Keith Srakocic/AP Photo
Workers lay a new stretch of asphalt on Interstate 79 in Cranberry Township, Pennsylvania.
A year ago, Democrats pulled a rabbit from their hat, passing the Inflation Reduction Act and the CHIPS and Science Act to cement the Biden administration’s industrial-policy focus. And for all the sparring over whether the attempt to fulfill multiple policy goals simultaneously would hamper either the clean-energy transition or the effort to reshore critical supply chains, it’s probably best to just look at the results.
There have been $231 billion of IRA and CHIPS-related investments in just the first year, and a projected 170,000 clean-energy jobs. Nearly all of these jobs are being located in areas with weekly wages below the national average, and most of them in areas with poverty rates above the national average. Manufacturing construction is making its biggest contribution to gross domestic product in over 40 years. The boom is at least partially responsible for what the Atlanta Federal Reserve estimates as a 5.8 percent annualized GDP increase this quarter, an almost unthinkable number for a country economic observers thought would be mired in recession a short while ago.
The IRA has even spurred other countries to devise their own subsidy schemes for clean-energy industries. Renewables will overtake coal globally as the largest source of electricity in the next two years, and already there’s substantially more investment worldwide in renewable production than fossil fuel production. As climate is a global challenge, the world is finally pitching solutions at roughly the scale of the problem.
That’s great news for the planet, and for government’s ability to act with purpose. The country is bumping up against some constraints: constraints for skilled workers to engineer and build factories, constraints for equipment and raw materials to get them running, and constraints from the political threat of a Republican takeover of government next year that would likely overturn the entire project. There are answers to these constraints inside the industrial-policy laws: incentives for apprenticeship pipelines to build the workforce, domestic development of things like minerals and manufactured components, incentives for good jobs, and the potential for what Kate Aronoff calls “pool party progressivism,” meaning measures that would provide tangible benefits to communities. Whether that all happens is what we’ve been tracking for the past few months.
Through an elegant study that isolates procurement processes, Liscow and his co-authors demonstrate that increases in government capacity translate into substantial cost reductions.
Others point to constraints in the permitting of clean-energy projects, or constraints from higher wages that reduce how much can get built. That has produced a debate which has generated a lot of heat and substantially less light. But I do think it’s important to identify impediments that can be fixed and trade-offs that can be made to speed the transition. That’s why a research paper released last month demands some more attention.
Zachary Liscow, a professor at the Yale Law School who until recently was the chief economist in Biden’s Office of Management and Budget, co-authored the paper with William Nober of Columbia University and Cailin Slattery of UC Berkeley. It finds significant cost drivers to building in an unheralded place: the procurement process, which accounts for one-quarter of all U.S. government expenditures.
Through an elegant study that isolates procurement processes, Liscow and his co-authors demonstrate that increases in government capacity, which can invite more bidders into a project and foment competition, translate into substantial cost reductions, which subsequently leads to faster and more robust projects. While one shouldn’t translate the study into an overarching monocausal theory for how to accelerate building, it’s a piece of the puzzle that is woefully understudied and should enter the conversation.
THE PAPER LOOKS AT STATE ROAD RESURFACING PROJECTS, which is a good way to reduce the variables on cost overruns. In general, citizens don’t clamor to stop projects that fill potholes, and permitting is usually straightforward. It’s also where the money goes in much of our infrastructure spending: $187 billion was spent on highways in 2018, and little of that on new highways. Yet there is substantial variation in cost. The authors point to the neighboring states of Georgia and South Carolina; Georgia spends about $189,000/mile to resurface on average, while South Carolina spends almost twice as much ($376,000/mile).
To understand why, the authors built three separate datasets. First, they surveyed state Department of Transportation (DOT) procurement officers in all 50 states, as well as contractors for those states. Then they collected project-level data costs state by state. Finally, they used administrative data to account for other cost drivers, like the weather, road usage, and labor costs. From this, they determine what professionals involved in the process think drive costs, and here the numbers stand out.
Between 1997 and 2020, while state public-sector employment rose and population increased, employment of state DOT highway officials decreased by 20 percent. About 88 percent of survey respondents said their department was moderately or severely understaffed.
Because road resurfacing work remains somewhat constant, this has led to a rise of outside consultants to perform the tasks needed to bid out and design projects. “It’s easy to tell a story here, you almost get into a negative feedback loop,” said Liscow in an interview. “You lose a worker and have to outsource more to consultants. That increases demand and pay for consultants. That draws more workers away from the state workforce.” Certainly, private-sector wages have outpaced government payrolls.
States whose DOT officials reported concerns about consultants correlate with significantly higher costs. Consultant-driven projects have fewer details, leading to more changes in project scope, which adds risk to the contractors and delay to projects. Consultants also have less incentive to create competition in bidding, and indeed competition on these resurfacing projects has been shrinking; almost 70 percent of states saw reductions in construction firms for road resurfacing between 2007 and 2017. The lack of manpower also just makes it less likely that a state DOT can encourage new bidders into the process.
According to the paper, just one extra bidder on a project drives costs down 8.3 percent. An increase in bidder outreach, to get more companies to try for a project, translates to 17.6 percent lower costs. And changes to scope of work, which correlates with more outside consultants, is seen as a leading cost driver. Overall, an increase in agency capacity leads to 16 percent lower costs, and a higher-quality workforce, as defined by the state DOT officials surveyed, corresponds to a 30 percent reduction in cost per mile of repaving.
In other words, once the authors control for all variables, the two main factors in expensive road resurfacing projects are weak state capacity directly, plus limited competition, which is also ultimately caused by weak state capacity.
THE PAPER PUTS HARD NUMBERS to a natural set of expectations. If fewer firms are available to work on a project, they’re going to charge more, in part because all the contractors know their competitors and have a good sense of what they will bid. If there isn’t anyone inside the government to deal with the project, there will be delays. If you involve management consultants, they’re going to want to build in their own profit and they won’t be as sensitive to cost overruns. This is all intuitive; now we have some quantifiable data.
It’s important to emphasize what it does not say. It makes no claims about environmental reviews or citizen participation in infrastructure projects. It simply finds that, in this one particular case, which happens to be one of the main things state governments spend money on, capacity and competition are driving factors in whether that spending is done well. There’s evidence outside highways as well; the NYU Transit Costs Project also cited the proliferation of consultants as a hindrance to mass transit projects.
Liscow believes there are lessons to draw. Larger mega-projects “all go through procurement,” he said. “If state capacity hampers road repaving, one would think it would be harder for mega-projects. If it’s hard to find competition for something that’s easy to do, you would think it would be more so for something specific and hard.”
The investment in building capacity pays off in the long term, and as we re-enlist in industrial policy, that will likely become more pronounced.
Perhaps the biggest lesson is that the Reagan Revolution degraded government capacity, and the neoliberal turn in the 1990s, combined with the Great Recession’s savagery to government head counts, kept government permanently hobbled. Even in areas as routine as road repaving, the effect is substantial.
The logic of neoliberalism was that reducing the scope of government would reduce spending. What the paper makes clear is that the result was the exact opposite: increased spending for the things we need government to do. The investment in building capacity pays off in the long term, and as we re-enlist in industrial policy, that will likely become more pronounced.
The IRA did the showiest version of rebuilding state capacity in its investments in the IRS, designed to bring back more revenue over the long term. Behind the scenes, Liscow said that the Biden administration is taking parallel approaches in its industrial policies. The effort to relieve bottlenecks in permitting has focused on “having the right bodies to do the permits,” Liscow said. “I know from being in meetings, there is a laser focus to getting people hired and projects permitted, with appropriate community input.”
That’s an area of implementation that gets overlooked; ensuring a robust and high-quality workforce has material benefits. It would be great to get more research to quantify this, but this paper is a great start. And it offers perhaps an area of agreement in the supply-side debates. Increasing the supply of government manpower, to reduce the tyranny of consultants, to ensure better competition in procurement, and to generally accelerate the pace of government action, pays serious dividends well beyond the budgetary cost.
