Book Review: Discrimination and Disparities
In his provocative and meticulously researched book Discrimination and Disparities, economist and social theorist Thomas Sowell challenges the widely held belief that discrimination is the primary cause of disparities between racial and ethnic groups in America. He presents a wealth of data and historical context to demonstrate that disparities can arise between groups independently of discriminatory practices. Sowell's dispassionate and data-driven approach challenges readers to confront hard truths about the origins of group differences. His analysis sheds light on the complex nature of disparities, offering valuable insights for policymakers and individuals seeking to grasp the nuances of these issues more deeply. Even those who disagree with some of his conclusions will find their assumptions productively tested by his analysis.
Sowell's powerful insights into the complex nature of disparities are grounded not only in his rigorous academic training and empirical research, but also in his own remarkable life experiences. Growing up in the segregated South and later in Harlem, Sowell directly confronted the challenges and barriers faced by marginalized communities. His personal journey from humble beginnings to becoming one of America's most distinguished intellectuals has undoubtedly shaped his unique perspective on issues of race, inequality, and social policy.
Background on Thomas Sowell
Sowell was born in 1930 in rural North Carolina. His father died before his birth, and his mother died while giving birth to his younger brother a few years later. He was adopted by his great-aunt and moved to Charlotte, then to Harlem, New York City, at the age of eight. Sowell dropped out of high school and worked various odd jobs before being drafted into the Marines. After his military service, he pursued his education relentlessly, taking night classes that led him to Harvard, where he graduated magna cum laude in 1958. After earning a master’s degree in economics, he went on to obtain a doctorate in economics from the University of Chicago under Nobel laureate George Stigler.
Now a senior fellow at Stanford University's Hoover Institution, Sowell has authored nearly 50 books across various disciplines, including history, political science, and the development of late-speaking children. His writing is for the general public, exemplified by his textbook Basic Economics, eschews technical jargon in favor of accessible explanations. This unique perspective informs his analysis of complex social issues, including the nature and causes of group disparities.
Disparities are Natural and Pervasive
One of the most insightful methods economists use is the practice of flipping questions on their heads. Economists don’t ask why countries are poor; they remind us that all countries started poor, and ask why some became rich. Similarly, Sowell suggests that, when discussing disparities, the question shouldn’t be why there are differences between groups, but rather, why there is an expectation of uniformity in the first place.
Indeed, in parts of modern society, disparities are often taken as proof of discrimination. This is perhaps the fundamental assumption behind the antiracist movement spearheaded by those such as Ibram X. Kendi. In Stamped from the Beginning, Kendi says, “When you truly recognize that the racial groups are equals, then you also recognize that racial disparities must be the result of racist policy.”
In Discrimination and Disparities, Sowell surveys an extensive range of data, uncovering disparities that pervade every level of society and nature. Disparities exist not only among individuals but also across various groups and institutions, ranging from families and businesses to entire nations. He posits that disparities are, in some sense, the natural order of things.
Disparities Among Individuals
Disparities begin at the individual level, with one notable example being the impact of birth order. Research consistently shows that firstborn children tend to score higher on IQ tests than their siblings:
A study of National Merit Scholarship finalists, for example, found that, among finalists from five-child families, the first-born was the finalist more often than the other four siblings combined. First-borns were also a majority of the finalists in two-child, three-child, and four-child families. IQ data from Britain, Germany and the United States showed that the average IQ of first-born children was higher than the average IQ of their later-born siblings. Studies with sample sizes ranging in the hundreds of thousands have found similar results.
He highlights numerous examples of disproportionate achievements of first-born children, from becoming astronauts to doctors to Members of Congress. While individual anecdotes or studies might not be persuasive on their own, collectively, they provide compelling evidence. There’s something different about first-born children. For one, they are far more likely to be members of the rationalist community. Scott Alexander’s analysis of Slate Star Codex survey data found that out of the 2965 respondents from two-child families, “a full 2118 of those were the older of the two. That’s 71.4%. p ≤ 0.00000001.” (Note that there is a follow-up post here.) As Scott said, “Birth order effects are totally a real thing.”
Discrimination almost certainly can’t be the cause of these disparities because people can’t easily tell if someone is first-born. It’s not evident just by looking at them or their employment applications. Sowell notes that the only obvious benefit for firstborns is the undivided attention they receive from parents in their early years. While one could argue that this preferential treatment is a form of discrimination, it would be a markedly different use of the term 'discrimination' than its common meaning.
Sowell concludes by posing the following question:
If there is not equality of outcomes among people born to the same parents and raised under the same roof, why should equality of outcomes be expected—or assumed—when conditions are not nearly so comparable?
Geographic and Natural Disparities
Disparities apply to people more broadly. He gives examples of disparities caused by geography:
Geographic differences are one among other factors that make for a skewed distribution of outcomes. Coastal peoples have long tended to be more prosperous and more advanced than people of the same race living farther inland, while people living in river valleys have likewise tended to be more prosperous and more advanced than people living up in the mountains.
Most of the most fertile land in the world is in the temperate zones and little or none in the tropics. Areas that are both near the sea and in the temperate zones have 8 percent of the world’s inhabited land area, 23 percent of the world’s population, and 53 percent of the world’s Gross Domestic Product.
Disparities are evident not only in human affairs but also in the natural world. Indeed, Sowell argues that in any scenario that depends on multiple prerequisites, we should expect significant disparities. Sowell gives the example of tornados:
Multiple factors have to come together to create tornadoes, and more than 90 percent of all tornadoes in the entire world occur in just one country—the United States. Yet there is nothing startlingly unique about either the climate or the terrain of the United States that cannot be found, as individual features, in various other places around the world. But all the prerequisites for tornadoes do not come together as often in the rest of the world as in the United States.
These examples from the natural world demonstrate that disparities are not just a common occurrence, but the natural state of affairs.
Change Over Time
Similarly, in human endeavors, success often hinges on a combination of key factors: intelligence, effort, supportive institutions, and luck. These elements can be seen as essential prerequisites for success, and lacking any one of them can significantly reduce the likelihood of achieving one's goals. Therefore, it is unrealistic to expect an even or random distribution of success, whether among individuals, groups, institutions, or nations. This also means that, as conditions and opportunities change, so too can the groups that are most successful. Sowell tells us that “backwardness in a given era does not mean backwardness forever," citing the Scots and Jews as examples.
He points out that though Scotland was once considered economically and educationally backward, a disproportionate number of Britain's leading intellectual figures in the 18th and 19th centuries were of Scottish ancestry, such as John Stuart Mill and Adam Smith. Similarly, despite facing persistent discrimination and having minimal involvement in science and technology before the early Industrial Revolution, Jews made remarkable advances from the mid-20th century onwards. Sowell highlights that, “With Jews constituting less than one percent of the world’s population, they have received 22 percent of the Nobel Prizes in chemistry, 32 percent in medicine, and 32 percent in physics.”
Disparities Are The Default
Sowell summarizes the lessons from disparities as follows (emphasis in the original):
What can we conclude from all these examples of highly skewed distributions of outcomes around the world? Neither in nature nor among human beings are either equal or randomly distributed outcomes automatic. On the contrary, grossly unequal distributions of outcomes are common, both in nature and among people, in circumstances where neither genes nor discrimination are involved.
What seems a more tenable conclusion is that, as economic historian David S. Landes put it, “The world has never been a level playing field.” The idea that it would be a level playing field, if it were not for either genes or discrimination, is a preconception in defiance of both logic and facts.
Discrimination
Discrimination is an odd word in that it has different, nearly opposite meanings. One can be discriminating in their tastes and gain a reputation for selecting the finest red wines. Conversely, one could discriminate against red wines and reject them all without giving them as much as a sniff. So we need to be clear about exactly what we’re saying when we talk about discrimination.
Sowell creates two categories of discrimination:
Discrimination 1: “an ability to discern differences in the qualities of people and things, and choosing accordingly”
Discrimination 2: “treating people negatively, based on arbitrary assumptions or aversions concerning individuals of a particular race or sex, for example”
Discrimination 1 is evidence-based, whereas Discrimination 2 stems from antipathy. Discrimination 2 is the more commonly used meaning and is the reason for anti-discrimination laws.
In the ideal, Discrimination 1 would mean judging each person as an individual, regardless of what group that person is part of. But in the real world, conditions are never ideal. Thus we can see that Discrimination 1 is a broad category, and could be split to be more precise. Sowell creates the following distinction:
Discrimination 1-A: Basing decisions on evidence about individuals
Discrimination 1-B: Basing decisions on evidence about groups
Discrimination 1-A represents the ideal form, while Discrimination 1-B is less than ideal. However, both are distinct from Discrimination 2, which is rooted in “unsubstantiated notions and animosities”.
Sowell provides an example to explain the difference:
“To take an extreme example of Discrimination 1-B, for the sake of illustration, if 40 percent of the people in Group X are alcoholics and 1 percent of the people in Group Y are alcoholics, an employer may well prefer to hire only people from Group Y for work where an alcoholic would be not only ineffective but dangerous. This would mean that a majority of the people in Group X —60 percent in this case—would be denied employment, even though they are not alcoholics.”
[...]
This is certainly not judging each job applicant as an individual, so it is not Discrimination 1 in the purest sense of Discrimination 1-A. On the other hand, it is also not Discrimination 2, in the sense of decisions based on a personal bias or antipathy toward that group. The employer might well have personal friends from Group X, based on far more knowledge of those particular individuals than it is possible to get about job applicants, without prohibitive costs.
Thus it is Discrimination 1-B because it is based on evidence, but not about the specific individuals.
Costs of Discrimination
Sowell emphasizes the importance of evaluating actions in terms of costs. There are costs we incur by going about our daily lives. Discrimination can increase or decrease those costs. He poses a scenario to illustrate his point: “If you are walking at night down a lonely street, and see up ahead a shadowy figure in an alley, do you judge that person as an individual or do you cross the street and pass on the other side?”
Not discriminating would be to continue to walk, no matter the evidence, even if the figure makes threats and pulls out a knife. Discrimination 1-A would be to not judge the shadowy figure until you get to know them. Ask them about their life and, if they pull out a knife and threaten you, avoid them.
But most people would opt for Discrimination 1-B—avoiding all shadowy figures in all alleyways. Sowell’s point is that although we might prefer to always judge people as individuals, there is a cost to consider, and there are situations where the costs of ignoring statistical differences between groups are too high. In this situation, the cost of not discriminating is significant because, as Sowell puts it, “a mistake on your part could be costly, up to and including costing you your life.”
This principle applies not only to personal safety but also to business decisions and public policy. In another of his books, The Quest for Cosmic Justice, published in 1999, Sowell, tells the story about how the relative of a member of the San Francisco Board of Supervisors tried to order a pizza:
[He] was told that the company does not deliver pizza where he lives, which happened to be in a high-crime neighbourhood. Immediately there was an outburst of moral indignation for which San Francisco—and especially San Francisco politicians—are all too well known. A law was passed immediately, decreeing that anyone who makes deliveries to the public in any part of the city must make deliveries all over the city.
At first glance, the San Francisco Board of Supervisors' decision to require pizza delivery across the entire city might seem like a moral victory, ensuring equal access to services for all residents. However, Sowell argues that this perspective fails to account for the full costs of the policy and who bears them. While the Board members can celebrate their perceived win for social justice, they are not the ones who will face the increased risks associated with delivering pizza in high-crime neighborhoods. Instead, it is the delivery drivers who will bear the burden of this decision. Sowell emphasizes that the costs of such policies extend beyond mere financial considerations, noting that "in this case, the increased costs would include dead truck drivers."
Costs of Discrimination 2
We can also consider the costs associated with Discrimination 2, which have been immense throughout history. Legalized forms such as slavery and Jim Crow laws imposed massive economic, physical, and psychological harms on black communities for generations, severely restricting opportunities and stifling upward mobility. Even after such overt discrimination was outlawed, pervasive societal racism, unequal treatment by institutions like banks and the criminal justice system, and disproportionate police violence against black people have continued to impose substantial costs.
Sowell knows the pain of these costs all too well. In Maverick: A Biography of Thomas Sowell, author Jason L. Riley says “[Sowell had] been turned away from restaurants and housing because of his skin color. He’d felt the pain and humiliation of racism firsthand throughout his life. He needed no lectures from anyone on the evils of Jim Crow.”
But Sowell builds on the work of economist Gary Becker and suggests it's worthwhile to not only acknowledge the direct costs borne by the victims of discrimination but also to analyze the costs that discriminators incur. Although it might seem odd to consider these costs, recall that the point is not to make moral claims but to develop a model to better understand discrimination. For example, one such cost is that discriminators might face economic disadvantages, such as reduced market competitiveness due to limiting their pool of talent based on arbitrary criteria. Another cost would be legal repercussions including penalties and fines.
This model will help us understand how certain policies could affect discrimination. By ensuring that these costs cannot be easily avoided, society can create strong deterrents against discriminatory practices. Conversely, it is important to recognize that some policies might inadvertently relieve discriminators of these costs if not carefully designed and implemented.
Let’s take the example of laws enforcing segregation. The most direct costs of legally mandated racial segregation fell overwhelmingly on black Americans, while the white policymakers and voters who imposed this discriminatory system were largely insulated from its harms.
Black individuals consigned to separate and starkly inferior schools, housing, and public facilities faced severely constrained opportunities for economic advancement. They also endured immense dignitary harms and psychological trauma from being officially designated second-class citizens. Perversely, anti-black disenfranchisement meant those most burdened by segregation were denied political recourse to dismantle it. White politicians paid no price for ignoring black interests and indeed were often rewarded by white voters for maintaining racial subjugation.
Society as a whole paid a cost because of the underutilization of talent and potential within a large portion of the population, but because such costs are diffused among the population (no one voter suffers a great deal from them), they can easily be overshadowed by racial animosity. While society as a whole suffered because of the underutilization of talent and potential within a large portion of the population, these costs are diffused among the population (no one voter suffers a great deal from them), they can easily be overshadowed by racial animosity.
Indeed, one of the few groups, aside from the black population, that incurred costs were the companies forced to implement these policies. Train companies, for instance, famously opposed segregation laws. Their opposition stemmed from a straightforward economic calculation: operating a single, full train was more cost-effective than running two half-full trains segregated by race.
In sectors where efficiency was key to profitability, companies often opposed discriminatory laws that hindered their operations and increased their costs. Thus, Sowell argues that the main counterforce to discrimination was the market itself.
Discrimination vs The Market
By emphasizing the economic irrationality and inefficiency of discrimination, Sowell shows how market forces can serve as a natural check against discrimination. He suggests evaluating this claim by examining a historical context where racial discrimination was not only explicit but legally enforced:
To avoid endless and inconclusive debates about the presence or magnitude of racism, we can test our hypotheses about the costs of discrimination in a context where there is no ambiguity on the subject—namely, South Africa during the era of apartheid, ruled by a white minority government, elected with the black majority denied a vote, and openly promoting white supremacy.
Sowell cites economist Walter E. Williams’ analysis in South Africa's War Against Capitalism to demonstrate how even stringent racist policies could not completely suppress market forces.
In apartheid-era South Africa, the government implemented laws that severely restricted the employment of blacks in certain industries, mandating racial discrimination. Yet there were numerous instances where black employment exceeded legal limits.
Is this because the employers were friendly and less racist than the voters? No, it’s because South African firms, like all firms, thought in terms of costs. These laws imposed a cost to businesses; exceeding the set quotas on black employees could, and often did, result in fines. However, compliance also carried a cost: if the most qualified candidate for a job was black and a firm chose not to hire them in order to comply with racial quotas, they missed out on a valuable employee. This dilemma often led firms, especially in competitive industries, to defy the laws, choosing to pay fines rather than forgo the best talent.
Even racists want to make the most money they can. Sowell concludes, “While racists, by definition, prefer their race to other races, individual racists, like other people, tend to prefer themselves most of all.”
In short, many racist policies failed largely because even racism couldn’t fight a greater force—the invisible hand of the market. Many racist hiring practices in South Africa were undermined because the economic cost of discrimination was too high. Sowell notes that racist policies tend to have the largest impact in environments where there is no significant market pressure to counteract them. Thus, in sectors or regions where market forces are weak or absent, discriminatory practices can persist and have more severe impacts. This is not to say that discrimination does not exist in competitive markets, just that there’s a higher cost to be paid for it. Indeed, in Markets and Minorities, published in 1981, he says, “The competitiveness of the market puts a price on discrimination, thereby reducing but not necessarily eliminating it.”
The complex relationship between racism and market economies is also evident in the United States, particularly in the post-Civil War South. In his book, The Declining Significance of Race, sociologist William Julius Wilson highlights ways in which white Southern landowners sought to keep down the earnings of black workers. However, Sowell points to work by economist Robert Higgs, who found that organized efforts by white employers and landowners in the postbellum South to control wages and conditions for black workers often failed due to competition between the landowners. Despite the vulnerability of newly freed blacks, market competition prevented these exploitative arrangements from lasting. In agriculture, which dominated the southern economy, the pressing need to plant crops each spring created urgency. Landowners who offered better terms quickly secured the best workers, enhancing their crop yields. Those who adhered to the restrictive agreements, or who underpaid and cheated workers, ended up with less skilled labor, diminishing their agricultural productivity.
What Explains the Disparities?
Sowell views disparities as a natural state, which raises the question about the extent to which we should expect to explain them. However, he seems to think that while some level of disparity is to be expected, this does not preclude further examination or explanation of its causes.
Throughout his work, Sowell recognizes that various forms of discrimination, whether recent or historical, have played a role in shaping disparities. However, he argues that discrimination alone is insufficient to fully account for the differences we see today. Rather than attempting to quantify the exact contributions of various factors, Sowell focuses on identifying what appear to be the primary influences. His approach does not seek a definitive answer but aims to determine the main trends driving these disparities.
When examining disparities, Sowell starts by asking what magnitude of differences we should expect to see naturally. He looks for appropriate benchmarks and comparisons to put the observed disparities into context. For example, in his latest book Social Justice Fallacies, published in 2023, he analyzes various types of disparities, including income differences between black and white populations in the US, and compares them to differences seen between other groups:
Median black American family income has been lower than median white American family income for generations. As regards the magnitude of the difference, official government data going back as far as 1947 show that the disparity has not been as large as 2:1 in any of those years. How does that particular disparity compare to disparities among other groups in the United States, or among groups in other countries?
Within the United States, the median per capita income of such Asian ethnic groups as those of Chinese, Japanese, Indian and Korean ancestry is more than twice as high as the median per capita income of Mexican Americans. These Asian groups also have higher median per capita incomes than the median per capita income of white Americans. Asian Indians have nearly three times the median per capita incomes of Mexican Americans, and a median per capita income more than $15,000 a year higher than the median per capita income of white Americans. Among full-time, year-round male workers, Asian Indian males earned over $39,000 a year more than white male full-time, year-round workers.
Is this the “white supremacy” we are so often warned about in some quarters?
Sowell is not engaging in moral judgments. His argument is not that existing disparities are acceptable or inacceptable; rather, he posits that disparities are a natural occurrence within societies. He contends that the disparities observed between black and white populations fall within the normal range seen across different demographic groups.
Is it Present Day Discrimination 2?
While disparities alone do not prove discrimination, it is important to consider whether discrimination is contributing to observed differences between groups. The existence of Discrimination 2—treating people negatively based on arbitrary group characteristics—is not in question. It is a real and ongoing problem. The key question, which Sowell seeks to investigate, is the extent to which Discrimination 2 drives current disparities.
Sowell scrutinizes the data for patterns that would be consistent with Discrimination 2 being the primary cause. One revealing example he explores is the gap in labor force participation rates between young black and white males. He notes that this disparity is widest among 16- and 17-year-olds and narrows with age. If Discrimination 2 were the dominant factor, we would expect to see a consistent gap across all age groups, as the race of these individuals remains constant over time. The fact that the gap shrinks as the cohorts get older suggests that other factors are at play. Perhaps employers are accurately assessing a lack of necessary skills in these individuals, which would be Discrimination 1. As Sowell puts it, Discrimination 2 "cannot explain the age difference in discrimination of young black males who do not change races as they get older."
He considers other commonly asserted explanations for the disparity in labor force participation rates:
Alternative explanations for these changing patterns of racial differences—such as racism, poverty or inferior education among blacks—cannot establish even correlation with changing employment outcomes over the years, because all those things were worse in the first half of the twentieth century, when the unemployment rate among black teenagers in 1948 was far lower and not significantly different from the unemployment rate among white teenagers.
To illustrate the limitations of a focus solely on discrimination, in Social Justice Fallacies, Sowell points to the example of predominantly white "hillbilly" communities mired in persistent poverty:
What we learn from persistent and severe poverty in hillbilly communities can be helpful in sorting out factors involved in the poverty and lagging progress of other peoples, including racial minorities. If, by some miracle, we could get to zero racism, it is by no means certain how much effect that would have. People in low-income American hillbilly counties already face zero racism, because these people are virtually all white. Yet they have lower incomes than blacks.
Ultimately, Sowell does not offer a precise breakdown of how much of current disparities can be attributed to Discrimination 1-A (individual assessments), 1-B (group averages), and 2 (arbitrary bias), or what proportion stems from historical versus ongoing discrimination. However, he does contend that Discrimination 1-A likely plays a significant role in the specific case of employment, with employers making decisions based on the actual skills and qualifications of individual applicants.
Is It Historical Discrimination?
Even if Discrimination 2 is not the main factor driving current disparities, there's clearly some lingering effect from historical discrimination. Many Jim Crow laws remained in effect in the South until the Voting Rights Act of 1965, and their impacts did not dissipate overnight.
The questions of how much and how quickly groups can recover from severe, long-standing discrimination are complex and emotionally charged. For Sowell, however, these are ultimately empirical questions that should be addressed through data and historical analysis. In his 1984 book Civil Rights: Rhetoric or Reality, Sowell examines the case of the Chinese in Southeast Asia, among other groups, to illustrate the potential limits of discrimination's impact:
The civil rights vision focuses on groups adversely affected in statistical disparities. Here the relationship between discrimination and economic, educational, and other disadvantages is taken as virtually axiomatic. But if this apparently obvious proposition is taken as an hypothesis to be tested, rather than an axiom to be accepted, a very different picture emerges. Groups with a demonstrable history of being discriminated against have, in many countries and in many periods of history, had higher incomes, better educational performance, and more “representation” in high-level positions than those doing the discriminating.
Throughout southeast Asia, for several centuries, the Chinese minority has been—and continues to be—the target of explicit, legalized discrimination in various occupations, in admission to institutions of higher learning, and suffers bans and restrictions on land ownership and places of residence. Nowhere in Malaysia, Indonesia, Vietnam, Thailand, or the Philippines have the Chinese ever experienced equal opportunity. Yet in all these countries the Chinese minority—about 5 percent of the population of southeast Asia—owns a majority of the nation’s total investments in key industries. […] In Malaysia, where the anti-Chinese discrimination is written into the Constitution, is embodied in preferential quotas for Malays in government and private industry alike, and extends to admissions and scholarships at the universities, the average Chinese continues to earn twice the income of the average Malay.
Nor are the Chinese minorities in southeast Asia unique. Much the same story could be told of the Jews in many countries around the world and in many periods of history. A similar pattern could also be found among East Indians in Africa, southeast Asia and parts of the western hemisphere, or among Armenians in the Middle East, Africa, and the United States. Italian immigrants to Argentina in the late nineteenth and early twentieth centuries also encountered discrimination, but nevertheless rose from poverty to affluence, surpassing the Argentine majority.
Japanese immigrants to the United States also encountered persistent and escalating discrimination, culminating in their mass internment during World War II, but by 1959 they had about equaled the income of whites and by 1969 Japanese American families were earning nearly one-third higher incomes than the average American family.
Sowell's analysis does not deny that present or historical discrimination can impact economic outcomes, but rather highlights the complex and context-dependent nature of this relationship. He suggests that while the effects of past discrimination may linger, its influence on current disparities may be more limited than often assumed.
If discrimination were the primary driver of group differences we would expect to see a more consistent pattern of disadvantaged groups underperforming relative to their discriminators. However, the fact that some groups who have faced significant discrimination now outperform the majority population in many economic measures suggests that the link between discrimination and outcomes is not always straightforward.
This is not to say that the impact of discrimination should be dismissed or that it cannot contribute to disparities. And, of course, I should add another reminder that none of these statements are meant to moralize—none of this makes discrimination “OK”. The success of some discriminated-against groups does not negate the reality of the hardships they faced.
Is it Genetic?
The extent to which genetic differences between racial groups contribute to observed disparities is an extremely contentious and often taboo topic. However, for Sowell, who is no stranger to controversy, it is ultimately an empirical question to which we should apply the standard approach of hypothesis testing and a careful weighing of evidence.
While Sowell acknowledges that his analysis hasn’t definitively excluded genetic factors, he argues that the claim of significant genetic differences requires evidence, and the available evidence doesn’t seem to match the hypothesis. In particular, he notes that data on the disproportionate achievements of first-born children across racial groups are hard to square with a genetic explanation. Similarly, the fact that previously "backward" nations like Scotland and Japan have risen to become economic leaders suggests that it’s not a good explanation.
(For more on Sowell’s thoughts on the relationship between race, IQ, and genetics, see his review of The Bell Curve.)
Sowell’s Cultural Explanation
Instead, he turns his attention to the role of culture in shaping outcomes for different groups. In his view, cultural factors, rather than innate biological differences, are the primary drivers of the observed disparities between whites and blacks in the United States. In his 2006 book Black Rednecks and White Liberals, Sowell argues that cultural factors, particularly the influence of "redneck culture," are the primary drivers of disparities between whites and blacks.
He contends that this culture, marked by a disdain for education, work ethic, and social responsibility, originated among certain groups in England and was brought to the American South by immigrants. Enslaved Africans, stripped of their native cultures in America, often adopted the prevalent "redneck culture" of the South. As blacks later migrated throughout the country, they spread this culture, perpetuating its negative impact on both black and white Southerners.
Sowell highlights historical distinctions within the black population in the US, particularly between native-born blacks and West Indian immigrants. Despite a shared history of enslavement, these groups exhibited significant disparities in areas like entrepreneurship, education, and incarceration rates, which he attributes to cultural differences:
Blacks in the United States noted that “the Negro immigrants, particularly the British West Indians, bring a zest of learning that is not typical of the native-born population.” While black Americans have long been over-represented among people in prison, a study of the racial composition of New York State’s Sing Sing prison in the early 1930s found that black West Indians were under-represented relative to their share of the population, at a time when native-born black Americans were over-represented several-fold among Sing Sing inmates.
[...]
Neither race nor racism can explain such differences. Nor can slavery, since native-born blacks and West Indian blacks both had a history of slavery. Studies published in 2004 indicated that an absolute majority of the black alumni of Harvard were either West Indian or African immigrants, or the children of these immigrants. Somewhat similar findings have emerged in studies of some other elite colleges. With blacks as with whites, the redneck culture has been a less achieving culture. Moreover, that culture has affected a higher proportion of the black population than of the white population, since only about one-third of all whites lived in the antebellum South, while nine-tenths of all blacks did.
Furthermore, Sowell contends that the expansion of liberal policies and the welfare state in the 1960s exacerbated the disparities between blacks and whites. He argues that blacks had been making progress in reducing these disparities, but their progress was undone by the rise of the welfare state and the War on Poverty in the 1960s. He articulates his critique of these policies in response to Nicholas Kristof’s New York Times article “When Whites Just Don’t Get It, Part 4”, in a piece titled “A Legacy of Liberalism”:
Nearly a hundred years of the supposed "legacy of slavery" found most black children being raised in two-parent families in 1960. But thirty years after the liberal welfare state found the great majority of black children being raised by a single parent.
The murder rate among blacks in 1960 was one-half of what it became 20 years later, after a legacy of liberals' law enforcement policies. Public housing projects in the first half of the 20th century were clean, safe places, where people slept outside on hot summer nights, when they were too poor to afford air conditioning. That was before admissions standards for public housing projects were lowered or abandoned, in the euphoria of liberal non-judgmental notions. And it was before the toxic message of victimhood was spread by liberals. We all know what hell holes public housing has become in our times. The same toxic message produced similar social results among lower-income people in England, despite an absence of a "legacy of slavery" there.
If we are to go by evidence of social retrogression, liberals have wreaked more havoc on blacks than the supposed "legacy of slavery" they talk about. Liberals should heed the title of Jason Riley's insightful new book, Please Stop Helping Us.
Sowell's critique of liberal policies and the welfare state in the 1960s highlights his broader concern with the unintended consequences of well-intentioned government interventions. In his view, these policies not only failed to achieve their stated goals but actually exacerbated the very problems they were meant to solve, ultimately harming the black community. This perspective informs his wider analysis of government efforts to combat racial discrimination.
Evaluating Government Policies
Sowell evaluates various government policies aimed at reducing racial discrimination through the lens of an economist, focusing less on the intentions of policymakers and more on the actual consequences—both intended and unintended. He finds that government policies often fail to account for their unintended consequences, leading to suboptimal or even counterproductive results. An illustrative example of this is the impact of minimum wage laws.
By setting wage floors above the market-clearing level, minimum wage hikes tend to increase the number of job applicants, while simultaneously decreasing the number of workers hired because labor has become more expensive. This creates a surplus of qualified applicants, which allows employers greater freedom to choose among them. In such scenarios, employers may indulge their biases, selecting candidates based on arbitrary preferences, such as racial animosity (Discrimination 2). As Sowell explains, “On the most basic economic principles, such a situation makes racial or other discrimination far more affordable by employers, and therefore more sustainable, than in a situation where wage rates are determined by supply and demand in a free, competitive market.”
Sowell extends this line of reasoning to critique policies that restrict employers' access to information, such as bans on criminal background checks. He contends that depriving companies of relevant data does not eliminate their underlying incentives to screen applicants, but instead forces them to rely on cruder proxies such as demographic stereotypes. He’s also critical of government organizations that attempt to implement such policies. For example, the Equal Employment Opportunity Commission (EEOC) sued an employer for conducting background checks on the grounds of disparate impact on black people. The EEOC eventually lost the case. Sowell argues that these well-intentioned policies lead to unintended consequences, which ultimately reduce employment opportunities for young black males.
To illustrate, consider a firm hiring for a sensitive role, such as a company treasurer, where the opportunity for theft is significant. The employer wants to ensure that candidates do not have a criminal history, and thus may ask about criminal records on the application. This approach, which Sowell refers to as Discrimination 1-A, allows the employer to make informed decisions based on relevant data about each individual.
However, if background checks were prohibited, the employer would not be able to access this information. Consequently, the employer might resort to making assumptions about which applicants are more likely to have criminal backgrounds based on the limited information available, Discrimination 1-B. Thus, Sowell’s framework predicts that if the government makes the ideal form of discrimination— Discrimination 1-A—illegal, more employers will conduct the nonideal form—Discrimination 1-B.
One example of such a policy is the "Ban the Box" (BTB) initiative, which has been adopted by many states in recent years. BTB policies prevent employers from asking about job applicants' criminal records until later in the hiring process. An article in The Atlantic titled When Banning One Kind of Discrimination Results in Another talks about the effects of these policies. It details many studies where laws were passed to reduce the amount of information employers would have, primarily in the hopes of helping young black males. According to Sowell’s framework, which, again, is not new and is based on Becker’s work from the 1960s and 1970s, these laws would have the opposite effect. Here are the results of the studies referenced in the piece:
Holzer, et al., 2006: “We find that employers who check criminal backgrounds are more likely to hire African American workers, especially men.”
Wozniak, 2014: “Adoption of pro-testing legislation increases black employment in the testing sector by 7-30% and relative wages by 1.4-13.0%, with the largest shifts among low skilled black men.”
Bartik and Nelson, 2016: “We study recent bans on employers' use of credit reports to screen job applicants […] We find that the bans reduced job-finding rates for blacks by 7 to 16 log points.”
Agan and Starr, 2018: “[O]ur results support the concern that BTB policies encourage racial discrimination: the black-white gap in callbacks grew dramatically at companies that removed the box after the policy went into effect. Before BTB, white applicants to employers with the box received 7% more callbacks than similar black applicants, but BTB increased this gap to 43%.”
Doleac and Hansen, 2020: “BTB policies decrease the probability of employment by 3.4 percentage points for young, low-skilled black men.”
In an interview on William F. Buckley’s 'Firing Line', Sowell expresses his exasperation: “I haven't been able to find a single country in the world where the policies that are being advocated for blacks in the United States have lifted any people out of poverty. I've seen many examples around the world of people who began in poverty and ended in affluence. Not one of them has followed any pattern at all like what is being advocated for blacks in the United States.”
Sowell is also particularly critical of the legal concept of 'disparate impact,' which features prominently in anti-discrimination laws and has influenced decisions up to the Supreme Court level. Disparate impact refers to practices or policies that, while seemingly neutral, have a disproportionately adverse effect on a protected group. For example, a company's hiring process that includes a test might be considered to have a disparate impact if it results in a significantly lower percentage of minority candidates being hired, even if the test itself does not explicitly discriminate based on race. Sowell argues that the disparate impact standard effectively reverses the presumption of innocence, placing an undue burden on defendants to prove their innocence rather than on the accuser to prove guilt.
Solutions
For readers seeking neat policy prescriptions that will eliminate disparities, Sowell's work may be disappointing:
Many people may expect discussions of economic and social disparities to end with “solutions”—usually something that the government can create, institutionalize, staff and pay for with the taxpayers’ money.
The goal here is entirely different. There has never been a shortage of people eager to draw up blueprints for running other people’s lives. But any “solution,” however valid as of a given moment under given conditions, is subject to obsolescence at some later time under changed conditions.
The hope here is that clarification is less perishable, and can be applied to both existing issues related to economic and social disparities and to new issues, involving the same subject, that are sure to arise with the passage of time. Given the limitations of prophecy, the point here is to seek to provide enough clarification to enable others to make up their own minds about the inevitable claims and counter-claims sure to arise from those who are promoting their own notions or their own interests.
Sowell does not claim to offer definitive solutions to societal issues, nor does he endorse the goal of achieving complete equality. In an interview, he said, “Nowhere in the world do you find this evenness that people use as a norm, and I find it fascinating that they will hold up as a norm something that has never been seen on this planet, and regard as an anomaly something that is seen in country after country.” So he has no solutions for ending racial disparities, much like he has no solution for ending disparities between the eldest child and their siblings, or the disparity that gay men earn 10% more than straight men.
This is not to say that Sowell believes nothing can or should be done to address the real hardships faced by disadvantaged groups. He acknowledges, for instance, the high costs borne by residents of high-crime neighborhoods and suggests that stronger law enforcement could help improve their situation.
Category Error
Although Sowell’s framework is illuminating, it is not without its limitations. He presents his categories of discrimination as distinct and exhaustive, but the boundaries between these categories are not as well-defined as he suggests. To see how, consider this real-world example told by renowned sociologist William Julius Wilson:
I am an internationally known Harvard professor, yet a number of unforgettable experiences remind me that, as a black male in America looking considerably younger than my age, I am also feared. For example, several times over the years I have stepped into the elevator of my condominium dressed in casual clothes and could immediately tell from the body language of the other residents in the elevator that I made them feel uncomfortable. Were they thinking, “What is this black man doing in this expensive condominium? Are we in any danger?” I once sarcastically said to a nervous elderly couple who hesitated to exit the elevator because we were all getting off on the same floor, “Not to worry, I am a Harvard professor and I have lived in this building for nine years.” When I am dressed casually, I am always a little relieved to step into an empty elevator, but I am not apprehensive if I am wearing a tie.
I get angry each time I have an experience like the encounter in the elevator.
Sowell responds with the following:
Professor Wilson’s sarcasm and anger were directed at people whose reactions reflected a greater concern for their own personal safety than for his sensitivities. His account suggests that they were not racists, for merely by wearing a tie he avoided tensions on both sides, even though wearing a tie did not change his race.
Sowell says they are not racists and this is Discrimination 1-B. While his point is valid to some extent, I think this example highlights a flaw in his model. If someone believes they are relatively more likely to be victimized by a casually dressed black man, and this belief is based on accurate evidence, then it would be classified as Discrimination 1-B. However, if their belief is based on an "unsubstantiated notion," then it would be considered Discrimination 2.
Determining the accuracy of such beliefs is not straightforward. Perhaps the assessment of higher risk is correct if state-wide statistics are used, but incorrect in that particular city. Additionally, arrest rates for young black men, which might be used as evidence, could themselves be the product of bias. Basing the definition of Discrimination 2 on the accuracy or inaccuracy of a belief is misguided.
Either way, there’s an even more important issue at hand. Even if, as Sowell suggests, it’s not Discrimination 2, this doesn’t reduce the impact of the discrimination. Although Discrimination 1-B stems from cold calculus rather than hot animosity, it still erects barriers to opportunity and inclusion for the disfavored group. If there is a widespread belief that young black males are more likely to commit crimes, regardless of whether this belief is accurate or not, it places a tremendous burden on them. They’ll encounter barriers in securing employment, housing, or educational opportunities as well as psychological suffering—not due to their personal qualifications or behaviors, but simply because of their physical characteristics. Moreover, these effects accumulate over time, further compounding their impact.
Michelle Alexander vividly depicts this dynamic in The New Jim Crow, chronicling how the cycle of arrest and re-arrest perpetuates what she calls a “racial caste system”, where laws, policies, and enforcement practices that appear colorblind actually have profoundly racial outcomes. Once labeled as felons, these individuals face extensive discrimination in employment, housing, voting, and education, essentially hindering their abilities to reintegrate into society.
These examples illustrate how both Discrimination 1-B and Discrimination 2 can create self-reinforcing patterns of inequality. Even if the discrimination is all Discrimination 1-B (and I’m not saying it is, just as a hypothetical), the cumulative effect can be to trap entire groups in a cycle of poverty and marginalization.
This is not something the market will sort out. This problem has persisted for decades, and there is no evidence to suggest that it will disappear in the near future. The pervasive nature of this issue is exemplified by a statement made by civil rights activist Jesse Jackson over 30 years ago: “There is nothing more painful to me at this stage in my life than to walk down the street and hear footsteps and start thinking about robbery—then look around and see somebody white and feel relieved.”
To be clear, Sowell never claimed that market forces alone would resolve these issues. However, given the magnitude of the problem, it certainly seems like something he could have discussed more extensively. Indeed, while Sowell offers compelling analyses of disparities in income and other economic outcomes, his book could have been improved by also exploring the disparities in incarceration rates.
Successful Interventions?
Although Sowell's focus is on the limitations and unintended consequences of government intervention, it still would have been helpful to learn more about successful interventions. No matter how clearly he says he’s not providing solutions, I’m still left wondering, “Isn’t there something the government can create, institutionalize, staff, and pay for with the taxpayers’ money, that would fix this problem?” Even after reading this book, the desire doesn’t go away.
It would be an exaggeration to say that Sowell doesn’t think there’s anything the government can do, but if you exclude tasks like, “undo the laws that made everything worse,” it doesn’t seem like much of an exaggeration. Unless I missed it, Sowell doesn’t provide a single example of an intervention gone well.
I get it, big government interventions are not his favorite things. But if he had to choose one medium-sized government intervention, which would it be? Isn’t there some instance where government action has achieved its intended goals without overwhelming drawbacks? Discussing such examples would have added to his analysis.
Even his framework seems to suggest that there should be some ways to truly benefit the intended groups. For example, aren’t there ways to help young black men counter the forces of Discrimination 1-B? Even if the cost is significant, the long-term benefits of breaking these cycles justify the investment.
One potential solution to address the effects of Discrimination 1-B on young black males and other affected groups could be to develop a system that provides credible signals of an individual's employability. This could involve a thorough character assessment conducted by a respected authority figure or institution, which would vouch for the person's suitability for employment, despite any superficial similarities to stereotyped groups. The assessment could be summarized in a statement along the lines of, "I have closely examined this individual's character and can attest that, despite sharing some characteristics with groups that face statistical discrimination, this person possesses the qualities and skills to be a valuable employee."
Such endorsements could come from various sources, including schools, churches, or community organizations, but the government could also play a role in establishing or supporting these programs. Similarly, for individuals with criminal records, a formal rehabilitation certification program could be developed, with a statement like, "Based on a rigorous evaluation process, I certify that this person has demonstrated genuine rehabilitation and is ready to reenter the workforce." These initiatives would aim to provide employers with reliable, individualized information about job candidates, helping to counteract the effects of group-level discrimination and opening up more opportunities for those who might otherwise be unfairly excluded from consideration.
I don’t want to imply that Sowell thinks nothing can be done. In the above case, my guess is he would propose getting rid of minimum wage laws, so employers would be able to hire them at lower cost. Those who proved to be valuable employees could demand higher wages before long.
On Proofs and Peer Review
While Sowell's early works were aimed at an academic audience, he has since largely eschewed publishing in scholarly journals, opting instead to write books and articles intended for a general readership. This choice has likely enabled him to reach a larger audience and, arguably, exert a greater influence on public discourse. I'm not sure why Sowell has never published much in academic journals, but I suspect it's because he doesn't think very highly of them.
That said, Sowell's decision to operate primarily outside of traditional academic channels leaves him open to criticism, and some readers may remain unconvinced by his arguments. The validity of such critiques is worth considering, and individuals must assess for themselves how significant they think this is. Had Sowell been able to draw upon a robust body of peer-reviewed research employing sophisticated statistical methods, such as instrumental variables estimation, to disentangle causal relationships from mere correlations, it could have fortified his arguments. While this additional layer of academic rigor still might not have persuaded everyone, it would have strengthened his position.
However, this is Sowell we’re talking about, and he would be disappointed if we didn’t mention tradeoffs—dedicating more time and energy to navigating the academic publishing process would have inevitably meant producing fewer interesting, accessible books for us.
Although I found his arguments largely compelling, the analysis of the changing labor force participation rates among young black men as they age struck me as providing only moderate evidence against the predominance of Discrimination 2, certainly not enough to definitively exclude its role in shaping disparities. Ultimately, readers must carefully consider the empirical evidence Sowell presents while remaining mindful of the inherent challenges of studying complex societal phenomena outside the controlled confines of a laboratory setting.
A Tough Person?
Critics sometimes paint Sowell as callous or indifferent to the struggles of disadvantaged black communities. But this charge is belied by the depth and intensity of his lifelong engagement with these issues. Far from ignoring the plight of the impoverished, Sowell has devoted his career to meticulously documenting the complex realities of racial disparities and tracing their tangled historical roots. His rigorous, empirically grounded analysis asks hard questions and follows the data where they lead, even when the answers prove unwelcome to ideologues of various stripes. That Sowell's conclusions often run afoul of conventional pieties testifies not to his lack of concern for the disadvantaged, but to his overriding commitment to pursuing the truth, however discomfiting it may be.
For Sowell, grappling honestly with the world as it is must precede fashioning effective solutions to deep-seated social ills. As he puts it, "Misdiagnosing the basis for discrimination produces more than a difference in words. It can produce policies less likely to achieve their goals, or even policies that make matters worse."
In an interview featured in the documentary Thomas Sowell: Common Sense in a Senseless World, produced by the Free to Choose Network, Sowell offers a characteristically incisive perspective on the ideological landscape: "If the world were the way the left conceives it to be, it would be a better world than the way the right conceives it to be. It just happens that the world is not that way."
Yet Sowell's critique of prevailing orthodoxies is often misconstrued as a call for pure self-reliance or an abdication of collective responsibility. In his essay "Blacks and Bootstraps," he confronts this mischaracterization head-on:
One of the things I have been falsely accused of many times over the years is advising blacks to lift themselves up by their own bootstraps. But you can look through the 21 books, dozens of articles and hundreds of newspaper columns I have written without finding any such statement. That is because I am not in the business of giving advice to individuals and groups, but rather in the business of discussing public policy and trying to show where one policy is better than another.
Sowell's historical analysis, however, does highlight the substantial progress made by black Americans before the advent of major civil rights legislation and affirmative action policies. As he notes, the poverty rate among black families fell from a staggering 87 percent in 1940 to 47 percent by 1960, a trend that continued through the 1960s.
Ultimately, Sowell's worldview is perhaps best encapsulated by his observation:
People say 'You're a very tough person.' I'm not tough. Life is tough. I'm merely trying to acquaint you with those facts.
Sowell’s Constrained Vision
In Discrimination and Disparities, Thomas Sowell offers a thought-provoking and empirically grounded analysis of the complex factors shaping group differences in modern societies. By challenging simplistic narratives that attribute disparities solely to discrimination, Sowell pushes readers to grapple with the multifaceted realities of unequal outcomes in a world that has never been a level playing field.
Perhaps the best way to understand Sowell's position is through his book A Conflict of Visions, where he describes two contrasting worldviews: the "constrained" and "unconstrained" visions. The “constrained vision” sees human nature as largely unchangeable and fundamentally flawed. This view, shared by thinkers like Burke, Hayek, and Friedman, suggests that the best policies work with human tendencies rather than trying to reshape them. In contrast, the "unconstrained vision," held by figures like Rousseau, Marx, and Mao, sees human nature as malleable and perfectible, leading to a belief in the possibility of near-utopias through the right laws and societal structures.
Sowell's approach to analyzing group disparities is firmly grounded in the constrained vision. His aim is not to provide a blueprint for perfect equality—a state of affairs that has never existed in any society—but rather to equip readers with a framework for critically examining the causes and remedies of group differences. By clarifying the complex realities underlying disparities, he seeks to promote clearer thinking about the tradeoffs and potential unintended consequences of proposed policies.
However, translating Sowell's insights into an actionable political agenda remains a daunting challenge. People say 'It takes a theory to beat a theory,' but he doesn't have a simple, overarching explanation for why any particular group is at the top or bottom of society. His constrained vision sits uneasily with the current fervor for bold government initiatives to reduce inequality, and his critiques of well-intentioned but misguided policies can be misinterpreted as a defense of the status quo. Moreover, in a society where concepts like systemic racism and intersectionality are widely accepted as explanations for disparities, Sowell's more nuanced perspective can be difficult to sell. For many people, systemic racism can explain everything, yet is required to predict nothing, and therefore can’t be falsified.
Not only does it take a theory to beat a theory, but it also takes a solution to beat a solution, and Sowell offers even less here. In fact, given his perspective, it’s not even clear what a “solution” for disparities would look like. Would that be a state where, no matter how government bureaucrats diced up the population, there would never be a disparity? This is not a goal that he would endorse. This makes his constrained vision a difficult sell, especially when compared to diversity, equity, and inclusion (DEI) initiatives, which are seen as concrete solutions, whether or not they work.
Nonetheless, engaging seriously with Sowell's analysis need not mean abandoning the moral imperative to expand opportunity and inclusion. On the contrary, it demands a redoubled commitment to evidence-based initiatives that break down barriers, build capabilities, and equip individuals to transcend the limits of statistical discrimination. While Sowell's work may not offer a simple, all-encompassing solution to inequality, it provides an essential foundation for grappling with the stubborn realities of group disparities in a diverse society.
For those seeking a foundation for grappling with the realities of group disparities in a diverse society, Sowell's oeuvre remains indispensable. Policymakers and citizens who absorb the lessons of Sowell's work will be better equipped to craft solutions that meet the world as it is, not as we might wish it to be. Doing so with wisdom and compassion will be a great challenge, but one that we must undertake if we hope to build a more just and inclusive society. By embracing the complexity and nuance of Sowell's analysis, we can move beyond simplistic narratives and work towards policies that genuinely improve the lives of those most affected by persistent disparities.