When PPPs fail

Research has investigated whether or not PPPs generate public value, but as of today, there is no clear, universally accepted metric for measuring the success or failure of such initiatives. For our part, we chose to focus on the latter, specifically asking two central questions: How can we define failure for an infrastructure PPP? What conditions make failure more likely?  

The PPPs we refer to here are agreements for funding, building and managing infrastructure (such as transportation, electricity, telecommunications and water supply systems). These partnerships represent a distinctive form of long-term governance that calls for more private integration into public administration compared to transactional contracts.  Such projects produce illiquid assets (as they have a limited secondary market); they are highly capital-intensive and extremely difficult to evaluate. Inspired or required by reforms associated with what’s known as New Public Management (NPM), PPPs build infrastructure that fosters long-term economic growth, and often they are thought to relieve governments from the burden of searching out alternative financing sources or other operating facilities. PPPs, in this sense, represent a key component of economic and social development strategies. 

Although statistically speaking, contract cancellation is rare with PPPs, the probability of this happening depends on both factors linked to the political environment and specific features of the project in question. With our recent study, we hope to help pinpoint the causes underlying the failure of infrastructure PPPs. To achieve our aim, we used the “Political Coase Theorem,” which holds that political institutions tend to obtain better outcomes for society regardless of what political coalition is in power.  In other words, when there are a number of checks on the work of political players, this increases the likelihood of establishing stable relationships with private suppliers. The end result is a lower risk of contract cancellation.  

We study public-private partnerships for infrastructure projects based on two working hypotheses: 

1. The most extreme case of failure for an infrastructure PPP results in cancellation. As far as measuring failure, using the cancellation of the partnership as a variable has the advantage of being symmetric, verified, and externally validated. 

1. For PPP infrastructure projects, the more “veto points” there are, the less likelihood of failure. What’s more, PPPs have a better chance of succeeding when institutional accords limit the power of political actors to do the following: expropriate the project, renege on their commitments, promote collaborations or initiatives without broad political consensus, or amend legislation. In countries where these institutional protections are lacking, PPPs have a higher probability of failure. Our focus on veto points is an approximative metric for these protections, but it allows us to assess the likelihood of project failure across a wide range of countries. 

To test our hypotheses, we structured our study in two phases.  

1. For the first exploratory qualitative phase, we conducted semi-structured interviews with PPP experts who work in international organizations tasked with funding and promoting these partnerships. We integrated these results with an analysis of documentation and guidelines from these organizations. This phase led us to conceptualize various types of failures, and then to select cancellation (of the partnership) as our dependent variable.  

1. The second phase consisted of a statistical analysis of a data set from the World Bank on more than 4,000 projects launched from 1990 to 2015 in 89 low- and middle-income countries  

Thanks to our findings, we were able to test the theory that the use of “checks and balances” in a given political system diminishes the likelihood of PPP cancellation. Specifically, the main result of our quantitative analysis is that the probability of cancellation drops to near zero as the number of veto points rises relative to the government in question. 

The approach we utilized, based on a mixed-methods research design (qualitative & quantitative), explored the notion of failure and identified the conditions under which it tends to occur. Our results provide robust empirical evidence to support the hypothesis that it is less likely that PPPs will be cancelled in countries that have a greater possibility to control their political institutions and restrain the actions of their politicians. 

To come up with an effective way to measure failure, we drew a distinction between project failure and sources of risk for public and private parties that account for failure. This is where the novelty of our research truly lies with respect to extant academic literature: we shift our focus to measuring the failure of infrastructure projects when one or both partners cancel the underlying agreement. If it’s no longer worthwhile for a partner to participate in a project, they abandon it. (Data show that this is almost always the private partner.) Despite the various limitations we discuss in our article, utilizing the cancellation of the partnership gives us a simple, observable measure of failure.  

Our findings offer a more solid foundation for understanding the role of political risk in evaluating the success or failure of infrastructure PPPs. Specifically, we provide concrete proof that there is a relationship between PPP cancellation and the political context of the partner country. 

Overall, our study presents a useful way to frame the cancellation of infrastructure PPPs as a rare but calculable event, breaking new ground in exploring the reasons behind failures concerning one of the cornerstones of New Public Management.