The Washington PostDemocracy Dies in Darkness

Message from new census data: Keep the recovery on track and leave the ACA alone!

Source: U.S. Census Bureau, my analysis. The notch at the end of the series represents the change in survey methods for this income variable.

This morning, the U.S. Census Bureau released its annual report on median income, poverty and health insurance coverage. Here’s my take:

Health-care reform is proving to be tremendously effective in delivering affordable health coverage to millions of Americans. The economy, on the other hand, is not adequately reaching the poor and middle class.

Poverty in America was 14.8 percent in 2014, as 47 million people lived in households with incomes below the federal poverty line, about $24,000 for a family of four with two children, according to today’s release by the Census Bureau. Since 2010, poverty has remained at about 15 percent, well above its pre-recession rate of 12.5 percent rate in 2007. Thus, while 2014 was the fifth year of an economic recovery, the poverty rate remains elevated.

Similarly, middle-income households failed to gain ground last year as the median income was also unchanged. Though a change in survey methods led to an increase in the level of income (see data note below), a consistent measure between 2013 and 2014 shows no gain in real terms for the median household.

Clearly, the improving economy and falling unemployment have yet to adequately lift the living standards of middle- and low-incomes. The census data show that almost 3 million more people were working year-round in 2014 than in 2013, yet real median earnings were unchanged for both men and women. Poverty remains higher and median incomes lower than before the recession, and this pattern — taking longer in the upturn to make up the losses from the downturn — seems dangerously embedded in the economy.

On the other hand, one of the most important, positive and striking findings from today’s report is the drop in the number of Americans lacking health coverage, a clear sign of the Affordable Care Act at work.  Last year, 8.8 million more people got coverage, leading to yet another decline in the uninsured rate, from 13.3 percent in 2013 to 10.4 percent in 2014. That’s a record low for these data, and a comparison using a range of health coverage surveys demonstrates that these are the largest single-year declines on record, with data back to 1987.

In other words, the Affordable Care Act is clearly having its intended effect as the premium subsidies, the exchanges and the Medicaid expansion (in states that accepted it) are significantly broadening coverage. As Matt Broaddus points out below, had more states taken the expansion and experienced the same decline in the uninsured rate as expansion states, 2.6 million more people would have health coverage.

But at the same time, the economy is failing to deliver enough growth to too many households. More than five years into this expansion, neither the poor nor the middle class have recovered their pre-recession losses, nor — according to these data — are making much progress in that direction. That fact has important policy implications. Both the Federal Reserve and the Washington policymakers need to be committed to ensuring that the recovery stays on track so as to give those left behind a fair chance to catch up. This implies accommodative Fed policy, no unforced politically-motivated errors — like a government shutdown, a debt-ceiling showdown, adherence to sequestered spending caps wholly unrelated to need, continued attacks on health reform despite its historic progress — and the pursuit of budget and wage policies that will help reconnect low- and middle-income prosperity to the growing economy.


Starting with the good news, insurance coverage expanded for both the privately insured and those with coverage from public sources. As my CBPP colleague Matt Broaddus pointed out to me: “A simultaneous rise in private and public health coverage of any size, let alone a rise of this year’s magnitude, hasn’t happened since 2000 and reflects health reform’s balanced approach to public and private coverage.  Notably, employer-sponsored coverage rates stayed flat, countering the claim that health reform would lead to substantial reductions in employer-based coverage.”

Broaddus also finds that these data affirm the importance of state participation in the ACA’s Medicaid expansion: The 27 states that expanded Medicaid under health reform had a 3.4 percentage point drop in their uninsured rate as compared to only a 2.3 percentage point reduction other states.  If the uninsured rate had fallen in non-expansion states at the same rate as in expansion states, an additional 2.6 million Americans would have gained coverage in 2014.

On the other hand, there’s the plodding recovery and the fact that it’s not lifting the poor and middle class. The census reported that real (inflation-adjusted) income of the median was unchanged between 2013-14 (using consistent survey methods — see note below). The 2014 median was 6 percent, or $3,700 below the level of 2007 (in 2014 dollars). For African American households, that loss was a larger 9 percent.

Notably, since median income from this data source grew slowly in the 2000s, the 2014 real median household income was 7 percent lower than the median in the year 2000. That’s 14 years and two economic expansions, and yet the median household income is down by about $7,000 since 2000, a remarkable example of the extent to which the economy has left the middle-class behind.

Poverty rates, as noted, remain around 15 percent (14.8 in both of the last two years) overall, with higher rates for children (21 percent), African Americans and Hispanics (26 and 24 percent respectively).

The table below was designed to trace losses and progress in income and poverty since right before the great recession took hold in 2008. Look, for example, at the column for white non-Hispanics. Their income fell and their poverty rose, as you’d expect, after they were hit hard by the recession. By 2011 (see table note), their income had bottomed out, and it’s up about 3 percent since then. But for each racial/ethnic group in the table, median income is lower and poverty is higher than before the downturn.

What explains this economic disconnect between growth, income and prosperity? While longer-term trends — globalization, technology, the absence of full employment, low bargaining power for many workers — have been in play for decades now, in recent years, fiscal policy has been insufficiently supportive of growth, and, in our age of increased income inequality, it takes longer for expansions to reach middle and low-income households. The figure at the top of this post makes this point using the median household income.

The cyclical nature of the series is clear: It falls in real terms in downturns and climbs in recoveries. I’ve drawn horizontal arrows from pre-recession peaks to show how long it took to regain the income peaks after the downturns. But in the 2000s expansion, the median never regained its peak. In the current expansion, the median remains well below its pre-recession peak.

Other income series that include more income sources than these census data provide a somewhat more optimistic picture, particularly for the poor (these census data leave out the value of non-cash benefits, tax credits, and the value of health benefits). But the story is roughly the same in other data sources, especially post-2000, as wage, income and wealth inequality has engendered persistent gaps between growth and more broadly-shared gains.

Moreover, while fiscal policy initially pushed back hard against the demand contraction caused by the great recession between 2011 and 2014, the pullback in federal spending shaved growth off of GDP, significantly slowing the pace of the recovery and leaving gaps in employment and output that remain to this day.

The policy implications of these observations are straightforward, both in terms of Federal Reserve policy and broader fiscal, anti-poverty, and workplace policy debates that are ongoing in D.C.

The Fed committee in charge of setting interest rate policy is currently meeting, and they may decide — either at this or at a subsequent meeting later this year — to raise rates slightly in response to concerns that the economy will soon be generating inflationary pressures. However, from the perspective of middle- and low-income households still making up losses from the downturn, or those struggling to make ends meet in a job market with pockets of underemployment and weak real wage growth, preemptive rate hikes are counterproductive and ill-advised.

On the other hand, policies that directly connect the fortunes of working households to the growing economy are particularly important right now. These include higher minimum wages, implementing the new overtime rule at the threshold proposed by the president (about $50,000, meaning salaried workers paid less than that would be covered by overtime rules), and ensuring that refundable tax credits for low-income, working families remain robust.

Given the survey methodology changes and the yearly noise in these data, it is important not to over-interpret them. Some analysis suggests this census report may be an outlier, given other data showing more positive 2014 outcomes(see Figure 2 here). But it is fair to observe longer-term trends like those in the income figure showing how long it’s taking for median incomes to make up their losses, and, conversely, the historically positive health coverage trends. These are trend-based results and they can be trusted, for better (coverage) or worse (income/poverty).

Data note: Because of some changes in survey methodology, there are some inconsistencies in longer-term trends in the census data released today. Any comparisons between 2013-2014 above use consistent methods, but some longer-term series can be affected by the change. Most notably, under the new methodology, median income is significantly higher. Thus, a consistent methodology would show a smaller uptick in 2013/14 over 2012 in the figure.