life, the unelected in life, which nevertheless to which I have to rise.

So I see you, Michael, have raised a very important issue for us in your opening statement here.  In addition to harm and so forth or long range impacts and so forth, what are the impacts in our whole understanding of the relationship of parents to children, mates to one another, is the issue to which we return here.

CHAIRMAN KASS:  Frank and then robby Ó- sorry.  Bill and robby.

PROF. FUKUYAMA: Well, this may not answer Michael's question of why this is chilling, but I have my own take on why this is important as an issue, which is that this is a textbook case of a negative externality where you have a decision based on medical technology that is individually rational for the parents, but has a negative social cost and has a population level effect.

This is not a moral.  You know, this is, I think, a good answer to people who say, you know — use this libertarian argument that says, "Well, the only thing wrong with eugenics was that it was state sponsored, and as long as it's individual parents that are doing it, we don't have to worry because, you know, they love their children and they know what's best, and so forth."

And so this just seems to be — and this may not be the most important issue, but it does seem to me it's a model for other decisions that will be possible in the future where you could have perfectly individual, rational decisions that will lead to population level effects that will be bad for society as a whole, which then, you know, I think even for the most libertarian economist, I mean, you know, is grounds for some kind of social regulation.

  Now, the troubling desire, you know, I have this mental exercise of a different case of this kind of selection that would be a lot more troubling to a lot of people.  Let's say that there is a gauging or, you know, you can actually in the privacy of your doctor's office, in effect, select the sexual orientation of your child and speculate as to what the population level effects of that kind of power would be

And my private, you know, privately held opinion is that even the most tolerant, you know, person with plenty of gay friends, and so forth, in the privacy of their doctor's office is very likely to, all other things being equal, you know, if they had that choice available, you know, they would avoid having a child with a proclivity to gayness, and it doesn't have to be genetic.  You  know, it could be just, you know, done through drugs or some other thing.

If you had that kind of technology cheap and relatively simple, I would think that you'd get population level effects, you know, that would seriously affect the number of, you know, distribution of gays within the society.  It could happen, you know, within a generation.

So the importance is really not the sex selection issues per se, but just the fact that this does demonstrate that there are new technologies that can have, you know, larger social consequences on the basis of individual choice.  So that eugenics is not troublesome simply because it's state sponsored, but can be quite troublesome as a result of disbursed, decentralized individual parental choices.

CHAIRMAN KASS:  Bill Hurlbut and then  robby, and then I think we'll have to break.

DR. HURLBUT:  I want to ask you a quick question, and then I want to make a comment.

Are there statistics on the gender realities of abortion in America?

DR. EBERSTADT:  Not that I'm aware of.  There may be some data of some sort kept.  Dr. Haney, you may know this better than I, but since the total estimates for the number of terminated pregnancies or abortions in any given year aren't exactly fixed, there correspondingly are other sorts of data one might want to get.

DR. HURLBUT:  I mean, it seems apparent if you add up the huge numbers of lost females in your international statistics that that must be driving the abortion rate strongly in certain countries.

I've read statistics concerning India, certain hospitals where 99 percent of the abortions are female fetuses.

DR. EBERSTADT:  I've read the same sorts of accounts.

DR. HURLBUT:  The second question I want to ask you:  is there a correlation with being noted with crime rates as there are disproportions of males to females?

DR. EBERSTADT:  Internationally?

DR. HURLBUT:  I know it's a hard one to weigh.

DR. EBERSTADT:  Not that I am aware of.  Everyone has — in all societies one always has impressions about the way things worked in the good old days and the way that things are, you know, falling apart now.

And certainly in most of the places in East Asia that we describe, they're not seen as high crime settings.  I wouldn't have the data to support that sort of generalization right now.

DR. HURLBUT:  It seems to me that one likely consequence of this in the regions of the world where there's a strong disproportion maybe 18 years later is going to be a high rate of prostitution and maybe increased rates of sexually transmitted diseases, which may be local to some extent, but has international implications.

DR. EBERSTADT:  I think that's a very good question to ask.

DR. HURLBUT:  What strikes me about this is if you add some of this up, you might actually say that there could be a social advantage overall.  This is the individual versus the state.  There might be a social advantage by governmentally imposed sex ratio of the opposite dimension, that at least in our country crime rates correlate with the number of young males more than they do with anything else.

They've tried to correlate this with drugs and post war and so forth.  It always comes out that it's the number of young males, and so you could theoretically hypothesize a better society by doing something like sex ratio engineering.

That leaves certain questions unanswered like who's going to marry whom, but I just bring it up because when I think about Michael's very deep and important questions, it seems to me that there's something going on under the surface about nature, and we somehow relate this question of God and prayer and nature all together as a single unit.

You start out with the interesting and ironic comment about God's provision for the balancing of the sexes because of male recklessness and death and war and so forth.  And it was kind of funny; it seemed an anachronism.

Nevertheless, underneath the surface that assumption seems to be holding in a broader way for both religious and nonreligious people; that there's something about the way nature does things that seems to be better, wiser, more balanced.  Even if we could produce a better society by decreasing the number of males, therefore decreasing crime or something like this, that it wouldn't be right even then and that, therefore, when we pray for one particular sex or another, there's the element that Bill May was suggesting, that there is a humility in our prayer.  Whereas in our technologies the humility seems to be lost.

DR. EBERSTADT:  I think that Sussmilch did not intend his comment about the Creator at all to be ironic.  I think he was completely serious in his description.

It sounds a little antique to some people today, but I think he was completely serious in his description.

And as for speculating about young males, if there were a technology which could cryogenically freeze our children at the beginning of adolescence and unfreeze them at the end of adolescence, people might pay a lot for that.  I don't know what the ethical implications of that would be.

CHAIRMAN KASS:  robby, and then we'll break.

PROF. GEORGE:  Dr. Eberstadt, a quick question of clarification, and then I want to discuss the matter that Michael Sandel raised.

In the charts that gave us information about the expressed preferences of women in various countries with regard to additional children or having children or additional children, the question was put to married women or was it married women of childbearing age?

DR. EBERSTADT:  It was married women of childbearing age.

PROF. GEORGE:  Of childbearing age.  Okay.


PROF. GEORGE:  I didn't think that was indicated.  Thanks.

I'm sorry Michael is not here.  I thought Michael was right or is right to warn us not to assimilate concerns we may have about sex selection entirely to familiar concerns that have been expressed in other domains, and the two that he mentioned were discrimination and abortion and infanticide.

But I wonder if that means we need to find that when we think the thing through we'll find a new set of concerns, a new set of concerns laying somewhere deep in our consciousness or whether it might not be yet another familiar concern, but not one of the two.

And the concern I have in mind, particularly in listening to Bill and Gil is the concern that Leon expressed many years ago and influenced some of us.  I realize not all of us accept this, and that is the concern about turning procreation into production or manufacture.

It was a concern that was voiced by some of us, including Leon, at the very beginning of our deliberations about human cloning and, of course, has been expressed by critics of IVF and other reproductive technologies.

And it does sound to me when I hear Bill and Gill, and their points sound very persuasive to me, that there is something about sex selection which seems incompatible with a posture toward a child, a posture toward the coming to be of a new child that really does treat that child not as a product of manufacture, but as a gift to be received.

And I think that if that distinction holds, and if we  can also distinguish, as I think Christians and Jews would want to do — and I'm not saying that other religious traditions would think about this differently,  I just don't know enough about them to say — but if we would do what Christians and Jews do about distinguishing prayer from magic so that we're not trying to use prayer as an efficient means.  It's just the best one we happen to have to this productive end.

Then I think we can begin to understand that there really is a distinction between praying to have a boy or praying to have a girl and an act of the will, a choice, the use of a technology, and in the process willing that we're going to have a boy or going to have a girl such that the lack of success in that enterprise would constitute a failure of our effort in a way that a Christian or Jew would never say our prayers were a failure.  They didn't produce the result that we aimed to produce.

CHAIRMAN KASS:  Thank you.

Dr. Haney, Dr. Eberstadt, any final comments that you'd like to make?

I want to thank you both for a really very informative and instructive and interesting morning.  We've run over, as is our habit.  We were supposed to start at 1:45.  It gives members an hour and 15 minutes for lunch.

(Whereupon, at 12:27 p.m., the meeting was recessed for lunch, to reconvene at 1:45 p.m., the same day.)

  - The President's Council on Bioethics -  
Home Site Map Disclaimers Privacy Notice Accessibility NBAC HHS