Introduction and Contribution

American bureaucracies are frequently criticized for impeding the delivery of both public infrastructure and private development projects, wrapping them in the “red tape” of regulatory procedures. Of course, bureaucratic oversight is necessary to ensure public accountability, enable stakeholder input, and respond to market failures. It is hard to imagine how, for example, the natural environment could be protected from exploitation without regulation. Yet disagreement persists about whether the current regulatory landscape strikes the right balance between oversight and efficiency.

The National Environmental Policy Act (NEPA) of 1970 is one of the most significant federal administrative laws and is foundational to American environmental law. Per its objective of protecting the environment, NEPA requires agencies to assess the environmental impacts of major federal projects. These include construction, energy, water, and public lands projects. 

NEPA’s highest level of review is the Environmental Impact Statement (EIS), which requires substantial time and scientific expertise to complete. Accordingly, the EIS process has been criticized for causing project delays. This critique is plausible at first glance: average EIS reviews now take over four years to complete and often produce documents over 600 pages long (over 1,000 pages including appendices). 

Despite criticism over NEPA procedures, little is known about why some EISs take longer to be approved than others. Answering this question is important for assessing whether American bureaucracies can adequately balance the oversight-efficiency tradeoff.

In “A hazard analysis of federal permitting under the National Environmental Policy Act of 1970,” Michael Bennon, Daniel De La Hormaza, and R. Richard Geddes present findings on the correlates of EIS permitting. Using a dataset of nearly 1,300 EISs, the authors show that major construction projects and those occurring in states with more restrictive environmental laws are associated with longer permitting times. By contrast, privately financed projects are permitted more quickly. These findings will likely cohere with readers’ intuitions, namely that (a) more demanding regulations increase approval times and (b) private economic interests have the resources to navigate regulatory environments.

The authors’ findings are relevant not only for those interacting directly with NEPA. Indeed, as illustrated in their analysis of the 2009 American Recovery and Reinvestment Act, lengthy environmental reviews can make it difficult to quickly undertake new infrastructure projects when they are needed — as was the case following the Great Recession.

Data, Methods, and Hypotheses

The authors’ dataset includes 1,269 EIS reviews. Of these, 1,161 reached a Record of Decision (ROD), the federal agency’s final decision. The average duration of time from a Notice of Intent (NOI — i.e., the intent to undertake an EIS) to ROD was 4.5 years, and the median was 3.6 years. This underscores the time required for projects to complete environmental review under NEPA.

The authors include variables for (1) major construction and infrastructure projects, which comprise about half of the data set; (2) projects backed by private investors; (3) states with restrictive environmental laws, defined, for example, in terms of whether uninjured parties can challenge projects in court or whether private actions require environmental review; and (4) inclusion in the “federal permitting dashboard,” which was created in 2015 with the goal of improving the efficiency of large projects.

To measure the correlates of EIS permitting, the authors use a “Cox proportional hazards model,” a statistical method for estimating how different variables affect the time until an event happens. For example, hazard models are often used to measure how different medical treatments affect patient mortality. For the authors’ purposes, a lower hazard rate indicates a longer permitting duration. 

The authors first hypothesize that larger projects will take longer to approve. This may be the case because such projects tend to attract greater stakeholder engagement, opposition, and litigation. Projects in states with more restrictive environmental laws, as well as those requiring supplemental EISs, are also hypothesized to take longer to approve. In addition, private financing ought to lead to faster approvals because private funders have the incentives and resources to expedite permitting. So too should the permitting dashboard decrease permitting times, consistent with its efficiency-maximizing objectives. 

Findings

As mentioned, the authors find that major construction and infrastructure projects have significantly longer permitting times, averaging 5.2 years.
 

Conversely, privately financed projects are indeed found to be approved more quickly, taking an average of just 3.7 years. Meanwhile, projects in states with restrictive environmental laws, such as California, New York, and Washington, are found to take longer. Contrary to the authors’ expectations, however, projects included in the federal permitting dashboard face longer permitting times. This may reflect selection bias, or the fact that projects placed on the dashboard tend to be especially large or complex. In terms of page count, longer EISs are associated with longer permitting times and longer delays. 

Finally, the authors examine permit completion rates by federal agencies, finding an average of 3 RODs per week. This rate is generally stable over time, though it spiked during Donald Trump’s election in 2016 and his inauguration in early 2017. Ultimately, the findings suggest that permitting duration is shaped not only by technical regulations but also by differences in economic resources, political conflict over projects, and federalism.

*Brief prepared by Adam Fefer.

Bennon is a research scholar at CDDRL and program manager of its Global Infrastructure Policy Research Initiative. He also teaches Global Project Finance to Stanford graduate students. He led sessions on public-private partnerships, infrastructure regulation, project assessment and safeguards, corruption, and energy-sector reform, linking research on infrastructure governance and finance to the choices practitioners face when developing and overseeing projects.

Participants applied these concepts to real-world infrastructure cases from Indonesia, Uganda, Canada, Brazil, Uzbekistan, and Ecuador, taking the roles of public officials and investors to work through contested procurements, project turnarounds, public pushback, and corruption. Rather than a single correct answer, each case asked participants to weigh political constraints, institutional capacity, and the competing interests of governments, financiers, private partners, and local communities.

Group work is a hallmark of LAD training. Each day, teams set aside time to work through a real-world PPP or infrastructure problem brought by program participants, using the week's lessons to build an implementation plan. Teams presented their recommendations to faculty and peers on the final day.

Participants said they expect to apply what they learned through the LAD program to their everyday work, and many pointed to the range of experience in the room as a resource in itself. One participant described the week as providing both practical frameworks and a network of professionals facing similar challenges. Another, reflecting on two decades in infrastructure finance, said the course served as a reminder of how much there is still left to learn.

The IFC-Stanford cohort now joins LAD's global community of nearly 2,700 alumni from 116 countries. Across 88 workshops organized with 31 institutional partners, LAD has built a network of government officials, international development specialists, and private-sector leaders who continue to learn from one another as they design policy and carry out reforms. That network gives this cohort a source of experience and advice as they apply the week's lessons to infrastructure decisions in their own institutions and countries.

Introduction and Contribution

Economies depend on infrastructure: roads enable the movement of goods and workers, electricity powers businesses and factories, and communications technologies allow for the exchange of information and services. Yet in our connected world, roughly 1 billion people live more than two kilometers from an all-season road, and approximately 1.2 billion have no access to electricity.¹

For decades, institutions such as the World Bank and McKinsey Global Institute have promoted the idea that closing a trillion-dollar gap between planned global infrastructure investment and the estimated level of investment needed to drive growth will alleviate poverty, harness the potential of developing nations to increase global GDP, and enrich foreign investors. But despite the fact that unrealized gains from greater infrastructure investment in emerging market and developing economies (EMDEs) outstrip those in the rich world, comparatively little capital flows into the infrastructure of poor countries.

In “The Global Infrastructure Gap: Potential, Perils, and a Framework for Distinction,” Camille Gardner and Peter Blair Henry reject the received narrative that flooding poor countries with capital simply to reach a target level of infrastructure investment will yield the desired gains. Instead, they urge investors and policymakers to move from thinking about infrastructure “gaps” to prioritizing infrastructure returns. Specifically, they propose the first “dual hurdle” framework. The framework uses cross-country differences in returns to identify efficiency in cross-country investments, incorporating the incentives that suppliers and allocators of capital must have to finance productive projects.

Applying the framework to historical data, Gardner and Henry debunk the notion that poor countries offer ubiquitous opportunities for productive investment. Only 13 percent of countries in the data set cleared the dual-hurdle tests of efficiency for projects in both roads and electricity. But for countries that did clear the hurdles, the return on infrastructure was large: more than 10 times greater in some cases than the return on rich-country private capital.

This underscores the need for updated and more accessible data to drive decisions and prioritize projects. Differences across countries, sectors (e.g., roads vs. electricity), and locations (urban vs. rural) must be taken seriously to avoid a wasteful allocation of global infrastructure capital.

Gardner and Henry expand our understanding of when cross-country infrastructure investments are efficient and why poor countries have struggled to catch up. And as labor forces in poor countries grow rapidly, the problems associated with infrastructure scarcity are likely to become more acute, increasing migration pressures. Effective investment can mitigate risks while generating mutual gains for poor and rich countries alike.

When Infrastructure Investment Pays Off

The authors’ dual hurdle framework reveals that public infrastructure investment in a poor country is efficient only if it yields higher returns than the social rate of return on private capital (a) within the country and (b) in rich countries. Countries that clear both hurdles are deemed strong candidates for more investment.
 

Image

Figure 4. The Dual-Hurdle Framework

Note: For a given poor country and type of infrastructure, the dual-hurdle framework sorts each country-infrastructure observation into one of four quadrants according to whether it clears the hurdle for efficient: (a) domestic investment, and (b) foreign investment.
 

Gardner and Henry apply their framework to 53 countries that belong to the only known data set of comprehensive cross-country estimates of the social rates of return on infrastructure (roads and electricity) in poor countries, found in a paper commissioned by the World Bank that is based on data from 1985.² In their analysis, Gardner and Henry show that while some countries have opportunities for high returns (especially for paved roads), many others do not. In very few countries can the case be made for investing across multiple kinds of infrastructure (only 7 of 53 countries cleared the dual hurdles for both roads and electricity), and for some countries there is no case to be made.

To be clear, infrastructure investment is not generally inefficient. Indeed, worthwhile projects to increase or upgrade infrastructure in poor countries can yield much higher gains than are available from private investment in both poor and rich countries. This is because infrastructure improvements raise productivity across the economy. The problem is that these high-return opportunities are difficult to identify, especially without updated, readily accessible, quality data.

Different types of infrastructure also vary in their impact depending on the context. For example, investments in electricity often yield low returns in rural settings but high returns in urban settings (where firms are concentrated and struggle due to unreliable power). By contrast, roads more often yield higher returns because they support market activity overall. Finally, the authors show that efficient infrastructure also depends on the interaction of private capital, human capital, and good institutions. This underscores the limitations of simply building more infrastructure in isolation. 

Challenges to Infrastructure Investment

Gardner and Henry document extensive variation in efficiency when comparing private versus public capital allocation (in roads and electricity) across rich and poor countries. The social rates of return on public capital vary much more than with private capital, which is the case in both rich and poor countries, and inefficiencies in public capital allocation are greater in poor countries than in rich ones. This variation is driven not just by economic factors but also by political and institutional constraints.

The concept of “frozen capital” helpfully describes the outcome of problems facing prospective investors. For one, because infrastructure requires complementarities (e.g., electrifying an area requires the preexistence of reliable roads), individual projects may be unprofitable unless others are simultaneously financed and coordinated.

Additionally, foreign investors may distrust the governments of the countries where they invest, or they may be wary of the quality of domestic firms tasked with completing a given project. This will certainly be true where transparency and accountability are lacking. Especially in poor autocracies or semi-democracies, governments may arbitrarily change the law, fail to honor contracts, or face political and social instability, leaving investors uncertain about the prospect of repayment and return on their investments.

Ultimately, infrastructure investment offers great promise, but it must be evidence-based and approached with caution. The challenge is not to simply raise the level of investment but, as “The Global Infrastructure Gap” shows, to use cross-country differences in returns to identify the right infrastructure projects, in the right places, to maximize the efficiency of cross-country capital allocation.

*Brief prepared by Adam Fefer.

_References:

_{1. Canning, David, and Esra Bennathan. 2000. “The Social Rate of Return on Infrastructure Investment.” World Bank Policy Research Working Paper 2390.}

_{2. Rozenberg, Julie, and Marianne Fay, eds. 2019. Beyond the Gap: How Countries Can Afford the Infrastructure They Need While Protecting the Planet. Washington, DC: World Bank.}

gaps persist between stakeholders. This is exacerbated by the lack of clear, standardized specifications from NATO, leading to inconsistent implementation and operational friction. Second, at its core, this is an intelligence-sharing challenge: member states operate under different national frameworks, threat perceptions, and priorities, which influence what information they are willing (or unwilling) to share. Third, the fragmentation of sharing systems is not merely a technical hurdle but often a deliberate choice made for operational and security reasons, reflecting concerns over sovereignty, data protection, and strategic advantage.

While technological advancements can enhance interoperability, they alone will not drive adoption. Our research highlights that the underlying issue is one of trust and incentives—NATO must establish mechanisms that encourage collaboration beyond just technological solutions. Without a strong foundation of mutual trust, transparency, and shared benefits, even the most advanced systems will face resistance. Creating sustainable incentives for participation—whether through policy alignment, risk reduction assurances, or value-added intelligence sharing—will be essential in fostering a more effective and unified cyber defense posture within NATO.

A key component of our project involves conducting in-depth research and stakeholder interviews focused on Mongolia's "Billion Trees Initiative," where we will be seeking actionable insights to scale up the initiative effectively and sustainably.

influencers work with politicians to coordinate these influence operations and shape public opinion. Our policy recommendations will address how governments and civil societies can work together to tackle this issue.

Australian Space Agency and U.S. Defense Innovation Unit, the goal of this project is to leverage Australia's strategic geographic position and investment in reentry infrastructure to mitigate the hurdles that U.S. commercial startups experience accessing military ranges for reentry. Our project aims to create a robust foundation for the development of orbital return capabilities in Australia, fostering greater commercial and national security collaboration between the U.S. and Australia.

to computational resources acts as a barrier to adoption. India has a thriving tech sector, and openly available AI models have the potential to democratize access to this trailblazing technology. At the same time, AI is expensive to build and deploy, and access to the specialized computational resources needed to do so is limited even for top Indian companies. Our team aims to develop solutions in partnership with Digital Futures Lab that can help bolster the AI ecosystem across India.

that contribute to this issue. During our fieldwork, we will explore community- and government-driven programs aimed at preventing and combating trafficking. Our final report will analyze existing policies, pinpoint gaps, and propose community-led interventions to address them.

The "Meet Our Researchers" series showcases the incredible scholars at Stanford’s Center on Democracy, Development and the Rule of Law (CDDRL). Through engaging interviews conducted by our undergraduate research assistants, we explore the journeys, passions, and insights of CDDRL’s faculty and researchers.

How has infrastructure development been used to gain influence in diplomacy? How has our understanding of that tool changed since BRI, and how successful has it been for China?

Infrastructure development has always been a problematic tool for amassing geopolitical influence; it builds friendships when loans are going out, then creates enemies once they’re issued. A recent example is the 1997 Asian financial crisis when Western countries had invested in power plants throughout the continent, only for many countries to default and expropriate. This has happened repeatedly throughout history.

While China’s done quite well at protecting its economic interests in infrastructure projects, it's a mixed bag due to the prevalence of moral hazard, public backlash, and the tarnishing of diplomatic ties. With the state being so heavily involved in BRI, China intervenes when countries want to default or expropriate, protecting its interests and those of state-owned enterprises effectively. However, this can lead to a moral hazard problem because these enterprises feel too protected by China and act without the appropriate caution while building risky projects.

Today, many countries that have received BRI lending have serious relational problems with China, if not at the government level, then among the public. People tend to push back and feel taken advantage of when a foreign country comes in and builds projects, especially with rumor mills churning out narratives about China’s 'debt-trap diplomacy.' These diplomatic challenges were true long before the BRI and persist today.

Why do countries, through BRI or other means, decide to take on infrastructure projects they obviously can’t afford?

Countries often don’t behave rationally — politics, corrupt officials, and conflicting interests all affect policymaking. Also, everyone builds infrastructure projects that may bankrupt them, partly due to an ingrained optimism bias in the infrastructure sector.

We’re in the worst developing country debt crisis in modern history, and countries are having a difficult time navigating a changing infrastructure lending landscape. China is now the largest bilateral lender, and its absence from international organizations like the Paris Club prevents the unified action needed to allow countries to emerge from debt crises. Even the International Monetary Fund (IMF) is struggling to help them, as it is cautious about issuing aid to countries with murky BRI loans to pay back.

Funding for infrastructure development can be used as an incentive for democratization through conditionality on loans. However, many countries are turning away from traditional Western lending institutions in favor of alternative lenders with fewer conditions. How can we balance the importance of conditionality and incentivizing democratization while preventing the decreased reliance on Western institutions?

Conditionality can be positive in promoting democratization, but there have to be limits to it, especially because it becomes less effective when alternative lenders like China exist. Conditionality began as limited to policies that promote democratization, development, and liberalization but has metastasized to the point where lenders are pushing a wide range of policies on borrower countries. Many of these conditions, such as environmental or social protections, are good policies but can be viewed as a manifestation of Western imperialism within these countries. These programs also become futile when countries become simply incentivized to seek Chinese loans instead, which have virtually no conditionality.

Is the turn away from Western lending institutions an inevitable shift, or can policy changes encourage its prodominance again, if that’s something that we want?

Western institutions are better for infrastructure development, as organizations like the World Bank are the best at protecting human rights and preventing environmental disasters. There are also strategic and security reasons for promoting Western institutions — for example, we don’t want Chinese technology companies building telecommunications grids in other countries.

The bigger question is, what would a return to a Western-dominated model of investment look like? Pre-BRI, there was an open, liberal system of direct investment from private companies. BRI represented a pivot to more state-driven investment. Should the US shift to a similar model, or return to private direct investment fueling infrastructure development? The Biden administration’s alternative to BRI for state-driven investment was the Partnership for Global Infrastructure and Investment (PGII). Despite mutual investment in telecommunications and renewable energy, PGII focuses on developing very different sectors than BRI, building social impact projects like healthcare infrastructure.

What is the most exciting or impactful finding from your research, and why do you think it matters for democracy, development, or the rule of law?

I’m focusing on how liberal democracies can get building again, so I researched flaws in domestic infrastructure projects within the US. It revealed how the judicial system was an engine fueling how infrastructure projects are conducted; I realized the extent to which permitting regulation and environmental litigation had been driving my own incentives when I was a bureaucrat. Decisions are often made in response to case law and to ‘litigation-proof’ projects, which can incentivize inefficient and expensive project management. I believe democracies are perfectly capable of building infrastructure projects well, but problems in current building initiatives, from the California High-Speed Rail to our housing crisis, are rooted in the outsized effects of the threat of lawsuits.
 

How do you see your research influencing policy or contributing to real-world change?

I do research on practical public-private partnership policy in the United States, working with the Build America Center and the Department of Transportation to directly supply the government with research when needed.

There are policy changes that must occur to promote effective infrastructure development. The US will have to reform institutions that predated BRI to adapt to today’s post-BRI world. The three key institutions are the World Bank, the IMF, and the World Trade Organization (WTO). I hope that my ideas can influence their restructuring. For domestic development, I’m continuing my work with the Build America Center on how governments can more efficiently procure infrastructure projects and help public officials adopt best practices.

Lastly, what book would you recommend for students interested in a research career in your field?

The first book, which is extraordinarily boring but crucial to infrastructure development, is The Strategic Management of Large Engineering Projects: Shaping Institutions, Risks, and Governance. Written by Miller, Lessard, Michaud, and Floricel, it includes the perspectives of MIT engineers on infrastructure project case studies to understand why so many have failed. For some great history, the economist Raymond Vernon’s book Sovereignty at Bay develops the idea that relationships sour over international investment and that it’s not an effective diplomatic tool.