Poverty fell, middle-class incomes rose, and the share of Americans without health coverage ticked down to a historical low last year, according to data released Tuesday morning by the Census Bureau. This trifecta of gains for poor and middle-income households, as well as the uninsured, shows that the seven-year expansion, along with the Affordable Care Act, has continued to lift the living standards of many American households. However, last year’s gains were even stronger for those at the top of the income scale, suggesting that the inequality of economic outcomes continues to grow in America.
Building on these gains will require stabilizing and expanding the reach of the ACA, keeping the economic recovery going, and making sure our anti-poverty policies remain robustly funded. This last part will require pushing back hard on plans of the congressional majority to cut spending on programs that are demonstrably helping economically vulnerable households.
The share of the poor, officially defined as people in households with incomes below a minimum threshold, declined from 13.5 percent to 12.7 percent last year, a rate close to that at the last economic peak in 2007 (see note below on how a 2013 change to the survey affects such comparisons). Real median income — the inflation-adjusted income of the household smack in the middle of the income scale — rose 3.2 percent last year, or $1,800 dollars, to about $59,000, again close to its pre-recession peak. While these census data refer to 2016, as economic conditions have continued to broadly improve, similar progress may be occurring this year as well (though the impact of the 2017 hurricane season will need to be assessed, of course).
Changes in poverty and mid-level family incomes are the function of many factors, including the strength of the economic expansion, employment and unemployment, wage trends, inflation and, particularly for low-income families, anti-poverty policies, such as nutritional support and wage subsidies.
Though the official poverty rate leaves out the impact of many of these policies, the census analysts have, in recent years, overcome this limitation by releasing a “supplemental poverty measure” (SPM), which not only accounts for most anti-poverty policies, but also allows analysts to quantify the number of people helped by them. Last year’s SPM came in at 13.9 percent, and it showed that 8.2 million people were lifted out of poverty by refundable tax credits and 3.6 million by the Supplemental Nutrition Assistance Program (formerly food stamps), neither of which are counted in the official measure (SPM poverty thresholds are also updated relative to the official measure, which is one reason this metric generates higher poverty rates than the official measure, even after factoring in more income sources; it also subtracts certain medical costs and work expenses from income).
The 2016 job market clearly helped to support income gains last year. Payrolls expanded at a solid clip of about 187,000 per month in 2016, and the unemployment averaged just under 5 percent. Compare that to 2009, amid the last recession, when employment fell by over 400,000 jobs per month on average and the unemployment rate topped out at 10 percent in October of that year. Poverty rose sharply and real median incomes fell.
By 2016, the current expansion was more than seven years old, and though economic momentum has been slow to reach non-wealthy households in this era of heightened economic inequality, incomes have been rising and poverty falling for a few years now. In fact, earnings data from today’s report reveal that for full-time, full-year workers (ft/fy), median earnings were essentially unchanged last year. However, about 2 million more men and women moved into ft/fy work, suggesting that middle-class income gains came more from added work than from higher pay. This is consistent with labor market data showing that while real wages have gone up recently, they’ve climbed less than would be expected given the low unemployment rate.
As noted, inequality grew last year. For 95th-percentile households, real income rose from $217,000 to $225,000 last year, while for the lowest income households (10th percentile), income was essentially unchanged ($13,400 to $13,600). Taking a longer view*, from 1989 to 2016, high incomes rose by a third in real terms, while low incomes were unchanged. (Note: This data leaves out both taxes and benefits, which would somewhat boost low incomes, and realized capital gains, which would boost high incomes; more complete data sources show similarly large increases in inequality over this period).
The first figure below shows that the official poverty measure is down significantly from its peak of 15.1 percent in 2010 to 12.7 percent last year (the other line is the historical SPM series, discussed below). The next figure shows real median household income, which, after falling sharply in the downturn, has made back its losses in recent years (a change in survey methods in 2014 is responsible for the break in the series).
These metrics of the living standards of poor and moderate-income families posted particularly strong improvements in 2015, in large part because inflation hardly rose at all that year. Last year, price growth was still subdued, but at 1.3 percent, it led to smaller gains than in 2015. That said, the two-year gain of 8.5 percent in real median household income, 2014-2016, is the largest on record.
The SPM poverty measure in the first figure above was prepared by my colleagues at the Center on Budget and Policy Priorities, and it is extremely important for policymakers and poverty advocates to understand and appreciate the difference between the SPM and the official series. For example, the official rate was higher last year than it was in the early 1970s, which has led some observers to suggest that despite our many anti-poverty interventions since then, the share of the poor hasn’t fallen over time. But the SPM series shows the opposite trend, with poverty down about five percentage points since the early 1970s.
A key reason for the difference is that the official measure ignores the very anti-poverty measures that have expanded the most over this period. It does not count, for example, the strong pro-work, anti-poverty impacts of the earned income tax credit and other refundable credits, which, as noted, lifted 8.2 million people, including 4.4 million children, out of poverty last year. Nor does it count the value of SNAP. The SPM accounts for these changes, and as such, it is the only measure of the two capable of rendering any substantive judgments about the effectiveness of our anti-poverty interventions.
In fact, as the figure shows, the SPM rate has fallen by about 44 percent since 1967 (from about 25 to about 14 percent). That doesn’t mean our anti-poverty work is done — not by a long shot. Its current rates of 13.9 percent overall and 15.1 percent for children, while historically low, are still far too high for an economy as wealthy as ours, and higher than similarly measured rates in most other advanced economies. Also, there’s been little progress in the SPM since the late 1990s.
In fact, that more recent trend raises another important point about the SPM: The anti-poverty interventions that ramped up to offset the damage to vulnerable families from the Great Recession were extremely effective. Note that in those years, circled in the chart, the official poverty rate rose sharply, while the SPM was flat. That difference implies that the “countercyclical” benefits that rose to meet the historically large market failure of those years (the worst recession since the Great Depression) are not captured in the official measure. In short, the SPM reveals that the safety net worked when we needed it.
The census data also confirm expectations that, while the large health coverage gains achieved by the ACA were maintained, they did not fall much further last year. Much of the “low-hanging fruit” regarding expanded coverage was picked already, particularly through the Medicaid expansion to 32 states (including the District) and the spread of the marketplace exchanges, where people without insurance through other sources can purchase non-group coverage that, for most recipients, includes a premium subsidy.
The coverage results, and the potential to build upon them by stabilizing and expanding the ACA, pose an important reminder about the role of policy in these poverty, income and health coverage outcomes. State-level increases in the minimum wage, for instance, give a needed to boost to the incomes of households with low-wage workers, putting downward pressure on working poverty (research by economist Arin Dube finds that the 10 percent increase in the minimum wage reduces the poverty rate of working-age households by 5 percent).
The SPM results showing the millions lifted out of poverty by policies targeting economically vulnerable families, along with its highly effective countercyclical protections, underscore the importance of protecting those programs from budget cuts. This observation is particularly germane in the context of current fiscal discussions. President Trump and congressional Republicans have proposed regressive tax cuts that are, at best, partially paid for. If enacted, they will increase the deficit by trillions, at which point, Republicans will call for spending cuts to offset them. Based on their past proposals, those spending cuts will invariably hit the very programs that the SPM shows to be highly successful in poverty reduction (they have lasting poverty-reduction impacts as well).
Middle-wage workers would benefit from updating/raising the salary threshold that is a key determinant of who is eligible for overtime pay after 40 hours of weekly work. This weekly salary threshold is 56 percent below its 1975 level, after adjusting for inflation, but instead of updating this important labor standard, the Trump Labor Department has begun a process to revise the rule to make it cover fewer workers. That means less income for millions of middle-class families who are, by definition, working extra hard.
Finally, the nation’s central bank, the Federal Reserve, has a very important role to play in sustaining the gains in today’s report. The mandate of the bank is to use monetary policy that raises and lowers the benchmark interest rate it controls to balance employment pressures against price pressures. Because unemployment is now a relatively low 4.4 percent, the Fed has been slowly raising rates, tapping the economy’s brakes to forestall future inflation. But as noted above, inflation remains low and steady, even at low unemployment, implying more slack in the job market than implied by the unemployment rate. To keep the expansion percolating along at a pace that will build on the gains for poor and moderate income families in today’s report, the Fed should pause in its rate hike campaign.
In summary, the 2016 U.S. economy continued to deliver real income gains to many who’ve long been left behind. Policy-wise, we must build on this progress in two ways. First, we must do no harm, either with wasteful, regressive tax cuts paid for by spending cuts on successful anti-poverty programs or with unwarranted interest rate hikes. Second, we must push back on the inequality that’s still very much embedded in our economy. That means building on the policy gains of the ACA, strengthening and protecting our anti-poverty policies, and bolstering wage policies such as the minimum wage and overtime.
*Data note: A change in the census survey in 2014 significantly increased the amount of income it recorded, as can be seen in the median income series by the break in the series. Thus, the longer-term changes cited above should be considered rough estimates that are somewhat biased up by the survey change.