POVERTY & EQUITY NOTES AUGUST 2018 · NUMBER 8 Distributional Effects of Investments in Road Infrastructure: The case of Colombia’s 4th Generation Concession Program Carlos Rodríguez-Castelán, Eduardo A. Malásquez, and Jorge Franco Estimating ex-ante distributional impacts of road infrastructure is increasingly important to better understand the poverty effects of these investments. This note presents such analysis for the ongoing 4th generation (4G) road concessions program in Colombia, which involves the concession of 40 new roads and the transformation of 8,170 kilometers of road network. An ex-ante evaluation of the program suggests that 180,000 jobs will be created and that moderate and extreme poverty would decline by 0.5 and 0.3 percentage points, respectively. Improving transportation infrastructure contributes to many developing countries because of poorly designed enhance competitiveness, promote economic growth, roads. reduce unemployment, and decrease socioeconomic gaps by easing access to education, health services and The Government of Colombia (GoC) aims to transform opportunities (Brenneman and Kerf, 2002; Calderon and its road networks through the fourth generation (4G) of Serven, 2010; Titheridge et al., 2014). road concessions. The project includes the concession of 40 new roads and the transformation of 8,170 Businesses located in areas with poor transportation kilometers of road network, with more than 1,200 infrastructure tend to be less productive, meaning less kilometers of double roadways, and a planned total employment and lower wages for their workers in the investment of $47 billion COP (US$24.4 billion). long run. However, investing in roads can lower transportation costs due to higher fuel consumption This note discusses the methodological approach to and delays due to congested traffic, saving significant assess the potential distributional impacts of the 4G time and effort for users, and freeing time and energy program launched by the GoC in 2012. These estimates to be spent in other purposes. In turn, lower were included as part of the Poverty and Social Impact transportation costs can reduce the prices of goods and Analysis (PSIA) of recent developing policy financing services relative to income for low-income families, projects such as the Productive and Sustainable Cities making goods and services easier to access. Moreover, II 1 and the Second Sustainable Growth and Income improving the design of roads for higher traffic safety Convergence.2 In addition, the 4G project is an example can lead to a decrease in traffic-related accidents of the Maximizing Finance for Development (MfD) (Brenneman and Kerf, 2002), a frequent problem in strategy promoted by the World Bank Group.3 1 3 Colombia - Development Policy Loan for the Second Programmatic Mobilizing private finance for development in Latin America and Productive and Sustainable Cities Operation, World Bank, 2014. the Caribbean, World Bank, 2018. 2 Colombia - Second Programmatic Sustained Growth and Income Convergence Development Policy Loan (DPL) Program, World Bank, 2015. from the localization of the new infrastructure. In Ex-Ante Evaluations of the Distributional addition, the type of transport infrastructure being Impacts of Investments in Infrastructure provided can also have unequal impacts across firms. For example, some firms may benefit more from the construction of railroads than from the construction of Ex–ante analysis and evaluation of distributional highways if the latter provides a more expensive impacts involves quantitative techniques to predict the alternative for their distribution chain; while some firms effects of policies on poverty and inequality may benefit more from having access to highways than (Bourguignon and Pereira da Silva, 2003). The analysis to railroads. Thus, the importance of identifying the allows to assess what would happen if certain policies different users and potential beneficiaries of each type were introduced or some shock affecting the economy of transport investment, and to consider the was bound to happen—e.g., an improvement in road externalities firms will face as a result (Eugenio, 2010). infrastructure. An ex-ante analysis often requires the use of a model to The Case of Colombia: The 4G Network create a counterfactual sample, which intends to Investment Project simulate the population of interest under the counterfactual scenario in question. The simplest arithmetic approach to perform an ex–ante analysis In 2012 the GoC launched the 4G road concessions does not assume any “policy response� by the agents , (Figure 1) to improve the existing national road network such as a pure accounting approach that computes only through public-private partnerships and help close first-round effects and disregards any second-round Colombia’s infrastructure gap. The 4G project involves effects attributable to behavioral responses. In contrast, the concession of 40 new roads and the transformation behavioral approaches take second-round effects into of 8,170 kilometers of road network, with more than account. For instance, an individual may decide to work 1,200 kilometers of double roadways and a planned more if new transport infrastructure improves his/her total investment of $47 billion COP (US$24.4 billion). connectivity to the city; parents may decide to send Estimates from the GoC suggest that with increased their children to school or college if new urban infrastructure investments the country’s economy could transportation reduces travel time and is cheap enough reach a long-term growth trajectory of 5 to 5.3 percent, to reduce the opportunity cost of attending school. as opposed to 4.6 percent without the 4G program. Heterogeneous impacts The GoC identified a pipeline of road transport The geographic localization of routes and the dynamics concessions (approximately 8,170 kilometers of the of the local markets and social structures partially primary road network), mandating the application of determine the size of the effects of new infrastructure the following principles: (i) efficient project structuring on socioeconomic outcomes (Eugenio, 2010). New to accelerate investments in infrastructure; (ii) a transport infrastructure can affect the localization of selection process that promotes transparent firms and workers, and increase or reduced participation; (iii) contract management focused on concentration of workers in different areas. Changes in results; and, (iv) adequate risks assignment between the the spatial distribution of labor could lead to changes in public and private sector. As part of this policy, the GoC the spatial location of goods and services provision. has also identified a set of “early wins� (4 projects, Infrastructure investments could display heterogeneous representing 551 kilometers), which were advanced in impacts across socioeconomic groups, due to individual terms of structuring and implementation, to generate preferences for transportation or externalities derived August 2018 · Number 8 2 demonstration effects and secure private investor employment, growth and productivity. The project is appetite for subsequent projects. estimated to generate from 180,000 to 450,000 jobs that will be distributed across the 24 departments where Figure 1: Poverty effects of Colombia’s 4G road network the different corridors will be built. However, the micro-simulations conducted used the lower bound-conservative estimates (180,000 jobs). The 4G infrastructure investments could be classified in two groups: (i) Short-term investments (ongoing investments that are expected to be developed in a 2 to 5-year window), and (ii) medium- term investments. According to DNP (2013) the first will create around 32,000 jobs, while the second will create approximately 148,000 jobs. The results of these simulations are based on data of the 2012 labor force survey (GEIH). The micro-simulation model distributes the total number of jobs at the national level by Departamento using as a driver the proportion of road kilometers of 4G in each Departamento relative to total kilometers planned in the entire 4G project. To assign jobs to the Source: Rodriguez-Castelán et al. unemployed a logistic model is computed for each (2015) Departamento. The model uses as a binary dependent variable that defines if an employed individual belongs The resulting investments in infrastructure are expected to the construction sector. Based on the model’s to have a direct and positive impact on the economy, coefficients the likelihood of being hired by the boost public-private partnerships, and employment construction sector is computed for the generation across the country, particularly in the unemployed. Finally, two variables are used to construction sector. In addition, the investments are determine the draw of the jobs by Departamento. The expected to have a positive impact on the reduction of first is a binary variable that flags unemployed inequality by increasing connectivity, boosting tourism individuals that have previous experience in the and jobs in traditionally isolated regions, and improving construction sector, and the second one is simply the access to markets and the supply of quality public probability of being hired. The micro-simulation sorts services in rural areas. The overall positive impact on by both variables; it will hire the first unemployed who growth and income inequality suggests infrastructure have experience in the construction sector, and with the development can be extremely effective to combat highest probability of being hired within the same poverty. sector. Once individuals are hired, the micro-simulation imputes wages based on hot-deck methods and re- Expected impacts on social outcomes estimates household per-capita income and poverty In 2013 the Departmento Nacional de Planeación (DNP) status. used a Computable general equilibrium (CGE) model to estimate the effect of the 4G investment on It is also important to mention the main assumptions of this approach: (i) the number of workers in the August 2018 · Number 8 3 construction sector is low in the labor survey. (ii) The and connectivity. Transport infrastructure provides simulation assumes that all employment is created physical access to new markets, reduces geographic simultaneously and is not a smoothing process. (iii) Job impediments to trade, and unleashes processes of creation is lineal with respect of expected number of urbanization that increase productivity by creating km. in each Departamento. economies of scale. Adequate transport infrastructure also enables access to jobs and social services, Results suggests that new jobs will lead to a reduction increasing opportunities, incomes and welfare. of moderate and extreme poverty by 0.5 and 0.3 percentage points, respectively. Under the more The GoC launched the 4G road concessions to improve optimistic scenario, which considers the realization of transport infrastructure. The 4G project has the both the expected short and medium-term potential to boost competitiveness by reducing costs of infrastructure projects, moderate and extreme poverty transport and logistics, and improve social outcomes by are expected to fall by 1.1 and 0.5 percentage points. promoting growth, employment, higher real wages and easing access to education and health services. Since investments are expected to take place throughout the country, poverty reduction is also The resulting infrastructure investments are expected to expected to occur nationwide. The simulations reveal have a direct and positive impact on the economy, that poverty reduction exceeds 0.2 percentage points in boost public-private partnerships, and generate 20 of the 24 Departamentos. Although the returns to employment across the country, especially in infrastructure investments are expected to benefit the construction. The ex-ante evaluation suggests that entire country, these are likely to have a greater positive 180,000 jobs will be created and moderate and extreme development impact on Departamentos that were poverty will fall by 0.5 and 0.3 percentage points. In previously disconnected from the national and addition, the proposed investments are also projected international economies. In other words, the 4G project to have a positive impact on the reduction of inequality is expected to promote regional economic by increasing connectivity, boosting tourism and jobs in convergence. Even after controlling for other traditionally isolated regions, and improving access to determinants of GDP growth at the Departamento level, markets and the supply of quality public services in rural such as education, health and other public services, the areas. results of a prospective conditional convergence model show that implementation of 4G investments would increase the rate of income convergence between poor ABOUT THE AUTHORS and rich Departamentos over the period of operation. Carlos Rodríguez-Castelán is a Senior Economist at the World Bank and co-Lead of the Market and Institutions GSG. Conclusions Eduardo A. Malásquez is an Economist at the World Bank. Jorge Franco is a Consultant at the World Bank. Transport infrastructure can play a transformative role in socioeconomic development by providing mobility This note series is intended to summarize good practices and key policy findings on Poverty-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank Group, its board or its member countries. Copies of these notes series are available on www.worldbank.org/poverty August 2018 · Number 8 4