The Son Also Rises Page 3
Only if genetics is the main element in determining economic success, if nature trumps nurture, is there a built-in mechanism that explains the observed regression. That mechanism is the intermarriage of the children of rich and educated lineages with successful, upwardly mobile children of poor and uneducated lineages. Even though there is strong assortative mating—because this is based on the social phenotype created in part by luck—those of higher-than-average innate talent tend to mate with those of lesser ability and regress to the mean. Similarly, those of lower-than-average innate talent tend to marry unlucky offspring of higher average innate talent.
If nature does indeed dominate nurture, this has a number of implications. First, it means the world is a much fairer place than we intuit. Innate talent, not inherited privilege, is the main source of economic success. Second, it suggests that the large investment made by the upper classes in the care and raising of their children is of no avail in preventing long-run downward mobility: the wealthy Manhattan attorneys who hire coaches for their toddlers to ensure placement in elite kindergartens cannot prevent the eventual regression of their descendants to the mean. Third, government interventions to increase social mobility are unlikely to have much impact unless they affect the rate of intermarriage between levels of the social hierarchy and between ethnic groups. Fourth, emphasis on racial, ethnic, and religious differences allows persistent social stratification through the barriers they create to this intermarriage. In order for a society to increase social mobility over the long run, it must achieve the cultural homogeneity that maximizes intermarriage rates between social groups.
What is the significance of these results for parents socially ambitious for their children? The practical implication is that if you want to maximize your children’s chances, you need to pay attention not to the social phenotype of your marriage partner but instead to his or her status genotype. That genotype is indicated by the social group your potential partner belongs to, as well as the social phenotype of their siblings, parents, grandparents, cousins, and so on to the nth degree of relatedness. Once you have selected your mate, your work is largely done. You can safely neglect your offspring, confident that the innate talents you secured for them will shine through regardless. If, that is, the theory on the source of status persistence conjectured here is correct.
I want to emphasize that this book is not a jeremiad. Despite the low reported rates of social mobility, despite the importance of lineage in determining current outcomes, and despite our inability to significantly influence underlying rates of social mobility, this book takes cheer from the completeness of social mobility. Thus the title The Son Also Rises. For the evidence of the book is that social position is likely determined by innate inherited abilities. The social world is much fairer than many would expect. And the evidence is that in the end, the descendants of today’s rich and poor will achieve complete equality in their expected social position. This equality may require three hundred years to come about. Yet why, in the grand scheme of societies, is three hundred years for convergence any more significant an interval than thirty years?
But an important corollary to the finding that social outcomes are the product of a lineage lottery is that we should not create social structures that magnify the rewards of a high social position. The justification for the great inequalities we observe is often that reward is the required stimulus for achievement. But we see in the various settings studied in this book, as in figure 1.6, no correlation between inequality and underlying rates of social mobility. If social position is largely a product of the blind inheritance of talent, combined with a dose of pure chance, why would we want to multiply the rewards to the lottery winners? Nordic societies seem to offer a good model of how to minimize the disparities in life outcomes stemming from inherited social position without major economic costs.
It should also be emphasized that the concentration in the book on the patrilineal line of inheritance, which is only one of many lines of descent once we look across many generations, is driven purely by the fact that in the societies studied here, surnames until recently were overwhelmingly inherited from fathers. It does not reflect any belief that women are unimportant: it merely results from the fact that until the last few generations, women’s status largely reflected that of their husbands. But there is no indication that were we to measure status persistence rates through the matrilineal line, we would observe more mobility. In the modest number of cases where we observe, for example, the correlation between fathers and sons-in-law, it is just as high as between fathers and sons.8 It is notable, however, that the emancipation of women in recent generations has had no influence on social mobility rates. Emancipated women mate as assortatively as before and transmit their status to children as faithfully as in the patriarchal societies of the past.
1 An online search of books and articles containing the phrase social mobility yields 244,000 items.
2 Appendix 1 explains these concepts in more detail.
3 The most famous Pepys, Samuel, did not contribute himself to this distinguished lineage, as he has no known descendants.
4 Ironically, given Big Brother’s reputation, Sir Peter is also the descendant of Louis’s son Sir Joseph Bazalgette, the nineteenth-century designer of the London sewer system.
5 Given the power of the results shown in this book, it is surprising that the systematic use of surnames to trace social mobility has been so little used in the past. The only author to pursue this line of inquiry was Nathaniel Weyl, whose Geography of American Achievement (1989) uses surnames to measure the status of groups of different ethnic origin. Weyl was a racist and was seeking by these means to show the presumed permanent superiority of those of Jewish and northern European descent.
6 The term status genotype does not imply here that genes do in fact transmit status, just that the process looks similar in character to genetic transmission.
7 Strictly speaking, the process is first-order Markov.
8 See, for example, Olivetti and Paserman 2013.
PART ONE
SOCIAL MOBILITY BY TIME AND PLACE
TWO
Sweden
Mobility Achieved?
IN EXPLORING SOCIAL MOBILITY USING SURNAMES, we begin with Sweden for two reasons. First, by conventional measures, modern Sweden has social and economic mobility more rapid than that of either the United Kingdom or the United States. And Sweden is representative of a group of Nordic countries—Denmark, Finland, Iceland, Norway, and Sweden—believed to have achieved low inequality, widespread educational attainment, and fast social mobility. In recent years these societies have been cited as a reproach to the economic model of the United Kingdom and the United States. Both have greater inequality in outcomes and lower apparent rates of social mobility. These contrasts are evident in figures 1.3 and 1.4 above. The rapid measured social-mobility rates of Nordic countries imply that very little of their citizens’ current income and educational attainment can be predicted from parental income or education. The Nordic social world is made anew each generation. These societies seem to offer profound equality in life chances for the children of rich and poor, educated and uneducated.
A recent study of four generations of families in Malmö suggests that intergenerational earnings and education correlations in Sweden have been at the modern level for at least three to four generations. The initial generation in the Malmö study was born between 1865 and 1912.1
Such mobility would suggest that Swedish institutional arrangements—the support for public education, for example, and the progressive taxation of wealth—play a vital role in determining rates of social mobility. The implication, as discussed in chapter 1, is that the lower rates of social mobility observed in countries such as England and the United States represent a social failure. The life chances of the descendants of high- and low-status ancestors can be equalized at low social cost. Sweden is, after all, one of the richest economies in the world.
However, this ch
apter shows that the persistence rates reported for these countries, unless carefully interpreted, lead to a false interpretation of Nordic social reality. Estimated using surnames, the intergenerational correlations for measures of status such as occupation or education are much higher.2 And in the modern era they are as high as the rates of the eighteenth century. Whatever the short-run mobility shown by earnings or education, there is considerable persistence of status—measured through earnings, wealth, education, and occupation—over as many as ten generations in Sweden.
The finding that social mobility is much slower than conventional estimates comes from the study of the frequency of two historically elite sets of surnames among modern high-status groups compared to their frequency in the general population. If this ratio, the relative representation, is greater than one, the surname group still constitutes an elite. If it is less than one, it forms an underclass. The speed with which the relative representation approaches one for any group reveals the social mobility rate of the society.3 The first set of names comprises uncommon surnames associated with the Swedish nobility. The second is the surnames of the educated elite of the seventeenth and eighteenth centuries. Both sets of names are still overrepresented among modern Swedish elites—physicians, attorneys, university students, and members of the Royal Academies—showing that true social mobility rates in Sweden have been low. By looking at the rate at which their overrepresentation in these groups has declined over the last two or three generations, we can measure mobility rates up to 2012. The results are summarized in table 2.1. They show that current social mobility in Sweden is very slow—no faster, as we shall see, than comparable estimates in the United Kingdom or the United States, and no faster than social mobility in eighteenth-century Sweden under monarchic rule.
TABLE 2.1. Estimates of status persistence rates by occupations, Sweden, 1700–2012
Swedish Surnames
Surprisingly, despite Sweden’s reputation as a model social democracy, a class of nobles is very much alive and functioning. The country has a formal guild of noble families, the Riddarhuset (House of Nobility) (figure 2.1). Though noble families have existed since medieval times, the modern Riddarhuset was created in 1626. From 1668 to 1865, it functioned as one of the four governing estates of the kingdom, analogous to the House of Lords in England. Since 2003 the Riddarhuset has been a private institution that maintains the records of the Swedish noble families, and lobbies on their behalf. In spite of Sweden’s advances in gender equality, only men vote in the Riddarhuset, and only sons transmit titles to their offspring.
FIGURE 2.1. The Riddarhuset, headquarters of the Swedish nobility, in downtown Stockholm.
FIGURE 2.2. The history of ennoblement in Sweden: families enrolled in the Riddarhuset.
The families enrolled in the Riddarhuset occupy three hierarchical ranks: in descending order, these are counts, barons, and “untitled” nobility. More than two thousand families have been enrolled, though only about seven hundred have living representatives.4 The timing of these ennoblements is summarized in figure 2.2. As the figure shows, almost all extant noble families were enrolled before 1815: indeed, a large fraction of all noble families was created in the period 1658–1721, when Sweden’s territories encompassed Finland, Estonia, and some north German states. From 1680 onward the nobility gradually lost its privileges, starting with the reclamation by the crown in 1680 of much of the land granted to nobles in previous years. By 1866 the nobles had no economically significant privileges.
When Swedish families were enrolled in the Riddarhuset, they typically adopted a new surname embodying status elements such as gyllen (gold), silfver (silver), adler (eagle), leijon (lion), stjerna (star), creutz (cross), and ehren (honor). Thus we get names like Leijonhufvud, Gyllenstjerna, Ehrensvärd, and Adlercreutz. Rosencrantz and Guildenstern, the two unfortunate Danish nobles in Shakespeare’s Hamlet (written around 1600) have forms of two common noble Danish and Swedish surnames. One-tenth of the aristocrats participating in the Danish coronation of 1596 supposedly bore one or the other of these names.5
Many Swedish noble surnames, however, are German in origin, reflecting the number of German military commanders rewarded with ennoblement for their service to the Swedish crown in the seventeenth century: hence names such as von Buddenbrock and von Köningsmarck. Scottish, English, French, and other foreign surnames also appear: Douglas, Maclean, Bennet, de la Gardie. Some noble surnames are, however, quite common and held by many people probably not descended from these noble families, such as Björnberg or Hamilton. The analysis below is therefore restricted to noble surnames now held by four hundred or fewer people, of whom a large fraction of current holders likely descended from an original ennobled family.
One privilege that the nobility obtained through the Names Adoption Act of 1901 was a ban on anyone else’s adopting their surnames.6 Thus, apart from foreign imports and name changes before 1901, the surnames of the enrolled nobles in the Riddarhuset uniquely identify the lineage of these noble families. Such surnames constitute a small Swedish elite. Only sixteen thousand individuals currently hold the surnames of count and baronial families (defined according to the criterion specified above). A further forty thousand people hold rarer surnames associated with the untitled nobility.
Signs that these surnames are mostly derived by descent from those ennobled many years ago come from the stock of these names as a share of the population. Figure 2.3 shows the incidence of a sample of aristocratic surnames among male deaths from 1901 to 2009, and among male births from 1810 to 1989. From 1810 to the present, these noble surnames have accounted for the same share of surnames.
The second class of surnames of interest are latinized surnames. In the preindustrial era, when most Swedes had impermanent patronyms, clerics, academics, and some merchants adopted surnames sometimes derived from Swedish names but typically ending in -ius or -eus, which became characteristic of an educated class. These include the names of a number of famous Swedish scientists of the seventeenth and eighteenth centuries: Carolus Linnaeus (1707–78), Anders Celsius (1701–44), Jöns Jakob Berzelius (1779–1848), and Olaus Rudbeckius (1630–1702). Typical examples of these surnames now are Aquilonius, Arrhenius, Berzelius, Boethius, and Cnattingius.
FIGURE 2.3. Percentage of aristocratic surnames at death for men born 1810–2009.
Only a small fraction of the modern population bears such latinized surnames. Of those dying between 2000 and 2009, for example, only 0.5 percent bore a surname ending in either -ius or -eus. However, in the late nineteenth and early twentieth century, significant numbers of people adopted newly created latinized surnames. Figure 2.4 shows that of men born 1810–1990 the proportion with latinized surnames doubled over time.
To avoid including newly adopted surnames, the latinized surnames employed here are restricted to those that existed before 1800. One quick way to identify such long-established surnames is to consider only latinized surnames held by forty or more people in 2011.7 They are overwhelmingly held by those who inherited them from their parents as opposed to adopting them, perhaps because of the restrictions imposed on name changes in the Name Regulation Law (släktnamnsförordningen) of 1901 and the Naming Law of 1982, which now requires the Swedish Tax Agency to approve all surname changes. These surnames constitute 0.2 percent of the current population.8 The share of these older latinized surnames in the population is close to stable between 1810 and 1989.
FIGURE 2.4. Percentage of latinized surnames at death for men born 1810–1979.
The most common Swedish surnames are patronyms, surnames formed from the first name of the father and ending in -son. These were the predominant type of surname in preindustrial Sweden. A sample of seventeenth-century parish marriage records, for example, shows that 93 percent of those who married bore such patronyms.9 In early Sweden, such patronyms changed from generation to generation. In the eighteenth and nineteenth centuries, their use declined, and families adopted more permanent surnames. The 1901 Name Re
gulation Law called for each family to have a surname that would remain unchanged across generations.
The decline of patronyms has continued to this day. Figure 2.5 shows estimates by twenty-year periods of the number of Swedish men born with and dying with a patronym as surname. After the year 2000, only 40 percent of males who died in Sweden bore a patronym. But for those dying before age 10, the share was even lower, approximately one-quarter.10
FIGURE 2.5. Percentage of men with surnames ending in -son, by date of birth and death.
We can observe the sources of this decline if we consider the percentage of patronyms occurring among all men born in 1950–51 dying by 2009. Of this cohort, half of those dying before age 10 but only a third of those dying between ages 50 and 59 had a patronym. By implication, nearly one-third of men born with patronyms changed their surnames, with most of the changes occurring by age 30. Thus, although patronyms in Sweden are associated with low social status, we have to be careful when using them to measure social mobility, since such patronyms are only selectively retained by the modern population.