How many workers are left out of it?

How many workers do have access to retirement plans but decline to contribute to them? And why?

Data from the 2025 Survey of Income and Program Participation (SIPP) was just released, so we have newly-updated answers to these and many other questions.1

The headline numbers

About 76 million workers — nearly 52 percent of working Americans ages 18 to 64 — lack access to an employer-provided retirement plan. But not every worker works for an employer, so we should probably cut out the self-employed from that total.

Excluding people who are self-employed and focusing only on the 132 million Americans ages 18 to 64 who work for an employer, about 46 percent of them, or 61 million, lack access to an employer-provided plan.2

Figure 1 further breaks down the access numbers, showing that workers in the private sector, which employs the majority of workers, are less likely to have access than government workers, and part-time workers are less likely to have access than full-time workers.3

Who gets a match?

A wide range of retirement plan types exist across the world of work. Only 37 percent of all workers receive an employer contribution or match. For those workers who do, the median amount received is about $3,000 per year. (The mean is about $5,800.)

Gaps by race, ethnicity, and education

While 82 percent of workers with less than a high school diploma and 64 percent of workers with a high school diploma lack access to a plan, only 31 percent of employees with a bachelor’s degree lack access.

Figure 2 offers details on retirement plans by race and ethnicity. The access gap is biggest for Hispanic and Black non-Hispanic workers, and smallest for Asian workers.

Income, access, and employer contributions

As would be expected, access to tax-advantaged retirement savings also differs starkly by household income, reflecting enormous differences between the highest and lowest earners.

In the lowest earnings decile, about 88 percent of private-sector employees lack access to an employer-provided retirement plan. Among the top decile of earners, only 14 percent lack access.

It is not just that access is better for high earners. At high income levels, meaningful portions of the workforce participate even if they do not receive an employer contribution to compound the benefits of savings.

Among workers who receive no employer contribution, those in the top half of the earnings distribution are about five times more likely to participate in an employer-provided retirement plan than those in the bottom half (19.9 percent versus 3.9 percent).

Higher-earning workers have greater likelihood of access to these plans, lower barriers to saving, and can more easily access the main federal tax benefit associated with retirement saving: deferring one’s income tax bill for retirement contributions until old age, when they will likely be in a lower tax bracket.

Another way to understand the same data: Expanding access to retirement plans for workers at the bottom of the income distribution is a necessary first step in addressing the future of retirement, but access alone will have a limited impact if it is not paired with a match or additional contribution.

What the tax code pays for

Both businesses and workers make their own decisions about how they spend and how they save, so the gaps that result may appear to simply reflect a range of preferences.

But the government is not neutral in how it influences those decisions. The tax code does not treat retirement savings as a cash-equivalent source of compensation; it incentivizes retirement savings.

In effect, taxpayers subsidize the retirement savings of high earners through foregone collection of tax revenue. My colleagues and I estimate that tax-advantaged retirement savings through income and payroll taxes cost between $256 billion and $356 billion in forgone annual federal revenue, measured on a present-value basis for savings made in 2024.4

When foregone tax revenue is understood as spending, subsidized retirement savings ranks among the largest expenditures in the federal budget.

That spending flows predominantly to retirement accounts that roughly half of workers cannot access via an employer-provided plan. The 48 percent of workers who are fortunate to have an employer-provided retirement account capture 97.5 percent of the tax benefit.

The Saver’s Match

Beginning in 2027, the Saver’s Match will replace the nonrefundable Saver’s Credit with a refundable government match of up to $1,000 (half of the first $2,000 a worker contributes) to be deposited directly into a retirement account, and phased out by income and filing status.

About 30 million workers (20.1 percent) are income-eligible for a Saver’s Match. But roughly 21 million of them lack a qualifying account — a 401(k)-type account or IRA — to facilitate the receipt of the Saver’s Match.

This puts the Saver’s Match, as currently designed, in a difficult place. The wedge between eligibility and access is not something the Saver’s Match is currently built to solve.

In April 2026, a White House executive order established TrumpIRA.gov, with the aim of connecting workers without an employer plan to low-cost individual retirement accounts eligible for the Saver’s Match.5 While these plans will not be able to automatically enroll individuals, it is possible that TrumpIRA.gov will meaningfully help workers to access the much-needed, user-friendly retirement savings vehicles that can also facilitate access to the Saver’s Match.

So what are the facts?

America’s tax-advantaged retirement savings system is a powerful wealth-builder, but it is not one that all workers have access to. If we can find ways to close the access gap, we can help tens of millions of workers reach a dignified retirement as a just reward for years of work. How many could be helped?

Right now, 55 million private-sector employees — nearly half — are left without access to an employer-provided retirement plan.

Roughly 41 million private-sector employees without a bachelor’s degree — about six in ten — have no employer-provided retirement plan.

21 million low- and moderate-income workers lack a retirement savings plan that would allow them to access the Saver’s Match that they are eligible for — about seven in ten.

Of the hundreds of billions of dollars spent by the government in the form of forgone tax revenue to subsidize retirement, 72 percent flows to households in the top 30 percent of the income distribution, 55 percent to the top 20 percent, and 33 percent to the top 10 percent.

The problem is clear. The task now is to decide how we solve it.

For more facts about the U.S. retirement system, including full methodological details, see our extended analysis.

A new national survey from EIG and Echelon Insights finds that American workers across partisan divides think the president’s economic agenda has failed them — and on the single issue they care about most, the cost of living, workers say the president’s policies have made their problems worse.

Conducted September 1–8, 2026, with a representative sample of 1,507 American workers, the survey addresses not only economic policy issues but also workers’ assessments of their own incomes, savings, and career prospects in the age of AI.

Below we highlight just a few of the survey’s key and surprising findings, concluding with the big takeaways for the midterm elections. To see the comprehensive findings presented in a user-friendly slide deck, click here.

Affordability: Still Waiting

The costs of housing and everyday essentials continue to dominate workers’ concerns, with 68 percent saying these are the issues policymakers should be most focused on. Unlike any other issue we surveyed, affordability unites workers across political, generational, educational, racial, and gender divides.

Consistent with recent data showing that real wage growth has stalled over the past year, 68 percent of workers say their cost of living has risen since last September. A mere 11 percent say that things have become more affordable.1

And it’s on this paramount cost-of-living issue that workers disapprove of the president’s performance the most. Only 22 percent of workers grade President Trump’s performance on inflation and the cost of living as good or excellent, compared to 62 percent who say it’s been poor. Even among those who voted for Trump in 2024, only 46 percent give him positive marks.

Tariffs: No More, Please

President Trump’s signature tariff agenda is widely unpopular with American workers.

Perhaps the survey’s single most telling result is the collapse of support for tariffs among blue-collar workers — the policy’s intended beneficiaries.

While in our 2025 survey these workers favored President Trump’s tariffs by 3 points, blue-collar workers now oppose them 58 percent to 36 percent.

The reason for their opposition is clear. More than two-thirds of workers believe that tariffs have made their cost of living much or somewhat higher in the past year. A majority of workers hold that view across each and every demographic, educational, political, and occupational group.

Among Trump’s 2024 voters, it should be noted, support for tariffs has been resilient, with 69 percent continuing to back the policy. But not with much intensity: Fewer than a third of Trump-voting workers strongly support tariffs. And among GOP-voting workers, only 3 percent select tariffs as their top priority.

Support for tariffs is also highly conditional: Among the 34 percent of workers who say they approve of the president’s trade agenda, the majority say that tariffs should be pursued only if doing so does not lead to a higher cost of living and/or fewer choices for American consumers.

Immigration: When Legal, We Still Like It

Other Trump policies have proven similarly unpopular. Most strikingly, workers have massively shifted towards trusting the Democrats more than the Republicans on immigration, just as they have on trade (and taxes).

Workers display nuanced, favorable views toward legal immigrants in general and toward high-skilled workers in particular. Fifty-one percent now say that legal immigrants are definitely or probably an opportunity for workers’ own personal careers, an 11 percentage point increase from last year. Only 19 percent view legal immigrants as a threat.

High-skilled immigration is especially popular among workers. Sixty-five percent support more of it, including 62 percent of those who voted for Trump in 2024. And 51 percent of workers agree that allowing more highly skilled immigrants would “boost the U.S. economy and enable companies to expand their hiring, making things better for American workers.”

The partisan split on immigration is a big one, though. Nearly three out of four GOP-voting workers rate the president’s performance on immigration as good or excellent, compared to only one out of four Independents and just 14 percent of Democratic-voting workers.

Company Stakes: No Thanks, You Keep ’Em

A new issue in this year’s survey is government ownership of companies, with the Trump administration having taken stakes in dozens of firms including Intel, U.S. Steel, and MP Materials. (This is not a one-party position, as the idea has also been supported by prominent politicians on the left, including Senator Bernie Sanders, Governor Gavin Newsom, and Representative Alexandria Ocasio-Cortez.)

Big majorities of workers across all parties believe the government “should not be able to take an ownership stake in American companies even if it’s in a key industry,” with only a small share believing the government should be able to take ownership stakes at the discretion of the president.

There was more disagreement between workers of different generations, with Gen Z workers the most open to the idea (though support is still a minority view for them).

AI: A Threat (But Not When I Use It)

American workers have a favorable view of technological progress overall, with 60 percent believing that “new technologies” are an opportunity for their personal careers, compared to only a quarter who believe they are a threat.

But views on artificial intelligence specifically are much more mixed, with 43 percent of workers calling AI an opportunity and the exact same share calling it a threat.

Politically, workers appear to be more aligned against AI than aligned with it.

Forty-nine percent of workers say that a candidate’s pledge to fight against AI companies would make them more likely to vote for that candidate, while just 17 percent say such a pledge would make them less likely to vote for that candidate.

Data centers are similarly unpopular, with a majority of workers opposing construction in their own communities. Fifty-three percent of workers say that a candidate’s pledge to fight against data centers would make them more likely to vote for that candidate. Only 14 percent say such a pledge would make them less likely to.

But these results emerge in the context of limited real-world AI usage among workers. Only 39 percent of workers say they use AI tools for their jobs. Even among white-collar workers, just a small majority (55 percent) use AI at work. But usage is clearly rising, especially for personal tasks.

Importantly, among those who actually have used AI at work, the large majority say it has helped them be more efficient, has improved the quality of their work, and has enhanced their job satisfaction. These findings suggest the possibility that opposition to AI may soften among workers if usage continues rising.

It’s worth underscoring that white-collar workers, the group that experts believe is most exposed to AI disruption, are the least likely of any occupational group to call AI a threat, the most likely to actually use it, and the most likely to want the pace of AI development to speed up. And among users, white-collar workers are by far the most likely to say it has improved their work quality.

Perhaps most surprising of all is that, for all the concerns about AI-driven job displacement and opposition to data centers, a plurality of workers (49 percent) actually supports maintaining or accelerating the pace of development. As with other issues related to AI, the “speed up or slow down” issue reveals deep partisan divides: 65 percent of GOP-aligned workers want to keep up or increase the current pace, compared to only 38 percent of Democratic-aligned workers and 44 percent of Independents.

I Like My Job, but Will I Have to Work the Same One Forever?

Workers’ pessimistic evaluations of the economy, job market, and cost of living do not always extend to their own circumstances.

Forty percent of workers say they are more optimistic about their personal financial situations than they were this time last year, while only 29 percent say they are less optimistic.

About two-thirds of workers say they are satisfied with their current jobs, versus just 14 percent who say they are unsatisfied. And vastly more workers are satisfied with their careers and the amount of money they earn than are dissatisfied.

But there is an outlier: retirement savings.

More workers are dissatisfied with the amount of money they have saved for retirement than are satisfied, a surprising contrast to their overall satisfaction with the benefits they get from their jobs.

Despite their broad work satisfaction, when asked about finding new jobs, workers are less sanguine. Half of workers say it would be hard to find a good job with good pay for them personally, and even higher shares believe it would be hard for someone in their community or American workers in general to find a good job with good pay. This negative sentiment has grown over time.

Socialism Is Cool, Except for the Socialist Parts

Workers overall still prefer Capitalism to Socialism. But Democratic-voting workers now have a stunning +31 net favorable view of Socialism alongside a net unfavorable view of Capitalism, possibly showing the rising influence of the Democratic Socialists of America.

Intriguingly, however, Democrats also overwhelmingly believe that their taxes are too high (84 percent) and that the government should not take stakes in private companies (69 percent).

We don’t know what to make of it, but we find the results amusing enough to pass along.

Midterm Takeaways

These survey results should be sobering for Republicans.

When asked which party they think has the better approach on the cost of living, workers now favor Democrats 44 percent to 27 percent.

In our survey ahead of the 2024 election, for comparison, workers favored the Democratic Party’s approach by just 1 percentage point — 37 percent to 36 percent. But now, ahead of the midterms, even two out of five workers who voted for Trump in 2024 do not favor Republicans on this question.

Asked for a comprehensive evaluation, workers overall now favor Democratic Party candidates at 54 percent to 39 percent. Among workers who are “extremely” likely to vote, 59 percent favor Democratic candidates versus 39 percent for Republicans.

Black and Hispanic workers now favor Democrats 69 percent to 27 percent and 63 percent to 30 percent, respectively. White and AAPI workers favor Democrats too, but by slimmer margins. The vaunted multiracial working-class coalition that the Republicans assembled in 2024 has, it seems, been undone by the party’s management of the economy.

In terms of household income, the only workers who favor Republicans are those making between $30,000 and $50,000.

Among workers with household income above $200,000, Democrats are preferred 67 percent to 28 percent.

When workers are asked which party they think has an approach that is better for American workers’ interests, they favor Democrats over Republicans 44 percent to 29 percent — a significant swing in their direction from our poll two years ago. Democrats now have the advantage here among white-collar workers (+14 points), blue-collar workers (+11), and service workers (+16).

When asked to rate President Trump’s performance on the economy generally, 54 percent of workers express that his performance has been poor, versus 28 percent who rate it good or excellent.

American workers are dissatisfied with the direction of the economy and dissatisfied with economic policies that fail to deliver higher wages and greater purchasing power. The lesson of recent elections is that these sentiments spell bad news for the incumbent party — even when other economic indicators, such as job growth and the stock market, are robustly positive.

It’s the cost of living, stupid. Always was.

For the full slide deck that accompanies this post in addition to all updates of the American Worker Project from the Economic Innovation Group, please click here.

Quick — without scrolling down, what stories about the American labor market would you tell from this chart of real wage growth since December 1982?2

You might first draw attention to the highest line, the 90th percentile wage, which has pulled away from all the other wages in the last 44 years, seeming to confirm narratives about rising inequality throughout this period.

Or maybe you would point out that every single wage percentile has climbed since at least the middle of the 1990s, dispelling the idea that wages have been stagnant. They are now almost as high as they have ever been.3

Those are the obvious stories and they are not wrong or insignificant. But they do leave out a lot, especially about the many evolutions of the economy during such a long and eventful period.

To really surface the lessons in the chart requires us to zoom in and out of it, breaking down this long era into a handful of mini-eras, searching for the commonalities and differences between them. It requires capturing the big turning points and remembering the unique pressures that drove each of them.4

The stories that emerge — six of them — offer a fuller portrait of an economy that has been experienced in radically different ways across generations of workers.

Story 1: Sustained tightness in the labor market really matters for widespread wage growth.

Here is another view of the same chart, but with the shaded bars in the background representing periods when the unemployment rate was at or below 5 percent.

The broad-based positive wage growth of the past three decades took place almost entirely during periods in which low unemployment was sustained for at least three years.

Unemployment of 5 percent or less is not, of course, the only available proxy for a tight labor market. Others would show slightly different dates for when the tightness started and stopped. But the same story will hold for most of them.5

Our read of the data, by the way, is that a tight labor market is massively helpful but not always sufficient for widespread wage gains. The events of the past year, with unemployment staying low but real wage growth turning negative, shows that other variables matter too. More on that later.

Story 2: For a majority of this 44-year period, labor markets have not been tight.

Of the 524 months shown in the chart, unemployment has been above 5 percent for 324 of them, or 62 percent of the time. Most of those months came during two exceptionally long stagnant periods. The first includes most of the 1980s through the first half of the 1990s, and the second begins with the early 2000s dotcom bust and ends roughly halfway through the 2010s.6

As would then be expected, the substantial growth of real wages in these decades has taken place during the other three relatively short periods: the late 1990s IT boom, the tightening labor market of the late 2010s, and the post-COVID recovery.

Figure 3 shows all five eras — the two long stagnations and the three wage booms — and the duration, annualized growth in the median wage, and average unemployment rate for each. (See Footnote 6 at the end of this post to read more about the exact dates for each era and why we chose them. We fully accept that others might choose different dates.)

Unsurprisingly, the high average unemployment during the two long stagnations coincided with abysmal median wage growth.

What effect might the sporadic, unreliable timing of positive wage growth have had on how the entire post-1970s period has been perceived by workers and their families?

We know from the work of economists such as Ulrike Malmendier and Stefanie Stantcheva that life experiences can have lasting effects on behavior and attitudes. A given worker’s understanding of how wages have changed over time is therefore likely to be influenced by the specific years in which their professional life started and stopped.

Unavoidably, as with inequalities of compensation and wealth, the luck needed for a worker to overlap time in the workforce with an era of strong wage growth is unevenly distributed.

Older workers who have been in the labor market since the early 1980s and are now entering retirement age will have lived and worked through both long stagnations, enjoying positive wage growth for barely more than a third of their working years.7 Some of these workers might come to believe that wage growth has been worse than it really was or perhaps even stagnant for the entire period because most of the time it was in fact stagnant.

This framing also helps explain the plight of Millennials, the cohort born between 1981 and 1996, during their young adulthood. Millennials first started entering the labor market in the early 2000s, just as the bursting of the dotcom bubble inaugurated a brutally long period of wage stagnation. By the end of that period in the mid-2010s, they were known for their cynical views on social trust and political institutions. Their disposition might have been caused in whole or in part by any number of things — smart phones, social media, populist politicians — but it would not be surprising if much of their distrust was the result of economic frustrations.

The objective fact that the short stints of rising wages have more than offset the long periods of stagnant or negative growth, leading to positive and significant real wage growth overall through the decades, should absolutely be celebrated. But it may not be enough to contradict the subjective feeling among some workers that wage stagnation is all they have lived through. Their impression is not accurate, but it is understandable, as the psychological impact of the inconsistency of wage growth has perhaps gone underappreciated.

Story 3: Wage inequality has climbed mainly during the two long periods of overall wage stagnation.

Of the five eras, only the two long wage stagnations have produced sharp rises in wage inequality.

Figure 4 zooms into the First Long Wage Stagnation from December 1982 to 1996:

A complete disaster — negative growth for the 10th percentile wage and hardly any growth for the three middle wage categories. Meanwhile the 90th percentile wage climbed away from all the others despite itself growing by less than 1 percent a year.

Figure 5 shows the more recent Second Long Wage Stagnation.

Again, terrible, and recent enough that a majority of today’s workforce experienced it.8

The first half of this era was dominated by the phrase “jobless recovery,” with the unemployment rate never falling back to its 2000 low before the economy slammed into the Global Financial Crisis of 2007–09. Unemployment after the crisis spiked to 10 percent, followed by a terribly slow return to health for the American labor force.

For the entire era, which lasted nearly 14 years, the median wage averaged just 0.36 percent growth per year (compared to the robust 2.6 percent annual growth during the preceding boom).

Growth in the 10th and 25th percentile wages was barely positive, and both of these wages finished the era lower than their levels from early 2003, more than a dozen years earlier.

And because the growth of the 75th and 90th percentile firmly outpaced the growth of the others, wage inequality climbed sharply.

No wonder that by the middle of the 2010s, inequality had become the dominant theme of so much research and commentary and of so many books. Capital in the 21st Century, the Thomas Piketty doorstopper that pundits everywhere pretended to have read, arrived in 2014. It was accompanied by inequality books from Joseph Stiglitz, Angus Deaton, Branko Milanovic, Anthony Atkinson, and many others.9

Academic and popular publications featured deep explorations of the relationship between inequality and populist politics, inequality and financial crises, inequality and the business cycle. Inequality was all anyone could talk about at the annual meetings of the American Economic Association, the big nerdfest gathering of economists every January.

Ambitious and seductive policies to alleviate inequality and reinvigorate the economy also became newly popular around this time. Universal Basic Income. A $15 federal minimum wage. A federally guaranteed job for everyone who needed one. The heterodox economic framework called Modern Monetary Theory gained traction with the media and public. Larry Summers reintroduced Alvin Hanson’s concept of Secular Stagnation.

The two of us have no interest in relitigating the merits of these intellectual fashions, though we admit that most of them are not our brand of bourbon. Our only point here is that they did not come from nowhere. They emerged from a severe, protracted ordeal for workers, when wages stagnated and inequality visibly climbed. Such times leave scars.

Thankfully, they also don’t last forever. Throughout history it has often been the case that the moment when intellectual currents catch up to a trend is exactly the moment when the trend changes. Something happened near the end of 2014, just as the chatter about inequality was reaching its feverish peak: Healthy wage growth returned.

Story 4: The last decade has been much better for real wage growth. (Aside from the Terrifying Global Pandemic interlude.)

From 2015 through the start of 2020, wages climbed all across the distribution, and the widening between the 90th percentile wage and the other wages slowed.

The escape from wage stagnation once again arrived as the labor market continued tightening. By 2019 unemployment had fallen to its lowest level since the 1960s. The number of job openings exceeded the number of unemployed people for the first time since the data became available in the year 2000. The prime-age employment ratio also returned to its highest level since just before the dotcom recession nearly two decades earlier.

The bottom-wage percentiles received a boost from a rising number of state and local hikes in the minimum wage.10 Leaving aside comparisons between minimum wage hikes to other wage-boosting ideas, one thing we would note is that most of them benefit from tight labor markets. Raising minimum wages is easier, for instance, when the economy is strong and overall wages are climbing, making it relatively less costly for employers to comply. A healthy macro environment can be a welcome stage for experimentation with micro ideas.

The late 2010s era of solid wage growth then crashed into COVID. The initial wage spike during the first few months of the pandemic, followed by a partial reversal in 2021, was the clear result of a recessionary composition effect. We therefore index Figure 7 to January 2022, by which point the composition effect had been mostly smoothed away:

The events of the past half-decade are easy to trace in the chart. Real wage growth was initially restrained as a result of post-pandemic inflationary pressures, which themselves were caused by some combination of lingering supply chain constraints, volatile economic shifts from services to goods consumption and back to services again, and the stimulative fiscal and monetary policies of the time.

But the labor market also re-tightened quickly, with unemployment falling to a shockingly low 3.4 percent in April 2023 before drifting back up. And as inflation gradually fell, real wage growth turned sharply positive again.

Story 5: Way too early to know, but we might have already begun a new era.

Recent indicators of labor market health have been mixed. The low-hire, low-fire dynamic of the past few years remains, and young adults in their early 20s have notably been struggling. But the unemployment rate is quite low at 4.2 percent, while the monthly pace of job creation, the prime-age employment-to-population ratio, and the prime-age labor force participation rate have all moved sideways.

Zooming in even closer to the post-COVID era, however, it is clear that real wage growth experienced a sharp stop around March 2025.

What to make of the conflicting trends? It is quite possible that the post-COVID era has already transitioned into an entirely different, murky new moment.

We noted earlier that other variables besides a tight labor market can affect real wage growth. Policies that undermine price stability and economic dynamism can counteract the positive wage effects of a relatively healthy labor market — policies such as higher tariffs, destabilizing attacks on key economic institutions, new bureaucratic impediments added to the legal immigration system, and war with Iran.

Perhaps these policies, in addition to the hazy visibility into the effects of Artificial Intelligence, are combining to signal the start of something new.

Story 6: The median wage has lagged.

Real growth of the median wage has trailed behind the growth of the other wage categories since 1982. Both the higher and lower parts of the distribution have outperformed it.

What explains the lag? We have no single confident answer.

Common explanations for the relative underperformance of middle-class wages have included technological shocks, globalization, slower productivity growth, rising monopsonistic pressures, financialization of the economy, and many others. Scholars have also done work on occupational polarization as a result of how technology has evolved.11 What most of these explanations fail to offer, however, is a reason for the stop-and-start nature of this divergence between the median and the rest of the distribution.

What we can say for sure about this 44-year period of history is that the top part of the wage distribution has outperformed both the middle and the bottom. And for roughly the last decade, the middle and the bottom have actually been converging.

Figure 10 shows the ratios between the 90th percentile wage and the median, the 90th and the 10th, and the median and the 10th.

Regardless of the mystery, a point worth emphasizing is that workers bounce in and out of the different wage categories throughout their careers. Just because median wage growth has underperformed does not mean that current middle-wage workers have themselves always lagged. Their wages climb up or fall down the distribution from switching jobs or careers, returning to school, scaling back hours to care for children or elderly parents, or shifts in luck — the unavoidable vagaries of life.

This fluidity is the reason that widespread and positive wage growth matters more than the performance of any single part of the distribution. That the real median wage has climbed 40 percent since the early 1980s despite its status as the laggard shows again that the labor market has enjoyed such growth — even if it has been far slower and at times less inclusive than anyone would like.

Now what?

How to interpret the wage trends of the last four decades is a frequent subject of debate among economists and economic commentators.

To be simplistic, one side of the debate argues that economic growth has powered American wages to nearly their highest levels ever, a clear sign that the American Dream endures. The other side responds that wage growth has been both unequal and also weaker than in the heady postwar decades, and that workers — many of them, at least — can be forgiven for not exalting the economic environments they have lived through.

We think both sides have a point.

Wages really are near their historic peaks, and it would be silly to believe that the American economy is fundamentally broken and in need of radical policies to overhaul it, which would risk undermining the strong parts of its foundations.

But it is just as clear that wage growth often disappoints for unacceptably long stretches, and that accidents of birth have often determined the prospects and prosperity enjoyed by some workers and not by others.

More generally, we believe it is perfectly compatible to recognize progress while arguing against complacency.

Another useful lesson of the past four decades is simply that fast wage growth and inclusive wage growth can clearly coexist. Indeed they hardly ever exist apart. The eras in which the median wage climbed at the steepest angle were those in which all the other wages climbed with it. And the two long stagnations, when the median wage mostly failed to grow, were those when the lowest wage percentiles suffered the most.

Even if just a coincidence, the idea of a tradeoff between growth and inclusion seems a misguided and ahistorical way to think about the labor market.

Careful readers might have noticed that the era of real wage growth we have not yet zoomed into is the single most successful one: the late 1990s boom, when the median wage climbed an annualized 2.6 percent. We do so in Figure 11.

So steep, so inclusive. Can policymakers return the economy to such a healthy place?

Other than demonstrating the clear relationship with a tight labor market, we have deliberately not said much about the causes of fast, widespread wage growth. Identifying them is not an exact science, and once again there are no easy or confident answers.12 But the late 1990s era at the very least has shown what is possible.

We also have not looked at the relationship between wages and productivity growth through the decades, or at the question of whether that relationship has disconnected. Nor have we discussed wage growth by educational attainment, gender, race and ethnicity, or geography. The details that emerge could have important consequences, if only through composition effects, for any simple narrative about wage growth over a specific period of time.

We will explore some of these dimensions in future posts, in addition to answering questions from readers. We did not intend this analysis to be the definitive statement on such a complex topic, as if such a thing were even possible. Our main goal has been to complicate and hopefully advance the conversation about American wages, not to end it.

According to our research, the most affordable metropolitan and micropolitan areas1 (or CBSAs) in the U.S. today are concentrated in Wisconsin, Iowa, Indiana, Ohio, and western Pennsylvania. In 2024, nine of the ten most affordable CBSAs were Midwestern, seven had a strong manufacturing core, and all ten ranked in the top half nationally for household income. These places are middle-class, with median household incomes around $75,000 and mean household incomes well into the six figures. They have modest population growth, low unemployment, and low poverty. Young adults with ordinary jobs are buying homes without breaking the bank. In many ways, these are communities where the American Dream is still alive.

Take Le Mars, Iowa. Dubbed the ice cream production capital of the world, it was the most affordable housing market in the country in 2024: the typical resident spent just 17 percent of their income on housing costs that year. It is also comfortably middle-class, with a median household income of roughly $83,000, and thousands of residents employed by the Blue Bunny ice cream manufacturing plant. Warsaw, Indiana, is another affordable place for both renters and owners, with a growing industrial economy. More than a third of its jobs in 2024 were in manufacturing, and the region recently secured a $400 million investment for a new factory specializing in electric pickup trucks.

These communities are economically thriving and homes are still attainable. As the map above shows, while the Midwest is full of affordable communities, across much of the rest of the country they’re rare. The West Coast and the Northeast corridor are famously expensive. But we find that even in Sunbelt states like Florida and Texas, which have a reputation for cheap housing, every one of their most populous metros is unaffordable too. What has enabled the Midwest to succeed while so much of the rest of the country struggles?

What makes places unaffordable?

When housing costs start climbing, people often blame income or amenities. Both explanations are about demand: who wants to live in a place, and what they can afford to pay once they’re there.

The income story comes in two versions. In one, high-paying jobs in industries like tech and finance attract workers who bid up home prices for everyone. This is indeed one reason that superstar metros like Los Angeles and New York are so unaffordable.

In the other, the loss of good-paying jobs results in incomes too low to cover even the existing housing stock, as in unaffordable distressed metros like Monroe, Louisiana, and Laurinburg, North Carolina. But across all CBSAs, the link between median household income and cost burden is only weakly positive, and plenty of middle-income places — like Eureka, California — are highly unaffordable too.2 Incomes alone — whether high or low — don’t explain unaffordability.

Another theory is that the most expensive housing markets are simply the most desirable places to live, whether because of cultural amenities or gorgeous weather. Many of the least affordable places in the U.S. today are coastal, like the beach towns of Florida and Southern California, where good weather, natural beauty, and strong tourism and hospitality sectors keep drawing transplants. Key West, the least affordable market in the country in 2024, is the archetype. The typical household there spends 34 percent of its income on housing — twice the share in Le Mars. Year-round sunshine keeps demand high and the economy runs on hospitality jobs that pay far less than finance or tech. As an additional burden, property insurance costs are rising fast.

Demand, though, is only half of the housing equation. Basic economic theory predicts that housing will be least affordable where the demand for homes most exceeds the available supply. And that mismatch can be measured.

The supply gap

Up for Growth, a pro-housing advocacy group, has estimated the rate of housing underproduction. In their methodology, it’s the gap between the homes a region needs and the habitable homes it actually has. Need includes both existing households and those that would likely form if they could, like young adults still living with their parents. By comparing these estimates with cost burdens across CBSAs, we find that housing shortages track closely with unaffordability. The least affordable places have both the largest and the longest-running shortages.

In 2023, the typical unaffordable market was short by an amount equal to 2.4 percent of its existing housing stock. Meanwhile, affordable markets were short by only 0.6 percent, and for much of the past decade they ran a surplus. While shortages have emerged throughout most of the country since 2012, unaffordable places have deeper gaps.

Los Angeles has run a shortage of 7 to 8 percent of its housing stock every year of the past decade, roughly 334,000 missing units. Even fast-growing metros that have built aggressively haven’t kept up with demand. Miami’s shortage spiked past 8 percent in 2019 before easing to 4.5 percent by 2023. Naples, Florida, grew its housing stock by a quarter over the decade and still crossed from a surplus in 2012 into a shortage by 2023, as demand outpaced rapid construction.

Where the shortage is money, not homes

The relationship between shortages and cost burdens holds in nearly every type of market, with one exception. When we sort unaffordable markets into quartiles by median household income, it holds everywhere except the bottom quartile. Low-income unaffordable places generally don’t have a housing shortage, and many run a surplus. Their unaffordability is a result of economic distress: cost burdens are high because incomes are so low, not because of housing underproduction. This result isn’t surprising. Where low incomes and weak demand are what hold back population growth and household formation, a shortage of homes isn’t the binding constraint.

Monroe is one such place. The birthplace of Delta Air Lines, the northeast Louisiana metro was once thriving, but has since thinned out. Its population is shrinking, the typical household earns about half the national median, and a quarter of residents live in poverty. Monroe has a housing surplus equal to more than 3 percent of its stock, yet its residents suffer from unaffordability. Macon, Georgia, has a similar story, with nearly 15 percent of its homes sitting vacant today. In places like Monroe and Macon, building and housing reinvestment won’t happen on their own since new construction largely doesn’t pencil out. The result is that these CBSAs suffer from a mismatch between incomes and housing costs, despite the overall housing surplus.

This does not imply that place-based policies incentivizing preservation and new construction in distressed places are inherently unhelpful, of course. A distressed zip code or Census tract next to a high-income, high-opportunity one is fundamentally different from a distressed region far from any jobs; adding homes in the first kind of place can help to ease affordability concerns as opportunities attract newcomers and add cost pressure on existing residents. Additionally, even if there is a surplus of homes overall, there may still be a shortage of certain types of homes. Several low-income and unaffordable metro areas with a housing surplus are large college towns, like Gainesville, Florida, that have a serious mismatch between the population of single renters and the stock of apartments. In these cases, allowing for more construction of the kinds of homes that better fit local needs could help to ease costs.

With those caveats aside, low-income metropolitan and micropolitan areas account for only 3 percent of the national population. For nearly everyone else, affordability is linked to how well supply is meeting demand. This is a straightforward lesson that affordable places in the Midwest today should not take for granted.

Storm clouds on the horizon

For much of the past decade, the Midwest lost residents to other regions as young people migrated to major job centers on the coasts and in the Sunbelt. Its overall population grew only modestly, due to births and international migration. However, there’s been a brewing pressure in the background: like much of the country, the composition of households in the Midwest has been changing. New households have been forming faster than population growth. As seniors age into smaller empty-nester homes and younger adults spend more years living alone and having fewer children, the same number of people is splitting into a greater number of households. In Wausau, Wisconsin, the population has grown just 4 percent since 2014, but the number of households has grown 11 percent. Definitionally, household growth — not population growth — is the metric that matters for housing.

But now a second source of demand is arriving. In 2025, the Midwest recorded positive net domestic migration for the first time in a decade, as more Americans moved in from other states than left. It’s a small shift so far, but it signals that the same demand pressure that made other regions unaffordable may now be moving to the heartland. For the Midwest, the Mountain West region offers a warning. Though it once had many of the same affordability characteristics as the Midwest, the Mountain West now looks a lot like the West Coast. After a large pandemic-era demand surge, Colorado and parts of Idaho, for example, now have housing affordability problems reminiscent of those in California, Oregon, and Washington. The Midwest has not yet had to build fast, which is one reason its shortages have stayed small.

But in a number of Midwestern CBSAs the problem is starting to surface. Appleton, Wisconsin, ran a housing surplus in 2012 equal to about half a percent of its stock; by 2023 it had swung to a shortage of nearly 6 percent, the largest gap among affordable places. Jefferson City, Missouri, still had a surplus as recently as 2019, but now runs a shortage close to 3 percent. Wausau followed the same path, and even slow-growing Sheboygan, Wisconsin, has recently tipped from a surplus into a modest shortage. In each case the mechanism is identical to the costlier regions of the country: household growth is outpacing new construction.

What it will take

The evidence that supply lowers costs is abundant. Study after study finds that building more housing slows price growth. The obstacle is that in most places, restrictive land-use rules make building enough of it overly burdensome or outright illegal.

Some of that is beginning to change. A wave of state-level zoning reforms and the passage of the federal 21st Century ROAD to Housing Act are real steps. In the Midwest, major citywide reforms in Minneapolis and Columbus are poised to help. But the country is still millions of homes short.

Rather than mandating specific outcomes from the top down, a promising policy approach rewards the places that actually permit housing and lets each community meet its own needs — the idea behind EIG’s Right-to-Build Zones. Building more won’t fix everything; distressed places need place-based policies that bring investment and jobs more than they need housing permits. But for most of the country, the binding constraint is simply that there aren’t enough homes. It’s time for America’s last affordable places to heed the warning.

What our data shows is that many small town and rural communities are finally enjoying some wind in their sails to support revitalization efforts, as population growth in the nation’s least populous areas finally turned positive in 2020 — a small silver lining of the pandemic.1

The use of relocation incentives surged in the early 2020s. A few years in, however, several pioneers mentioned in the Times piece have abandoned their programs after local leaders determined that their opportunity cost was too high. Programs were not particularly expensive, but local governments need to be able to justify every dollar they spend, and the incentives simply attracted too few permanent new residents to be worth continuing.

And yet interest in relocation incentives is likely to remain high in this age of declining birth rates, falling international migration, and slowing domestic migration. In this piece we analyze recent trends in county-level population growth and dig deeper into the uneven population recovery in rural areas. We then discuss why population loss matters and can leave communities searching for tools to regain some control over their demographic trajectory. We finish by exploring the costs and benefits of relocation incentives, which can give communities a leg-up when structured well but are ultimately zero sum in nature.

Two big things

Let’s start with some broader context. There are two major developments in county population trends over the past several years:

1. Slowdown in large urban county population growth. The large urban county population slowdown has now lasted about a decade. Big cities had enjoyed the strongest population growth of any category coming out of the Great Recession, but their advantage started to fade by 2016. The pandemic then hit large urban counties hard (remember the donut hole?) before they regained some of that lost ground as international migration picked up from 2022 to 2024. But then came the 2025 immigration crackdown, and big cities fell back to their low population growth trendline.

2. Shift to population growth for rural areas. The other big reversal, this one in a positive direction, is for rural areas. After declining and then stagnating for much of the 2010s, rural population growth accelerated in 2021 and remains elevated compared to pre-pandemic growth rates. As illustrated in the figure below, the rural population increased by 2.1 percent from 2019 to 2025, up 2.3 percentage points from the six years prior. This rebound extended to small-town areas as well, where growth rates improved by 1.2 percentage points to 3.3 percent from 2019 to 2025.

One finer point

The pandemic-era tailwinds did not reach all rural areas, though.

Rural portions of metropolitan areas drive most of the post-pandemic rural population growth. Despite making up roughly one-quarter of America’s rural population in 2019, rural counties within metro areas account for over two-thirds of total rural population growth since then.

Metro rural counties grew 5.5 percent from 2019 to 2025 — in line with suburban counties — while non-metro rural counties grew just 0.9 percent during the same period. Non-metro rural areas did reverse their long-term population decline after 2020, but their population remains below 2010 levels.

Age and place

Over 55s are driving rural population growth in rural counties both within and outside metro areas. This was true before the pandemic, and it remains true after. What is different is that rural areas closer to urban centers now enjoy outright growth in younger segments of the population. In addition, the outright loss of young and working age residents in rural counties outside metro areas slowed dramatically, but not sufficiently to result in a return to growth in these demographic groups.

There are clear regional differences in the age groups driving rural population growth, as the below map shows. In the Mountain West and East Texas, prime working age adults helped power the rural resurgence. In the Northern Great Lakes and Southeast, the over-55 cohort dominates.

Why this matters

Trends in population growth and decline carry major economic and social implications. Population decline in particular matters because it often makes the work of economic development harder. As our colleagues Sarah Eckhardt and Adam Ozimek have written:

Declining populations often coincide with shrinking economic opportunities, aging workforces, and the erosion of vital local services such as hospitals, public schools, and infrastructure built for larger populations…

…One of the most important consequences of population loss is lack of dynamism. A growing body of scholarship shows that demographic decline is linked to lower rates of business formation. Fewer residents doesn’t only mean fewer entrepreneurs, but also a smaller labor force. Both issues make it less likely that a community will be able to generate startups [and attract investment]…

…Another unavoidable result of demographic decline and lack of investment is that the tax base erodes. This in turn causes other deteriorations in quality of life. Population loss can also lead to local vacancies as some neighborhoods empty out and businesses shutter without replacement.

In short, local population decline is often a trap from which communities struggle to break free.

So what’s a community to do?

Given the momentous implications of population growth, communities have an interest in shaping their demographic trajectory.

Communities basically have two levers under their control to influence population growth: retention and attraction.

To retain residents, places need to maintain a strong value proposition: a high quality of life supported by strong job opportunities with a competitive cost of living. For many large urban counties, the pandemic delivered a negative shock to this proposition, with quality of life deteriorating just as strong job opportunities geographically decoupled to other places with the rise of remote work. The Wall Street Journal recently called out Seattle and Washington, DC, as two cities that have bucked the wider demographic trends affecting large urban counties by maintaining their appeal for families, in particular, with a range of quality schools and cost-competitive housing markets relative to neighboring suburban jurisdictions.

When it comes to attracting new residents, communities can compete passively or actively.

Competing passively involves trading on the same basic value proposition that drives retention — the amenities and opportunities an area offers for a given price (cost of living). Many of the rural areas that boomed during the pandemic found they were newly or increasingly competitive on the merits following the pandemic and remote-work shocks. Sun Belt growth poles have long competed relatively passively on price and quality of life while setting themselves up for success by permitting more housing, too.

Relocation and remote worker incentives represent the newer strategy of competing much more actively on the market, or paying for someone to move. Communities might choose to do this to abate population losses. They might do this to stock the talent pipeline. And they might do this because each prime-age white collar digital nomad offers a tidy, tiny little stimulus to the local economy. Regardless of the motivation, this represents a much more active intervention in the market for residents.

Remotely interested

Relocation incentives are a topic we have thought a lot about. Back in 2021, EIG conducted the first full economic impact analysis of the Tulsa Remote program. Tulsa Remote was an early mover in this space, and the program stood out for three main reasons:

• The scale at which the community has tried to attract talent (thousands over only a few years);

• The wraparound services the program offers new remoters both to increase the likelihood that incentive recipients stay and to juice each one’s expected local economic multiplier; and

• The extent to which the program fits into a broader constellation of local economic development initiatives, all working towards a shared local vision of economic development.

In short, Tulsa combined a people-based incentive with a constellation of other place-based investments to bolster its attractiveness for migrants in an intentional, well-rounded strategy.

As DePillis noted, Tulsa Remote is really the gold standard for these types of programs. Even Tim Bartik — traditionally an economic development incentive skeptic — speaks positively of the program, finding that every $1 it spent attracting a remote worker generated $4 in value for local residents. That compares starkly to the bad deal that most communities get when they pay for companies directly to relocate jobs (which can cost hundreds of thousands of dollars per job).

Nevertheless, three reflections should inform how communities approach such incentives.

1. Relocation incentives rest on strong assumptions about the multiplier effect — meaning the wider economic impact that the marginal highly-educated, highly-paid worker brings to a local economy. Increasing an area’s population one person at a time is a slow process, but communities utilizing relocation incentives are typically betting on quality over quantity, aiming to attract highly-educated, high-earning individuals predisposed to start businesses or engage civically. For a state, a remote worker acts much like an export earner, selling their labor to an out-of-state buyer and bringing income, spending, and tax receipts back home.

2. Any marginal population changes generated by these relocation incentives risk being dwarfed by larger shocks and shifts affecting where Americans live and work. Recall what communities have navigated over the past few years: COVID-era migration shifts catalyzed by the existential push factors of a pandemic and the sudden and widespread adoption of remote work; the surge and then dramatic curtailment of international migration; and the aging of the population and continuous decline in birth rates, which has weakened the contribution of the”natural increase” tailwind to population growth. Even in Tulsa, population growth has flatlined over the past couple of years as events overtake even the best local intentions.

3. There’s a leaky bucket caveat to local migration policy. On their own, relocation incentives only open a spigot for population inflow; they do nothing to stem the leak at the bottom (residents leaving). That’s why the most effective programs treat these incentives as merely one component of a broader development strategy. Again, this is where Tulsa is due a lot of credit: Tulsa Remote is one element of a whole suite of initiatives to create buzz and spur local economic growth.

Domestic migration is zero sum. You know what isn’t?

Competition across communities is healthy. It forces places to continuously evaluate and invest in the value proposition they offer residents and businesses. But domestic migration is ultimately a zero-sum game. Every in-migrant to one community is an out-migrant from another. The pandemic increased the pool of migrants nationwide, but only temporarily; the domestic migration rate has since fallen back to earth. Americans’ proclivity to move to opportunity remains near its all-time low.

Which is why American communities have long embraced international migration as a demographic lifeline. Immigrants have driven the revitalization of Rust Belt cities and stabilization of rural hamlets. The slowdown in international migration is putting more communities under demographic and economic stress. The crackdown on high-skilled immigration is fiscal lunacy. The negative repercussions will only deepen the longer the slowdown persists.

A proposal such as EIG’s Heartland Visas — a new visa category that jurisdictions staring down demographic decline could opt into — would give locales another tool for taking control of their demographic fates. It would deliver a healthy fiscal bump to communities without poaching talent from their neighbors. And while Heartland Visas might function best in legacy cities where the embers of agglomeration still burn, the tool would be useful in rural areas, too: More than 200,000 immigrants settled in non-metro rural counties between 2021 and 2025, representing more than one-quarter of all net-migration into non-metro rural areas.

Until the window of opportunity for a more enlightened approach to international migration opens again, communities in the market for residents may find worker relocation incentives an increasingly attractive tool. The situation could be worse — they are less expensive and less dubious than chasing smokestacks — but it could be better, too.

See our GitHub with replication code here.

America’s tax-advantaged retirement savings system helps tens of millions of workers achieve a dignified standard of living in old age. But for all of its success, the system leaves tens of millions of other workers on the outside looking in.

With a bevy of new facts, figures, and charts, EIG senior economist Ben Glasner published in September the latest entry of EIG’s Retirement Fast Facts, a vital guide to who is able to save what for retirement and how. Here’s what Ben finds:

• Access to employer-sponsored plans is split almost right down the middle, with about 48 percent of American workers aged 18–64 having such an option and the other 52 percent lacking one.

• The 52 percent of workers without access number about 76 million.

• Moreover, only 37 percent of all workers receive the boost of an employer contribution or match. Among workers who do have an employer contribution or match, the median sum is about $3,000 per year.

As Ben explains, the existing system tilts to the advantage of college-educated, high-earning workers. Breaking down the findings by demographic and income characteristics suggests that, while useful, the system results in some perverse outcomes, especially if we consider tax advantages as forgone tax revenue.

• About 69 percent of private-sector employees with bachelor’s degrees or higher have access to employer plans, compared with just 36 percent of employees with a high school diploma. Among Black employees with a high school diploma, 29 percent have access to an employer plan; among Hispanic employees with a high school diploma, just 25 percent.

• The income gradient is even steeper. Within the private sector, 84 percent of employees in the top income quintile have access to an employer plan, while only 14 percent of employees earning in the bottom quintile do.

• Tax-advantaged retirement savings made in 2024 cost the federal government more than $250 billion, measured on a present-value basis.

• The bottom half of workers receives only about 10 percent of retirement tax benefits.

Beginning in 2027, the Saver’s Match, passed in the SECURE 2.0 Act, will offer a government match of up to $1,000 to about 30 million workers. While the plan has some kinks, the April 2026 White House executive order establishing the TrumpIRA program could help facilitate the program’s success.

Read the latest installment of EIG’s Retirement Fast Facts here.

Research & Analysis

• EIG research director Nathan Goldschlag published his detailed and comprehensive investigation into entrepreneurship trends and whether AI is already influencing them. His key findings:

• Business applications for startups likely to employ workers were flat in 2025 and started rising in 2026, but fewer of them are of the type most likely to hire employees.

• Applications for likely nonemployer startups rose sharply in 2025 and early 2026, but they are probably side-gigs because they don’t show up as individuals’ primary source of income.

• AI doesn’t seem to have much to do with any of this beyond the surge of likely nonemployers in the information and professional services sector. (Which, again, are likely side-gigs.)

• Sam Peak and Jason He analyzed the likely effects of the Department of Homeland Security’s proposed $100,000 fee on Optional Practical Training (OPT), a program that allows international students to work in the United States after graduating. “Regardless of whether this fee falls on schools, employers, or the students themselves,” they write, “it will be the American people who bear the costs. OPT constitutes America’s most important pipeline for retaining the world’s most promising scientists and inventors.”

  ---

Policy

• As states finalize the next round of Opportunity Zone nominations, EIG’s new guide, OZ 2.0: What Comes After Zone Designations, offers a practical playbook for turning OZ status into lasting economic impact. Drawing on lessons from OZs 1.0, Kenan Fikri and Catherine Lyons outline strategies for state and local leaders to cultivate local investment ecosystems, prepare communities for investment, help high-impact projects get across the finish line, and connect local residents, workers, and entrepreneurs to the opportunities that follow.

• Jiaxin He and Sam Peak submitted a public comment criticizing DHS’s proposed $103,265 fee on H-1B visas. They point out numerous mathematical and conceptual flaws in the fee rule’s analysis, finding that the fee could lead to the H-1B program being undersubscribed by 27,000 and raising $2.8 billion less revenue than the agency projects. He and Peak also point out that the rule undermines the White House’s stated goal of “attracting and retaining top-tier global talent in critical national security science and technology fields” because the rule advantages older, less skilled workers at the expense of retaining today’s most promising young professionals. The rule, they point out, is explicitly designed under the assumption that “an employer’s willingness to shoulder the fee is proportional to how much their H-1B hires are underpaid.”

  ---

Around the Horn

• In Politico, Kelsey Brugger and Brian Faler quoted EIG president John Lettieri on the blue-collar jobs that data-center construction can create in Opportunity Zones.

• In the New York Times, Lydia DePillis quoted EIG senior fellow Kenan Fikri on Americans’ declining propensity to move across the country.

• In Bloomberg Law, John Kreighbaum quoted EIG policy manager Sam Peak on the Trump administration’s $100,000 H-1B fee.

• In Plan Advisor, Valentina Baez highlighted a new report from EIG senior economist Ben Glasner on American workers’ access to employer-sponsored retirement accounts.

• In National Review, John Fund quoted EIG research from Kenan, Ben, and Sarah Eckhardt on American communities’ reliance on government transfer income.

• In the Washington Sun, Jade Lozada cited work from Ben, John, and Adam Ozimek on housing and Opportunity Zones.

• In City Journal, EIG fellow Jordan McGillis reported on Alaska’s blue-collar labor market.

• Also in City Journal, Jordan celebrated the renter-friendly provisions in the ROAD to Housing Act, passed by Congress this summer.

• And in Vital City, Jordan penned an analysis on the status of middle-class families in New York’s tight housing market.

  ---

EIG Charts of the Month

Nathan Goldschlag’s analysis of entrepreneurship trends includes this chart, which shows that business applications are falling for the two kinds of startups — those that register as corporations and those that give a date for when they will start paying wages — most likely to eventually become actual job-creating businesses.

From the OPT analysis by Sam Peak and Jason He, a chart showing the share of foreign higher-education graduates of American universities end up staying in the United States after graduation.

How much will Artificial Intelligence boost productivity? Will unemployment rates rise and, if so, by how much? Will AI transform the economy quickly or, like previous general purpose technologies, will it take many years or even decades?

These and so many other questions about the impact of AI on the economy remain unanswered. Or at least it is hard to feel confident in answering them. But there is one exception, at least for me: I am extremely confident that AI will increase entrepreneurship.1

And it will matter enormously for the economy. Each year, roughly 400–500,000 new firms enter the market. Few survive, but from those that do, a small subset will grow quickly and contribute disproportionately to overall job creation. Unburdened by the weight of legacy systems, processes, and technologies, young firms introduce new ideas, products, services, and ways of doing things.

Those are the startups with employees. Other startups, known as nonemployers or “solopreneurs,” will have no employees and be run entirely by their founder, either as the founder’s primary source of income or as a sidegig.

Recent stories in the New York Times, Washington Post, Wall Street Journal, The Economist, and other outlets suggest that AI has already started boosting entrepreneurship. It would be unsurprising, as startup activity often climbs during times of rapid technological change. And I would love to believe that it is already happening.

I have just published a new EIG analysis scrutinizing the available federal statistics in detail. And unfortunately, I find that the trend is simply not yet reflected in the data. In fact, separate from the question of AI’s effects, entrepreneurship itself may no longer be enjoying the post-COVID spike that raised hopes of a sustained new era of economic dynamism. My key findings are:

• Likely employer business applications were flat in 2025 and started rising in 2026, but fewer of them are of the type most likely to hire employees.

• Likely nonemployer applications rose sharply in 2025 and early 2026, but they are probably side-gigs because they don’t show up as individuals’ primary source of income.

• AI doesn’t seem to have much to do with any of this beyond the surge of likely nonemployers in the information and professional services sector. (Which, again, are likely side-gigs.)

This may change soon. I hope it does. But for now, an entrepreneurial renaissance just isn’t jumping out of the data. Again, the new analysis — which includes a detailed investigation of the strengths and weaknesses of the different methodologies that monitor this trend — can be found here.

OPT constitutes America’s most important pipeline for retaining the world’s most promising scientists and inventors. While most international students leave the United States upon graduating, a minority of them end up staying to launch their careers here, usually through OPT. The students who remain are disproportionately likely to hold an advanced degree, usually in a STEM field — indicating that OPT helps filter for those with the most sought-after skills.

OPT lasts for a maximum of three years, but those who continue to showcase their value in the labor market get sponsored by their employer for an H-1B visa afterward — and many of these H-1B holders will eventually also get sponsored for a green card. The process of sponsoring an H-1B and green card already requires employers to incur thousands in legal and administrative costs, further filtering for top talent.

International students comprise 5 percent of total students pursuing higher-education degrees in the United States, a smaller share than the average for OECD countries and much smaller than countries like Canada and Australia.

The small fraction of international students who do end up working after graduation tend to have valuable skillsets that complement American workers. For example, when DHS extended OPT from 29 months to 36 months for STEM graduates back in 2016, job opportunities increased for U.S. college graduates. In localities with more professionals working on OPT, employers were better able to expand their operations and increase their hiring of U.S. workers. The average employment gain was equivalent to roughly 450 added American workers per 100,000 residents — with no overall drop in wages.

Importantly, DHS’s expansion of OPT occurred less than one year before President Trump took office in 2017 and presided over one of the strongest labor markets in American history. Unemployment fell to its lowest point in roughly half a century in 2019 — the very same year that participation in the OPT program also peaked.

President Trump himself has long acknowledged the importance of retaining talented students, noting that many top international students “can’t even make a deal with their company because they don’t know if they’re going to be able to stay in the country. That is going to end on day one.” The $100,000 OPT fee would make the President’s goal impossible.

International students increase college enrollment for Americans.

A key way in which international students increase opportunities for Americans is that they expand college enrollment slots for U.S. students. International students pay roughly two to three times more in tuition than Americans. This additional revenue subsidizes in-state tuition and financial aid for domestic students. At the graduate level, an increase of 10 international students leads to the enrollment of 8 additional U.S. students.

In addition to reducing tuition costs for Americans, international students ensure that university STEM programs across the country remain open. Nationwide, international students represent over one-third of students studying math and computer science at the master’s level. For engineering, international students comprise over one-fourth of master’s students at U.S. universities.

Given that over half of these students work in the U.S. on OPT after graduating, adding a six-figure tax to the program removes a big draw for them to study in the U.S. in the first place. If universities are forced to pay a $100,000 fee for each graduate working on OPT, overall international student enrollment at the master’s level would fall by at least 35 percent. Universities would have no choice but to reduce enrollment slots and raise in-state tuition for American students.

But such an enrollment drop wouldn’t just cause STEM programs across the country to shrink; many of them would be eliminated. At the University of Cincinnati, for example, nearly 85 percent of master’s students in information technology from 2022 to 2025 were international. Such a program would likely cease to operate without international students, forcing approximately 300 enrolled U.S. students to pursue their studies elsewhere.

By making STEM graduate programs less abundant and affordable, the $100,000 OPT fee makes it much harder for young Americans to move up the income ladder. College graduates enjoy a large earnings premium over those who did not attend or graduate college. STEM graduates do particularly well, regardless of whether they land a job in STEM or not. The proposed fee, in other words, would undermine one of the most reliable means of upward mobility for American students.

The OPT fee further rigs the H-1B lottery to favor outsourcers.

With only 85,000 H-1B visas available every year, employers must compete for scarce slots in the H-1B lottery. IT outsourcing firms from India have historically dominated the lottery. In 2021, for example, half of the top 30 H-1B employers were outsourcing firms — a fact the Trump administration has cited as proof the program needs to be reformed. Without the OPT program acting as a stepping stone for international students to obtain an H-1B, however, multinational IT outsourcing companies may take the majority of H-1Bs.

This would be disastrous for America’s pipeline for skilled immigration. While international students who transition to H-1B status are more likely to earn high wages and add good-paying jobs to the American labor market, H-1Bs who work for IT outsourcers earn lower wages and tend to have less experience working in the United States. Relative to H-1Bs who started as international students, H-1Bs employed by outsourcers earn $9,000 less per year despite being six years older, on average.

In addition to paying lower wages, IT outsourcing firms are also known to violate H-1B rules and have been found liable in court for discriminating against American workers during both the hiring and firing processes.

Furthermore, DHS has proposed a $103,265 fee on H-1Bs in addition to the $100,000 OPT fee. Taken together, this means that international students would need to clear a $203,265 barrier to remain in the United States, meaning that IT outsourcing firms — the employers infamous for using H-1B to undercut American workers — would effectively get a $100,000 discount.

DHS’s previous attempts to reform how H-1Bs are selected have not improved the situation. Last year, DHS implemented a weighted lottery to ensure that H-1B applicants with higher “Wage Levels” were more likely to be selected. The problem with this approach is that wage levels aren’t the same as actual wages — they simply mean higher wages relative to a person’s location and occupation. For example, an experienced exercise physiologist in Jacksonville, Florida, earning $55,000 per year would have four times the odds of earning an H-1B as an early career software developer in San Francisco earning $155,000.

The Wage Level system also opens the door to gaming the system. An IT outsourcer in DeKalb County, Georgia, may intend to hire a software developer for $100,000. But if the employer categorizes this worker as “software developer,” they would only be certified at Wage Level I. To improve their chances at the lottery, the outsourcer would be inclined to miscategorize this individual as a “computer programmer,” an occupation where average salaries are lower. This effectively doubles their odds of success without paying a higher salary.

And because these Wage Levels also don’t take age or tenure into account, a worker at an IT outsourcing firm who is underpaid for his experience often has a greater chance of success in this new weighted system than a far more economically valuable young electrical engineer who recently entered the labor force. The change to a weighted lottery increases the expected number of H-1Bs awarded to IT outsourcing firms by 7.4 percent. And when combined with a recent Labor Department proposal to raise the wage level thresholds across the board, outsourcer participation in the H-1B program would rise by over 3 percentage points.

The OPT fee undermines the White House’s national security objectives.

Adding a $100,000 fee on OPT will also harm the White House’s strategy of “attracting and retaining top-tier global talent” in fields critical to national security. America’s ability to attract global talent has long been essential to its global technological leadership, with foreign inventors co-authoring roughly one-third of patents in semiconductors and pharmaceuticals. OPT is America’s best existing tool for retaining such top talent from U.S. universities.

Although international students comprise less than 5 percent of overall higher education enrollment, they account for 22 to 55 percent of doctoral STEM students, depending on the field. Around 80 percent of these international STEM PhDs use OPT to find employment during and after graduation.

A major reason why international students comprise such a large portion of the U.S. STEM graduate programs relative to the rest of the education system has to do with the fact that America’s advanced STEM programs are considered the best in the world — luring the best and brightest from across the globe. Without foreign STEM PhDs in particular, U.S. universities would lack the manpower to research chip design, 6G, and other critical innovations, ceding ground to adversaries like China.

OPT holders also contribute to AI, aerospace, and undersea engineering — the three critical areas needed to maintain a battlefield advantage identified by the White House in its National Security Science and Technology Strategy published in August of 2026. OPT holders play an especially prominent role in AI, where international students participating in the program are 10 times more likely to work in the field than the average U.S. college graduate.1

If America won’t import top talent, it will import Chinese innovations.

When DHS previously made it more difficult for companies to hire workers on H-1B status, companies based in the United States responded by offshoring more of their operations abroad and enabling countries like China, India, and Canada to reap the benefits of new jobs and inventions that would have otherwise been created on American soil.

The proposed $100,000 fee on OPT will likely do far greater harm than good, depriving American workers of the opportunity to be at the forefront of the latest technological revolution reshaping the global economy. China is resorting to exit bans to keep its technical talent from leaving the country. Why, then, should the U.S. actively push away talented international students and squander its advantage as a freer, more prosperous society? America has a choice: It can import the talent that produces the world’s most consequential innovations, or it will be forced to import those innovations from adversaries like China instead.

Despite its nominal commitment to advancing U.S. innovation as a means to greater prosperity and global competitiveness, the Trump administration is implementing immigration policies that directly counteract those aims.

Specifically, the administration has proposed a levy of $100,000 on international students graduating from U.S. universities who apply for the Optional Practical Training (OPT) program and has reasserted its $100,000 fee on applicants for the H-1B visa (revising it upward to $103,265).

As Sam Peak and Jiaxin He explained in the wake of the OPT fee announcement, deterring graduates from U.S. universities will not only reduce their probability of entering the U.S. labor force, it will also entrench some of the perverse outcomes of the existing skilled-immigration system, biasing it in favor of less-skilled workers and the notorious “outsourcers” that hire them. The OPT program is a vital bridge that connects new grads to the labor market on a short-term basis, to the wider benefit of the American economy.

The H-1B program, though flawed, can best be improved through reforms that filter for the immigrants who create the most economic value. The most important changes pertain to the lottery system that allocates H-1B visas. Though the administration has made changes to the lottery, we contend that its “Wage Level” reforms are vastly inferior to an alternative approach, “Experience Benchmarking.” Experience Benchmarking, as Sam and Jiaxin wrote in a letter to the Department of Labor in May, “ensures all certified H-1B workers earn more than the median wage of comparable native-born workers with equivalent education and experience.”

While both the OPT program and the H-1B visa require reform to maximize their potential (see our other ideas here, here, and here), the administration’s punitive approach is a frontal attack on them. Curtailing, rather than improving, these skilled-immigration pathways undermines the immense productivity, wage-growth, business-formation, and technological benefits that America derives from its historic position as the world’s top destination for talented workers.

Research & Analysis

Understanding AI and Productivity

Chad Syverson, an economist who specializes in productivity and the George C Tiao Distinguished Service Professor of Economics at the University of Chicago Booth School of Business, contributes a guest essay to EIG’s American Worker Project looking at what is known to this point about the relationship between Artificial Intelligence and productivity growth.

An excerpt:

So, do the early data offer any hints as to which combination of these multiple considerations about AI’s employment effects may apply going forward? Early empirical studies of AI applications in production actually offer minimal evidence either way. At least over the first couple years of its use, AI has had remarkably little employment effect, positive or negative. While that may seem on its face to be a non-informative result, the very fact that there have been no obvious employment effects yet does say something about the combined effects to this point.

The New Bazaar

Top Five Unpopular (But Totally Correct) Beliefs

EIG Chief Economist Adam Ozimek joins host Cardiff Garcia to share his five favorite unpopular beliefs. Adam believes these ideas are true, even if others may find them contentious, perhaps even contemptible.

Adam tries to persuade Cardiff and New Bazaar listeners that…

• Prediction markets are going great

• Economists should ignore the efficient market hypothesis in their own lives

• The United States should let Chinese cars in

• One Billion Americans is absolutely a goal the country should pursue

• Consumers should spend less money on cars

Adam and Cardiff also consider when contrarian thinking is useful and when provoking disagreement becomes a brand rather than a way of getting closer to the truth.

Around the Horn

• In City Journal, Adam Ozimek argues alongside IFP’s Jeremy Neufeld that the Trump administration’s proposed $100,000 fee on Optional Practical Training applicants is a clumsy misstep that will squander immense economic value.

• In Reason, Billy Binion quotes work from Sam Peak and Jiaxin He on the same topic.

• In Forbes, Teresa Ghilarducci cites research from Ben Glasner and Sarah Eckhardt on worker retirement plans.

• In Bloomberg, Erin Schilling cites the EIG Opportunity Zones Coalition’s comments to the Treasury Department.

• In Forbes, Marie Sapirie quotes Kenan Fikri on OZ designations.

• In Realty Today, Sebastian Niles cites EIG data in its New York OZ coverage.

• Jiaxin He and Adam Ozimek draw citations from Stuart Anderson in Forbes and Billal Rahman in Newsweek for their paper critiquing the empirical basis of the Trump administration’s H-1B fee.

• And in a Financial Times letter, Jordan McGillis rebuts columnist Rana Foroohar’s claim that America’s economy only benefits the rich.

EIG Chart of the Month

Via EIG research director Nathan Goldschlag’s dashboard, the share of businesses using AI in any business function is just north of 20 percent:

At the microeconomic level, productivity is one of the best predictors of businesses’ survival and growth prospects, the wages their workers earn, and the prices their customers pay. At the macro level, to quote Paul Krugman, “Productivity isn’t everything, but in the long run, it is almost everything.” (And I am not sure the proviso is necessary.)

Sustained long-run growth in income per capita requires productivity growth. It essentially sets the speed limit on how fast material standards of living can grow. Faster productivity growth means faster per capita income growth. Slower productivity growth means bad news.

But what about its relationship with AI? Well, it is still early, but a balanced assessment of the data indicates we do not yet know for sure. A few case studies have offered convincing evidence that AI has raised productivity in specific business settings. At a broader, economy-wide level, the combination of AI’s novelty and the vagaries of measurement mean we cannot yet be certain if it is having an effect, or how big any such effect would be (though I discuss below what can be seen and what it might mean).

Regardless of what is in the data now, though, it is not hard to imagine AI and productivity could have a very important relationship in the future. The reason is simple: AI is plausibly the newest general purpose technology.1 A general purpose technology has many applications across all sorts of production settings. Think the internal combustion engine, electricity, or the computer.

Past general purpose technologies spurred productivity accelerations that facilitated faster growth in economic well-being. The questions at hand are whether AI will do so too, how large and long-lasting that acceleration will be, and how the resulting productivity growth would be distributed through the economy and its people (for instance, how will its impact be spread across sectors, capital versus labor owners, and the income distribution).

What the Data Say Now

Let us take a look at current (U.S.) economy-wide productivity. Like most economies in the world, the U.S. had been experiencing a productivity growth slowdown since the middle-to-late 2000s. Labor productivity (output per worker-hour) grew throughout the 2010s by about 1.5 percent per year, roughly half of its growth rate during the productivity boom of 1995–2004.2 Figure 1 shows labor productivity since 2013. As can be seen, the slow growth of the 2010s continued to the onset of the Covid pandemic. There was then a large, up-and-then-down blip in productivity during the height and eventual ebb of the pandemic. This blip was probably mostly a spurious measurement phenomenon due to composition and utilization effects rather than a true change in the economy’s productive capacity. After the blip unwound, however, productivity from mid-2022 on has maintained a faster-than-2010s trajectory involving annual growth of about 2.2 percent.

Could this acceleration be due to AI? Perhaps. The timing leans against AI being the sole initial cause. The acceleration began when AI investments were still quite small relative to the size of the economy. Additionally, there were well-documented increases in economic dynamism (labor market churn and business formation) during the pandemic emergence whose timing matches the acceleration’s start. On the other hand, total AI-related infrastructure investments grew very quickly and by 2025 were unmissable even in aggregate investment statistics. Business adoption of AI tools and AI-enabled processes have sped up as well. And of course even if other factors started the acceleration, it could well be that AI is starting to contribute now. Regardless of AI’s current effect, the longer the aggregate productivity acceleration continues, the more plausible it is that AI is an important driver.

As for the magnitude of any effect, a sustained increase from 1.5 to 2.2 percent annual productivity growth would be substantial (after a decade, GDP per capita would be 7 percent higher than otherwise), but hardly unprecedented. The 1995–2004 productivity boom saw annual productivity growth of nearly 3.0 percent per year, and other past general-purpose-technology-related productivity boosts saw labor productivity growth in excess of 2.5 percent for a decade or longer.

Another approach to examining whether AI is currently affecting productivity is to look across sectors. For many reasons, the intensity of AI use already differs substantially across sectors. The highest adoption is in the information, financial services, and professional services sectors. At the other end of the spectrum are mining, construction, and accommodation and food services. If productivity growth has risen more in high-AI-adoption sectors than in low-adoption ones, this would be a sign of AI pushing productivity gains already.

Figure 2 shows this comparison. It plots the post-2020 change in a sector’s contribution to economy-wide productivity growth (that is, the 2019–25 contribution minus the 2007–19 contribution) versus a measure of the sector’s AI adoption rate. A sector’s contribution to economy-wide productivity growth is the product of two numbers over the respective period: the sector’s average annual productivity growth and its average share of total worker-hours in the economy. Essentially, this metric size-weights the sector’s productivity growth, so between two sectors experiencing the same productivity growth rate, the larger sector increases economy-wide productivity growth by more. A contribution value of 0.1, for instance, implies that a sector’s productivity growth raises economy-wide productivity growth by 0.1 percent. To measure a sector’s AI adoption rate, I use the data from the Census BTOS survey administered in 2025.3 The measured adoption rate reflects the fraction of companies (weighted by firm employment) that have adopted AI in their business practices. If the relationship between these two values is positive, the sectors that adopted AI more intensively by 2025 saw their productivity growth rise more.

The results, shown in Figure 2, are mixed. The correlation between the two values is positive, so higher-adopting sectors saw a larger pickup in size-weighted productivity growth. What is less clear is whether one can reject the possibility that this relationship is only due to statistical chance. Taking in all the data, standard statistical methods say the answer is no: the positive relationship is weak enough, relative to the number of sectors, that one cannot rule out chance as its source.

That said, this result is substantially driven by the fact that the retail sector saw a large productivity acceleration across the two periods even though its AI adoption rate is rather low. (That is why its dot is up and to the left on the figure.) We are treading on thin statistical ice if we simply ignore this observation and concentrate on the remaining sectors. However, to live dangerously for a moment, if we were to focus on all sectors other than retail, the positive correlation between a sector’s productivity acceleration and its AI adoption intensity is more than twice as large as with retail included, and it is considerably less likely that this relationship would be due to mere chance alone.

Employment Effects

Any conversation about AI’s productivity effects inevitably leads to speculation about its employment effects, often with a particular focus on AI-induced unemployment.4

Is unemployment a likely outcome if AI substantially boosts productivity growth? Well, recall that productivity is the ratio of outputs to inputs. If productivity rises, fewer inputs would be needed than before to make the same output. If that is the entire story, then yes, productivity growth would imply less employment, because fewer workers would be needed to make the same output.

But that is clearly not the entire story, for several reasons.

For one, that scenario assumes output is held fixed. This is highly unlikely to be the case after a productivity growth surge. Because productivity growth means fewer inputs are required per unit of output, the costs of making goods and services fall. Lower costs in turn mean lower prices, and lower prices mean people want more of the good (that is, demand curves slope down, as an economist would say).5 That extra output demand requires more workers to produce, counteracting the employment effect from productivity gains requiring fewer workers per unit output. Does making this extra output need so many more workers that employment will actually rise when productivity goes up? Not necessarily; the net employment change depends on the relative size of the two opposing effects. However, it does mean that if productivity, say, doubles, employment will not fall by half. Additionally, there is a related effect if productivity growth also facilitates the creation of entirely new products (which, historically, it always has). Consumers will want these new products too, creating new demand for labor.

A second consideration is that products are made with multiple inputs. AI is one type of input and workers another (or another several types, as different types of workers can be thought of as different kinds of inputs). AI’s employment effects will depend on how workers are used alongside it in the production process. If workers, or certain types of workers, are substitutes for AI, then AI adoption will reduce demand for them. On the other hand, if they are complements to AI, AI adoption will increase demand for them.

Third, AI’s productivity effects are unlikely to act evenly across industries and sectors. Sectors with larger employment-reducing effects from productivity growth will shed more workers, but if demand for products made by sectors with slower productivity growth remains high (or perhaps is driven still higher as consumers shift spending share away from sectors with rapidly decreasing costs and prices), slower-productivity-growth sectors could absorb workers displaced from faster-productivity-growth sectors. This sort of cross-sectoral shift due to differential productivity growth has plenty of historical precedents. Advanced economies have seen enormous, broad-based shifts away from agriculture and toward manufacturing first, and later, away from manufacturing and toward services. There have been many other shifts operating at narrower sectoral levels.

Finally, any mass unemployment effects will depend on how fast productivity grows and how quickly the economy can reallocate resources in response to the productivity surge. Past GPT-driven productivity booms have raised annual productivity growth by 1 to 1.5 percent for a decade or two. The economy has been easily able to absorb any aggregate employment effects of these shifts. If AI were to raise productivity growth by 5 to 10 percent per year for a few years, on the other hand (a rate claimed by some boosters but that has not been observed before at an economy-wide level), that may outstrip the economy’s flexibility and absorptive abilities.

So, do the early data offer any hints as to which combination of these multiple considerations about AI’s employment effects may apply going forward? Early empirical studies of AI applications in production actually offer minimal evidence either way. At least over the first couple years of its use, AI has had remarkably little employment effect, positive or negative.6 While that may seem on its face to be a non-informative result, the very fact that there have been no obvious employment effects yet does say something about the combined effects to this point.

Productivity J-Curve

One phenomenon at play in measuring AI’s productivity effects is the productivity J-curve. The J-curve implies that when a new general purpose technology emerges, there will be a period — possibly of considerable length — during which resources are spent but with little incremental observed output. As a result, measured productivity is depressed relative to true productivity, and measured productivity growth may slow or even turn negative during the new technology’s initial diffusion. Later, however, the mismeasurement reverses itself and there is a period where measured productivity growth is overstated.

Here is how that works. The undermeasured-then-overmeasured productivity pattern arises because fully harnessing general purpose technologies requires a lot of intangible investments. Companies cannot simply plug and play AI and experience the new technology’s productivity benefits. To use AI effectively, companies must make many complementary investments like training workers, restructuring organizations, designing new business processes, and so on.7 These investments require real resources to build. Conceptually, they are capital investments — that is, outputs of the production process that will be themselves valuable in the future for making more output.

These intangible investments are rarely measured as such, however. Instead, standard accounting systems treat them as invisible in terms of outputs while still counting the resources spent to produce them as business expenditures. For example, paying a team of employees to redesign an organization for AI efficacy builds intangible capital. Production of capital goods should be counted as output, and if those employees instead built a new office building, standard accounting would count that building as investment output. But intangibles like organizational capital here are not tracked in most accounting systems. As a result, output is undercounted, implying productivity (output per unit of inputs) is undercounted too.

That output undermeasurement causes the initial dip in the J-curve. However, it is just one of two measurement effects resulting from the failure to track intangibles in productivity measurement. The second arises because once installed, those intangible investments become part of the intangible capital stock. Like all capital, they are inputs used to make output (as a more effective organizational structure would, in the example above). However, being intangible, they are not measured as such. Once these intangible inputs are put to use, they become productive inputs. But because they remain unmeasured, the input denominator in the productivity ratio is understated. This leads to measured productivity being overstated, which creates the hump after the dip in the J-curve. We have understatement of productivity growth early in the diffusion of a general purpose technology and overstatement later.

This phenomenon in the context of AI raises the obvious questions of how large and how long the periods of productivity growth under- and over-measurement will be. Based on work I did with coauthors Erik Brynjolfsson and Daniel Rock, we found other technologies’ J-curves could have understatement periods that extend well beyond a decade, and cumulative productivity mismeasurement of double-digits percent. As a result, it is worth remembering even persistently slow measured productivity growth in the face of broad adoption of AI is not an indication that AI isn’t having (unmeasured) productivity effects now and that measured productivity will accelerate later in response. Where are we on the AI J-curve now? It is too early to know that with much precision (and it might depend on how dangerously you want to live when you interpret the results in Figure 2).

What Next?

As even this brief discussion has conveyed, AI is a technology with enormous potential to shape our economic futures. Economic theory points us in directions to look to try to predict something about those futures, and empirical work has given us a very early — if fuzzy — look at them. If AI adoption continues at the rapid rate it has seen over the past couple years, we will quickly learn more about how large AI’s productivity effects are and how broadly they spread.

Back in 2015, when we at EIG published the original paper on Opportunity Zones, we knew that OZs had unique potential to boost economic growth in distressed parts of the country. OZs became law in 2017, catalyzing a wave of investment in communities all across America.

Newly-released data from the Treasury Department on the success of OZs proves just how valuable the policy’s incentives have been.

Here’s what we now know:

• The average state benefited from more than $2 billion in OZ investment through the end of 2024.

• There are more than 12,000 active OZ funds operating nationwide. These funds aggregate qualifying investment capital from 41,000 different taxpayers, 35,000 of which are individuals and 6,000 of which are corporations.

• In total, the funds hold $116 billion in total assets and $112 billion in tangible assets at work on the ground.

• The OZ tract designations we helped governors with in 2017 and 2018 translated into investment for 77 percent of all OZ census tracts across the 50 states and the District of Columbia.

• Rural tracts were no less likely to receive investment — a proportional 77 percent of rural tracts registered OZ investment.

With OZs having been made a permanent feature of the tax code last summer, Congress has now called for a new round of OZ census tracts to be designated.

Here is what we recommend to governors going forward:

1. Get moving now. While some states have actively communicated the latent value of OZs to potential investors, others need to get in gear as the 90-day designation window has now opened.

2. Set a statewide economic vision. OZs should be an integrated part of a growth strategy, not an afterthought.

3. Designate a lead coordinating entity within state government and hold it accountable.

4. Engage local partners strategically. State strategy only works when it’s tied to local needs and opportunities.

5. Balance economic need and investment potential. The most distressed areas may not necessarily reflect the best mix of characteristics to maximize outcomes.

6. Combine both quantitative and qualitative insights. There’s an art to OZ designation that goes along with the science.

7. Embrace purposeful transparency.

8. Align OZ nominations with supportive policy tools

OZ designation is one of the most powerful economic development tools at a governor’s disposal and the ball is now in governors’ courts to bring the immense benefits of targeted investment to the corners of their states that need it.

For EIG’s full analysis of the new Treasury Department OZ success report, click here.

And for EIG’s full OZ designation breakdown guidance, click here.

Policy

• Treasury's data update wasn't the only big OZ news this month. EIG's Opportunity Zones Coalition submitted its priorities to Treasury for the successful implementation of OZ 2.0. The Coalition also submitted a letter responding to Treasury's recently published transitional guidance for Opportunity Zones. In it, the coalition thanks the agency for providing helpful clarification on several matters, and raises concerns related to Treasury’s assertion that property acquired in zones set to expire in 2028 cannot qualify as Qualified Opportunity Zone Business Property after December 31, 2026, except under limited circumstances. The letter recommends that Treasury push the effective date for the requirements to the end of 2028 to allow existing OZs to use their full designation, among other recommendations.

Research & Analysis

Measuring the Economic Effects of AI
“How many firms are using Artificial Intelligence?,” writes Nathan Goldschlag. “What are they using it for? How many workers are using AI, and how are they using it? To track and understand the effects of AI on the economy, researchers will need accurate, detailed, comprehensive answers to these fundamental questions.” In a comprehensive new report, Nathan details what policymakers and the nation’s stats agencies should do to better answer those questions.

The Last Ten Per Cent
Estimating the exposure of a given task to being automated by Artificial Intelligence is the only the first step to understanding AI’s potential effects on the labor market. Josh Gans, an economics professor at the Rotman School of Management at the University of Toronto and chief economist of the Creative Destruction Lab, explains the remaining nuances in a new guest essay for The American Worker Project.

America’s most prosperous neighborhoods are barely 25
“Here’s a striking observation as the nation celebrates its 250th birthday: its most prosperous neighborhoods are barely 25 years old,” writes Kenan Fikri in a provocative new study. “While America’s well-off places come in many forms and have many birthdates, when a neighborhood was built turns out to be a pretty good predictor of the level of economic well-being enjoyed by its residents. Generally speaking, our best-off neighborhoods are those with the shortest histories.”

Exceptions to the American manufacturing jobs decline
In response to a reader’s request, Jiaxin He scrutinizes the data on manufacturing employment to find the subsectors that have been most resilient: electrical components (AI) and transportation equipment (aerospace). Both sectors have not only grown since the post-COVID peak but also have continued their growth into 2026.

Around the Horn

• In a Senate hearing on Capitol Hill, Mercatus Center fellow Liya Palagashvili invokes both an Agglomerations essay by Adam Ozimek and an EIG guest essay by Google economists Zanna Iscenko and Fabien Curto Millet on AI and employment.

• In a paper from Stanford’s Institute for Economic and Policy Research, former head of the Bureau of Labor Statistics Erika McEntarfer and co-authors apply work from Nathan Goldschlag and Sarah Eckhardt on separating AI hype from real productivity improvements.

• In the New York Times, Ben Casselman quotes Nathan on how to measure AI’s economic impacts.

• In City Journal, Sam Peak and Jiaxin He argue for tailored changes to the H-1B program to improve high-skill immigration outcomes.

• Also in City Journal, Jordan McGillis discusses Alaska’s frontier resource economy.

• In the Wall Street Journal, Anne Tergesen and Veronica Dagher cite research from Kenan Fikri on retirees starting businesses.

• And in Inc., Georgia Fearn quotes Kenan on the importance of contextualizing healthcare job growth across the country.

• In Newsweek, Giulia Carbonaro draws upon EIG data in a piece on the Trump administration’s new retirement plan.

• For Kiplinger, Daniel Goodwin promotes EIG’s Opportunity Zones Activity Map.

• In Forbes, Damini Sood cites EIG research on rural economic growth.

• In Canada’s Globe and Mail, Mike Moffat cites EIG work on the departure of middle-class taxpayers from big cities.

• And in the American Spectator, Jordan explains how the Declaration of Independence sparked 250 years of economic dynamism.

The New Bazaar

Sex, Drugs, and Death

Economist Alvin Roth, winner of the economics Nobel in 2012 for his work on market design, joins host Cardiff Garcia to discuss his new book, Moral Economics: From Prostitution to Organ Sales: What Controversial Transactions Reveal About How Markets Work.

The book is about morally contested markets: transactions that participants would like to engage in but that others think they shouldn’t be allowed to, even when it’s hard to identify measurable harms to those objectors. It’s also a great way to understand how markets interact with the law, politics, social norms, income inequality, human psychology, and so much more than just monetary exchange.

In this chat, Alvin and Cardiff discuss:

• Medical aid in dying and the economics of “repugnant” markets

• Sex work, trafficking, and the unintended consequences of criminalization

• Plural marriage and the challenge of designing new legal protections

• Sports gambling and the rise of prediction markets

• Kidney exchanges, the transplantation shortage, and the opposition to cross-border exchanges.

Throughout the conversation, they confront the trade-offs involved in both allowing controversial markets and trying to ban them.

EIG Chart of the Month

EIG research director Nathan Goldschlag oversees a number of frequently updated dashboards on his personal site, including one that features this chart showing that young adults, both college and non-college, are struggling:

But an EIG reader asked on X for a more-detailed breakdown of how specific subsectors within the manufacturing sector are doing, and because EIG is very much in the business of fan service, we are happy to oblige.

The short version of what we found is that nearly every manufacturing subsector has lost jobs since the post-COVID peak roughly three years ago.3 But we also discovered a few intriguing exceptions, and each of them tells a useful story about recent trends in the U.S. economy.

Start with Figure 1, which shows net job gains or losses in each subsector since the recent peak in 2023 through the end of 2025, the last available date for which data on subsectors is available.4 To get a sense of the most recent momentum for job growth in each subsector, the chart also shows how they did in 2025 alone.

Four manufacturing subsectors have enjoyed positive net job growth since the most recent peak for overall manufacturing employment. Two of them suffered reversals last year:

• Food, beverage, and tobacco product manufacturing added an impressive 34,800 jobs on net since the peak. But that headline figure obscures a reversal last year, possibly reflecting Chinese retaliatory tariffs against American food products and immigration enforcement pressures on industries that rely heavily on migrant labor.

• Petroleum and coal products manufacturing have eked out a tiny net job gain since the 2023 peak, then gave back ground in 2025, though the 2026 oil price spike may have since staved off that slide.

Jobs in two other subsectors — electrical components and transportation equipment — not only have grown since the peak but also have continued their growth throughout 2025. The rest of this piece will focus on them.

Figure 2 graphs these two subsectors against all other manufacturing subsectors. After jobs all across manufacturing bounced back following the pandemic, only these two sectors have enjoyed consistent and (for now) sustained growth since 2023.

As of the end of last year, electrical components manufacturing employed 9.1 percent more workers than in 2019, while transportation equipment employed 5.8 percent more.

Jobs in the rest of the American manufacturing sector, in contrast, were 2.9 percent below their pre-COVID employment level. The recovery and brief expansion of 2022 and 2023, when the sector climbed above pre-pandemic employment, proved short-lived.

What accounts for the resilience of these two subsectors? We can zoom in even closer to look at the specific industries within each subsector.5 We show the results in Figure 3.

Job growth in the transportation subsector is almost entirely driven by the boom in aerospace.6 In fact, aerospace leads all manufacturing industries in job growth since 2023. Last year it also posted the largest nominal trade surplus of any manufacturing industry, at $116.7 billion.

Driving the growth has been strong commercial and passenger airline demand, including a record-high order backlog. Boeing, the world’s leading airplane manufacturer, returned to profitability and high delivery volumes in 2025 after being plagued by safety scandals, worker strikes, and financial losses the prior year. Demand has also rippled downstream: Plane engine orders have surged, likely pushing engine makers such as GE Aerospace and Evendale to expand their workforces.

Aerospace’s continued growth reflects the United States maintaining its historical edge in high-tech, high value-added manufacturing.

The employment surge in the electrical components subsector tells a different story. It owes its strength to two industries: electrical equipment manufacturing; and batteries, wires, cables, and miscellaneous electrical components.7 8 These are among the industries that bore the brunt of the China Shock and suffered acute decline in the 2000s.9 Green energy and electric vehicles likely contributed to their initial recovery, raising demand for batteries and electrical infrastructure. But their continued growth even after the cancellation of EV subsidies in 2025 points to a different, potent, and entirely unsurprising driver: Artificial Intelligence.

AI’s manufacturing footprint runs the length of the supply chain. Dedicated data centers more than doubled their real consumption of electrical equipment and components from 2019 to 2024.10 Software firms building and deploying AI show an even larger rise, likely propelled by onsite infrastructure upgrades.11 Further downstream, data center buildup has lifted demand from engineering designers and construction contractors, while surging energy requirements have pushed governments to expand electricity generation and transmission.12

In Figure 4 below, we summarize the sources of demand (from other industries and sectors) for these two industries:

Rising AI-related demand has driven up prices, giving manufacturers a strong incentive to expand. The six industries and sectors in the figure above increased their real consumption of electrical components by 15.4 percent from 2019 to 2024, while nominal consumption rose 51.9 percent. That 36.5 percentage point gap far exceeds the roughly 20 percent broad inflation over the same period, pointing to price appreciation specific to these products.

Protectionists may interpret the new electrical manufacturing jobs as a tariff success, given the duties on copper wires and electrical grid equipment that remain in place. They would be mistaken. Imports of electrical equipment and components did not decline in 2025 relative to 2024, and metals tariffs have raised input costs for the very manufacturers supposedly being protected. Their growth owes to AI-driven demand, not trade policy.

The AI supply chain’s impact on manufacturing extends beyond electrical components. Architectural metals and cooling equipment rank among the fastest-growing manufacturing industries since 2023 and sustained gains through 2025, even as their broader subsectors contracted.13 14 15 The top five industries and sectors expanding real purchases of these products are all related to information technology, led by software companies and data centers, which collectively raised their demand for pipes and cooling systems by 8.9 percent in real terms from 2019 to 2024.

AI-led job creation in manufacturing complicates the popular narrative that data centers don’t create many jobs. A Corning-NVIDIA partnership is opening three fiber optic cable factories. Electrical and cooling equipment-makers are benefitting from Virginia’s data center buildup. GE Vernova received more electrical equipment orders from data centers in the first quarter of 2026 than in all of 2025. The spotlight falls on AI labs, chip designers, and computing hubs, but the AI ecosystem reaches the factory floor. Demand for energy infrastructure and climate control apparatuses represents an opportunity for significant industrial growth.

The post-COVID manufacturing boom is fading, and current policy is not arresting the decline. The rapid erosion of job gains in chips and autos is particularly concerning. Tariffs have proven insufficient to reverse the broad-based retreat in manufacturing employment, and the remaining steel and aluminum tariffs continue to impose a cost burden on manufacturers. Even in the industries that are expanding — aerospace and electrical components — the binding constraint is supply struggling to keep up with demand, whether measured by airplane order backlogs, defense production delays, or surging components prices.

A durable industrial renaissance will require cutting input and energy costs and lowering barriers to capital formation, whether the goal is sustaining leadership in aerospace, building out the AI-related manufacturing ecosystem, or stemming job losses elsewhere.

See our GitHub with replication code here.

In May 2026, Dario Amodei, the chief executive of Anthropic, offered a compact way to think about artificial intelligence and work: “If you automate 90 percent of the job, then everyone does the 10 percent of the job.”1

Consider a lawyer who no longer needs to spend hours reviewing routine documents, leaving more time to understand a client’s objectives, test a strategy, and exercise judgment. The service improves and, because the job is done more efficiently, its price falls and demand for it might even grow. Amodei’s point was that the remaining 10 percent of work could well expand to fill the day while workers become far more productive.2 The remark drew attention partly because it sounded more optimistic than Amodei’s earlier warnings about white-collar job loss.3

Of course, something else might happen. Lawyers who have 90 percent of their jobs automated might simply end up with less legal work to do, and thus fewer of them needed to do it.

Amodei’s comment raises real economic questions: Under what conditions does the last ten per cent become a better and more productive job? When does the job merely become smaller and performed by fewer people?

Task-level analysis remains the right place to start for understanding AI exposure, or which of the tasks that make up a job might be automated. Trouble begins when a job itself is treated as a checklist and the exposed tasks are crossed out one by one, appearing to simply shrink the amount of work needed to do the job. But once some tasks disappear, the rest of the job may evolve and change character entirely, with the remaining tasks expanding or new tasks emerging to replace the old.

Tasks differ widely. They may be substitutes, complements, bottlenecks, filters for expertise, objects of worker motivation, or parts of an organisational bundle that firms redraw after automation. How a job changes after some of its tasks are automated and new ones emerge depends largely on the nature of those tasks and, especially, how the tasks relate to each other.

Tasks are a useful starting point

The task-based approach used by economists Daron Acemoglu and Pascual Restrepo in their 2019 paper starts with a useful observation: Technology rarely replaces a job title in a single stroke. Instead, it changes who, or what, performs particular activities within production.4 Their framework separates a displacement effect (capital takes over tasks previously assigned to labour) from a reinstatement effect (innovation generates new tasks in which people have a comparative advantage). Productivity gains may also increase demand for the resulting goods and services.

This approach is more informative than asking whether an entire occupation is “safe” or “at risk.” An accountant, software engineer, teacher, or nurse performs a bundle of activities. A technology may be well-suited to some activities and poorly-suited to others. Studies of large language models have measured occupational exposure by assessing the tasks for which these systems could reduce completion time.5 The resulting measures give us a useful map of technical applicability.

But the map still leaves a difficult step. Once we know where a technology might apply, we need to ask what happens to the human work that remains and its effects on wages and employment. The answer will reflect displacement, reinstatement, and demand. It will also depend on the job’s internal structure.

What happens to expertise?

Economists David Autor and Neil Thompson focus on the type of task that disappears.6 Was it among the more expert parts of the occupation, or among the less expert parts? (Autor and Thompson define expert tasks as “problem-solving and decision-making tasks requiring specialized knowledge and training.”) The same activity can be an expert task in one setting and a routine task in another. Once that activity is automated, a worker who remains in the occupation no longer needs the expertise associated with it.

Suppose software removes routine bookkeeping from an accounting clerk’s day. The remaining activities may involve exceptions, problem-solving, and communication with clients. The occupation has become more demanding. Workers with the relevant expertise may become more valuable, while fewer people qualify for the role. Wages can rise as employment falls.

In another occupation, technology might remove the most demanding diagnostic or administrative activity. The remaining role becomes accessible to a broader pool of workers. Employment may rise as wages fall because expertise is less scarce.

For practical analysis, the warning is clear. Wages and employment can move in opposite directions. A task labelled “routine” in one setting may play a different role elsewhere. Autor and Thompson show how automation can replace experts in one context and complement expertise in another. Their framework asks how automation changes the expertise threshold for the remaining tasks.

When the remaining work matters more

Another mechanism appears when the quality of one task affects the value of the others. Michael Kremer’s O-ring model captured this idea: Production can depend on several essential activities, and a weak link reduces the value of the whole effort.7 In my joint work with Avi Goldfarb, we apply that reasoning to automation when a worker has a fixed amount of time to allocate across essential tasks.8 A manager who spends less time scheduling can devote more attention to a difficult negotiation. A radiologist relieved of routine triage can spend more time on integrative diagnosis. A lawyer freed from document review can devote greater care to strategy.

Here, automation has a second effect. A machine takes over one activity, freeing up time for the remaining activities. If the quality of human work depends on attention, the quality of those residual tasks rises. We call this the focus mechanism. At the level of an occupation, it can look like augmentation even though it begins with the automation of a particular task.

This is one way to read Amodei’s final ten per cent. The tasks left to people may be the bottlenecks that determine service quality. As auxiliary tasks disappear, those bottlenecks can become more valuable.

Demand matters too. Lower costs may attract more customers, while a better service may draw in people who previously went without it. The history of bank tellers after the spread of automatic teller machines is a useful recollection that work reorganisation and demand expansion often arrive together.9

Three cautions follow.

First, average exposure can obscure the job’s architecture. Two occupations with the same average exposure may look quite different once we see whether the exposure is concentrated in a few tasks or spread across the whole bundle.

Second, adoption can be lumpy. Automating one task changes the return to automating another, so firms may adopt a package of tools rather than move through a tidy sequence.

Third, better automation can increase the value of retained labour, even as a human bottleneck remains necessary. If full automation becomes credible, the wage effect is less predictable; the worker’s contribution may become more valuable even as the firm’s outside option improves.

Jobs can be redrawn

Occupations themselves are moving targets. When AI removes part of a job, firms can redesign the human work that survives. Related work on endogenous task bundling studies this organisational margin directly.10 Consider a radiologist whose pre-AI role combines image classification with physician consultation, while quality assurance and model governance are handled by a separate technical role. If AI begins to read images, the old radiologist’s role changes. The hospital has to decide how to recombine the surviving activities.

The hospital might create a broad clinical role that bundles consultation with model governance. Or it might split the work, with a senior clinician focusing on consultation and a governance specialist monitoring the model. The broad role rewards breadth. The specialist arrangement rewards comparative advantage. A skill that carried little weight in the original role can become central after automation, while a skill that anchored the old wage may disappear from the schedule.

For measurement, this matters greatly. A study that holds occupational boundaries fixed can estimate the effect of automation within the old bundle and still miss much of the effect on workers. Where occupations are split into specialist roles, a fixed-bundle estimate may understate the return to comparative advantage. Where tasks are recombined into broader roles, the bias can run in the other direction. Changes in occupational classification belong in the analysis because they record part of the economic response.

What payroll records leave out

Jobs can change even when wage records barely move. Consider a software engineer who uses AI to reduce time spent on documentation, support tickets, and routine debugging, and then devotes additional time to refining a tool they find intrinsically rewarding. Payroll may show little change. The job itself has changed substantially.

Work on enjoyable tasks and contracting shows why this can matter.11 Workers sometimes expand their effort on tasks they value. Two jobs can then have the same hourly wage, paid hours, and wage bill while differing in total time worked and task mix. In some settings, AI reduces recorded hours. In others, it intensifies work, extends the working day or moves effort towards activities beyond formal requirements. Empirical work on AI exposure and the length of the working day points to the importance of this intensive margin.12

Enjoyment is one reason for extra effort. Promotion incentives, deadlines, professional norms, and managerial pressure can produce it as well. The practical point is that productivity gains do not automatically translate into shorter hours or better jobs. Payroll records cannot reveal the full quality of the work experience.

Turning exposure into a forecast

Exposure measures still have a clear use: They tell us where to look. Turning exposure into a forecast requires a second layer of analysis. For any occupation, decision-makers should ask at least six questions.

1. Which tasks are exposed? Are they peripheral activities, expert activities, or essential bottlenecks?

2. What expertise remains? Does automation raise or lower the threshold for entering the occupation?

3. Does freed attention improve quality? Can workers use the time freed up by AI to perform residual tasks more effectively?

4. Will firms redraw the job? Are surviving tasks likely to be rebundled into broad roles or split among specialists?

5. How elastic is demand? Will lower costs and higher quality expand the market enough to support additional employment?

6. What do the available data miss? Are actual hours, work intensity, worker well-being, and changing job boundaries visible?

The answers are likely to vary across occupations. In professional services, medicine, and management, judgment can remain a bottleneck, leaving room for focus effects. In clerical work, displacement may be more direct, although much still depends on whether AI removes expert or inexpert tasks. In software and finance, job boundaries may move quickly as firms separate oversight, client interaction, and technical execution. In delegated creative and professional work, payroll may remain stable even as the texture and intensity of work change.

For researchers, this points towards a fairly concrete data agenda. Occupation-level exposure scores should be paired with task-level time use before and after adoption, quality measures for the remaining tasks, evidence of lumpy implementation, vacancy data showing new specialist roles, and worker-reported measures of intensity and well-being.

No single model will fit every setting. The task is to work out which mechanism matters in each case.

Conclusion

What remains after automation can change in several ways. The final ten per cent may become more valuable, more demanding, easier to enter, harder to perform, or newly divided among specialists. AI may displace tasks and reinstate labour elsewhere. It may remove routine work and increase the value of expertise. It may free scarce attention for human bottlenecks, prompt firms to redraw occupational boundaries, or change the intensity of work in ways that payroll does not capture.

For policymakers and business leaders, a useful discipline is to keep asking questions after an exposure estimate is released. What remains? How do the remaining tasks fit together? Who can perform them? How will firms organise the work? Will demand expand? An exposure measure is a starting point for these questions, not the answer.

While America’s well-off places come in many forms and have many birthdates, when a neighborhood was built turns out to be a pretty good predictor of the level of economic well-being enjoyed by its residents. Generally speaking, our best-off neighborhoods are those with the shortest histories.

Youthful exuberance

Using measures of educational attainment, workforce participation, poverty, incomes, housing vacancies, and employment and business trends, EIG’s Distressed Communities Index (DCI) allows us to compare economic well-being across nearly every zip code in the United States.1 In the DCI’s methodology, a neighborhood’s distress score is equivalent to a percentile based on its average rank across the seven metrics. Each quintile is then assigned a label from “prosperous,” for the top 20 percent, to “distressed,” for the bottom 20 percent.

Plotting distress scores against community ages reveals a distinct relationship between neighborhood newness and prosperity. As a rule, economic well-being runs higher in newer places and lower in older ones.

To approximate a neighborhood’s age, I used the American Community Survey to obtain the year its median residence was built — meaning an equal number of residences in the area are older and an equal number are newer. The line in the graph below shows the average distress scores for zip codes according to those median years, grouped by decade.2

The average zip code built out in the 2010s posts a distress score of only 16.7, firmly in the prosperous quintile. It stands more than 35 points ahead of the average zip code built out in the 1980s, and it enjoys a nearly 45-point lead over the oldest neighborhoods.3

Of course, new places aren’t the only ones that are well-off. There are plenty of prosperous zip codes that were built out in earlier decades. The median home was constructed in the 1970s or 1980s in 51 percent of all zip codes nationwide and in 42 percent of zip codes that score as prosperous on the DCI. Zip codes that combine both old and new housing stock can also register mid-century medians.

But the relationship between the age of homes in a neighborhood and the economic well-being of its residents is strong. The scatterplot below shows the year in which the median home was built for the average zip code in each percentile on the DCI.4 Newer places are more likely to be thriving communities in which the population enjoys high levels of well-being.

Heritage assets

The age of the housing stock offers a proxy for where the country channels its growth and investment. True, Agglomerations readers may be more likely than most to live in older neighborhoods that combine high-levels of well-being with a historical housing stock. But such neighborhoods are the exception. Nationally, for every one prosperous “historic” neighborhood — which I’ll define as a zip code in which the median residence was built in the 1940s or earlier — there are three categorized as distressed.

Stately Massachusetts manages to have more prosperous historic zip codes than distressed ones, although there are more housing units in distressed and historic parts of Fall River, New Bedford, and Springfield than there are in heritage corners of tony Brookline, Marblehead, or Newton. In California, the ratio between prosperous and distressed historic zip codes is balanced, with every Berkeley matched by a Stockton, and every Presidio by a Skid Row.

The picture is very different across much of the rest of the country.

Consider the city of Chicago. Of its historic zip codes, one prosperous spot lies at the heart of downtown and three “comfortable” zips (second quintile on the DCI) line Milwaukee Avenue as it extends northwest out of the city. Covering Wicker Park, Logan Square, and adjacent neighborhoods, these zip codes encompass some of the city’s trendiest enclaves. Yet they are far outnumbered by the 16 deeply distressed zip codes arrayed across the city’s South and West Sides with housing of similar vintages.

Or consider Ohio, where more than half of the state’s nearly one hundred historic zip codes are outright distressed. Only eight are prosperous, and only two of those are urban, covering the Hyde Park and Oakley neighborhoods in Cincinnati. Pennsylvania has 14 distressed historic zip codes for every one prosperous one, and Missouri 18. Florida’s only heritage zip code, just north of downtown Jacksonville, is distressed.

The determining factor, of course, is investment. Some historic neighborhoods have retained their value and received ample investment over the years in upgrades and rehabilitations. But this feat does not come easily for the typical neighborhood as its housing stock fades into obsolescence.

Building the American dream

The other way investment flows into a community is new builds. By definition, an old housing stock is synonymous with a lack of new housing construction. Where new housing gets built follows a distress-prosperity gradient, as people seek out amenities, opportunities, and high levels of well-being. But it also reinforces the gradient, by channeling new investment, new residents, and new incomes into already-growing and well-off places.

Consider two emblematic cities that have approximately the same population (~275k people) but couldn’t be more different: Buffalo, New York, and Gilbert, Arizona. The former is one of the most distressed cities on the DCI and has experienced long-term population decline, while the latter is one of the best-off boomtowns. Only 4 percent of all housing units in Buffalo have been built since 2000, and the median home in 15 of its 17 zip codes was constructed before 1940. In Gilbert, by contrast, 56 percent of all homes have been constructed since 2000.

In some ways, the comparison is unfair. Gilbert is part of the country almost completely unencumbered by history, while Buffalo suffers from the weight of it — saddled with legacies of segregation and deindustrialization. But it also points to a key fact: Unencumbrance makes it easier for new Sun Belt cities to thrive, just as the burdens of the past — which include zoning codes and other administrative accretion — make it harder for their Rust Belt cousins to renew. Housing is not the root cause of Buffalo’s travails, but speak with Rust Belt civic leaders, and a lack of the type of housing people want usually comes up quickly as a key barrier to their regeneration. A quality housing/quality workforce chicken-and-egg problem is one of several obstacles standing between them and renewal.

Economically distressed neighborhoods have less housing overall, and the housing they do have skews older.5 The below graph shows the count of all housing units constructed in each decade across all prosperous and distressed zip codes in the country (about 5,000 zip codes each). Very few housing units have been constructed in distressed communities in the past 15 years, and one-fifth of distressed community housing stock predates 1940. Top quintile communities are skewed in the other direction, with a preponderance of housing units built in more recent decades, and 36 percent since the turn of the millennium.

Where we build thus reinforces geographic inequality. Forty-one percent of all housing units built since 2020 have risen in prosperous communities. Only 8 percent have risen in distressed ones. These figures likely would have been even more skewed away from distressed communities absent the federal Opportunity Zones tax incentive.

The phenomenon does not only operate between cities or metro areas, but within them, too. Across the greater Dallas-Forth Worth metro area, 1.16 million new housing units have been constructed since the turn of the millennium. Of those, 677,000 have risen in prosperous zip codes, compared to just 72,500 in distressed zip codes. Suburban Dallas is home to five of the 10 fastest growing cities in the country, all of which are considered prosperous on the DCI.

Even in the country’s most vibrant growth pole, neighborhood age and well-being are closely linked. The median residence was built in 1974 in the typical distressed zip code in Dallas-Fort Worth, compared to 2001 in the typical prosperous one. Growth has occurred in distressed zip codes — the average one has welcomed nearly 2,100 new housing units since 2000 — but at about a quarter the magnitude of well-off parts of the metro area.

Rehabbing Pleasantville

Housing growth is part of the recipe that sustains high levels of well-being in a community.

Americans have always had a predilection towards the new, and new housing is no exception. New builds command a premium in the market. But when that preference for newness grafts onto a map of already-steep geographic inequality, and widespread variation in where new building is even possible, it becomes a dynamic that reinforces social and economic divides. Where we build impacts where we grow. Where we build affects where opportunity rises. The fact that building is so hard in many corners of the country is part of what contributes to the lack of economic opportunity in them.

Consider the country’s quintessential mid-century city, Los Angeles. Only five zip codes have had at least one-third of their housing stock constructed since 2000. Three cover downtown, representing the sorts of densely urban distressed zip codes that have absorbed growth in recent years. The fourth is Koreatown, an at-risk zip code (fourth quintile) that has densified its commercial corridor. And finally there’s Playa Vista, a master-planned district that arose on a massive old industrial site and was reborn as one of the most prosperous mixed-use corners of the city.

Such second leases on life are rare, even at much smaller scales than Playa Vista. But they do exist: in Highland Park, Denver, where single family neighborhoods abut riverine mixed-use districts; in Over the Rhine, Cincinnati, where reinvestment has revived one of the nation’s finest concentrations of Italianate architecture; in South Bend, Indiana, where new builds along the St. Joseph River are mirrored by new student housing adjacent to the University of Notre Dame; and in Washington, DC, where more housing units have risen along the Southwest Waterfront in the years since 2000 than were constructed in the 1960s.

But for every Georgetown or Beacon Hill, in which prosperity has endured across generations, the fact remains that the highest levels of well-being offered by the U.S. economy are much more typically found in wholly new places, like Playa Vista, Gilbert, or Dallas’s northern exurbs. That our best-off communities tend to have the shortest histories is in some ways a triumph of the suburbs, a uniquely American social innovation. But it also reflects a failure of more mature communities to retain the capacity to grow. Millions of Americans are mired in low-investment, low-opportunity neighborhoods, and many aging neighborhoods find themselves battling deteriorating social and economic conditions, because they are structurally and administratively unable to renew.

The tight connection between place and opportunity that pervades our society at the nation’s semiquincentennial surely would have shocked the nation’s founders. But patterns of growth and investment are not inevitable. Public policy choices, not just private economic calculations, help explain why the U.S. economy is much better at building prosperous communities from scratch than it is at sustaining prosperity in legacy neighborhoods. As the nation turns 250 years old, there’s much it can learn from its youngest places — and the older ones that have managed to rejuvenate.

***

The Distressed Communities Index (DCI) is EIG’s flagship data tool for exploring geographic inequality in the United States. To learn more about the DCI, explore the interactive map, or download the data for your own use, visit: eig.org/dci.

Artificial Intelligence appears destined to reshape the labor market and how firms produce goods and services. Workers, firms, and policymakers will need timely information about how the economy is changing, answering questions such as:

• How many firms use AI? How many workers do?

• How do firms that use AI differ from those that do not?

• What industries are expanding and what industries are contracting?

• What skills and occupations are in high demand, and which are increasingly less valuable?

• What types of workers are switching jobs and which are losing jobs and stuck in unemployment or falling out of the labor force?

• What types of new businesses are being created and how fast are they growing?

High-quality data enables displaced workers to find a new career, students to choose the right skills for the new labor market they will face, and firms to choose which investments to make and which operations to expand or shrink.

The good news is that there is already widespread agreement that something needs to be done.1 Not only that, but much of the statistical infrastructure needed to measure AI’s impact on the economy — surveys that can be expanded, administrative data to combine, human capital within the federal statistical system — already exists.

Policymakers simply need to make some targeted investments and scale what is already working.

Table 1 below summarizes my recommendations. For the full discussion of which measures of AI’s impact are currently available; a summary of findings thus far; and a detailed description of what can be done to improve measurement of AI, see the full report here.

To generate the economic data that policymakers and the public rely on, the government’s statistical agencies use various methods to navigate a delicate balance between precision and the protection of privacy. As EIG research director Nathan Goldschlag explains in a new Agglomerations post, the Commerce Department has now banned the use of one such method, making the production of quality data more difficult.

On June 4, Commerce banned a practice called “noise infusion,” a vital tool that statistical agencies use to provide granularity without violating the privacy of the businesses and individuals from whom it collects information. Noise infusion, Goldschlag explains, is the practice of slightly altering data in order to preserve both the general accuracy of the data and the identities and proprietary information of private-sector actors.

In the example he offers, noise infusion can be used to “fuzz” a business’s employment numbers to preclude their rivals from gaining competitive intel. He presents the hypothetical of a county with a small number of breweries. If the owner of one brewery is able to infer the number of employees its rivals have by a process of elimination, that owner might be able to operationalize an advantage. Noise infusion changes the specifics ever-so-slightly in order to give a useful picture for researchers and policymakers without disclosing too much on any given business.

The motivation behind the ban is likely that Commerce wants to improve the precision of the data that statistical agencies generate. Removing noise infusion, however, will have a different effect. Because the agencies are legally bound to confidentiality, they will have to resort to broader aggregations, such as at the state level, rather than at the county or municipal level. Businesses themselves may also end up withholding a lot more information in their survey responses on the front end. Perhaps worst of all, the statistical agencies will be forced to stop publishing certain data series altogether.

Noise infusion is a vital way to balance between granularity and privacy. Commerce’s blunt ban — “Noise infusion shall not be used for any statistical product” — undermines the ability of the agencies to provide researchers and policymakers with the best data they can. Policymakers should take note and read Nathan’s post.

Policy

John Lettieri, President and CEO of the Economic Innovation Group, was quoted in the New York Times responding to the passage of the 21st Century ROAD to Housing Act, arguing that the bill is a welcome step towards addressing the nation’s entrenched housing shortage — but far from the final step:

“I think the American people are running low on patience broadly on this issue, and that has been true for years...”

Mr. Lettieri said the bill was not “the magic fix-it-all button” on housing “that people would love to see pushed here.” But he said the measure was still a rare and significant achievement that would “show demonstrable progress.”

EIG also issued a statement praising the bill, encouraging Congress to continue playing a central role in helping communities build at the scale the moment demands:

EIG remains committed to advancing research and policy ideas, such as Right to Build Zones, that make housing abundance a national priority and expand access to opportunity. The 21st Century ROAD Act is an important first step, but there is still work to be done.

Tina Lee, Jess Remington, and Adam Ozimek recapped a convening about Right to Build Zones hosted by EIG in March of this year. “The goal of the convening,” they write, “was to strengthen the policy’s structural design with input from leading experts, practitioners and industry.” The guests included several members of EIG’s housing advisory board: Scott Alter, Alex Armlovich, Bobby Fijan, Arpit Gupta, Paul Williams, Emily Hamilton, Michael Novogradac, and Mike Kingsella.

Around the Horn

• John Lettieri was also quoted by Bruce Golding in The Independent about the 21st Century ROAD to Housing Act.

• The Economic Innovation Group is proud to be a partner of this year’s Abundance Conference.

• In Newsweek, Billal Rahman, highlights a new report that examines earnings linked to high-skill immigration using EIG data.

• In The Hill, AEI’s Brent Orell cites work from EIG research director Nathan Goldschlag and former research associate Sarah Eckhardt on AI’s labor-market impact.

Research & Analysis

AI versus the China Shock
Adam Ozimek, Jason Harrison, and Nathan Goldschlag ask: Will AI do to laptop workers what the China Shock did to so many factory workers? “Hoping to avoid the same mistakes of the past,” they write, “researchers and policymakers will be tempted to say yes. But after a closer look at the details of the China Shock, combined with our understanding of workers exposed to AI, we arrive at a different conclusion. The real lessons of the China Shock emerge not from its similarities with the AI Shock, but from its differences. And these lessons should mitigate concerns about a massive upcoming disruption, not exacerbate them.”

AI Shocked workers are likely to be better educated, have higher incomes, and live in more places with high levels of human capital — all of which contribute to far greater resilience than their China Shocked predecessors.

The New Bazaar

• AI Shock vs China Shock: Adam Ozimek and host Cardiff Garcia chat about Adam’s new analysis (see above) contrasting the characteristics of China Shocked with those of workers likely to be affected by the AI Shock. They also discuss how to think about automation, the collection of tasks that make up a job, the resilience of “messy” jobs, and much more.

• A City from Scratch: California Forever has a bold and ambitious plan: to erect an entire new city in California, from scratch, about a one-hour drive north of San Francisco. It has bought about 70,000 acres of land in Solano County, and it was the city to be two things: 1) an urbanist, walkable city of about 400,000-500,000 people, zoned for density instead of suburban sprawl, and located in a place where you would not expect to find it; and 2) a tech manufacturing center, with new factories making actual physical stuff, and a big shipbuilding operation. Cardiff is rooting for it to succeed, but he also had questions. Jan Sramek, Founder and CEO of California Forever, was exactly the guest to answer them.

• Debt and Deficits: Time to Worry?: Martha Gimbel, executive director of The Budget Lab at Yale, returns to the New Bazaar to chat with Cardiff about all things debt and deficits. Martha argues that the fiscal choices of the past decade have already led to higher costs on mortgages, car loans, and small-business debt. She and Cardiff discuss what might happen next given the frightening projections for future borrowing.

EIG Videos of the Month

Ben Glasner comprehensively summarizes the new retirement policy from the Trump administration to expand access to retirement accounts:

Also, did you know that the New Bazaar now also does video episodes? Here is Cardiff Garcia’s chat with Martha Gimbel, available on EIG’s YouTube page, where you’ll find all full episodes of The New Bazaar in addition to shorter clips:

The policy bans the use of something called noise infusion, a necessary tool used by statistical agencies, in particular the Census Bureau, to publish granular data about individuals and businesses without disclosing their identities, thereby protecting their privacy.

The likely reason for the policy change is that the Trump administration wants to improve the precision of certain data series. But the likely result of the policy will be that the government publishes less data, weakening the public’s ability to understand what is happening in their local communities and the American economy.

This consequence would be particularly damaging because individuals, businesses, and policymakers rely on these data to make decisions. Workers use these data to decide whether or where to move, which skills to acquire, and which jobs to seek. Businesses use these data to decide whether and what type of investments to make. And policymakers use these data to design policies and evaluate their effects on constituents. Without the data, they are all flying blind.

I agree with the goal of making data more precise. This just isn’t the way to do it.

Disclosure Avoidance Primer

Disclosure avoidance refers to the set of methods used by the federal stats agencies to protect the privacy of individuals and businesses that contribute to statistical estimates. The agencies are required by law to protect the data of respondents to their surveys.1 In practice, this means they must ensure that it is impossible to back out the information of a specific individual or business from any statistical product.

An example is the County Business Patterns (CBP) data, which provides statistics on business activity by industry and geography. Consider the following three scenarios:

• There is only one brewery in a small county. If the CBP published the exact count of brewery employees in that county, it would be disclosing the information of one business (how many workers it employs), a clear violation of the law.2

• There are two breweries in a small county, and the CBP again publishes the exact count of brewery employees. If I own one of those breweries, I could learn how many employees my competitor has, again violating the law.

• There are more than two breweries in a small county, but the CBP chooses not to publish the total number of brewery employees out of concern that it might compromise the privacy of the businesses. If I’m a prospective brewery owner, I may deem the project too risky to pursue without information about the market I’m entering.

These hypotheticals illustrate a core tradeoff of disclosure avoidance: Granular data is often more useful than aggregate data, but it also risks disclosing private information.

The stats agencies have several tools to manage this tradeoff, including coarse aggregations (releasing less data), cell suppression (removing certain statistics from the data that is released), and noise infusion (fuzzing the data).

It is the third option, noise infusion, that is being banned by the Commerce Department. When I say that noise infusion means “fuzzing the data,” I’m not using a technical term. What I mean is that noise infusion is a method of slightly altering granular data in such a way that the privacy and identity of individual people and businesses are protected in the statistics derived from them. The data are not so much altered, however, that they become unhelpful.

Another example helps to explain how it works. The CBP uses a type of noise infusion known as multiplicative noise, in which the employment of each business in a county is multiplied by a random number, increasing or decreasing it by a small amount. Go back to our hypothetical county, and this time there are 20 total breweries. Noise infusion allows the CBP to publish the total number of brewery workers in that county because it is not based on each brewery’s exact employment. But the data on total brewery employment in the county will be close enough to its actual value for it to be useful to anyone who needs this information to make decisions.

For any statistic that covers a large number of people or businesses, the noise cancels out and the statistic ends up very close to the actual value. For a specific industry in a small county with only five or ten businesses, the statistical and actual value may differ. But the differences tend to be small, and even in those cases it is nonetheless better to have some data rather than none. In the absence of noise infusion, for example, the stats agencies might be forced to only give the count of brewery employees at the state level. This information is far less useful to local policymakers and businesses — or to a prospective employer looking to build a new brewery in a specific county with unique business conditions that may differ from other parts of the state.

The practice of noise infusion became contentious around the 2020 decennial census because of the Census Bureau’s introduction of a new disclosure method called differential privacy, which is another type of noise infusion.

Differential privacy is similar to multiplicative noise infusion in that it adds noise to protect the privacy of people and businesses. But differential privacy is fundamentally different in that it gives the Census Bureau a better understanding (through complicated math) of how changes in the amount of noise added will affect the risk of disclosing too much private information.

The worry on the part of data users was that the Census Bureau would use — or misuse, in their view — differential privacy to err too far in the direction of protecting privacy, thus compromising the precision of the data.

Importantly, as I’ve already explained, differential privacy is not the only type of noise infusion used by the stats agencies. The County Business Patterns, Business Dynamics Statistics, and Quarterly Workforce Indicators have been using the simpler multiplicative noise infusion for decades. But the new policy from the Commerce Department would ban multiplicative noise infusion in addition to differential privacy.

Why I’m Worried

Noise infusion is a vital tool in an agency’s toolkit to release more data than it otherwise would. An excerpt from Abowd, Stephens, and Vilhuber (2005) makes this point nicely for the Quarterly Workforce Indicators (QWI) data:

“Because of the fine detail offered by the published statistics and the confidential nature of the micro-data used to compile the statistics, confidentiality protection is a critical and integral part of the design of the QWI system. Only the application of state-of-the-art protection methods allows the Census Bureau to publish these statistics.”

Noise infusion makes possible the granular tabulations of QWI data, which are used to inform the public about employment, hires, separations, and earnings by industry, geography, and worker characteristics including age, sex, race, and education.

If the Commerce Department policy stands and noise infusion is banned, which disclosure avoidance tools could the agencies use?

The policy states: “Coarsening shall be the preferred category of Disclosure Avoidance methods for all statistical products.”

And: “Suppression shall be permitted as a last resort.”

As a reminder, coarsening means releasing less granular data, and suppression means no longer publishing certain statistics altogether.

And that is why you should be worried. Either the agencies decide it’s actually okay to publish data for the county with only a handful of breweries, or, more likely, they decide that there will simply be no county-level brewery data at all. The data might get rolled up to the state level, which is far less detailed and helpful.

Now what?

Remember again the fundamental tradeoff.

On the one hand, the statistical agencies must adhere to strict legal requirements that ensure the privacy of individuals and businesses. Those privacy guarantees, importantly, encourage survey responses by giving respondents confidence that their data won’t be made public.

On the other hand, the statistical agencies are charged with producing quality data about the nation’s people and economy. Excessive focus on disclosure risks can degrade the quality of data, reducing its value to the individuals, businesses, and policymakers that rely on them.

Outside of clearcut cases like our small county with just one or two breweries, the statistical agencies exercise significant discretion in terms of how strict the disclosure-avoidance protocols should be. In the brewery example, how many breweries does a county need for the data to be considered safe? If one brewery was much bigger than the others, could the smaller ones back out the data of the bigger one?

This ambiguity leaves a lot of room for choice. The agencies can choose to publish as much data as possible given legal constraints, or they can avoid the risk and design rules that reduce the amount of data released. The agencies often choose the latter.

The 2022 proposal to change the Current Population Survey Public Use microdata files is an instructive example. In early 2022, the Census Bureau announced a plan to more than double the population threshold for suppression and round wages. By December of 2022, the Bureau walked back those changes due to pushback from data users, finding alternative approaches less damaging to the usability of the data.

Given this tendency towards limiting data release, which I myself observed over years working within the statistical system,3 it is entirely reasonable for policymakers to rebalance the scales and place greater emphasis on producing more and higher-quality data.

But if the administration’s goal is to pursue that rebalancing, banning a specific disclosure avoidance method is not the way to do it.

Without changing the laws governing privacy protections at the statistical agencies, or promoting a culture and leadership more focused on the quality and volume of data they produce, removing noise infusion as a disclosure avoidance technology will lead to less data overall.

Worse yet, the prospect of less data comes at exactly the wrong moment. There is growing bipartisan agreement that we need significant investments in our statistical system to improve our measurement of AI’s impact on the economy.

I hope to be proven wrong, and that the rollout turns out to be more sensible than I’ve imagined it here. But this new policy has me worried.

The China Shock refers to the widespread job loss that certain workers and communities suffered because of the rapid rise of trade with China in the 2000s.

The AI Shock refers to the same thing happening to white collar workers from the rise of Artificial Intelligence.

For now, the AI Shock is merely a potential shock. But a number of economists, journalists, and other experts have drawn worrying parallels between these two shocks.

We understand why. Across a large body of literature, researchers have found that the China Shock had big and enduring impacts on the local economies and workers who were most exposed to it.

Even more concerning, the extent of the China Shock’s damage was a surprise to economists and other experts. It had long been known that expanding trade will produce winners and losers, but economists tended to think that jobless workers would adjust by moving to the better job opportunities created in a growing economy. Outmigration and flexible wages would restore balance to local labor markets, while many factories would be repurposed to make different products or provide services.

For too many communities, however, especially certain manufacturing towns in the Midwest and South, those adjustments never came. Jobs were lost and not replaced. Jobless people who stayed in their towns simply stopped looking for work. Health, crime, and family formation deteriorated.1

Will AI do to laptop workers what the China Shock did to so many factory workers? Hoping to avoid the same mistakes of the past, researchers and policymakers will be tempted to say yes.

But after a closer look at the details of the China Shock, combined with our understanding of workers exposed to AI, we arrive at a different conclusion.

The real lessons of the China Shock emerge not from its similarities with the AI Shock, but from its differences. And these lessons should mitigate concerns about a massive upcoming disruption, not exacerbate them.

But to understand how we discovered these lessons, we first need to explain how we compared the experiences of China Shocked workers to the potential outcomes for AI Shocked workers. We begin with our measures for each set of workers.

China Shock Exposure

The workers most exposed to the China Shock were highly concentrated in a select few manufacturing industries. In fact, the vast majority of American workers had no exposure at all.

Here is how we know.

A paper published in 2021 by David Autor and his co-authors, building upon earlier work, produced a measure of trade exposure by industry — specifically, the increase in Chinese imports as a share of each industry’s output.

We largely follow this approach. But to arrive at a measure of China Shock exposure for workers, we did two things. First, to measure industry exposure, we chose the time period of 1991 to 2007, a window that captures both the 1990s rise in Chinese import competition and the acceleration after China’s 2001 WTO accession. Second, we adapted the Autor measure — again, the rise of Chinese imports within each industry — by assigning exposure to individual workers based on the industries they worked in.

Our findings:

• For 82 percent of American workers — including everyone outside of manufacturing — there is no measured exposure. They worked in industries in which there was no rise in Chinese imports as a share of their industries’ output between 1991 and 2007.

• Another 14 percent of American workers faced positive but still modest exposure. The rise in Chinese imports was less than 15 percentage points as a share of their industries’ output.

• In contrast, roughly 5 percent of workers were in industries where the growth in Chinese imports was quite large — 15 percentage points or more — relative to output in their respective industries. For example, Chinese imports climbed by 15 points in electrical machinery and tires, 54 points in computers and related equipment, 70 points in toys and sporting goods, and 77 points in footwear. All are manufacturing industries.

It is this last category — the 5 percent most exposed workers — that we use as our comparison group in the rest of this analysis.

In other words, when we compare workers exposed to the China Shock to workers exposed to the AI Shock, we are specifically referring to the 5 percent of American workers who were most exposed to Chinese import competition between 1991 and 2007.

This group numbers about 6 million workers. They all worked in manufacturing, representing about a third of all workers in manufacturing industries. We explore their other characteristics in the sections to come.2

AI Shock Exposure

There is no directly comparable dollar based exposure measure for AI.

Instead, we use the approach in a 2024 paper by Tyna Eloundou and co-authors to identifying the set of workers, based on their occupations, that are most exposed to AI.

Eloundou and co-authors begin by measuring whether the use of AI, potentially with additional specialized software, could reduce the time to complete a task by at least half. To do this they prompt both expert human reviewers and an early version of OpenAI’s ChatGPT-4 to classify tasks based on this time reduction rubric, finding a high degree of agreement between them. For our analysis, we will refer to such tasks as those that can be “done by AI.”

Eloundou and co-authors then aggregate from the task level to the occupation level, computing the share of tasks that can be done by AI.

What is useful about this measure is that it focuses on what could be automated, not what has been automated today. Because the AI shock is mostly yet to come, a forward looking exposure measure is more appropriate.

Another complicating factor is that it is not yet possible to know how much task exposure will lead to actual job loss. For the China Shock, empirical estimates show that the 5 percent most exposed workers were at serious risk of job loss. For the AI shock, however, there is no evidence about the share of a job’s tasks being done by AI that would constitute such risk.

Even if a very high share of an occupation’s tasks can be automated by AI, the occupation itself may survive as a new and different combination of tasks.3 As there is no obvious cutoff for which workers will face displacement (job loss) pressures, we consider multiple cutoffs for the pool of workers that are AI exposed. Table 1 below shows the percentage of the workforce employed in occupations at each respective amount of task exposure to AI.

Our narrowest definition of high exposure assumes that the AI shock only affects the workers at the most extreme end, where AI can replace 90 percent of their tasks. These workers represent less than 1 percent of the workforce.

Our broadest definition assumes the AI shock disrupts a much wider swathe of jobs, including everyone where at least half of their tasks could be done by AI, representing closer to one out of every four workers.

But as our analysis will illustrate, the results of our comparison between China Shocked workers and AI Shocked workers are consistent regardless of where we draw the line.

AI Shocked Workers: More educated, better paid

With highly exposed groups for our two shocks defined, we can explore how they are similar and how they differ. Two differences are especially notable.

1. Compared to China Shocked workers, AI shocked workers are far more educated.

Figure 2 shows that the majority of China Shocked workers have an education level of a high school degree or less, while only 6 percent have an advanced degree.

Using the broadest sample of AI Shocked workers — at least half of their tasks are exposed to AI — the majority of them have a bachelor’s degree or higher. Workers in this group are more likely to have an advanced degree (18 percent) than just a high school diploma or less (17 percent).

And the more concentrated the AI shock, the more educated the group becomes. Workers in the most exposed AI Shock group, representing less than 1 percent of the workforce, are three times more likely to have an advanced degree than to have a high school diploma or less.

In short, regardless of how broad or narrow we define the AI shock, it will affect a group of workers who are substantially more educated than workers who were most exposed to the China Shock. As AI exposure rises, so does educational attainment.

One of the reasons for the disparity is simply that average education rates increase over time, with the China Shock reflecting an early 1990s workforce. Between the early 1990s and early 2020s, the share of workers with a high school degree or less fell from about 48 percent to just over 30 percent. But even if we redo the analysis comparing shocked workers in the same period, the educational gaps persist.4

2. AI Shocked workers also have higher earnings than China Shocked workers.

Regardless of which AI Shock exposure group we use, its share of the top third of all earners is much higher than the China Shock share, while the reverse is true for the respective shares of the bottom third of earners, as shown in Figure 3.

Only about a third of China Shocked workers were in the top tercile (top 33 percent) of overall earners during the time of that shock, whereas 43 percent of the broadest group of AI exposed workers made the top tercile. Meanwhile, when looking at the most exposed group of AI Shocked workers, nearly two thirds of them were in the top tercile.

In contrast to the heavy concentration of AI Shocked workers among the top tercile of earners, China Shocked workers were fairly equally distributed across the wage spectrum, with around a third of workers in each tercile.

Not all workers are equally vulnerable

Why does it matter that AI shocked workers are more educated and have higher earnings? The reason is that education and earnings greatly affect the ability of workers to adapt to job loss.

One piece of evidence supporting this view is that the least educated are more likely to be unemployed. Census data going back over 80 years shows that unemployment for the least educated third of workers is always above, and usually at least double, that of the most educated third.

Another piece of evidence is that even among those workers who lost jobs from the China Shock, not all of them suffered equally.

Difficult adjustments and longlasting effects were concentrated among workers with lower initial wages.5 Low wage manufacturing workers in highly exposed industries lost about 1.2 years of initial earnings over the following 16 years relative to comparable workers in less exposed industries. But middle wage workers lost much less, and the estimated effect for the highest paid third of workers was essentially zero.

High wage workers were able to leave exposed firms and industries without large earnings losses. Lower wage workers, on the other hand, were more likely to remain at their initial employer until the firm underwent mass layoffs, pushing those workers into worse jobs or forcing them to exit the labor force entirely.

The proof that more educated and higher paid workers adapt better to shocks is not just limited to the China Shock research. A similar pattern occurred during the Great Recession. Workers with more education were less likely to lose their jobs initially, and when they got a new one it was less likely to be part time or to pay lower earnings. Looking at the longrun effects of local labor market shocks from the Great Recession, Danny Yagan found that the workers with the highest initial earnings were least likely to be jobless in 2015, long after the recession had officially ended.

The relationship between education and adaptability to local shocks has also been found in the decline of coal mining.

One reason that better educated, higher paid workers are more able to adapt is they are more willing to move. Another reason is that their skills are more transferrable.

Whatever the cause, the evidence on a wide range of economic shocks illustrates that when job loss is concentrated among lower wage and less educated workers, it is likely to generate different adjustment problems than a shock concentrated among workers with higher pay and more college degrees. We consider this an important reason to worry less about the AI shock mirroring the China Shock.

Where the shocks are

The disruptiveness of an economic shock is determined not just by which people get hit but also which places. Research has shown that one reason the China Shock was so disruptive was that it was geographically concentrated.

Local labor markets adapt to small losses of employment all the time. Even outside of recessions, roughly 30 million jobs across the United States are lost every year due to layoffs or business closures. But the national unemployment rate stays low as workers reallocate from shrinking sectors or firms to growing ones. This activity is simply part of the everyday churn of the economy.

When job losses are large and concentrated in a few specific places, in contrast, it becomes much more difficult for those local economies to adapt. Rather than Schumpterian creative destruction, the job loss is large relative to the local economy. The reallocation will take longer, and the struggle to adapt also can generate demand shortfalls as local spending pulls back and spills over into other sectors. Nationally set monetary policy will do little to stabilize local economies suffering from demand problems that arise from concentrated shocks.

If the few million jobs lost from the China Shock had been scattered evenly throughout the country, they would have represented just a small fraction of normal economic churn, and thus workers would have bounced back more easily. Instead, the shock was heavily concentrated in manufacturing towns in the Rust Belt and South.

The good news is, the evidence we have suggests that the AI shock will be far less geographically concentrated. Figures 5 and 6 below show the respective place level exposures to the China Shock and the AI shock across Commuting Zones (CZs). China Shock exposure is concentrated in a much narrower set of manufacturing heavy labor markets. AI exposure, in contrast, is more evenly dispersed throughout the country.

The difference is also obvious when looking at relative exposure in the hardest hit places, as shown in Figure 7. The 10 Commuting Zones most exposed to the China Shock average about 5.8 times the exposure of the average one; the next 40 about 3.2 times the exposure of the average one; and the next 50 about 2.1 times.

AI exposure is much flatter, with the top 10 CZs only about 1.2 times the average CZ.

Which types of places

How a place adapts to a shock also depends on the characteristics of the place.

Research shows that local economies whose workers have higher average levels of education adapt much more easily to economic shocks than places with lesser educated workers. The reasons for their resilience go beyond the fact that higher educated workers themselves are more resilient. Places with more human capital tend to have faster rising populations, which helps to boost dynamism and the growth of underlying demand growth. A more educated population will also be more innovative and have more entrepreneurs who identify next best uses for displaced capital and labor.

When we compare China Shocked places to AI shocked places, it is clear that education levels are a major differentiator. Figure 8 below looks again at exposure ranked Commuting Zones, but it compares the average share of workers with a high school degree or less against the average share of those workers for all Commuting Zones in the relevant year.6 (We use 1990 education data for the China Shock and 2024 education data for the AI Shock.)

The most China Shocked places had populations with lower than average education levels at the time, while the opposite is true for the most AI exposed.

Looking at the share of the population with bachelor’s degrees or higher, we see a complementary pattern. China Shocked places are slightly below average, while AI shocked places are way above average.

Missing Something?

We have presented the evidence for why the China Shock was far worse than the AI Shock is likely to be. What might the opposite case look like? We’re happy to play devil’s advocate, both because it’s a good intellectual habit and also because the case ends up looking so much weaker than ours.

One possibility is that the AI shock materializes much faster than the China Shock. If AI agents prove to be as capable as their creators suggest, displacing a worker whose primary tasks occur on a laptop could be as simple as installing and initializing the AI agent.

A second possibility is that the AI Shock ends up displacing a much bigger share of workers than we expect. In our most aggressive AI Shock scenario, 27 percent of workers would be exposed, 5.7 times larger than the 5 percent exposed to the China Shock. If most or all of these exposed workers lose their jobs — especially if, coinciding with the first possibility, they lose their jobs quickly — an utterly massive share of the labor force will find itself suddenly unemployed. The job losses and subsequent demand destruction could be enormous.

That’s the worst case scenario: AI hits faster and hits bigger than even pessimistic forecasts. We find this outcome unlikely, for two reasons.

First, rolling out AI at such scale requires significant computing hardware, which takes time to build. Data center construction is a real constraint and is already causing policy blowback. The prices of information processing equipment have also surged, rising at the fastest pace since 1959. The current AI buildout is already facing big obstacles. Imagine how much bigger they would be for the kind of buildout required to replace millions of workers.

Second, if the AI Shock were to produce vast job losses, the magnitude of the effect itself makes economic policy more powerful in offsetting the effects on demand. Recall that monetary policy, for instance, was of limited use for helping local economies destroyed by the concentrated China Shock. If the AI Shock proves to be bigger and more widespread, monetary policy regains in potency.

What is more, significant losses of jobs because of AI would be accompanied by rapid economic growth. Both economic growth and AI augmentation are helpful for smoothing the economic adjustment in a variety of ways, including boosting the demand for other types of workers.

Nevertheless, we admit that an AI Shock large enough to upend the labor market overnight would be unprecedented, which means that reasoning about it involves uncertainty in both positive and negative directions — about adjustment costs but also benefits, layoffs but also progress, unemployment but also innovation. The economy would be entering uncharted territory.

In such a world, comparisons to the China Shock, or to any other great historical shock, would therefore be pointless.

A tale of two shocks

It is true that the impacts of the China Shock were more negative than many economists expected. But it is also the case that those negative effects are no longer a mystery. Thanks to a growing body of research, quite a bit is known about which types of people and places were more affected than others.

In sum, the China Shock was disruptive because it 1) disproportionately hurt workers with less education, 2) was geographically concentrated, and 3) inflicted its worst damage on places with lower levels of human capital.

It is still early days for AI and so a large dose of humility is appropriate. But so far as we can tell, the AI Shock is aimed at the kinds of people and places that actually weathered the China Shock just fine.

The old cliché is that history does not repeat itself but does often rhyme. If the AI Shock turns out to be as bad as the China Shock despite such different underlying characteristics, it would turn the cliché on its head: an instance in which history did repeat itself but skipped the rhyming. It’s not impossible, but we find this outcome unlikely.

Appendix

In the preceding analysis, we compare the education levels of the workers most exposed to the China Shock to those who are most exposed to the AI shock. One complicating factor is that these two shocks are occurring at different points in time, and education levels among workers overall climbed.

Absolute increases in education over time are relevant for resilience, but we can also compare relative education rates by comparing each group of exposed workers to the average education rates of the overall workforce from the same time period. In other words, we can compare the China Shocked workers to the average education of workers between 1991 and 1993. And we can compare the education of AI Shocked workers to average education levels of all workers between 2021 and 2022.

The results are directionally the same: Workers exposed to the China Shock are still less educated, while the Al exposed groups are more educated.

Specifically, compared to the overall 1991 to 1993 workforce, workers exposed to the China Shock are 10.4 percentage points more likely to be high school graduates or less and 5.9 percentage points less likely to have a bachelor’s degree or higher.

In contrast, the most AI Shock exposed groups are substantially more educated than the 2021 to 2022 workforce. Depending on the exposure group, they are 13.7 to 24.1 percentage points less likely to have a high school diploma or less, and 13.1 to 29.9 percentage points more likely to have bachelor’s degrees or higher.

Taking the z scores for each group shows consistent results, with China Shock exposed workers below their period average and every exposure level of AI shock above its period average.

While overall national jobs numbers continue to inch upward, more granular data at the state and sectoral levels reveal a startling trend across the country: healthcare jobs in many states are booming, but most other sectors are stagnant.

“And no state appears as reliant on the healthcare and social assistance sector to support its labor market,” write EIG senior fellow Kenan Fikri and research associate Thomas Cronin, “as the nation’s most populous and the world’s fourth largest economy: California.”

New research from Fikri and Cronin finds that since March 2022 California has increased employment in these fields by a massive 25.3 percent, but has seen negative private-sector jobs growth otherwise.

California stands out from the pack, but it is not the only state where healthcare is masking labor market weakness overall. Georgia, New York, Oregon, New Jersey, Maryland, and Colorado all have experienced healthcare job growth exceeding 17 percent while employment in all other sectors in these states has expanded by 3 percent or less.

Nationally, jobs in healthcare and social assistance increased by a strong 16.3 percent, compared to just 2.9 percent in the non-health segments of the economy.

Amplifying EIG’s publishing of this research at Agglomerations, the Wall Street Journal printed a story in its Thursday May 14 edition anchored by these stunning findings. “EIG analyzed a three-year span of the most recent Labor Department data to cover a mixed period for the labor market, which saw a dramatic cooling off after a postpandemic hiring boom,” Harriet Torry and Josh Ulick conveyed to Journal readers.

While healthcare and social assistance jobs are not a bad thing to add to a state’s employment mix, the wages associated with them tend to be lower than in fields in which these states have suffered losses. Moreover, California, Oregon, and other states may see the healthcare sector struggle going forward with the dissipation of federal Medicaid spending.

“If this federal support starts to dry up,” Fikri told the Journal, “we’re going to soon discover whether the healthcare boom has legs sufficient to stand independently.”

Policy

• EIG submitted a comment letter arguing that the Department of Labor’s proposal to raise prevailing wages for H-1Bs and green card applicants is based on flawed premises. EIG found that the proposal fails to account for differences in worker education, experience, and qualifications when estimating prevailing wages. The letter also argued that the proposal would continue to advantage large outsourcing firms and fail to select for top talent. EIG urges the Department of Labor to adopt the alternative proposal, known as Experience Benchmarking, to correct these flaws. EIG also noted, however, that fixing flaws within the H-1B program is a suboptimal solution overall and recommends market-based immigration reforms, such as selecting H-1Bs with the highest salary offers.

• EIG issued a statement in response to the House passage of the 21st Century ROAD to Housing Act. John Lettieri, President and CEO of the Economic Innovation Group. “America cannot lower housing costs without building substantially more homes of all types,” said John Lettieri, EIG President and CEO. “This legislation takes meaningful steps toward that goal, and Congress should move quickly to get it across the finish line.”

Research & Analysis

Smells like Teen Unemployment
The overall labor market appears reasonably healthy, but real weakness has seeped in and virtually all of it is concentrated in the young. Employment has fallen not just for recent college grads but also for noncollege young adults and for teenagers, making it unlikely that the cause of these problems is AI replacing worker tasks. Something else is going on. But what?

How Students and Recent Grads are Responding to AI
Sarah Eckhardt and Nathan Goldschlag find that undergraduates “are flocking towards the most-AI-exposed degrees, with enrollment in those degrees up 8 percent last year compared to 2017. This trend holds despite a notable decline in Computer Science degrees, one of the most-AI-exposed degrees, but whose decline is more than offset by increases in other exposed degrees like Engineering.” One reason why is that the jobs these students are likely to be offered still pay very high wages.

The Problem of Partial Noncompete Bans
Partial noncompete bans can be net positives if they are carefully written and apply to a wide swath of workers, but too often the exemptions — some of them quite absurd — lead to mass confusion, chicanery on the parts of employers, and unenforceability. The evidence is clear, write Thomas Cronin and Ben Glasner: “Banning noncompetes for certain job types or income levels helps curb some abuses, but it leaves intact many of the mechanisms by which noncompete agreements hurt most workers and stifle innovation and entrepreneurship. The best way to achieve a stronger, fairer, and more dynamic economy is for states to ban noncompetes entirely.”

Around the Horn

• For the National Bureau of Economic Research, Nathan Goldschlag and coauthors have issued a working paper examining the relationship between patents and economic growth.

• Kyla Scanlon compiles ways to respond to the rise of AI, including “Be AI native, not AI proof” citing the post on students and young graduates from Sarah Eckhardt and Nathan Goldschlag.

• In City Journal, Jordan McGillis explains the fraught labor dynamics hindering autonomous trucking in California.

• In National Review, McGillis argues against a federal subsidy for an Alaskan pipeline.

• In The Dispatch, McGillis applies EIG’s Distressed Communities Index to family housing decisions.

• And also in National Review, McGillis has published a longform reported essay on the U.S. Geological Survey’s Earth Mapping Resources Initiative.

• In Newsweek, Billal Rahman quotes Sam Peak on increased scrutiny of the H-1B program.

• In an essay for the Washington Post, the University of Notre Dame’s Alexander Kustov highlights EIG’s survey work showing public support for high-skill immigration.

• Also in the Washington Post, reporters Shira Ovide and Rachel Lerman drew input from Nathan Goldschlag in their coverage of AI’s effects on the labor market.

• Coverage of the Trump administration’s new retirement plan continues, with reporters citing Ben Glasner and Sarah Eckhardt’s pivotal research on the issue in USA Today and Forbes.

EIG Chart of the Month

Via Kenan Fikri’s post on the dominance of health care jobs in the California labor market:

There is another data point making me skeptical that AI is behind the labor market woes of young people: The unemployment rate for teenagers (ages 15–18) has risen more than for any other category of young workers. From its post-pandemic low, teenage unemployment has climbed 4 percentage points, compared to the 1.5 percentage point increase for recent college graduates.

Is it plausible that the cause of these trends is primarily AI? In other words, how likely is it that AI is the main cause of employment struggles not just for recent college graduates, but also noncollege youths and even teenagers?

What do teenagers do?

When looking at the kinds of entry-level, white-collar “laptop” jobs that recent college graduates tend to do, it is natural to wonder if AI is playing a role in the lack of hiring. The same is not true when looking at the kinds of work that teenagers do.

The industries that hire teenagers are simply not the ones that AI is likely to disrupt by displacing a large share of workers. Of the teens who were employed from 2022 through 2024, for example, a combined 75 percent of them worked in Accommodation and Food Services; Retail Trade; Arts, Entertainment, and Recreation; and Health Care and Social Assistance. ChatGPT can do a lot of things, but it can’t help if you want fries with that.

Measures of AI exposure at the occupation level tell a similar story. According to three separate measures of occupational AI exposure from influential research papers, only a small share of teens — between 3.8 and 8 percent, depending on the measure — are in the most AI-exposed occupations.

To whatever extent the struggles of teenage workers have the same cause as the employment struggles of other young adults, that cause is almost certainly AI.

Side note: Are the kids alright though?

Closer scrutiny of the teen labor market reveals an interesting wrinkle in the story. By one key metric, conditions are simply back to what they were before COVID.

The employment rate for teenagers, which is the share of all teens who are employed, started surging in 2021 as the pandemic recovery took hold and climbed far above its pre-pandemic high.1 It finally peaked in early 2023 before subsequently falling to where it is now, roughly consistent with its level from before the pandemic.

Put another way, the labor market for teens has indeed weakened, but only as a reversion to a historically normal level, and a pretty strong one at that.

Contrast the teen story with employment for young adults. The employment rate for both college and noncollege young adults (ages 22–25) barely recovered to its pre-pandemic level before stagnating and then falling again in recent years.

Why teen employment rose to greater heights during the pandemic than young-adult employment is itself an interesting question. Part of the story likely has to do with post-pandemic policies, which by design made it easier to not work for people who benefited from them — a group that largely excluded teens. Stimulus checks went to their parents. Teens mostly never qualified for expanded unemployment insurance. And because most teens live with their parents, they get nothing from an eviction moratorium. Nor do people who haven’t been to college yet benefit from pauses in student loan payments.

Still, the relative resilience of the teen labor market is consistent with the fact that it has deteriorated in recent years, just as the labor markets for recent college graduates and noncollege young adults have deteriorated. Something is damaging the labor market for young people. If, as I have already argued, that something is unlikely to be AI, then what is it?

Low Hire, Low Fire

I have to admit that current labor market conditions are kind of weird. This is not a normal business cycle.

In addition to the struggles of young workers, another sign of weakness for the overall labor market is the slowdown in real and nominal wage growth. The quits rate, which tends to climb during strong labor markets as workers leave their jobs for better opportunities, is down from its post-pandemic highs. And interest rates are probably still above neutral, meaning that the Fed still has some pressure on the brakes of the economy as it tries to fight inflation.

On the other hand, the prime employment rate, which is the share of all adults aged 25–54 who are employed, has moved down only slightly from its peak. Total unemployment has inched up, but only a bit. Taken by themselves, and in the context of previous historical periods, these two measures would suggest a healthy environment for jobs.

One way to summarize this bizarre labor market is that real weakness has crept into it, but it seems concentrated almost entirely on the youth. Just why the youth are bearing the brunt is unclear, an ongoing economic mystery.

For teenagers, one possible contributing factor is that because their employment rate climbed so high after the pandemic, it thus had the farthest to fall. But that point doesn’t resolve why youth of all kinds, including groups that didn’t see a post-pandemic surge, have also experienced these deteriorating conditions. Teenagers are hardly the only canary in this coalmine.

Time to Zoom Out

Somewhere between the economic measures showing weakness and those showing strength is the middling reality of the low hire and low fire economy, in which employers are not laying off workers but neither are they hiring many of them. The lack of visibility into the economic future has left businesses paralyzed by a generalized uncertainty.

There are plenty of good reasons for it: tariffs that move up and down violently, wars abroad and the resulting shocks to energy prices, interest rates that might go up or down, high levels of deportations, and — yes, it’s fair to include — the unknowable effects of AI on the economy and society.

But while AI is a possible cause of uncertainty, one of several, it’s important to recognize that employers not hiring because they are frozen by AI uncertainty is not the same thing as AI replacing the tasks of young workers. Those are two different stories.

Economists and researchers and journalists are mostly focused on the story of task displacement. They are constantly trying to assess how real it is, whether it has started, how far it will go, which jobs are vulnerable to it, and so on.

I think it’s time to zoom out and instead look at AI uncertainty and also the many other types of possible macro uncertainty the economy now confronts. Given the mystery surrounding youth labor-market weakness, it would be helpful to widen our scrutiny beyond the question of which tasks performed by recent college grads might someday be taken by AI, and instead investigate the strange macroeconomy we are now living through.

See our GitHub with replication code here.