Your Surname Is Your Score
Six
hundred years of data from Florence prove that meritocracy is a lie. The glass
floor is real, and your
great-great-great-great-great-great-great-great-great-great-great-great-great-great-great-great-great-great-great-great-grandfather’s
tax return still predicts your paycheck.
The
Number That Should Haunt You
Let us
begin with a number so small that most economists would dismiss it as
noise: 0.04.
That
is the long-run earnings elasticity between the richest families of Florence in
1427 and their descendants alive today in 2011. It sounds like nothing. A
rounding error. A statistical ghost that disappears the moment you adjust for
measurement.
But
here is the thing. Standard economic models predict that number should be zero.
Not 0.04. Zero. The reason is pure math. Over twenty generations—roughly six
hundred years—every financial advantage should have been diluted into oblivion.
Marriages scatter wealth. Children split inheritances. Wars, plagues,
expropriations, and simple bad luck wipe out the careless and the unlucky
alike. The math is brutal and unforgiving. If each generation has just two
surviving children—a conservative estimate for pre-modern Europe—a single
couple’s wealth is divided among more than one million descendants by the
twentieth generation. A fortune of 80,000 florins, the recorded wealth of
Gabriello di Bartolomeo Panciatichi, the fourth-richest Florentine householder
in the 1427 Catasto, becomes less than one-tenth of a florin per person.
That is not wealth. That is dust.
And
yet.
The
same surnames that sat atop Florence’s tax rolls in 1427 are still sitting at
the top today. The Strozzi. The Guicciardini. The Rucellai. The Panciatichi.
The Bardi. A family that was at the 90th percentile of the earnings
distribution in 1427 is still, on average, 5 percent higher in
earnings than the mean Florentine today. For wealth, the persistence is even
stronger: the gap remains above 10 percent. That is not mobility.
That is slow-motion feudalism with spreadsheets and regression analysis.
The Catasto: A War Ledger That Became a Time Capsule
The Catasto of 1427 was not a kindly census
undertaken for the benefit of historians. Florence created it because the city
was nearly bankrupt. The cause was a debilitating war with Milan, which had
drained the republican treasury and exposed the old tax system as a corrupt
joke. The existing system relied on forced loans called prestanze,
which fell hardest on the poor while the wealthy evaded through political
connections. Something had to change.
What Florence did next was radical for its time. The
government surveyed every head of household in the city and
its surrounding countryside—9,780 families in the urban core alone—and
recorded, in painstaking detail, their assets, debts, business interests, real
estate holdings, occupational status, and even the number of dependents, or
"mouths to feed." The resulting ledger was a weapon of fiscal
survival. It was also, entirely by accident, a six-hundred-year time capsule of
social structure.
The numbers from 1427 are obscene by any standard. The
top 137 households, just 1.4 percent of the population, held 29.8
percent of all private wealth. These were the wool and silk guild
members—the Arti Maggiori—along with bankers, international
merchants, lawyers, and notaries. They lived in the palazzi that
still line the Via Tornabuoni and the Piazza della Signoria. They married their
daughters to each other’s sons. They sat on the city’s governing councils.
At the bottom, 1,431 households—14.6 percent of
the population—held no net assets at all. They were unskilled laborers,
specifically recorded in the Catasto as working in the
combing, carding, and sewing of raw wool. These were the ciompi,
the wool workers who had staged a bloody revolt in 1378, only to be crushed
back into destitution. Their surnames were recorded alongside their poverty.
When the economists Guglielmo Barone and Sauro Mocetti of
the Bank of Italy digitized the Catasto and lined it up
against Florence’s 2011 tax records, they found roughly 900 of the
original 1,210 surnames still present in the city. The rich names
stayed rich. The poor names stayed poor. The gap between them after six
centuries of wars, plagues, revolutions, and economic transformations: 12
percent.
Barone and Mocetti gave this persistence a name that has
stuck. They called it the glass floor. The children of the rich do
not fall through it. Not in three generations. Not in twenty. The floor is
invisible, but it is there, and it holds.
Gregory Clark’s Uncomfortable Constant: The 0.75 Rule
Florence is not a special case. It is not an Italian anomaly
explained by pasta, popes, or patronage. The same phenomenon has been
documented, with almost identical statistical magnitude, across England,
Sweden, China, and India. The economist Gregory Clark has spent two decades
tracking rare surnames through historical records—tax rolls, probate records,
university admissions registers, medical licensing lists, and even wills. His
method is simple, even brutal: find a distinctive surname that was wealthy in
the past, and see where those same surnames appear today.
The result is a universal constant of
social mobility. Or rather, the lack of it.
Clark’s headline number is 0.75 per generation.
That is the intergenerational elasticity of status—the fraction of a parent’s
relative advantage that is passed to the child. An elasticity of 0.75 means
that if your parents were at the 90th percentile of the income distribution,
you are expected to be at approximately the 75th percentile. That is not zero.
That is not even close to zero. It is a staggeringly high degree of
persistence.
At 0.75 per generation, an elite family does not return to
the population average in three generations. It takes ten to fifteen
generations. Three hundred to four hundred and fifty years. The practical
implication is brutal: the grandchildren of the rich are still rich. The
great-grandchildren of the poor are still poor. And this is not because of any
obvious failure of policy or education. It is because the mechanisms of
persistence are baked into the structure of family, marriage, inheritance, and
information.
Clark’s evidence across countries is remarkably consistent.
In England, surnames that held wealth in the 1800s—such as those found in
probate records or among Oxford and Cambridge matriculants—still appear
in Who’s Who and the medical register at rates far above their
population share. In Sweden, despite a century of social democratic
redistribution, high taxation, and universal welfare, the same pattern holds.
The surnames that were rich before the welfare state remain rich after it. In
India, the traditional landowning castes remain economically dominant even
after formal legal equality and affirmative action policies. In China, the
surnames associated with the Qing-era gentry have re-emerged as the entrepreneurial
elite after the post-Mao reforms.
The glass floor is not a Florentine quirk. It is a global
structural feature of stable societies. And it operates with the indifference
of a physical law.
The Four Pillars: Why Splitting Fails
Why does the math of splitting—the exponential
multiplication of heirs—fail to destroy dynasties? Because four countervailing
forces overwhelm division. These are not conspiracies. They are not secret
societies meeting in Florentine palazzi to plot surname
preservation. They are rational behaviors, aggregated across centuries, that
produce an emergent equilibrium of persistence.
First pillar: Assortative mating.
Rich people marry rich people. This is not merely a matter
of taste. It is a structural necessity in societies where social circles are
tight, marriage markets are stratified, and families guard their status through
strategic alliances. In Florence, the elite married within a tight pool of
other elite families for generation after generation. The Strozzi married the
Medici. The Medici married the Salviati. The Salviati married the Rucellai. The
effect is multiplicative, not additive. Two wealthy heirs marrying each other
doubles the starting capital per child compared to a random marriage to a
person of average wealth.
The statistical consequence is straightforward. In the
absence of assortative mating, the intergenerational elasticity of wealth might
be 0.4. With a spousal wealth correlation of 0.8—conservative for historical
elites—the effective elasticity approaches 0.7 before any other mechanisms are
added. Assortative mating resets the mean upward every generation. It is the
original wealth multiplier.
Second pillar: Productive assets held whole.
A textile mill, a bank, a portfolio of urban real estate, a
fleet of trading ships—none of these has to be divided at death. The law of
primogeniture, still common in much of European history, gave the entire estate
to the eldest son. Younger sons were sent into the church, the military, or
colonial administration, but they did not take a saw to the family mill. In
more recent centuries, trusts, foundations, and corporate structures have
performed the same function. The principal remains intact. Only the income is
split.
Consider the compounding math. A family holding a mill worth
1 million florins in 1427, earning a 5 percent annual real return, would see
that mill grow to about 2.65 million florins after twenty years. If the mill is
never divided—if it is held in a perpetual trust or passed whole to a single
heir—the capital base compounds indefinitely. The surname stays rich because
the capital stays whole. The splitting math, so devastating in theory, simply
does not apply to the core productive asset.
Third pillar: Diversification as a shock absorber.
A single baker with one oven goes bankrupt when the oven
breaks or the neighborhood changes. A diversified family with land, trade
credit, real estate, bonds, and ecclesiastical patronage survives shocks that
would destroy a concentrated portfolio. Over six hundred years, Florence
experienced the Black Death (1348), the Ciompi revolt (1378), the Medici exile
(1494), the siege of 1529-1530, the rise and fall of the Grand Duchy,
Napoleonic occupation, Italian unification, two world wars, and the flood of 1966.
A non-diversified family would have been wiped out in the first century.
The elite families survived because their assets were
uncorrelated. When the wool trade collapsed due to English competition, their
landholdings held value. When plague reduced the labor supply, their urban real
estate became more valuable as survivors consolidated. When a bank failed,
their ecclesiastical bonds were still honored. The mathematics of
diversification is unforgiving in the other direction: the probability of a
diversified portfolio losing everything in any given generation is the product of
each asset’s individual failure probability. With five uncorrelated assets,
each with a 20 percent chance of total loss in a generation—a very high
risk—the family’s ruin probability is 0.2 raised to the fifth power, or
0.00032. Over twenty generations, the survival probability is 99.4 percent.
That is not luck. That is mathematics.
Fourth pillar: Information and early access.
Wealth buys better information. In 1427, Florentine elite
families sat on the Catasto assessment committees. They knew
which trade routes the Signoria was about to subsidize. They
knew which neighbors were overleveraged and would soon default. They knew which
properties would be rezoned for commercial use or targeted for public works. In
modern terms, this means board seats, private equity pre-IPO rounds, venture
capital funds that are closed to the public, real estate purchases ahead of
public infrastructure announcements, and admission to elite universities
through legacy preferences and development offices.
The edge is small. Perhaps 1 to 2 percent extra annualized
return. Perhaps a 10 percent lower entry price into a new asset class. But over
twenty generations, 1 percent per generation compounds into a 22 percent higher
terminal wealth. More importantly, early entry means buying at a fraction of
the later public price. The middle class discovers opportunities when they are
already expensive—when the IPO has happened, when the neighborhood has
gentrified, when the degree has become a credential arms race. The wealthy
discover them at dinner parties. That is not insider trading. That is just
having the right address.
The Escape Valves That Failed
Conventional wisdom holds that three forces break dynasties:
revolutions, geographic mobility, and mass education. The evidence suggests
otherwise.
Revolutions. China in 1949, Russia in 1917,
Vietnam in 1975, Cuba in 1959, Cambodia in 1975—each revolution violently
expropriated the old elite. They killed or exiled the surnames that had sat
atop the tax rolls. They redistributed land and capital. They promised a new
world of equality and opportunity. And within two generations, new elites
formed. China’s current billionaires are disproportionately the children of
Communist Party cadres who got first access to state assets in the
privatization drives of the 1990s. Russia’s oligarchs are ex-Komsomol officials
and former Soviet ministry bureaucrats. The surnames change. The elasticity
does not. Revolutions reassign who sits above the glass floor. They do not
remove the floor itself.
Geographic mobility. Historically, moving from a
stagnant village to a booming city was an escape route. Manchester in 1820,
Detroit in 1910, Shenzhen in 1990—these growth centers were underpopulated
relative to opportunities. Land was cheap. Labor was scarce. Incumbent networks
were weak. A talented migrant with nothing but ambition could outperform the
locals. Today, every El Dorado is full. Housing costs in San Francisco, London,
Shanghai, and Sydney require a down payment that itself requires family wealth.
The poor cannot afford to live where the growth is. Tech hubs and finance
centers hire through internships and alumni referrals—networks that advantage
the already advantaged. There is no empty frontier left. The geography of
opportunity has been claimed, zoned, and priced.
Education. Mass education was supposed to be the
great equalizer. Instead, it has become a credentialing bottleneck that filters
by family background. Top universities admit children from the top income
quintile at rates five to ten times their population share. Unpaid internships
in elite law firms, investment banks, and media companies require a parental
safety net—someone to pay the rent while the graduate works for free. Medical
school tuition in many countries favors families with savings. The brilliant
middle-class child cannot afford the gap year, the networking trip to London,
the summer of unpaid labor at a magazine, or the low-paid fellowship at a think
tank. These are the signals that gatekeepers use to sort candidates. The
credentialing system does not lift the poor. It filters them out.
The Biological and Psychological Depths
The glass floor is not purely economic. It is partially
biological. Assortative mating means that cognitive and non-cognitive traits
associated with high earnings—conscientiousness, numeracy, health, even
height—cluster within dynasties. Polygenic scores for educational attainment
correlate with parental income. The rich are not necessarily smarter as
individuals, but they have stacked the genetic deck across generations. Clark
found that elite English surnames from 1800 still had above-average life expectancy
and educational attainment in 2012, even after controlling for current income
and wealth. Something non-economic is being inherited. Something that survives
taxation, redistribution, and divorce.
The psychological dimension is equally powerful and less
discussed. Poor children internalize their position. They under-match in
college applications: they do not apply to elite schools even when their test
scores and grades qualify them. They avoid risk because they cannot afford to
fail. A startup, a freelance career, a graduate degree in the humanities—these
are luxuries for people with safety nets. Poor children have shorter time
horizons because financial insecurity prioritizes immediate income over long-term
investment. Rich children have a psychological safety net. They can take the
low-paid fellowship, the startup gamble, the artistic detour, the unpaid
internship. That optionality compounds into higher lifetime earnings not
because of superior talent, but because of superior patience and superior
insurance against failure.
The Exceptions That Prove the Rule
Only catastrophic disruptions have temporarily broken the
glass floor. The Black Death of 1348 created labor shortages that raised real
wages for peasants and killed a disproportionate share of the elite—the wealthy
were more densely clustered in cities, where plague spread faster. The
disruption lasted about 150 years, until the old families reasserted themselves
through land purchases and strategic marriages. The World Wars and
hyperinflations of the early twentieth century destroyed physical capital and expropriated
enemy elites across much of Europe. The disruption lasted about forty years.
The Japanese asset bubble burst of 1990 reduced land values by 80 percent and
temporarily displaced old zaibatsu families. The disruption
lasted about twenty years.
In every case, the original elite reasserted itself within
two to three generations unless physically eliminated, as in China or Russia.
Japan’s Mitsui and Sumitomo families are still atop corporate Japan. England’s
Norman-descended aristocracy—the Grosvenors, the Howards, the Percys—still owns
a third of the land. The glass floor is not fragile. It is elastic. It bends
under catastrophe and snaps back when the catastrophe ends.
The Uncomfortable Bottom Line
The glass floor is not a conspiracy. No secret society meets
in a Florentine palazzo to preserve surnames across centuries.
The persistence is emergent—an equilibrium arising from rational behaviors that
make sense for a single family in a single generation. Marry within your class.
Preserve capital in trusts. Diversify across uncorrelated assets. Leverage
information asymmetry. Each decision is individually rational. Aggregated
across twenty generations, those decisions produce a world where your surname
predicts your score.
This is uncomfortable because it contradicts the liberal
faith in meritocracy and the Marxist faith in revolution. The glass floor does
not care about effort or ideology. It cares about structure. In stable
societies—without plague, world war, or revolution—the surnames that were rich
six hundred years ago will be rich six hundred years from now. Not because of
superior talent. Because of superior mathematics.
The only honest response is not outrage, but attention. The
first step toward any remedy is acknowledging that the game is rigged, that the
rigging is self-replicating, and that the usual policy solutions—education
spending, anti-discrimination laws, progressive taxation, minimum wage
increases—have barely dented the 0.75 constant. That does not mean they are
worthless. It means they are insufficient. Without structural interventions
that directly target the four pillars of persistence—breaking assortative mating
through integration, taxing undivided capital, insuring against diversification
failure, and democratizing information—the glass floor will outlive every
reformer.
And that is not pessimism. That is just reading the tax
records.
References
Barone, G., & Mocetti, S. (2021). “Intergenerational
mobility in the very long run: Florence 1427–2011.” Bank of Italy Working Paper
No. 1360.
Clark, G. (2014). The Son Also Rises: Surnames and
the History of Social Mobility. Princeton University Press.
Herlihy, D., & Klapisch-Zuber, C. (1978). Tuscans
and Their Families: A Study of the Florentine Catasto of 1427. Yale
University Press.
Chetty, R., Hendren, N., Kline, P., & Saez, E. (2014).
“Where is the land of opportunity? The geography of intergenerational mobility
in the United States.” Quarterly Journal of Economics, 129(4),
1553–1623.
Comments
Post a Comment