The work of Thomas Piketty and Emmanuel Saez on the evolution of top income shares has yielded a lasting and iconic image of American inequality: a long historical curve that starts high in the early years of the twentieth century (with the top 10 percent claiming about half of all income and the top 1 percent claiming about one-fifth), drops precipitously with the political innovations of the New Deal, and then climbs again—returning, by 2012, to its early-century heights—as those innovations are dismantled. This curve provides a dramatic and telling baseline for the trajectory of particular policies (like labor law and financial deregulation, for example) and for the political history of the last century.
In a new paper, Mark Price and Estelle Sommeiller have followed this methodological lead and developed estimates for top incomes shares, from 1917 through 2011, for American states and regions. This work provides texture and detail to the national story, highlighting both common elements and important differences across states. The full results are plotted below. For each year, the states (the light blue dots) are plotted by top income shares. The middle of the shaded box marks the median state (half are lower on the chosen measure, half are higher), and then the states are broken into quartiles: the middle quartiles (marked by the boxes) surround the median, the outer quartiles run along the lines between the boxes and the outer tick marks.
Across the full sweep of the last century, the pattern is telling. Into the 1930s the variation across states is fairly dramatic. On the eve of the Great Depression, the top 10 percent in the industrial (DuPont) fiefdom of Delaware claimed almost 85 percent of state income, the top 1 percent claim almost two-thirds, and the top .01 percent (that’s about ten tax returns in Delaware) hoarded over one-third. In the lightly developed frontier setting of Alaska the same year, by contrast, top earners took home not much more than their share: the top 1 percent of tax filers claimed a modest 5 percent of income in 1928.
Once the policies of the New Deal—collective bargaining, retirement and unemployment security, financial regulation, progressive taxation—were entrenched, the income share claimed by top earners fell and the variation across states shrank. This is evident in the box-and-whisker plot above, and in the comparison of any single state or region with the national trends. Perhaps the most striking aspect of the Price and Sommeiller data is how similar (as the graphic below underscores) state patterns are to those of the nation as a whole. Less developed and less populated states in the West and Midwest show less concentration of income at the top in the early decades. But overall, the arc of inequality—from state to nation—is remarkably consistent. Again, this is especially true in the middle years of the last century, when strong federal policies trumped (or overcame) the economic and political differences among the states.
Since 1979, as Price and Sommeiller show, both inequality and the disparity across states has widened again. In four states (Nevada, Wyoming, Mississippi, Arkansas), only the top 1 percent saw any income gains between 1979 and 2007 (the last year before the recession); in another fifteen states, the top 1 percent captured over half of all income growth over that span. Among those states whose 1 percent pulled furthest ahead of the pack, local growth in financial services (New York, Connecticut, New Jersey, Illinois) or information technology (Massachusetts, California, Washington) seem to be the driving force.
The map below illustrates this nicely. Here the states tip from green to red when the top 10 percent’s share of income exceeds one-third. Of the 16 states to top this threshold in 1972, the only ones outside the South were the tri-state home of big finance—New York, New Jersey, and Connecticut. As we move forward from there, the states in which the top 10 percent claim less than a third of total income gradually diminish in number, disappearing entirely by 1989. By 2011 the top 10 percent are claiming almost 60 percent of income in New York and Connecticut, and over 40 percent in all but three states (Iowa, Nebraska, and South Dakota).
The still-unfolding story is how much these gaps widened across the last business cycle, and especially during the first two years of the recovery. Thirty-eight of fifty states saw overall income growth over 2009–11, but in seventeen of those states only the incomes of the top 1 percent grew—and the bottom 99 percent actually lost ground. In seven more states, the top 1 percent captured over half of all income growth. In nine of the twelve states where overall incomes fell during the recovery, the top 1 percent still claimed net gains.
Mark Price and Estelle Sommeiller’s new paper traces the trajectory of top incomes in American states and regions from 1917 through 2011. Mapping this data across the continental United States and over the last century suggests both important similarities across states, and some key differences.
On the map below, the states tip from green to red when the top 10 percent’s share of income exceeds one-third. In the early years, inequality (as measured by the high income share of top earners), is starkest in the Northeast. This inequality is generalized by the impact of depression and war in the 1930s and 1940s, but once the policy innovations of the New Deal (collective bargaining, retirement security, labor standards, and financial regulation) take hold, inequality eases: by the middle 1950s, only New York and the Deep South are still colored red.
The pattern across the last generation is just as telling. Of the sixteen states to top this threshold in 1972, the only ones outside the South were the tri-state home of big finance—New York, New Jersey, and Connecticut. As we scroll forward from there, the states in which the top 10 percent claim less than a third of total income gradually diminish, disappearing entirely by 1989. By 2011, the top 10 percent are claiming almost 60 percent of income in New York and Connecticut, and over 40 percent in all but three states (Iowa, Nebraska, and South Dakota).
In The Increasingly Unequal States of America: Income Inequality by State, Mark Price and Estelle Sommeiller develop estimates for top income shares, from 1917 through 2011, for American states and regions. The national version of this story is now quite familiar: the iconic Piketty and Saez curve that starts high in the early years of the twentieth century (with the top 10 percent claiming about half of all income and the top 1 percent claiming about one-fifth), drops with the political innovations of the New Deal, and then climbs again—returning, by 2012, to its early-century heights—as those innovations are dismantled. So what does the state and regional breakdown tell us?
The full results are plotted below. For each year, the states (the light blue dots) are plotted by top income shares. The middle of the shaded box marks the median state (half are lower on the chosen measure, half are higher), and then the states are broken into quartiles: the middle quartiles (marked by the boxes) surround the median, the outer quartiles run along the lines between the boxes and the outer tick marks.
A few patterns stand out. First, the general sweep of the graph echoes the national story. The arc of inequality from Gilded Age to New Deal and back again is experienced across every state, not just in a few of them. Second, the policy innovations that dampened inequality—collective bargaining, retirement and unemployment security, labor standards, financial regulation, progressive taxation—also narrowed the variation across states. In the middle years of the last century strong federal policies trumped (or overcame) the economic and political differences among the states. And third, the erosion of those policies, beginning in the 1970s, saw both inequality and the variation across states widen again.
The American system of unemployment insurance is a remnant of Jim Crow. While national in its reach, the program’s administrative details are left to the states, a bargain struck in the 1930s as the price for Southern support for New Deal–era social programs. Since then, both eligibility for unemployment insurance and the generosity of the benefit have varied widely (and intentionally) by race, region, occupation, and earning history.
The loosening of occupational exemptions and earnings thresholds closed these gaps somewhat during the last half of the twentieth century. But in recent years they have begun to widen again as many states—foremost among them the states of the old Jim Crow South—have reasserted themselves as laboratories of austerity.
All of this is underscored by the expiration of federal emergency unemployment insurance benefits on December 28. The federal program, which has been invoked during every economic downturn since 1959, extends eligibility beyond the twenty-six weeks typically provided by state programs. In each of those past business cycles, federal benefits were phased out when the long-term unemployment rate (the share of the workforce unemployed six months or longer) fell into the 1 to 1.5 percent range. This time? With long-term unemployment still above 2.5 percent, the December 28 deadline cast about 1.3 million Americans off the eligibility rolls, and another 72,000 are expected to lose their benefits each week through the first half of 2014.
Leading this effort—in Congress and in the States—is the South. Even before the expiration of extended federal benefits (see map below), six of the eight states in which fewer than a quarter of the unemployed were receiving benefits fell below the Mason-Dixon line. The racial patterns are clear. African-Americans make a much larger share of the labor force across the South. African-Americans suffer unemployment rates about double those of whites. Over-represented on the margins of the labor market, blacks are about 25 percent less likely to qualify for jobless benefits than whites. Southern legislators—whether by paring benefits at the state level or leading the Congressional charge to slash the federal share—have the motive and the means to roll back the clock.
The expiration underscored and exposed the unevenness of state programs. After December 28, the national share of the unemployed receiving benefits plunged to 25 percent (this rate, since this data was first recorded in 1946, had never fallen below 30 percent before). The share in the states ranges from 11 to 35 percent. In the most generous settings, barely a third of the unemployed are getting any help. Of the eleven states of the Old Confederacy, nine cover fewer than 25 percent of the unemployed and seven cover fewer than 20 percent.
While the Senate is poised to reinstate the extended federal benefits, cooperation in the House—without big cuts elsewhere—looks unlikely. And, even if Congress is willing to cobble together another three months of aid, there is no assurance that Southern statehouses—following the lead of North Carolina—won’t refuse to cooperate altogether.
For the first time in a long time, the jobs report offered mostly good news. The last month saw decent job growth (just over 200,000, alongside an upward revision to the October numbers), and not just in low-wage sectors. An uptick in hours worked and a small decline in the number of involuntary part-time workers (those that would take full-time work were it available) suggest that we are moving in the right direction. Unemployment dipped to 7 percent even as labor force participation increased—meaning that the rate was falling because people got jobs, and not because they gave up looking.
But the very fact that we are celebrating 7 percent unemployment—a threshold not seen in five years—is itself grounds for caution. We are still a long way—about 8 million jobs by one estimate—from recovering recessionary losses and keeping up with growth in the labor market. An accounting of all the missing workers, who are unemployed or out of the labor market altogether due to the weakness of the labor market, would push the unemployment rate over 10 percent.
This persistent weakness is captured in the graphic below, which places the current business cycle in a longer historical perspective and highlights other measures of labor force weakness—including the share of part-time work, the share of “involuntary” part-timers, and the underemployment rate (which includes the unemployed, those no longer counted as unemployed but who would look for work if conditions were better, and those working part-time because they can’t find full-time work). Of special note here is the persistently high rate of long-term unemployment—over 37 percent of the unemployed (2.6 percent of all workers), a rate that actually inched up in November.
This brings us to the cautions. Some of the eagerness for solid job numbers comes from those who are looking for a signal (or excuse) to shift policy: to rein in the Federal Reserve’s moderately expansionary fiscal policy and to put an end to extended unemployment benefits for the long-term unemployed. Either move would be a serious mistake.
In the absence of decent labor standards or union density, full employment is our best hope for bidding up wages and dampening inequality. As new work from Dean Baker and Jared Bernstein makes clear, we have ample room to push down the employment rate without risking inflation. Even at the point at which expansionary fiscal policy might begin to push up wages and prices, the benefits of full employment would still outstrip the costs.
By the same token, it is the wrong time to think about paring pack unemployment benefits. Long-term unemployment, the target of the extended federal program, shows no sign of abating. In each of the last few business cycles, there has been a point in the recovery at which extended benefits are pared back. But as Jared Bernstein points out, this decision has historically been made when long-term unemployment is less than half its current levels. If extended benefits expire as scheduled, about 1.3 million workers would be cut off in the last week of December (happy holidays!), and another 1.9 million would be cut off in the first six months of 2014.
The past couple of weeks have offered glimmers of hope on the minimum wage front. Successful ballot measures in New Jersey and SeaTac, Washington (the Seattle suburb surrounding the airport) marked the latest efforts of state and local governments to push local minimums above the federal floor of $7.25 an hour. And, in the wake of the election, the President threw his support behind Tom Harkin’s Fair Minimum Wage Act—which would raise the federal minimum to $10.10 but has languished on the Senate calendar for nine months. Opponents, just as quickly, have dusted off the usual suspects and hit the airwaves and op-ed pages with tired claims that a $10 minimum would be a radical and intrusive departure, that it would slow growth and kill jobs, and that it would hurt the very workers it claims to help.
Let’s look at this more closely. By any reasonable measure, of course, there is nothing particularly dramatic or game-changing about a $10 minimum wage. In inflation-adjusted dollars, this still falls short of just returning the minimum wage to its level of a generation ago. More to the point, as John Schmitt has shown (here and here), a minimum wage indexed to a more meaningful benchmark—like productivity or the median wage—would fall closer to the $15 to $20 range.
The workers that would benefit from an increase, in turn, scarcely resemble the suburban teenage hamburger-flippers so often invoked by conservatives. The graphic below summarizes the distribution (across various measures) of hourly workers and of hourly workers paid at or below the minimum wage in 2012. Toggling between the two universes of workers underscores those demographics, occupations, and industries in which low-wage workers are overrepresented.
Unsurprisingly, a higher share of minimum wage workers are young—a fact reflected in the age and marital distributions. But over three-quarters of those working at or below the minimum are not teenagers. Indeed, recent work by David Cooper and Doug Hallunderscores that almost 90 percent of those who would benefit from an increase (a group that includes those working below, at, or near the minimum) are at least twenty years old. Almost 45 percent of minimum or sub-minimum wage workers have better than a high school education. And about 42 percent work more than thirty hours a week. African-Americans, Latinos, and women are all disproportionately represented in the minimum wage workforce.
But perhaps the starkest gaps are those revealed by industry and occupation. The leisure and hospitality industry accounts for just 13 percent of hourly employment but over half of all minimum wage employment. Food preparation and service account for less than 10 percent of hourly employment, but nearly 44 percent of workers earning at or below the minimum. This is why the fast food industry (in which low-wage, no-benefit, part-time employment is not just a pattern but a business model) has drawn such close scrutiny recently. This includesthe rolling strikes that began last summer and a devastating new report from the UC Berkeley Labor Center which finds that more than half of the families of front-line fast-food workers depend upon public social programs (twice the rate for the workforce as a whole), at a cost to taxpayers of nearly $7 billion annually.
Finally, the benefits of raising the minimum clearly outweigh any costs. In an exhaustive survey of the economics literature on this question, John Schmitt establishes pretty convincingly that the negative employment effects (layoffs or slower hiring in response to an increase) are either negligible or so small as to be undetectable, and that the adjustments made by workers and firms—including reductions in turnover, improvements in productivity, or wage compression—are mostly good things.
Job numbers, especially as we emerge slowly from the Great Recession, are an important marker of economic recovery. And for politicians of every stripe — especially those who have set goals for job creation — they are an equally important marker of political success. For these reasons, our attention is drawn to the monthly national jobs report, and to the state numbers that follow a few weeks later. And, for these reasons, it is important to get the numbers right.
National and state numbers come from the Establishment Payroll Survey, conducted each month by the Department of Labor (DOL) in cooperation with State Departments of Labor or Workforce Development. The DOL calls workplaces (private and government), asking how many people are employed. The survey results are tweaked to accommodate seasonal adjustments and sampling issues, and then yield a count of nonfarm jobs broken down by sector. Often the monthly release includes adjustments to the numbers from previous months.
The result is a measure of net job creation (adding up all the gains and all the losses), the number behind the typical headline: “Iowa added [fill in number] jobs in [fill in month].” By this straightforward measure, Iowa lost about 62,000 jobs during recession (December 2007 to January 2010) and (through August 2013) added about 70,000 jobs during the recovery — for a net gain of just under 8,000 jobs. Since Governor Branstad took office in January of 2011, we have added 56,600 jobs. Since the beginning of this year we have added 12,500 jobs.
Counting the number of jobs lost or gained is a pretty coarse measure of how we are doing, in part because that assessment depends on a lot of other moving pieces. This is especially true when we are aiming for certain targets — such as “creating” a certain number of jobs, or “getting back” to prerecession levels of employment. For this reason, our monthly “Iowa JobWatch” accounts for both the raw job numbers and for change over time in the state’s labor force. Our current jobs deficit — the number of jobs we would need to recover from the recession and keep up with population growth — is 55,100.
Let’s review: Using the establishment survey to calculate net job growth yields a rough measure of jobs gained and lost over any given period of time. Over any extended period of time (such as the current business cycle), it makes sense to account for changes in the labor force as well — so in a growing economy the targets are actually a little higher.
So how does all of this relate to Governor Branstad’s much-advertised goal of creating 200,000 jobs during his term, and the accompanying claims that we are halfway or three-fourths of the way there?
The administration invented a new measure of “Gross Over-the-month Employment Gains” in which Iowa Workforce Development adds up the gains in sectors that were adding jobs while ignoring those sectors that were losing jobs. Such a formula has no statistical or real world relevance. It is like counting deposits to your savings account but not the withdrawals; or like counting the touchdowns your team scores but not the ones that the other team scores.
But, for the sake of argument, if you add up the numbers in that line on IWD’s downloadable nonfarm jobs spreadsheet, you’d come up with 122,500 jobs created since January 2011, when the Governor took office. Yet, the Governor’s office is throwing around an even higher number, 163,500 jobs — the source or calculation for which remains a complete mystery.
We have been following these numbers every month for 10 years. It’s important that Iowans have real, reliable and independent data to understand the economic context for public policy debates and choices. New state numbers will come out November 22. If you want to know what the economy is doing, go on that day to the Iowa Policy Project website — www.iowapolicyproject.org. We will have it added (and subtracted) properly for you.
The congressional tantrum over Obamacare ended—as most such outbursts do—with lingering sniffles of discontent, general weariness, and stern recriminations. But it also ended, somewhat surprisingly, with most of the adults in the room talking quietly about a grand budgetary bargain centered on “entitlement reform.” For Tea Partiers, this was the last desperate target of the shutdown; for centrist Republicans and Democrats it marks the reprise of a monotonous—and utterly false—conviction that Social Security is some sort of time bomb ticking away in the bowels of the Treasury. “Washington has yet to address the main threat to the nation’s solvency, as USA Today concluded breathlessly last week, “the growth in entitlement programs.”
These attacks are as old as the law itself, and their arguments—that Social Security is an affront to American values, a burden on employers, and a false promise to workers—can still be heard today. But given the pension program’s spectacular success and popularity, opposition is now almost always couched as a simple matter of fiscal realism and responsibility: “We love Social Security (and Medicare/Medicaid), we just can’t afford it.” This argument rests on a series of willful and persistent myths:
Social Security has “exploded” as a share of the federal budget. Yes, social security has grown: the pension program represent about 1 percent of GDP in 1950, and now is about 5 percent of GDP (Medicaid and Medicare account for another 6 percent). But revenues—96 percent of which come from a dedicated payroll tax–have grown alongside the spending. The growth of social security is not some fiscal accident. It is a sustained political choice informed by thedemonstrable benefits of a program that lifts 22 million Americans—most of them seniors, but over a million of them children—out of poverty.
Social Security faces a “grey tsunami” as the baby boomers retire. In this view, social security’s prospects rest almost entirely on the shifting ratio between current workers and retirees. First, this is scarcely a surprise: politicians and the social security trustees have recognized this demographic challenge for years. Second, there are lots of moving pieces here—and equitable wage and productivity growth are far more important to the program’s future than any demographic bulge. There is nothing preventing us from matching projected wage growth with modest tax increases to sustain a valued and successful program.
Social Security is going “broke” or “bankrupt.”This claim willfully misrepresents the relationship between program revenues and program commitments. Social security outlays have exceeded revenues since 2010, and—at current rates—the trust fund itself will be exhausted in about 20 years. But that doesn’t mean everything grinds to a halt. In the absence of any changes, current revenues could still cover three-fourths of scheduled benefits after 2033. In the big picture, the shortfall is actually pretty modest (about 1 percent of GDP over the over the next 75 years)—and, as I trace below, easily addressed by modest reform.
Entitlements are out of control. The slippage here is both persistent and intentional. The combined costs of Medicare, Medicaid, and Social Security are employed to argue for cuts in all three. But, however you stack up those lines, the trajectory of Social Security as a share of GDP is essentially flat. The problem with entitlements is not—as it is often implied—the fact that they are entitlements. The problem is that we spend twice as much per capita on health care than any of our peers. Using that fact to argue for cuts in Social Security is like responding to a hike in gas prices by throwing out your bicycle.
With the debate framed in this way—in which social security is off the rails, out of control, or just plain broke—the accompanying solutions are almost always confined to real or effective program cuts—including lower monthly benefits cuts (accomplished directly or by tinkering with the calculation of inflation behind the cost-of-living increase) or a higher retirement age.
The alternative—barely voiced outside a few progressive policy shops—is to do what we have always done: adjust the tax rate and tax base to ensure long term solvency. The graphic below summarizes these adjustments over the history of the program. The green line traces the increase in the payroll tax rate since 1937; the blue line traces the increase (in real, inflation adjusted dollars) in the taxable base. So, in 1950, employers and employees split a 3 percent rate on the first $25,000 in earnings; in 1980, the rate was 9.04 percent on the first $69,000 in earnings; in 2013, the rate was 10.6 percent on the first $117,000.
There is substantial room here to raise both tax rates and the tax cap. The tax rate has not budged since the 1990s, and the latest Social Security Trustees report projects wage growth sufficient to absorb a modest increase. And the tax cap (which is indexed to average wage growth) has failed to keep pace with widening inequality. As the share of wages flowing to high earners grows, the social security wage base suffers. Raising that base, or having it kick in again once incomes reach a certain point, would yield little hardship and considerable new revenue. These options, and their impact, are summarized in the graphic below. The red line is spending (outlays) as a share of GDP; the blue line is revenues. The grey area to the left summarizes program history since 1973; the area to the left projects revenues based on a menu of reform options. These options (drawn from a 2010 report by the Congressional Budget Office) are not meant to suggest a clean solution, but simply to illustrate the impact of immediate or staged increases in the tax rate, the taxable wage base, or both. The upshot is clear: if we get past the hyperbolic myths and misconceptions, it actually doesn’t take much to weather the entitlement crisis, ride out the demographic tsunami, or defuse the fiscal timebomb.
A new report from the Economic Policy Institute traces the erosion of job-based health coverage across the last decade—and the slowing of that decline in the last year. Nationally, as EPI Director of Health Policy Research Elise Gould underscores, job-based health coverage has shed almost 14 million non-elderly Americans since 2000. But slow improvement in the national economy over the last year, coupled with key components of the Affordable Care Act (most notably the provision that young adults are allowed to stay on or join their parents health insurance policies) have slowed those losses since 2011.
The other important trend across this decade is the dramatic growth in public health insurance. Medicaid and CHIP (Hawk-I in Iowa) have picked up coverage for at least some of those who have lost job-based coverage. While losses (2000 to 2012) in employer-sponsored insurance were greater among children than among non-elderly adults, for example, the share of children without any coverage actually fell 1.8 percentage points.
Much the same pattern has played out in Iowa. In 2000-2001, Iowa’s rate of job-based coverage for those under the age of 65 was 76.9 percent—one of the highest rates in the nation; by 2011-2012, that had fallen to 64.5 percent—closer to the middle of the pack among states. As the graphic below shows (Iowa is the red dot), this was one of the steepest rates of loss in the nation (coverage down 12.4 percent) and represented a net loss in job-based coverage of over 200,000 non-elderly Iowans. Of those losing coverage, about half (97,075) were working age adults and about half (111,839) were kids (indeed, the share of kids losing job-based coverage in Iowa, at 16 percent, was the highest in the country).