the left

The Case Against Supply-Side Liberalism Is Weak

Photo: John Moore/Getty Images

In 2023, meeting liberalism’s goals requires building lots of stuff.

Progressives want to swiftly decarbonize the U.S. economy without lowering Americans’ living standards. That entails rapidly erecting vast constellations of solar panels, wind turbines, geothermal plants, mineral mines, and high-voltage transmission lines, among other things.

And liberals also want to bring down inflation without increasing unemployment. One way to do that is to make it easier for the supply side of the economy to grow in response to consumer demand. For example, a key driver of contemporary inflation is shelter costs. One way to slow rent growth is to raise interest rates until the economy enters recession, and workers have less money to spend on homes. But another, better way to lower rents would be to expand the nation’s scarce supply of housing units through zoning reforms and public investment.

If meeting modern liberalism’s objectives requires facilitating rapid construction, however, some of the ideological tradition’s tendencies aren’t conducive to that end.

In the 1970s, American liberals developed a skepticism toward development and sought to empower individuals and communities to challenge it. This was a rational response to the environmental depredations of the postwar era and highly inequitable approaches to “urban renewal.” Nevertheless, this turn ultimately served to multiply the veto points facing ambitious construction projects and empowered unrepresentative individuals and groups to block worthy initiatives.

Meanwhile, contemporary Democrats’ desires to deliver for a wide array of interest groups can lead them to layer burdensome requirements onto public contracts. For example, in San Francisco, the city is prohibited from building affordable housing with the aid of contractors who earn more than $7 million a year. This is meant to promote small businesses. But it has the effect of reducing the amount of affordable housing that San Francisco can build since the law makes it harder for the city to find contractors while barring it from availing itself of the most successful builders in the area, who have high revenues by definition.

Ezra Klein, Matthew Yglesias, and other commentators have called for liberalism to reorient its priorities in light of its present challenges. They argue that we need a new “supply-side liberalism” or a “liberalism that builds,” which would prioritize rapid decarbonization and housing abundance over other goals. In practice, this means scaling back regulatory obstacles to energy development that some environmental groups value and restricting the capacity of localities or self-appointed community representatives to override the will of democratically accountable officials.

This call for “supply-side liberalism” has attracted some sharp critiques. In a couple of recent pieces, the American Prospect’s David Dayen has suggested that Klein’s outlook is antithetical to both upholding core progressive values and building the coalitions necessary for enacting Klein’s own goals. Many scholars at the progressive Roosevelt Institute have seconded Dayen’s objections.

Although I’m sympathetic to Klein’s perspective, I think supply-side liberalism warrants critical engagement. There is plausibly some tension between narrowing progressivism’s priorities and maintaining a majority coalition. And given how often apparent tradeoffs in economic policy have proven illusory, those that Klein encourages progressives to accept should be scrutinized.

But the most prominent critiques of supply-side liberalism have misrepresented its arguments while evading its strongest points. We deserve a better debate over Klein’s program than the one we’ve had so far.

What supply-side liberals do (and do not) believe.

The latest round of the supply-side liberalism debate began in April when Klein published a column titled, “The Problem With Everything-Bagel Liberalism.” The piece begins by examining San Francisco’s peculiar difficulties in building affordable housing. Infamously, the city chronically fails to house a large percentage of its low-income population. But a building with 145 units of permanent, supportive housing for the chronically homeless recently went up in the city’s SoMa neighborhood. That project took three years to build and cost $400,000 per unit.

This qualified as a triumph of efficiency. A typical affordable housing project in San Francisco takes six years to build and costs as much as $700,000 per unit. What enabled this new building to sprout up at such a discount in time and money? The fact that the government didn’t build it.

The Charles and Helen Schwab Foundation provided $50 million for the project, which enabled it to skirt the myriad requirements that the city attaches to affordable housing projects built with public money.

As mentioned above, those requirements include contracting exclusively with small businesses. But the city also requires projects to hire locally, get power from the city’s San Francisco Public Utilities Commission, submit to a review by the Arts Commission, and pass a review by the Mayor’s Office of Disability (even though every project must already be in compliance with the federal Americans With Disabilities Act). All these processes add years and millions of dollars in cost to affordable housing projects. With tax dollars scarce, this means that the city is building less housing for the homeless and at a slower rate than it would if it set fewer goals for its affordable housing program. The end result of the government’s failure to prioritize has been private funders stepping in and providing core government functions.

Klein sees this as a perverse and self-defeating place for liberalism to end up. And he raises the concern that a similar dynamic could compromise the Biden administration’s industrial policies.

Through the CHIPS and Science Act, the White House aims to seed a thriving domestic semiconductor manufacturing industry. Achieving that goal will likely require lowering the cost of producing chips in the United States. Right now, building and operating a fab in the U.S. is about 30 percent more expensive than it is in Taiwan, South Korea, or Singapore.

And yet, the administration has paired its subsidies to domestic chipmakers with a long list of asks that are sure to increase their costs. These include the drafting of “an equity strategy,” providing “child care for facility and construction workers,” including minority-, veteran-, and female-owned small businesses in their supply chains, creating a “climate and environment responsibility plan,” and making community investments in transit and affordable housing, among many other things. Klein believes that this list is self-defeatingly ambitious.

A cursory read of his column might lead one to believe that he is arguing for liberals to forfeit their commitment to promoting high wages, child care, and equity, in favor of a single-minded focus on production. But the piece explicitly disavows that idea.

Klein allows that the administration’s guidelines aim to advance “good goals,” but asks whether they “are good goals to include in this project?” The question isn’t whether liberals should pursue a wide range of social reforms, but whether it is wise to pursue them by attaching cost-increasing requirements to public funding for housing and semiconductor production.

Further, Klein writes that some of the administration’s requirements for CHIPS are worth preserving:

Making sure jobs in semiconductor factories are good jobs is worthwhile. But there is a cost to accumulation. How many goals and standards are too many? And why is subtraction so rare?  

Klein reiterates that he is not objecting to efforts to ensure high job quality in America’s chip industry elsewhere in the piece, writing, “We’re not, of course, going to undercut Taiwan or South Korea on labor costs.” And his column’s overriding metaphor further underscores that it is not an argument against any and all multitasking in public programs. “Everything bagels are, of course, the best bagels,” Klein writes. “But that is because they add just enough to the bagel and no more.”

Given that America wants to build a semiconductor industry that is both globally competitive and exceptionally high-paying, however, Klein suggests that any additional, cost-increasing requirements should clear a high bar. Otherwise, the administration risks saddling the CHIPS Act with so many disparate goals that it fails to achieve its core one.

We don’t need a liberalism that builds strawmen.

In May, Dayen published a lengthy critique of supply-side liberalism in general and Klein’s column in particular titled, “A Liberalism That Builds Power.”

In it, Dayen makes a few distinct arguments against Klein’s perspective. First, he insists that the fundamental obstacles to rapid decarbonization, housing abundance, and other “supply-side liberal” goals is the progressive movement’s limited power, not overly burdensome regulations or administrative guidelines. After all, America’s natural-gas industry managed to rapidly expand into a world-beater over the past decade, despite its copious need for new infrastructure. The sector’s awesome political and economic power overwhelmed all barriers in its path. Therefore, liberals should worry less about changing meddlesome rules than building power, since doing the latter can render the former toothless.

Dayen further posits that setting priorities in the manner that Klein advises is antithetical to building such power. Guarding against a backlash to the green transition requires ensuring that communities benefit from clean energy projects and that the renewables industry pays good wages. Rather than being mutually undermining or contradictory, Dayen sees all progressive goals as “mutually reinforcing.” Mandating high labor and environmental standards builds the coalition for decarbonization. And unions can facilitate recruitment for the industries that progressives wish to expand.

Dayen is a sharp thinker and great reporter. And his essay includes many cogent points. But he never forthrightly addresses the biggest challenges that supply-side liberals identify. Worse, he wildly misrepresents the argument that he’s trying to rebut.

According to Dayen, supply-side liberals insist that the goals of “domestic manufacturing, decarbonization, union jobs, and economic justice” must not be combined, and that “any barrier to the national abundance we need” must be torn down. Dayen continues:

Between the lines of their arguments is the idea that increasing manufacturing capacity should take precedence over making sure somebody has a good job. Clean-air laws that prevent 230,000 needless deaths a year must take a back seat. Safety laws that have reduced workplace injuries nearly fourfold are too onerous. Democracy — the ability for the public to express their views on matters that will affect them and get a hearing — isn’t worth the trouble.

“You’re telling people they’re not going to get anything out of it and their lives will be made worse,” said Marshall Steinbaum, an assistant professor of economics at the University of Utah. “That’s the tell that it’s not designed to build a political movement. That’s something Silicon Valley billionaires believe.”

Here, Dayen invents disreputable positions and then attributes them to his adversaries. Neither Klein nor Yglesias nor any other prominent supply-side liberal has called for repealing clean air laws, eroding workplace safety standards, or subordinating democracy to expert planners. As noted above, the column Dayen is responding to explicitly states, “Making sure jobs in semiconductor factories are good jobs is worthwhile.” Yglesias and Klein have both written extensively on the harms of air pollution and the necessity of regulating it. Klein also advocates for raising labor standards through the promotion of unions and sectoral bargaining.

As for democracy: A core premise of the supply-side liberal argument is that America’s existing regulatory processes enable unrepresentative, well-resourced individuals and groups to constrain the discretion of elected officials who are actually accountable to the broad public, in defiance of democratic principles. Klein’s “everything bagel” column, meanwhile, specifically lamented that San Francisco had rendered itself so ineffective at building affordable housing that the private sector had started impinging on the prerogatives of democratic government. One can argue that the supply-side liberal analysis is wrong and that empowering privately-financed, publicly unaccountable “community groups” to veto development is more democratic than making it easier for elected officials to implement ambitious plans. But to make that case persuasively, you need to do more than smear supply-side liberals as anti-democratic.

Meanwhile, if you find yourself summarizing an argument for building more housing and clean energy infrastructure as a call for “telling people they’re not going to get anything out of [liberalism] and their lives will be made worse,” I would say that this is a pretty good “tell” that you are not making a good-faith effort to fairly represent the perspective you’re rebutting.

Unfortunately, Dayen’s tendency to misrepresent the supply-side liberal perspective has proven persistent. Last month, Klein wrote a belated reply to Dayen’s piece. In it, the Times columnist argued that Dayen’s account of the relationship between priority-setting and building power was incomplete: Perhaps narrowing the objectives of discrete projects could alienate some coalition partners. But failing to set priorities could alienate the broader public from Democratic governance by preventing blue cities and states from building projects on time and budget. Indeed, blue states currently struggle with precisely this, and the U.S. electorate’s confidence in the public sector’s competence is low.

In making this argument, Klein wrote the following:

I’d like to see a stronger labor movement in America, which is one of many reasons I support sectoral bargaining and the mandated appointment of workers to corporate boards. Plenty of countries with stronger unions than America complete transit projects more rapidly and more affordably than we do. But one of the major challenges of pro-labor politics in the United States is the public perception that unions often slow construction rather than make it better, and that perception is rooted in real failures of real projects in places where liberals and unions hold real power. I don’t think that’s impossible to fix, but it is impossible to fix if liberals refuse to admit that it is true and refuse to make the policy changes necessary to prove that it’s untrue.

Here, Klein states explicitly that strong unions are both desirable and entirely compatible with the government building things more effectively. But he argues that the failures of Democratic governments in areas with high union density promote the perception that organized labor undermines construction. In context, the column attributes such failures to Democrats’ refusal to set priorities, such as by easing regulatory burdens to development.

Klein further indicates that he is not arguing that Democrats must buck unions by describing organized labor as a “natural ally” for a “liberalism that builds.” He also signals that he is warning about a damaging perception about unions inherently slowing construction — rather than a reality — by placing a hyperlink over the word perception that leads to a study finding that high transit construction costs are not attributable to unions.

As an example of how Democrats can expedite construction by setting priorities, Klein approvingly cites Pennsylvania governor Josh Shapiro’s decision to “use union labor” for the reconstruction of I-95, while goring “a lot of other interests and processes.” This enabled Shapiro to promptly rebuild the fallen highway while countering the impression that unions undermine government efficacy.

And yet, in his response to this piece, Dayen writes that “a chorus of supply-side progressives” are arguing that “worker wages are fundamentally not that important,” and that Democrats must forfeit their commitment to “good jobs” in order to “solve the problems that matter in America.” He continues:

As much as Klein protests, that’s what he’s saying through his emphasis. It reflects the Silicon Valley billionaire perspective of getting all this building done with non-union labor.

As with “between the lines,” Dayen uses the vague qualifier “through his emphasis” as a means of attributing reactionary views to Klein without the burden of providing any textual support for his charge. And once again, the view he is imputing to Klein — that Democrats should not seek to promote good jobs, high wages, or unions — is one that Klein explicitly disavows in the very piece Dayen is responding to.

This said, it is not unreasonable for Dayen to worry that supply-side liberal arguments about the need for a more efficient public sector could be used to undermine labor standards. There is a real tension between maximizing the amount of green energy or housing one can produce per federal dollar and mandating high wages on publicly financed construction projects. Given that every ideological tendency has more conservative and progressive variants, there is likely a subset of supply-side liberals who would question the merits of, say, Biden’s push to strengthen wage standards for public construction projects. But there is no contradiction between supporting such standards and accepting supply-side liberalism’s core premises. As Klein argues, and Governor Shapiro just demonstrated, there are plenty of ways to expedite the buildout of socially desirable infrastructure without shortchanging workers.

Stigmatizing an argument is not a substitute for refuting it.

Dayen makes a strong case against sacrificing all progressive objectives to the sole imperative of maximizing production. But most supply-side liberals aren’t arguing otherwise. Rather, they posit that there is a limit to how many disparate goals the government can layer onto any single program without undermining its efficacy. And they also argue that rapidly expanding the supply of clean energy and housing requires constraining the capacity of localities, groups, and individuals to stymie development.

Critics of “supply-side liberalism” have not persuasively rebutted either of these premises. Dayen neither concedes that it is possible to load up a program with too many goals nor offers an explanation for why overkill is not a genuine risk. Does Dayen agree that San Francisco should try to build housing for the homeless more efficiently, even if that means allowing the city to work with large contractors? If he does agree that San Francisco’s affordable housing program is one legitimate example of the government trying to do too many things at once, what rules of thumb can we use to ensure that other programs don’t cross a similar threshold? Neither of his pieces provides answers to these questions.

Perhaps, supply-side liberals’ most contentious claim is that abetting the green transition requires making it harder for conservationists to obstruct renewable energy installations and transmission lines.

This argument is rooted in a simple observation: The scale of infrastructure construction required for decarbonization is almost unimaginably vast. Right now, the largest solar power installation in the U.S. generates 585 megawatts of electricity. To meet a “middle of the road” renewable energy target, the U.S. will need to bring two new 400-megawatt solar power facilities online every week for the next 30 years.

Building carbon-free grids also requires a tremendous number of transmission lines, so renewable energy can be transferred from where it is generated to where it’s needed. The U.S. has never built more than 4,100 miles of transmission lines in a single year. To meet our climate targets, we need to double that amount and then keep hitting it year after year.

We are nowhere close to moving at the necessary pace. And one way to accelerate the buildout is to reform permitting in general, and the National Environmental Protection Act (NEPA) in particular.

The NEPA requires the drafting of environmental impact statements (EIS) before new projects are approved. When the law was first enacted, those statements could be as short as ten pages and individuals did not have the right to obstruct projects through litigation due to objections to an EIS. But courts subsequently established such a right. And under the threat of litigation, environmental impact statements grew radically more burdensome. As of last year, the average EIS ran 600 pages and took 4.5 years to complete.

Many environmental organizations value the NEPA’s capacity to frustrate projects. For a long time, green groups were primarily concerned with stymying environmentally damaging forms of development, rather than abetting the rapid buildout of clean energy. Many green legal practices, meanwhile, developed specialties in exploiting NEPA to thwart fossil fuel projects.

But supply-side liberals have argued that, in the context of the climate crisis, the costs of a permitting system biased toward delay outweigh the benefits. NEPA doesn’t merely enable environmentalists to obstruct genuinely destructive development but also helps rich NIMBYs fight offshore wind power and solar installations.

Building a clean energy economy requires constructing gargantuan amounts of new infrastructure. Maintaining a carbon economy, by contrast, requires building scarcely any, since the existing energy system is built around the needs and capacities of fossil fuels. A permitting system biased toward the status quo is therefore one that is biased toward carbon energy. As of 2021, NEPA reviews were holding up twice as many green projects as fossil fuel ones.

Dayen does not adequately address these arguments. His rebuttal to the supply-side liberal critique of the NEPA is twofold. First, he notes that increasing hiring at agencies that conduct NEPA reviews could reduce wait times. This is true. And yet, if hiring more reviewers could speed projects, surely doing that and streamlining the review process would expedite clean energy development even more. Perhaps, doing the former will be sufficient to reach the pace of development necessary for hitting our climate targets. But Dayen makes little effort to prove as much.

His second argument is that NEPA has not prevented the natural-gas industry’s rapid expansion. Therefore, if liberals build power behind clean energy, the law shouldn’t be an issue.

And yet, one reason why permitting hasn’t been a problem for the natural-gas industry is that gas drilling is categorically exempt from NEPA, while the federal government has centralized siting authority over natural-gas pipelines. These facts surely reflect the gas industry’s power. But making that observation leaves the fundamental question unanswered: Should liberals demand similar permitting exemptions for renewable energy?

It’s fine to say that liberals need to build power. But as an answer to the question of how liberals should go about accelerating the green transition, it’s also a bit of a non sequitur. Observing that it is necessary to attain power does not tell us what to do with it. And if one believes that reducing regulatory obstacles to clean energy is indispensable for decarbonization, then persuading others of that point is surely a precondition for “building the power” to make the relevant reforms.

The NEPA debate is less pressing now than it was when Dayen published his first broadside against supply-side liberalism. As part of the debt ceiling deal in June, Congress implemented many of the reforms championed by supply-side liberals, imposing page limits on environmental impact statements and time limits on NEPA reviews. But other proposed reforms, such as imposing a statute of limitations on legal challenges to approved projects, remain unimplemented.

Meanwhile, the federal government still lacks centralized authority to site transmission lines and allocate its costs. Dayen actually agrees with supply-side liberals that this needs to change. But any bipartisan deal to expand federal authority over transmission would almost certainly require Democrats to acquiesce to further NEPA reforms that are desired by Republicans. Supply-side liberals have historically been favorable to such a trade, while their critics have typically opposed the concept.

It would be helpful for the latter camp to present a detailed case for how we can exponentially increase the pace of the green transition without either making any further reforms to the NEPA or giving any concessions to Republicans in exchange for permitting reforms around transmission lines.

But few of supply-side liberalism’s critics have attempted to make this argument. Rather than demonstrating that priority setting is not necessary for hitting America’s climate targets — or stipulating that hitting those targets is unfeasible or unimportant — they’ve evaded the issue through misrepresentations and pejoratives.

Dayen is not alone in this. The Roosevelt Institute’s Todd Tucker is typically an incisive analyst of economic policy. But some of his critiques of supply-side liberalism do less to rebut that outlook than to stigmatize it.

One implication of Klein’s analysis is that there is a conflict between the interests of some conservationist NGOs and the imperatives of decarbonization. Tucker has signaled opposition to this idea. But he has not explained in detail why the alleged tradeoff between expediting decarbonization and preserving regulatory obstacles to development is illusory.

Instead, he describes Klein’s argument as a “call to a Sister Souljah moment for the planetary emergency.” This is a reference to Bill Clinton’s decision to repudiate an African American activist during his 1992 campaign, in a bid to increase his appeal with moderate white voters. Progressives generally hold that tactic in contempt. Analogizing Klein’s argument to Clinton’s gambit is clearly intended to discredit the former. And yet the analogy makes little sense: Klein isn’t calling for Democrats to performatively disown environmentalists for the sake of winning over voters who resent them. He is saying that achieving the progressive movement’s consensus goals requires implementing a few policies that some environmentalists oppose.

Rather than rebut this empirical claim, Tucker asks rhetorically whether “Klein’s survivor island approach to coalitions — first unions and Dems team up to vote enviros off the island, and then Dems turn on labor” will ultimately move American politics in a progressive direction. Beyond the demagogic tenor of this language (which equates deprioritizing a coalition partner’s objectives with ostracizing them), Tucker’s argument reiterates Dayen’s misrepresentation of Klein: No one is arguing for the Democratic Party to “vote labor off its island.” Positing that the progressive coalition has some internal conflicts and taking a side in one of those discrete disputes is not tantamount to arguing that Democrats should disown environmentalism or pro-labor politics.

The debate over supply-side liberalism is important and touches on many genuinely thorny questions. And Klein’s critics may well have persuasive answers. But when they attempt to discredit the supply-side liberal perspective through misrepresentations and name-calling, instead of forthrightly explaining why there is no risk of Democrats overloading programs with too many objectives or of regulatory processes thwarting timely decarbonization, they do not inspire confidence.

The Case Against Supply-Side Liberalism Is Weak