Social Capital Francis Fukuyama
Social Capital Francis Fukuyama
Social Capital Francis Fukuyama
FRANCIS FUKUYAMA
Delivered at
Brasenose College, Oxford
May 12, 14, and 15, 1997
FRANCIS FUKUYAMA is Omer L. and Nancy Hirst Professor of Public Policy at the Institute of Public Policy at
George Mason University and director of the Institutes
International Transactions Program. Educated at Cornell,
he received his Ph.D. from Harvard University. He has
been a member of the Political Science Department of the
RAND Corporation, where he is currently a consultant,
as well as a member of the Policy Planning Staff of the
U.S. Department of State. He is a member of the American Association for the Advancement of Slavic Studies
and the Council on Foreign Relations; the editor, with
Andrez Korbonski, of T h e Soviet Union and the Third
W o r l d : T h e Last Three Decades (1987) ; and book review
editor at Foreign A f a i r s . His publications include T h e
End of History and the Last M a n ( 1 9 9 2 ), which received
the Premio Capri and the Book Critics Award (from the
Los Angeles T i m e s ) , and Trust : T h e Social Virtues and
the Creation of Prosperity (1995), which was named business book of the year by European.
[377]
378
boy, his mafioso father made him climb a wall and then invited him to jump, promising to catch him. H e at first refused,
but his father insisted until finally he jumped - and promptly
landed flat on his face. The wisdom his father sought to con3
[FUKUYAMA]
Social Capital
379
vey was summed up by these words: You must learn to distrust even your parents. 6
The Mafia is characterized by an extremely strong internal code of
behavior, l'omert, and individual mafiosi are spoken of as men
of honor. Nonetheless , these norms do not apply outside a small
circle of mafiosi; for the rest of Sicilian society, the prevailing
norms can be described more as take advantage of people outside
your immediate family at every occasion because otherwise they
will take advantage of you first. Obviously, such norms do not
promote social cooperation, and the negative consequences for
both government and economic development have been documented extensively.7
The norms that produce social capital, by contrast, must substantively include virtues like truth-telling, the meeting of obligations, and reciprocity. Not surprisingly, these norms overlap to
a significant degree with those Puritan values that Max Weber
found critical to the development of Western capitalism.
It is clear that the norms that produce social capital are partible: that is, they can be shared among limited groups of people
and not with others in the same society. While social capital exists
in all societies, it can be distributed in very different ways. Families are obviously important sources of social capital everywhere.
But family structure differs from one society to another and the
strength of family bonds differs, not simply from family ties in
other societies, but relative to other types of social ties. In some
cases, there appears to be something of an inverse relationship between the bonds of trust and reciprocity within kinship groups and
between kin and nonkin; while one is very strong, the other is very
weak. What made the Reformation important for Weber was not
6 Diego Gambetta, T h e Sicilian Mafia: T h e Business of Private Protection
(Cambridge: Harvard University Press, 1 9 9 3 ) , p. 35.
7 See, for example, Edward C. Banfield, The Moral Basis of a Backward Society
(Glencoe, III.: Free Press, 1958); and Robert D. Putnam, Making Democracy Work;
Civic Traditions in Modern I t a l y .
380
[FUKUYAMA]
Social Capital
381
382
[FUKUYAMA]
38 3
Social Capital
T HE P ROBLEM
OF
M EASUREMENT
38 4
[FUKUYAMA]
Social Capital
385
386
[FUKUYAMA]
Social Capital
387
388
Divorce
Chart 1 shows the U.S. divorce rate from 1920 to 1994. Discounting a war-induced uptick in divorces, the rate of change increases very dramatically after about 1967. While it has leveled
off in the 1980s, this does not reflect so much an increase in marital
stability as the passing of the baby boom cohort through the age
when they are most likely to divorce. Approximately half of all
marriages contracted in the 1980s in the United States could be
expected to end in divorce. The ratio of divorced to married persons (chart 2 ) has increased at an even more rapid rate due also
to a parallel decline in marriage rates; for the country as a whole
this rate has increased over fourfold in the space of just thirty
years. 13
It has been widely recognized that American social patterns
are exceptional and differ substantially from those of other industrialized countries.14 The United States has always been significantly more individualistic, had a smaller welfare state, and
generally exhibited higher levels of social deviance than most
European states, particularly those of continental and northern
Europe. Many American observers trace the breakdown of the
nuclear family to specifically American cultural factors, like the
12
This argument is made, for example, in Ladd, The Myth of Moral Decline.
13
390
youth counterculture and the crisis of authoritative American institutions in the wake of the Vietnam War and Watergate scandals.
Charts 3 through 7 show that virtually every country in Europe
has experienced a massive increase in divorce levels, and in many
cases the rate of change has been as high as that of the United
States. After settling down from wartime disruption in the 1950s,
families began breaking apart in Europe at around the same time
as in the United States. There are some individual variants
Germany and France have relatively lower rates and the Nordic
countries as well as Britain have higher ones. Other European
Catholic countries like Italy, Spain, and Portugal did not legalize
divorce until rather late in this period (1970, 1981, and 1974, respectively);15 once divorce was legal, however, divorce rates began
to climb rapidly in those societies as well, as tables 1 and 2 indicate.
A s chart 7 indicates, the United States started with and ended
up with a higher divorce rate than any country in Europe. Chart 3
indicates another outlier within the larger group of OECD coun-
[FUKUYAMA]
Social Capital
391
392
[FUKUYAMA]
393
Social Capital
1971
1975
1980
1986
1988
0.19
0.6
0.21
0.21
0.29
0.44
0.44
TABLE 2. D IVORCE
AND
IN
SPAIN,
1981-88
1981
1985
1988
18291
25046
22449
33240
tries: Japan, whose rate has increased only modestly over the postwar period.
Illegitimacy
Higher divorce rates have meant larger numbers of children
growing up without both parents at home. In addition, steadily
increasing proportions of children have been born out of wedlock.
Births to unmarried women as a proportion of live births for the
United States climbed from under 5 percent to 31 percent between
1940 and 1993.16
A number of observers have pointed out that the reason that
the percentage of out-of-wedlock births has increased so dramatically is less because of an increase in the number of births to unmarried women than as a result of a steep drop in the fertility of
married women.17 This fact is sometimes adduced to argue that
the relatively high present-day U.S. illegitimacy rate should not be
1 6 U.S. Bureau of the Census, Statistical Abstract of the United States (Washington, D.C.: U.S. Government Printing Office, 1 9 9 6 ) , p . 79.
17
See, for example, McLanahan and Casper in Farley, State o f the Union, p . 1 1 .
394
of concern. It is not really clear why the fact that those women
best able to properly care for and raise children have decided to
have fewer, while those less able to do so are having more, should
be considered reassuring. The increase in fertility of unmarried
women is not trivial, moreover, having moved from about 12 live
births per 1,000 women to about 50 between 1950 and 1990.18
Furthermore, it is clear that illegitimacy is not spread evenly
among communities in the United States. Rather, it tends to be
concentrated in certain particular sectors where fatherlessness has
become a virtual norm. Chart 8 traces the number of births to unmarried women in the United States from 1970 to 1993 as a percentage of all births. While African Americans started out this
period having greater absolute numbers of illegitimate children
than whites, whites have caught up rapidly. Fatherlessness is the
condition of a significant majority of black American children, and
in certain areas of concentrated poverty having a father married to
[FUKUYAMA]
Social Capital
395
ones mother can be an extremely rare occurrence. (It is interesting to note, incidentally, that the overall illegitimacy rate for the
United States as a whole is now higher than the rate for the black
community in the mid-1960s when Daniel Patrick Moynihan wrote
his famous report on the black family.) 19
If we turn from the United States to the OECD world we
find that the United States is no longer such an outlier and that
virtually all industrialized countries, again with the exception of
Japan, have experienced extremely rapid rises in illegitimacy rates
(see charts 9-11). While some countries like France and the
United Kingdom saw increases in their rates somewhat later than
the United States, the increases when they came were even more
dramatic. In Scandinavia, illegitimacy rates are the highest in the
world and significantly greater than in the United States. (The
meaning of illegitimacy in Scandinavia and other parts of Europe
is different than in the United States insofar as unmarried cohabitation is much more common; many illegitimate children grow
up in a household with their biological parents.) Again, within
Europe, Germany has relatively low rates, while Italys rates are
lower still, reflecting, apparently, the delayed liberalization of
divorce laws.
It should be noted that from the standpoint of child welfare,
being born to a single mother is only part of a broader social problem. Of children born to married parents in the United States in
the 1990s, about 4 5 percent will see their parents divorce by the
time they are eighteen.20 In subcommunities like African Americans, the percentage is much higher, making the experience of continuously living with two biological parents throughout ones
childhood a relatively rare experience. These kinds of rates are
not without historical precedent -in colonial America, fewer than
19 Daniel Patrick Moynihan, T h e Negro Family: A Case for National Action
(Washington, D.C.: Office of Planning and Research, U.S. Dept. of Labor, March
1965).
20 Sara McLanahan and Gary Sandefur, Growing Up with a Single Parent:
What Hurts, W h a t Helps (Cambridge: Harvard University Press, 1994), p. 2.
396
[FUKUYAMA]
Social Capital
397
half of all children reached the age of eighteen with both of their
biological parents still living.21 The difference, of course, was that
in the eighteenth century loss of parents was overwhelmingly due
to disease and early mortality, while in the late twentieth century it
has come about largely as a matter of parental choice.
CRIME
398
[FUKUYAMA]
Social Capital
399
400
[FUKUYAMA]
401
Social Capital
heroin use in the United States between 1974 and 1994. The rates
for alcohol use have also declined marginally for all age groups
over this period.
EXPLANATIONS
FOR THE
GREAT DISRUPTION
402
direction of causality between phenomena that appear to be statistically correlated. One consequence is that various groups are
able to emphasize their favored causal factor and adduce plausible
evidence showing that theirs is the primary factor involved. The
evidence we have available will not permit us to establish causal
relationships that will satisfy all observers, but the data do permit
us to evaluate and in some cases eliminate certain causal claims as
insufficient to explain more than a small part of the phenomenon.
SOCIAL NORMS AND T H E N E W SCIENCE OF H U M A N NATURE
See Clifford Geertz, The Interpretation of Cu ltu res (New York: Basic Books,
1973), p p . 330.
[FUKUYAMA]
Social Capital
403
vironmental stimuli after birth, a view extended even to the acquisition of langu age. 24 Neoclassical economics, for its part, has a
highly developed branch -game theory -devoted to the generation of cooperative social norms by rational actors. The belief in
the social construction of norms was particularly strong in discussions of gender-related issues, where feminist scholars tended to
assert that observed differences between males and females reflected not differences in genetic endowments, but the cumulative
effects of generations of patriarchy and norms established by
male domination.
Over the past generation, however, a tremendous amount of
research has been carried out in the related fields of evolutionary
psychology, sociobiology, and evolutionary genetics that cumulatively has had the effect of restoring a concept of human nature as
the starting point for any social science .25 Evolutionary psychologists, for example, start from the view that any given set of genetic
endowments in an organism exists by virtue of its adaptive significance at some earlier stage in the organisms evolution. The
characteristics subject to genetic influence are not simply physical
ones, but mental attributes as well, many of which produce behavioral regularities not simply across human cultures, but across
diff erent animal species. Evolutionary psychology has produced
a series of testable hypotheses about human behavior that have
received substantial empirical confirmation. While most of these
studies have not been able to connect these behavioral regularities
to specific parts of the human genome, there is good reason to
expect that this will occur over the next several decades. 26
24 This view of the human mind as a tabula rasa goes back at least to Locke.
For a critique of Skinner, see Lionel Tiger, The Manufacture of Evil: Ethics, Evolution, and the Industrial System (New York: Marion Boyars, 1 9 8 7 ) ,pp. 19-21.
25For an overview, see Robert Wright, T h e Moral Animal: W h y We Are the
W a y W e Are: T h e N e w Science of Evolutionary Psychology (New York: Vintage
Books, 1994).
26 Already there are highly suggestive studies linking behavioral regularities to
particular, highly specific parts of the brain, as well as studies in molecular biology
404
Evolutionary psychologists have uncovered behavioral regularities that have a direct bearing on the question of social capital.
Take, for example, the question of altruism. Richard Dawkins has
made us all aware that, contrary to Karl Marx, man is not a species being, i.e., does not have a genetic predisposition to sacrifice
individual interests for the sake of the survival of the species as a
whole, or even for the sake of larger collectivities like nations or
tribes.27 However, there is substantial evidence that altruism is the
rule among individual organisms sharing genes or gene sequences,
and indeed that the degree of altruism is strictly proportional to
the degree of genetic overlap. Thus parents can be expected to
exhibit unreciprocated altruism to their children and to their siblings; they practice a lesser degree of altruism toward cousins and
nephews with whom they share half as many genes, but still a
greater degree than toward second cousins or complete strangers.
The power of this approach is illustrated by the work of the
anthropologists Martin Daly and Margot Wilson on homicide.
Daly and Wilson argue that from a neo-Darwinian point of view
the incidence of homicide should be inversely proportional to the
genetic relatedness of the murderer and the victim. After an exhaustive study of police records, they showed that contrary to the
popular perception that most murders take place within families,
the rate of homicide is in fact lower among biological relatives, as
the theory would suggest. That is, given the same opportunities to
commit murder, murder of nongenetically related spouses or steprelatives is far more common than the murder of children by parents, brothers by brothers, etc.28
linking individual genes or gene segments to inherited traits like breast cancer or
obesity. Given the rapidity of progress in the area, we can only expect more empirical evidence to appear in coming years explicating the biological origins of a good
deal of what we have assumed to be social behavior.
27Richard
[FUKUYAMA]
Social Capital
405
406
far lower level of parental investment to get their genes into the
next generation, and their strategy therefore involves spreading
their genes as widely as possible.
These diff ering reproductive requirements lead to diff erent psychological proclivities that correspond to our commonsense perception of the difference between women and men. Females necessarily tend to emphasize the quality of their mates over sheer numbers, since they cannot increase their fertility past a certain natural
limit and because the mates quality-often measured by economic resources -will be critical to the survival of the mother's
offspring. Males, by contrast, can get their genes into the next
generation with a minimum of parental investment, simply by mating with as many females as possible. This difference in strategy
accounts for a wide variety of phenomena: the fact that males tend
to be more promiscuous than females; the fact that male and
female jealousy differs (males will care more if their mates had sex
with another partner, since this will imply nonpaternity, while
females will be more concerned with emotional involvement, since
this may mean loss of economic resources) ;31 and the fact that
high-status males (and not high-status females) will tend to accumulate multiple partners.
One of the staples of postwar cultural anthropology was the
seemingly infinite variety of kinship relations across human cultures, and indeed the changing nature of kinship- across time within
a single culture like that of Western Europe. Cultural relativism
was born in a climate that saw only difference and particularisms
among all of these varying practices.
Contrary to Clifford Geertz, however, there are in fact many
cultural universals that appear to be grounded in nature rather
than culture.32 As Lionel Tiger and Robin Fox have pointed out,
31 On this point, see Martin Daly and S. J. Weghorst, Male Sexual Jealousy,
Ethnology and Sociobiology 3 (1982) : 1127.
32For example, Robert Wright points out that the tendency for males to be
more promiscuous than females is not only true in virtually all human cultures, it is
[FUKUYAMA]
Social Capital
407
while kinship practices vary widely between cultures, the evolutionary function of kinship does not.33 The function of kinship is
to reconcile competing male and female reproductive strategies
through an exchange of fertility (controlled by the female) for
economic resources (almost always controlled by the male). This
exchange can take the form of the bourgeois family of Western
Europe and the United States in the 1950s, but need not. From
the womans standpoint, the chief objective is protection and resources for herself and her offspring through the reproductive
cycle, and it doesnt matter whether those benefits come from a
husband, or the husbands male relatives, from her brothers, or
from a larger kinship group. The variety of actual kinship relationships thus masks a commonality of function in terms of
reproduction.34
An understanding of the natural substrate on which social relationships are built permits us to begin to build causal connections between some of the phenomena documented above. The
most clear-cut is the one between family breakdown and child
abuse. Martin Daly and Margo Wilson were puzzled by reports
of rising rates of child abuse and suspected that the real story had
to do with substitute parents, since perhaps the most obvious prediction from a Darwinian view of parental motives is this: Substitute parents will generally tend to care less profoundly for children
than natural parents. 35 The data that Daly and Wilson found
also true across primate species and indeed, with two or three exceptions, across all
vertebrate species. Wright, T h e Moral An im al, pp. 39-49.
33 Lionel Tiger and Robin Fox, T h e Imperial Animal (New York: Holt, Rinehart, and Winston, 1971).
34 This Darwinian perspective explains why there are so many male polygamists
(either simultaneous or serial) across different human cultures and so few female
ones, and why conversely there tend to be more unattached low-status males than
low-status females. Since the kinship relationship is an exchange of fertility for
economic resources, it follows that those males with greater resources and higher
status will be able to support a large number of partners, leaving some males without partners. A wealthy or high-status female, by contrast, wouldnt need the economic resources of multiple partners, nor would multiple partners do much to increase her rate of fertility.
Daly and Wilson, Homicide, p. 8 3 .
408
from the United States and Canada showed that children were
anywhere from 10 to over 100 times more likely to suffer abuse at
the hands of substitute rather than natural parents.36
Family breakdown is also closely associated with crime more
generally, though more weakly than in the case of child abuse.
Crime, needless to say, has many environmental causes that have
been documented by criminologists over the years, including,
among other things, the nature of police enforcement and deterrence. But it also has a biological dimension: crimes are overwhelmingly committed by young males, in virtually all societies
around the globe. 37 There is substantial evidence that in addition
to being more promiscuous, young males are significantly more
aggressive and violent than females and than older males and that
this is the result of their underlying psychological makeup rather
than culture. Another function of kinship that cuts across virtually
all human cultures is the need for a society to control its young
males, and the male initiation ceremonies that occur from Polynesia to the U.S. Marine Corps are essentially means by which
older males socialize younger ones into the rules of their respective
societies. Just as male promiscuity needs to be controlled by the
institution of marriage, male aggressiveness needs to be controlled
by paternal authority. When this does not happen- when, for
example, large numbers of young males are being raised in femaleheaded families -then a society is, in Daniel Patrick Moynihans
phrase, looking for trouble. According to William Galston and
Elaine Kamark, The relationship (between family structure and
crime) is so strong that controlling for family configurations erases
the relationship between race and crime and between low income
and crime. This conclusion shows up again and again in the litera36See data ibid., pp. 86-87; also M. Daly and M. Wilson, Children as
Homicide Victims, in R. Gelles and J. Lancaster, eds., Child Ab use and Neglect:
Biosocial Dimensions (New York: Aldine de Gruyter, 1987).
37On this point, see James Q . Wilson and Richard J. Herrnstein, Crime and
Humun Nature (New York: Touchstone Books/Simon and Schuster, 1 9 8 5 ) ,
p p . 104-47.
[FUKUYAMA]
Social Capital
409
I:!
Wilson and Herrnstein, Crime and Human Nature, pp. 213- 85.
410
ships with other adults.40 While it is hard to find evidence supporting this directly, it would be surprising if the sharp declines
in expressed trust from the survey data were not in some way related to the breaking of primary trust relationships in families.
There are a number of other social indicators that are linked
to changes in family relationships that can only be touched upon
briefly. The decline in the educational achievement of American
primary and secondary students has been widely documented and
has been the source of endless studies over the past two generations. The causes for this decline are multiple, and the recommendations for how to reverse it equally diverse. A long series of
empirical studies beginning with the famous Coleman Report of
the 1960s has indicated, however, that the single factor that can
be reliably correlated with educational achievement is not per capita
spending on education, standards, curriculum, computers in the
classroom, teacher training, vouchers, or any of the other panaceas
offered up as public policy solutions, but rather the parents involvement in their childrens education.41 While educational
achievement is not directly linked to family structure, family structure is highly correlated with parental investment and socioeconomic status (as will be discussed in the section immediately following) and therefore has consequences for educational achievement
as well.42
The decline of the nuclear family and the failure to perform
the economic and socialization functions it provided appear, then,
to be related in varying degrees to many of the social indicators
that have turned negative in many developed societies in recent
years. Each one of the problems listed here has many differing and
4 0This point was made in Coleman, Social Capital; see also McLanahan and
Sandefur, Growing U p with a Single Parent, pp. 5 6 .
41James S. Coleman, Equality o f Educational Opportunity (Washington, D.C.:
U.S. Dept. of Health, Education, and Welfare, 1966).
42A summary of some of the evidence for this is presented in McLanahan and
Sandefur, Growing U p with a Single Parent, pp. 3 3 3 5 .
[FUKUYAMA]
411
Social Capital
OF
C HANGES
IN
F AMILY N ORMS
412
[FUKUYAMA]
Social Capital
413
414
independent variable explaining poverty. This would only confirm what we would assume from common sense: that households
with only one parent lose economies of scale and can draw on only
half as much income and labor compared to those with two (actually, less than half given lower average earnings for women).
They will consequently be able to devote less money, time, and
energy to their childrens education and socialization. This is in
fact borne out by the data: following divorce, households experience substantial drops in income, regardless of the parents predivorce socioeconomic status.46
Conservatives, for their part, have made two different types of
argument to explain family breakdown. The first, associated with
Charles Murray, is the mirror image of that of the left: it maintains that it is the perverse incentives created by the welfare state
itself that explain the rise in family b r e a kdown.47 The primary
American welfare program targeted at poor women, the Depressionera Aid to Families with Dependent Children (AFDC), provided
welfare payments only to single mothers and thereby penalized
women who married the fathers of their children.48
As noted above, the comparative data at first glance give greater
support to the Murray hypothesis than to its left-wing counterpart: high-benefit welfare states like Sweden and Denmark have
higher rates of divorce and illegitimacy than low-benefit ones like
Japan. There are numerous anomalies, however, beginning with
the fact that the United States, which has a substantially lower
level of welfare benefits than, say, Germany, has much higher
divorce and illegitimacy rates. Detailed studies of welfare benefits
in the United States have found similar discrepancies when corre46See the evidence summarized in McLanahan and Sandefur, Growing Up with
a Single Parent, pp. 79-94.
47This argument was made originally in Charles Murray, Losing Ground (New
York: Basic Books, 1984). It had actually been made earlier by Gary Becker in
A Treatise on the Family, enlarged ed. (Cambridge: Harvard University Press, 1991).
48The disqualification of married women was actually ended in many states during the 1980s.
[FUKUYAMA]
Social Capital
415
416
ments empirically the nature of the values shift that has been
going on across a wide variety of cultures, as have Daniel Yankelovichs opinion serveys.52 The question, however, is whether this
value shift was truly autonomous or whether it was in turn shaped
by other, more basic socioeconomic forces that make it to some
extent a dependent variable.
There is good reason to treat the view that the Great Disruption was caused by an autonomous shift in values with some caution. Everything we know about cultural variables is that they tend
to change very slowly when compared to other kinds of factors,
such as shifts in economic conditions, public policies, or ideology.
In those cases where cultural variables have changed quickly, such
as in rapidly modernizing Third World societies, cultural change
is clearly being driven by socioeconomic change and is therefore
not an autonomous factor. So with the Great Disruption: it is
hard to believe that people throughout the developed world simply
decided to change their attitudes toward such elemental issues as
marriage, divorce, child-rearing, authority, and community so as to
completely alter the nature of the family in the space of two or
three decades. Those explanations that link changes in cultural
variables to specific events in American history like Vietnam,
Watergate, or the counterculture of the 1960s betray an even
greater provincialism : why then were social norms disrupted in
other societies from Sweden and Iceland to New Zealand and Spain ?
To explain the Great Disruption, we need to look for largescale social changes that correlate with it both geographically that is, that would account for the varying levels of disruption in
different societies -and over time, that is, by explaining why the
disruption began in the 1960s rather than a generation earlier or
later.
52See Ronald Inglehart, Modernization and Postmodernization: Cultural, Economic, and Political Change in Forty-three Societies (Princeton, N.J.: Princeton University Press, 1997); and Daniel Yankelovich, How Changes in the Economy Are
Reshaping American Values in Henry Aaron et al., eds., Values and Public Policy
( Washington, D.C.: Brookings Institution, 1 9 9 4 ) , pp . 18ff.
[FUKUYAMA]
Social Capital
BIRTH CONTROL
417
AND WORKING W O M E N
418
The significance of birth control was n o t that it lowered fertility, since fertility had been on the decline in many societies prior
to the widespread availability of birth control or abortion. 54 Indeed, if the effect of birth control is to reduce the number of unwanted pregnancies, it is hard to explain why its advent should
have been accompanied by a rise in the rate of abortions, 55 or why
use of birth control is positively correlated with illegitimacy across
the OE C D. 56
The main impact of the Pill and the sexual revolution that followed was to dramatically alter calculations about the risks of sex
and thereby to change male behavior. The reason that the rates of
birth control use, abortions, and illegitimacy all went up in tandem
is that a fourth rate -the number of shotgun marriages -declined substantially at the same time (see chart 17). Since the Pill
54On
55 I
56 U.S. Dept. of Health and Human Services, Report to Congress on O u t - o f Wedlock Childbearing (Hyattsville, M d . : USGPO, 1995), p. 72.
[FUKUYAMA]
Social Capital
419
permitted women for the first time to have sex without worrying
about economic consequences, men felt liberated from norms requiring them to look after the women whom they had gotten
pregnant.
The second factor altering male behavior is entry of women
into the labor force. That female incomes should be related to
family breakdown is an argument associated most closely with
economist Gary Becker.57 Increasing female earnings make it possible, of course, for women to support themselves and their children without husbands. The more subtle consequence, however,
was to further weaken the norm of male responsibility. The weakening norm of male responsibility reinforced, in turn, the need for
women to arm themselves with job skills so as not to be dependent
on decreasingly reliable husbands.
There is considerable empirical evidence that Becker is right
about the importance, to put it crudely, of husbands as economic
commodities in marriage markets. Chart 18 shows the changing
rate of male v. female labor force participation in the United
57Becker,
4420
2
States 1960 and 1995. Not only does female participation jump
from 35 to 55 percent in this 35-year period, but male participation
actually drops from 7 9 to 71 percent. Chart 19 shows changes in
male and female median incomes in the United States between
1947 and 1995. It is interesting to note that male median incomes,
after dropping slightly in the immediate postwar period, rise
steadily through the late 1960s, after which they stagnate. Indeed,
male real median incomes were higher in the early 1970s than they
are today. Female incomes, however, after dropping slightly in the
late 1940s as female workers left wartime employment, have risen
steadily from the early 1960s to the present.
Chart 20 plots the change in the ratio of female-to-male median incomes over this same period. I noted earlier that the immediate postwar period was one in which divorce rates fell, fertility rose, and family stability actually improved. It is probably
something more than coincidence that the female-male median
income ratio actually shifted in favor of men in this period. How-
[FUKUYAMA]
Social Capital
421
422
[FUKUYAMA]
Social Capital
42 3
424
[FUKUYAMA]
Social Capital
425
426
SOCIAL N ORMS
AND
ECONOMIC I NCENTIVES
[FUKUYAMA]
Social Capital
427
dren through multiple partners. The fact that human males with
high status and/or wealth have done this readily in cultures from
China to Turkey to present-day corporate America suggests that
this drive is still very much with us (the difference being, of course,
that American corporate executives have their wives and children
serially rather than simultaneously as in the case of Ottoman
pashas or Chinese mandarins) .
It is easy to see how the change in economic conditions that has
occurred in recent decades has altered the norm of male responsibility. With birth control and easier abortion, the economic consequences of sex decline dramatically for women, so they can
afford to be much less selective in their choice of partners. With
the movement of women into the paid labor force, abandonment
of a wife and children does not have the same dramatic negative
consequences as it once did. Many women prize the greater freedom that economic independence brings -hence the feminist
movement -and their assertion of independence releases men
from the norm of family responsibility. Males do not have to be
persuaded to behave less than responsibly toward their families ;
there are plenty of biological forces pushing them in this direction
as it is. The growth of male irresponsibility then reinforces the
female drive for independence: even if a girl wanted to grow up
to be a dependent homemaker today, she would be ill-advised not
to equip herself with job skills, given that her marriage partner is
more likely than not to end up abandoning her and her children
or else will have difficulties providing a family income. When
these tendencies toward greater independence on the part of both
men and women are reinforced by a general Western culture that
prizes individualism, and by a specifically American culture that
denigrates the importance of virtually all inherited social duties
and obligations, it is easy to see why the United States ends up
with a serious degree of family breakdown and declining levels of
social capital.
428
n 1..t
[FUKUYAMA]
Social Capital
429
SC = (cn) 1..t
SC = (r p cn) 1..t
430
[FUKUYAMA]
431
Social Capital
(1.4) SC=
( ( l / rn ) r p c n )
l..t
To some extent, we could expect that c and rn might be positively correlated with one another. That is, internal cohesiveness is
often based on strongly shared norms and values within a group:
both the Marines and the Mormon Church are examples. But the
very strength of those internal bonds creates something of a gulf
between members of the group and those on the outside. Latitudinarian organizations like most contemporary mainline Protestant denominations in the United States, by contrast, easily coexist
with other groups in the society, and yet are capable of a much
lower level of collective action. Ideally, one would like to maximize the c and minimize the rn.values: such would be the case, for
example, in a professional organization that socializes its members
into the values of its particular profession, while not at the same
time breeding distrust of other professions or being closed to influences from them.
432
OF
SOCIAL CAPITAL
[FUKUYAMA]
Social Capital
433
observed growth, and I believe that cultural factors play an important role in accounting for some of that residual. Precisely how
much of a role they play we may never be able to say because of
the measurement problems alluded to in the first lecture. As I
argued in Trust, the primary impact of social capital on the particular economies I studied seemed to be less on aggregate growth
itself than on sector specialization and the role of the state in
guiding the economy.
Many social scientists will accept the importance of exogenously generated norms in the capitalist economy, and many further believe that the latter plays a role over time in depleting social
capital that has been built up through noneconomic means. There
is a significant and interesting literature on the cultural contradictions of capitalism, which argues that capitalist development ultimately undermines itself by producing insufficient norms, or else
by producing norms at odds with those necessary for the operation
of markets and innovation. Perhaps the most famous exponent
of this view was Joseph Schumpeter, who argued in Capitalism,
Socialism, and Democracy that capitalism tended to produce a class
of elites over time that was hostile to the very forces that had
made their lives possible and that they would eventually seek to
replace market economies with socialist ones.1 Daniel Bell and
others have written about the problems of affluence in undermining the work ethic and other kinds of norms necessary to the capitalist order.2 It is a staple of the contemporary literature about
corporate layoffs and downsizing that the global economy has become an enemy of community by disrupting families, localities,
and the loyalties that at one time pervaded the workplace.3
1 Joseph A. Schumpeter, Capitalism, Socialism and Democracy (New York:
Harper Brothers, 1950).
2 Daniel Bell, T h e Cultural Contradictions o f Capitalism (New York: Basic
Books, 1976); see also John K. Galbraith, T h e Affluent Society (Boston, Mass.:
Houghton Mifflin Co., 1958).
3 See, for example, Michael J. Sandel, Democracys Discontent; America in
Search o f a Public Philosophy (Cambridge, Mass.: Harvard University Press, 1996).
434
The problem with this literature, apart from the fact that capitalism has not yet collapsed or otherwise undermined itself, is that
it is extremely one-sided. That is, it sees capitalism as a destructive, disruptive force that breaks apart traditional loyalties and
obligations, but fails to see the ways in which capitalism also
creates order and builds new norms to replace the ones it destroyed. Indeed, it may be possible that capitalism is a net creator
of norms. What I intend to argue in the remainder of this lecture
is that there are reasons for thinking that social capital and informal norms will become even more important as we move from
an industrial to a postindustrial or information-age economy and
as the complexity and technological level of an economy grows.
Whether these norms will actually be produced to meet this demand, and how this will come about, is a separate subject that I
will deal with in the final lecture.
[FUKUYAMA]
Social Capital
435
436
[F UKUYAMA ]
Social Capital
437
438
tems created by organized religions. A nongovernmental organization like Amnesty International or the National Organization for
Women also achieves coordinated action on the basis of shared
values; as in the case of friends or members of a religious organization, the behavior of the organizations individual members
cannot be explained on the grounds of economic self-interest alone.
A society like the United States is characterized by a dense, complex, and overlapping set of networks (see chart 2 9 ) .
Note two features of this definition. A network is different
from a market insofar as networks are defined by their shared
norms and values. This means that economic exchange within a
network will be conducted on a different basis than economic
transactions in a market. A purist might argue that even market
transactions require some shared norms (the willingness, for example, to engage in exchange rather than violence being one) ;
but the norms required for economic exchange are relatively minimal. Exchange can occur between people who dont know or like
one another, who speak different languages; indeed, it can occur
anonymously between agents who never know each others identi-
[F UKUYAMA ]
Social Capital
439
440
of whose members share a common corporate culture. While networks can coexist with formal organizations, they can, at one
extreme, also replace formal organizations.
It should be clear that when networks are overlaid on top of
formal organizations, the results are not necessarily beneficial and
indeed can be the source of a good deal of organizational dysfunction. Everyone is familiar with patronage networks, based on kinship, friendship, love, or some such factor. Members of such a
network share important norms and values with one another (particularly reciprocity) that they do not share with other members
of the organization. Chart 31 illustrates a patronage network
superimposed over an organization with a common corporate culture. Within the patronage network, information passes readily,
but its outer boundaries constitute a membrane through which information passes much less readily. Patronage networks are problematic in organizations because their structure is not obvious to
those outside of them, and they often subvert formal authority relationships. Common ethnicity may facilitate trust and exchange
among members of the same ethnic group, but it inhibits exchange
between members of different groups. Often, the bonds of reciprocity between members of a network are undertaken for noneconomic reasons and stand in the way of economic rationality, as
when a boss is unwilling to criticize or fire an incompetent subordinate who is a protg or personal friend.
The other problem with informal networks is that there is an
inverse proportion between the strength of the values or norms
linking the community (and therefore the degree of coordination
they can achieve) and their openness to people, ideas, and influences from outside the network. Being a member of the U.S.
Marine Corps or the Mormon Church involves much more than
membership in a formal organization; one is also socialized into a
strong and distinctive organizational culture that creates a high
degree of internal solidarity and potential for coordinated activity.
However, the cultural gap between Marine and civilian, or Mor-
[FUKUYAMA]
Social Capital
441
442
C H A N G I N G METHODS
OF
C OORDINATION
1974).
[FUKUYAMA]
Social Capital
443
444
[FUKUYAMA]
Social Capital
445
firms needs to hire hotshot economists or software engineers, because their skills and performance are much harder to define in
formal terms.
Finally, hierarchies can be less adaptive. Formalized systems of
control are much less flexible than informal ones; when environmental conditions change, they are often more visible to lower
levels of the organization than to higher-level ones. Hence overcentralization can be a particular liability in areas of rapid environmental change, such as the present-day information technology
industry.
The reason that networks, defined as groups sharing informal
norms and values, are important is that they provide alternative
conduits for the flow of information through an organization.
Friends do not typically stand on their intellectual property rights
when sharing information with each other and therefore do not
incur transaction costs. Friendships thus facilitate the free flow of
information within the organization. Nor do friends usually spend
a lot of time strategizing over how to maximize their relative
power positions vis--vis each other. Someone in marketing knows
someone in production and tells that person over lunch about customer complaints concerning product quality, thereby bypassing
the formal hierarchy and moving information to the place where
it is most useful more quickly. A corporate culture ideally provides an individual worker with a group as well as an individual
identity, encouraging effort toward group ends that again facilitate information flow within the organization.
Social capital is also critical to the management of highly
skilled workers manipulating complex, diff use, tacit, or hard-tocommunicate knowledge and processes. Organizations from universities to engineering, accounting, and architectural firms generally do not try to manage their professional staffs through detailed bureaucratic work rules and standard operating procedures.
Such workers are usually trusted to be self-managing on the basis
of internalized professional standards. Doctors presumably will
44 6
[FUKUYAMA]
Social Capital
447
incentives, whether rewards or punishments, and were readily interchangeable with one another, Reacting to this system through
their unions, the blue-collar labor force demanded formal guarantees of their rights and the narrowest possible specification of
duties, Hence the rise of job-control unionism and labor contracts
that were as thick as telephone books.15
Taylorism was an effective means -perhaps the only means of coordinating the activities of a low-skill industrial labor force.
In the first two decades of the century, half of Fords blue-collar
workers were first-generation immigrants who could not speak
English, and as late as the 1950s 80 percent did not have a high
school education. But Taylorism ran into all of the problems of
large, hierarchical organizations, with slow decision-making, inflexible workplace rules, and an inability to adapt to new circumstances.
There is a certain parallel between the Taylorite factory and
the concept of the modern, rational, bureaucratic state. And it is
no accident that the Tayloritefactory is undergoing rapid change
today just as the modern state appears to have reached its limits.
The limits of Taylorism have been discussed extensively in the
existing literature, and it is not necessary to detail them here, nor
to describe the flatter, less hierarchical forms of management that
have replaced it. The point I want to stress is that the move from
a hierarchical Taylorite organization to a flat or networked one
involves off loading the coordination function from formal bureaucratic rules to informal social norms. Authority does not disappear
in a flat or networked organization; rather, it is internalized in a
way that permits self-organization and self-management.
A lean or just-in-time automobile factory is an example of a
flat, post-Fordist organization. In terms of formal authority, many
of the functions previously assigned to white-collar middle managers have now been undertaken by blue-collar assembly-line
workers themselves acting in teams. It is the factory-floor work
15 Harry Katz, Shifting Gears: Changing Labor Relations in the U.S. Automobile Industry (Cambridge, Mass.: MIT Press, 1985).
448
[FUKUYAMA]
Social Capital
449
450
[FUKUYAMA]
Social Capital
451
Ibid., p . 3 3 .
2 1 Thomas
452
[FUKUYAMA]
Social Capital
453
454
[FUKUYAMA]
Social Capital
455
456
[FUKUYAMA]
Social Capital
457
45 8
for governments to intervene to regulate polluters or other producers of negative externalities, because the parties negatively impacted will have a rational incentive to organize and buy off the
miscreant. Social regulatory norms, in other words, will simply
arise out of self-interested interactions of individual agents and do
not have to be mandated through law or formal institutions.
While the zero-transaction-cost requirement is almost never
met in any real world situation, economists have by now studied
quite a number of intriguing cases of self-organization, whereby
social norms have been created through a bottom-up process. Andrew Sugden describes rules for the sharing of driftwood, 3 and
Robert C. Ellickson gives numerous examples of spontaneous economic rules, including his own detailed field research on the interactions of ranchers and farmers in Shasta County, California.4
There are many other fascinating examples that bear looking into,
including the ways in which trust norms developed among members of the New York Stock Exchange, the ways in which oil companies share excess capacity, and the way in which railroads developed a mechanism for sharing boxcars.
However, game theory is limited as a general explanation for
how social norms are generated. The problem goes straight to the
heart of the model of human behavior underlying modern neoclassical economics: namely, that few human beings enter into economic exchange without bringing with them a host of previously
existing social norms that strongly influence their willingness and
ability to coopera te.5 Indeed, there is a long literature associated
with Emile Durkheim and Karl Polanyi that argues that market
exchange itself is dependent on the existence of prior social norms
3 Andrew Sugden, Spontaneous Order, Journal of Economic Perspectives 3
(1989) : 85-97; Sugden, T h e Economics of Rights, Co-operation and W e l f a r e
(Oxford: Basil Blackwell, 1986).
4 Robert C. Ellickson, Order without Law: H o w Neighbors Settle Disputes
(Cambridge, Mass.: Harvard University Press, 1991).
5
On this point, see Jon Elster, Social Norms and Economic Theory, Journal
[FUKUYAMA]
Social Capital
459
460
retic process, since reciprocity turns out to be in the individual organisms selfish interest. But individuals do not necessarily arrive
at reciprocity through use of their individual reason; their rational
self-interest appears to receive a powerful boost from an instinctive drive in that direction. The biologist Robert Trivers actually
ran a Prisoners Dilemma game in which cooperation between the
prisoners was induced by the presence of a hostile third party in
the room -a factor that seems to reflect some inherited predisposition to cooperate, and not the calculations of coolly rational
agents.8
The problem posed by game theory could thus easily have been
stated differently: how, given the fact that there are strong cultural and biological impulses toward cooperative behavior, does it
come about that cooperation often breaks down ?
In reviewing the game-theoretic literature, it is remarkable how
little empirical evidence there is showing that an actual social
norm of some significance to a given society was actually generated
in a game-theoretic manner. There are, of course, many post hoc
ergo propter hoc arguments, which first seek to discern an economically rational reason for the existence of some social norm and
then go on to assert that the norm must have been generated in a
game-theoretic manner.
Take, for example, the Confucian rules on filial piety and inheritance, which, as I have shown elsewhere, have a significant
impact on the development of the Chinese economy. It is possible
to argue retrospectively that these rules were a rational adaptation
to the environmental conditions of traditional Chinese life.9 That
is, sons were seen as an economic resource that would be particularly important as a form of old-age insurance in a society where
8 Robert Trivers, The Evolution of Reciprocal Altruism, Quarterly Review of
Biology 46 (1971): 35-56.
9 I made a version of this argument in Trust (New York: Free Press, 1 9 9 5 ) ,
p p . 69-82.
[FUKUYAMA]
Social Capital
461
462
A TAXONOMY OF NORM-GENERATION
1 . Institutionally constructed norms. By institutionally con-
a. rational
b. irrational
2 . spontaneously constructed
a. rational-game theoretic
b. irrational-common law model/complex adaptive systems
3 . exogenously constructed
a. religion
b. ideology
c. culture and shared historical experience
4. natural
a. kinship
b. race and ethnicity
[FUKUYAMA]
Social Capital
4 63
464
[FUKUYAMA]
Social Capital
465
lar to those of the game-theoretic rational spontaneous constructivists, Friedrich Hayek and other Austrian economists argued that
social norms are the result of a long-term spontaneous evolution.12
Hayek stressed, however, that this process is not a rational one and
quoted Hume approvingly to the effect that the rules of morality . . . are not conclusions of our reason. 13 By taking this
position, Hayek sought to debunk the rational constructivism he
saw as the core of the socialist project. With the model of English
common law in mind, Hayek argued that social norms are generally not legislated through a formal political process, but are
rather the result of the repeated interactions of individuals seeking to achieve common aims. H e argued that while formal norms
(i.e., law) are necessary for modern economies, they are by and
large a modern invention and do not account for the vast majority
of actual norms that constitute a civilization. Because they evolve
spontaneously as a result of the continuing interaction of communities with their environment, such norms are highly adaptive.
Political freedom is necessary, in Hayeks view, to protect the complex web of evolved social norms that emerge in an open society.
Hayeks description of social norm-formation anticipated the
rise of studies in so-called complex adaptive systems in the natural
sciences.14 Over the past fifteen years, there has been a great deal
12 Friedrich A . Hayek, Law, Legislation, and Liberty (Chicago: University of
Chicago Press, 1983); T h e Fatal Conceit (Chicago: University of Chicago Press,
1988).
13
14 There
466
[FUKUYAMA]
Social Capital
467
(1996) 312.
468
[FUKUYAMA]
Social Capital
469
ample, are bound by a common religion and ethnicity, but are also
shaped by common experiences of persecution that create solidarities of a different sort. Many cultural phenomena can have relatively recent political or economic roots: hence the postwar German central banks emphasis on a strong Deutschmark and a tough
anti-inflationary policy is said to be a direct outcome of the German experience with hyperinflation during the Weimar period.
4 . Norms rooted in nature. Despite the changes in family
structure described in the first lecture, kinship remains the most
powerful form of social relationship in contemporary societies. As
I indicated in Trust, the importance of kinship relative to other
kinds of social structures varies considerably from one society to
another, but there is no society in which it has completely withered
away.
Beyond the family, classical liberalism argues that human beings in the state of nature are isolated individuals who come together with other human beings in civil society only as a means of
satisfying selfish appetites. While differing in many particulars,
Hobbes, Locke, and Jean-Jacques Rousseau all deny that human
beings are sociable by nature and see duties to others as derivative
of rights. Modern liberal politics and modern neoclassical economics both begin from this individualistic premise.18
There is considerable evidence from the natural sciences that
suggests that sociability is in fact natural not only to humans but
to many other primate species as well. While this sociability is to
a degree rooted in and reinforced by culture, culture itself is not
unique to Homo sapiens, and there is evidence that certain basic
tendencies toward sociability are hard-wired into the genetic code.
The anthropologist Lionel Tiger has suggested that male bonding
18 On this point, see Mark Granovetter, Problems of Explanation in Economic
Sociology, in Nitin Nohria and Robert G. Eccles, eds., Networks and Organizations: Structure, Form, and Action (Boston, Mass.: Harvard Business School Press,
1 9 9 2 ) , p. 38; see also Mary Ann Glendons account of the lone rights bearer in
Rights Talk: T h e Impoverishment of Political Discourse (New York: Free Press,
1 9 9 1 ) , pp. 4 7 7 5 .
470
20 James
[FUKUYAMA]
Social Capital
471
law, religion, and spontaneously generated social constraints simultaneously. Much of the discussion of the origins of norms is politically loaded: the left has traditionally argued that norms are institutionally grounded and can be shaped as a result of deliberate
policy by states and other institutions. The right, for its part, tends
to argue that norms arise from religion, culture, or nature and that
these norms are themselves a constraint on the ability of states or
institutions to shape human interactions. Needless to say, there is
some truth in both points of view; what is important is to understand the full range of sources for norm-generation and to understand the opportunities and limits of each. A s it turns out, the
choice of source for social norm generation is quite important; institutional means cannot readily substitute for spontaneous or exogenous norm-creationand vice versa.
RATIONALITY AND FAMILY DECLINE
472
[FUKUYAMA]
Social Capital
473
To an economist, economic goods are fungible: if those resources necessary to provide for a mother and her dependent children are not forthcoming from biological fathers in the context of
traditional nuclear families, then they can come equally well from
the mothers independent income or from the state. Unfortunately,
there is a good deal of evidence that indicates that this isnt so. A s
David Popenoe has documented at great length, fathers play an
extremely important role in the raising of both sons and daughters,
and they are much more likely to play that role well if they are the
biological father of the child in question.23 Hence it may be the
case that men and women, interacting in a game-theoretic manner
as rational maximizing agents, are developing new cooperative
social norms for family life and the raising of children. But while
these new norms of fatherlessness and male irresponsibility may
serve to maximize the interests of the mothers and fathers, it is by
no means clear that they are in the interests of children. A gametheoretic interaction might therefore lead to an overall result that
is far from socially optimal.
Hayek might argue that the process of spontaneous normgeneration is still going on and that postindustrial societies may
yet generate new norms that come closer to a social optimum given
contemporary economic conditions. A neo-Darwinian might assert
that if in fact existing norms were suboptimal for the life-chances
of future generations, then those parents exhibiting this behavior
would have fewer and weaker children, ultimately reproducing
themselves less well than other parents with different preferences.
Neither of these propositions is, needless to say, a comforting
thought from the standpoint of near-term social policy.
474
spontaneous generation or else through what was labeled exogenous generation -as the byproduct of religion, culture, or other
forms of shared historical experience. Social capital is created
daily in a host of situations, from two individuals who become
friends and come to share a norm of reciprocity, to a firm that socializes workers to a common corporate culture as described in
the previous lecture, to a political movement started by a group of
activists.
Social capital is in some respects a public good: its benefits are
often widely dispersed and therefore not readily appropriable by
the agent responsible for creating it. This suggests that there will
be limits to the degree to which market actors will have an incentive to supply social capital. T o the extent it is a public good, it
must be supplied either by a government agency or by a nongovernmental actor outside of the market economy like a church or a
voluntary association.
Public agencies have created social capital in a variety of ways.
One of the most basic functions of a state (to which any neoclassical economist would assent) is the protection of property rights
through a transparent rule of law. In situations where states fail
to perform this function, as historically in southern Italy or today
in Russia or Ukraine, private organizations enter the protection
business and private trust plummets.
In addition, public agencies can foster more specific kinds of
social capital. Military organizations, for example, deliberately
foster group solidarity through an intensive process of socialization, because unit cohesion is critical to military effectiveness.24
Agricultural extension officers were frequently responsible for
creating webs of community solidarity in the rural areas in which
they operated during the first part of the twentieth century. And
public school systems from the United States to France to Japan,
24 On the role of unit cohesion in World W a r II, see Martin Van Geveld,
Fighting Power: German and U.S. Army Performance, 19391945 (Westport, Conn.:
Greenwood Press, 1982).
[FUKUYAMA]
Social Capital
475
476
[FUKUYAMA]
Social Capital
477
This trend obviously reflects the fact that there has been a great
deal of gender discrimination in the past. Gender is very different
from race, however: while the biological evidence for serious and
consistent differences between human races is small to nonexistent,
the evidence for serious gender differences is legion. Men and
women are different in their underlying psychological makeup ;
they face different incentives with regard to family life and childrearing; and they pose different risks to society in terms of aggression and asocial behavior. Public policies that seek to be genderblind in the way that they are supposed to be color-blind are therefore bound to have unanticipated consequences.
One way to become more gender-specific is simply in the way
that we describe the problem. For example, while it may have
been the movement of women into the paid labor force that lies at
the heart of the Great Disruption, the real behavioral problems
have all been on the side of men. T o talk neutrally about family
breakdown ignores the fact that men and women are not equally
complicit in creating the social problem. Women, even working
women with high-powered careers, still tend to invest more of
their time in child-rearing than men. The real problem is men,
who today feel that they have been released from the obligation to
stay with their wives and particularly with the children they father.
There is no deficit of mothers and motherhood; there is, however,
a serious deficit of fathers and fatherhood. Policies therefore need
to be very much aimed at changing male behavior and incentives
through new cultural values or different forms of socialization. I
will illustrate this principle with regard to two areas of public
policy: welfare and worker retraining.
1 . W e l f a r e reform targeted at fathers rather than mothers.
In 1996 the U.S. Congress finally passed a major welfare reform
bill that ended the federal entitlement to unlimited welfare benefits and closed down the Depression-era Aid to Families with Dependent Children (A FD C) program. One of the critical justifica-
478
tions for this reform cited by its backers was the negative effect
of AFDC on family stability among the poor.
The most significant critique of the 1996 welfare reform bill
was made by James Q. Wilson in a speech to the Manhattan Institute in 1995.25 Along the lines of the analysis above, Wilson argued
that the real behavioral problem encouraged by the current welfare
system is not single mothers on welfare who dont want to work,
but rather the fathers of their children who take no responsibility
for their offspring once they are born. However the current welfare bill changes incentives for poor families, they will operate
entirely on the mothers side. There is an unstated hope that by
removing incentives for the women to become single parents, the
mens behavior will also change -presumably as a result of the
women denying sex to the men unless they return to the traditional
exchange of fertility for stability and resources. It is extremely
dubious that anything like this will actually occur. Moreover, from
the standpoint of young children, it is not clear that it is desirable
for the mothers to be working.
Far more important is to restore fatherhood, something that
requires a positive act of resocialization so that the men recover a
sense of responsibility for the children they produce. Doing this
requires, in turn, a solution to the problem that the fathers are unemployable and therefore have no resources to offer the mothers as
their part of the bargain. It is not clear to me that whatever little
money will be spent on job training and job creation as mothers
move from welfare to work would not be better spent on providing incentives to the fathers and making them employable.
2 . Worker training. The disappearance of high-wage low-skill
jobs in sectors like automobiles, steel, meatpacking, and the like is
a universal problem for all industrialized countries. This job loss
is what accounts for continuing high rates of unemployment in
Europe and for falling real working-class wages in the United
25 This speech was reprinted as Welfare Reform and Character Development,
City Journal 5, no. 1 (Winter 1995): 56-69.
[FUKUYAMA]
Social Capital
479
480
The American system of vocational education is very weak compared to that of central Europe; for the most part, low-skill
workers are simply left to fend for themselves in acquiring the
skills needed to remain employed.
Adding urgency to the worker retraining issue is the shifting
age composition of all OECD countries. Owing to steady drops
in fertility almost all OECD countries will be losing population at
a dramatic rate by the middle of the twenty-first century. Using
the low-growth fertility assumptions of the U.N. Population Offices estimates for global population, countries like Germany,
Italy, and Japan will be losing population at a rate of about 1 percent per year by the middle of the next century. Even more worrisome is the fact that the median age will drift steadily upward,
moving to fifty-eight years for Italy by 2050, fifty-three years for
Germany, and fifty-four years for Japan. Most economists are
aware of the huge social security liability problem this creates.
Given advances in the health of the elderly population, it would
seem to be inevitable that working careers are going to have to be
extended and retirement ages pushed back. If people routinely
retire at seventy-five or eighty rather than sixty-five, it would seem
all the more mandatory that training and education be seen as an
ongoing process rather than something that is accomplished before
the age of twenty-five. The likelihood that job skills are still going
to be valuable fifty or sixty years after they are acquired seems, to
say the least, implausible.
SOCIAL CAPITAL AND POSTINDUSTRIAL SOCIETY
I began the first lecture by noting that many criticsof the Enlightenment argued that a society founded solely on the principles
of reason, without help from religion, tradition, and culture,
would ultimately become self-undermining. Though they did not
use the term social capital, they were concerned with the issue
of whether a political system that limited the powers of the state
and depended heavily on institutions to regulate the pursuit of
[FUKUYAMA]
Social Capital
481
482
taxonomy of norm-generation suggests that apart from rational institutional sources, norms can be generated spontaneously through
the repeated interactions of individual agents or exogenously
through the introduction of a new set of moral norms. Hayek suggests that informal social norms are constantly in the process of
evolution and that over time free individuals, if left to their own
devices, will arrive at ones suitable to their circumstances. Stated
in this fashion, Hayeks assertion is unfalsifiable; any problematic
set of norms is necessarily in the process of evolving to something
different and potentially better. The question is how long this evolution will take, what mechanisms might intervene to speed the
process, and how much damage will be done in the meantime.
We can draw some comfort by looking at other historical
periods and noting that societies have been able to regenerate social capital in the past. Historical data on social trends are necessarily sketchy, but it is clear that the levels of social deviance evident in the developed world today are not historically unprecedented. The Industrial Revolution in England had, by the early
nineteenth century, produced a host of social pathologies including
high levels of murder and robbery, family breakdown, abandonment of children, alcoholism, and the like.27 A number of sources
suggest that deviance rates rose steadily through the middle of the
nineteenth century and thereafter began a long, slow decline.
The causes of that decline are complex. Religion clearly played
a role: the Wesleyan movement in the first half of the century, as
E. P. Thompson has shown, had a profound influence on the British working class and was able to arrest the spread of social decay.28 Victorian morality, which has come in for extraordinary
criticism in the late twentieth century, was in many respects understood as a deliberate effort to counteract the breakdown in norms
that had occurred earlier in the century.
27 For an overview, see Gertrude Himmelfarb, The De-Moralization of Society:
From Victorian Virtues to Modern Values (New York: Knopf, 1995).
28 E. P. Thompson, The Making of the English Working Class (London: Penguin, 1963).
[FUKUYAMA]
Social Capital
483
48 4
cades at the end of the nineteenth and the beginning of the twentieth centuries. American religious life, as is well known, has
never been on a one-way descent toward secularism, but rather has
experienced periodic upsurges from the Great Awakenings of the
eighteenth and nineteenth centuries to the Pentecostal revival in
the twentieth. All of this suggests that the Great Disruption will
call forth countervailing efforts to remoralize and restore the social
norms broken by economic change.
If the field of sociology is one long commentary on the transition from Gemeinschaft to Gesellschaft, then it is clear that this
transition has entered a new phase with the transition to postindustrial society in which work has been feminized and gender relations put on a new basis. The great classics of social theory
Karl Marx, Ferdinand Tnnies, Max Weber, Emile Durkheim,
and Georg Simmel-need to be written anew to take account of
what is easily as momentous a shift as the one that they experienced during their own lives.
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: