Mark Green, the new head of USAID, with a favorite picture from an event he was part of in Africa. (Photo by Bill O’Leary/The Washington Post)

Last month, Ambassador Mark Green became the new administrator of the U.S. Agency for International Development (USAID). Green is widely respected within the global development community and has made clear that business as usual at USAID is not an option.

On his first day inside the Ronald Reagan Building, he told USAID staff, “I believe the purpose of foreign assistance should be ending its need to exist. … Each of our programs should look forward to the day when we can end it. And around the world we should measure our work by how far each investment moves us closer to that day.”

But neither USAID nor any other major aid agency is going to “put itself out of business” until they figure out how to help the world’s poorest countries build stronger government institutions.

This is far easier said than done — and that’s the problem.

Diagnosing weak government institutions is a lot easier than fixing them

There are many reasons to focus on reforming weak, inefficient and corrupt government institutions in failed and fragile states. It is difficult for aid agencies to achieve long-lasting gains in the countries where they work unless their host governments can independently maintain law and order, collect taxes, deliver public services, and create an environment in which the private sector can flourish.

Weak states are also destabilizing to the international order — and a threat to U.S. national security. The 2002 U.S. National Security Strategy concluded that “America is now threatened less by conquering states than we are by failing ones.” Subsequent versions of this strategy document (in 2006, 2010 and 2015) echoed this point, noting that terrorism, complex humanitarian crises, disease outbreaks, and drug and arms trafficking often accompany state failure and fragility.

But there is little agreement about what outside actors such as aid agencies can do to help poor countries build strong institutions. Three leading scholar-practitioners put it this way: “Development projects have, by and large, been successful at building physical stuff: schools, highways, irrigation canals, hospitals and even building the buildings that house government ministries, courts, and agencies. But some building is hard. … [B]uilding capabilities of … human systems — including that human system called ‘the state’ — has proven much more difficult.”

Do aid agencies help or hinder institution-building?

In a new study published in International Studies Quarterly, we argue that aid agencies are actually making the challenge of institution-building more difficult.

We examined a sample of World Bank projects approved between 2003 and 2011 and identified all of the project goals related to building institutions. We categorized each of these 1,280 targets according to whether they measured institutional form or function — that is, whether they measured how institutions are organized, or the ability of institutions to do the things necessary to put international development agencies out of business.

Here’s what we found: countries able to access grants and highly subsidized loans are more likely to select cosmetic measures of success in institution-building. But countries that receive close-to-market-rate loans are more likely to select meaningful measures of institutional performance. Our findings hold true even after we control for many different factors, including per capita income and baseline institutions.

These results call attention to a cruel irony: Many of the countries most in need of more effective public-sector institutions are also the least likely to invest in the creation of such institutions.

Why would development agencies take steps that make the problem worse?

Aid agencies are under intense pressure to produce specific, measurable near-term results. These pressures create perverse incentives for developing countries to pursue shallow but highly visible reforms — for example, the establishment of a nominally independent anti-corruption agency or the passage of anti-human trafficking legislation.

The adoption of cosmetic reforms allows everyone involved — project managers and executive leadership teams at funding institutions, contractors and NGOs responsible for project implementation, and recipient government officials — to “declare victory.” But whether an agency or law exists is not a meaningful measure of success. What really matters is whether laws are enforced and government agencies are able to independently discharge their core functions.

If past is prologue, proceed with caution

The evidence we collected is specific to the World Bank, but there are few reasons to think that USAID is any less vulnerable to unproductive pressures to “game the system.” Andrew Natsios, USAID’s administrator from 2001 to 2006, attempted to “refocus attention on institution building as the central thematic focus of aid programming” during this tenure, and considered it an article of faith that “[a]ll construction or service delivery projects should be subordinate to the larger institution-building task.”

But he learned that Washington suffers from “Obsessive Measurement Disorder,” which creates little space for USAID to focus on programs that will deliver benefits only after many years of patient investment. Here’s how Natsios summarized the essence of the problem: “Those development programs that are most precisely and easily measured are the least transformational, and those programs that are most transformational are the least measurable.” He also warned that the measurement disorder was creating a wide array of “perverse, unintended consequences.”

For these reasons, we think that Green may be walking into a trap. If he acquiesces to the demands of his political, technical and financial overseers in the executive and legislative branches, USAID will find it difficult to prioritize institutional strengthening in weak states — and probably will have to focus instead on programs that produce specific and measurable results over short time horizons.

However, if he focuses on the long game and patiently invests in programs that can create indigenous capacity and produce more sustainable results, USAID will probably face fierce criticism for being insufficiently results-oriented.

Helping weak states begin the long process of getting unstuck is a worthy endeavor. But our research suggests that success will remain elusive until USAID, its partners and its overseers end the “pick-an-easy-target-and-call-it-success” charade.

Dr. Brad Parks is the executive director of AidData, a research lab at the College of William & Mary. You can follow AidData’s work on Twitter @AidData.

Dr. Mark Buntaine is an assistant professor at the Bren School of Environmental Science & Management and the Department of Political Science at the University of California at Santa Barbara.

Benjamin Buch is a PhD candidate in political science at Stanford University.