ECO 450G Notes

Fall 2006 Semester

University of Kentucky

© Professor Aaron Yelowitz

Chapter 8 – The Underclass: Culture and Race

I. The Culture of Poverty

II. Wilson’s Underclass Theory

III. The Racial Inferiority Theory

I. The Culture of Poverty

When describing the origins of poverty, debate has focused on whether an underclass exists that is poor and socially isolated, so that it doesn’t respond to mainstream prosperity, incentives or values.

This chapter explores issues about the existence and explanations for an underclass.

Many Americans believe there is abundant economic opportunity, hence the persistence of poverty relies on “flawed character” arguments such as welfare mothers preferring to have kids over jobs, etc.

Page 141 – “Young men would rather ‘hustle’ on street corners than accept a regular job. ... Because they are trapped by this culture ... (they) don’t behave in ways that would help them escape from poverty.” The Freakonomics book explains in detail the economics of drug-dealers. One of the key points of the chapter in Freakonomics is that drug-dealing is a profession where the rewards are concentrated at the very top, and young men may simply be responding to those incentives.

Norms Versus Traits

A key issue in determining “flawed character” is that we do not directly observe preferences (such as a person’s work ethic), or as Schiller calls it, “culture.” Instead we observe behaviors, such as lack of work or low savings. Observed behaviors are combination of different preferences and different budget constraints.

Example:

A welfare recipient may be observed to reject a “reasonable” job offer.

● One conclusion relates to preferences: she is lazy.

● Another relates to the budget constraint: Childcare and transportation responsibilities relative to the wages from the job, or the lack of advancement at certain jobs.

● A final one relates to transfers: Welfare benefits could be reduced with higher earnings.

Schiller says (p. 142) “It must be shown that the norms and aspirations – not just behavior – of the poor are different and that these differences impede the escape from poverty.”

● Since preferences are unobserved, then it will be difficult to make conclusions like this.

● Simply because something is difficult to prove (or disprove) – such as differences in preferences or differences in innate intelligence – doesn’t mean that we should necessarily discount it.

● The policy implications are vastly different if poverty results from differences in preferences or the budget constraint.

Anthropological Studies

Schiller cites a 40-year-old study by Oscar Lewis, who examines Mexico and Puerto Rico, that argues that there are a number of behavioral traits that explain poverty, such as the non-use of banks, museums, etc.

Others focus on the alleged self-indulgence of the poor. It is observed that middle class people feel the need and desire to save, where the poor do not have such motivation. This also shows up, perhaps, in investment toward children.

● How would you try to learn the answer to these questions?

A Question of Opportunities

Does observed behavior relates to preferences or budget constraints? Schiller lays out several conditions for preferences to drive the behavior:

● The satisfaction being deferred must be equally important to the poor and non-poor. For example, the time horizons must be equal (life expectancy & retirement savings). This implies the monetary benefits from “good” behavior will be the same.

● Equal opportunity to defer satisfaction. For example, equal access to universities, retirement planning, or financial institutions.

● The poor and non-poor must equally suffer from deferment. For example, going to school is equally as expensive.

● The probability of obtaining gratification at the end of the deferment period must be equal for both groups

Essentially, the entire budget constraint must be identical in order to attribute differences in actions to preferences. By definition, the poor and non-poor face difference budget constraints because income is different.

The requirement to take a “good” action in a typical cost-benefit analysis are less stringent than what Schiller lays out. A typical cost-benefit analysis assumes that one takes an action (such as investing in education) if the net present value is positive. In economics we worry about the net benefits, e.g., the benefits minus the costs. There is not an assumption that each of the individual components has to be the same across the poor and non-poor.

● Schiller’s schooling example: In higher education, the costs for the poor are lower because of financial aid. The benefits may be lower too.

● Savings: Consider the intertemporal budget constraint. Are there ways of backing out preferences versus opportunities?

Schiller – “there is no foundation for inferring cultural inadequacies on the basis of observed differences in savings.”

II. Wilson’s Underclass Theory

William Julius Wilson has emphasized external causes that lead to an underclass - mainly restricted opportunities rather than flawed character.

As the U.S. economy has evolved, “good jobs” have moved away from the inner city (where many poor, primarily black families live). This is the “spatial mismatch” hypothesis – jobs are not where the poor are. The relatively advantaged families have left the inner city in favor of the suburbs, increasing social isolation.

● A natural question is why the labor market doesn’t adjust. If jobs are scarce in the inner city, market wages should fall, which in turn will attract employers there.

Testing the Spatial Mismatch Theory

How large is the underclass? Since the definition of who is really the “underclass” is open to debate, the size of the “underclass” varies tremendously. More stringent definitions focus on long-term poor who exhibit deviant behavior, more liberal definitions focus on anyone living in neighborhoods with some fraction of households who are poor.

Direct and Indirect Tests of Aspirations

Direct tests rely on direct questioning about aspirations versus expectations

● Table 8.1 shows that many (70%) of welfare mothers had white-collar aspirations for their oldest child, while a smaller fraction had expectations that their child would achieve this.

● When poor and nonpoor families of the same ethnic background were asked about aspirations, no significant differences were found.

● This led Leonard Goodwin to conclude that “welfare recipients do not differ markedly from other Americans with respect to general personality characteristics or with respect to the work ethic and basic life goals.”

What are the problems with kind of direct questioning?

● Still difficult to separate out preferences, because the stated responses are costless to give. Who would really want to admit that they are lazy or unmotivated?

● Are there more objective questions? For example, whether the person plays the lottery or expectations about their child being a professional athlete?

In environmental economics, there are similar problems with “contingent valuation” studies.

Indirect tests examine different groups, and see if they change their behavior when environmental circumstances change. If their behavior changes, this is suggestive that opportunities, rather than preferences, are different.

● Savings budget constraint: Should behavior change if preferences are different? Yes – differences in preferences do not imply complete irresponsiveness.

● Schiller asserts that “the poor have demonstrated a marked ability to move out of poverty when economic opportunities have improved. ... opportunities are seized when they arise ... Such evidence is consistent with the notion that structural forces rather than cultural value systems are the decisive factors in shaping underclass behavioral patterns.”

The time-series trends for poverty does not change dramatically when the economy improves, however. Even the introduction to this chapter says “Even during the strong economic expansions of the 1980s and 1990s the poverty population shrank only modestly.”

Assessment

Even those who advocate the poverty culture do not argue that 100% of behavior (or lack of it) is due to preferences, or that it applies to all poor.

III. The Racial Inferiority Theory

The culture of poverty hypothesis often has distinct racial overtones. There are clear differences in income and poverty rates across races, but it is unclear whether this is

because of “flawed character” (e.g. culture of poverty) versus “restricted opportunities.”

● The President’s National Advisory Commission on Civil Disorders found that whites favored the racial explanation by 3-to-1. Less than 1-of-6 whites thought discrimination was a serious problem.

Discussion oftentimes leads to the relationship between race and intelligence – which is charged, has undertones of racism, and oftentimes fruitless.

● The debate was recharged in 1994 when Richard Hernstein and Charles Murray, in The Bell Curve, argue that there are differences in innate intelligence across race, represented by a test score known as the AFQT score.

● Innate intelligence is unobserved – what we observe is a combination of innate ability and actual investment. The same could be made for physical abilities.

Relationship between intelligence and economic circumstances seems to be strong: Some twin studies, where the genetic component is identical, have shown that even twins separated at birth tend to end up in similar socioeconomic circumstances.

IQ Scores

The “normal” score for an average person on an IQ test is 100. Many studies have shown that blacks tend to score worse on IQ tests than whites, often by substantial amount – 15 to 20 points.

● IQ tests, and the AFQT tests used in “The Bell Curve,” represent a combination of genetic intelligence and environmental experience. There is no easy way to measure the genetic component.

● IQ scores get much closer between children of different races once socioeconomic status is controlled for.

● IQ scores have risen for 50 years – which suggests it measures more than genetic endowment, unless we believe that genetics are improving.

Even if IQ scores do not measure innate ability, they might better measure cognitive ability – that is abilities that are likely to affect one’s ability to succeed in the marketplace.

● Many colleges discuss changing entrance requirements to reduce the importance of the SAT or ACT – arguing not only that the variation represents cultural differences, but that the scores are uncorrelated with college success or success in the market place. That is, some college administrators argue that test scores like these do not even measure cognitive ability.

● Some students, often the ones doing poorly in a class, will argue that test scores do not measure their true understanding of the material.

Schiller, page 153: “The controlled IQ difference between blacks and whites, however, is only five to ten points. To suggest that this relatively small difference could account for an existing income disparity of over $20,000 a year would be extremely tenuous. One would then be arguing that a 5 to 10 percent difference in intelligence could account for a 40 percent disparity in income.”

● Schiller is not being forthcoming about statistical analysis. The word “controlled” means that the IQ scores are adjusted for other factors like family structure, geography, external opportunities, etc. These environmental factors affect IQ scores, but there still appears to be a racial difference.

● If you’re using “controlled” IQ scores, you should control for those same factors in income. Schiller uses the raw difference in incomes. It is not surprising that the role of intelligence would then be implausibly high. It is clear that factors like family structure and geography also affect incomes, and the income differences would converge, too, if these factors were controlled for.

Other Complications

Median family incomes for blacks (relative to whites) vary by geography, which means the differences in income are not totally driven by genetics.

● But is the argument about genetics being the sole factor?

Definition of “race”: There are anywhere between 3 and 30 definitions of race, and separating race from ethnicity, nationality, and religion is sometimes difficult.