ERS Charts of Note

Subscribe to get highlights from our current and past research, Monday through Friday, or see our privacy policy.
See also: Editor’s Pick 2017: Best of Charts of Note gallery.


Natural gas price shocks increase the probability of food hardship

Friday, July 14, 2017

Poor households often lack the savings, assets, and income to protect themselves from unexpected increases in energy prices (called energy price shocks). A recent ERS study explored the relationship between energy price shocks and food hardship, including food insecurity—not having resources to acquire enough food for some or all household members. The study found that price shocks in gasoline, natural gas, and electricity increased the probability of households becoming food insecure and/or experiencing two other food hardship measures, with a larger response for low-income households compared to the average response for all households. Natural gas price shocks had the most consistent effects. Over 2000-14, annual price increases for natural gas ranged from 7 to 25 percent, and some years posted price declines. The study found that increases in natural gas prices above these typical increases, i.e. unexpected, large price rises estimated to average 41 percent above prior years’ prices, raised the probability of needing more money for food by 1.0 percentage point for all households in the data set and by 1.4 percentage points for low-income households. The unexpected, large price increases also increased the probability of food stress by 1.2 percentage points for all households and 2.2 percentage points for low-income households. Natural gas price shocks increased the probability of food insecurity by 2.3 percentage points for low-income households, more than double the response for all households. This chart appears in "Unexpected Hikes in Energy Prices Increase the Likelihood of Food Insecurity" from ERS’s Amber Waves magazine, July 2017.

Poorest U.S. households spent 33 percent of their incomes on food in 2015

Friday, May 19, 2017

While households spend more money on food as their incomes rise, food expenditures represent a smaller portion of income as households allocate additional funds to other goods. In 2015, U.S. households in the highest income quintile spent an average of $12,350 on food—both from grocery stores and eating out. This spending accounted for 8.7 percent of their incomes. Middle income households spent an average of $5,799 on food, or 12.4 percent of their incomes. Households in the lowest income quintile spent less for food on average—$3,767 in 2015—but their food expenditures accounted for 33 percent of their incomes. Two years earlier, the lowest income quintile spent 36.2 percent of their incomes on food. The share of income spent on food depends on several factors, including food prices and incomes. While retail food price inflation was relatively low in 2013, income levels were also lower than in 2015, contributing to the higher percent of income spent on food in 2013 by the lowest income households. Food expenditures as a share of income could fall in 2016 and 2017 across income levels due to declining retail food prices in 2016 and a continued trend downwards in prices for some foods in 2017. This chart is from ERS’s Selected charts from Ag and Food Statistics: Charting the Essentials, 2017, released April 28, 2017.

Share of income spent on food by poorer U.S. households is relatively large and volatile

Tuesday, September 13, 2016

Over the past two and a half decades, U.S. households in the lowest income quintile (the poorest 20 percent of households) spent between 28.8 and 42.6 percent of their annual before-tax income on food, compared with 6.5 to 9.2 percent spent by households in the highest income quintile. Before-tax income includes earnings and other money income, public assistance, Supplemental Security Income payments, and Supplemental Nutrition Assistance Program (SNAP) benefits. The share of income spent on food is more volatile for poorer households reflecting their less stable incomes and changes in grocery store (food-at-home) prices. Higher at-home food prices disproportionately affect the spending behavior of low-income households and often require them to allocate a larger share of their incomes to food. The lowest income households saw their share of income spent on food drop from 41.1 to 28.8 percent over the years 2001 to 2007, rise to 35.5 percent in 2009, rise again to 37.8 percent in 2013, and return to 35.5 percent in 2014. Over the same period, the highest income households saw relatively minor yearly swings of 0.5 to 1.0 percentage points. This chart appears in “Percent of Income Spent on Food Falls as Income Rises” in the September 2016 issue of ERS’s Amber Waves magazine.

Food insecurity more common for households that include adults with disabilities at each income level

Thursday, September 1, 2016

Disability has emerged as one of the strongest known factors that affect household food insecurity. Food-insecure households are those that lack consistent access to adequate food for one or more household members. Because they face higher expenses for health care and adaptive equipment, households affected by disabilities require higher incomes to meet their basic needs than do households without members with disabilities. Even households that have incomes greater than three times the poverty level have a relatively high likelihood of being food insecure if they include an adult with a disability. In 2009-10, an estimated 13 percent of households that included a working-age adult not in the labor force due to a disability, and had incomes at least three times the Federal poverty line ($22,113 for a family of four), were food insecure. About 9 percent of households in that income range with an adult with a non-work-preventing disability were food insecure. In comparison, about 4 percent of households in that income range with no working-age adults with disabilities were food insecure. This chart appears in the May 2013 Amber Waves feature article, ?Disability Is an Important Risk Factor for Food Insecurity.?

During economic downturns, more children receive free and reduced-price school lunches

Thursday, September 1, 2016

USDA?s National School Lunch Program (NSLP) is the Nation?s second largest food and nutrition assistance program. In fiscal year 2012, expenditures totaled $11.6 billion and an average 31.6 million children participated in the program on a typical school day. While total participation is closely linked to school enrollment and does not vary with economic conditions, the share of children receiving free or reduced-price lunches rises during economic downturns. During the Great Recession (December 2007 to June 2009) and continuing through 2010, the share of NSLP participants receiving free or reduced-price meals grew from 59 to 65 percent. Preliminary data for fiscal years 2011 and 2012 show that this share continued to rise even after the unemployment rate started to decline, suggesting that economic conditions did not improve enough to raise people out of poverty and the need for assistance remained high. This chart appears in the May 2013 Amber Waves article, ?Economic Conditions Affect the Share of Children Receiving Free or Reduced-Price School Lunches.?

Incomes remained lower for the poorest rural families with children in 2014

Thursday, September 1, 2016

By 2014, average income (adjusted for inflation) for all U.S. families with children exceeded prerecession levels, and average income had almost completely recovered for all rural families with children as well. For the bottom 25 percent of rural families (when ranked by income), however, average income remained considerably below its prior peak. In 2003, the average income for families in this lowest income quartile was $17,200 (in 2014 dollars) and it fell by 6.0 percent between 2003 and 2007, despite the fact that the U.S. economy was growing. Not surprisingly, incomes for the bottom quartile fell by another 4.6 percent between 2007 and 2010, due to the Great Recession (December 2007-June 2009). When economic growth resumed, however, it did not immediately translate into growth for these low-income rural families: by 2012, their average income had fallen by another 10.1 percent. Average income for the bottom quartile rebounded somewhat between 2012 and 2014, but remained 13.4 percent below the 2003 level. This chart is based on the Amber Waves feature, ?Understanding Trends in Rural Child Poverty, 2003-14.?

Changing rural demographics contributed to rising child poverty

Thursday, August 11, 2016

Using data from the Census Bureau?s Current Population Survey and a modified official poverty measure, ERS researchers found that rural child poverty rose from 18.7 percent in 2003 to 22.1 percent in 2014. The bulk of this 3.4-percentage point increase?3.2 percentage points?was due to rising income inequality, and not a decline in average incomes. A portion of this increase in inequality, in turn, was driven by changing rural demographics. An increase in the number of children in the average rural family raised poverty by 0.6 percentage points, while declines in the number of adults of prime working age and in the share of household heads that were married raised rural child poverty by 0.9 and 0.7 points, respectively. A slight increase in the average age of the household head helped reduce rural child poverty by 0.5 percentage points. The most beneficial demographic change was a rise in the share of rural household heads with a college degree, which rose from 15.8 to 19.5 percent, helping to reduce child poverty by 0.9 percentage points. The net impact of all these demographic changes was to contribute 0.9 percentage points towards the increase in rural child poverty. This chart is based on a data table found in the May 2016 Amber Waves feature, ?Understanding Trends in Rural Child Poverty, 2003-14.?

Linking administrative and survey data shows SNAP reaching more of the poorest households

Thursday, October 8, 2015

In general, households are eligible to participate in USDA’s Supplemental Nutrition Assistance Program (SNAP) if their gross monthly incomes do not exceed 130 percent of the poverty line, and if they meet other net income and asset criteria. Households with disabled or elderly members can qualify with higher incomes, and some States have adopted higher income thresholds. ERS researchers recently linked American Community Survey (ACS) data to SNAP administrative records from the State of New York to get a more complete look at whether SNAP benefits are reaching the poorest households. When researchers adjusted the households, as defined by the ACS, to reflect the SNAP definition of a household, they found that 71.6 percent of New York households receiving SNAP had annual incomes at or below 130 percent of poverty, compared with 63.1 percent using just the ACS data. Fifty-three percent of the New York SNAP households with incomes above 130 percent of poverty had an elderly or disabled member or a child. New York allows households with dependent care expenses to qualify for SNAP with gross monthly incomes up to 200 percent of poverty. This chart appears in “Linking Administrative and Survey Data Provides a More Complete Picture of Whether SNAP Benefits Reach the Poorest Households” in the September 2015 issue of Amber Waves magazine.

Percent of residents receiving SNAP benefits in 2014 declined in many States

Tuesday, June 2, 2015

USDA’s Supplemental Nutrition Assistance Program (SNAP) served an average of 46.5 million people per month in fiscal 2014. The percent of Americans participating in the program declined from 15.0 percent in 2013 to 14.6 percent in 2014, marking the first decline in the percent of the population receiving SNAP since 2001. Between 2013 and 2014, 42 States and the District of Columbia saw a decrease in the percent of residents receiving SNAP benefits, while 8 States experienced small increases. The percent of State populations receiving SNAP benefits ranged from a low of 6.1 in Wyoming to a high of 21.9 in Mississippi, reflecting differences in need and in program policies. Southeastern States have a particularly high share of residents receiving SNAP benefits, with participation rates of 15.8 to 21.9 percent. Utah had the largest decline from 2013 to 2014 and joined Wyoming and North Dakota as States with less than 8 percent of the population receiving SNAP benefits. This chart appears in the ERS data product, Ag and Food Statistics: Charting the Essentials, updated May 7, 2015.

Poorest U.S. households spent 36 percent of their income on food in 2013

Monday, May 4, 2015

While households spend more money on food when their incomes rise, food expenditures represent a smaller portion of income as households allocate additional funds to other goods. In 2013, U.S. households in the middle income quintile, with an average 2013 after-tax income of $43,592, spent an average of $5,728 on food, or 13.1 percent of their incomes. The lowest income households—those with annual after-tax incomes of $10,092 and below in 2013—spent $3,655 on food on average, or 36.2 percent of their incomes. Since 2009, rising food prices and falling incomes put pressure on food budgets. In pre-recession 2006, households in the lowest income quintile spent 32 percent of their incomes on food and middle income households 12.8 percent. Between 2006 and 2013, average incomes for the lowest quintile rose only 1.2 percent and fell 0.5 percent for middle income earners. This chart is from the Food Prices and Spending section of ERS’s Ag and Food Statistics: Charting the Essentials data product.

Share of WIC participants with incomes at or below poverty has increased since 2000

Friday, January 30, 2015

Income eligibility for USDA’s WIC program is capped at 185 percent of the Federal poverty level. Applicants who demonstrate current eligibility for Medicaid automatically meet income eligibility for WIC and do not have to document their incomes when they apply for WIC. With a large and increasing share of WIC applicants reporting participation in Medicaid at the time of WIC certification (73 percent in 2012, up from 66 percent in 2010), there has been concern that if States continue to raise their Medicaid income cutoffs to greater than 185 percent of the Federal poverty level this might result in WIC serving a larger share of participants with incomes above the intended cap. However, that has not happened. The share of WIC participants with incomes at or below poverty has been steadily increasing since 2000. The share of WIC participants with incomes at or below 50 percent of the poverty line has increased from 26.5 percent to 33.4 percent in 2012, and the share with incomes between 51-100 percent has grown from 29.1 percent to 33.2 percent in 2012. A version of this chart appears in the ERS report, The WIC Program: Background, Tends, and Economic Issues, 2015 Edition, released January 27, 2015.

Transfer payments to residents in nonmetro areas outpace payments to metro residents

Wednesday, December 5, 2012

Government transfer payments comprise a large share of personal income for both nonmetro and metro residents. In 2011, transfer payments to individuals accounted for 24.8 percent of total personal income in nonmetro areas and 16.3 percent in metro areas. Per capita in 2011, nonmetro residents received more government transfers than metro residents: $8,236 vs. $7,022. Since 1978, nonmetro per capita transfer payments (adjusted for inflation) have risen faster than payments in metro areas. Most of this increase comes from the rising cost of government programs that provide medical benefits, such as Medicare and Medicaid. Because nonmetro areas have an older population and a higher proportion of persons with disabilities than metro areas, nonmetro areas receive more transfer payments. Between 2010 and 2011, per capita transfer payments fell, primarily due to a decline in unemployment insurance compensation, which fell 25 percent in both metro and nonmetro areas as benefits expired. This chart is found in the Rural Poverty & Well-being topic page on the ERS website, updated November 2012.

Food insecurity rates highest for single mother households, lowest for elderly

Monday, September 24, 2012

In 2011, 14.9 percent of all U.S. households were food insecure, meaning that they had difficulty at some time during the year providing enough food for all their members due to lack of resources. About 41 percent of households with incomes below the Federal poverty line were food insecure. Rates of food insecurity were substantially higher than the national average for single-parent households, as well as Black and Hispanic households. Food insecurity was more common in large cities and rural areas than in suburban areas and other outlying areas around large cities. Rates of very low food security--a more severe form of food insecurity--were higher for these demographic groups as well. The statistics for this chart are from Household Food Security in the United States in 2011, ERR-141, released on September 5, 2012.

SNAP participation up more in some States than others

Tuesday, July 24, 2012

In 2011, nearly one in seven Americans (14.3 percent of the total population) lived in a household that received food assistance benefits from USDA's Supplemental Nutrition Assistance Program (SNAP). This represents a 3.4 percentage point increase from 2009 when 10.9 percent of Americans were receiving SNAP benefits. The rate of increase varied by State with 10 States, plus the District of Columbia, showing percentage point increases above 4.5 and 8 States with increases below 2.5. A combination of economic and policy factors account for differences among the States. All of the States with the slowest growth in SNAP caseloads had unemployment rates in May 2011 below the national average, and only two of them had enacted policies to expand program eligibility. In contrast, 8 of the 11 jurisdictions with the fastest growing SNAP caseloads had unemployment rates above the national average and all 11 had adopted policies raising SNAP income and/or asset eligibility cutoffs for many households. This chart is one of the new maps in ERS's Food Environment Atlas, updated on June 29, 2012. For more information on SNAP and other nutrition programs, visit the Food and Nutrition Assistance topic page on the ERS website.

Participation in USDA's Supplemental Nutrition Assistance Program varies by State

Tuesday, July 17, 2012

The Supplemental Nutrition Assistance Program (SNAP) is the cornerstone of USDA's nutrition assistance programs. The program accounted for 73 percent of all Federal food and nutrition spending in fiscal 2011, serving an average of 44.7 million people per month, or about 14 percent of Americans. The percent of the population receiving SNAP benefits to purchase food varies across States reflecting differences in need, as well as differences in program policies. The Southeast stands out as a region where all States have a high percent of residents receiving SNAP benefits, with participation rates of 16 to 21 percent. In 2011, 6.4 percent of Wyoming's population received SNAP benefits-the only State with less than 8 percent of the population receiving SNAP benefits. In 2009, there were 12 States with less than 8 percent of their populations participating in SNAP. This chart is one of the new maps in ERS's Food Environment Atlas, updated on June 29, 2012.

SNAP benefits mitigated increases in severity of child poverty during the 2007-09 recession

Tuesday, July 10, 2012

The poverty rate for children grew from 17.4 percent in 2006 to 20.7 percent in 2009, as many families' incomes fell during the 2007-09 recession. Official poverty estimates do not include non-cash assistance, such as benefits from USDA's Supplemental Nutrition Assistance Program (SNAP), as income. A recent ERS study found that including SNAP benefits in family income would have lowered the poverty rate for children to 18.7 percent in 2009. SNAP benefits also ensured that the severity of child poverty increased only slightly from 2006 to 2009 despite worsening economic conditions. Severity of poverty was measured using an index that reflects the square of the poverty gap, defined as the average distance of poor families' incomes below the poverty threshold. Without SNAP, the severity measure would have increased from 5.5 in 2006 to 6.8 in 2009, an increase of almost 24 percent. With SNAP benefits, severity rose by only 11 percent. This chart appeared in "SNAP Benefits Alleviate the Incidence and Intensity of Poverty" in the June 2012 issue of ERS's Amber Waves magazine.

Middle-income households made biggest cuts to food spending during recession

Wednesday, September 14, 2011

During the 2007-09 recession, inflation-adjusted food spending by U.S. households fell 5 percent-the largest decrease in at least 25 years. Spending patterns differed by income level, with middle-income households curbing expenditures the most. Households in the middle quintile of income decreased their inflation-adjusted food expenditures by 12.5 percent from 2006 to 2009. Households in the lowest quintile cut spending 1.8 percent, while the highest quintile reduced food spending 5.7 percent. This chart appeared in the September 2011 issue of Amber Waves magazine.

Nonmetro families in deep poverty reached a 20-year high in 2009

Tuesday, August 16, 2011

The 2007-09 recession was particularly severe in the depth of poverty it engendered. The number of nonmetro residents living in poverty increased more than 8 percent between 2007 and 2009 to 8.1 million (a 16.6-percent poverty rate). Of these, an estimated 3.3 million (6.7 percent of the nonmetro population) were in deep poverty in 2009, up nearly 13 percent from 2007. The number of nonmetro residents living in poverty rose markedly in the previous two recessions as well (as measured by 1989-91 and 2000-02 annual data). Yet, the change in the deep poverty count was substantially higher during the 2007-09 recession. This chart appeared in the June 2011 issue of Amber Waves magazine.

Growth in nonmetro poverty is tied to recessionary increases in unemployment

Monday, March 14, 2011

Historically, poverty grows during recessionary periods as a result of job losses, difficulty finding new work, and reductions in work hours or wages. This is evident in both nonmetro and metro poverty rates, which follow the same trends as unemployment over time. The dramatic increase in unemployment during the 2007-09 recession is tied to growth in the number of nonmetro residents living in poverty. In 2009, the nonmetro unemployment rate was 9.0 percent, which represents a 3.4-point increase from 2008. This is the largest annual increase in unemployment in more than 35 years. The percent of the population living in nonmetro areas that were poor in 2009 was 16.6 percent, compared with 15.1 percent in 2008. This is the most significant year-to-year increase in nonmetro poverty since the early 1980s recession. This chart is from the December 2010 issue of Amber Waves.

Charts of Note header image for left nav