Resources | About EPI

The workers’ think tank: A history of the Economic Policy Institute

Download PDF

EPI stands astride history and shouts, ‘Why not?’ to economists who say, ‘No.’

It says, ‘Where’s your data?’

– Jared Bernstein, Chair of President Biden’s Council of Economic Advisers

For the past four decades, one small and scrappy think tank has played a key role in reviving and reinventing U.S. liberalism, and returning its concerns to the well-being of U.S. workers. By providing the analysis and the hard numbers that demonstrated the failure of the neoliberal policies that Democrats as well as Republicans had turned to after the 1970s, the Economic Policy Institute has helped create an updated version of New Deal economics, while also addressing the concerns of race and gender that the New Deal had largely overlooked. 

Since its founding in the late 1980s, EPI has practiced an empirically rooted institutional economics that documented how unchecked corporate power and its accompanying agenda—the destruction of unions, the offshoring of industry, the suppression of wages, the discrimination against women and minorities—led to the squeezing of America’s middle class, the marginalization of its working class, and the rise in its deaths of despair. 

It’s only recently that EPI’s long-embraced advocacy for policies to rebuild America’s once broadly shared prosperity have begun to be adopted. EPI’s founders—economists Lester Thurow and Barry Bluestone, economic policy writers Robert Reich and Robert Kuttner, former Labor Secretary Ray Marshall and EPI’s founding president Jeff Faux—called for the kind of industrial policy that only now has come into being. For a long time, EPI was the only economic institution to both predict and then document the calamitous effects that corporate trade agreements would have over U.S. industry and workers—a stance that was once almost universally denounced by economists and elite media, and for which it is now not-quite-universally but very widely respected. 

Over the subsequent decades, EPI economists and leaders like Larry Mishel, who revealed the widening gap between rising productivity and stagnating wages, and Josh Bivens, who led the advocates for the massive stimulus that enabled the nation to dodge a recession at the outset of the Biden presidency and lift millions of adults and children out of poverty, have had a profound influence in the fight to create a fairer and more vibrant economy. In recent decades, EPI has also been at the hub of a network of state and regional organizations fighting for more equitable economic policies at the state and local level.

In recent years, EPI has continued to enter many of the nation’s most contentious arguments—providing critically important work that documented the rising corporate profit margins that fueled the nation’s inflation, and revealed the economic impact that draconian abortion restrictions have had on women, their families, and their communities. EPI was also one of the first of the nation’s think tanks to devote major resources to documenting and analyzing the ongoing economic effects of systemic racism and sexism, providing a roadmap for other institutions seeking to undertake their own complementary studies. 

Throughout the years, EPI has also incubated and advanced the careers of many of the nation’s leading economic policymakers, including prominent Biden administration officials and congressional staffers.

Today, EPI is helmed by two powerful women leaders, one a passionate economic scholar and the other a longtime crusader for worker rights. President Heidi Shierholz and Executive Vice President Naomi Walker lead a growing team of economists and researchers focused on bolstering race and gender equity and pushing federal, state, and local governments to lift up workers. In this era of misinformation, EPI continues to provide national and local policymakers and grassroots activists with the numbers and analyses that enable them to push for a more equitable and prosperous America.

How did this organization, which began as a handful of economists who contested virtually every neoliberal tenet in the midst of the Reagan eighties, and revealed the fallacious assumptions that propped up all the received “truths” of the economic establishment, become the most trusted go-to source for the economic policy of the new progressive mainstream? Herein hangs the tale.

We needed data

The origins of EPI

In the 1970s, a new American Right began to take shape. Its seeds were sown in a 1971 paper written for the U.S. Chamber of Commerce, the “Powell memo,” which called on big business to devote much more funding to lobbying, to political campaigns, and to a new infrastructure of policy-oriented think tanks.

Business answered Powell’s call—promoting institutions and policies well to the right of the modest conservatism of the Eisenhower and Nixon administrations. In 1973, direct mail pioneer Paul Weyrich and beer baron Joseph Coors founded the Heritage Foundation. In 1977, oilman Charles Koch provided the funding for a start-up libertarian think tank, the Cato Institute. Emboldened by the decade’s run of stagflation, which conservatives attributed to the failures of Keynesian economics and too much liberal spending, Milton Friedman promoted an anti-government libertarianism, and Ronald Reagan won the White House proclaiming that government was the problem.

A softer version of these doctrines seeped into the Democratic party as well. A new generation of Democrats, representing more affluent, historically Republican suburban districts that went Democratic in the wake of Watergate, fashioned themselves as post-New Deal pragmatists—liberal on social issues, not very liberal on bread-and-butter economics. For its part, the Carter administration deregulated a host of industries in the belief that empowered markets would cure whatever ailed the economy. The cult of the market grew so pervasive that Ray Marshall, Carter’s Secretary of Labor (and one of the few liberals in his administration) recalls that “even our economists at the Labor Department were caught up in free market fundamentalism.”

In the eighties, the movement toward deregulated markets became a stampede. (The government did intervene in the market on behalf of corporations and the wealthy, however, through measures like extending copyrights and patents). Freed from government’s guardrails, Wall Street preached the gospel of profits über alles, and corporate America responded by doing everything in its power to maximize its profits by minimizing its labor costs. Emboldened by Reagan’s mass firing of striking air traffic controllers, CEOs tore up the New Deal’s social contract between capital and labor and began mass firings of their own workers when they opted to strike, as well as routinely disrupting and defeating organizing campaigns through tactics that violated the National Labor Relations Act but went effectively unpunished. They offshored work to low-wage nations; they sent work out to low-wage sub-contractors; they employed low-wage temps—and they wielded these strategies as ways to compel their remaining U.S. employees to work for less, for diminished or no benefits, for crazy hours. 

The Democrats’ drift rightward continued apace. The Brookings Institution, Washington’s most venerable think tank, shifted during this period from espousing a mildly center-left economic perspective to one that was more center-right. Market-oriented “Atari Democrats” championed the rise of the deregulated tech industry; the Democratic Leadership Council promoted Democrats—most notably, Bill Clinton—who combined a faith in markets with more centrist stances on social issues. Behind this move rightward was a shift in the party’s funding base. As Thomas Edsall documented in his 1984 book The New Politics of Inequality, the financial contributions to all 1982 Republican congressional candidates came from right-wing institutions and organizations (corporate and conservative PACs) over left-wing institutions and organizations (union and liberal PACs) by a ratio of 33-to-1. The right-to-left ratio of contributions to the Democrats in 1982, Edsall found, was 1-to-1. 

This was the terrain on which—and against which—the Economic Policy Institute was founded. 

“By the mid-eighties, there was a lot of discussion among liberals about the lack of an intellectual infrastructure, of institutions on the left that would provide progressives with the kind of information and analysis that the right had developed by that time,” Robert Reich recalls. “There was literally nothing, a spectacular lack of thinking outside the neoliberal box.”

Reich was one of several progressive economists, public intellectuals and activists who began discussing filling that void with a think tank that would provide them with the kind of information and analysis they could wield in the battle of ideas—above all, hard data on the consequences that deregulating U.S. capitalism had on the U.S. people. He was soon to join such notable dissidents from the zeitgeist as economists Barry Bluestone and Lester Thurow, economics writer Robert Kuttner, former Carter administration Labor Secretary Marshall, and progressive policy proponent Jeff Faux in discussions on how to establish a Washington-based think tank. They agreed that Faux was the person with the time, energy, smarts, and track record to build it and lead it. 

Faux brought to the task a combination of Washington experience, having run a grassroots economic development organizing division of Lyndon Johnson’s War on Poverty in the Sixties, and economic policy development, having worked alongside Gar Alperovitz on such projects in the Seventies. In the early eighties, he’d collaborated with longtime progressive activist Roger Hickey in an effort to set up an institution that would promote industrial policy – a common concern of EPI’s eventual founders as plant closings ballooned in the early years of Ronald Reagan’s presidency. Faux’s efforts proved to be about 40 years too early (it’s only in the past couple of years, as trans-oceanic supply chains broke under the weight of a global pandemic, that official Washington came to realize the necessity of industrial policy). Sufficient funding for that Faux-Hickey venture was not forthcoming, but as the Reagan administration began dismantling as much of the New Deal order as it could get away with, the need for a progressive economic think tank grew even more urgent.

At the same time, the unions bearing the brunt of offshoring and heightened corporate and governmental opposition to their very existence saw declines in membership, which meant declines in their budgets. At their height, unions like the United Auto Workers had what nearly had amounted to economic think tanks of their own, but those days were either going or gone. “They were cutting research departments and policy staffs,” Faux recalls. “They weren’t hiring policy economists as they had in the past; what remained was focused more on collective bargaining.” 

Ray Marshall, the one Carter administration alumnus who understood the importance of unions, had been “prevailed upon,” he recalls, “by some foundations to form a think tank of some kind, which became the National Policy Center. But the work was too broad; we needed to narrow the focus to economic policy. Around that time, Jeff was thinking of forming EPI. I thought it would be a good idea to discontinue the National Policy Center. It had done good work, but what Jeff was planning was better.”

Faux had known Marshall when he’d worked in the War on Poverty. “I went to Ray,” Faux says, “even before meeting with Bob [Kuttner] and Barry [Bluestone]. He was a great asset. He was critical in expanding our network beyond our group of liberal economists.” 

Marshall proved particularly critical in lining up EPI’s seed money, which came from labor. The first union to back Faux’s proposal was AFSCME, whose policy director, Rob McGarrah, brought it to the attention of Gerry McEntee, who’d just been elected AFSCME’s president. McEntee immediately understood the importance of the project and agreed to help organize other union leaders. Marshall and Bluestone then accompanied Faux to a meeting with Lane Kirkland, president of the AFL-CIO, and helped persuade Kirkland to sign off on the project. 

With that, McEntee invited a group of union presidents—the UAW’s Owen Bieber, the Steelworkers’ Lynn Williams, the Clothing and Textile Workers’ Jack Sheinkman, the Machinists’ William Winpisinger, SEIU’s John Sweeney, and the Communication Workers’ Morty Bahr—to a dinner where Faux presented the proposal. (Rich Trumka, then President of the Mineworkers, couldn’t attend the meeting, but was enthusiastic about EPI from the beginning and quickly joined the group. He later became Chair of the EPI Board when he succeeded Sweeney as president of the AFL-CIO.) 

“You guys are getting killed,” Faux told the presidents. “The Heritage Foundation had an 800-page plan for Reagan. We had nothing for the Democrats.” (Indeed, the main economic policy advanced by the Democrats’ 1984 presidential nominee, Walter Mondale—raising taxes to reduce the growing budget deficit—was the brainchild of Goldman Sachs executive Robert Rubin.)

Each of the union presidents then made a three-year pledge of support for an economic think tank that would center its research and analysis on workers and their families.

From the beginning, the labor movement accepted that EPI had to be independent. There were a few times when one union or another was unhappy with EPI’s analysis. But with only minor exceptions, they understood and accepted that EPI would never establish or retain its credibility if it couldn’t present its research even when it didn’t show what its funders wished. Despite the caricature of unions as narrowly self-interested institutions, they understood not just the importance of EPI’s mission but of its independence as well. Moreover, they continued to support EPI financially even as their memberships and dues declined and times grew harder for them. 

It was 1986, the middle of the dismal eighties. Faux came up with a “vanilla name” for the outfit while Hickey found an office. The founders knew what they wanted and needed from the fledgling outfit. “We were all writing about structural changes to the economy,” says Reich, “but we needed data; we needed more and better analysis.” 

All of the Institute’s founders were “institutionalists” in their approach to economics, following in the tradition of John Kenneth Galbraith, Robert Heilbroner, Adolph Berle, John R. Commons, and Thorsten Veblen. They focused on the structures of power behind the veil of free-market fundamentalist theory that was then dominating economics.

That required them to look at specific sectors and industries. Conventional economics, by contrast, focused on only two perspectives—the individual alone in the market or the entire macro economy. Mainstream economists wouldn’t or couldn’t see anything in between; that would be altogether too messy for economics as they conceived and practiced it, too full of class interests, social bias and politics that were beyond the scope of supply and demand curves. Asked whether the government should consider having a policy that bolstered key industries, Michael Boskin, who then chaired George H.W. Bush’s Council of Economic Advisers, made clear just how far beyond the pale of proper economics and proper economic policy such considerations were. “Potato chips, computer chips,” he said, “what’s the difference?”

The neoliberal policy wonks wouldn’t admit it, but their “data-based” science began with the conclusion that free markets worked best, which their number then “proved.” EPI, by contrast, would plunge into the workings of the actual economy, where power, class, race, and gender shaped actual outcomes. It would look at the economy through the updated versions of the institutionalist lenses that its great institutionalist predecessors had used to describe the actual economies of their own times. It would do so by uncovering and analyzing economic data that conventional economics ignored and dismissed.

Challenging the stranglehold neoliberalism had on economic policymaking seemed impossibly ambitious. “You’ll never make it,” a sympathetic, seasoned Washington economist told Faux. “The establishment has too much money and power for you to overcome.” Moreover, it would require building an institution that could incubate and influence succeeding generations of activist economic analysts. EPI could never count on the corporate and foundation deep pockets that were available to support the conservative agenda of the “Powell memo.”

To have a chance against these odds, EPI had to first establish its credibility by showing that the economics professions own quantitative methodology supported the EPI criticism.

 “Numbers are the currency of economic and political debate in this town,” says Faux. I knew the only way we were going to break through was to be good with the numbers.” 

“That’s when I hired Larry Mishel.”

Good with the numbers

EPI defines itself

In 1987, when Faux hired him as EPI’s initial research director and chief economist, Larry Mishel had already worked for the UAW—the gold standard for union economic policy shops—and then the AFL-CIO’s Industrial Union Division. “Larry was young, quantitative; he liked playing with numbers and understood what we were about,” Faux says. 

What Mishel was about, particularly in EPI’s early years, was providing the numbers that challenged the conventional economic wisdom. “The Commerce Department kept insisting there was no decline in U.S. manufacturing,” Faux recalls, “but that ran counter to everything we were hearing and seeing. Larry dived into those numbers and concluded after a month that they were mismeasuring manufacturing output and ignoring a rise of imported inputs. At first, Commerce denied that they were miscalculating, but then a former BLS [the Bureau of Labor Statistics] productivity expert said that Larry was right.”

“We had to overcome the ‘stigma’ of being a labor-supported think tank,” says Mishel, “so I wanted to do that by paying the closest attention to the data. When my first report, ‘Manufacturing Numbers’ was published in April 1988, the conventional wisdom was that the U.S. had not deindustrialized. I argued that the BEA [the Commerce Department’s Bureau of Economic Analysis] was wrong. And by October, BEA had suspended its data series and committed to revising it.” This also began to dispel a huge misunderstanding of the U.S. economy. 

 “In the world of economic statisticians,” says Faux, “this was a big deal. It showed we were serious about the numbers.”

With its credibility as an organization that could decipher, document and analyze the real economy steadily growing, EPI became a big deal not just for statisticians but for its union backers (the Steelworkers’ Lynn Williams called the union’s EPI contribution “the best investment I’ve ever made”), for foundations (Ford became an early supporter) and liberals more generally. In 1988, each of the Democrats seeking the presidency cited EPI’s numbers in their presentations of where—and how, and why—the economy had gone astray. 

Also in 1988, on Labor Day, EPI published its initial State of Working America (SWA). “It established our mode of working for many years: It did not have anything about policy; we just wanted to overwhelm readers with the data, with 90 tables, that showed the condition of the middle and working classes,” Mishel says. “It became EPI’s brand; it made clear our focus on getting the data right and highlighting what was happening to jobs, wages, job quality and inequality.” 

“For some people on the left, the State of Working America became a bible; it was the icing on our statistical cake,” says Faux. “The press relied on it, congressional staffers used it; when I walked into congressional offices, it would be sitting on the shelf.”

On the subsequent editions of SWA (which EPI continued to publish through 2012), Mishel had a stellar array of EPI co-authors. In the institute’s formative years, he hired a number of young economists and policy advocates—among them, Jared Bernstein, Thea Lee, Heather Boushey, Dean Baker, and John Schmitt—who were to become some of the nation’s most influential policy analysts and formulators. 

Mishel hired Bernstein as a research assistant in 1992. Bernstein could program microdata, and EPI purchased the computer equipment that Bernstein could use to view and analyze BLS survey data.

“It was life-changing for EPI,” Mishel says. “We could get and analyze data” that no one else was looking at. “Bill Clinton was running for president saying, “What you earn depends on what you learn”—that education was the answer for a struggling workforce. But our data was the basis of a front-page New York Times story with the headline ‘Wages for College Grads Not Doing Well.’”

In a sense, the Times story conveyed EPI’s first public salvo against the then-prevailing notion that the nation’s rising inequality was the result of a “skills gap” among U.S. workers. Much of EPI’s work over the past 35 years has centered on demonstrating the real causes of that inequality: slack economies, deunionization, structural racism, and trade accords that undermined domestic living standards—public policies that had been enacted and could therefore be reversed. 

“Larry’s big investment was to buy computer equipment that filled up a whole chunk of a room,” says Bernstein. “I was crunching these big magnetic tapes. We realized that a distributional analysis of hourly wages did not exist. BLS didn’t produce it. So we crunched hourly wage data by income percentile and level of education. At the time, Democrats and the center-left said the remedy for wage stagnation was education. But we crunched, and found that people with four year of college had stagnant wages, too.”

“We were contesting a very individualistic view of what was driving economic inequality,” Bernstein continues, “that the blame belonged to people who weren’t smart enough. That ignored all the macro forces that were shaping the economy, like workers’ institutional bargaining power.” 

“Some of what we were doing was the result of new technology that enabled us to work with microdata from the Current Population Surveys,” says Dean Baker. “It was hard to work with, but Larry made a commitment that we’d do this. No one else was doing this in a timely fashion. Harvard and MIT had the resources to do this—and they didn’t.” 

The 1994 edition of the SWA contained the chart that Mishel says will be engraved on his tombstone. It contains two lines, one representing the rise in productivity since the late 1940s, the other representing the rise in median hourly pay over the same time period. Through the 1970s, the lines showed that productivity and the pay of the typical worker grew in tandem. Beginning in the eighties, however, they showed productivity continuing to rise (albeit more slowly than before), while wage growth tanked.

The gap between productivity and a typical worker’s compensation has increased dramatically since 1979: Productivity growth and hourly compensation growth, 1948–2024

Year Productivity Pay
1948q1   100 100
1948q2 101.2224 100.7585
1948q3 100.7825 100.859
1948q4 102.1177 102.478
1949q1 102.2019 105.3803
1949q2 101.3389 106.6297
1949q3 104.7471 109.0284
1949q4 104.7364 110.3131
1950q1 108.8723 111.8143
1950q2 110.1335 112.3882
1950q3 112.1586 111.6846
1950q4 112.8913 112.1322
1951q1 111.5826 110.2063
1951q2 111.6458 111.2497
1951q3 114.4287 113.5507
1951q4 114.3274 113.8062
1952q1 114.3807 114.4272
1952q2 115.0211 115.8473
1952q3 115.3531 115.8715
1952q4 117.5749 117.4075
1953q1 120.0173 119.7173
1953q2 120.5348 120.7118
1953q3 120.2853 121.8617
1953q4 119.7519 122.7065
1954q1 120.3647 123.7449
1954q2 121.8411 124.5355
1954q3 123.7844 125.3718
1954q4 126.1801 127.1677
1955q1 128.206 127.8147
1955q2 130.5995 129.5961
1955q3 131.4314 130.7226
1955q4 131.8896 132.3935
1956q1 132.5184 134.2646
1956q2 132.3767 134.5564
1956q3 132.4324 134.9433
1956q4 134.4478 135.9284
1957q1 135.9806 136.5261
1957q2 135.4518 136.9394
1957q3 135.7934 137.2891
1957q4 136.6831 138.6305
1958q1 135.1496 137.6791
1958q2 136.422 138.1341
1958q3 139.7677 138.9453
1958q4 142.0154 140.1919
1959q1 143.5943 142.1337
1959q2 144.3773 143.1391
1959q3 144.6854 142.9913
1959q4 144.6406 143.4921
1960q1 148.9546 145.2161
1960q2 146.0806 145.7559
1960q3 146.3742 146.375
1960q4 145.1088 146.6526
1961q1 146.7386 147.7571
1961q2 151.2943 148.604
1961q3 152.9593 148.6978
1961q4 154.7853 149.9996
1962q1 155.8676 151.192
1962q2 155.7629 151.7841
1962q3 158.4039 153.2191
1962q4 159.1433 153.5229
1963q1 160.1264 154.2017
1963q2 161.2386 155.2373
1963q3 164.0396 155.5606
1963q4 164.8935 156.7498
1964q1 167.6906 158.5101
1964q2 167.2588 159.062
1964q3 169.84 160.773
1964q4 168.174 160.1893
1965q1 170.5501 161.3186
1965q2 170.8476 162.1148
1965q3 175.3211 163.6557
1965q4 177.9659 164.5323
1966q1 178.8131 164.8556
1966q2 178.4471 165.9094
1966q3 178.9881 166.1527
1966q4 179.7122 166.2315
1967q1 180.3977 167.3538
1967q2 181.2156 168.3047
1967q3 181.8779 169.2548
1967q4 182.6768 169.8076
1968q1 186.6931 171.2016
1968q2 188.7502 172.4089
1968q3 188.2602 172.8439
1968q4 189.137 174.2238
1969q1 190.1281 175.703
1969q2 189.1847 175.8024
1969q3 190.774 176.8568
1969q4 189.2837 177.1482
1970q1 190.5522 177.0935
1970q2 192.3818 177.5397
1970q3 195.7578 179.2568
1970q4 194.3567 179.5898
1971q1 200.5013 181.8354
1971q2 201.3067 182.8491
1971q3 203.7236 184.2475
1971q4 203.0908 185.6951
1972q1 205.1834 190.3615
1972q2 209.7812 192.4523
1972q3 211.5681 193.9578
1972q4 213.4589 194.0056
1973q1 215.3432 194.6084
1973q2 214.4144 192.921
1973q3 211.9744 192.4998
1973q4 212.3932 190.784
1974q1 207.2693 188.3682
1974q2 207.7652 187.9351
1974q3 206.6166 188.3849
1974q4 207.6962 187.493
1975q1 209.3333 187.2972
1975q2 212.9105 188.4237
1975q3 215.0174 187.8178
1975q4 215.4609 187.5831
1976q1 217.304 188.2701
1976q2 218.8501 189.7046
1976q3 218.2819 190.3936
1976q4 218.8123 191.1914
1977q1 219.5919 191.9768
1977q2 219.4248 192.7663
1977q3 221.3454 193.6298
1977q4 221.118 194.5295
1978q1 220.3615 196.0737
1978q2 223.5137 196.0318
1978q3 223.3659 195.9118
1978q4 224.8815 196.4933
1979q1 222.5492 196.1878
1979q2 221.5493 194.0663
1979q3 219.8124 193.3174
1979q4 217.7292 191.7634
1980q1 216.6058 189.3787
1980q2 213.3814 188.9364
1980q3 213.2651 188.5993
1980q4 215.2463 188.4181
1981q1 218.8552 187.8761
1981q2 217.3573 188.0737
1981q3 219.7654 187.9574
1981q4 216.5601 187.2807
1982q1 216.0183 188.1088
1982q2 215.7346 188.6548
1982q3 215.4963 188.3419
1982q4 216.5423 187.8983
1983q1 219.368 190.0454
1983q2 221.9851 189.4936
1983q3 223.6938 189.4451
1983q4 225.8788 189.3539
1984q1 226.15 188.8803
1984q2 227.6229 188.5421
1984q3 228.9141 188.4745
1984q4 229.4068 188.3055
1985q1 230.6319 188.1152
1985q2 230.6769 188.1922
1985q3 233.286 188.4558
1985q4 233.0346 188.1202
1986q1 234.9234 188.3092
1986q2 237.875 190.2938
1986q3 238.3172 189.9818
1986q4 237.4557 190.2002
1987q1 235.2406 188.7349
1987q2 235.6077 187.6395
1987q3 235.8354 187.2748
1987q4 237.7856 187.3136
1988q1 238.4391 187.2439
1988q2 238.4714 187.03
1988q3 239.4969 186.7535
1988q4 240.3885 187.5959
1989q1 240.5406 187.7474
1989q2 240.6685 186.9796
1989q3 241.6523 187.5564
1989q4 241.2308 188.0957
1990q1 242.147 187.4809
1990q2 244.0834 187.3034
1990q3 242.6241 185.8919
1990q4 238.6094 184.7695
1991q1 239.516 185.2578
1991q2 243.2864 186.3642
1991q3 245.2206 187.2
1991q4 246.0497 187.3464
1992q1 249.9206 187.8119
1992q2 251.6125 188.1735
1992q3 254.1083 188.4343
1992q4 255.623 187.9503
1993q1 254.0963 189.0897
1993q2 252.7075 188.2978
1993q3 253.441 188.4688
1993q4 255.1901 188.6102
1994q1 256.5796 189.8571
1994q2 256.1244 189.2323
1994q3 254.8452 188.9591
1994q4 257.284 189.1304
1995q1 255.981 188.5205
1995q2 256.0143 188.251
1995q3 255.7288 188.3237
1995q4 257.5326 188.6312
1996q1 259.1014 188.4575
1996q2 259.8923 188.263
1996q3 259.7266 188.8626
1996q4 259.6332 188.6513
1997q1 259.8197 188.9004
1997q2 263.016 190.1508
1997q3 264.9808 191.0714
1997q4 265.685 192.413
1998q1 266.4552 193.7973
1998q2 267.161 195.1811
1998q3 270.3272 195.8496
1998q4 271.7016 196.6668
1999q1 274.6505 197.8043
1999q2 273.2147 198.0771
1999q3 274.6204 198.6224
1999q4 276.8704 198.66
2000q1 274.9271 198.4793
2000q2 279.5904 199.5756
2000q3 279.5187 199.9269
2000q4 282.4208 201.4682
2001q1 280.1394 201.9444
2001q2 283.4774 203.1679
2001q3 283.9923 204.4302
2001q4 287.5201 206.543
2002q1 292.5072 207.7645
2002q2 292.2707 207.8738
2002q3 293.8224 209.065
2002q4 294.1808 210.6484
2003q1 295.9907 210.7718
2003q2 301.746 212.1644
2003q3 305.8693 211.9317
2003q4 309.1752 211.6063
2004q1 309.1799 211.5796
2004q2 311.7492 211.3924
2004q3 313.4896 211.5273
2004q4 314.7269 210.848
2005q1 318.1473 211.7963
2005q2 318.6223 212.1476
2005q3 317.9135 210.6053
2005q4 318.5955 210.6313
2006q1 321.1005 211.0853
2006q2 320.3473 211.6171
2006q3 318.5199 211.573
2006q4 321.7783 214.2278
2007q1 323.3931 214.0092
2007q2 323.2545 214.5643
2007q3 325.1677 215.287
2007q4 324.6322 214.3371
2008q1 321.1595 214.1922
2008q2 321.2409 213.7276
2008q3 318.8241 212.5401
2008q4 323.8696 219.1567
2009q1 329.3948 224.0722
2009q2 333.0931 224.173
2009q3 335.7491 223.7632
2009q4 339.1306 223.7858
2010q1 340.7556 225.6749
2010q2 342.6806 226.5948
2010q3 346.6712 226.931
2010q4 346.7416 226.4612
2011q1 343.0123 225.4128
2011q2 341.196 223.7981
2011q3 340.1887 222.9265
2011q4 340.6023 223.1113
2012q1 342.1459 221.6012
2012q2 343.9474 222.1198
2012q3 341.9148 222.2
2012q4 341.6058 221.5144
2013q1 344.4001 223.0594
2013q2 345.5619 224.7863
2013q3 346.9207 225.221
2013q4 349.0968 225.711
2014q1 346.2669 225.4386
2014q2 348.0385 225.2024
2014q3 351.0635 225.6106
2014q4 350.6203 226.9267
2015q1 354.2249 229.7838
2015q2 355.8439 229.4826
2015q3 356.8215 229.8854
2015q4 355.7143 231.6308
2016q1 356.702 233.2616
2016q2 356.4604 232.8107
2016q3 357.4862 233.3005
2016q4 359.9804 233.1388
2017q1 359.9192 232.9396
2017q2 360.0144 234.1445
2017q3 363.5001 235.0553
2017q4 364.8114 234.5054
2018q1 366.2557 235.0642
2018q2 366.7028 235.7902
2018q3 367.7332 236.9186
2018q4 367.5066 238.5109
2019q1 369.4537 239.6746
2019q2 370.7367 240.0619
2019q3 373.9044 241.6013
2019q4 375.7408 241.8278
2020q1 375.401 243.091
2020q2 387.6417 255.7427
2020q3 393.6397 250.122
2020q4 391.0226 249.9645
2021q1 395.5453 250.3091
2021q2 395.76 249.1343
2021q3 394.0284 249.0241
2021q4 395.3104 248.0879
2022q1 389.8198 246.2414
2022q2 386.0947 243.8393
2022q3 384.7867 243.7039
2022q4 386.6525 244.4187
2023q1 386.7251 244.9451
2023q2 387.4294 246.1366
2023q3 391.6656 246.7619
2023q4 393.3331 247.9148
2024q1 393.8268 248.1665

 

ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Data are for compensation (wages and benefits) of production/nonsupervisory workers in the private sector and net productivity of the total economy. “Net productivity” is the growth of output of goods and services less depreciation per hour worked.

Source: EPI analysis of unpublished Total Economy Productivity data from Bureau of Labor Statistics (BLS) Labor Productivity and Costs program, wage data from the BLS Current Employment Statistics, BLS Employment Cost Trends, BLS Consumer Price Index, and Bureau of Economic Analysis National Income and Product Accounts.

Copy the code below to embed this chart on your website.

Conventional, and at the time, all but consensual, economic theory said that wage increases track increases in productivity; and if wages stagnate, that’s because productivity is ebbing, too. Those whose wages were not faring well, in this view, had skills inadequate for the information age. This is what Charles Schultze, a Brookings Institution fellow who had chaired President Carter’s Council of Economic Advisers, and others at Brookings were arguing: Lowering deficits, increasing savings, and reversing the productivity slowdown were what workers needed.

In a sense, Schultze was merely restating President John F. Kennedy’s famous assertion that “a rising tide lifts all boats.” The EPI chart, however, suggested that the tide’s boat-lifting mechanisms had been a consequence of institutional economic factors that loomed large during the years of Kennedy’s early sixties presidency, such as the bargaining power of a substantially unionized workforce in a not yet globalized economy. That power that had waned as union membership accelerated its decline in the eighties, and in the face of competition from low-wage nations. In the absence of those mid-century factors, the emphasis on growth that characterized both the Clinton administration and Brookings did not necessarily translate to economic gains for the median worker. Worse yet, the economic orthodoxy at Brookings and the White House held that expansive fiscal and monetary policies to engender full employment were impermissible because they ran the risk of inflation. That meant not only that the tide wouldn’t lift all boats but also that the tide itself wouldn’t amount to much. EPI countered that the actual numbers showed that fears of inflation were excessive, reflecting instead the anxieties of Wall Street bondholders.

EPI built the chart around a key distinction that lies at the center of much of its work. The focal point was not on the experience of the “average” worker, but instead the experience of what statisticians would call the “median” worker and what EPI usually refers to as the “typical” worker. The median worker’s wage is the wage of the worker right in the middle, the one earning more than the bottom half of workers but less than the top half of workers. In an economy with U.S. levels of inequality, the wage earned by this median worker is a much better description of the economic circumstances facing the typical than the much more commonly cited “average” wage, which is pulled up by the wages of a relatively few, very high earners.

The fact that the chart tracked median hourly wages rather than, say, median annual incomes or average wages (which include all earners from CEOs to janitors) is further testimony to EPI’s focus on profiling the lived experience of working people. Workers may be able to augment their income by working overtime or taking a second (or third) job (or, today, a gig), but they do so at the expense of other aspects of their lives, like spending time with their children and being engaged with their communities.

That’s why the hourly wages of a typical or median worker offer the most revelatory and telling measure of the state of worker well-being.

According to Baker, “no one else was even looking at the people who made the median wage, and so didn’t understand how much they were hurting. When Alicia Munnell [then assistant secretary for economic policy in Bill Clinton’s Treasury Department] met with Jared, she’d only been looking at average wages, not median wages. So, she was surprised.” 

By rooting economic policy in improving the lot of median-wage Americans, EPI was also rooting economic argument in empirical data rather than in the aggregate approach (which focused on the GDP) that almost completely dominated macroeconomics and economic reporting at the time. In that sense, EPI prefigured by more than 20 years the revival of empiricism in the practice of academic economics sparked by the work of Thomas Piketty and Emmanuel Saez. In going first, EPI was challenging what was then the otherwise unchallenged economic mainstream, its fundamental methodology and its shibboleths. 

“Any time someone wants to do something progressive, a bunch of economists say why you can’t,” says Bernstein. “EPI would always get into that argument and ask, ‘Are you sure? Where’s your data? If we raise the minimum wage by five cents, there will be job loss? That’s a testable hypothesis.’” 

“That’s the thematic behind everything we did.” 

That thematic informed the work of EPI economist Dean Baker, and later, of EPI economist Josh Bivens, both of whom charted the rising share of the nation’s income going to capital and the declining share going to labor. It also informed EPI’s revelations about the nation’s wealthiest residents, including CEOs—something that most other economists considered impolite (or at least impolitic) to focus on. In the mid-nineties, EPI began developing the data and the chart that displayed median CEO pay and median worker pay as far back as the early 1960s. Since 2012, EPI has issued annual reports showing the ratio between the two. As this chart shows, U.S. CEOs went from making 21 times that of the typical worker in 1965 to 344 times that in 2022. 

CEOs make 344 times as much as typical workers: CEO-to-worker compensation ratio, 1965–2022

year Realized CEO compensation Granted CEO compensation
1965 20.7 15.4
1966 21.9 16.3
1967 23.2 17.2
1968 24.4 18.1
1969 24.1 17.9
1970 23.9 17.7
1971 23.6 17.5
1972 23.3 17.3
1973 23.0 17.1
1974 24.6 18.2
1975 26.2 19.4
1976 27.7 20.6
1977 29.3 21.7
1978 30.9 22.9
1979 33.6 24.9
1980 36.2 26.9
1981 38.9 28.8
1982 41.6 30.8
1983 44.3 32.8
1984 47.0 34.8
1985 49.6 36.8
1986 52.3 38.8
1987 55.0 40.8
1988 57.7 42.8
1989 60.4 44.7
1990 75.9 56.3
1991 91.5 67.9
1992 107.1 79.4
1993 108.2 99.2
1994 87.2 116.5
1995 117.4 130.5
1996 150.3 176.7
1997 223.1 233.0
1998 307.8 304.1
1999 278.4 288.7
2000 380.9 395.8
2001 213.4 326.3
2002 185.4 233.1
2003 229.1 226.4
2004 262.2 231.4
2005 318.6 244.8
2006 323.6 237.7
2007 325.9 242.0  
2008 199.5 216.5
2009 175.0 181.2
2010 209.4 203.0
2011 238.1 210.9
2012 367.4 206.0
2013 318.6 210.7
2014 317.6 219.1
2015 318.8 216.6
2016 271.2 220.0
2017 299.2 232.5
2018 289.7 228.3
2019 311.0 228.0
2020 372.3 220.3
2021 389.0 254.3
2022 344.5 229.0
ChartData Download data

The data below can be saved or copied directly into Excel.

Notes: Average annual compensation for CEOs is for CEOs at the top 350 U.S. firms ranked by sales. Typical worker compensation is the average annual compensation (wages and benefits of a full-time, full-year worker) of production/nonsupervisory workers in the industries that the top 350 firms operate in.

Notes: Average annual compensation for CEOs at the top 350 U.S. firms ranked by sales is measured in two ways. Both include salary, bonus, and long-term incentive payouts, but the “granted” measure includes the value of stock options and stock awards when they were granted, whereas the “realized” measure captures the value of stock-related components that accrues after options or stock awards are granted by including “stock options exercised” and “vested stock awards.” Projected value for 2022 is based on the percent change in CEO pay in the samples available in June 2021 and in June 2022 applied to the full-year 2021 value. “Typical worker” compensation is the average annual compensation (wages and benefits of a full-time, full-year worker) of production/nonsupervisory workers in the industries that the top 350 firms operate in.

Source: Authors’ analysis of data from Compustat’s ExecuComp database, the Bureau of Labor Statistics’ Current Employment Statistics data series, and the Bureau of Economic Analysis NIPA tables.

Copy the code below to embed this chart on your website.

By the early ‘90s, EPI had begun to resemble a small, bustling, out-of-the-mainstream but relentlessly accurate economics department. Faux and Mishel’s early hires included Bernstein, Baker, Thea Lee, Bill Spriggs, John Schmitt, Eileen Appelbaum, and Heather Boushey, all of whom were to play major roles in eventually convincing center-left and liberal policymakers and advocates to adopt more egalitarian and social democratic economic perspectives. This recruitment of young, obscure economists who were then to play key roles in reshaping both public and elite understanding of the economy continued when Mishel succeeded Faux as EPI president in 2002, as exemplified by Mishel’s hire that year of Josh Bivens and, a year later, Elise Gould. Bivens would go on to become research director and chief economist, while Gould would become EPI’s key expert and commentator on wages—taking those reins from Mishel—in the mid-2010s. 

“It was an exciting place to work,” recalls Baker. “We shared what we’d found in weekly seminars, but on any day, you could just go down the hall to find out something new.” 

It helped that EPI practiced what it preached. It encouraged the creation of the staff union. And Faux—an employer—was asked to speak at the national convention of the parent union’s national convention.

From its earliest years, EPI put full employment at the center of its argument for reducing inequality and meeting the economic needs of the lower and middle classes. “I spent years in the mid-’90s working with Dean on full employment,” recalls Bernstein, “on how the structure of the economy led to slack labor markets most of the time.” The way to take up that slack, EPI argued, was public investment. Before the 1992 Democratic presidential primaries, Faux argued that the party needed to address three deficits: fiscal, trade, and public investment. EPI produced papers calculating the job impact of infrastructure and other public investment; Faux co-authored a book with Todd Schafer on the topic that outlined a policy agenda on which Democrats could run and win. 

“We always said that full employment was essential to combating inequality, and our work showed it was essential in combating racial and gender inequality in particular,” Mishel says. The tight labor markets of a full-employment economy bring employment opportunities and wage increases to those groups historically at the economy’s margins when it’s not running at full strength. EPI’s understanding of full employment’s salutary effects—influenced by such earlier advocates as Michal Kalecki and Michael Harrington—underpinned its support for more stimulus spending to revitalize an economy walloped by the 2008 financial panic. The same understanding informed EPI’s opposition to the fiscal austerity and budget cuts proposed by the 2010 Bowles-Simpson Commission and many mainstream Democrats and center-left think tanks in order to reduce the deficit, even as unemployment levels remained high in the wake of the Great Recession. 

Through the first five presidencies that overlapped EPI’s existence—Reagan’s Clinton’s, both Bushes,’ and Obama’s—EPI’s numbers came to be viewed as consistently reliable by growing numbers of Democratic elected officials and by policy wonks and economists of all tendencies. Its policy recommendations, however, were routinely rejected by the same political and economic establishments.

Bill Clinton’s 1992 campaign mantra that Americans were working harder for less, for example, came from EPI’s numbers on stagnant wages. As president, though, Clinton surrounded himself with Wall Street financiers and the pro-business Democratic Leadership Council who blamed stagnant wages on the workers themselves—their lack of skills and education, and their unwillingness to relocate.

In an economic policy briefing a month after Bill Clinton’s election, Robert Rubin of Goldman Sachs, who went on to dominate Clinton’s economic policy, used an early version of an EPI chart to show how bad things had been under Republicans. But then, to Faux’s shock, Rubin concluded that the chart showed that the answer to the wage problem was more education and training, the exact opposite to what the chart implied. 

This emphasis on workers’ ostensible deficiencies enabled mainstream Democrats to vow they would “do something” for workers, even if that just meant subsidizing worker training for corporations without engaging issues of diminishing worker power and the increasing maldistribution of income. 

This establishment refusal to acknowledge the policy implications of the very data the establishment did not contest was a common occurrence throughout EPI’s early years. When centrists praised EPI, it came with an invariable caveat: Their numbers are honest and accurate even though I disagree with their conclusions. Of course, EPI’s policy conclusions were based on the numbers. 

By the late 2010’s, however, EPI’s once heretical economics had become the common sense of most Democratic policymakers and Democratic voters. When Joe Biden became president and confronted an economy still in pandemic shutdown, he proposed the American Rescue Plan, which the Democratic Congress then enacted. The plan kept millions of Americans from plunging into poverty, saved hundreds of thousands of small and mid-sized businesses, and created the fastest recovery, with the greatest gains for the lowest-paid workers, in the nation’s history.

“For decades, the government had worried about inflation and ignored workers and wages when the economy was in trouble,” Mishel says. “This was a major change”—one that EPI, often as a voice in the wilderness, had been persuading the Democrats to make virtually since its inception. 

The leftward movement of Democratic party on economic fundamentals from Clinton’s presidency to Biden’s is due chiefly to Americans’ growing experience and awareness of the economy’s dysfunctions and its towering levels of inequality. Occupy Wall Street, the Fight for 15 and Bernie Sanders’ presidential campaigns all played a crucial role in boosting that awareness. But the movement of the centrist and center-left economists who play a large role in shaping economic understanding, discourse, and policy is in no small part due to the credibility of EPI’s work since the mid-1980s.

“I first became aware of them in the early ’90s,” says Harvard economist Lawrence Katz. “Larry and Jared and the State of Working America put them on the map for an academic like me. Clearly, the empirical work they presented was very carefully done; the comprehensive and transparent presentation of publicly available facts made them clear and credible.” 

Even when EPI’s politics were far from the economic mainstream, says David Card, a Nobel laureate in Economics who chairs the economics department at UC Berkeley, “I’ve never heard anyone say they manipulate the data from the population surveys. They’re invariably reliable.” 

Another economist who attests to EPI’s reliability is Jason Furman, who once was the chief economist for the Hamilton Project, which was founded (and funded) to advance a Robert Rubinesque perspective on the economy. “In 2004, I was working for the [John] Kerry [presidential] campaign, and I needed some minimum wage numbers,” Furman says. “I’d thought of EPI as some lefty organization, but when I called an academic labor economist, he said Jared Bernstein runs the data better than anyone. That was the first time I learned to rely on the data analysis they produce, and I’ve relied on and used their data for the past 20 years, even when I may disagree with them on a particular position. On the state of working America, on CEO pay, on a lot of key issues, they don’t make mistakes. They’re really good.” 

“The transformative nature of the data they put out was very important,” says Card. “That graph—showing poor wage performance relative to productivity—really helped build the new mainstream.” 

Among economists, EPI’s reputation has extended well beyond the various factions of the Democratic party. Kevin Hassett, who directed economic policy at the American Enterprise Institute, served as chief economic adviser on the McCain and Romney presidential campaigns, and chaired President Trump’s Council of Economic Advisers from 2017 to 2019, called EPI’s work, “Really reliable, professional, carefully thought-out.” 

Faux had been right. EPI succeeded because it had been good with the numbers—better, in fact, than just about anyone else.

Everybody wins except for most of us

By the mid-1990s, EPI was known chiefly for two things: its revelation that wages had stagnated even as productivity was rising, and its critique of and opposition to the free trade agreements—at that point, chiefly NAFTA —that Republicans and corporate Democrats were promoting and enacting. 

The North American Free Trade Agreement had first been proposed by Ronald Reagan in the 1970s but was dropped for lack of support in the Congress when he became president. When George H.W. Bush became president, NAFTA’s time had come. Bush negotiated a three-way deal with Mexico and Canada that would allow U.S. capital to build factories in Mexico and ship the production to the U.S. virtually duty free. In effect, the deal forced Americans to compete with exploited workers whose wages came to one-tenth of theirs.

NAFTA was something new. Until it became a governmental priority in the last years of the Bush presidency, “our only trade agreements were with [the comparable economics of] Canada and Israel,” says Thea Lee, whom EPI hired in 1991 and put on the trade desk the following year. “We’d never had a trade agreement with a developing country like Mexico.”

Following his election in 1992, Bill Clinton made passing this Reagan/Bush initiative a major priority of his presidency. The proposal was anathema, however, to much of the Democratic base—not only to unions (particularly those that represented workers in manufacturing), but also to environmental groups who deplored its absence of any enforceable environmental standards. Liberal Democratic House members, especially those from manufacturing-heavy districts, led by Michigan’s David Bonior, strongly opposed the deal.

Yet NAFTA was almost universally praised by the economic policy pundits, whose catechism held that free markets would be “win-win” for all three countries. Mexicans, the pundits argued, would take low-skilled jobs in the production chain, while Americans would move up to even higher-paying technically advanced work.

But as EPI saw it, this completely ignored the vast institutional differences between these two economies. It also hid the implicit racism that underpinned the conventional wisdom—that Mexicans were incapable of meeting the skill levels of U.S. workers.

At the time, Mishel had already compelled the BEA to concede that the level of manufacturing in the U.S. had declined. “Offshoring was a visible explanation for that,” Faux recalls. “So, we jumped into that. As for academics and think tanks—well, there wasn’t anybody else.” 

“EPI was already grappling with how much of the emerging wage gap and deindustrialization was caused by trade,” says Roger Hickey. “There weren’t many experts in this field; it was really only EPI that even considered it to be a field.”

One of those very few experts was Richard Rothstein, who’d had to deal with factory relocation to Mexico when he’d headed the clothing workers union in Southern California in the 1980s. With Faux, Rothstein co-authored an EPI paper entitled “Fast Track Fast Shuffle,” which pointed out not just the deficiencies in NAFTA, but also in the “fast track” process in which the tri-national agreement could not be subjected to congressional amendment but rather be presented as a fait accompli for an up-or-down vote.

As far as nearly every economist was concerned, the merits of free trade were self-evident. “The economic establishment treated us as complete ignoramuses,” Mishel recalls. It was ignoring the fact that “basic economic theory says that trade will lead to inequality, and you don’t get the benefits of trade unless it does,” Mishel adds. “That meant you had to deplete the number of jobs of workers without college educations, or lower their pay. But [NAFTA’s advocates] wouldn’t acknowledge that. [President] Clinton’s economic advisers would deny it was so.” 

After his paper with Rothstein was published, says Faux, “I became our trade person. But I didn’t even have an M.A. in economics. So, we hired Thea.” 

“In 1991, I was basically not finishing my dissertation [in economics at the University of Michigan], and teaching at UDC [the University of the District of Columbia],” says Thea Lee. “I’d read about EPI in Dollars and Sense, where I’d been an editor. A job opened up there as a research assistant, but Jeff and Larry decided I was overqualified, so Larry created a trade economist job.” 

Initially, the scope of that job was confined to two industries already moving abroad: textiles and steel, on which Lee worked with the University of Maryland’s Rob Scott. “Then NAFTA was thrown into the mix,” she recalls, as it became clear that Bush was determined to sign the accord with Mexico and Canada, and that Clinton, his successor, was equally determined to push it through Congress. “I became part of a group with Richie [Rothstein] and Mark [Levinson, then the clothing workers union’s economist] insisting that trade deals needed enforceable labor standards.”

“We were treated as nut cases. That made the job fun.”

The trade “experts” whom Lee debated, by contrast, didn’t find it fun at all. “I was a no-name grad student,” Lee says, “who ended up debating Carla Hills [the U.S. Trade Representative under Bush] and James Tobin [Yale’s Nobel laureate economist]. “She debated Tobin at Yale and killed him,” Mishel recalls. “These guys just assumed that free trade must be good. She’d say, ‘Have you read the treaty? It’s just about protecting high profits for our corporations’ foreign investments.’ She called out the special ‘courts,’ which weren’t part of any country’s judiciary, that would rule on disputes. She was a star. She was great on TV.” 

“We were never comfortable with a nationalist argument,” Lee says. “We made clear that NAFTA didn’t benefit—wasn’t designed to benefit—Mexican workers. We built a trinational coalition with workers and environmentalists from Mexico and Canada. What we insisted on was that we wanted rules for globalization that actually developed economies so that they benefited workers. Instead, what NAFTA would do was let corporations bully governments so they couldn’t set standards.” It established private pro-business “courts” to rule on cases that companies could bring against the governments in all three nations when they believed that governmental regulations might impede NAFTA’s guarantees of free trade. 

The U.S. establishment’s strategy was to use NAFTA as a way to ensure that Mexico would remain a docile client of Washington. In 1988, leftist candidate Cuauhtémoc Cárdenas had lost the Mexican presidential election very narrowly to the business-friendly Carlos Salinas (though most reporting at that time and since has held that Cárdenas actually won but for fraudulent vote tallying). The prospect of the pro-labor Cárdenas winning the presidency had alarmed both Wall Street and Washington. Faux recalls that during a pause in their debate over NAFTA before members of Congress, a high Clinton trade official told him that they had to keep the left from power in Mexico. To that, one Democratic member added. “You may be right about the economics, but we have to support Salinas. He is our man down there.”

NAFTA was designed by corporate and financial powers to establish their permanent foothold in Mexico regardless of changes in its government. Once it passed, tariffs on their Mexican-made goods would be a thing of the past, and Mexico’s status as a low-paid, nonunion element in the production chain would be set in stone.

While Lee was rebutting her fellow economists, Faux was making that case to the larger Democratic universe, addressing countless union gatherings and liberal groups.

Their efforts persuaded many congressional Democrats, but not enough to offset the lobbying of corporations, Wall Street and the Clinton administration. In the end, although Senate Democrats narrowly voted against NAFTA (27 nos, 25 ayes) and their House Democratic counterparts rejected it more decisively (156 nos, 102 ayes), NAFTA was enacted with heavy support from Republican members who joined forces with the pro-corporate Democrats.

After the NAFTA fight, EPI worked with like-minded researchers in other countries. With assistance from the Ford Foundation, EPI helped foreign researchers and activists develop their own national versions of EPI. Among the results were a State of Working Mexico (in Spanish), a State of Working Canada, and State of Working Brazil (in Portuguese). 

There was an ugly side to the reaction to labor’s and EPI’s critique of NAFTA: the charge that opponents were anti-Mexican. In fact, it was NAFTA’s advocates who made a nativist argument that the trade deal would benefit the U.S. by reducing immigration from south of the border. EPI’s critique of trade deals has always been based on the harm they do to workers in all countries. Indeed, a poll of Mexican public opinion showed that a majority there opposed NAFTA, accurately fearing it would devastate that country’s agriculture in the face of heavily subsidized corn and other farm exports from the U.S.

Within a few years after NAFTA went into effect, the logic and data on EPI’s side of the argument began to erode some of the economic profession’s knee-jerk support for such trade accords. Within a couple years, both MIT economist Dani Rodrik and Nobel laureate Joseph Stiglitz came forth with the same critiques of corporate free trade that EPI had leveled. EPI’s three-part argument—the damage such deals did to labor standards, to environmental standards, and to the democratic sovereignty of governments by curtailing their capacity to enact such standards—became the baseline critique of trade skeptics. In 1995, it informed the congressional opposition to establishing the World Trade Organization, which gave global corporations the power to override national governments. It was EPI that provided opponents with arguments and data. 

With each successive piece of free-trade legislation that came before Congress, the number of Democratic opponents in the House grew steadily larger. In 1997, Lee was hired away by the AFL-CIO to coordinate its trade work, and EPI hired University of Maryland economist Rob Scott to take her place. Scott was soon to provide the job-loss projections that informed both Lee and Faux’s presentations against the Clinton administration’s proposal to establish permanent normal trade relations (PNTR) with China, which came before Congress in 2000. Scott’s projections proved to be prescient, as did the arguments others advanced against the administration’s contention that opening China to global capitalism would inevitably turn China into a democracy—a contention that was to prove not merely wrong, but fundamentally and spectacularly wrong. 

If there is a summary statement of EPI’s stance on the trade deals of the past quarter century, it’s EPI’s 2008 book by economist Josh Bivens, with the inspired title Everybody Wins Except for Most of Us. Here, as in EPI’s earlier work, Bivens looked at the effects of these agreements not on the nation’s overall GDP, which factor in the increased income (from the lower costs of offshored labor) that redounds to large shareholders in global corporations, but on Americans at the level of the median wage and on non-college-educated workers overall. 

Using conventional models, including one introduced by Paul Krugman in the mid-’90’s, Bivens showed that globalization reduced the wages of the median wage earner by $1,400 in 2006, and a household with two wage earners by $2,500.

The error that establishment economists and politicians make, Bivens wrote,

is conflating national income and national welfare. As global integration generally raises national income (gross domestic product, or GDP), many economists rest arguments for it there. If pressed, they often argue that concern about how this national income is distributed post-integration is just not their business—that’s politics, not economics. This is a dodge, pure and simple. Economists expressing a preference for global integration are, in fact, expressing a purely political or ethical preference, not a scientific judgment. Nothing in the discipline of economics tells us that a policy that raises national income but leaves some individuals better off and some worse off is one that raises national welfare. 

How this national income is distributed has always been EPI’s business, of course. In time, as Democrats began to understand how greatly free trade had weakened its hold on much of the working class, and as pandemic-era Americans began to understand the nation’s vulnerability to on-again off-again global supply chains, EPI’s emphasis on domestic full employment and global worker rights became the dominant emphasis of many Democratic elected officials. It became a particular emphasis of the Biden administration, where a number of EPI alumni serve in key positions, among them Jared Bernstein as chairman and Heather Boushey as a member of the president’s Council of Economic Advisers and Thea Lee as the Labor Department’s Deputy Undersecretary for International Affairs.

Even as the Democratic mainstream has come to accept EPI’s analyses and conclusions, the economics establishment has increasingly been compelled to do so, too. In Everybody Wins, Bivens had calculated the amount by which trade with China had lowered wages for non-college-educated workers. In 2016, MIT economist David Autor co-authored a widely publicized paper on what he termed the “China Shock.” In it, Autor and his co-authors calculated the economic effects that the relocation of U.S. manufacturing to China, propelled by PNTR, had on U.S. workers.

His numbers were identical to those Bivens had published eight years before! 

Tragically, the failure to heed EPI’s argument that NAFTA and PNTR would squeeze and anger the U.S. working class was a political as well as a policy error. In his 2016 presidential victory, Donald Trump swept previously Democratic working-class districts in the industrial Midwest with his attacks on free trade in general and NAFTA in particular. By 2016, acknowledging the public’s backlash against such deals, Hillary Clinton had changed her mind and become a NAFTA critic. By then, however, it was too late.

Class and schools

EPI and education

Beginning in the 1980s, critics on the American Right launched a new line of attack on the nation’s public schools: Their budgets were rising, but their standardized test scores were declining. Clearly, something had to be wrong with the public schools that more funding wouldn’t help. The most prominent of these critics was William Bennett, who served as Secretary of Education during Ronald Reagan’s second term as president. 

Reagan’s successor as president, George H.W. Bush, extended this critique with the claim that the United States “lavishes unsurpassed resources on schooling,” thus demonstrating, presumably, that the failings of U.S. education couldn’t possibly be attributed to insufficient funding. A 1990 paper by Mishel and Edith Rasell, however, documented that the U.S. investment in K-12 schools was, in fact, surpassed by 12 of the 16 largest industrialized nations. And shortly thereafter, EPI was to embark on the most extensive research on school funding. 

In the late ‘80s and early ‘90s, Richard Rothstein had worked as a budget analyst for the elected school board of the Los Angeles Unified School District, the nation’s second largest. Like the schools’ right-wing and centrist critics, he looked at budgets and performance—and came to distinctly different conclusions, which EPI began publishing as a series of books and articles in 1995. Beginning with his work with Jeff Faux and Thea Lee on trade policy, Rothstein’s status was that of an EPI research associate and later an EPI fellow; he was never on EPI staff. By any measure, his was an extremely productive association. 

A 1994 Brookings Institution report on schools typified both the establishment and the right-wing critiques that then were current. “Despite increasingly large amounts of resources devoted to schools,” it proclaimed, “student performance has shown few tangible improvements…[O]ur schools are demonstrably inefficient.” 

Rothstein undertook detailed case studies of school districts to determine how much spending had grown and in what areas. With the assistance of a small flock of research assistants, Rothstein canvassed the budgets of nine school districts and found that the share of those districts’ budgets devoted to regular education had actually declined from 80% in 1967 to 59% in 1991, while the share spent on special education had soared, and that spent on school lunches and other services had increased as well. Spending on regular education—what was expected to lead to better test scores—had risen by far less than the public school critics claimed. From 1995 through 1999, Rothstein documented this budgetary shift in three successive EPI studies, beginning with 1995’s Where’s the Money Gone? 

In other work for EPI during this period, Rothstein also compared student achievement in the postwar decades to its current level, concluding that when factoring in the greatly more socioeconomically diverse and non-English-speaking student bodies of recent decades, schools had continued to do a good job educating their pupils. Looking at the international landscape, he showed in a paper co-authored by Stanford economist Martin Carnoy that U.S. public schools had done a better job improving student achievement than such presumably stellar nations as Finland, once the socioeconomic backgrounds of their students were taken into account. 

Rothstein’s work was so persuasive that the New York Times contracted with him to write its regular education columns from 1999 through 2002, after which he lectured at Columbia University’s Teachers College through 2005. 

“Richard was our guru,” says Larry Mishel. “He documented how a host of factors— housing, neighborhood, health—were key to understanding variables in student achievement.” 

That analysis achieved its fullest expression in Rothstein’s book Class and Schools: Using Social, Economic, and Educational Reform to Close the Black–White Achievement Gap, which EPI published in 2004. As Rothstein wrote in the book’s introduction:

For nearly half a century, the association of social and economic disadvantage with a student achievement gap has been well known to economists, sociologists, and educators. Most, however, have avoided the obvious implication of this understanding—raising the achievement of lower-class children requires amelioration of the social and economic conditions of their lives, not just school reform.

Aggregating a wealth of data, Rothstein laid out the social and economic disadvantages that most “education reformers” had opted to overlook. At the time, Rothstein recalls, the idea that the source of disparate educational outcomes was the schools themselves wasn’t confined merely to the right. “A lot of people on the left, like Jonathan Kozol, focused on there being something fundamentally wrong with the schools,” Rothstein says. “I argued that the achievement gaps depended on the kind of neighborhoods where children grew up, on the whole social and economic environment.”

Some on the right derided Rothstein’s attention to varying levels of student health, such as toothaches—but his observations struck thousands of teachers as accurate reflections of what they saw in their classrooms every day. “Richard’s work was very important in debunking the unexamined prejudices of the education reform types,” says Leo Carey, who until recently headed the Albert Shanker Institute. “It had a particularly wide influence in the education policy and research world.” 

Rothstein’s Where’s the Money Gone? studies also inaugurated EPI’s ongoing work on teacher compensation, a subject that Mishel turned to in the 2000s with Sylvia Allegretto and Sean Corcoran. These studies showed that teachers earned substantially less than comparably educated and experienced workers, and that this pay penalty (even after incorporating benefits) widened steadily after the mid-’90s. Readers of EPI’s ongoing documentation of teachers’ comparative under-compensation should not have been surprised when the “Red for Ed” wildcat teacher strikes swept a number of Republican-governed states (Arizona, Oklahoma, West Virginia) in 2018. In late 2022, an EPI report co-authored by senior economist John Schmitt correlated the post-pandemic teacher shortage with the low levels of teacher pay and the high levels of teacher stress. 

The influence of Class and Schools on teachers themselves was evident in the big city strikes, beginning in Chicago in 2012, that saw teacher unions demanding not just the standard increases in pay and benefits, but also greater funding for a host of community services. This approach, which has been called “Bargaining for the Common Good”—in which teachers and other employees bargain not just for their own wages and benefits, but also for improvements to the communities they serve—is rooted in the arguments that Rothstein made in his EPI publications. 

Advocacy, outreach, and growth 

EPI’s mission expands

Throughout its first 20 years, 1986 through 2006, EPI’s work centered on documenting and analyzing the actually existing economy—not just its growth and contractions (the traditional objects of economists’ research), but also the distribution of its growth and contractions, its impact on classes and quintiles, on races and genders, and on median-income Americans.

During this time, EPI’s focus did not regularly include policy advocacy, though there always was some, on such issues as raising the minimum wage. The great exception to policy reticence during these years came, of course, on trade, where EPI’s opposition to such initiatives as NAFTA and PNTR made it something of a pariah to the economics establishment, and certainly cost it funding from some established foundations.

Beginning around the time that PNTR passed Congress in 2000, some within EPI wanted the organization to increase the accessibility of its work, to more regularly add stories to its numbers, as Faux and Lee were doing on trade, and to state more clearly its differences with the economics of mainstream academics, corporate Democrats and Republicans. Economist John Schmitt, who worked at EPI from 1995 through 2001 and then again, as vice president and senior economist from 2018 until recently, wrote an internal memo in 2000 giving voice to those sentiments. “The State of Working America,” he wrote, “conveys a vast amount of information.”

The book, however, fails to communicate the “story,” the lay person’s “model,” the explanation that a reader can apply to new evidence and new circumstances and then deploy against a brother-in-law at a family barbecue. EPI has done a masterful job fighting the other side with numbers. But, as economists are fond of saying, it takes a theory to beat a theory. We need to start offering a coherent story that stands in opposition to the one that the other side is selling—successfully—and to many of our allies in the Democratic party. 

In coming years, EPI would indeed add stories and frameworks to its numbers. Most importantly, it would add advocacy to its analyses. If EPI’s first two decades primarily exposed the bankruptcy of neoliberal economic policies, its next two decades have been directed at establishing the alternatives to such policies.

Some of EPI’s early work had included significant advocacy on issues before Congress. In 2002, attorney Ross Eisenbrey came over to EPI from the Occupational Health and Safety Administration and, before that, Senator Edward Kennedy’s staff. “My job was lobbying and policy,” Eisenbrey says. “I’d bring policy issues to the staff; we’d do what we could do on it, and take it to the Hill.” 

“In 2002, after the dot-com crash,” Eisenbrey continues, “millions were still unemployed and the rate wasn’t going down. We lobbied for extending UI [unemployment insurance] benefits and we won. Later, Bush’s Labor Department changed regulations so that millions of workers could no longer qualify for overtime. Jared and I calculated the number of people who’d lose their coverage, while the administration was insisting their change would actually increase the number of eligible workers. A bill blocking the change carried in both houses, but it died in conference.” 

EPI was far from alone in its efforts on policy issues, of course; the labor movement always had a much heftier presence on the Hill than EPI did. But it was with Bernstein’s and Eisenbrey’s numbers that labor convinced legislators to back their position on overtime eligibility. 

It was in 2006 that EPI really made policy advocacy more central to its mission. One proximate provocation was the formation of the Hamilton Project, staffed by economists close to former Clinton Treasury Secretary Robert Rubin, which announced that it would provide an economic agenda for the candidates running in the 2008 Democratic presidential primaries. Hamilton also profiled interesting policy ideas from academic economics but didn’t undertake any efforts to provide comprehensive policy solutions to health care, inequality, or other problems. To counter that compendium of neoliberal nostrums, EPI concluded, it would produce, for the first time, an agenda of its own.

Over the course of 2006 and 2007, EPI unveiled its “Agenda for Shared Prosperity,” a panoply of progressive policies that could reshape the terrain on issues ranging from worker rights to full employment to affordable housing to health care. Under the co-directorship of EPI Vice President Eisenbrey, and economist Mark Levinson (then on leave from UNITE HERE), it published articles not only by staff and associates but also by such noted figures as Joseph Stiglitz, Paul Krugman, James K. Galbraith, Ray Marshall, Heidi Hartmann, Richard Freeman, Thomas Kochan, Beth Shulman, Teresa Ghilarducci, Harley Shaiken, and Susan Helper, among many others. Ghiularducci’s proposal on Guaranteed Retirement Accounts helped shape retirement policies in various states and launched EPI’s ongoing work on retirement insecurity. 

The agenda also included political scientist Jacob Hacker’s first detailed proposal for a public option in health care, in which individuals or employers could opt to purchase a plan run by Medicare, which, minus the profit motive, would be a more affordable and comprehensive offering than those of private insurers. EPI also commissioned and published a detailed financial analysis of the plan from the Lewin Group.

EPI provided multimedia presentations of its agenda, as well as publishing a detailed summary of its elements. It also held a series of forums on agenda particulars, which featured prominent members of Congress, leading advocates, and such notable figures as writer Barbara Ehrenreich. One measure of the agenda’s success was that a version of Hacker’s public option was part of the initial Affordable Care Act (ACA) that the Obama administration sent to Congress in 2009, which had majority support in the Senate but failed to clear the 60-vote cloture hurdle when Connecticut’s Joe Lieberman said he’d not give that 60th vote to the ACA if the public option was part of it. Other features of the agenda—such as reviving U.S. manufacturing—had to await the coming of the Biden presidency in 2021. 

The Agenda for Shared Prosperity wasn’t a one-shot. The Obama presidency afforded EPI a number of opportunities to advocate for and against policies, and it most vociferously made the case against cutting spending and balancing the budget that was pushed by the Bowles-Simpson Commission, as well as a number of figures within the administration, the chatterati of editorial pages and even center-left think tanks late in Obama’s first term. Papers by Mishel and Josh Bivens predicted that such policies would condemn the recovery that followed the 2008 financial panic to move at a snail’s pace—as it turned out, an entirely accurate prediction.

During this period, EPI also began to issue shorter publications that made it easier for its progressive and union allies to convey the case against austerity to public officials and the media. With the inauguration of its Working Economics Blog in 2011, EPI was better able to put forth its data, arguments and analysis of current events in even more concise form. 

In 2017, EPI inaugurated its Perkins Project on Worker Rights and Wages, named after Frances Perkins, FDR’s Labor Secretary, who was the first woman Cabinet member and a key architect of the Social Security law. Led by Heidi Shierholz, then EPI’s director of policy, and Celine McNicholas, then EPI’s counsel, the project tracked the Trump administration’s wage, labor, and employment policies, documenting the many ways that Trump’s economic policies took a toll on the very working Americans he claimed to champion. 

More changes came to EPI in 2017, when Thea Lee returned from the AFL-CIO to succeed Mishel as EPI president. During Lee’s three years as president (she left in 2021 to handle international issues at the Biden administration’s Labor Department), EPI’s yearly budget grew from $6 million to $11 million, and its staff grew to roughly 50. The additional funding was coming largely from the foundation world that had considered EPI too radical and too close to labor in earlier years. EPI was no less radical or close to labor than it had been, of course; it was the political priorities of the foundations, and the general political climate, that had changed.

In its first years, roughly 90% of EPI’s funding came from unions. By 2018, while unions still supported EPI at the same levels they always had, their contributions, thanks to the greater foundation support, amounted to roughly a third of EPI’s total income. 

In a sense, Lee was building on the credibility that EPI had attained under Faux and Mishel. By 2017—one year after Bernie Sanders’ primary race had demonstrated the growing support for social democratic and pro-labor perspectives within the Democratic party—it had become apparent to policymakers, foundation officials, and many within the mainstream media that EPI’s analysis of the U.S. economy had been more accurate and germane than those of the economics establishment, and its prescriptions for change more on point. “By the time I came back,” says Lee, “we’d won the argument on the growth of inequality, on raising the minimum wage, on the effects of our trade policies.” 

The sense of anger and abandonment that white working-class voters had expressed in casting their ballots for Donald Trump gave a particular urgency to this belated acknowledgment of EPI’s focus on the plight of nonaffluent Americans—on the state, as it were, of working America. The mountain, or more precisely, the mainstream, had come to EPI. (Well, the center-left mainstream.) 

Taking advantage of the increased financial commitment from foundations, and responding to the threats to U.S. democracy posed by the Trump presidency and the rise of neofascism, Lee took a series of actions that sharpened EPI’s focus.

Upon taking office, Lee initiated a staffwide project to devise a strategic plan for the organization. It came to include the considerable expansion of two projects that had begun some years earlier: the Economic Analysis and Research Network (EARN), which linked EPI to a host of state-based advocacy organizations that drew on EPI’s expertise, and the Program on Race, Ethnicity, and the Economy (PREE), which dealt chiefly with the economic state of Black and brown America and the policies that could improve it. Both projects had been initiated on a far smaller scale under Mishel’s leadership: EARN in 1997 and PREE a decade later. Each will be discussed in the following separate chapters. 

Before he’d stepped down, Mishel had strengthened EPI’s capacity for advocacy. He’d formed a policy department, hiring Heidi Shierholz to direct it in early 2017. Shierholz had first come to EPI in 2007 as a labor economist, in 2014 moving to the Obama administration’s Labor Department as its chief economist. At the same time, Mishel had also hired Celine McNicholas, who’d been handling congressional liaison for the National Labor Relations Board during Obama’s second term, to succeed Eisenbrey as EPI’s director of governmental affairs and its general counsel. Like Eisenbrey, McNicholas maintained close contacts with Democratic legislators, but with two signal differences: The new policy department had a bigger staff than Eisenbrey had had, and EPI was more proactive not just in its back-and-forth with legislators, but also in putting forward its own policy proposals, which the policy department developed in collaboration with EPI’s economists and researchers.

“Rather than just wait for a call, for requests from the Hill, we’d send up proposals,” says McNicholas. “Our research could be targeted for a particular debate or for members who’d advance certain arguments. We did more outreach; we developed more relationships. We did weekly reports on how policies supported by the Trump administration would hurt workers. We did the preparatory work for bills to raise the minimum wage and for the PRO Act. Thea really prioritized this policy work.” 

Lee also made sure that the work was more widely accessible. She expanded EPI’s communications department, bringing on communications director Eve Tahmincioglu, and worked on making communications both clear and concise. “Think tanks can be those falling trees in a forest,” Lee says. “They can do amazing work, but if nobody reads them, it doesn’t help anything. Some of our researchers were turning out 75-page papers. Our communications team worked with them to boil down the takeaways. That helped us turn out punchier products.”

What all of this has meant is that EPI is now more directly engaged than ever before with both a wider range of readers and a wider range of groups and constituencies. With the expansion of the policy staff and of the EARN staff, too, “We’re now more oriented to what the Hill, and what state capitols, are interested in,” says EPI Chief Economist Josh Bivens. This lengthened to-do list is now addressed by a much larger, broader, and more diverse EPI.

In the states

EPI and EARN

In recent years, EPI has greatly expanded its work with state and local economic research and advocacy groups, but its history of involvement with local policy work and campaigns for economic justice goes back to the 1990s. In 1995, when city contract workers in Baltimore became the first in the country to campaign for and win a living wage, EPI and University of Massachusetts economist Bob Pollin provided research and analysis making the case for that raise. When living wage campaigns and then minimum wage campaigns began springing up in states and localities later in the nineties, EPI added a staff member to provide technical assistance to those efforts. 

In 1997, Mishel joined with University of Wisconsin academic and activist Joel Rogers, and several state groups to convene a meeting that explored setting up a network, anchored at EPI, that would work on economic issues at the state level. This meeting led to the creation of the Economic Analysis and Research Network (EARN), connecting EPI directly with a small group of progressive policy research organizations across the country, including Joel Roger’s Center on Wisconsin Strategies (COWS), Pennsylvania’s Keystone Research Center headed by Steve Herzenberg, Policy Matters Ohio then led by Amy Hanauer, and Washington state’s Economic Opportunity Institute, which was explicitly modeled by its founder, John Burbank, to be a state-level version of EPI. Some of these groups already belonged to a network affiliated with the Center on Budget and Policy Priorities, which focused on using government tax and transfer policies to help the poor.

EARN’s focus, like EPI’s, would be on policy issues affecting working people, especially the working poor and middle-class, such as wages, working conditions, access to health care and retirement security. Today, EPI’s EARN program helps build and strengthen the progressive economic justice infrastructure at the state and local levels, providing funding and technical assistance to nearly 60 state groups, facilitating cross-state collaboration and learning through a very active listserv, webinars, and large annual conferences, where policy ideas are debated, skills are taught (e.g., Data 101), and successful initiatives are profiled, so other states can adopt them.

One of EARN’s first efforts was to collaborate with the Wisconsin group that Rogers had founded (the Center on Wisconsin Strategy, or COWS) on the first state version of EPI’s signature publication—in this case, the State of Working Wisconsin. This has led to dozens of such reports produced on a regular basis by EARN’s state partners to this day.

The EARN network remained relatively small for nearly two decades, doing important work in the states that had affiliates. Its Virginia affiliate, for instance, began publishing annual State(s) of Working Virginia during President Obama’s first term, which helped lay the groundwork for the state’s minimum wage increases when the Democrats won trifecta control of state government in 2018. 

In 2018, Lee hired Naomi Walker to direct EARN—a move that was key to the expansion of both EARN and EPI. Walker had directed state government relations for the AFL-CIO and then coordinated AFSCME’s coalition work with progressive groups around the country. Much of the higher levels of financial support that EPI has received in recent years is the result of Walker’s fundraising; during her time running EARN, the project’s staff grew in size from 1.5 employees to nearly a dozen.

With Walker at the helm and with EARN staffed up, the network became a national resource and training ground for policy scholars and activists across the country. In 2021, Walker became EPI’s vice president, and she was succeeded at EARN by David Cooper, a longtime EPI staffer who’d been EARN’s senior economic analyst and then deputy director. 

The election of Donald Trump changed the political and economic landscape in ways that made EARN more central to the progressive project. With Trump in the White House, “there was no opportunity to enact good economic legislation on the federal level,” says Walker. Lee and Walker both understood the opportunity this created to both deepen and broaden EPI’s work with state organizations—states suddenly becoming the sole level of government where legislation like minimum wage increases could be enacted. Walker presented this proposal to a number of foundations that found it compelling. With the additional funding and responding to the focus on states, the EARN network has grown since 2016 to include 57 “think-and-do-tanks” in 42 states. 

“Many of the state-level groups had been focusing on their states’ tax structures, budgets and safety net—issues that the CBPP network worked on,” says Walker. “They weren’t doing work on labor standards, on worker power, on race and gender equity, on public employee collective bargaining—the issues EPI has worked on. We had the resources to give them the capacity and the space to do that.” 

Virginia’s Commonwealth Institute for Fiscal Analysis—which Levi Goren, its director, describes as a “think-and-do” organization—joined the EARN network in 2010. Its experience working with EPI typifies that of the various state organizations. “EARN provides us with the expertise, data, and thought partners that help us explain the policies that have shaped the state’s economy and the consequences for all kinds of Virginians,” Goren says. “They gave us the tools to do that analysis, so we could show how workers and farmers were doing, how many were unemployed, how Virginians at all income levels were making out—so we could tell this story in ways meaningful to a range of different groups and activists.” 

During the Commonwealth Institute’s years-long campaign to raise the state minimum wage, Goren says, “EPI enabled us to show the impact of the different proposals, how many workers would benefit from each, breaking the numbers down by race, gender, and industry.” That material helped shape the bill that the legislature finally passed in 2020, which phased in major hikes to the state’s minimum wage, which currently has risen from $7.25 to $12 and will go to $15 in 2026. 

EARN’s role in the Virginia minimum wage fight wasn’t exceptional: The minimum wage tracker that Cooper developed for Virginia has been adapted to their own state’s conditions by groups in many states and localities to document the problems that low wages inflict on specific populations and to highlight the gains those populations would make with different levels of higher minimum wages.

When Walker came to EPI and EARN in 2018, she pushed to have the network focus more on issues of worker power and racial justice. In several states, the issue of establishing collective bargaining for public employees, at either the state or local level, was before the legislature, and EARN worked with its Virginia and Colorado affiliates to show how organized workforces—both more stable and more productive—could more than pay for themselves and create a more efficient government in the process. Goren notes that once local collective bargaining was established in Virginia, EPI has worked with their organization and others to make that case in the localities where public employees were unionizing. 

Such campaigns reached well beyond the various state think tanks that belonged to the EARN network, and so the EARN network expanded accordingly. “We moved beyond just the research groups,” says Goren, “as we moved into campaigns more explicitly about worker power. We added grassroots organizations to our network, and EPI provided the support that enabled us to do that.” 

In Alabama, says Walker, “our affiliate worked with groups like the Adelante Worker Center in Birmingham. It interviewed workers and integrated their stories into the State of Working Alabama. We worked to create the space for research groups and community groups to work in coalition, to create the trust that can cement a coalition for the long haul.” In Virginia, Goren’s institute worked with New Majority Virginia, for instance, on the issue of pretrial detention and cash bail, tallying the number of Virginians who’d lost their jobs while behind bars for crimes for which they’d then be acquitted. 

In structuring that campaign, Goren says, they learned from the experience of their counterpart group in Oklahoma. That, in turn, was the result of EPI’s decision to form regional networks of their EARN member groups in both the South and Midwest. “Our Southern affiliates needed something specifically for them,” says Walker. “They’d say, ‘We’re not facing the same things they face in New York and California; we need a space to talk about the challenges particular to the South.”

EARN created that space and now convenes regular conferences for its Southern and Midwestern affiliates. For its part, EPI is now also producing research and analysis specifically for those regions. Its work on the South, says Goren, “emphasizes how policy has affected different races there, with specific socioeconomic breakdowns of the consequences. It’s more useful to us than a national analysis.” 

EPI’s embrace of advocacy as well as research has been particularly notable in its work in EARN. Last year’s annual conference of all EARN affiliates, held in Detroit, very intentionally focused on showing how EPI and the state research groups could work on a diverse group of issue and legislative campaigns, and highlighted examples from states where that had already happened.

“As with EPI broadly,” says Walker, “we’re now partnering more directly with community groups and unions on policy campaigns. Previously, some EARN affiliates may have provided important data but stood back from involving themselves more fully in the campaigns. Now, they’re full partners in those policy campaigns as well.” 

The exploitation of Black, female, and immigrant workers 

EPI and PREE

EPI never looked at the economy solely through the prism of class. From its earliest days, from its first papers breaking down how workers at different income levels were faring, those papers also included tables showing how Black workers and Hispanic workers were faring relative to each other, to whites, and to the national population. Those papers also measured how women workers were making out compared with their male counterparts. EPI’s studies often addressed policy areas that disproportionately affected women, including part-time/contingent work, minimum wages, child care, the need to raise pay in “caregiver” work, balancing work and family, paid leave, and sick leave. The workers whom EPI studied in its papers and reports had class, racial, and gendered identities as invariable complements to their combined “all-worker” identities. 

Such comparative measurements and analyses informed a number of EPI’s early publications, among them Richard Rothstein’s book Class and Schools: Using Social, Economic, and Educational Reform to Close the Black–White Achievement Gap, which EPI published in 2004. That said, EPI began to deepen its focus on race in the 2006, and in 2008 it formed its Program for Race, Ethnicity, and the Economy (PREE), with guidance from EPI Secretary-Treasurer and board member Julianne Malveaux, Howard University professor and AFL-CIO economist Bill Spriggs, and activist/scholars Deepak Bhargava and Manuel Pastor. At the same time, EPI substantially increased the diversity of its board and staff, which led to a shift in governance and personnel, and an expansion of its areas of work. 

PREE’s work also involved a deeper engagement with a range of civil rights and community-based organizations. As it had with the labor movement, EPI’s work thus came both to reflect and inform the perspectives of those communities, as well as connecting them in new ways to the labor movement.

PREE’s first director was Algernon Austin, and its early work focused on detailing the wage, income and employment inequities faced by Blacks, Hispanics, Asian Americans, and Native Americans. PREE was also a pioneer in documenting the racial and ethnic disparities in unemployment at the state level. 

At the time of its 2008 founding, PREE was something new in the think tank world: EPI was the first such organization to establish a research program specifically focused on race, ethnicity, and economy policy. “Only in the last three-to-five years have we seen other think tanks do this sort of thing,” says Valerie Wilson, who has headed PREE since 2014. “Several of those organizations reached out to us to see how to do this kind of work.” 

Wilson, an economist, came to EPI and PREE in 2014 from the Urban League, where she’d headed the league’s research initiatives. “It was the perfect job for me,” she says. “It enabled me to continue to work on race, ethnicity, and economics, at an organization distinctly focused on economic research and policy.”

Wilson’s work began with a study she co-authored that looked at various states’ cutbacks to unemployment insurance, whose raison d’être was that they were supposed to increase employment levels. Wilson’s study showed that they did nothing of the sort. 

Her largest early project looked at Black-white wage gap levels between 1979 and 2015. Disaggregating the causes for its persistence, she was able to measure the relative weight of educational and regional disparities and of racial discrimination, finding that discrimination was the most important factor. The study also showed that policy mattered, “When policies [minimum wages, collective bargaining] made it more difficult to pay less, wage gaps narrowed,” she says. The converse, unfortunately, was also true. 

Since the study’s release, EPI has continued to provide the data and analyses of the wage gaps—both by race and by gender—online, through the State of Working America data library. “There are so many ways that inequality shows up in the workplace,” Wilson says. More recent PREE studies have focused on the disparity in available work hours between Blacks and whites, on the particular importance of a full employment economy for racial minorities and women, and on the disparate income challenges for women with children in the absence of affordable child care.

In the past half-decade, the staffing for PREE, like that for EARN, has grown. Under the leadership of Thea Lee and now Heidi Shierholz, EPI has devoted time and resources to increasing funding specifically for these projects. “We’ve invested more not just in these work products,” Wilson says, “but organizationally into diversity, equity, and inclusion. The [EPI staff] union made recommendations on how and where to post job openings,” to which management added its own. Departmental hiring committees worked with management to recruit a less preponderantly white and male workforce. Since 2017, even as EPI has been rapidly growing, it has been growing steadily more diverse. 

PREE’s expansion coincided with EPI’s growing emphasis on policy formation and advocacy. “We now have a policy agenda on our website, some of which focuses on issues of gender and race,” says Wilson. Several years ago, PREE was able to hire a designated policy analyst, Adewale Maye, who works with EPI’s policy department as well. In 2019, PREE hosted a series of workshops on its analysis of racial disparities, the issues they raised and the policies that could mitigate them, which was attended by several dozen researchers from other think tanks. It followed this up with an anti-racist research and policy guide based on the workshops, which provides the data and analyses of racial disparities in labor markets, health outcomes, incarceration rates, wealth accumulation—“lots of information in one place for researchers and advocates,” Wilson says. A recent study by Maye charted the disparities that have persisted or even grown in the 60 years since the 1963 March on Washington. It documented, for instance, that the wealth of the typical white family is eight times that of its Black counterpart. 

EPI’s projects inform one another. EARN’s national conference in 2022 was planned in conjunction with PREE, and featured a plenary session put together by PREE on “the complex history of unions and race,” says EARN director Dave Cooper, and the important role unions have played in the fights for racial and economic justice.

One of EPI’s most notable contributions to the understanding of American racism was a result of the institute’s open-ended support for Richard Rothstein’s research. “In Class and Schools,” says Rothstein, “I was writing about disparities in the individual characteristics of students—the ‘dental theory of education,’ critics called it, since there were disparities in dental care. But schools concentrated children with problems like this, and if almost every child in a school had one or more such problems, the school was overwhelmed. If most of its students were stressed about violence or had levels of lead poisoning, or asthma, or other similar challenges, student achievement fell: The school couldn’t handle the concentration of these problems in its neighborhood. So residential segregation was the root cause of lower student achievement.” 

“I had this epiphany in 2007, when I was still under my EPI contract.” In that year, the Supreme Court issued a ruling about school districts in Louisville and Seattle, which had tried to integrate schools across city lines to overcome the residential segregation between cities and their suburbs. “[Chief Justice John] Roberts wrote the majority opinion,” Rothstein continues, “saying that the segregation of neighborhoods was de facto, not de jure. And since it wasn’t the government that had created it, the government couldn’t address it.” 

“Then I remembered [longtime civil rights activists] Carl and Ann Braden. They had lived in Louisville. An African American family who were friends of Carl and Ann wanted to move to the suburbs, and Carl [who was white] bought a home for them,” as no one in the suburb would sell to the family. “The house was dynamited—and then, the state of Kentucky tried and convicted Carl of sedition for provoking a riot by selling a home in a white neighborhood to a Black family. This wasn’t de facto segregation. It was de jure.” 

“It was widely accepted on the left as well as the center and right that residential segregation was de facto. It wasn’t. The left said it was bank redlining and the norms of realtors. I told Larry I wanted to look into the roots of residential segregation as my next EPI project. Larry said, ‘go ahead.’” 

Rothstein’s project took a decade to research and write, and its findings were so important and, in his words, “so explosive,” that Mishel and he decided to take it to a trade publisher, rather than have EPI publish it, as it had done with all but one of Rothstein’s previous books. The Color of Law, which was published in 2017, brilliantly and comprehensively documented the government policies that had segregated U.S. cities and suburbs. It was on the New York Times nonfiction bestseller list for the better part of a year, and was also on the Times’ list of the year’s 50 best books. Five years later, Mishel says, royalties from sales of the book have brought EPI more than one million dollars.

During George W. Bush’s second term as president, at roughly the same time that EPI was expanding its advocacy and adding PREE to its departments, the most intensive effort to enact immigration reform since the mid-1980s was floundering in Congress, which ultimately failed to pass it.

Partly in response, EPI produced its own plan for comprehensive immigration reform, based largely on the work of former Labor Secretary and longtime EPI board member Ray Marshall, who adapted it, in collaboration with EPI vice president Eisenbrey, into an article that first appeared in EPI’s Agenda for Shared Prosperity. In 2009, EPI followed up, publishing Immigration for Shared Prosperity, a short book by Marshall which addressed and, where necessary, reconciled the needs of immigrants seeking citizenship and worker rights with those of the labor movement to which many of them belonged. The elements of the plan outlined in the book became the unified plan of the previously divided labor movement for immigration reform. 

The “Marshall plan,” as some referred to it, was submitted to the Obama administration and friendly members of Congress for consideration—with parts of the plan appearing in proposed bills from progressive immigration reformers—and became an ongoing part of union advocacy on immigration. Like EARN and PREE, this initiative put EPI directly in touch with groups of activists with whom it had scant contact before. 

“When the reform initiative pushed by [George W.] Bush in his second term fell apart, unions got blamed for not being on board, but the criticism was unjustified” says Daniel Costa, who is EPI’s director of immigration law and policy research. “Most of the unions were critical only of the bill’s guest worker provisions because they would’ve vastly expanded the number of workers who would have very few workplace rights, who’d be indentured to their employers, and who’d never be able to become citizens.”

Costa was hired as a policy analyst to support EPI’s immigration work in 2010 and a few years later began leading that work. The son of immigrants, he’d received degrees in international and comparative law from Syracuse and Georgetown—becoming only the second lawyer EPI ever hired, with Eisenbrey being the first—and had worked on immigrant integration in California, and before that, on developing the legal frameworks for disaster response and humanitarian relief for the Red Cross and Red Crescent in Geneva, Switzerland.

Once he came to EPI, he was heavily involved in the immigration reform efforts of Obama’s second term, in which the Senate passed the bipartisan “Gang of Eight” legislation only to see it never even get a vote in the Republican-controlled House. When the Gang of Eight legislation was being developed in the Senate, Senator Chuck Schumer outsourced one key component of the bill jointly to the AFL-CIO and the U.S. Chamber of Commerce, tasking them to come up with an agreement that would enable low-wage migrant workers to come to and work in the U.S. without undermining existing labor standards. With EPI economists and Eisenbrey, Costa produced research and data used by the Federation. After an agreement was reached, Schumer told the New York Times that the guest worker issue “has always been the deal breaker on immigration reform, but not this time.” 

Some of Costa’s first work for EPI also included a paper on the State Department’s J-1 visa program, under which foreign exchange students are supposed to spend four months working in the United States as a form of “cultural exchange.” Costa’s work coincided with the strike of J-1 exchange students who’d somehow ended up working long hours for a subminimum wage on assembly-line jobs in Hershey Chocolate’s Pennsylvania factories—a story that ended up on the front page of the New York Times. “No one had ever written about J-1 jobs before,” Costa says, “and how they’d gone from a student exchange program to indentured servitude.” Costa’s paper helped put EPI on the immigration-issue map, and led to new State Department rules for its exchange program. 

Since then, he has been working extensively on work visa programs dealing with farm labor, landscaping, traveling carnivals, construction, and computer occupations—“just about everything about immigration and labor,” as he puts it. In 2020, EPI published a paper he co-authored analyzing wage and hour data for farmworkers over a 20-year period. A new EPI paper he authored reports that federal spending on enforcing laws to deport immigrants is ten times what the government spends on enforcing labor standards for all of America’s workers. “There are 80,000 federal workers employed on immigration enforcement and fewer than 10,000 employed across all the agencies seeking to ensure that labor laws are followed in the United States,” Costa notes. 

Costa’s writing and testimony in hearings have also informed the updating of farm labor standards, such as overtime laws, enacted in blue states like New York. At the federal level, by contrast, such victories are few and far between. “On immigration matters, it’s really hard to get good stuff accomplished,” he says, “but at least we’ve had the ability to slow down or stop the bad stuff that corporate lobbyists push for that would harm immigrant and American workers alike.” 

Winning the battle of ideas

EPI’s trajectory from heretic to oracle 

On Friday, April 7, 2023, the New York Times ran a story on the disjuncture between an economic problem and its supposed economic cure. The problem was inflation; the cure that the Federal Reserve was pursuing was raising interest rates, which was intended to diminish investment, hiring, and consequently, wages and spending. 

The disjuncture, as many economists have noted, was that wages, though having risen, didn’t seem to be the cause of inflation. The war in Ukraine, the supply chain kinks, and a host of other causes seemed the more likely culprits behind the rise in prices. That said, the Times didn’t present any data, any real numbers, pointing to alternative causes of inflation. 

Until, halfway down the piece, it began to quote Josh Bivens, EPI’s chief economist. Here’s what followed: 

Josh Bivens, the chief economist at the Economic Policy Institute, a liberal think tank, said that while higher unemployment generally curbs wage growth and price pressures, wages “are no guide at all” right now with “the unexpected and weird shocks we’ve had over the past couple of years.” 

According to Mr. Bivens’s research, profit markups have “relented a bit”—accounting for about a one-third share of price increases in the fourth quarter of last year, down from more than half in the comparable period in 2021 — but are “still quite high relative to a baseline,” which has been closer to 13 percent in previous business cycles. 

What Bivens provided was precisely what EPI’s founders had hoped their fledgling institute could add to the analytic, policy and public discourse: real numbers that could dispel unfounded beliefs and show policymakers what really ailed the economy. 

What had changed over the nearly four decades since the founders’ first deliberations was the prominence that mainstream institutions gave to what were once viewed as EPI’s heresies. In recent years, the media have frequently—at a number of media outlets, routinely—gone to EPI staffers for their data-based analyses of economic events: Elise Gould on the monthly employment figures, Rob Scott on trade, Valerie Wilson on economic issues in minority communities, and Heidi Shierholz on—well, just about anything.

This is not because EPI has moved toward what was once the neoliberal consensus. It’s because much (by no means all) of the economics establishment, a growing share of economics journalists, and most Democratic policymakers have been persuaded by the force of EPI’s arguments—and the growing irrelevance of neoliberal doctrine to the actually existing economy—to move towards EPI. 

“It’s good to feel that you’ve had some influence on the way that people now talk about economic issues,” says Wilson, attributing this shift not just to EPI’s message but to its messaging as well. “Some of the value EPI brings to the discussion is to simplify how we talk about these things: telling the story plainly, with statistics and data that support the story. We’re good on creating graphics that are easily understandable”—the famous productivity and hourly wage graph being a prime example. 

Over the years, EPI has produced papers and studies identifying the causes behind the lagging wages that its pay vs productivity graph so forcefully illustrated. That process culminated in a Bivens/Mishel study in May of 2021 that disaggregated and quantified the particular causes of wage stagnation—which, the authors argued, was in fact a misnomer. “We refer in this analysis,” they wrote,

to wage suppression rather than wage stagnation because it was an actively sought outcome—engineered by policymakers who invited and enabled capital owners and business managers to assault the leverage and bargaining power of typical workers, with the inevitable result that those at the top claim a larger share of income. These policy changes and the change in business practices they enabled have systematically undercut individual workers’ market (exit and voice) options and the ability of workers to obtain higher pay, job security, and better-quality jobs. These corporate and policy decisions had the most adverse consequences for low- and middle-wage workers, who are disproportionately women and minorities, the groups whose legacy of being discriminated against in labor markets means that they especially need low unemployment, unions, strong labor standards, and policy supports for leverage when bargaining with employers.

Neither slow productivity growth nor inevitable economic forces can explain U.S. wage problems. Rather, wage suppression reflects the failure of economic growth to reach the vast majority. It was a “failure by design” (Bivens 2010), engineered by those with the most wealth and power. The dynamics are primarily located in the labor market and the strengthening of employers’ power relative to their rank-and-file workforce (which increasingly includes those workers with a four-year college degree). In other words, the dynamics that have challenged the growth of living standards for the vast majority are based on workers not sharing in economic gains, not, as some have argued, on consumers suffering from monopolistic prices. Changes in product market monopoly and corporate structures have had an impact, but primarily by squeezing supply chain profits and wages rather than by spurring higher consumer prices through much wider profit margins. 

The Bivens/Mishel paper examined the role that, according to neoliberal theory, the skills gap played in lagging wages, and found it so bogus as to be inconsequential. They then looked at a series of policy decisions—the interest rates and deficient public investment that drove up unemployment rates, the increased trade with low-wage nations, the all-too-successful war on unions and workers’ ability to collectively bargain, the failure to raise the federal minimum wage, the mislabeling of workers as independent contractors, and the spread of forced arbitration and noncompete agreements, which curtailed workers’ means of redress and even of exit.

It was these factors, whose effect on lagging wages they quantified based on publicly available data, that were responsible for the wage stagnation—which, since it was the result of corporate and governmental policy, they termed “wage suppression.” Combined, these factors contributed heavily to the redistribution of the nation’s income to the wealthiest 1%. Combined, they accounted for three-quarters (76.1%) of EPI’s signature gap between the levels of productivity and median hourly compensation that rose relentlessly between 1979 and 2017 (the most recent year with available data). 

Looking back on his (still ongoing) work at EPI, Bivens highlights both his wage suppression paper with Mishel and a solo effort he produced in April 2011, entitled “Abandoning What Works.” The paper was written as a response to the report of the Simpson-Bowles Commission, which had appeared at the end of 2010. That commission had been charged by President Obama to suggest ways to reduce the federal government’s deficit, which had swelled in response to the deep recession triggered by the bank collapses of 2008. Responding to the cries of Republicans and centrist Democrats to reduce the deficit by cuts to spending, particularly spending on Social Security and Medicare, the report laid out how those cuts could—and should— be made. 

In response, Bivens noted that unemployment was still at its highest levels in 30 years and that the recovery was proceeding at a glacial pace. “The key to jobs recovery,” he argued, was “stabilization through fiscal support.” Obama, he wrote, needed to support more and bigger stimulus programs—old ideas that had worked to curtail previous recessions and get the nation closer to full employment. Absent such policies, he showed, it would be minorities and the young who would pay the price in lost jobs and reduced incomes. Bivens’s paper prodded the Congressional Progressive Caucus to move to a full-throated endorsement of greater stimulus spending. It preceded—and in a sense, predicted—the challenge to austerity and the case for egalitarian economics that was soon to erupt in the Occupy Wall Street movement and, later, in Bernie Sanders’s presidential campaign.

By the time COVID-19 shut down the economy in 2020, a deep downturn that persisted into 2021, many mainstream economists and most Democratic policymakers had taken Bivens’s argument—which he forcefully renewed in 2020—to heart. “Josh was key to organizing the center-left coalition to fight the recession with a genuine stimulus,” says Shierholz, referring to the trillion-dollar-plus American Rescue Plan passed at the outset of the Biden presidency. “Josh deserves a lot of the credit for this recovery being five times as fast as the one that followed the 2008 collapse.” 


‘We used to quote Joan Robinson’s line that ‘the reason we study economics is to avoid being deceived by economists,’ Shierholz says. ‘That’s far less the case today.’

The establishment economists had long defined economics for the economic policy world. They were the gatekeepers determining what idea could be let in. Thus, as Bivens adds, “EPI used to spend a great deal of time fighting with academic economists. That was important since academic economics was really pretty dreadful. Now, that’s changed: We have fewer fights, and more of the time, we’re acting as transmission belts from academia to policymakers.” 

What’s changed, of course, is the substance of mainstream academic economics—not all of it, but a lot. During its first 25 years, EPI had been the odd-man-out in U.S. economics, as an internal memo that economist John Schmitt wrote to his EPI colleagues in 2000 made clear. “It is sad but true,” he began, “that EPI has become one of the last intellectual homes of the great American ‘institutionalist’ school of labor-market analysis. We have a responsibility to keep the flame alive, at least until it can catch again more widely in the academic world.” 

Over the past decade, it has indeed caught on more widely. “The field [of economics] itself has changed,” says Shierholz. “There are now lots of young economists, like those on our newly formed academic advisory board, who are at elite universities and who understand that policy decisions, institutional factors, and power relations are key drivers of economic outcomes. They’ll be around for decades.” 

Even more important than the conversion of much of the economics establishment is the conversion of Democratic policymakers and the Democratic rank-and-file. “The shift from Clinton’s and Obama’s economic policies to Biden’s is remarkable,” Shierholz adds. “It took a long time, but it’s so gratifying to see our fingerprints all over the ways that Democrats now write and talk about the economy.”

“The wonderful things that Jeff and Larry started has continued and expanded under Thea and Heidi,” Heather Boushey adds. “EPI wasn’t a one-trick pony. It’s delivered solid analysis and built an institution around that work. It needs to continue for generations to come.” 

American Prospect founding co-editor Robert Kuttner was also one of EPI’s founders and has served on its board since its beginnings. In a recent Prospect column headlined “What Comes After Neoliberalism?” he wrote, “We are winning the battle of ideas. We have a long way to go before we win the politics.” 

It’s in no small measure EPI’s doing that we’re winning the battle of ideas. Winning the politics remains the task of all of us.


See related work on About EPI

See more work by Harold Meyerson