A Modest Proposal for Improving Supervision in World Bank Infrastructure Projects

Infrastructure funding is a massive component of international development—in 2014, the World Bank alone allocated $24 billion to infrastructure, amounting to roughly 40% of its total lending. Yet as has been widely documented (see here, here and here), infrastructure construction and development projects are particularly susceptible to corruption. Compared with other areas of development lending, such as education and public administration, large construction projects require more specialized contractors and consultants, increasing the points of access for corruption or collusion schemes. Furthermore, labor-intensive industries like construction are often captured by organized crime, which increases their susceptibility to corruption.

Corruption schemes in infrastructure projects often take the following form: a contractor pays government officials a bribe to secure a contract, and in an effort to preserve profits, the bribe-paying contractor compensates for the expense of the bribe by failing to build the project to specification. The supervision consultant—the person or entity responsible for evaluating whether the project has in fact been built to specifications—therefore plays a critical role in stopping or enabling infrastructure construction.

However, when the World Bank funds an infrastructure project, whether through a grant or a loan, the recipient country’s government is responsible for hiring the project’s contractors and consultants—including supervision consultants—subject only to arm’s length World Bank supervision. While this process is also subject to the World Bank’s procurement guidelines, these have been criticized as ineffective in addressing corruption (as previously discussed on this blog). Under the current system, if a project has not been adequately completed because of a corruption scheme, government officials have every incentive to retain inspectors willing to mask the abuse of funds. And if the Bank does discover fraud or corruption after the fact, its remedies are limited: the Bank can suspend or bar contractors from future contracts, and can refer matters to national prosecuting authorities, but successful convictions amount to fewer than 10% of sanctioned parties.

The World Bank must therefore prioritize prevention of these situations. Given the existing system, one measure that the World Bank could take to help prevent corruption in infrastructure projects, is to fund independent supervision consultants.

Consider the following example that, though extreme, indicates the shortcomings of the status quo. In 2006, the World Bank finalized an $84.6 Million loan to the Indian government for a hospital improvement project in Orissa, India to improve the quality of the state’s health services delivery. The completion report was prepared by construction supervision consultants hired by the government of Orissa. Their report indicated that the project’s outcomes were “Moderately Satisfactory,” and certified that 38 hospitals had been completed to specification. In reality, however, when the World Bank sent its own inspectors to Orissa, after being alerted to the situation by locals, it found that  93% of the hospitals funded by the project had problems like leaking roofs, crumbling ceilings, molding walls, and non-functional water, sewage, or electrical systems.

Though the corruption of the hospital project in Orissa occurred 10 years ago, situations like this remain regrettably common. For example, a 2011 World Bank Report admitted that due to fraud and corruption, infrastructure projects like roads and bridges cost far more to build than they should, and are of suboptimal quality. The report suggests some measures for improving the existing procurement process, but does so within the existing framework of the recipient countries managing procurements, which in practice does not differ from the system that led to the egregious abuse of funds in Orissa.

Conversely, a truly independent supervision consultant could have helped to prevent this situation. The looming threat of independent supervision consultants would mean that contractors would not be able to cover their corruption-related costs by failing to construct projects to specification. To ensure their independence, the funding of independent supervision consultants should be altogether separated from the loan or grant funding the project. Currently, supervision consultants are simply a part of the program’s overall administrative costs, administered by government officials. As a separate program directly administered by the World Bank, there could be separate procurement guidelines emphasizing the priority of quality and subject matter expertise of inspectors over minimizing overhead costs.

The World Bank’s country office or regional office should directly manage the hiring of these consultants. In theory, as Bank staff do not directly manage the project funds, they are less likely to be complicit in a corruption scheme, and would have no incentive to hire consultants willing to mask substandard construction. Further, the World Bank can institute sector-wide guidelines ensuring a higher degree of monitoring and reporting to officials without a direct stake in the project. Consultants directly hired by the World Bank would also be subject to greater World Bank control, and it would be simpler for the World Bank to discipline them or terminate the contract through internal procedures.

Admittedly, hiring independent supervision consultants would add to the Bank’s operating costs, and the use of such independent consultants is no panacea. After all, there is no guarantee that these consultants could not be captured or coerced by corrupt parties into submitting fraudulent reports. And for the World Bank to insist on its own independently-procured supervisors could be viewed as a violation of the implementing country’s sovereign authority, leading to resistance to this plan not only by member governments, but also by the World Bank’s own regional officers, who might view this plan as encroaching their domain. Nevertheless, on balance this plan would be a marked improvement over the status quo. Shifting the procurement of supervision consultants away from recipient governments fundamentally realigns incentives, and would likely reduce cover-ups abuse of funds in infrastructure projects. Though this proposal might generate significant bureaucratic pushback, the ultimate goal of the World Bank is to alleviate extreme poverty, and any effort that can concretely reduce the abuse of funds of development funding is worth pursuing.

8 thoughts on “A Modest Proposal for Improving Supervision in World Bank Infrastructure Projects

  1. Great post, Dan! I definitely agree with the suggestion of funding independent supervision consultants as a solution for potential corruption, collusion, or fraud. However, such initiative has to be coupled with other policies initiated by governments, including reforming the country’s Competition Law (and its Anti-cartel Regulations) and ultimately increasing the level of supervision over the procurement process (e.g. through mandatory pre-bidding meetings, complaint mechanisms that are open to contractors and public, and even greater participation by civil society monitors).

    The World Bank published http://siteresources.worldbank.org/INTDOII/Resources/Roads_Paper_Final.pdf in 2011, one of the examples of short-term and long-term plans to fight corruption, collusion, or fraud in the roads sector.

  2. Creating separation between the implementing actors and WB supervision makes a lot of sense. However I wonder if it does not cause duplication with good practice monitoring and evaluation. Why not expand the role of the monitoring by the WB of its loans to include more rigorous corruption reviews?

  3. I love this suggestion. It seems like such a straightforward way of addressing what is apparently a significant problem. The way you present everything here makes it seem so eminently…reasonable: here’s a real problem, here’s a simple thing we can do to address it. I really hope this post gets the attention it deserves.

    (Your reference to possible sovereignty concerns makes sense as a way to cover your bases/acknowledge possible pushback, but at least in my eyes, it’s a fair part of the bargain when accepting a World Bank project.)

    • I agree, Katie – Daniel, your good recommendation is highly reasonable. However, I would not underestimate the sovereignty concerns and I think that, ultimately, they really do jeopardize the viability of the proposal. I would think, like you, Katie, that accepting certain restraints on government project management control is a “fair part of the bargain” for funding. After all, the Bank’s current ability to audit is just such a concession. But, for several reasons, the Bank has an interest in maintaining the current structure. First, the Bank is ultimately comprised of governments. This odd director/client status of country governments vis-a-vis the Bank makes for some perverse incentives and seemingly irrational limits on the Bank’s power. It is one reason (among many others) the Bank effectively can’t sanction government officials/departments that may be repeat bribe solicitors. Of course, as he says, Daniel’s proposal is modest and doesn’t go anywhere near that far. But the point is that the Bank can only demand so much before it starts to step on powerful toes. Second, the Bank does not contract directly with providers and I wonder if doing so would augment liabilities and/or transaction costs to an unacceptable degree. As Daniel and others have pointed out in interesting ways, there are creative ways to control these costs. But the World Bank is a bank, not a development company (in the infrastructure sense). Third, although the Bank’s mission is poverty alleviation and a means of achieving that end is through infrastructure development, another important component is capacity building. While it seems absurd to empower corrupt government departments to continually manage funds in a corrupt manner, I’m not sure the solution is to take that power wholly away from them. Having the Bank more involved in projects would be a very good, potentially necessary, solution in the short run but what about the long term?

      Don’t get me wrong, Daniel, you point to a very real, painfully apparent issue and you offer a solution that would help a lot. For the practical and normative reasons I discussed above, though, I may side with Cheyanne and say that we should strengthen the Bank’s existing monitoring power. The Bank could borrowing from other models of contractual relationships between lenders, contractors, and governments in fields like oil and gas. For one, it might consider including an approval provision in its contracts (I feel like this already exists for certain projects but I’m not sure on the details). It could also create a pre-approved list of supervisory engineering firms. The Bank could also include a clause that allows the lender to step in in instances of material breach. These ideas are half-baked at the moment but there could be a lot to learn by looking at more structurally risk-averse lenders.

  4. Pingback: A Modest Proposal for Improving Supervision in World Bank Infrastructure Projects | Anti Corruption Digest

  5. Like the rest of the commenters, I really like this suggestion. I also think it’s interesting that the example you give was discovered because of reporting by the locals. Is there anyway to enhance the reporting channel between the people these projects are designed to assist and the World Bank? Maybe then independent inspectors could be sent in in a targeted fashion, which could (potentially) cut down on cost. I assume something like this is already being done, but maybe it could be escalated?

  6. Like Courtney, I think there is an opportunity to embrace this proposal at a lower cost than your post might suggest. The very possibility that an independent supervision consultant might review a project could reduce corruption on the margin, even if no consultant is ever assigned to the project. Alternatively, the World Bank could assign independent supervision consultants to every project of a certain size for a few years, and then alter its practice to only assign consultants randomly to a certain percentage of those projects after that period of time. A contractor would have to factor in the risk that its project would be supervised by an independent consultant, which might deter corrupt behavior going forward. Of course, targeting the use of consultants to certain projects (as Courtney suggests), or adopting a randomized assignment program to reduce costs runs the risk of that assignment system being corrupted and defeating the purpose.

    • As others have said, this is a great suggestion. It seems to me some amount of independent oversight could be implemented for less than the amount of cost savings it would realize. Nathan, your suggestions about how to make the proposed oversight lower cost yet effective are insightful. On the random model, I would wonder how the risk of (and potential punishment because of) the consultant visits could be adequately communicated, particularly to those who may be participating in a project funded by the Bank for the first time. Certainly knowing of some risk is better than knowing there is virtually no risk.

      On another note, the way I pictured this in my mind included massive infrastructure works, like bridges, literally crumbling after poor-quality construction. Where Bank-funded hospitals and roads both support lives and could threaten lives if completed in a haphazard way, there seems to be a stronger imperative for implementing a system of supervision. What type of impact would a tragedy that could have been prevented with oversight have for the credibility of the World Bank and its work?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.