Quantifying the Effects of Corruption on the Water and Sanitation Sector in Latin America and the Caribbean

Adam, I., Fazekas, M., Regös, N. and Tóth, B. (2020). Quantifying the Effects of Corruption on the Water and Sanitation Sector in Latin America and the Caribbean. Technocal Note No IDB-TN-02055. Inter-American Development Bank, Washington, DC.

The importance of transparency and governance as determinants of efficiency, effectiveness, and quality of service undoubtedly occupy a key place among the most relevant lessons learned from the reforms implemented in the Water and Sanitation sector (W&S) in Latin America and the Caribbean (LAC) in the last two decades.

In order to support data-driven policy reform aimed at increasing efficiency in public investments, the present analysis focuses on estimating direct financial costs in terms of contract award prices and direct social costs in terms of project delivery quality (measured as frequency of delays and cancellations). Given data constraints in LAC W&S, the following questions are explored by this study:

  • What are the scale and types of corruption affecting W&S services?
  • What is the effect of corruption in terms of tender completion
  • What is the financial impact of corruption for W&S services providers, such as high cost of infrastructure development?

These questions were investigated using data from six LAC countries– Colombia, Ecuador, Jamaica, Mexico, Paraguay, and Uruguay– covering the 2006-2018 period, albeit data availability varies by country (see data section for details). Selection of countries was based on data availability and quality, while also aiming to have broad geographical coverage of the region.

In order to measure the impacts of corruption in the W&S sector, additional data was collected on two main outcome groups identified in the conceptual framework:

  • prices (unit prices or relative prices) and
  • quality (delays and cancellations).

However, each of these outcome indicators was only available in one or two countries and the datasets are of varying quality, with Jamaica having the less complete dataset and Mexico only containing federal contracts, missing local W&S contracts (this is a major shortcoming, as local water utilities are among the biggest spenders in the sector). As a result, the findings are not comparable across countries; however, considered together they give an overview of the different impacts corruption can have on financing the W&S sector.

The methodology for proxying corruption was centred on a composite Corruption Risk Index (CRI) capable of providing a comprehensive assessment of corruption risks in the W&S and other sectors. This methodology has been applied to over 40 countries around the world, suggesting its applicability to other countries in the LAC region where appropriate administrative data is available. The CRI methodology rests on identifying a range of risk factors, validity testing them and pulling them together into a composite score. Risk indicators (or red flags) considered included noncompetitive, direct procedure types or single bidding. Similar to data on outcomes, the list of corruption proxies available varied per country, making direct cross-country comparisons unreliable.

Due to the varying availability of data and the range of available variables in the six countries covered, the combination of corruption risk indicators is different from country to country. In addition, the proportion of different types of contracts also differs across countries (e.g. some have far fewer construction contracts). Despite these differences, applying the same corruption risk assessment methodology provides valuable insights. For example, it is informative to compare countries according to one widely available elementary risk indicator, “single bidding”, which has already been used in cross-country contexts in Europe (Fazekas & Kocsis, 2017). It is shown that while contracting risks are significantly lower in the W&S sector in Colombia, Ecuador, Jamaica and Uruguay, compared to the whole public procurement sector, the share of single bidding tenders in W&S in Mexico and Paraguay is slightly above their national averages. This finding points out the need for a further analysis in these countries to
help develop recommendations improving the governance of procurement processes in the W&S sector

Full PDF