The finding, from an annual NAR survey of homebuyers, set off a social media firestorm and earned coverage in virtually every national media outlet. It also played into a burgeoning narrative that young Americans are screwed. Softening labor markets are colliding with long-simmering problems like student debt. Consumer sentiment among Americans under 35 is worse today than during the toughest days of the Great Recession. NAR’s survey provided yet more validation that young people are falling hopelessly behind.

But there is one problem with NAR’s data: It’s wildly wrong.

The National Association of Realtors likely overstates the age of the typical homebuyer by 15 to 20 years, while overestimating the age of the typical first-time buyer by between 5 and 10 years.

Much larger government-run surveys like the gold-standard American Housing Survey and American Community Survey show no increase whatsoever in the age of the typical homebuyer since the pandemic, when NAR’s median age estimate went truly vertical. My analysis of American Housing Survey data finds the median buyer in 2023 was 42. A similar estimate using 2023 American Community Survey data shows a median age of 41, unchanged for well over a decade.

While the National Association of Realtors’ survey was conducted more recently than those two, other recent data also cast doubt on its estimates. An analysis of mortgage data from the American Enterprise Institute and Cato Institute found that the median first-time homebuyer in 2024 was, at most, 34, unchanged since at least 2018. A large national survey of homebuyers conducted this summer by Zillow found the median overall homebuyer was 42 — 17 years younger than NAR’s estimate.

NAR’s survey and these other data do not measure precisely the same things. For instance, mortgage data would not capture all-cash transactions, whose market share has modestly increased over the past few years. But the American Housing Survey and American Community Survey cover all buyers, whether paying in cash or financing, so cash purchases don’t seem to explain the enormous divergence between NAR’s results and the other sources.

A representative of the National Association of Realtors offered another possible explanation to the Washington Post: that young buyers priced out of markets where they want to live are instead buying investment properties in cheaper markets and renting them out. They therefore don’t show up in NAR’s survey, which only covers primary residences. But NAR offered no concrete evidence that this was a widespread phenomenon, let alone one big enough to account for its claim that the typical buyer today is a full decade older than in 2023. And in my own analysis of ACS and AHS data, I filter only for owner-occupied units, making this point moot.

So what explains the disagreement? It could be caused by low response rates, a problem nagging survey-takers across the world. Adding to the baseline difficulty of getting responses from households, NAR’s survey is also very long — it asks 120 questions — and is conducted purely by mail. Its latest edition surveyed more than 170,000 households, but got responses from just 6,100. That is a rate of less than 4 percent, far lower than government surveys like the ACS or AHS. The NAR’s old-school survey-taking is exactly the kind of methodology that no longer holds up in a digital age.

Whatever the explanation, it’s certainly no mystery why the NAR results garnered far more attention than these other, high-quality estimates: We live in an information ecosystem where the most pessimistic data goes viral regardless of its quality. If a statistic directionally reinforces a compelling narrative — even if the degree of that narrative is wildly overstated — its flaws are largely ignored.

While NAR’s specific claims are massive outliers, housing affordability is indeed a growing problem, particularly for young people. Millennials and Gen Z are well behind prior generations in their rates of homeownership. The typical age of a first-time homebuyer does not appear to have risen nearly as much as the NAR estimates, but it has increased over the past few decades. This is in large part thanks to communities’ refusal to allow enough new housing construction.

Nationwide, the rate of new housing construction per capita has fallen by half since the 1970s, and has still not yet returned to pre-Great Recession levels. As a result, home prices in the 100 largest metros are five times the median income, pricing young families out of the market. Since 2020, the number of children under 5 has fallen a staggering 19 percent in New York City, 15 percent in San Francisco, and 14 percent in Los Angeles. Once engines of upward mobility, these so-called “superstar” cities have become so expensive that many families are simply leaving. Meanwhile, booming metros across the Sun Belt, where rules on new construction are generally more relaxed, are doing far better at attracting families and investment.

The most pessimistic, headline-grabbing housing market data points may be wrong, but the reaction they elicit speaks to something true: America simply isn’t building enough housing. Until that changes, the housing market is going to leave more young people behind.

President Trump’s showdown with Mexico and Canada this week was shaping up to be an early test of just how far he has shifted the Right’s views on trade policy over the last decade. The last-minute decision to delay the proposed 25 percent tariffs instead put off this fight for another day. The loose coalition of trade policy heretics, who arrive at support or sympathy for tariffs for a diverse range of reasons, will hold together for now.

But as the president and his team develop a trade policy agenda for the rest of his term, fissures in this coalition are likely to grow over time. Trump will ultimately pick one general course — are tariffs a temporary negotiating tool or a permanent revenue-raiser? — leaving some factions victorious and others unsatisfied.

These divides are fascinating. While the Right has become broadly more supportive of protectionism over the last decade, the range of thought within this coalition remains wide. On Saturday, I sketched out what I view as the six main pro-tariff camps. I elaborate on them below.

The Reciprocity Camp: Tariffs can be leveraged to help us achieve freer trade.

The first camp’s motives can be summed up in one word: reciprocity. The United States has signed 14 free trade agreements with 20 countries and is a member of the World Trade Organization. No major, rich economies have double-digit effective tariff rates; the holdouts with average tariff rates above ten percent are almost exclusively poor African countries.

Nevertheless, the U.S. does have longstanding, industry-specific trade disputes with many friendly countries. Famously, the European Union puts a ten percent tariff on cars imported from the United States, which in turn applies only a 2.5 percent tariff to European cars. But these days, with average tariff rates so low, it’s arguably non-tariff barriers to trade where the real action lies. The U.S. has long sought to push Canada to take a more free-market approach to its lumber and dairy industries, for example, arguing that Ottawa’s backing undercuts American competitors. In theory, tariff threats could yield concessions on these issues.

A traditional free trader like Milton Friedman would argue for unilateral tariff abolition regardless of what other countries do. Another country wants to make its citizens and businesses pay more for imported goods? That’s their problem, not ours. Yet this line of thinking has not caught on, either in the U.S. or abroad. Domestic firms want their government to fight for their ability to export abroad. The Reciprocity Camp wants us to get back to playing hardball in defending their interests, and free trade in general.

The Negotiator Camp: Tariffs can extract other, non-trade concessions in areas like migration or drugs.

This camp likely feels vindicated after this week, as President Trump’s tariff threats against Mexico and Canada extracted promises from the leaders of both countries to send more resources to their respective borders with the United States. Last week, a threat of tariffs on Colombian imports pushed its president to back off a demand to stop returning deportees to the country on military aircraft.

Like the Reciprocity Camp, the Negotiators see no problem using the threat of tariffs to get what they want. But they want to go further, using the threat of tariffs to achieve goals unrelated to traditional trade policy, like drug trafficking, migration, or human rights.

Both of these camps’ support for tariffs is merely tactical, and neither may be comfortable with a more permanent protectionist regime.

The Friendshoring Camp: Tariff are a tool to shift supply chains to allied countries, with whom we should integrate further.

Further down the protectionist spectrum are the “De-riskers” and “Decouplers,” both of whom see tariffs as a tool for extricating ourselves from the Chinese economy. The former group wants heavy tariffs on China to nudge strategic supply chains toward allied countries, remaining indifferent as to whether China continues to manufacture our apparel or furniture. The latter group wants broader tariffs on China to separate our economies entirely.

The Friendshoring Camp doesn’t think we are ever going to produce all the electric vehicles consumed by American consumers. They would accept an outcome in which the U.S. reshores some EV production but sources much more from South Korea or Japan.

This group does not support universal tariffs, just tariffs on imports from adversarial countries. In fact, many want deeper integration with friends like Canada or the European Union. A prolonged trade war with fellow democracies could alienate this camp.

The Builder Camp: Tariffs may make Americans marginally poorer. But they will help us reshore manufacturing, a national security imperative.

Many hard-nosed national security hawks convinced a hot conflict with China is imminent may fall into this camp. America’s defense industrial base is ill-prepared for a potential war over Taiwan. CSIS estimates that U.S. forces would run out of some key ammunition within a week. China’s shipbuilding capacity dwarfs that of the United States by a factor of more than 200. An inability to quickly ramp up defense manufacturing in the event of a war could have devastating consequences.

The Builder Camp may acknowledge that tariffs will impost costs on Americans, including businesses. But they argue that this is a necessary step to reshore sectors that simply cannot be left to foreign countries. Whether tariffs are likely to accomplish this goal in practice is an exercise I’ll leave up to the reader.

The McKinley Camp: Tariffs were the secret to 19th century American prosperity and would do the same today. Impose them permanently.

From writers like Michael Lind to groups like the Coalition for a Prosperous America (CPA), the McKinley Camp offers an alternative history of why the United States got rich in the first place, putting tariffs at the center of the story. Nineteenth-century America indeed had far higher tariffs than we do today, though average rates did steadily decline between the end of the Civil War and the outbreak of World War I.

As opposed to the Builder Camp, which concedes that tariffs have both costs and potential benefits, the McKinley Camp views tariffs as much closer to a free lunch. CPA, in particular, is adamant that tariffs will increase not just U.S. manufacturing output, but productivity. Its modeling team asserts that an across-the-board 10 percent tariff would increase inflation-adjusted incomes by 5.7 percent. The empirical assumptions within CPA’s model have come under heavy criticism.

The Depreciation Camp: The U.S. dollar’s status as the world’s reserve currency is the fundamental issue. Tariffs can disrupt or ameliorate the problems with this system.

The final camp accepts mainstream economists’ view that it is the U.S. dollar’s place at the center of the global financial system that drives our persistent trade deficits, not tariff policy.

My friend likens the dollar’s role to a “resource curse.” Demand for American financial assets drives up the value of the currency, making our exports uncompetitive and incentivizing “financialization at the expense of investment in the real economy.” These points have also been fleshed out in Michael Pettis and Matthew C. Klein’s, Trade Wars Are Class Wars, whose framework has earned devotees on both the Left and Right.

Before the pending tariffs on Canada and Mexico were scrapped, Sam suggested they could be the opening gambit in a campaign to de-dollarize the global financial system. Trump’s incoming CEA Chair Stephen Miran wrote a notable paper in November floating the idea of a “Mar-a-Lago Accord” to weaken the Dollar. While Trump appears to be using tariffs to negotiate on other issues for now, I would not be surprised if this final camp proves ascendant at some point over the next few years.

Ultimately, this is still Trump we are talking about, so we may end up with some unpredictable synthesis of the above camps, or a new strategy entirely. Regardless, this typology should provide a useful way of thinking about how different factions within the Right are approaching trade policy, and how they might react as Trump’s tariffs agenda comes together this year.

Elon Musk and Vivek Ramaswamy’s “Department of Government Efficiency” (DOGE) is shaping up to be a high-stakes, high-profile effort at overhauling the federal government. Whether it proves to be serious and effective remains an open question—so far the signs are decidedly mixed. 

As folks toss around their ideas for turning DOGE into a force for a leaner, more effective federal government, I’ll offer one of my own: Pay government to trim itself by cutting bureaucrats in on savings. 

Here’s how it would work: When Congress passes a rescission package that revokes prior appropriations for a particular office or program, federal employees within those offices get a cut of the savings, up to some cap. If HUD’s Office of Community Planning and Development, for instance, finds it can accomplish the mission given to it by Congress with $1 million less than what it was appropriated, its employees can share (ten percent?) those savings, divided up into equal-sized bounties. 

Today, when the clock is running out on Congressionally-appropriated funds, bureaucrats face a “use it or lose it” incentive. Funds that an office or agency doesn’t spend (or commit to spend) in the prescribed window essentially evaporate. This chart from the Congressional Research Service demonstrates how this works:

If an office can accomplish its mission with merely 90 percent of what Congress gives it, appropriators will come back with a budget the following year that is ten percent smaller. So offices have a completely rational incentive to spend everything Congress gives them, even if those last few dollars are effectively set on fire.

For an incoming, reform-minded administration intent on cutting waste, the hundreds of offices and agencies responding to this incentive can be hard to track. Their responses are even harder to combat and corral into regular rescission packages sent to Congress.

Bounties would counter the incentive to get every last dollar out the door by giving rank-and-file federal employees a direct, personal stake in saving their agencies and offices money.

Broadly-shared bounties—as opposed to larger rewards for program or agency directors—would have a few additional benefits:

First, sharing the benefits of cutting waste aligns everyone’s incentives. Directors overseeing an office who have either a sincere motivation to trim costs or a large (un-shared) personal reward for doing so can easily be resisted by hard-to-discipline subordinates. Offices making unpopular cuts may still face resistance, but the marginal rebel can be “bought off” by the promise of a cushy bonus.

Second, shared bounties take advantage of the deep knowledge and experience widely distributed throughout the federal workforce. GS-9-level analysts and program managers in every agency are likely aware of small but common instances of waste that higher-ups know nothing about. Indeed, as writes, Al Gore’s Reinventing Government initiative owed much of its success to relying on and coordinating with the civil service, rather than business world outsiders, to source reform ideas.

Finally, creating a bottom-up incentive to identify savings should make it easier to build and coordinate rescission packages in the first place. This is an Executive Branch muscle that has largely atrophied. Congress has not adopted a rescission package proposed by the President since the Clinton administration. Uncooperative offices can simply run out the clock or slow-walk directives from a politically-appointed boss to identify potential savings. In a world where bureaucrats get paid to save the government money, they would be more likely to source ideas from their subordinates and pass them up the chain.

Such an incentive scheme would yield an initial flurry of cuts. But just as importantly, keeping it in place long-term would gradually change the cultural norms within federal agencies from the bottom-up. Thrift would be rewarded across government, while identifying ways to give money back to taxpayers would become just another common, boring, day-to-day activity of the federal bureaucracy.

Reason magazine is out with a short documentary on the new set of degree requirements for childcare workers in the District of Columbia. I recommend you go watch it. New rules mandate early childhood education bachelor’s degrees for childcare center directors, associate’s degrees for teachers, and certificates for assistant teachers.

I find this policy particularly frustrating for two reasons. First, I am a District resident who would like to raise children here. Driving up the cost of my future kids’ care will directly cost me money. Second, early childhood education policy advocates tend to get undue deference from the public, even when advocating for measures that will drive up families’ costs without clear benefits.

Here we have just such a policy. Saddling childcare workers with additional educational requirements will make care more expensive for D.C. families. Yet there is little reason to believe it will improve outcomes for the District’s kids.

D.C. childcare is the most expensive in America. Not coincidentally, families with young kids are leaving the city.

It is hardly an understatement to say the childcare market in Washington D.C. is in crisis. Center-based care for a single toddler in D.C. costs more than $24,000 per year, according to an analysis from the Annie E. Casey Foundation. In the closest state, Massachusetts, annual costs are $4,500 lower. For home-based care, the cost gap between the District and the most expensive state is $6,000.1

In my work with the Economic Innovation Group, I regularly report on the geography of young families. In the early years of the pandemic, families with young kids left major cities in droves. D.C. was no exception. Between April 2020 and July 2022, the number of kids under five nationally fell 3.3 percent, as the country gradually continued to age. But in D.C., the population of young kids fell 7.8 percent over just 27 months, indicative of many young families heading elsewhere. The high cost of childcare in the District likely contributed to this exodus.

Pairing high childcare costs with a negative supply shock is a terrible idea.

Any negative supply shock, such as new regulations with which many current staff can’t comply, will only widen the cost gap between childcare in D.C. and the rest of the country. Surely the city will grandfather existing staff into new requirements, ensuring a smooth transition, right?

In 2022, Martin Austermuhle reported that fewer than half of care center teachers had met the new requirements, six years after they were first approved:

According to the D.C. Office of the State Superintendent of Education, which oversees and licenses child care centers and workers, 78% of directors of child care centers are now meeting their new educational requirement. The number is lower for teachers and assistant teachers, though, at 40% and 34% respectively, and around 50% for caregivers at home-based daycares. Still, an agency spokesman said everyone but directors have until Dec. 2023 to come into compliance, and that this year OSSE funded scholarships for 427 workers to get their Child Development Associate’s credential.

The city is scrambling merely to re-certify current childcare workers. Even those longtime teachers exempt from new rules are facing problems getting their waiver requests approved by the city. Reason highlights the story of one longtime childcare worker:

Ami Bawa, lead teacher and assistant director at a nursery school in northwest D.C., exemplifies the unintended consequences of the regulation. Although she has been working in the field for over 20 years, Bawa may now be forced out of her job. "Even though I have a lot of experiential learning, I don't meet what is now the current standard," she explains.

As a veteran teacher, Bawa is technically eligible to apply for a waiver to continue working, but she's been waiting for five months for a response from the city. "All of these roadblocks make it harder. We're going to lose a lot of really good teachers," Bawa says.

District parents have entrusted Bawa with their kids to provide quality care for two decades, relationships endangered by the city’s failure to process her paperwork on a reasonable timeline. The District will mandate new teachers spend two years obtaining an additional degree of questionable value but cannot itself give longtime teachers an hour to review their waiver applications.

The benefits of early childhood education are social, not academic. That’s bad news for childcare credentialing advocates.

Perhaps we could excuse saddling families with higher childcare costs or the city’s sloppy rollout if new rules were likely to yield great benefits to D.C. children. Unfortunately, we have little reason to believe requiring additional degrees will improve kids’ educational outcomes.

It is true that research finds early childhood education likely has positive, long-lasting effects on kids’ high school graduation rates, future criminality, and substance abuse. Importantly, it also increases the odds that kids ultimately graduate from college. Yet famously, the purely academic effects of the federal preschool program Head Start—measured by test scores—fade away entirely by the first or second grade. As Vox’s Kelsey Piper writes:

In the past few years, early childhood education has taken a beating in studies of its effects a few years down the road. The Department of Health and Human Services commissioned a massive study of Head Start, the flagship early childhood education program, and found “the benefits of access to Head Start at age four are largely absent by 1st grade for the program population as a whole.”

In the 2008 to 2009 school year, when Tennessee had to assign spaces in their early childhood education program by lottery, it created the conditions for the perfect natural experiment. Researchers found, if anything, negative effects: “the control children caught up with the pre-k participants on [kindergarten and subsequent] tests and generally surpassed them.”

There are studies out there which have found lasting benefits to test scores. But in general, the better-conducted the study, the more discouraging the results.

These results are not inconsistent with early childhood education having positive, long-term benefits for kids. Rather, they strongly suggest that the mechanism by which kids’ future outcomes are improved is not the substantive, academic content of childcare programs (which can be tweaked through staff education), but something else.

Perhaps the benefits come from teachers’ nurturing and close attention. Maybe it is being removed from chaotic homes, also enabling parents to work consistently, that helps kids from low-income families. It could simply be the freedom to play, explore, and socialize with other kids all day. Regardless, the benefits seem to be social, not academic. As millions of American parents prove every year, it is indeed possible to create a nurturing, healthy, and stimulating environment for young children without having a college degree.

Unfortunately, the case for additional childcare worker credentialing rests primarily on the idea that it is the academic content of early childhood education that helps kids in the long run. If teachers are more well-read in history or Shakespeare2, the thinking goes, or are equipped with certain expert-approved classroom exercises, maybe they’ll be better prepared to impart their knowledge to kids and improve their test scores permanently. Yet none of that seems to matter. Head Start teachers’ educational credentials and curriculum choices are “not correlated with better outcomes.”

Nevertheless, the nation’s capital of degree inflation persists. Teachers will take out loans and dutifully attend their gen-ed history courses, evidence of relevance to two-year-olds notwithstanding. If you’re like me and plan on sticking around, you’d better start saving now.

Since its publication in 2013, Joe Studwell’s, How Asia Works has quickly become one of the most influential books written about industrial policy in the 21st century. Studwell compares rapid economic growth and industrialization in Northeast Asia (Japan, South Korea, and Taiwan) with failed industrialization schemes in its neighbors to the south (Indonesia, Thailand, Malaysia, and the Philippines). 

In just a few decades, Japan, South Korea, and Taiwan not only grew into wealthy modern economies but did so while often rejecting orthodox free-market recommendations from the International Monetary Fund and the United States. Today, they are home to innovative, globally competitive firms like Hyundai and TSMC which were born out of active state intervention and planning. Studwell argues that this intervention—pushing state champions to compete in global markets and culling those who failed to become competitive—was central to Northeast Asia’s economic rise.

Today, as the United States spends tens of billions on industrial policy drives in areas ranging from semiconductors to green energy technology, what lessons might Northeast Asia have to offer us today? Is industrial policy inherently more difficult for countries like the United States already operating at or near the technological frontier? 

Studwell was kind enough to offer some thoughts over e-mail reflecting on these questions and others. 

This Q&A has been very lightly edited for grammar and clarity. 

Behind Northeast Asia’s industrial policy success stories were highly capable public agencies like MITI that were able to foresee the direction of global markets and make coordinated bets with a long-term outlook. Yet Taiwan, Korea, and Japan were all poor countries at the time with far smaller stocks of highly-trained people. It’s hard to imagine many agencies in the U.S. today pulling off something so ambitious. What accounts for these countries’ especially high state capacity?

Those countries developed state capacity, but they had the advantage they were playing catch-up so they could learn the very similar development routes that all developed countries have taken. The East Asian countries also had relatively more educated populations, and elites, at the outset than was the case on a continent like Africa. So having sufficient initial human capital, committed political leadership and a regional leader to learn from—Japan—were all important. If we then think of developed countries presently toying with the idea of renewed industrial policy—the UK or the US—the challenge is more difficult for two reasons. First, the bets you take in industrial policy are riskier the closer to the technological frontier you get. It can be done—as with the Danish wind industry—but you do risk picking the wrong technology, as has happened in a number of countries in areas like biotechnology. The second problem is that countries like the UK and the US no longer have sufficient industrial policy expertise, not to mention banking sector expertise in long-term project finance, to deliver industrial policy. Protectionist industrial policy was born in the United States in the 19th century under leaders like Hamilton, and passed from there to countries like Germany, but those days are long gone—not even known to the average American politician.

In How Asia Works, you argue that industrial policy is appropriate for developing countries, but that rich countries should ultimately phase it out. A decade later, do you still generally believe this? 

As I wrote above, the risk-reward trade-off becomes less attractive as you get closer to the technological frontier. This is why it makes sense to move away from industrial policy. However, there are still moments when a government may decide the risk is acceptable in order to nurture a sector of the economy that private enterprise won't develop unaided. Indeed, there is always some of this going on—just look at what NIH does in the US.

FDI can be a double-edged sword. Nippon Steel’s investment in POSCO helped South Korea climb the value chain, but its deal with Perwaja left Malaysia reliant on Japanese steel expertise. [Note: Perwaja was a state-led steel venture in Malaysia which Studwell argues floundered because of a failure to produce for global markets.] How big a role does company culture play here and can policy make a difference in that? Does the calculus change as a country develops further?

Company culture and leadership are everything. POSCO was established as a learning project that started with purely mechanical processes. Perwaja was a fantasy leap into an untested technology whereby the analogy would be to send your primary school child to college to see how much they learn. Not much is the answer. Effective policy makes a difference in the sense that it recognises that learning is step by step; you don't leapfrog to knowing everything. 

How important are foreign experts, immigrants, and returning diaspora in helping countries execute industrial policy effectively? It seems key to the story of Taiwan's semiconductor industry.

All of these are vectors of knowledge. Effective developmental states learn any way that they can. Taiwan was particularly lucky because so many technical personnel had gone to work on the west coast of the US. They came back not just with technology, but with processes and in some cases with smarter ideas about how to do things than they saw in the US. 

In your book, you write that Taiwan was the weakest of the three Northeast Asian success stories (on industrial policy, at least), while Malaysia was grouped with Southeast Asian failures. The former seems to have caught up with its peer group. Malaysia, taking a more open, neoliberal approach to FDI, is by no means rich, but it's roughly now on par with Poland. Did either of these surprise you?

Taiwan did not take the classic industrial policy route, but as a small country, it just needed to get one scale business right, which it did with semiconductors. TSMC turned out to be probably the best company in the whole of Asia and that sets the tone for the most important sector in the Taiwanese economy. Malaysia failed in most areas of industrial policy for indigenous firms but really applied itself to making the country a good home for FDI-based manufacturing, especially electronics. When you compliment Malaysia for its GDP per capita, however, you have to remember they have a lot of oil and gas. 

I understand you’re writing a book on African development now. Are there any recent cases you’ve seen there or elsewhere of countries doing industrial policy the right way?

Africa has one tiny country, the island of Mauritius, that ran industrial policy very effectively and became prosperous. Over the past couple of decades, the second most populous African state—Ethiopia—copied agricultural policy from East Asia and moved on to industrial policy. The results were extremely gratifying until a recent civil war. We must wait to see if Ethiopia gets back on track. I suspect it will because the foundations that have been laid are robust. More generally, however, Africa is a very different story to East Asia for reasons that will be explained in [my forthcoming] book. Perhaps the biggest difference is demographics.

The list of reasons to support radical energy abundance is very long. Let’s add another to the list: avoiding what I’ll call “energy gentrification.”

The popular conception of gentrification in the context of housing sounds something like this: developers put up new “luxury” units in low-income neighborhoods, causing prices to rise and involuntarily driving locals out of their long-time neighborhoods. This is why while so much of NIMBYism is resistance to development by parochial self-interest, much of it is also a genuine concern for low-income people.1

Yet we know that this popular conception is wrong. Housing scarcity, rather than growth, displaces low-income families. Picture an influx of wealthy outsiders buying and fixing up dilapidated brownstones; incumbent homeowners sell and earn a windfall, but without an outlet like new construction to absorb this new demand, existing renters or prospective buyers with fewer resources are trapped in a bidding war they stand no chance of winning. There are no completely independent affordable and high-end housing markets; all housing markets are connected. That’s why even building so-called “luxury” housing quickly leads to more abundant housing options for low-income renters through the filtering effect.

Kate Pennington, an economist at the Census Bureau, has a clever working paper looking at the effect of new construction in San Francisco, taking advantage of random new construction induced by building fires. She finds that rather than pushing existing residents out into poorer neighborhoods, new construction actually reduces displacement and lowers nearby rents. Change, not stasis, ends up being the source of stability and security.

Why the long housing windup for a blog nominally about energy? Rapid advances in artificial intelligence could quickly and substantially increase energy demand in the United States, both through the explosive growth in data centers directly used to train and run AI models, and through accelerated economic growth. The Boston Consulting Group estimates that data centers’ share of U.S. electricity demand will triple by 2030:

But these are just the mere short-term, first-order effects. What if AI, as many believe, turns out to be a game-changer for economic growth in the medium to long term? I’m not referring to wild projections of explosive growth on the order of 20 or 30% per year (though you should check out this Asterisk Mag debate on this possibility). But say AI sparks a productivity boom that brings growth up to 5% for a sustained period.

In such a scenario, the race between energy efficiency and economic growth that has kept U.S. electricity use relatively flat since the mid-2000s would likely end, with growth emerging victorious. Add this to the challenge of electrifying transportation and other fossil fuel-guzzling sectors and we could see huge growth in absolute electricity demand.

We should expect this potential, massive growth in demand to vary widely between regions. The growth in huge, energy-hungry data centers will accelerate in exurbs or suburbs on the outskirts of larger metro regions featuring affordable, buildable land. Needing only a handful of security staff and maybe a technician or two on-site, these facilities won’t be especially tied to clusters of skilled labor. The landing spots for many of these new data centers intersect significantly with emerging remote work migration patterns, potentially putting these two trends on a collision course.

Here’s the problem: as with housing, we have made it very difficult to quickly build new energy infrastructure in the United States. And that’s in a world of steady-but-slow 2% GDP growth year after year, where meeting growing energy needs rarely requires big expansions of energy infrastructure. In a future of substantially faster growth, the costs of highly inelastic electricity supplies may start to compound quickly.

This raises not just the possibility of escalating data center NIMBYism, but also the prospect of a wave of energy cost-induced displacement. In a housing market in which government has barred expansions in supply, low-income and working-class households are pitted against wealthier newcomers with much deeper pockets. As we’ve covered, that’s a bidding war the former will lose. A similar scenario could play out if energy demand from AI proves truly explosive. Rather than facing off against high-income newcomers, families may find themselves bidding for a fixed energy supply against even richer corporations—either AI companies themselves or businesses whose productivity is turbo-charged by integrating AI into their processes. Tamping down the energy demand of the marginal crypto mining operation is one thing, but OpenAI? Not so fast.

As in housing, the answer is abundance. Just as building more housing is the key to ensuring families can stay in the neighborhoods they want, building an energy grid that can more easily scale will be necessary to do the same in a world of more rapid, AI-induced growth. Relentless R&D to further bend cost curves, radically expanding interregional transmission capacity, and yes, permitting reform, are all in order.

If AI is “eating the world,” then it is, indeed, time to build.

Washington is collectively aghast at the pending sale of U.S. Steel to its much larger Japanese rival, Nippon Steel. While Nippon quickly pledged to honor all labor agreements with U.S. Steel’s workers and is set to make major investments in the United States going forward, the backlash was nevertheless swift and vicious. Now, the transaction faces months of federal reviews that may ultimately scuttle the sale. 

The New York Times hails this sale as a major test of the Biden Administration’s reindustrialization efforts. Perhaps this is true. But I think this slightly misses the mark. A new bipartisan consensus has emerged that the United States ought to engage unapologetically in activist industrial policy to support strategically important sectors and supply chains. I think it’s more accurate to say that this moment represents a test of whether this consensus (of which I am a supporter!) is going to be based on cold, rigorous, and consequentialist assessments of the national interest or mere vibes and nostalgia. The initial reaction to U.S. Steel’s sale suggests we may be in for more of the latter than the former.

Are the “national security concerns” in the room with us right now?

Opponents of the deal overwhelmingly flagged it as a risk to national security. Indeed, the idea that a once-iconic industrial giant so symbolic of American power might fall into the hands of a foreign-domiciled corporation sounds a bit scary. But what precisely is the worry here? I’m having a hard time finding clear answers.

Japan is one of the United States’ closest allies. Tens of thousands of American troops are stationed across the country. Further, Japan has at least as much to fear from an ascendant and aggressive China as we do; they are as firmly in the “American bloc” as you can get.

If firms from a country as closely aligned with American foreign policy and strategic interests as Japan cannot invest here, who can? Deriving opponents’ first principles here is challenging. Even their most generous interpretation implies a far more radical and disruptive decoupling from global supply chains than anyone in the political mainstream has seriously lobbied for. 

Sen. Marco Rubio, an opponent of the sale, correctly points out in Newsweek that while Japanese and American leadership have been closely aligned for many decades, it is not inevitable that American and Japanese interests will always and everywhere align. It’s fair to go a step further than that. No American alliance, with Japan or otherwise, is guaranteed to last.

Rubio points to a Bureau of Industry and Security report from 2018 arguing that maintaining a strong domestic steel supply chain is critical for meeting the military’s demands. High-quality, low-cost steel is also an upstream component key to all sorts of other manufacturing sectors.1 While the staunchest market fundamentalists may outright reject industrial policy per se, for the rest of us, the case for supporting domestic steel is worth a hearing.

But here is an obvious question that nobody seems to be asking: If a strong, innovative, and productive domestic steel industry is an urgent national security imperative, is the current leadership of U.S. Steel even remotely up to this task?

A failure to innovate doomed U.S. Steel, not vague notions of “neoliberalism.”

The fall from grace of U.S. Steel is staggering by any measure. Today, U.S. Steel produces 14.5 million tons of steel per year, barely one-third more than it did in the year of its founding, 1902 (global production increased roughly one hundred-fold since then). Compared with its peak of 35 million tons in the 1950s, production is down nearly 60 percent. Having once produced a full two-thirds of American-made steel, the company now accounts for less than 20 percent of a shrinking market. Once far and away the world’s largest producer, the company now barely cracks the top 30.

As the Institute for Progress’ Brian Potter writes, the fall of U.S. Steel lies squarely in the company’s hostility to new technology and consistent inability to adopt competitors’ innovations:

“Arguably, US Steel has been a disappointment since the day it was formed. It was created as a fundamentally conservative reaction to the vicissitudes of the steel industry, and this guided its early years and shaped its culture. The economies of scale it achieved were never passed on to the consumer, and instead it used its size to bully other steelmakers and extract money from consumers. When this stopped working, it used its political influence to prevent consumers from buying low-cost foreign steel. Improving the efficiency of its operations was something it did as a last resort when left with no other options.”

“…being transformed into a lean, competitive company doesn’t seem to have changed its fundamental culture, a company that's content to be a follower, rather than a leader in technological development and pushing the industry forward.”

Potter plots one instance after another in which U.S. Steel either (at the height of its power) attempted to squash technologies that might disrupt the industry and threaten its status or (later on) simply failed to act on the threats it faced. 

As Potter points out, policymakers hardly stood by idly as the company slowly faded. Presidential administrations have granted the industry massive subsidies and protection from international competition for decades. In one episode, Potter writes, President Lyndon Johnson strong-armed European and Japanese producers to limit exports to the United States and give domestic steelmakers some breathing room to catch back up to the technological frontier. Rather than use this reprieve to tech-up and regain international competitiveness, steelmakers’ investment instead stagnated.

Congress has supplied decades of even more explicit support. The Buy America Act of 1933 directs public procurement of construction materials like steel to domestic producers, even if they offer higher prices than their peers. The Bipartisan Infrastructure Law strengthened these rules further in 2021, widening the scope of federally-funded projects required to use American-made steel. Further, domestic producers have long been protected by tariffs and import quotas, most recently the 25 percent “national security tariff” applied by the Trump Administration in 2018. As the Congressional Research Service reports, efforts to fight the scourge of affordable steel products have escalated further over the last 20 years:

Yet decades of guaranteed business from government at elevated prices, timely interventions when domestic producers were on the ropes, and successful lobbying for protective tariffs have evidently failed to create a thriving and innovative domestic steel sector. China produces more than ten times more steel annually than the United States. India and Japan are well ahead of American totals, too. An iconic brand like U.S. Steel remains but a minor player in global markets, apparently too vulnerable to survive on its own.

Isn’t there a lesson we should draw here?

Competition and industrial policy success go hand-in-hand.

Decades of American steel policy have failed to learn what I see as a fundamental lesson of industrial policy’s success stories: Any serious industrial policy to advance national or economic security interests must relentlessly chase productivity improvements and innovation, harnessing competition and global engagement to do so.

Libertarians may argue that the failure of American steel policy demonstrates the inevitable folly of industrial policy writ large. But this gets the story only half-right.

Take a look at the most recent rankings of the 50 largest steel producers in the world. A majority are Chinese-owned and nearly all are, on some level, creations of activist industrial policy, including Nippon Steel itself.

I recently re-read Joe Studwell’s, How Asia Works, which comes as close to an industrial policy Bible as one can find. In it, Studwell traces the rise of South Korean steel giant POSCO, setting its story against the epic state-led failure to create a domestic steel champion in Malaysia.

Studwell takes us on a tour of POSCO’s Pohang facility, where the firm scaled up its production and became the global powerhouse it is today. Set on the coast, the facility has convenient terminals to both directly receive inputs from its upstream suppliers and promptly export the finished product abroad. As Studwell notes, the Pohang plant exported between 30 and 40 percent of its output from the very beginning. POSCO received deep financial support from the state, and was heavily incentivized to export—competing against the world’s most advanced producers—rather than simply serve the smaller domestic market.

Through explicit support for exporters, Korea pushed POSCO to climb the value chain and gradually learn the most advanced production techniques. POSCO courted investment from larger Japanese firms, absorbing their technologies. But the firm insisted that Korean workers learn these new systems inside and out:

“…the firm showed a relentless application to the job of learning everything there is to know about a steel plant. During the first and second phases of its construction, POSCO management refused to employ the computerised control systems recommended by their Japanese consultants lest they did not fully understand the equipment they were buying.”

Contrast the experience of POSCO with Malaysian Prime Minister Mohamad Mahathir’s pet project, the Perwaja steel plant. The first mistake Studwell highlights is the plant’s location; whereas POSCO’s Pohang facility is right on the coast, designed from the outset to export, Perwaja was instead located across the Malaysian peninsula from both its suppliers and buyers. An effort to develop a particularly poor region of the country, Mahathir’s choice of location was political, not economic.

In contrast to POSCO, Malaysia’s Perwaja project faced no requirements to export. It, too, sought Japanese investment, but failed to learn the underlying technology and remained reliant on its foreign patrons. While POSCO is now one of the biggest steel producers in the world and a publicly traded independent company listed on the New York Stock Exchange, Perwaja remains a stark reminder of Malaysia’s failure to join the ranks of developed economies.2

The dueling tales of these two plants stand in for Studwell’s broader thesis: that manufacturing policy which insisted on producers exporting (learning from international competition) and which cut off support for those who didn’t succeed, was what differentiated successful industrialization drives in East Asia from failures.

Studwell’s book is taken largely as a set of lessons for developing countries seeking to climb the value chain, but the lessons for American steel and our new era of industrial policy should be hard to miss. State support for firms in key sectors may be necessary at least for a time, but it is hardly sufficient for producing world-leading firms.

I like to think about industrial policy in the way international relations specialists view economic sanctions. When coupled with tangible, achievable policy goals, sanctioning a misbehaving regime stands a decent chance of working, causing the targeted regime to either make concessions or fall. But at least two obstacles stand in the way. First, the sanctioning country may renege on the conditionality of its sanctions. The sanctioned country may concede to giving up its missile program, but the sanctioning country may try to push its luck, promising to maintain its policy until the sanctioned country gives up all its nefarious activities.

Second, a defiant regime may simply withstand the initial period in which the sanctions bite hardest. Domestic supply chains re-adjust. The hard, initial costs start to fade. The country is left poorer, but the regime survives and can rally domestic support in opposition to the sanctioning country trying to impose its will.

Similarly, industrial policy and state support can indeed forge world-leading producers at the bleeding edge. In fact, that’s how most of today’s global steel powerhouses earned their status. But support won’t work unless producers are forced, as Studwell writes, to compete in global markets, learn from their more advanced rivals, and ultimately lose state support if they fail.

One administration after the other was happy to protect U.S. Steel when it was most vulnerable and direct public procurement officers its way even when it cost taxpayers dearly. But this assistance did not come with the condition that the firm learn, upgrade, or innovate. It did little more than subsidize the culture of complacency that ended up being U.S. Steel’s undoing.

Sale or no sale, the challenge of building a productive, internationally competitive domestic steel sector remains no less daunting. And as this new bipartisan consensus further explores questions of economic resiliency and supply chain security, we will need serious ideas beyond nostalgia and one-off security reviews. If we are going to do industrial policy, we have some learning and innovating to do ourselves.

What makes some cities resilient to sudden economic shocks and others not?

Policymakers across the world are chasing varying notions of economic resilience, spurred on by pandemic-era supply chain woes and a deeper reckoning with deindustrialization in developed economies. National governments are revisiting old industrial policy and protectionist frameworks, seeking to duplicate key supply chain nodes at home or within allied trade blocs. Tariffs, local content requirements, and massive corporate subsidy schemes are back in vogue.

Yet for mayors or city-level leaders looking to make their region’s economy more resilient to trade or geopolitical shocks, there isn’t a readymade playbook. Economic diversification is an obvious answer, but prescribing this is a bit like just restating the ailment. Coordinating public investments in emerging sectors that fit with a city’s existing strengths might make sense, but this also requires a high degree of state capacity to execute well.

A recent working paper from economists Luisa Gagliardi, Enrico Moretti, and Michael Serafinelli suggests that maybe the mere accumulation of talent is the best insurance policy a city can buy.

Gagliardi, Moretti, and Serafinelli (henceforth GMS) examine nearly two thousand regional economies across six countries—France, Germany, Great Britain, Italy, Japan, and the United States—tracking the rise and fall of manufacturing hubs over the late 20th century.

GMS identifies the year in which manufacturing employment peaked in each country. The authors then take the top third of each country’s cities in manufacturing employment share in that year and designate those places as manufacturing hubs, whose trajectories they then trace through the subsequent era of deindustrialization.

On average, manufacturing hubs were hit quite hard by deindustrialization, which began as early as the 1970s in the United States and Britain and as late as the 1990s in Japan and Germany. GMS finds that a one standard deviation increase in a city’s manufacturing employment share corresponds with overall employment growth 2.71 percent lower in each subsequent decade. As you would expect, a shock to Western manufacturing sectors wreaked the most long-term havoc on those regions specializing in manufacturing.

But not all former manufacturing hubs went the way of Detroit, falling victim to a doom loop of population decline and perma-stagnation. Instead, GMS finds that 34 percent of former manufacturing hubs fully recovered, defined as ultimately returning to or exceeding national rates of employment growth.

The rate of manufacturing hub recovery varied quite dramatically between countries: in Germany, 47 percent of manufacturing hubs bounced back from the shock of deindustrialization to return to Germany’s national growth trajectory. In the United States, just 17 percent rebounded, the lowest rate of the countries examined. Narrowing the sample further to only those cities in the top 20 percent of manufacturing employment share in each country reveals an even grimmer picture for American hubs. Whereas 29 percent of all top quintile manufacturing cities recovered across the countries examined, only six percent of top quintile hubs in the U.S. ultimately bounced back. In other words, those U.S. cities most heavily specializing in manufacturing at the country’s industrial peak around 1970 almost universally failed to catch back up to the national growth trend. This helps explain in part why American economic geography is so strongly bifurcated between a handful of superstar metro areas and the rest of the country.

The authors identify a strong and accelerating relationship between cities’ initial levels of human capital at their country’s industrial peak and subsequent economic performance. Across all cities (not just manufacturing hubs), the authors calculate that a one standard deviation increase in the share of workers with a college degree resulted in 9.1 percent faster employment growth per decade.

What is notable here is how little college completion rates seemed to matter before deindustrialization. Cities with high college shares grew no faster than cities with the lowest rates in the decades before manufacturing began to decline in each country (denoted as year zero in the chart below). Only after deindustrialization began did the initial presence of college graduates begin to matter, and it grew more important as time went on.

There is a concern that initial levels of human capital and a city’s propensity to rebound from deindustrialization may be endogenous—subject to reverse causality. Perhaps, for instance, before the decline of manufacturing began, college graduates could foresee which cities were likeliest to grow and moved there. GMS allays these concerns by using cities’ distance from older nearby colleges—whose locations are unlikely to systematically correlate with economic shocks to manufacturing decades later—as an instrument for cities’ college graduate shares at manufacturing’s peak.1

Using this approach, GMS finds that in historic manufacturing hubs, a one percentage point increase in the share of workers with a college degree before deindustrialization began is associated with a 2.4 percentage point increase in subsequent per-decade employment growth. In other words, the presence of college graduates is what separated those manufacturing hubs which recovered from deindustrialization from those that did not.

Why did American manufacturing hubs perform so poorly? One likely reason is that the college degree gap between manufacturing hubs and non-hubs was greatest in the United States than in other industrialized peers. Their initial, (relatively) low levels of human capital made them uniquely vulnerable to trade shocks and less adaptable than peers in other countries.

Talent = Resilience

Cities can hedge against harmful economic shocks by being relentless about attracting and cultivating talent, a task no less important in periods of prosperity. Those manufacturing hubs which had a ready supply of smart and flexible skilled workers were quickly able to shift to the next thing as soon as heavy industry packed up and moved overseas. They remained attractive places to invest and the non-linear returns to clustered talent gave them a distinct advantage in sprouting new ideas and new firms. Meanwhile, those cities whose workforces and policy environments were entirely focused on meeting the needs of industrial mega-plants found themselves unable to adapt.

At manufacturing’s peak, the presence of college graduates didn’t appear to matter much to cities’ fates; indeed, there is no real relationship between employment growth and the share of local workers with a degree pre-deindustrialization. Yet while this reserve of human capital may not have been immediately valuable, it served as a highly effective insurance policy for those cities who built it.

It is important to remember that the measure used here, local college degree attainment, is merely a quick, available proxy for talent. You do not have to believe that more local students should be attending traditional colleges to understand the link between human capital and your city’s resilience to economic shocks. A 21st century agenda to better cultivate local talent might send students through a diverse array of alternative pathways, fellowships, or incubators. At the same time, it might also include measures like the Heartland Visa, which would attract skilled workers and entrepreneurs from abroad to pockets of the country suffering from economic and demographic decline, perhaps independent of their educational credentials.

Massive economic disruptions akin to the China Shock will happen again. They may just topple some of today’s superstar cities. Their shape and extent will be hard to predict, and explicit planning to avoid them can only go so far. Just as in the past, those places which adapt to these shocks will be those who attract the best and brightest, while those who rest on their laurels will watch them crumble beneath.

The Wizards and Capitals are fleeing for Virginia. The federal workforce remains largely remote, depriving downtown D.C. of business and tax revenue. Our eternally-flailing Metro system faces a $750 million deficit that could result in dramatic service cuts. In short, the vibes in D.C. are grim.

Is the District in a mere “vibecession” of sorts, or do the data really justify the dour sentiment of locals? As we near the end of 2023, I examine the health of the nation’s capital on key dimensions including demographics, public safety, and school quality.

1. DC has faced one of the steepest population losses among major cities post-pandemic.

After a decade of rapid population growth in which the District added more than 100,000 new residents, the pandemic and its aftermath took a substantial portion of that growth back. As a share of its estimated population on April 1, 2020, DC experienced the sixth-largest decline in population among the 25 largest cities.

As I’ve written of major cities, young families appeared to be leading the exodus. Between April 2020 and July 2022, D.C.’s under-five population fell by more than 3,200, or about eight percent. On this measure, D.C. is at least in line with some nearby peers, as declines in the number of young kids were of similar magnitudes in Philadelphia, New York, and Boston.

2. Homicide and violent crime are way up in D.C., in contrast to national trends.

It’s no secret that the District is experiencing a steep rise in gun violence, but the divergence between Washington and other major cities is striking.

The latest data from the city’s police department shows a more than one-third increase in homicides year-over-year, a two-thirds increase in robberies, and a near-doubling of carjackings.

How does this compare to the national picture? An association of police chiefs from the largest cities and counties in the U.S. and Canada collates data into uniform categories on a quarterly basis. In the first three quarters of 2023, this data showed D.C. had one of the highest year-over-year increase in homicide in the country.

On broader measures of violent crime, the picture in D.C. looks even worse. This same dataset shows a nearly 40 percent year-over-year increase in violent crime, far outpacing increases anywhere else.

District leaders made enormous progress tackling the city’s violent crime problem in the 2000s and 2010s. In two major categories of violent crime—sexual assault and assault with a dangerous weapon—there has been no major spike post-pandemic. But on homicide and robbery, the city has ceded 20 years of progress in a very short period.

Now, the city’s crime picture still looks nothing like it did in the 1990s. Indeed, increases in both violent and property crime have been concentrated in a few categories, perhaps suggesting they are relatively tractable problems. Still, the D.C. of the 1990s was also the murder capital of America. It’s little consolation that today’s D.C. compares favorably.

3. D.C.’s public schools: Truancy has skyrocketed and math scores have not recovered from pandemic-era declines.

On two measures of school performance, D.C. school quality has eroded in the aftermath of the pandemic.

First, the share of District students scoring “proficient” on PARCC and MSAA standardized tests has fallen substantially since 2019. While rates of proficiency in English/Language Arts are not far from pre-pandemic rates, the district has again given back years of progress on math proficiency.

The District’s schools have also made headlines for jaw-dropping rates of truancy. Already having a higher rate of chronic absenteeism that any state pre-pandemic, chronic absenteeism rose in 2022 to nearly half of all students (48 percent). While these numbers did improve in the most recent school year, they remain far above where the District (or any other state) was in the years before the pandemic.

4. The recovery in Metro ridership is in line with peer cities’ systems.

It’s been a wild few years for the Metro. In October 2021, WMATA had to yank all its Kawasaki 7000-series train cars from service—60 percent of its fleet—overnight after a derailment in Arlington. Only late this summer did WMATA finally return these cars to service in full, after more than a year of delays and hiccups.

WMATA now faces an impending $750 million budget shortfall. Its governing board is weighing drastic cost-cutting measures, including ending all service after 10 p.m. and closing ten stations.

Despite all this, the Metro’s ridership recovery has been surprisingly strong in comparison to peer systems’ across the country. Well ahead of the Bay Area’s BART system since early 2022, WMATA’s return rate passed the Chicago Transit Authority’s this summer, too. Average daily entrances in November were 60 percent of the November 2019 average.1 While New York City’s MTA remains well ahead of WMATA, DC’s ridership looks more impressive when you consider D.C. remains the top city in America for remote work.

While the below data include only a handful of the largest peer systems for comparison, anecdotes from smaller systems look no better. Rail ridership in Atlanta’s MARTA system remains nearly 60 percent below pre-pandemic levels as of September. In Los Angeles, Metro ridership is 25 percent below 2019 levels and 40 percent below 2018. Philadelphia’s SEPTA system is hovering around 60 percent of pre-pandemic service, in-line with WMATA as well.

5. New housing construction in D.C. continues to boom.

Now for some truly good news.

D.C. has continued to build new housing at a torrid pace. More than 7,700 new units were permitted last year, the District’s highest total since 1965. The city’s audacious goal of 36,000 new units by 2025, set in 2019, is well within reach. New housing growth has been one of the District’s greatest assets during its transformation of the last 20 years. To remain a competitive, affordable destination for skilled young people, it simply must be allowed to keep growing.

Conclusion

The District faces real challenges that should not be underestimated. The post-pandemic crime wave threatens to reverse decades of progress on public safety. Allowing this problem to fester will start to seriously affect families’ decision to live within the District. But there is also reason to be optimistic. Developers continue to build new housing at a near-record pace. People are indeed returning to the oft-maligned Metro. D.C. has overcome far worse before, and with serious leadership, may well do so again.

What if I told you there is a low-cost, highly effective tool for seeding new clusters of high-growth startups that your city is almost certainly ignoring?

This may be surprising, given that state and local governments spend tens of billions of dollars each year doling out economic development incentives. But these deals are largely wasteful or even outright harmful. Consider that most of this money is spent “chasing smokestacks,” attempting to lure established businesses in a zero-sum subsidy war. Very little is devoted to a far more important driver of cities’ economic fates: attracting and retaining world-class talent.

The tool I’m referring to is the Global Entrepreneur-in-Residence program, or Global EIR. Global EIR is a little-known wrinkle in the H-1B visa program that allows universities and non-profit research organizations to sponsor local entrepreneurs and avoid the annual H-1B cap.

The number of H-1B visas available to the private sector each year has been capped at 85,000 since 2006. While the national economy and the number of applicants have both grown substantially since then, our primary tool for recruiting skilled workers to the U.S. has not expanded accordingly. Randomly allocated, the H-1B looks much more like a lottery ticket with ever-dwindling odds than a rational way for a global power to attract talent.

For entrepreneurs or would-be founders, the issues with our immigration system are further compounded by the fact that the U.S. lacks a startup visa. While our universities are very open to international students, retaining them after graduation, either to work at American companies or to build their own firms here, is a growing challenge.

Global EIR is an elegant workaround of these roadblocks. Here’s how it works: universities identify and vet budding entrepreneurs, likely students on F-1 visas or recent graduates on OPT. Schools can hire part-time or accept volunteer commitments from applicants, often taking the form of student mentorship roles. Universities, as non-profits categorically exempt from the H-1B cap, serve as their sponsoring entities for the visa. So long as their company (both founders and key executives can use Global EIR) pay the applicant their relevant “prevailing wage,” they have certain access to a visa and don’t have to take their chances with the broken, clunky lottery system. Entrepreneurs then have a smoother pathway to take advantage of more permanent options to stay in the U.S. like the O-1 visa or employment-based green cards.

As the Center for Growth and Opportunity’s Josh Smith () notes, local leaders can do this all on their own:

A key advantage of the GEIR route is that today’s immigration rules allow the establishment of a GEIR. Universities don’t have to apply to a program or get permission from any government agencies. The only people who can say ‘no’ are the university officials who choose not to create one at their university. Unlike other immigration reforms, this one doesn’t depend on what lawmakers do in Washington DC. It’s something that universities can and should do today.

Yet sadly, Josh goes on to write, only seven—yes, seven—out of the country’s 2,500 four-year universities had a Global EIR program set up as of 2021.

Your city is almost certainly leaving free growth—and the revenue that comes with it—on the table.

The economic and fiscal benefits to setting up a Global EIR program and thereby keeping dozens or hundreds of high-potential entrepreneurs in your backyard should be obvious. We know, for instance, that the take-off of superstar cities is often highly contingent, shaped by the location decisions of one or two firms who end up being globally significant. But startup incubators, as Global EIR programs effectively are, do not need to attract the next Microsoft to fundamentally change their host community’s economic trajectory. Clusters of highly-skilled technical people, and of entrepreneurs in particular, build on themselves, attracting further investment and talent.

For universities considering setting up a Global EIR program for themselves, the hard part is in many ways already taken care of. By construction, Global EIR selects for entrepreneurs who are already raising funds, not mere hopefuls with a plan on paper.

While spinning up a Global EIR program is relatively cheap (if somewhat complicated), it’s easy to imagine how universities could turn it into a significant moneymaker, perhaps taking a small share of equity or IP in participants’ companies. For state or local governments, paying the operational costs of EIR programs run by local universities could very well be an investment that pays for itself over time.

Cities can use Washington’s dysfunction to their advantage.

While comprehensive immigration reform from Congress may be hard to imagine right now, cities frustrated by their inability to retain the students they train after graduation or spin up clusters of startups can take action now.

Between actively-enrolled international students and recent graduates on OPT, there were more than one million international students studying and working across the U.S. last year. Many thousands of these students are potential entrepreneurs. As the odds of winning the H-1B lottery continues to fall and other visa options buckle under the weight of bureaucratic delays, cities can step up with their own alternative, at least for startup founders.

Cities are not helpless in the face of a broken federal immigration system. They are not destined to sit on the sidelines bemoaning the state of national politics, or joining another bidding war for the next Foxconn boondoggle. Cities can rewrite their stale economic development playbooks for the better by refocusing on what truly drives their growth: people and their ideas. Activating the latent talents of entrepreneurs with a slate of new Global EIR programs is a great place to start.

The backlash to self-driving cars is shifting into fifth gear. Last month, the California Department of Motor Vehicles suspended Cruise’s autonomous taxis’ license to operate after a series of incidents and credible allegations the company withheld important evidence on at least one accident from state investigators. 

In a statement celebrating the suspension, Safe Street Rebel, a San Francisco anti-car activist group that drew national attention for purposely disabling robotaxis across the city over the last few months, say they are far from done. Waymo, Cruise’s chief competitor in the city, is next. Their mission is not about the errant practices of one dishonest company or a plea for higher, attainable safety standards, but a campaign for permanent prohibition, full stop. And in San Francisco, at least, they might just win. 

Safe Street Rebel is on the more extreme end of anti-car activism common in urbanist circles not just in San Francisco, but in practically every major American city. The organization’s homepage reads:

The last half-century has been a failed experiment with car dominance. They  bankrupt our cities, ruin our environment, and force working people to sacrifice an unacceptable amount of their income to pay for basic transportation. It is time to end car dependence and rethink our streets  around public transit, walking and bikes.

Indeed, they have a point. The 20th century saw urban highways cut through and destroy vibrant communities. Urban rail networks shriveled as the country suburbanized. Car dependency is deeply rooted in land use rules, too, as new housing and commercial space are typically required to purchase and set aside parking spaces for users and residents. In a display of just how ingrained car culture is in American politics, even modest congestion pricing in Manhattan–perhaps the community where such a policy makes the most sense on both economic and environmental grounds–comes with a vicious years-long fight. In many ways, car dependency is the original sin of American urban planning.

To lay my cards on the table: I’m an urbanist myself who desperately wants to see more bike lanes, bus rapid transit, and car-free streets where I live in Washington D.C. I do not own a car and don’t imagine that I will anytime soon.

But would the efforts of fervent anti-car activists to ban autonomous vehicles actually ameliorate the problems caused by car dependency, or make us any safer?

Let's start with safety, which is ostensibly the main issue in the debate on self-driving cars. Automobile accidents kill nearly 40,000 people per year in the United States. In 2021, nearly 8,000 of these were pedestrians. It is worth noting that the U.S. is a major outlier on road safety. Our rate of automobile fatalities is twice the average of a group of 28 peer countries, according to a CDC analysis, at 11 per 100,000 people. 

Even at this early stage of development, autonomous vehicles already appear to rival human drivers. A recent analysis from insurer SwissRe of nearly four million miles of driving data from fully autonomous Waymo vehicles found they accounted for 76 percent less property damage-causing accidents than human drivers. Despite some high-profile incidents in San Francisco and elsewhere, driverless cars may already be safer in general. (Notably, Cruise’s safety record has been consistently and substantially worse than Waymo’s.) But even if you don’t yet buy this claim, self-driving cars are almost certainly going to improve with further testing and more powerful internal models. Their human rivals, meanwhile, are not due for a software upgrade anytime soon. If safe street groups gain more traction against self-driving cars than against human-operated ones—which at this point seems very likely—their efforts may actually make our streets more dangerous.

Perhaps you’re playing the long game here and think that the growth in self-driving cars on our streets will further tilt the political playing field towards auto-centric infrastructure priorities. That’s an understandable concern given how entrenched car-centric politics have proven to be. But consider this: even an extreme anti-car group like Safe Street Rebel is proudly in alliance with San Francisco’s taxi drivers’ association. Existing taxi drivers obviously stand to lose from the transition to autonomous vehicles, having spent up to $250,000 on licenses to operate. Might salvaging the taxi lobby as a viable political force also serve to entrench auto infrastructure? Not even the most ardent believers can avoid the pitfalls of coalitional politics. 

Would banning self-driving cars even be a win against pollution and climate change? Unlikely: the fleets of both Cruise and Waymo are 100% electric, which neither the general driving public nor San Francisco’s taxis can claim themselves. Electric cars do not completely eliminate the air pollution from driving (their higher weight stirs up more dust and wears down tires faster), but evidence so far suggests they are a major improvement over gas-powered rivals.

Critically, bans on self-driving cars do nothing to make public transit alternatives any better. On the margin, prohibiting people from taking advantage of cheap AV rides might protect some of public transit’s market share. However, that transit needs such protection speaks more to its failure to deliver quality and reliable service than it does to the supposed evils of self-driving cars. Urban transit agencies will be no less dysfunctional with this new source of competition neutralized. The cost crisis afflicting American infrastructure–and urban metro expansions in particular–is still crippling and inexcusable. 

In other words, banning self-driving cars fails even on urbanist anti-car activists’ own terms. The damage done to our cities by an over-reliance on cars is deep and will take decades to root out. Doing so will require radical liberalization of land use to allow for denser development; a national effort to tackle the causes of spiraling infrastructure costs; and serious conversations about how we sustainably fund transit systems across the country. Stamping out AVs now, just as they are beginning to near viability for widespread adoption, does nothing on these fronts. 

In the near-term, self-driving vehicles have the potential to save many thousands of lives, reduce carbon emissions, and provide people with another affordable option to meet their unique transportation needs. They will make cities function better in many of the ways that matter most to urbanists like myself. In many neighborhoods in which robust public transit may never quite pencil out, self-driving cars can help solve systems’ last-mile problem, improving job access for those living in distressed communities. 

AVs may not deliver us from our cities’ most consequential mistakes, but they will quantifiably improve urban life along many dimensions. That sounds like “good news” to me.

Why do so many feel like American bureaucracy is deeply broken?

Noah Smith argues that our public agencies are underfunded and understaffed, lacking the capabilities to swiftly promulgate rules or make the decisions we ask of them. Others argue that the explosion of veto points has systematically empowered parochial private interests at the expense of the public good, as Will Rinehart catalogs. Ezra Klein’s famous “everything bagel liberalism” hypothesis suggests bureaucracies become dysfunctional when they lose focus, trying to accomplish everything, everywhere, all at once. 

Each of these three theories is true to an extent, but there is an underlying commonality being overlooked: the way in which we hobble public priorities–even narrow ones–from the very start. At times, the quickest way to doom a project to be over-budget and long-delayed is to make it an urgent public priority. Not because government is bad at executing its aims per se, but because added requirements and expectations that come with disbursing public money do not apply to equivalent actions taken by the private sector. Unfortunately, it is often those most enthusiastic about the benefits of government investments and interventions that support these rules, undermining the quality of government for all of us. 

Perhaps the most glaring examples of late have come from NEPA, or the National Environmental Policy Act. NEPA requires federal agencies to assess the environmental impact of any federal action, from buying office furniture to green-lighting oil and gas drilling. If a federal action is expected to have a significant environmental impact, agencies are required to write a comprehensive Environmental Impact Statement, which can take years and cost millions of public and private dollars. NEPA is not itself a set of environmental standards, merely a mandate to produce a study. Federal agencies finding that their intended action will cause environmental degradation are still free to go through with it.

The delays and costs associated with NEPA often halt worthwhile projects or send the message that building something that might activate NEPA rules is not even worth trying. Courts can rule agencies’ NEPA reviews insufficient, creating major risk and uncertainty for investments to which it might apply. In one famous example, an eight-year NEPA review held up an offshore wind project off Cape Cod that would have offset 880,000 tonnes of CO² emitted per year. But that was just the start. Activists using every tool at their disposal filed one bad faith lawsuit after another, arguing the review was insufficient. After 16 years in limbo, Cape Wind gave up and abandoned the project entirely, leaving Massachusetts reliant on fossil fuels for more than three-quarters of its electricity.

As the Commerce Department doles out CHIPS and Science Act subsidies to semiconductor manufacturers over the coming months, NEPA lurks in the background as a potentially devastating threat to the timely construction of new factories subsidized by the $50 billion pot set aside by Congress. Yet here’s the rub: NEPA only applies here because the federal government is subsidizing chip fabs. Ironically, only because Congress came together in bipartisan fashion to make a generational investment deemed critical to economic and national security do new fabs need to undergo this burdensome, costly step. If the semiconductor industry was not overwhelmingly concentrated on Taiwan–and therefore not an urgent security issue–TSMC or another chip giant seeking to expand in the United States would be free from such a review. Of course, we have no reason to believe the environmental impact of a privately-funded chip plant expansion would be any different than that of one receiving public subsidies. 

“Buy America” rules are an even more frustrating example. Public infrastructure projects or investments are required to use mostly American-made inputs. In the case of projects like bridges, tunnels, and roads, that means American-made steel, iron, and cement, regardless of whether they can be found at lower cost from other sources. But such rules even extend to government procurement of IT infrastructure, raising its cost by 25 percent. According to one estimate, Buy America rules alone raise the cost of buying a metro car for local transit systems by one-third. Government makes these investments in the public interest. However, if a company wanted to build its own internal transit system on a sprawling corporate campus, for example, no such Buy America rules apply. Only when new infrastructure is built because our elected governments deem it to be socially important and in the public interest are they forced to use more expensive equipment and materials. 

Proponents of Buy America rules often argue that these requirements are a critical source of demand for American manufacturing. Yet versions of Buy America requirements have been on the books since the 1930s and they have done little if anything to slow the decline of manufacturing as a share of the national labor market. The decline of such jobs is a global phenomenon that no government procurement rules can fix. Despite their evident failure to address this problem and widespread acknowledgement that our infrastructure costs are far too high, these rules are only getting stronger under the Biden administration.

Supporters of a more muscular, capable state undermine their own goals when they support policies like Buy America, NEPA, or the litany of other burdensome rules that only apply to government actions or projects. Rather than showing the public that government can in fact do good for the average citizen, they instead doom government to appear hopelessly incompetent when compared with the private sector. They may do as much to hobble the actual, on-the-ground effectiveness of government as any libertarian advocate of “starving the beast.” This is the paradox at the core of so many flaws of American bureaucracy. 

Government projects will inevitably have standards. The public does not want to see government pay employees starvation wages, nor does it want tax dollars buying materials made with slave labor in Xinjiang. But people do want better value out of public infrastructure funds. They do want new bridges, subways, and chip factories delivered on-time and at a decent price. As the United States enters into a period of mounting fiscal challenges, our spiraling infrastructure costs will become more salient and spark more outrage.

If those who believe in the power of government to improve people’s lives want to win over the public in an era of increasingly-intense fiscal trade-offs, they’ll have to recognize that their efforts to strengthen government’s impact are often self-defeating. Rather than asking how many problems a dollar of public money can solve, we should flip our presumption. Just as Ezra Klein warns against using narrow programs to solve all social ills, we should carefully scrutinize every additional requirement that gets in the way of our public priorities. We will continue to disagree over the proper size and scope of government, sure. But we should be able to agree that when democratic government mobilizes for reasonable aims, it should be allowed to deliver.

Suddenly, Japan was open to the world. In 1868, the conservative, isolationist shogunate that had ruled the country for two and a half centuries was overthrown by a group of enlightened revolutionaries seeking to re-empower the emperor and modernize the country. 

We call this event the Meiji Restoration and the regime it ushered in embarked on a swift and historic effort at industrialization. But it did not attempt to do this on its own. At first, the government employed a small army of Western experts, including former European bureaucrats and businessmen, to staff its government agencies and state-owned firms:

In the early years of Meiji, the new government hired from 300 to 600 foreign advisers in any year on a project contract basis, at considerable fiscal cost, to establish Western-style state-owned enterprises in railways, telegraphy, and silk reeling (Umetani, 1968). Some foreign advisers received salaries higher than that of the Japanese prime minister. Each project recruited a team of foreigners, usually of the same nationality, with various functions, who imported virtually all the materials required to create an exact replica of a foreign model (Kasuya, 2000). These were turnkey projects with a foreign director supervising his fellow countrymen and Japanese workers, with the Japanese side taking over operation and maintenance after project completion.1

Skilled Japanese workers, over time, would learn from these projects, allowing the country to climb up the value chain and engage in more sophisticated manufacturing. Foreigners staffed key posts at high-tech firms in the early days of Japan’s opening up, but only at first. In mere decades, the country became a preeminent economic and military power. The rapid advancement of Japan’s military might and industrial capacity enabled the country, in 1905, to shock the world by defeating the Russian Empire in the Pacific. 

Japan’s new leadership was attempting to answer a question that both bureaucrats and industrialists alike have contemplated for generations: How does a country build industries in which it has no particular expertise? 

We call a strategy to answer this question “industrial policy.” Wrong answers to this question have resulted in humanitarian catastrophe. The best answers have coincided with historic reductions in hardship and poverty. 

For decades, industrial policy has largely been the domain of developing countries seeking catch-up growth through manufacturing, but as the world re-fractures into competing economic blocs, rich countries are getting back in the industrial policy game, too. Getting it right means drawing on the two centuries of lessons from past experiments. And if there is one lesson that stands above the rest, it is the indispensability of foreign expertise.

The temptation of industrial policy is nearly irresistible. After all, the goal from the start is merely to replicate the kind of manufacturing happening in many other countries across the globe, not necessarily to break new ground or invent new technologies. How hard could that be? Imitate, then innovate. 

Advocates have a broad toolkit for channeling credit, subsidies, and educational institutions towards building up the target industry of choice, perhaps auto manufacturing, steelmaking, or microchips. Industrial policy may be hated by most market fundamentalists, but successful cases harness the unrivaled discipline of markets. Countries like South Korea pushed top firms in nascent industries like auto manufacturing to compete (export) in global markets, where competition is most fierce. Kept alive by state support while facing technologically superior foreign firms, Korean conglomerates learned from their competition and were eventually able to stand on their own. 

Yet it is important to remember that for all the attention to the virtue of “export discipline,” perhaps deserved, industrial policy success stories are fundamentally about people learning how to do new things from other people. Industrial policy has always been linked to immigration policy or the utilizing of foreign expertise, which many advocates of the former fail to see today. Foreign experts from countries at the technological frontier have been indispensable to developing nations chasing catch-up growth, imparting the kind of deep, tacit knowledge one can only impart face-to-face. Those who try to summon advanced manufacturing sectors from the ether inevitably fail. As Joe Studwell writes of post-War China and India in the throes of strict autarky: “Each time firms wanted something new they had to, as the saying goes, reinvent the wheel.” 

Industrial take-off is hard under the best of circumstances. Starting from scratch is insanity.

For perhaps the most famous example of foreign-born scientists and inventors getting a nascent industry off the ground in a new country, look to American aerospace. 

As Brian Balkus (another Roots of Progress fellow) writes in Palladium, American efforts to steal information and documents on Nazi Germany’s rocket program bore little fruit. U.S. officials discovered that capturing technical information was insufficient for reverse-engineering and harnessing Germany’s breakthroughs for themselves. Instead, they needed the actual experts working on the program. 

This would evolve into Operation Paperclip, in which hundreds of German scientists were brought to the United States to build an American aerospace program. This turned out to be the talent coup of the century, forming the backbone of both NASA’s space program and an immensely productive aerospace manufacturing sector. 

“There are huge chunks of technical knowledge that cannot be acquired by reading texts,” Balkus writes, “And history has shown that it is only possible to access this knowledge through the humans who possess it.” Failure to understand the nature of information necessary to produce some of the world’s most sophisticated technologies–and how people actually learn it–has been behind many failures of industrial policy. Again, the example of aerospace is instructive: 

Aerospace is one of the deepest branches of humanity’s technological tree. It is a telling fact that more countries have produced a nuclear bomb than have mass-produced a jet engine. Recent history illustrates how hard it is to build these capabilities. China has provided an estimated $71 billion dollars in funding to the Commercial Aircraft Corporation of China to develop a competitor to Boeing and Airbus, and 15 years later it has barely begun to produce its first operational commercial airline even while its engines, avionics, and other core systems are imported. In Japan, Mitsubishi Heavy Industries spent 15 years and nearly $8 billion attempting to build Japan’s first passenger jet before killing the project in 2023, stating “We didn’t have engineers with such know-how, and it was rather [hard to find] any in Japan.”

Successfully replicating jet engine production domestically is a Herculean task in the most generous and supportive policy environment. Attempting to do it alone is nearly impossible. As the United States and our allies usher in a new era of industrial policy targeting a set of key technologies, we cannot forget this lesson.

Immigration policy is a central tool in great power competition. Let’s use it.

At my day job with the Economic Innovation Group, I recently co-wrote a proposal with my colleague Adam Ozimek for a Chipmaker’s Visa, an immigration reform to pair with Congress’ $52 billion investment in semiconductor manufacturing and research. As we point out, just as American-educated members of the Taiwanese diaspora were critical to the early days of TSMC and its accompanying supply chains, so too will workers and managers at top firms in Taiwan, South Korea, Japan, and elsewhere be key to rebuilding our domestic chipmaking capacity today. 

The United States remains the world’s top economic and political power, but there is increasing evidence that we are lagging behind our chief geopolitical adversary in key technology areas. Whether here, in allied countries, or in a combination of the two, the next decade of international economic policy will prioritize building resilience and bypassing Chinese Communist Party-controlled production chokepoints. There may even be further, expensive sequels to the CHIPS and Science Act for other industries Congress deems strategically-significant. 

Regardless of whether you favor the kind of heavy-handed approach embodied by the CHIPS or something more light-touch, the importance of talent with on-the-ground experience and practical expertise cannot be ignored. We need to get serious about attracting the world’s top young scientists, engineers, and innovators, particularly those who specialize in areas of technology in which we will most intensely compete with adversaries. Without them, even our boldest and most well-designed industrial policies may flounder.

Consider the aggressive new efforts by the Chinese government to bring home Western-trained scientists and engineers in areas deemed critical to geopolitical competition. The government is offering generous subsidies for purchasing homes and bonuses of half a million dollars or more for Chinese experts returning to work in critical industries. Meanwhile, those same scientists and engineers recruited by U.S. firms to build out our semiconductor, artificial intelligence, or materials science industries could face a wait time of nearly two decades. Being a much freer, more open society certainly gives us some leeway here, but if we allow our skilled immigration system to deteriorate further, we will lose more top scientists, engineers, and inventors to our adversaries. 

The proper financial tools to wield as we chase technological superiority over China in critical fields may still be up for debate. Ambitious industrial policy with extensive state intervention on behalf of our national economic interests–and claims about what those interests actually are–deserve a healthy dose of skepticism. But there is one thing that should not be up for debate: the United States needs major skilled immigration reform if we are to maintain or recapture global technological leadership. While the sprawling factories whose construction industrial policy aims to spur may be home to immensely complicated machinery and equipment, it is ultimately brains that make them work.

Think of a society as a kind of investment portfolio of talent.

In capitalist economies, people are free to choose their paths, guided by market prices. Over time, this type of “portfolio” has far out-earned all the rest in the form of explosive economic growth and innovation. Yet even in the freest societies, the interests and careers people pursue are inevitably influenced by mediating institutions, social pressures, and outright fads. We hope–but are never sure–that enough of our best and brightest are working on the kinds of problems that really matter.

On October 2, we were reminded that even a purportedly well-balanced portfolio can indeed be fragile, subject to “black swan” failures of oversight or groupthink. On that day, Katalin Kariko was awarded the Nobel Prize in medicine for path-breaking work on mRNA technology that yielded highly effective Covid-19 vaccines. Her and her collaborators’ research may just mark the beginning of a new era of treatments for diseases ranging from malaria to cancer, with the potential to save tens of millions of lives.

Yet Katalin Kariko was ignored by mainstream science. For three decades, she bounced from one lab to another with little job stability and meager pay. Today we have the mRNA vaccines that saved so many lives mostly because of her dogged refusal to take “no” for an answer. The National Institutes of Health, the body that dominates biomedical research grantmaking, simply did not take her ideas seriously for many years. The issue was not inadequate funding to go around, but insufficiently diverse perspectives in the grantmaking ecosystem. As Jason Crawford points out in a Twitter thread, there are few outside sources of funding in Kariko’s field beyond the NIH. He ends with this:

The failure of the NIH to spot the potential of mRNA technology is itself interesting, but even a well-designed central grantmaking body is inevitably going to miss some breakthroughs. Single institutions are always going to have internal cultures with biases or flaws. The key is to have a diverse body of grantmaking organizations with differing perspectives. 

In other words, we need to hedge our bets. 

The Kariko saga raises not only important questions about how we fund biomedical research in the United States, but how we allocate talent across society. For our smartest students, budding academics, or researchers, are there enough pathways and funding sources to prevent a Kariko in another critical field from being ignored? Few seem to be asking this question. 

Institutional pluralism and diversity of perspective aren’t exactly compelling rallying cries, but there is potentially much more at stake here than having an effective Covid-19 vaccine. We ought to be more concerned that many dominant institutions in American society–particularly those through which nearly all of our most talented young people pass–tend to think the same way, whether on politics, scientific claims, or moral values. 

It is worth dwelling upon a society that had extremely little institutional diversity, instead betting nearly all its most valuable human capital on a system that ended up being immensely hostile both to freedom and technological progress: imperial China. 

How Chinese civil service exams squashed dissent–and an industrial revolution

The Keju civil service examination system’s purpose was first and foremost political control. Established under the Sui Dynasty in the sixth century, Keju exams would serve as the primary means of selecting government bureaucrats through the final days of the Qing Dynasty in the early 20th century. In a fascinating new book, The Rise and Fall of the EAST, MIT economic historian Yasheng Huang traces the effects of the Keju exam system on Chinese autocracy and attitudes towards change and technological progress through our current era.

The exams, Huang argues, were the fairest, most accessible means of upward mobility in imperial China. By pioneering double-blind grading schemes for local and provincial exams and cracking down on cheating, this system ensured that talented young Chinese subjects knew that if they worked hard enough, they could ascend the ladder of the Chinese state, enriching themselves and their families in the process. Huang writes: 

As a method for controlling society, Keju was extremely advanced for its time. The route to success and the path for preparation were laid out clearly for participants. The metrics were unambiguous and delineated, and in modern marketing language Keju had a deep channel penetration—it was accessible to all of the cross-sections of society, including its lowest rungs. All else being equal—and often all else was stacked in favor of Keju in any case—a rational person would likely choose Keju over other career paths that yielded uncertain payoffs. 

As the sponge that soaked up the vast majority of highly intelligent and motivated young people, the Keju-powered bureaucracy would direct the energies of the nation’s top talent toward the preservation of state power. Religious, commercial, or research institutions that could have been the source of economic or intellectual flourishing were instead starved of talent. The merit-based nature of selection added a hefty layer of legitimacy to an expansive and stifling bureaucracy. Keju would grow over time to diffuse threats to the imperial regime, to co-opt talent that might otherwise coalesce and push back on an authoritarian state, and to concentrate moral legitimacy in all-powerful emperors. 

The material exam candidates were required to learn deeply enforced attitudes deferential to hierarchy, tradition, and state authority, all of which characterize China’s feeble civil society to this day. The core canon that ultimately emerged was based on Neo-Confucianism, which Huang described in the following terms:

Neo-Confucianism is more unabashedly autocratic and statist than original Confucianism. It exalted eliminating human desires and a complete subjugation of the self. Summarizing a common view among historians, Peter Bol observes that Neo-Confucianism “provided a justification for seeking external authority in the ruler” and stipulated that the emperor alone was responsible for transforming the world. 

In other words, the material drilled into generations of elite bureaucrats was hostile to free thought and the relentless curiosity and skepticism that defined Enlightenment-era European intellectuals. This did not exactly yield fertile ground for an industrial revolution.

But for all the flack Confucianism gets for attitudes hostile to free thought and technological progress, these features were not necessarily the impetus for adoption. Other ideologies may very well have taken off as the basis for the exams instead. As Huang points out, the Confucian curriculum withstood long periods of rule by Buddhist and Daoist emperors. Why? He argues that the exams served as a kind of signaling device. With over 9,000 characters, the Confucian canon was far denser and more difficult than any potential alternatives. Students who made it through had to put in years of incredibly tedious work and rote memorization. Successful exam-takers proved more than just mastery of a particular school of thought; they also demonstrated a kind of dedication and loyalty that would yield dedicated yet servile bureaucrats. 

It was only over centuries that the effects of pushing the country’s top talent through the Confucian ringer that its ideals became deeply ingrained in the attitudes of those who steered the state and society. The technological lead that China had over Western Europe at the turn of the last millennium would gradually become a gaping deficit. Generations of both liberal and revolutionary thinkers saw the downsides of hostility to modernizing but could do little to uproot it. 

Echoes in modern academia?

We have no Keju exam system in the United States today, of course. Yet there seem to be many echoes of Keju in the status and signaling games that dominate the time and attention of talented young people today, particularly higher education. No, universities are hardly explicit tools of political control. But just as Keju monopolized nearly all of China’s top talent, most bright young Americans pass through traditional academia at some point during their formative years. Consider some of the other similarities. 

Keju emerged primarily for instrumental reasons; the signal from passing one or more exams was useful for a state seeking deferential but capable bureaucrats. You had to be smart to digest a dense Neo-Confucian curriculum, but you also had to be dedicated and thorough. Today, economist Bryan Caplan and others compellingly argue that much of, or perhaps most of, the value of a college degree comes from the signaling it sends rather than the content it teaches students. The dramatic expansion of access to higher education arose after World War II for highly practical reasons, just like Keju. The values it imbued were merely incidental at first. 

Modern academia certainly allows vastly more free thought than Keju, which militantly limited the scope of acceptable ideas. But the highly democratic elements of academia also means its scope of views and people permitted in its ranks is primarily consensus-based. If you are an outsider with few backers on the inside–perhaps one with an idea for an experimental technology that could change the world of vaccinations–you may struggle for funding or employment. 

Finally, consider the finding that political and social attitudes in academia are converging on a particular brand of progressive thought. A small minority of professors at American universities even self-identify as conservatives. Campus wars break out routinely over guest speaker invitations, episodes which often end with universities caving to activist demands. The degrowth movement and its skepticism of economic growth and technological progress is one primarily led by elite academics. We should hardly be surprised if, over the long run, funneling nearly all of our best and brightest young people through this institutional monoculture will have major effects on how future generations of elites view the world. 

One of the most popular ways to earn bipartisan plaudits in American politics is by promoting alternatives to four-year degrees. The costs are simply too high for a four-year degree to be worth it for the average student, critics say. But this may get the truth exactly backward. For the average student finishing his or her degree, college still pays handsomely, particularly at state schools. Tellingly, the kids of the elites who most decry universities may be those most certain to attend themselves. 

Instead, we should be thinking more carefully about alternative pathways for our brightest minds, the “exit” in Albert Hirschaman’s Exit, Loyalty, and Voice framework. It is not necessarily the case that academia itself is broken per se, nor is it necessarily true that the NIH is hopelessly rotten. Rather, we may not have properly balanced our talent portfolio. 

Relying on dominant institutions to cultivate and nurture talent–like holding an investment portfolio overwhelmingly made up of Ford stock–is an implicit bet that their worldview is correct. The biomedical research ecosystem is dominated by an NIH that was deeply skeptical of mRNA vaccines, nearly halting a revolution in medicine before it got started. The Chinese imperial bureaucracy bet heavily on a system that passed on values deeply skeptical of freedom and innovation; it is no coincidence that China remained mired in deep poverty long after the West. 

Today, rather than create early pathways to independent research or entrepreneurship like the Thiel Fellowship or Emergent Ventures, the default is to send our smartest young people through traditional universities, where degrowth and illiberalism hold great sway. We’d be fools not to think this will shape the attitudes of future elites. Unless you think these institutions have got it exactly right, there may be more Katalin Karikos in our future.

The triumph of science and reason that seeded the industrial revolution was as much a story of towering individual geniuses as it was of the small number of specific places that produced and nurtured them. The so-called “Great Enrichment” happened only once and began in only a handful of places; today, most world-changing innovations still emerge from a tiny number of especially-productive “superstar” cities.

Yet much like the lives of the generational inventors and scientists who walk their streets, these “Centers of Progress”–as Cato Institute author Chelsea Follett calls 40 cities that shaped the modern world in a new book–often do not last long at the absolute frontier of economic, scientific, or social achievement. As Matt Ridley noted in the book’s introduction, “global progress depends on a series of sudden bush fires of innovation, bursting into life in unpredictable places, burning fiercely, and then dying rapidly.” 

Schumpeter argues that both the creation and death of firms are key to driving innovation and growth in the long-run; so, too is the cycle of individual cities displacing leading centers of progress, only to later be usurped themselves. Fixing the policy mistakes of today’s superstar cities is indeed important, but perhaps we are neglecting the many second-tier regions across the country, some of whom will be tomorrow’s superstars.

Consider the dramatic extent to which the economic geography of the United States has changed over the past century. Of the ten largest cities by population in 1900, only three remain in the top ten today. Detroit was the wealthiest city in the nation in 1950 on a per-capita basis, while San Jose had yet to crack 100,000 residents. The next 50 years would see the hollowing out of once-thriving Rust Belt metros and the rise of new technology hubs on both coasts. Today, another great shift is taking place as wealth, talent, and investment flow to booming Sun Belt cities and their surrounding suburbs from coastal cities that have priced out families. 

The rise of the knowledge economy and its tendency to tightly cluster were deep, structural forces probably destined to reshape the country’s economic map in some way, but where new hubs took root was often a highly contingent outcome determined by the decisions of just a handful of people. As economist Enrico Moretti notes in his book, The New Geography of Jobs, Bill Gates and Paul Allen’s fateful decision to move Microsoft from Albuquerque to the Seattle area in 1979 changed these two cities’ economic trajectory for decades. On paper, Albuquerque and Seattle both had decidedly unremarkable economies in the late 1970s and the latter had been hit hard by population loss over the prior decade. Gates simply wanted to live closer to home. It is not far-fetched to imagine an alternate scenario in which Albuquerque emerges as a global hub of tech talent and Seattle remains a relative backwater. 

To the many of us who see continued economic progress and ending the “Great Stagnation” as moral imperatives, the failure of the most innovative metro areas to accommodate new people and the ideas they bring is blatant economic self-sabotage. Indeed, one famous estimate from Moretti and fellow economist Chang-Tai Hsieh finds new restrictions on housing construction since 1960 in just New York City, San Francisco, and San Jose are costing the average American family more than $3,600 per year. But while the housing policy failures of regions like the Bay Area are costing us all dearly, these cities’ place at the top of America’s economic hierarchy is not an inevitable fact of life. For those of us who do not live in a high-cost superstar region, campaigning to force them to remain open to new people, ideas and technologies is a tempting idea that would benefit us all, but we would also benefit from channeling at least some of that energy to our own backyards. 

The contingency of a take-off story like Seattle’s should not cause you to throw your hands up and conclude the take-off of future innovative hubs is mostly random. Rather, it should demonstrate the effect even very small changes can make in the long-run. Subtle differences in cultural attitudes or minute regulations may ultimately separate leaders from followers. 

Your city may be at just such a crossroads at this moment. The flexibility of your city’s housing policy towards densification may determine whether future innovators or entrepreneurs can cluster together sufficiently to share ideas. Your local university’s ability to truly integrate with the local economy will influence whether talented graduates leave or put down roots and build their professional and intellectual networks locally. Even city leaders’ attitudes towards new technology and experimentation may shape your city’s trajectory. Not all of these areas are viewed as “innovation policy” per se but are arguably just as important as cities’ efforts to forge new clusters themselves. 

There is good news: unlike in the Bay Area, where supervisors blocking new housing or self-driving cars are national villains and local debates are bitter and entrenched, odds are few people in your city are even paying much attention. You can likely make a much greater difference than you think.

The study of progress almost always focuses on the extremes–the stories of the lone inventor or the one-off lab. But “place” matters just as much, and world-shaping ideas, companies, or technologies often take root in unexpected places. The next one may very well originate in your own backyard, if you are willing to help.

President Biden rightfully earned the consternation of economists of practically every political stripe when in his State of the Union address on Tuesday night, he announced new steps “to require all construction materials used in federal infrastructure projects to be made in America.” Despite facing infrastructure costs that are in many areas well above those of peer wealthy nations, so-called “Buy American” provisions apparently remain a political winner with cheap, populist appeal.

But rather than dwell on the merits of requiring public infrastructure dollars to be funneled towards a domestic construction industry that hasn’t seen productivity growth in decades, as many others will do better than I can, how about something new and constructive (pun intended)?

The transit cost crisis seems to be most severe in especially complex, bespoke systems built in major cities, the most infamous example perhaps being the recently-opened Second Avenue Subway in New York, which was originally scheduled to be finished as early as 1958. What has always baffled me about spiraling transit costs of this type is the stunning lack of curiosity among public officials about why exactly it costs so much more per mile to build out new transit in the U.S. than seemingly anywhere else. One could easily tell a story about local capture, incompetence, or outright corruption and bribery, sure. Yet local leaders ultimately do have some incentive to keep costs (and taxes) low and deliver public services constituents actually want to use, even if construction timelines don’t quite align with electoral calendars.

So why, when, for instance, Barcelona builds 30 miles of an underground subway system for one-third less than the cost of a three-mile extension on the East Side of Manhattan, does public outrage not ensue? More pointedly, why doesn’t the Mayor of New York City or the Metropolitan Transit Authority call the folks who built such a system in Barcelona for so little and ask them how they did it?

I’m under no illusion that reducing transit costs is hardly that simple, of course. The U.S. is substantially wealthier than Spain and Italy, so per-hour labor costs are inevitably going to be higher. Complicated projects like transit build-outs each face their own unique challenges. But the difference between what we pay and what so many of our peer economies do is, at times, too staggering to be explained away by these factors.

Don’t take my word for it. The Transit Costs Project recently came out with an extensive report highlighting a litany of American practices driving up costs in so many cities. The four authors of the report conclude that a contributing factor in America’s uniquely high costs is inadequate state capacity, specifically civil service employees with little to no experience delivering cheap, timely transit projects. Losing the ability to build cheap transit systems is self-reinforcing: once you lose the institutional muscle memory, it’s incredibly difficult to recover. It doesn’t help that regional transit authorities—and agencies handling regional infrastructure generally—can be insular and complacent.

Insufficient knowledge diffusion, it appears, may explain a fair amount of the country’s inability to deliver high-quality infrastructure on-time and at a reasonable price tag. On some level, this may seem absurd. After all, Western local governments are more transparent than they have ever been, and the internet enables access to strategic reports and detailed construction plans for infrastructure projects in nearly every city you can name. But the diffusion of frontier knowledge and practices from leading firms to laggards has slowed in swaths of the private sector as well, with troubling implications for economic dynamism and productivity growth, so maybe this shouldn’t surprise you. A better question might be why we would expect public infrastructure authorities to be any different in this respect, given how much they outsource to private contractors and consultants.

So how might we induce local governments to learn from global peers?

The Rebuilding America Visa

One straightforward, budget-neutral tool for sparking knowledge diffusion is immigration. Did Barcelona (or Milan or Tokyo) build and operate a major piece of complicated infrastructure your city is now looking to build? Your mayor and city council should have no-strings attached infrastructure green cards to issue at-will to upper-level transit agency staff, key infrastructure employees, and developers or contractors with expertise and experiencing delivering infrastructure at low cost.

Cities exceeding a certain minimum population size should be granted an annual batch of “Rebuilding America Visas” for flexible allocation among prospective key government employees, advisers, or managers recruited from abroad to contribute to new infrastructure investments, maintenance, or management. Such visas should come with permanent residency that grants flexibility to move between employers of any kind, and ultimately a path to citizenship. Immigrants receiving these visas should not be subject to annual worldwide or per-country green card caps that leave migrants from some countries waiting for years in limbo. The terms should be generous and attractive while the process should be cheap and seamless for participating cities. Perhaps federal funds from the bipartisan Infrastructure Investments and Jobs Act (IIJA) could be supplemented with broader, plug-and-play access to the global pool of relevant talent.

This potential new tool is not meant to displace American bureaucrats, infrastructure administrators, or developers, casting them aside as irredeemably unproductive. In fact, a Rebuilding America Visa will bring first-hand knowledge of best practices from around the world directly to the many capable, talented American officials currently hamstrung by institutional sclerosis and political indifference. In the long-run, most Rebuilding America Visa recipients will move on to jobs in the private sector, likely bringing their cutting-edge knowledge and experience to a private domestic construction sector that is notoriously fragmented and stagnant.

Granting local governments the ability to easily tap into a pool of experienced experts and officials from across the globe is unlikely to cure most of what ails American infrastructure. We still desperately need permitting reform, for example, to cut down on bogus environmental and community reviews that often delay projects for years. No talented foreign administrator is going to get around this problem (though perhaps even here, knowledge of how peer countries do environmental review could help break this impasse). Projects will still face “Buy American” or “buy local” requirements like those announced this week that inflate costs and water down the value taxpayers get for their money. Rebuilding America Visas will not come close to closing massive infrastructure cost gaps.

Nevertheless, it’s a good place to start. And if w’ere serious about reinvigorating American infrastructure and once again building things the rest of the world envies, we literally cannot afford to continue being complacent.

Thanks for reading. Follow me on Twitter and subscribe below.

Subscribe now

If you’ve ever seen the movie In Bruges starring Colin Farrell, there’s a good chance you have sworn off ever visiting the city that serves as the movie’s namesake. Farrell’s character, a rookie hitman who botches an assassination attempt of a priest in England, instead killing a child, is sent by his boss to the city of Bruges to lay low and wait for further instructions. He is not very happy about it:

“There's a Christmas tree somewhere in London with a bunch of presents underneath it that'll never be opened. And I thought, if I survive all of this, I'd go to that house, apologize to the mother there, and accept whatever punishment she chose for me. Prison... death... didn't matter. Because at least in prison and at least in death, you know, I wouldn't be in fuckin' Bruges. But then, like a flash, it came to me. And I realized, fuck man, maybe that's what hell is: the entire rest of eternity spent in fuckin' Bruges.

Harsh.

In truth, Bruges is a charming little city, albeit one dominated by corny tourist traps. But it is indeed a bit depressing. Like other medieval European cities, you feel walking through the city center that you can precisely identify the era in which the city slipped into irrelevancy. And there’s a good reason for that. In the late Middle Ages, Bruges was one of the preeminent European trading hubs, drawing merchants from the far reaches of the Christian world. The city became the northwestern outpost of what historians call the Commercial Revolution, a dramatic revival of global trade that brought immense wealth to coastal ports able to meet merchants’ needs.

An excellent book recently recommended to me by the Charter City Institute’s Kurtis Lockhart satisfied my longstanding desire to learn more about what happened to Bruges, but also left me making some unexpected connections to problems cities face today. The book, Cities of Commerce: The Institutional Foundations of International Trade in the Low Countries, 1250-1650, fittingly by a Dutch economic historian named Oscar Gelderblom, traces the rise of Bruges, the subsequent usurpation as the region’s dominant port by Antwerp, followed by Amsterdam.

The economic institutions of Bruges, Antwerp, and Amsterdam underwent enormous changes through the period Gelderblom examines, first to attract foreign merchants and the lucrative growth they brought, then to retain them. The competition between the port cities of the Low Countries with its rivals—including London, Calais, and the various cities of the Hanseatic League in modern-day Germany—was intense. The close proximity of these ports to each other allowed merchants to take their business elsewhere with relative ease if cities failed to protect trade, settle disputes in a timely manner, or ensure contracts were enforced. Gelderblom’s thesis is thus relatively simple: the intensity of this urban competition incentivized individual cities to adapt legal systems conducive to frictionless, transparent, and open markets.

The extent to which this institutional competition between cities caused subsequent rapid growth in the Netherlands is a question better answered by professional economic historians, but there is obviously an interesting story here. And I am, to a great extent convinced by Walter Scheidel’s case laid out in Escape From Rome that the deep political, religious, and social fragmentation of Europe after the fall of the Western Roman Empire—even when it led to fierce and prolonged warfare—was a major factor in the continent’s adoption of more modern trade law and practices. So it’s certainly worth talking about what competition between Bruges, Antwerp, Amsterdam, and its rival port cities looked like. What were its key features? A few themes stand out in Gelderblom’s analysis.

Bruges, Antwerp, and Amsterdam continually revisited economic policy to adapt to market expansion.

As trade in the Low Countries, France, and Germany outgrew large seasonal fairs, merchants began to trade in cities like Bruges year-round. Early on, trade was often between merchants from the same jurisdiction. Merchants from the German Hanse would trade with fellow German conduits in Bruges, for example. Magistrates in Low Country ports had a solution to this: They would allow merchants to set up courts to resolve disputes in accordance with their own legal systems. In essence, foreign traders could appeal for a “foreign nation” (trade guild) creating a regulatory sandbox within each market for, say, English traders to settle suits under English law.

For the moment, this move was rather brilliant. Cities like Bruges didn’t need to risk picking the “wrong” set of rules for merchants to follow, which might alienate traders or foreign governments. But it has some obvious limits. When trade became more permanent and merchants began to live full-time in these fledgling commercial cities, they wanted to trade with merchants of other nationalities. An English trade guild could not effectively resolve disputes between a wholesaler operating out of Genoa and a merchant from London. Trade law had to be reclaimed by local authorities, which it ultimately was. Over time, Gelderblom notes, municipal Low Country courts took up some of the work of enforcing contracts and resolving disputes under a unified legal code.

The circumvention of foreign trade guilds is just one instance of many in which the port cities of the Low Countries adapted to the pressures the growth of the size and scope of trade put on key economic institutions. At other points, local governments would find clever ways to bypass potential bottlenecks on the growth of trade and broaden the types of evidence admissible in court to speedily resolve litigation.

The Low County ports were particularly open to adopting foreign merchants’ legal and business practices.

The adaptability of the Low Country ports’ legal institutions had additional benefits. Even after bypassing trade guilds and establishing specialized courts to deal with various areas of mercantile law, traders in these cities relied mainly on third-party arbiters. This meant that while cities’ legal systems gradually converged, they remained open to alternative rules and customs practiced elsewhere. Over time, a standardized system of mercantile law emerged in northwestern Europe, in part because cities remained so flexible and reliant on private solutions.

A prominent example of this is the adoption of double-entry bookkeeping, a key innovation that came out of the thriving maritime economies of Italian city-states in the 14th century. The use of double-entry bookkeeping proved incredibly useful for merchants running complex businesses and the practice gradually spread north. Notably, it also served as more ironclad proof of debits, credits, and orders than mere shipmaster accounts. Eventually, merchants in disputes over missed deliveries appealed to arbiters and courts for these documents to be admissible as proof in contract disputes. As the use of them continued to grow, not only did double-entry account books become acceptable evidence in legal disputes, but so did partnership agreements, IOUs, and other contracts hashed out between merchants. Rather than declaring a standard and implementing it top-down, Bruges, Antwerp, and Amsterdam allowed a type of “bottom-up” legislating that led to the use of business practices that were not only more efficient, but also more conducive to speedily resolving disagreements.

At key points, Low Country port cities crushed the power of rent-seekers in favor of growth.

Growing markets in Low Country ports put them on collision courses with key interest groups, such as the trade guilds operating legal sandboxes. Two particularly important junctures in the development of these cities were their conflicts with hostellers and brokers.

Before merchants bought their own compounds in trading hubs in cities like Bruges, hostellers served as key go-betweens for merchants, distributors, wholesalers, and workers. In effect, hostels were, as Gelderblom calls them, “one-stop shops” providing not only lodging but pricing information, storage for merchants’ goods, and lending products. Some hostellers, with permission, even bought and sold goods on merchants’ behalf. Similarly, brokers, with their deep knowledge of local markets and relationships with merchants, could charge fees to set up trades between merchants who otherwise didn’t have the information to do it themselves.

When merchants began to permanently settle in port cities, however, the usefulness of hostellers and brokers declined. Merchants didn’t need hostellers to execute purchases and sales for them anymore. Traders could get pricing information from local publications or their own employees on the ground. They could also store their own goods on their own properties.

Anticipating a future in which their services could be split up and purchased by merchants more efficiently elsewhere, hostel operators and brokers, like many rent-seeking interest groups throughout history, could have fought back and extracted legal protection from competition from local government. Indeed they did. Broker guilds were, for a time, able to artificially keep their numbers low, their fees high, and convince magistrates to mandate that merchants use licensed brokers when trading commodities.

Ultimately, however, these attempts by interest groups to extract rents consistently failed. Local magistrates, when forced to side with free markets or politically powerful interest groups, sided consistently with the former. Gelderblom writes that this choice was made much easier by the fact that trade was nonetheless growing rapidly. Interest groups were not fighting over slices of a shrinking pie. There was plenty to go around without dishing out special favors.

Ok, but what does this have to do with modern cities?

None of this is to say the late medieval Dutch model is something to be emulated, of course, and the Dutch Republic after this era entered into a period of prolonged stagnation that saw it fall behind England and become a mere second-tier player in European affairs. The period we’re discussing here is one of Smithian growth, not the Schumpeterian growth associated with the Industrial Revolution (to which the Dutch were latecomers). But the relative intensity of urban competition during this period and the mechanisms by which it produced consistent leaps forward in commercial policy—all while the major powers of Europe dumped untold resources into wars for control of the Netherlands—is fascinating, and, I think, raises some key questions for today’s American cities. Among them:

• How can cities build political cultures more friendly to outside practices and adapt to economic change? Relatedly, what’s stopping cities from experimenting with more radical institutional change?

• How can cities break rent-seeking bottlenecks on growth, or perhaps more importantly, prevent their formation?

American cities seem increasingly sclerotic and hostile to experimentation or learning. Major cities across the country are captured by tiny cliques of NIMBY landowners who make even minor changes in land-use regulation an enormous undertaking. Despite fellow developed countries like France and Spain consistently proving subway systems can be built far more cheaply than they are in the US, domestic infrastructure officials and transit agencies see absolutely no urgency to learn from the rest of the world, instead choosing to deliver outrageously expensive, mediocre services from favored contractors. Growing urban-rural political polarization has also meant that the real action in municipal elections has shifted from multi-party fall contests to lower turnout Democratic primaries, further narrowing the scope of urban policy debates. The global economy has changed tremendously over the last 50 years. Has the way we organize and govern cities really adapted commensurately with these changes?

How might we take a page from the nimble cities of the Dutch Golden Age and break out of the political malaise that’s infected so many cities? There are no easy answers, of course. But I have a few ideas, and at least one seems particularly compelling.

Feed hungry software.

A decade ago, venture capitalist Marc Andreessen wrote that software was eating the world. At the time, companies like Uber and Airbnb were shattering the status quo in key urban markets by circumventing powerful rent-seeking lobbies. In the decade since, this revolution has only accelerated. When economist Mancur Olson lamented the decline of American growth in the early 1980s, in part blaming decades of accumulated rent-seeking producing economic sclerosis, he proposed accelerating jurisdictional integration as a means of peacefully countering extractive interest groups. While achieving macroeconomic growth so rapid that rent-seekers become dispensable is unlikely anytime soon, expanding the scope and depth of markets is a viable alternative. Olson had in mind measures like free trade agreements and interstate occupational license compacts but software can accomplish much of this legal legwork too.

In 2020, Zoom radically upended regional white-collar labor markets, suddenly collapsing dozens of regional professional labor markets into one national talent pool. Suddenly employers could scour the country for top talent from a desk in New York and assemble a team scattered across the country. The medieval trade guilds of today are on notice.

Over the long-run, software will continue to integrate regional markets both within and outside the United States. As the magistrates of Bruges, Antwerp, and Amsterdam discovered, growing the depth and scope of the market will have two political effects. First, it will tilt the political environment against extractive interest groups, amplifying the potential gains from gutting their state-granted favors and protections. Second, even if cities resist these incentives, software will allow markets to circumvent them nevertheless, so long as national governments with the power to fight back instead nurture it. In Bruges and Antwerp, the proliferation of unlicensed brokers proved beyond the reach of the state to combat, so it simply gave up. Software will integrate markets by force and there’s little that cities themselves can do to stop its march. With this inflow of new talent, ideas, and tools, perhaps even stagnant, slow urban bureaucracies will show renewed vigor and creativity.

Just as the growth of international trade proved to be the key juncture that led Low Country ports to break free of hostellers and brokers, the maturation of the virtual economy with its capacity to seamlessly integrate markets across borders may also prove to be exactly the kind of jolt our cities need. 

Disagree? Did I get something completely, laughably wrong? Feel free to tell me in the comments below or on Twitter @cojobrien.

This is perhaps the most interesting moment in decades to be studying or otherwise thinking about the future of American cities. The social and economic shocks of the last two years have either fundamentally upended the economic trajectory of dense urban centers across the developed world or been a mere blip on an inevitable march toward continued urbanization, depending on whom you ask.

The potential and durability of remote work are likely the most important questions in this genre and touch on everything from transportation funding to the safety of dense downtown office corridors. Two years since its explosion, remote work and the freedom it provides have already started to shift where people choose to live, sending workers away from expensive cities and toward cheaper suburbs and resort towns. Sizable outmigration from major cities threatens once-reliable sources of tax revenue, so it’s no surprise that big city leaders like New York Mayor Eric Adams have practically begged office workers to return full-time.

Urbanists like myself have naturally been skeptical that remote work truly threatens the economic foundations of cities. After all, cities have survived far more daunting episodes of plague, war, riots, and economic collapse. To city lovers, the economic attraction and cultural allure of cities seem practically unshakable. But in recent months, I’ve come around to thinking that remote work will be a much more important factor shaping cities going forward than most urbanists seem to believe. No, it will not spark a mass exodus to the suburbs and exurbs on a permanent basis, at least not if us urbanists play our cards right. Instead, as I will argue in this piece, the rise of remote work will provide urbanists and YIMBYs an unprecedented opportunity to reshape the urban political landscape in our favor, aligning urban leaders’ interests with our goal of more vibrant, livable, affordable, and accessible cities. Rather than lament the rise of remote work, we should embrace it.

Remote work softens the agglomeration effects that pull in high-skilled workers

People migrate to cities for many different reasons, but they roughly boil down to two buckets. First, there are “amenities” in the broadest sense. Living in New York or Los Angeles, you have virtually unparalleled access to some of the world’s best entertainment options, food, and experiences you simply can’t have in a cookie-cutter suburb or small rural town. For young people, the dating pool is an order of magnitude larger. You also have better access to transportation that enables you to see friends and family both within the city and elsewhere.

The second bucket includes both the strength and depth of labor markets. People move to cities to work, but not necessarily for a particular job. The size of urban labor markets provides a kind of insurance policy to workers. If you’re fired or don’t like your employer, you very likely have the option to move to a different firm nearby in the same industry. Dense labor markets also provide a safety net for potential entrepreneurs, who will be less reluctant to take the enormous risk of starting a new venture if they know a strong job market awaits them if they fail. Finally, the proximity of highly-skilled workers to each other leads to cross-pollination or collaboration between firms and spontaneous interactions that lead to new ideas and new companies.

At its best, this labor market agglomeration looks something like AnnaLee Saxenian’s fascinating 1996 portrayal of Silicon Valley’s business environment, Regional Advantage. Workers are constantly bouncing between employers, starting their own highly specialized firms, and interacting with university and industry researchers on the cutting edge at industry forums, more structured R&D environments, or local bars. This force is why “superstar” firms have tended to concentrate in superstar cities, where they become more productive despite higher labor and real estate costs.

While Saxenian argues that proximity was a key driver of Silicon Valley’s success, remote work enables this physical proximity to be recreated virtually. High-quality video conferencing, collaborative coding platforms like GitHub, and rich online communities on social media platforms can, for many workers, rival what they would experience in-person, adjusting for cost. There is a growing body of evidence that the internet has substantially reduced the advantages of physical proximity in a host of technical, high-paying industries. Matt Clancy summarizes much of this research in his Case for Remote work paper from 2020. If you’re interested in a deeper discussion of the erosion of agglomeration effects on innovation, I’d strongly suggest this paper.

Remote work aligns the parameters of urban competition with urbanist and YIMBY demands

If labor market agglomeration effects are declining in importance (and they will likely not disappear), amenities and quality of life issues become relatively more important for people deciding where they want to live.

Cities have always had an incentive to attract workers. Stable or growing populations are better able to fund the high fixed costs associated with urban government like transportation, water, or port systems. The mid-20th century is rife with horror stories of American cities being hollowed out and verging on outright bankruptcy. But addressing quality of life and affordability issues is hard and requires taking on deeply entrenched interest groups. So for decades, cities have tried to attract workers through urban agglomeration effects. Cities throw subsidies, tax breaks, and special leasing deals at large companies willing to set up shop in their downtowns. Though an expensive and overrated tool of economic development, chasing firms yields visible, tangible accomplishments for mayors and council members and, to the extent it works, enables cities to paper over their other, deeper challenges.

In a world of widespread remote work, cities are forced to lure people, not firms. Livability and affordability will matter more. Cities like San Francisco or San Jose can no longer be bailed out by the fact that some of the most innovative firms in the world locate there if workers for those firms can live wherever they want. If major cities want to continue to thrive in an economy with a large share of completely remote workers, they need to address housing shortages, provide transportation systems that are convenient and affordable, and design streets safe for walkers and bikers.

In other words, the competition between cities will increasingly take place on turf friendly to urbanists and YIMBYs. Cities will have an incentive to align themselves with pro-growth advocates for permitting more market-rate affordable housing so that workers aren’t forking over 30 or 40 percent of their income each month in rent. If cities want to attract a new class of remote workers, they’ll also need to provide larger living spaces. While workers may have been willing to temporarily sacrifice half of a living room or bedroom to their employers, they’ll increasingly demand a second or third bedroom for a permanent office space as remote work arrangements set in. Again, cities must build to compete.

As workers spend less time commuting and more time in their own neighborhoods, they will demand better entertainment amenities like concert halls, theaters, clubs, and restaurants. They’ll want riverwalks, boardwalks, and permanent outdoor seating at their favorite bars. Stale office-dominant corridors like parts of downtown Chicago or Midtown Manhattan will need to transform their streetscapes to attract people with something other than a well-paying nine-to-five. They will need to make many of the pandemic-era adaptions like open, pedestrian-only streets and marketplaces permanent.

Cultural amenities, already a key feature of urban life, will take on increasing importance. Cities will want to be more conducive to art, music, and writing scenes. This means devoting more resources and attention to housing affordability, public spaces, and affordable storefront rents.

Urbanists should embrace the opportunity provided by remote work’s rise

In other words, while remote work is on one level a tool that enables people to leave pricey cities for lower-cost places, it may also force cities to finally confront long-simmering quality of life problems. As urbanists, it provides an opportunity to flip the terms of debate in our favor. If your city wants to remain competitive, you can’t simply throw taxpayer dollars at Amazon or Ford or Meta. You need to embrace density, growth, and change. Sprawling, bland, car-centric downtown office clusters are not what they once were. Cities are far from doomed, and talented young people will always want to cluster around other talented young people in some form or another. But it’s far from given that this clustering will continue to look like the clustering of past eras, especially as housing costs spiral out of control, transit systems become dysfunctional, and violent crime ticks up.

Urbanists tend to be on the leading edge of new cultural movements and new technologies. We love cities because we love diversity, variety, and change. Let’s not stick our heads in the sand and deny the obvious ways it’s impacting economic geography across the country. Let’s instead embrace this change and utilize the leverage it gives us to make our cities more affordable, more livable, more vibrant, and more of what made us fall in love with urban life in the first place. 

Welcome to Regions, an occasional newsletter that will cover, among other things, the history and future of regional competition, renewal, and progress.

A bit about me

I’m a Policy and Research Associate at the Economic Innovation Group, a small think tank in Washington focusing on policies to spur inclusive economic growth and boost dynamism across the American economy1. Originally from New Jersey, I’m a proud graduate of Rutgers-New Brunswick and the University of Chicago. Follow me on Twitter @cojobrien.

What I’m doing here

I want to keep this newsletter low-pressure, a place to jot down thoughts on what I’m reading or thinking about that won’t neatly fit elsewhere. These posts will be a mixture of reflections on what I’m reading in economic history or political economy and current policy issues.

I’m particularly interested in big picture questions around why certain regions (think: cities and sub-national units) are able to become such dynamic hubs of innovation, creativity, and high culture while others don’t. Further questions that interest me:

• Why have so many thriving cities throughout history collapsed or faded into irrelevancy?

• What differentiates those which dramatically rebound?

• To what extent do cities even have real agency in a world dominated by large nation-states and truly global markets?

Unfortunately I’m just A Guy Like You, floating through this unimaginably complex universe, trying to learn a bit along the way. I suspect most of what I write about these questions will be, to some degree, embarrassingly wrong. I hope you meet my wrongness with generosity. Learning in public, even in an age of intense online scrutiny and vitriol, is seriously underrated.

See you soon,

Connor

Subscribe now