Descriptive Title: Income inequality

Geographic Unit of Analysis: County

80/20 income inequality percentile ratio = $114,119/$21,175 = 5.4.
The distribution of income in San Francisco is illustrated in the graph below.

Income Inequality in the Bay Area (2006-2010)

County

Gini**

Gini MOE*

80/20 ***

80/20 MOE*

20th percentile

20th MOE*

80th percentile

80th MOE*

Alameda County

0.46

0.00

5.15

0.10

$26,669

$394

$137,449

$1,306

Contra Costa County

0.45

0.01

4.57

0.11

$32,724

$560

$149,503

$1,270

Marin County

0.50

0.01

5.11

0.22

$35,012

$1,241

$178,870

$5,810

Napa County

0.46

0.01

4.35

0.23

$30,006

$1,397

$130,528

$3,988

San Francisco County

0.51

0.00

6.38

0.16

$23,739

$812

$151,519

$2,405

San Mateo County

0.47

0.01

4.36

0.11

$38,170

$619

$166,527

$1,978

Santa Clara County

0.45

0.01

4.64

0.08

$35,759

$712

$165,795

$1,607

Solano County

0.40

0.01

3.96

0.12

$30,887

$687

$122,372

$2,103

Sonoma County

0.44

0.01

4.35

0.14

$27,370

$921

$119,138

$1,211

* MOE = 90% margin of error

** The Gini coefficient measures the distribution of income relative to the distribution of people--how much income do the poorest 10% of the population control, the poorest 20%, and so on. The Gini coefficient ranges from 0 to 1, with larger values indicating greater inequality.

*** The "80/20 percentile ratio" illustrates the ratio of income at the 80th percentile cutpoint to income at the 20th percentile cutpoint.

Why Is This An Indicator Of Health and Sustainability?

Numerous studies have shown that income inequality, a measure of the distribution of income, is strongly and independently associated with decreased life expectancy and higher mortality, as well as reduced self-rated health status.a  The effects of income inequality are likely mediated via public investments in shared goods and services and socially via social cohesion, intrapersonal trust, and reciprocity. Accordingly, places with relatively more egalitarian distributions of income would have a higher average expectancy irrespective of the average level of income.b

Interpretation and Geographic Equity Analysis

Income inequality metrics aim to describe inequalities in the distribution of income in a specific population. Some measures like the Gini coefficient are measures based on the entire distribution of income and others capture relative differences in incomes at specific points in the distribution or between different populations.  The tables and charts above illustrate income inequality in San Francisco compared to California and the United States by population quintile, by ethnicity, and by household type from 2006 to 2010.  The fourth table presents income inequality by Gini Coefficient and the 80/20 percentile ratio for the nine Bay Area counties during the same time period. 

All four charts illustrate that there is significant income inequality among San Francisco residents.  The Bay Area, and San Francisco County in particular, have some of the highest income disparities in the state of California.

The first chart and table show the average income for the populations of San Francisco, California and the United States by quintile, or fifth of the population.  The quintile with the highest income in San Francisco earns 22.6 times more than the quintile with the lowest income.  By comparison, the highest quintile in California and the United States earns 15 times more than the lowest quintile in California and the United States respectively. 

Although the median income and the second, third, fourth and highest quintiles of household income are lower in California than in San Francisco, the household income of the lowest quintile of San Franciscans is less than the household income of the lowest quintile of Californians.  This would suggest that very low income people in San Francisco are poorer on average than very low income people in California generally, even though household incomes for all other income quintiles are higher on average in San Francisco than in California.

The significant disparity between high and low income earners is further affirmed by the Gini Coefficient and the 80/20 percentile ratio in the fourth table (See methods section below for description of these two measures).  The Gini Coefficient in San Francisco (0.51) is higher than all other eight Bay Area counties, suggesting greater income inequality in San Francisco.  By comparison, the Gini Coefficient for Los Angeles is 0.48 and for the US as a whole is 0.47.   

The San Francisco household with earnings at the 80th percentile earns over 6.4 times more than the household at the 20th percentile ($151,519 vs. $23,739).  These estimates represent an increase in income inequality since 2000, when the Gini Coefficient for San Francisco was 0.44, for Los Angeles was 0.39 and for the US as a whole was 0.41; and the 80th percentile SF household earnings were 5.4 times more than households at the 20th percentile ($114,119 vs. $21,175).

IncomeEthnicity

The above table illustrates household incomes by different income brackets and ethnicity both in ACS population estimates and as a percentage of the ethnicity’s population.   Overall, the household incomes and ethnicity table illustrates that on average, people of color in San Francisco have lower household incomes than Whites/Caucasians. 

IncomeFamily

The table above illustrates household incomes by household type, comparing San Francisco estimates to those of California and the United States overall.  The table illustrates that roughly one of every three family households have children under 18 in San Francisco compared to one of every two in CA and the US, yet unlike CA and the US, families with their own children under 18 earn more on average than those without their own children.  Female-headed households (with no husband present) earn 46% and male-headed households (with no wife present) earn 58% of the household income of married- couple households in San Francisco.  Although male-headed households earn proportionally the same amount in California and the United States as they do in San Francisco relative to married-couple households, female-headed households earn proportionally less in California (44%) and the US (41%) compared to San Francisco.  Similarly, females living alone in San Francisco earn more on average (58% of the SF median household income) compared to females living alone in California (49% of CA median household income) or in the United States (46% of the US median household income).Comparisons of both the estimated number and the percentage are important for identifying income inequalities to demonstrate magnitude and relative inequity.  For example, although there are three times more White/Caucasian households earning less than $25,000 per year than Black/African American households (27,542 vs. 9418), roughly one of every two Black/African American households earns less than $25,000 per year (42.9%) compared to one of every six White/Caucasian households (16%).  Similarly, although White/Caucasian households represent 40% of all households earning less than $60,000, 74.2% of Black/African American households, 53% of Hispanic/Latino households, and  49.6% of Asian/Pacific islander/Native Hawaiian households earn less than $60,000 per year, in comparison to 35.1% of White/Caucasian households.  White/Caucasian households are more than seven times as likely as Black/African Americans, three times as likely as Hispanics/Latinos, and more than twice as likely as Asian/Pacific Islander/Native Hawaiians to be earning more than $200,000.

Across the United States, income inequalities have grown larger since the 1970s.  The U.S. Census provides the following description of income inequality in the United States: "Generally, the long-term trend has been toward increasing income inequality. Since 1969, the share of aggregate household income controlled by the lowest income quintile has decreased from 4.1 percent to 3.6 percent in 1997, while the share to the highest quintile increased from 43.0 percent to 49.4 percent. Most noticeably, the share of income controlled by the top 5 percent of households has increased from 16.6 percent to 21.7 percent.” (http://www.census.gov/prod/2011pubs/acs-16.pdf)  

Between 1987 and 2009, 35.5% of income gains went to the wealthiest 1% of Californians (or 144,000 taxpayers) and 71.3% of income gains went to the wealthiest 10% of Californians.  During this time, the average inflation-adjusted income of the top 1% of California’s taxpayers increased by 50.2% (from $778,000 to $1.2 million), whereas the average income of taxpayers in each of the bottom four fifths of the distribution lost purchasing power.  For more about the widening inequality in California, see the California Budget Project 2011 presentation: http://www.cbp.org/pdfs/2011/111101_A_Generation_of_Widening_Inequality.pdf

As noted by the Alameda County Public Health Department, income inequality in the United States over the past thirty years has been impacted by declining welfare benefits, regressive tax cuts, cuts to education and health programs, and the erosion of workers’ collective bargaining power.  These changes have resulted in rising wage inequalities and a widening of the rich-poor gap in life expectancy, despite initial gains from civil rights legislation and Great Society programs in the 1960s. Accessed at: http://www.acphd.org/media/53628/unnatcs2008.pdf

Methods

The “Gini Coefficient” measures the distribution of income relative to the distribution of people--how much income does the poorest 10% of the population control, the poorest 20%, and so on. The Gini coefficient ranges from 0 to 1, with larger values indicating greater inequality.   

The "80/20 percentile ratio" illustrates the ratio of income at the 80th percentile cutpoint to income at the 20th percentile cutpoint. Calculating the 80/20 percentile ratio for household incomes involves arranging household incomes from lowest to highest income, and then dividing the list of all incomes into five categories (quintiles) with equal numbers of households in each category. The income figure at the 80 percent cutpoint is divided by the income figure at the 20 percent cutpoint to generate a percentile ratio. The larger the percentile ratio, the greater the inequality.

Limitations

This indicator attempts to provide different methods of assessing income inequality by income level, ethnicity and household type, yet each of the methods rely upon reported household income, which is a widely used though limited measure of access to wealth and material resources. 

Household incomes do not necessarily account for differential access to public or private assistance.  Individuals who have independent wealth or financial support from families or friends may be better able to weather financial turmoil than those without that additional support.  Financially secure homeowners may be able to borrow money from the equity in their homes to help pay high medical bills, car accidents, college tuition, or other large financial burdens; whereas homeowners facing mortgage foreclosures may have additional financial burdens that further deplete current and later household incomes.  Individuals with good medical insurance coverage may be less impacted by health emergencies than those with no or poor coverage.  Financial stability and access to a self-sufficiency wage are impacted by many different factors including educational attainment, race/ethnicity, class, languages spoken, access to financial institutions, job training, financial literacy, inter-generational wealth or poverty, etc.

The data above provides an average of a five year time-span (2006-2010) during which there was a major economic recession.  The impacts of the recession upon household incomes today may not be accurately reflected in the data presented above.   Certain populations and industries were  more impacted by the recession, layoffs, and the mortgage foreclosure crisis than other populations, which would further impact the income inequalities described above.   

Data Source

U.S. Census.  American Community Survey 5 Year Sample (2006-2010).  Tables:

  • B19001, B19001B, B19001C, B19001D, B19001E, B19001F, B19001G, B19001H, B19001I: Household Income in the Past 12 Months (in 2010 Inflation-Adjusted Dollars).
  • B190081: Mean Household Income of Quintiles.
  • B19049: Median Household Income in the Past 12 Months (in 2010 Inflation-Adjusted Dollars) by Age of Householder.
  • S1903: Median Income in the past 12 months (in 2010 Inflation-Adjusted Dollars).  

Obtained via American Factfinder.  Charts and tables created by the San Francisco Department of Public Health.

  1. Lynch J, Smith GD, Harper S, Hillemeier M, Ross N, Kaplan GA, Wolfson M. Is income inequality a determinant of population health? Part 1. A systematic review. Milbank Q. 2004;82(1):5-99.
  2. Ichiro Kawachi, 2000. Income Inequality and Health. In Social Epidemiology. Eds. Lisa Berkman and Ichiro Kawachi. New York: Oxford University Press. Pp. 76-94.