writing

The Thermostat of Aspiration

15 min read

There is a dataset that I keep coming back to.

Raj Chetty and his team at Opportunity Insights matched the tax records of roughly twenty million Americans to the neighborhoods they grew up in. What they found was not subtle: a child born into a low-income family in one census tract of Seattle had an adult income comparable to the American average. A child from the same income bracket, growing up just a few miles away in a different tract, ended up near the bottom. The gap between these two kids, in dollar terms, is larger than almost any policy intervention we know how to run. And the cause was not their family. It was not their genes. It was the neighborhood — the specific square miles of the world they happened to inhabit during childhood.

This finding should disturb anyone who believes in individual agency. And in a way, it does disturb me. But the version of it that I find most interesting is not the deterministic reading — you are your zip code — but the exception. Because within every neighborhood, outcomes vary. Some kids from the worst-outcome tracts in America go on to extraordinary things. And some kids from the best-outcome tracts founder completely. The neighborhood explains a huge amount of variance, but it does not explain everything. The residual is human.

What explains the residual? That is the question this essay tries to answer. My argument, built from the economics literature on mobility and the psychology literature on self-concept, is this: the people and environments around you do not just shape your material opportunities — they shape what you believe you deserve. And the belief in what you deserve is, I think, the most underrated variable in the study of human outcomes.

Part I: What the Data Actually Shows

Let me first be precise about what Chetty's work actually demonstrates, because it is often misunderstood.

The Moving to Opportunity experiment, which randomly assigned housing vouchers to families in high-poverty public housing across five American cities from 1994 to 1998, is one of the cleanest natural experiments in social science. When Chetty, Hendren, and Katz linked MTO participants to tax records in 2016, they found something striking: children who moved to lower-poverty neighborhoods before age thirteen earned about 31% more as adults than those who stayed. They were more likely to attend college. Girls were less likely to become single parents. The neighborhood causally improved their lives.

But here is the part that gets less attention: children who moved to the same neighborhoods after age thirteen showed no significant improvement in adult earnings. At all. The same voucher, the same schools, the same neighbors — and basically no effect.

This age-of-exposure gradient is the most important empirical fact in the literature. It tells you that the neighborhood is not doing its work through the specific school you attend or the specific job market you enter. If it were, moving at sixteen would help just as much as moving at six. It is doing something developmental. Something that happens to you as you are becoming who you are.

A companion paper by Chetty and Hendren, which compared siblings within the same family who had moved at different ages, confirms this. Each additional year a child spends in a higher-opportunity county increases their adult income by roughly four percent. The effect is linear and cumulative, like compound interest. Moving at five gains you twice as much as moving at fifteen. The neighborhood is writing something on you, and it writes most legibly when you are young.

What is it writing?

Part II: The Standard Answers, and Why They're Incomplete

The usual explanations for neighborhood effects go roughly like this: better neighborhoods have better schools, lower crime, more stable social environments, and denser networks of employed adults. All of these things are true and all of them matter.

But they don't fully account for the age gradient. Better schools benefit a sixteen-year-old as much as a six-year-old. If the effect were purely about material conditions, we'd expect roughly equal benefits regardless of age of entry. The sharp dropoff after thirteen implies the mechanism is something specifically tied to development — to what happens to you as you are becoming who you are.

George Akerlof and Rachel Kranton's identity economics points in the right direction. Their argument is that people derive utility not just from outcomes but from acting in accordance with the norms of their perceived social category. Behaving inconsistently with your identity produces psychic costs that compete with whatever material benefit the inconsistent behavior might provide. If you see yourself as "someone like me," and someone like you doesn't negotiate their salary or apply to elite schools, then doing those things feels wrong in a way that is more fundamental than just being difficult.

This gets close to what I think is actually happening. But it still doesn't quite name it.

Part III: The Psychology of Deserving

Here is the distinction I want to make, and it is one the literature has been slow to fully articulate.

There is a difference between believing you can do something and believing you deserve to have it.

The first is what psychologists call self-efficacy, and it is well-studied. Albert Bandura's foundational 1977 paper established that self-efficacy — the belief in your own capability to execute a behavior — is one of the strongest predictors of whether people attempt and persist at difficult tasks. People with high self-efficacy for a domain approach it with more effort, more resilience in the face of setbacks, and less avoidance of threatening situations. The sources of self-efficacy are well-identified: your own past successes, watching similar people succeed, hearing encouragement from credible others, and interpreting your emotional states as signals of capability rather than incapacity.

The second — the belief that you deserve to have something — is murkier, less studied, and I think more important.

Deserving is not about capability. It is about legitimacy. It is the implicit answer to the question: is this for someone like me? And it operates below the level of conscious deliberation. You don't sit down and ask yourself whether you deserve to apply for the promotion. You just don't apply, experiencing the position as somehow not-yours, the way a stranger's apartment is not yours — not because you couldn't enter it but because the question of entering doesn't really arise.

Pierre Bourdieu, the French sociologist, had the sharpest analysis of this dynamic. He called the internalized sense of one's social place the habitus — a system of durable dispositions acquired through sustained exposure to one's social field, which then generates perceptions and behaviors consistent with that field without conscious calculation. The working-class child, Bourdieu argued, does not merely lack the resources to pursue elite outcomes. She has internalized a self-concept that experiences those pursuits as inappropriate — for others, not for her. He called this amor fati, a love of fate: you come to desire the life that is available to you, and stop desiring the life that isn't. The constraint becomes invisible because it becomes normal.

What Chetty's data shows, through the lens of habitus, is something heartbreaking and precise: the neighborhood is the primary site of habitus formation. During childhood — particularly before adolescence — the world around you is not just a context. It is the raw material from which you construct a self. What jobs the adults hold. What schools you see people like you attending. What happens to the ambitious ones. What the ceiling looks like from where you stand.

And here is the key thing: once formed, the habitus is not experienced as a constraint. It is experienced as common sense.

Part IV: The Self-Image Thermostat

Maxwell Maltz was a plastic surgeon in the 1950s who noticed something strange: when he fixed a disfigurement, some patients immediately blossomed — confidence, social life, career outcomes all improved. Others showed no change. Their face was different; they were not. He concluded that the limiting variable was the self-image: people had an internalized picture of themselves, and their behavior automatically self-corrected to match it. He published the idea in 1960 in Psycho-Cybernetics — the self-image as a thermostat, set to a target, automatically activating behavior to maintain that target.

The academic establishment ignored Maltz, and rightly questioned his rigor. But the mechanism has since been formalized. Hazel Markus and Paula Nurius's concept of "possible selves" — published in the American Psychologist in 1986 — showed that individuals hold specific, self-referential images of what they might become, and that these images directly regulate motivation and behavior. The content of your possible selves is acquired through social exposure: you can only imagine yourself as a lawyer if you have seen lawyers who look like you. Children who grow up in socioeconomically segregated neighborhoods — with few bridging ties to different social worlds, as Mark Granovetter's work on weak ties established — have a narrower range of conceivable selves not because they lack imagination, but because they lack the social raw material from which imagination draws.

The thermostat model integrates all of this. Your environment — through role models, teacher expectations, peer norms, and the social categories you learn to see yourself as belonging to — sets your self-image at a particular target. Your behavior then automatically calibrates to that target: you pursue opportunities consistent with it, self-correct away from opportunities inconsistent with it, and interpret your experiences through its lens. You are not consciously choosing your ceiling. Your ceiling is choosing your behavior.

The self-efficacy research fits within this model as the capability dimension of the thermostat. But the worth dimension — am I someone for whom this is possible? rather than can I do this? — is what I think really determines the height of the thermostat. And this is why neighborhood effects are so durable. The question is not whether you have learned enough. It is whether you have internalized a self-concept that includes the permission to want what your capabilities could actually obtain.

Part V: The Exception

Here is what keeps this from being a fatalistic argument.

Robert Rosenthal and Lenore Jacobson ran an experiment in an elementary school in San Francisco in the 1960s. They told teachers that certain students — randomly selected, unbeknownst to the teachers — had scored high on a test that predicted intellectual blooming. Those students gained significantly more IQ points over the subsequent year than their peers. The only difference was the teacher's expectation.

The mechanism Rosenthal identified runs through four channels: teachers in the high-expectation condition provided warmer emotional climates, more challenging material, more opportunities for students to respond, and more detailed feedback. They behaved differently because they believed differently. And the students — absorbing those behavioral signals continuously over a school year — internalized a different picture of what they were.

Claude Steele's stereotype threat research adds the flip side: when Black students at Stanford were told a verbal test was diagnostic of intellectual ability, they underperformed relative to white students with equivalent SAT scores. Frame it as a problem-solving exercise, and the gap disappeared. The mechanism is the same in reverse — sustained exposure to low-expectation environments may produce chronic vulnerability that depresses performance exactly when it most needs to be high.

Together, these findings identify the mechanism of the exception: the thermostat can be revised upward by people who expect more of you than your background predicts.

A single high-expectation teacher can partially counteract years of low-expectation environment. A mentor who treats you as someone destined for extraordinary things communicates not just "you can" but "this is for you." The MTO findings are consistent with this: moving to a better neighborhood before thirteen provides sustained exposure to adults, peers, and institutions that hold different expectations, which over years revises both the capability and worth dimensions of self-concept.

The exception is not random. It is produced by specific mechanisms. And understanding those mechanisms is the beginning of being able to engineer them.

Part VI: What I Think This Means

I want to be careful here, because this is where analysis shades into prescription, and prescription can easily become either fatalism ("your zip code is your destiny") or victim-blaming ("just believe in yourself"). Neither of those captures what I actually think.

What I think is this: the data is real, the mechanisms are real, and they point in a fairly clear direction for individuals who are trying to understand their own ceilings.

The first implication is diagnostic. If you find yourself consistently not pursuing things your capabilities would qualify you for — not applying, not asking, not negotiating — it is worth asking whether the constraint is efficacy ("I don't think I can do it") or worth ("I don't think this is for me"). The efficacy question has a standard answer: gather evidence, build mastery, find similar others who succeeded. The worth question requires something different: challenge the category membership itself. Ask whose permission you are waiting for, and where you learned to wait for it.

The second implication is environmental and relational. You cannot imagine yourself into categories you have never seen anyone who resembles you inhabiting — so deliberately put yourself in rooms where people are doing what you want to do, even when you feel like a fraud there, especially then. And find people who expect more of you than your history predicts. Not cheerleaders, but specific people who see a version of you that you haven't yet internalized as real. The Pygmalion research suggests a single such relationship, sustained over time, can partially rewrite what your environment wrote — not because belief is magic, but because someone who treats you as extraordinary gives you different material to work with.

The worth dimension, though, is the one I find hardest and most important. Capability is trainable. The belief that you deserve to deploy it ambitiously is a more fundamental revision. It requires something like a renegotiation of identity — a decision, made slowly and usually through accumulated evidence rather than a single moment, that the category "someone like me" now includes the possibility of the life you actually want.

This is not a decision that can be made purely by willpower. It requires the right inputs: exposure, expectation, and time. But it can be made. And the research suggests, tentatively, that making it is the thing that most distinguishes the people who exceed their environmental prediction from the ones who confirm it.

Coda: The Fifty-Three Percent

Chetty and Hendren estimate that about 53% of the cross-area variance in upward mobility in the United States is explained by causal neighborhood effects — by places themselves, not by the families who sort into them.

Fifty-three percent is enormous. It means that more than half of the variation in who rises and who doesn't in America is explained by geography — by the specific square miles a child happened to occupy during the years when the thermostat was being set.

The remaining forty-seven percent is not fully explained either. Some of it is family. Some of it is genetics. Some of it is chance — the teacher you got, the mentor you encountered, the moment someone told you something that revised your picture of yourself.

But some of it, I believe, is the work of people who decided to revise the thermostat deliberately. Who noticed the ceiling and questioned the source of it. Who sought environments that expanded what they thought was possible for someone like them. Who found, eventually, that the self-concept their neighborhood had given them was a first draft rather than a final one.

The data cannot tell us exactly how large that fraction is. But I think it is larger than we act like it is. And I think naming the mechanism — not just "work hard" or "believe in yourself" but the specific psychological structure of deserving, and the specific environmental inputs that revise it — is the first step toward making it more intentional.

That is what I have tried to do here.

References

Akerlof, George A. and Rachel E. Kranton. "Economics and Identity." Quarterly Journal of Economics 115, no. 3 (2000): 715–753.
Bandura, Albert. "Self-Efficacy: Toward a Unifying Theory of Behavioral Change." Psychological Review 84, no. 2 (1977): 191–215.
Bandura, Albert. Self-Efficacy: The Exercise of Control. New York: W. H. Freeman, 1997.
Bourdieu, Pierre. Distinction: A Social Critique of the Judgement of Taste. Cambridge, MA: Harvard University Press, 1984.
Bourdieu, Pierre. The Logic of Practice. Stanford: Stanford University Press, 1990.
Chetty, Raj, Nathaniel Hendren, and Lawrence F. Katz. "The Effects of Exposure to Better Neighborhoods on Children." American Economic Review 106, no. 4 (2016): 855–902.
Chetty, Raj and Nathaniel Hendren. "The Impacts of Neighborhoods on Intergenerational Mobility I: Childhood Exposure Effects." Quarterly Journal of Economics 133, no. 3 (2018): 1107–1162.
Chetty, Raj, et al. "The Opportunity Atlas: Mapping the Childhood Roots of Social Mobility." NBER Working Paper 25147 (2018).
Chetty, Raj, et al. "Where is the Land of Opportunity?" Quarterly Journal of Economics 129, no. 4 (2014): 1553–1623.
Granovetter, Mark S. "The Strength of Weak Ties." American Journal of Sociology 78, no. 6 (1973): 1360–1380.
Jussim, Lee and Kent D. Harber. "Teacher Expectations and Self-Fulfilling Prophecies." Personality and Social Psychology Review 9, no. 2 (2005): 131–155.
Maltz, Maxwell. Psycho-Cybernetics. Englewood Cliffs, NJ: Prentice-Hall, 1960.
Markus, Hazel and Paula Nurius. "Possible Selves." American Psychologist 41, no. 9 (1986): 954–969.
Putnam, Robert D. Bowling Alone. New York: Simon & Schuster, 2000.
Rosenthal, Robert and Lenore Jacobson. Pygmalion in the Classroom. New York: Holt, Rinehart and Winston, 1968.
Rotter, Julian B. "Generalized Expectancies for Internal versus External Control of Reinforcement." Psychological Monographs 80, no. 1 (1966): 1–28.
Steele, Claude M. and Joshua Aronson. "Stereotype Threat and the Intellectual Test Performance of African Americans." Journal of Personality and Social Psychology 69, no. 5 (1995): 797–811.