Wealth Accumulation and Communities of Color in the United States: Current Issues

by Jessica Gordon Nembhard Ph.D., Ngina S. Chiteji

Overview

"Congratulations to Drs. Nembhard and Chiteji and the authors included in this much needed volume of work! Their book offers the perspective and insight of scholars of color that are too often missing from information produced by the asset building field (people and organizations seeking to help low-income people develop assets). Communities served by the asset building field are disproportionately made up of people of color. This book captures work produced by scholars representing these communities and offers innovative and thought provoking analyses of wealth inequality. Decision-making on research, policy, and practice that fails to incorporate the knowledge of these and other asset accumulation experts of color runs the risk of being fatally flawed and irrelevant to the communities the asset building field intends to serve."
—Kilolo Kijakazi, Ph.D., The Ford Foundation

"An important contribution to the economics literature on wealth and to our understanding of racial and ethnic inequality. This book adds to our knowledge and understanding of the wealth positions of Latinos, Asian Americans, Hawaiians, and Native Americans and places this information in the context of black-white wealth inequality."
—Cecilia A. Conrad, Department of Economics, Pomona College

"This book does an outstanding job of introducing readers to a host of interesting questions related to racial and ethnic minority status and wealth composition and accumulation. The chapters on wealth accumulation among Native Americans, Latinos, and Asian Americans offer one of the few places where this information is readily available. The recent disaster in New Orleans has shown the nation that there is a strong interaction between wealth, race, and social outcomes. This book not only fills a void in understanding the black-white wealth inequality that was apparent after Hurricane Katrina, but it also provides great insight into the wealth status of other racial and ethnic minorities."
—Patrick L. Mason, Department of Economics, Florida State University

"This edited volume takes up an important, indeed, fundamental, topic, bringing together leading scholars to assess wealth accumulation among people of color. No other book or research report covers as many groups of color as appear in this volume, devoting chapters to African Americans, Latinos, Native Americans, Asian Americans, and Native Hawaiians. The result is a noteworthy achievement." —Michael Sherraden, Benjamin E. Youngdahl Professor of Social Development, Washington University in St. Louis





Jessica Gordon Nembhard is Assistant Professor and Economist, African American Studies Department, and co-founder of the Democracy Collaborative at the University of Maryland, College Park. Her work on the history of black cooperatives is well known in progressive circles.




Ngina Chiteji is Associate Professor of Economics, Skidmore College. She was a Visiting Assistant Research Scholar at The Democracy Collaborative, University of Maryland, College Park.

Product Details

ISBN-13:	9780472069583
Publisher:	University of Michigan Press
Publication date:	11/06/2006
Edition description:	New Edition
Pages:	376
Product dimensions:	6.00(w) x 8.90(h) x 1.00(d)

About the Author

Jessica Gordon Nembhard is Assistant Professor and Economist, African American Studies Department, and co-founder of the Democracy Collaborative at the University of Maryland, College Park. Her work on the history of black cooperatives is well known in progressive circles.

Ngina Chiteji is Associate Professor of Economics, Skidmore College. She was a Visiting Assistant Research Scholar at The Democracy Collaborative, University of Maryland, College Park.

Read an Excerpt

Wealth Accumulation & Communities of Color in the United States

CURRENT ISSUES

---

THE UNIVERSITY OF MICHIGAN PRESS

Copyright © 2006 University of Michigan
All right reserved.
ISBN: 978-0-472-09958-0

---

Chapter One

Wealth Measurement

Issues for People of Color in the United States

WILHELMINA A. LEIGH

Although income has long been used to compare the well-being of populations in this country, only since the mid-1960s (toward the end of the civil rights movement) has wealth been used for this purpose as well. The 1967 Survey of Economic Opportunity was the first government survey that allowed an examination of differences in wealth by race. Comparing Blacks and Whites only, the Survey of Economic Opportunity revealed marked disparities that persisted during the remainder of the twentieth century. In 1967 the ratio of mean (or average) Black household wealth, or net worth ($16,972), to mean White household wealth, or net worth ($90,507), was 0.188 (18.8 percent). Net worth for the average Black household had risen to $31,992 in 1984 with the comparable figure for White households at $123,677, for a ratio of 0.259 (25.9 percent). Nearly ten years later, in 1993, the ratio remained virtually unchanged at 0.26 (26 percent), with mean Black household wealth, or net worth, at $34,270 and mean White household wealth, or net worth, at $131,613 (Eller and Fraser 1995; Oliver and Shapiro 1989; Terrell 1971).

In 1998, the ratio of mean wealth for Black non-Latino households to mean wealth for White non-Latino households had fallen to 0.18 (18 percent) again. The comparable ratio for Latino household wealth vis-à-vis wealth for White non-Latino households was 0.25 (25 percent) (Choudhury 2001/2002). In that same year, however, mean wealth for Asian American households was 0.87 (87 percent) of the value held by White households (Wolff 2001).

Explanations for these persistent gaps have included, in particular for the Black-White disparity, the intergenerational consequences of the fact that in 1863 each emancipated slave did not receive the promised "forty acres and a mule." Advocacy for a massive capital transfer or reparations has developed around this historical oversight (Browne 1974; America 1990; Oliver and Shapiro 1995; Ogletree 2004). Limited inheritance of wealth by people of color also is cited to explain this gap (Blau and Graham 1990; Oliver and Shapiro 1995). Another often-suggested explanation is the failure of Blacks (or African Americans), other people of color (i.e., Asian Americans, Native Hawaiians and other Pacific Islanders, American Indians/Alaska Natives, and Hispanics [or Latinos]), and immigrants to invest in the stock market, a shortcoming attributed in part to the lack of familiarity with this market and to perceptions of its risk (Brimmer 1988; Wynter 1997). Yet another suggested explanation is the use of measures of wealth that exclude employer pensions and Social Security benefits, both of which are large proportions of the wealth of people of color (Smith 1995).

This chapter provides measures of wealth and details about the distribution of wealth by race/ethnicity to shed light on these summary statistics and the postulated explanations for them. The first section includes measures of, and notes critical distinctions among, these rubrics. Surveys that collect the most commonly cited and readily accessible wealth data also are described in this section. The second section discusses wealth by race/ethnicity-both the frequency with which assets are held and the value of assets. Special emphasis is placed on housing, one of the most frequently held assets and the asset with the greatest value across racial/ethnic groups. The third major section raises issues that arise when measuring wealth and collecting data about it for people of color. The chapter concludes with a summary of its findings.

WEALTH MEASUREMENT

HOW WEALTH IS REPORTED

Gross personal wealth is reported as the gross value of all assets before reduction by the amount of debts. Net personal wealth (or net worth) is the level of wealth (or worth) after all debts have been removed from gross personal wealth (U.S. Census Bureau 1990). Wealth is commonly reported either aggregated for selected population subgroups or per household for each of these groups. Total, or aggregate, net worth for a population group is the sum of the market value of all the assets owned by every member of every household belonging to this group minus the value of the corresponding household debts, or liabilities, both secured and unsecured (Orzechowski and Sepielli 2003).

Aggregate net worth for any population group also can be thought of as the product of multiplying the number of households by the average, or mean, net worth. Put another way, the average, or mean, net worth of a population group is obtained by dividing the total net worth of that group by the number of households in the group. Because of the way it is calculated, mean net worth is affected by all the actually observed values, including very low and very high net worth (known as outliers). Median net worth, on the other hand, is the central measure in a distribution of values (i.e., the dollar value that divides all households in a population into two equal-sized groups, one with net worth less than that amount and the other with net worth above that amount). Although mean net worth and median net worth both are used to reflect the wealth of a typical family or household, median net worth is a more appropriate measure for distributions that are skewed. If wealth were evenly distributed across a population, the average, or mean, net worth and the median net worth would be equal, and either could be used. Because wealth distributions generally are skewed, median wealth, or net worth, is reported more frequently in the published research that is the source for this chapter. Means also are provided, however, for comparison since in the social sciences the mean is the most commonly used measure of the central tendency of a distribution of numbers.

HOW WEALTH IS HELD

Social scientists are interested not only in the total amount of wealth held but also in the components of wealth holdings. Net worth, or wealth, is held in a variety of forms, and this mix of wealth holdings is referred to as portfolio composition or portfolio allocation. In addition to the level and mix of wealth holdings, rates of ownership are reported for assets held. These asset-ownership rates are the percentages of a population that hold any given asset (such as a bank account or a motor vehicle).

Within a wealth portfolio, the assets are commonly distinguished as either housing equity or nonhousing equity (Choudhury 2001/2002). This distinction is made because housing is the major source of wealth for all households in the United States (Caner and Wolff 2004). In addition, nonhousing equity is often characterized as both financial assets and tangible assets, with the major distinction between the two being the degree of liquidity. Financial assets are generally more liquid (i.e., can be converted to cash more easily and quickly) than tangible assets. Commonly used categories of financial and tangible assets are described here:

A. Financial Assets 1. Liquid assets, such as interest-earning assets at financial institutions (e.g., passbook savings accounts, money market deposit accounts, certificates of deposit, and interest-earning checking accounts). Other accounts at financial institutions that are not interest bearing, such as regular checking accounts, also are liquid assets.

2. Stocks and mutual fund shares

3. Bonds-municipal and corporate bonds, as well as U.S. savings bonds

4. IRAs (Individual Retirement Accounts) or Keogh plans

5. Other interest-earning assets, such as money market funds and U.S. government securities, as well as mortgages held from sales of real estate and amounts due from the sale of a business, unit trust, or other financial investments

B. Tangible Assets

1. Equity in motor vehicles

2. Equity in business or profession

3. Other assets (e.g., rental property equity and other real estate equity)

Although some surveys (e.g., the Consumer Expenditure Survey and the Survey of Consumer Finances) collect data on other forms of wealth, such as pensions, these data are not collected in all wealth surveys. Thus, the major forms in which employers provide pensions today (i.e., 401[k], 403[b], and thrift plans) are not included in the preceding list of financial assets. (See discussion of various wealth surveys in this chapter's appendix and in the following section.)

The assets that constitute wealth could alternatively be categorized as either consumable or income producing (Oliver and Shapiro 1989). Consumable assets do not necessarily generate income or future wealth. They include home equity and equity in motor vehicles. Income-producing assets, as the label implies, generate income and potentially generate future wealth. These assets include such items as liquid financial assets, rental property, and net business assets (O'Hare 1983). Net business assets include equity in business or profession, other real estate equity, and possibly "other assets." By either categorization (i.e., housing versus nonhousing equity, or consumable versus income-producing assets), the mix of assets comprising wealth determines likely future wealth accumulation.

WHAT ARE THE MAIN SOURCES OF WEALTH DATA?

The main sources of wealth data for the United States are surveys that have been conducted over time with different populations and differing sample sizes and including somewhat different asset measures. The surveys whose wealth data are most frequently analyzed in published literature include the Consumer Expenditure Survey (CEX), Health and Retirement Study (HRS), National Longitudinal Surveys (NLS), Panel Study of Income Dynamics (PSID), Survey of Consumer Finances (SCF), and Survey of Income and Program Participation (SIPP). Although only one of these (SCF) is designed to be a wealth survey, both the SCF and the PSID are the most widely used data sets for scholarly research about wealth. Each study/survey is described briefly in the appendix to this chapter. A comparison of the surveys follows.

The most commonly used sources of wealth data (CEX, HRS, NLS, PSID, SCF, and SIPP) differ from one another both in the characteristics of the populations from which they collect data and in the type of wealth and other data collected. Thus, for analytical purposes, the choice among wealth survey data sets should be made on the basis of the questions to be addressed. For example, HRS data are most useful for analyzing the wealth holdings of older populations (i.e., those born in 1947 or earlier). The NLS data enable analyses of both mature populations and younger populations (e.g., older men, young men, mature women, and young women). The nationally representative sample surveys (CEX, PSID, SCF, and SIPP) are the best sources for data about the nation as a whole. However, all these surveys tend to underestimate populations difficult to locate such as immigrants and young males, and these frequently excluded populations are likely to include sizable proportions of people of color (Oliver and Shapiro 1995).

The geographic residence of persons eligible to be sampled for the major wealth surveys varies considerably. For example, several surveys (CEX, HRS, and SIPP) exclude from their sampling universe residents of places outside the contiguous United States, such as the Commonwealth of Puerto Rico, U.S. Virgin Islands, and Guam. The HRS draws its samples only from within the forty-eight contiguous states and, thereby, excludes the many people of color in Alaska and Hawai'i and in the U.S. territories. Although the CEX includes all fifty states and the District of Columbia, it excludes the offshore territories from its sampling. The SIPP does not include residents from the Commonwealth of Puerto Rico in its samples. Clearly these sampling design decisions limit the possibility that members of the many diverse subpopulations of color will be represented in the wealth data collected by these surveys.

In the face of geographic sampling decisions that exclude selected populations of color a priori, the major wealth surveys employ various techniques to ensure representation of racial/ethnic subpopulations in their samples. The HRS oversamples African Americans and Hispanics in the forty-eight contiguous states. By oversampling Floridians as well, this survey captures many immigrants and other people of color. During its many years of implementation, the PSID also has oversampled not only for African Americans and Latinos but also for immigrant populations (including Asians), even though these sample sizes are still statistically very small. The SIPP indirectly oversamples racial/ethnic subpopulations because it oversamples for low-income individuals (among whom racial/ ethnic subpopulations are overrepresented). In direct opposition to this, however, the SCF oversamples the White population because it oversamples higher-income individuals (who are more likely to be White) in its effort to capture information about all the major forms of wealth. Based on their sampling, the PSID and SIPP would be expected to be the most robust sources of wealth data for people of color (although there are limitations with the use of these data as well).

Differences in the characteristics of the wealth data collected emanate primarily from differences in the purposes of the surveys. For example, the CEX collects data about expenditures, both consumable and durable. Like the CEX, the SIPP and the PSID were not designed primarily to be wealth surveys and, therefore, lack detail about the wealth holdings of persons in the upper tail of the U.S. wealth distribution. Surveys designed primarily to catalog wealth holdings (such as HRS and SCF) include a more comprehensive set of categories for these holdings and sample in such a way as to collect data about wealth in all of these categories. The PSID, for example, reports data for a small fraction of the categories of wealth for which SCF data are provided.

Another difference in the wealth data collected arises from the treatment of nonresponses. The use of bracketed follow-up queries with nonresponses has enabled both the HRS and PSID to convert between 75 percent and 80 percent of nonresponses into categorical responses. The SIPP does not use any bracketed questions, the SCF seldom uses them, and the CEX Arst used them in 2001 (Consumer Expenditure Survey 2003). Thus, the HRS has a much lower conditional nonresponse rate (9 percent) for stock values than other surveys (e.g., the nonresponse rates in SIPP and in the SCF were 42 percent and 26 percent, respectively) (Smith 1995). Table 1.1 summarizes the advantages and disadvantages of the six major data sets.

All of these differences in the populations from which data are collected and in the type of data collected are reflected in the reported values of net worth. If one compares the SCF with the PSID and the SIPP, for example, net worth measures differ for several reasons (Czajka, Jacobson, and Cody 2003). As mentioned previously, median net worth and mean net worth based on data from any single survey conducted in the United States are expected to differ because of the marked, unequal distribution of wealth in this country. However, mean net worth differs for proximate years of data collection from the SCF and from the PSID and the SIPP, with the SCF measure exceeding measures from the other two surveys because of differences in design and implementation. (See tables 1.2a and 1.2b.) The SCF primarily is intended to measure family wealth and thus captures asset holdings more comprehensively than the SIPP and the PSID. Median net worth from the SCF also exceeds these measures for proximate years of PSID and SIPP data. (See tables 1.3a and 1.3b.)

(Continues...)

---

Excerpted from Wealth Accumulation & Communities of Color in the United States Copyright © 2006 by University of Michigan . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---

Table of Contents

Contents

LIST OF FIGURES....................v
LIST OF TABLES....................vii
ACKNOWLEDGMENTS....................xi
INTRODUCTION & OVERVIEW JESSICA GORDON NEMBHARD & NGINA CHITEJI....................1
1. WEALTH MEASUREMENT Issues for People of Color in the United States WILHELMINA A. LEIGH....................23
2. ESTIMATING THE EFFECT OF RACE & ETHNICITY ON WEALTH ACCUMULATION & ASSET-OWNERSHIP PATTERNS NGINA CHITEJI & DARRICK HAMILTON....................67
3. DETERMINANTS OF INTRAGROUP WEALTH INEQUALITY AMONG WHITES, BLACKS, & LATINOS YUVAL ELMELECH....................91
4. WOMEN & WEALTH MARIKO LIN CHANG....................112
5. NATIVE AMERICANS' WEALTH JAY L. ZAGORSKY....................133
6. TROUBLE IN PARADISE The Economic Marginalization of Native Hawaiians PAUL ONG....................155
7. ASIAN AMERICANS & WEALTH PAUL ONG & R. VARISA PATRAPORN....................173
8. FINANCIAL MARKETPLACE PARTICIPATION & PENSION HOLDINGS OVER THE LIFE COURSE NGINA CHITEJI, ELENA GOUSKOVA, & FRANK STAFFORD....................191
9. BANKING & WEALTH ACCUMULATION IN THE ASIAN AMERICAN COMMUNITY Questions & Evidence from Los Angeles GARY DYMSKI, LISA MOHANTY, & WEI LI....................219
10. WEALTH CREATION IN LATINO COMMUNITIES Latino Families, Community Assets, & Cultural Capital BÁRBARA J. ROBLES....................241
11. LIVING WHERE THE NEIGHBORS ARE INVESTED Wealth & Racial/Ethnic Differences in Individuals' Neighborhood Home Ownership Rates RACHAEL A. WOLDOFF....................267
12. WEALTH, CIVIC ENGAGEMENT, & DEMOCRATIC PRACTICE JESSICA GORDON NEMBHARD & ANTHONY A. BLASINGAME....................294
AFTERWORD Trends & Trappings, Research & Policy Implications: An Unorthodox Policy Guide JESSICA GORDON NEMBHARD....................326
ABOUT THE AUTHORS....................343
INDEX....................349