The shape of the distribution is the variable. Everything else is a policy.
I. FOUR SHAPES, FOUR LOGICS
The class structure is usually described by its height. The variable that matters is its shape.
The pyramid. A mass base, a thin middle, a narrow top. The gap between base and top usually is unbridgeable within a lifetime. The only available distinction is categorical, because ordinal distinction requires near rungs and there are none. This is the shape that usually produces caste, and it is the shape the first essays described: the category does the work because the gradient cannot. Exceptions confirm the rule—a pyramid with unusually many avenues of ascent can soften, and a pear with a captured gate can harden—but the shape sets the default.
The hourglass. A fat bottom, a fat top, a hollow middle. The middle is squeezed, precarious, and anxious. Its status is revocable, so it must be continuously displayed and continuously defended. This is the shape that produces boundary-work, because the category is the anxious middle’s capital. It is also the shape that produces populism, credit-financed consumption, and the politics of the “squeezed middle.”
The diamond. A large middle, thin bottom, thin top. The standard idealization. It produces stability, and it produces closure: the middle is large enough to defend its own entrance, and the entrance becomes a gate. A gate is a filter, and a filter becomes a category.
The pear. A broad, comfortable base; a bulge just above it; a long, gradual, many-runged slope to a thin top. Adjacent rungs are close. The number of rungs is large. The slope is long enough that ascent is a career rather than an event.
The pear is the shape in which class usually stays class. The others convert class into caste, into anxiety, or into closure.
II. THE MACROECONOMIC TAKE
The pear maximizes four things at once.
Consumption stability. A comfortable base with discretionary income means aggregate demand does not depend on the top’s marginal propensity to save or on the bottom’s access to credit. In the hourglass, demand is financed by the top’s asset prices and the bottom’s borrowing, which is Rajan’s Fault Lines (2010) in one sentence: the political system substituted credit for income, and the substitution was the crisis.
Firm formation. The bulge supplies both the founders and the customers. The upper working class is the classic small-business stratum—the shop, the workshop, the restaurant, the haulage firm. A pear has a large stratum of people with a little capital, a local market, and a reason to take a risk. An hourglass has a hollow middle and therefore a thin market for the kind of firm the middle would found.
Labour elasticity. A base that is comfortable enough not to grasp can retrain, relocate, and change sector. Precariousness does not produce mobility; it produces immobility, because the cost of a failed move is existential.
Contestable incumbency. Competition at the top is what forces incumbents to invest. A static top is a rentier top. The pear’s long slope keeps the top contestable without making the bottom desperate, which is the only combination that produces innovation without producing panic.
III. THE FISCAL TAKE
The pear maximizes the tax base without maximizing the tax rate.
The bulge is large, comfortable, and not dependent on transfers. It contributes more than it receives. The base is not so poor that it consumes more than it contributes. The top is thin enough that progressive taxation is not existentially threatening to capital, because capital has nowhere better to go that offers the same market.
This yields a broad, shallow, stable fiscal structure. The hourglass yields a narrow, deep, unstable one: the tax base is two thin strata, the working poor who contribute little and the rich who can leave, so the state faces either high rates on few people or permanent deficits. The diamond yields a fiscally comfortable but closed structure: the middle can pay for its own status, and it will, and it will defend the gate that protects the payment.
The pear has a further fiscal property that is usually missed. It minimizes the cost of status maintenance. In an hourglass, the state ends up subsidizing the middle’s position—housing subsidies, credential subsidies, mortgage relief, the whole apparatus of keeping a precarious stratum in place. In the pear, the middle is not precarious, so the state is not paying for positional goods. The saving is large and it is invisible, because it is a saving on a subsidy that was never labelled a subsidy.
IV. THE STRATIFICATION TAKE
A category becomes a caste when two conditions hold: the gap is unbridgeable, and the gate is single.
The pear denies both. Adjacent rungs are close, so no gap is unbridgeable. There are many rungs, so no single rung is a gate. And there are many avenues of ascent—enterprise, profession, craft, sport, art, migration, invention, marriage—so no single avenue can be captured by the incumbents.
This is Tilly’s Durable Inequality (1998) run in reverse. Opportunity hoarding requires a closure mechanism. Many avenues mean no closure mechanism can be installed, because the hoarders would have to capture every door. Constant recycling means no dynasty can form, because a dynasty requires a stable transmission channel, and recycling destroys transmission channels.
The cost is real, and it should be stated. Recycling means people fall. A pear that recycles is a pear in which the middle is continuously disrupted, and disruption is unpleasant for the disrupted. The pear’s legitimacy problem is that it must persuade the middle that being recycled is normal. The diamond avoids this problem by not recycling, which is why the diamond is politically popular and structurally closed.
The pear also requires a band, not a point. The base must be comfortable enough not to grasp and far enough that the grasp is a real step. Too comfortable and mobility falls for lack of push; too precarious and boundary-work returns. The pear is a range, and the range is narrower than its advocates admit.
V. COMFORT AT THE BASE, COMPETITION AT THE TOP
The design principle is asymmetric on purpose.
At the base, competition is for security. It should be easy. A person should not have to fight for a home, a school, a clinic, a pension. Security competition produces anxiety, and anxiety produces boundary-work, because the category is the cheapest available status claim for someone whose position is precarious.
At the top, competition is for position. It should be hard. A person should have to earn the next rung, and the rung should be contestable. Position competition produces effort, and effort produces the thing the top is supposed to produce.
Reverse the two and you get the hourglass. Easy at the top—inherited position, captured gates, rentier incumbency—and hard at the base, where security is precarious. The hourglass is the pear inverted, and it produces exactly the politics the pear is designed to prevent.
The mechanism connecting shape to the earlier essays is this: boundary-work is the status claim of the precarious near-middle. The anxious middle defends the category because the category is its capital. If the near-middle is not precarious, boundary-work loses its function. The category does not need to be refuted. It needs to be made unnecessary.
VI. WHY THE UPPER WORKING CLASS DOESN’T KEEP UP
Positional competition is a function of reference-group proximity and status revocability.
The upper working class compares locally and longitudinally: to its own street, to its own parents, to its own past. The reference group is stable, the comparison is about security, and the status is local and durable. It does not need to be displayed, because it cannot be revoked by a stranger who does not know the street.
The middle and upper middle compare upward and laterally across a wide field, because their position is defined by credentials, and credentials are comparative by nature. A credential is worth exactly what the next credential is not. The status is achieved and revocable, so it must be continuously re-earned and continuously displayed.
This is Veblen’s Theory of the Leisure Class (1899) with a mechanism attached. Conspicuous consumption is not a universal human trait. It is a tax paid on revocable status. If your status can be revoked, you must display it. If it cannot, you do not.
Bourdieu’s Distinction (1984) supplies the register. The upper working class has necessary taste: the good that is good because it works. The middle has luxury taste: the good that is good because it distinguishes. The distinction is not vanity. It is the grammar of a status that has to be proved.
VII. THE NAIROBI POST
The example is sharp because it isolates the register.
A mzungu—a white man holding a Western European, British, or North American passport—upper working class, running a factory or a shop or a restaurant or a fashion house or a solar car park in Nairobi. The post he makes is: come see my shop. It is not: I bought a home for my family in Nairobi.
The middle posts the home. The home is the achievement, and the post is a benevolence claim—look how I provide, look how I love—which is a status move that uses the family as an instrument.
And the middle is whose wife most reliably is a mzungu herself. The middle-class mzungu in Nairobi is married to a Western European, British, or North American woman, and the marriage is a positional fact before it is an intimate one. She is legible. She is a consumption bundle. She signals arrival in the same way the house does, and for the same reason: both are comparable, both are rankable, both are types.
The upper working class mzungu is married to a Kenyan woman, and he does not post her the same way. Not because the marriage matters less, but because it is not a claim in the same register. The local wife is not legible as a position, so she cannot be deployed as one, and the family stays where it belongs—in the register of the private, which is the register of the person rather than the type.
And the honest remark must be made, because the difference in the posting is not the whole story. If he does post her, he has to give her agency. The post about the couple starts with we, not I—we opened the shop, we moved here, we built this—because there is no way to place her in the frame as a possession without turning her into the very thing the middle turns his wife into. The I is the grammar of the benevolence claim: I bought, I provide, I love, instead of I love her, I hope it’s still mutual. The we is the grammar of the partnership, and a partnership is a relation between persons rather than a relation between a person and a position.
Which is why the upper working class mzungu who posts his marriage posts it as a we, and why the middle mzungu cannot. The middle’s post requires the I, because the I is the claim. The we would dissolve the claim, since a partnership of persons has no rank to display.
The crass case is the middle-class mzungu who posts the identical post with a Black local wife and a mixed family. It feels crass because the post cannot work as written. I bought a home for my family is a positional claim, and a position requires a register in which it can be read. The register is the wazungu. A mzungu wife is legible in that register as a position; a Kenyan wife is not. So the middle-class mzungu who posts the home with a local wife is making a claim that depends on her being a position while denying that she is one—and denying it to a host country national, in her own country. He uses her as the instrument of the benevolence claim and strips her of the agency the claim requires. The post reads as benevolence and functions as expropriation.
Hierarchy is only legible among the wazungu. She is not among them. So the post can only place her as the object of the sentence rather than its author. That is the crassness—not the mixed family, which is ordinary, but the grammar, which is not.
Three things follow, and the third is the sharpest.
Consumption categorizes; production individuates. A consumption bundle is a type: legible as a position, legible precisely because it is comparable. A functioning factory is not a type. It is a specific thing with specific problems that a specific person solved. The same distinction the earlier essay made about the sensual protagonist: the achievement attaches to a person rather than a category, which is why it is worth more.
The upper working class mzungu is not seeking individuation through the category, and not through consumption either. He seeks it through production, the least positional and most individuating register available. The enterprise is the counter-example to the type, and counter-examples are what the type cannot answer.
The middle’s display extends to the spouse, and the grammar betrays it. The middle posts the home and the wife because both are positions, and a position must be displayed to be held. The wife is individuated as a person and categorical as a sign, and the middle needs the sign—which is why the post begins with I. The upper working class marriage is not a sign, so it does not need to be displayed the same way, and when it is displayed it is displayed as a we, because there is no position to claim and therefore no I to claim it with. The family remains secondary in the register of status because it was never in that register to begin with.
VIII. LIBERAL VALUES AS THE SOFTWARE OF THE PEAR
Liberal values are the software of an ordinal society. They assume mobility, and mobility assumes the category is an attribute rather than a cause.
Categorical racism is the software of a pyramid with an unbridgeable gap. It explains why the gap is permanent, and it makes the permanence feel like a fact about persons rather than a fact about the distribution.
The pear makes liberal values true enough to be believed, which is why it usually produces them rather than requiring them. You cannot preach a person out of categorical thinking in an hourglass. You can build a pear, and the category loses its function.
This is the strongest version of the argument, and it connects directly to the first essay’s mechanism. Contact individuates. Contact requires proximity plus mixing. The pear supplies both: the bulge is dense, the rungs are close, the avenues are many, and the schools and streets are shared. The hourglass supplies neither: the middle is hollow, the strata are separated, and the only contact is service contact, which is categorical by construction.
Alesina and Glaeser’s Fighting Poverty in the US and Europe (2004) and the public-goods literature on ethnic heterogeneity (Miguel & Gugerty, 2005; Habyarimana et al., 2007) converge on the same finding from different directions: heterogeneity suppresses redistribution. The finding is usually read as a fact about solidarity. It is better read as a fact about the cost of investigation. Where the classification is cheap and the encounter is thin, the classification substitutes for the person, and a polity composed of classifications will not tax itself on behalf of a category it has not individuated.
The pear does not abolish the classification. It makes the classification expensive to use. The bulge is dense, the rungs are close, the avenues are many, and the schools and streets are shared. The person in front of you is a neighbour, a colleague, a customer, and a class peer, and the classification has nothing to add to what you already know. You have to override the specifics to use it, and overriding is work. In the hourglass the classification is free; in the pear it is a cost. That is the whole difference, and it is enough.
The hourglass supplies neither proximity nor mixing. Its middle is hollow, its strata are separated, and its only contact is service contact, which is categorical by construction—the person is a server, a client, a tenant, a claimant, and the transaction is exhausted by the role.
A society that meets its others only in moderated spaces will classify them and know nothing about them, and then wonder why the racial category will not die.
IX. WHAT THE PEAR COSTS
The pear is not free, and the costs should be named.
Disruption. Recycling means falling. The middle must accept that its position is a lease, not a title.
Legitimacy. The pear must persuade the middle that recycling is normal, and it must do so against the middle’s own interest in closure. This is the hardest political problem in the shape.
The band. The base must be comfortable enough not to grasp and far enough that the grasp is a real step. Outside the band, the pear degrades into the diamond (too comfortable) or the hourglass (too precarious).
The visible seam. A pear with a high recycling rate is a pear in which status anxiety is high even though status is not precarious. Anxiety without precarity is better than anxiety with precarity, but it is not comfort. The pear’s middle is competitive. That is the design.
THE BOTTOM LINE
The shape is the variable. Height describes the distribution; shape determines what the distribution does.
Four shapes, four logics. The pyramid usually produces caste. The hourglass produces anxiety and boundary-work. The diamond produces closure. The pear usually keeps class as class.
The pear maximizes the macro take—consumption stability, firm formation, labour elasticity, contestable incumbency—because the bulge supplies both founders and customers and the base can afford to move.
The pear maximizes the fiscal take—broad, shallow, stable—because the bulge contributes more than it receives and the top is thin enough to be taxed without flight. It also minimizes the invisible subsidy of status maintenance, which the hourglass pays and never labels.
The pear denies both conditions of caste. Adjacent rungs are close, so no gap is unbridgeable. Many rungs and many avenues mean no single gate can be captured, and constant recycling means no dynasty can form.
The design principle is asymmetric. Security competition at the base should be easy; position competition at the top should be hard. Reverse them and you get the hourglass, which is the pear inverted.
Boundary-work is the status claim of the precarious near-middle. Make the near-middle comfortable and the category loses its function. The category does not need to be refuted. It needs to be made unnecessary.
The upper working class does not keep up with the Joneses because its status is local and durable. The middle and upper middle do because their status is achieved and revocable. Conspicuous consumption is a tax paid on revocable status.
Consumption categorizes; production individuates. The middle posts the home and the mzungu wife because both are positions, and a position must be displayed to be held—the grammar of the post is I, because the I is the claim. The upper working class mzungu posts the factory, and if he posts the marriage he posts it as a we, because a partnership of persons has no rank to display. The enterprise is a specific thing that a specific person made. It is the counter-example the type cannot answer.
Hierarchy is only legible among the wazungu. The middle-class mzungu who posts the identical post with a Black local wife and a mixed family is making a claim that depends on her being a position while denying that she is one, to a host country national, in her own country. It reads as benevolence and functions as expropriation. The crassness is not the mixed family. It is the grammar.
Liberal values are the software of the pear. They are true enough to be believed because the shape usually makes them true. Contact individuates, and the pear supplies the contact. The hourglass supplies service contact, which is categorical by construction, and then wonders why the category will not die.
The pear costs. It disrupts the middle, it has a legitimacy problem, it requires a band, and it does not produce comfort. It produces competition without precarity. That is the best available shape, and it is the only one in which the category is unnecessary rather than forbidden.
References
Acemoglu, D., & Robinson, J. A. (2012). Why Nations Fail: The Origins of Power, Prosperity, and Poverty. Crown.
Alesina, A., & Glaeser, E. L. (2004). Fighting Poverty in the US and Europe: A World of Difference. Oxford University Press.
Bourdieu, P. (1984). Distinction: A Social Critique of the Judgement of Taste (R. Nice, Trans.). Harvard University Press.
Bourdieu, P., & Passeron, J.-C. (1970). La reproduction: Éléments pour une théorie du système d’enseignement. Minuit.
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.
Clark, G. (2014). The Son Also Rises: Surnames and the History of Social Mobility. Princeton University Press.
Corak, M. (2013). Income inequality, equality of opportunity, and intergenerational mobility. Journal of Economic Perspectives, *27*(3), 79–102.
Easterlin, R. A. (1974). Does economic growth improve the human lot? In P. A. David & M. W. Reder (Eds.), Nations and Households in Economic Growth. Academic Press.
Frank, R. H. (2007). Falling Behind: How Rising Inequality Harms the Middle Class. University of California Press.
Frank, R. H., & Cook, P. J. (1995). The Winner-Take-All Society. Free Press.
Goldin, C., & Katz, L. F. (2008). The Race between Education and Technology. Harvard University Press.
Habyarimana, J., Humphreys, M., Posner, D. N., & Weinstein, J. M. (2007). Why does ethnic diversity undermine public goods provision? American Political Science Review, *101*(4), 709–725.
Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional well-being. PNAS, *107*(38), 16489–16493.
Kuznets, S. (1955). Economic growth and income inequality. American Economic Review, *45*(1), 1–28.
Lenski, G. E. (1966). Power and Privilege: A Theory of Social Stratification. McGraw-Hill.
Lindert, P. H. (2004). Growing Public: Social Spending and Economic Growth since the Eighteenth Century. Cambridge University Press.
Marmot, M. (2004). The Status Syndrome: How Social Standing Affects Our Health and Longevity. Times Books.
Merton, R. K. (1938). Social structure and anomie. American Sociological Review, *3*(5), 672–682.
Miguel, E., & Gugerty, M. K. (2005). Ethnic diversity, social sanctions, and public goods in Kenya. Journal of Public Economics, *89*(11–12), 2325–2368.
Milanovic, B. (2016). Global Inequality: A New Approach for the Age of Globalization. Harvard University Press.
Piketty, T. (2014). Capital in the Twenty-First Century (A. Goldhammer, Trans.). Harvard University Press.
Rajan, R. G. (2010). Fault Lines: How Hidden Fractures Still Threaten the World Economy. Princeton University Press.
Sen, A. (1999). Development as Freedom. Oxford University Press.
Simmel, G. (1903/1950). The metropolis and mental life. In K. H. Wolff (Ed. & Trans.), The Sociology of Georg Simmel. Free Press.
Sombart, W. (1906). Warum gibt es in den Vereinigten Staaten keinen Sozialismus? Mohr.
Tilly, C. (1998). Durable Inequality. University of California Press.
Veblen, T. (1899). The Theory of the Leisure Class. Macmillan.
Weber, M. (1922/1978). Economy and Society (G. Roth & C. Wittich, Eds.). University of California Press.
Wilkinson, R., & Pickett, K. (2009). The Spirit Level: Why More Equal Societies Almost Always Do Better. Allen Lane.