POVERTY & EQUITY NOTES TOOLS & METHODS JULY 2020 · NUMBER 27 Welfare and competition (WELCOM): a simulation approach Carlos Rodriguez Castelan, Eduardo A. Malasquez and Rogelio Granguillhome This note focuses on the Welfare and Competition (WELCOM) microsimulation tool. WELCOM, an easy-to-use Stata- based package with minimum data requirements, was conceived as part of larger World Bank efforts to better understand how competition policy can improve market efficiency and reduce poverty, especially in developing countries. Using examples from Mexico and Ethiopia in the telecommunications and food sectors, this note shows how WELCOM can estimate likely distributional effects—that is, decrease in price and poverty, and increase in product uptake—from expanding competition. Fully grasping the distributional effects of structure of certain industries: for instance, breaking up competition is central for policymaking. Households a total monopoly to have only one-half of market share. at the lower end of the income distribution are Using this simple data and competition targets, the tool commonly most negatively affected by market simulates likely direct distributional effects of concentration, since they tend to consume more competition mediated through changes in prices. homogeneous goods, have less opportunity to WELCOM allows users to estimate models using substitute consumption, and have restricted access to variation in three alternative market structures: markets. Therefore, policies promoting competition in 1 monopoly, oligopoly with a homogeneous good, and concentrated markets could help reduce poverty, and partial collusive oligopoly. increase growth and productivity. Lower prices resulting from higher competition in The Welfare and Competition tool (WELCOM) fills an production of any good benefits both current and 2 important analytical gap. While several studies have potential new consumers previously “priced out” of rigorously analyzed the casual relationship between a market. 3 To identify the likely entry effects of new market concentration and household welfare, consumers due to lower prices of goods and services, as conducting this analysis is demanding in terms of time, well as the likely welfare effects for those new entrants, data needs, and identification of event studies. the WELCOM tool has recently developed a new module on “Market Competition and the Extensive Margin Analysis.” This new module has been piloted in Ethiopia Methodology and Data (see case study on Ethiopia below). WELCOM estimates likely welfare effects of changes Case Study: Mexico in competition. The tool uses a simulation approach based on minimum data requirements. WELCOM combines expenditure data from typical national Mexico’s history of high market concentration representative household surveys used for poverty follows the privatization of State-Owned analysis with parameters on expected outcomes of monopolies in the 1990s. One of the best known cases competition reform, which would modify the market of privatized monopolies is the one of fixed line telecommunications. Another known case of market Figure 2: Poverty reduction due to competition in mobile comms concentration is the corn products market dominated by a duopoly. Both corn and telecommunications comprise 0 a significant percentage of household spending. Changes percentage points in poverty The sum of the expenditure share on these two -0.2 goods by poor households is higher than for richer rate households (Figure 1). The simulation assumes that the -0.4 mobile telecommunication market behaves as an oligopoly and that corn markets act as a partial collusive Point estimate Conf. Interval (95%) oligopoly (PCO). The simulations use price elasticities of -0.6 demand of −0.476 for mobile communications 4 and 4 6 8 12 … N −0.876 for corn products 5. Number of firms Source: Authors elaboration using WELCOM and ENIGH 2014. Figure 1: Expenditure share by income decile on mobile services and corn products Households in the richest deciles of the income distribution would benefit more in relative terms from 10 9.3 Mobile communications lower concentration in the mobile communications 7.6 Corn 8 market, since they spend a higher share of their 6.6 Total 6.0 incomes on telecommunications than on 6 5.2 consumption categories such as food and beverages. % household expenditure 5.0 4.5 4.1 4 3.6 2.5 Figure 3: Poverty reduction due to competition in the 2 corn flour market 0 0 Changes percentage points in poverty rate 1 2 3 4 5 6 7 8 9 10 -0.2 Income decile Source: Author’s elaboration based on ENIGH 2014. -0.4 Expanding competition in both the -0.6 telecommunications and corn products industry -0.8 would likely reduce poverty by almost one percentage point. Reducing the market share of the -1 Point estimate Conf. Interval (95%) oligopoly in corn products from 31.2 percent to 7.8 percent, and increasing competition from 4 to 12 firms -1.2 31.2 15.6 10.4 7.8 … 0 in the mobile telecommunications sector, would induce Oligopolistic Market share (%) a relative price drop of 38.5 percent for corn and 11.4 percent for mobile telecommunications. These price Source: Authors elaboration using WELCOM and ENIGH 2014. changes would reduce Mexico’s poverty headcount by an estimated 0.8 percentage points (Figure 2 and Figure In contrast, a decline in market concentration in corn 3), and could produce a drop in the Gini index, which products would benefit households at the bottom of the measures income inequality, of 0.32 points. income distribution relatively more, since these goods represent a higher share of their food expenditures. July 2020 · Number 27 2 Figure 4: Relative impact on household budgets by Breaking up the monopoly in telecommunications income decile would also spur technology adoption (extensive 4 margin), from users previously priced out of the Mobile communications 3.1 market. Using the expected relative decrease in prices Corn 2.4 of 25 percent due to increased competition, the “Market % average income Total 2.1 Competition and the Extensive Margin Analysis.” 2 1.7 1.4 module of the WELCOM tool estimates new user uptake 1.3 1.1 0.9 of 4,423,938 new users, equivalent to a 4.9 percentage 0.6 0.3 point increase in total individual users. 0 1 2 3 4 5 6 7 8 9 10 Households between the third and seventh decile of Income decile the expenditure distribution would likely make up Source: Authors elaboration using WELCOM and ENIGH 2014. the largest portion of new users of telecom services, with increases in the probability of expenditure of about 11.8 percent. Relative welfare gains (higher expenditure) Case Study: Ethiopia for new users is expected to be between 1.14 and 1.33 percent (lower and upper bounds in Figure 5, The telecommunications market in Ethiopia is respectively). Households in higher expenditure deciles dominated by the state-owned monopoly provider, would see the highest absolute welfare-enhancing Ethio Telecom. By 2019, Ethiopia’s ICT market was one effects from competition, since they are expected to of the most concentrated worldwide, according to the spend more on telecommunication services than poorer Herfindahl-Hirschman index (HHI) (GSMA 2019). households. High market concentration and high prices in the Figure 5: Competition in telecom would induce higher use telecom industry partially explain the low take-up of of mobile services telecommunications by poor households. Only 14 Bars represent relative percentage change in probability of use, dots percent of households in the poorest decile reported show relative welfare gain of old and new users as a percentage of average expenditure. spending on telecom services in the 2015/16 national household survey (HCES). The simulation conducted for 16.0% 2.0% this case study assumes that the mobile 14.0% monetary incidence average expenditure Absolute welfare 12.0% 1.5% Percentage of telecommunication market is a monopoly with price 10.0% elasticity of demand (PED) of -1.5. 6 8.0% 1.0% 6.0% 4.0% 0.5% Opening the telecom sector to competition would 2.0% 0.0% 0.0% have positive welfare effects for current consumers. 1 2 3 4 5 6 7 8 9 10 WELCOM estimates suggest that decreasing Ethio Expenditure decile Telecom’s mobile services market share from 100 to 45 Relative change in probability (percentage) Relative welfare incidence (Upper-bound) percent would induce poverty reduction of 0.2 Relative welfare incidence (Lower-bound) percentage points in the short-term and up to 0.8 in the Source: HCES, 2015/16. long-term, assuming no additional take-up of Notes: Upper-bound expenditure estimates were carried out using telecommunications services; this implies that welfare random imputation. Lower-bound estimates were carried out taking gains accrue only to current users of mobile average expenditure by primary sampling unit. communication services. July 2020 · Number 27 3 ABOUT THE AUTHORS Conclusions Carlos Rodriguez Castelan is a Lead Economist in the Africa Region and Global Lead of the Data for Policy Global Solutions Group in the Poverty and We applied the easy to use WELCOM tool to these case Equity Global Practice. studies in Mexico and Ethiopia. The results—decreasing Eduardo Malasquez is an Economist in the Poverty prices and increasing product uptake—show how the and Equity Global Practice. tool can help with policy making by estimating the effects on consumers and poverty reduction from Rogelio Granguillhome is a Consultant in the Poverty and Equity Global Practice. freeing sectors to more competition. Mexico and Ethiopia provide useful examples to inform development practitioners and academics about the potential distributional gains of reducing concentration in key industries in developing countries. 1 Creedy, J. and Dixon, R. (1998). “The Relative Burden of Monopoly on Households with Different Incomes.” Economic, New series, Vol. 65, No. 258 (May, 1998), pp. 285-293.; Urzua, C. (2013). “Distributive and Regional Effects of Monopoly Power.” Economía Mexicana, 22(2), pp.279-295. 2 For more information refer to: Rodriguez-Castelan, Carlos; Araar, Abdelkrim; Malasquez, Eduardo A.; Olivieri, Sergio; Vishwanath, Tara. 2019. Distributional Effects of Competition: A Simulation Approach. Policy Research Working Paper No. 8838. World Bank, Washington, DC. 3 Changes in consumption expenditure due to changes in the products entering households’ consumption basket. 4 Elasticity is from Hausman, J.A. and Ros, A.J. 2013. “An econometric assessment of telecommunications prices and consumer surplus in Mexico using panel data.” Journal of Regulatory Economics, 43(3), pp.284-304. 5 Elasticity derived from Deaton’s (1999) unit value model and the WELCOM elasticity of demand module. 6 Fuss, Melvyn (2005), “The Impact of Telecoms on Economic Growth in Developing Countries”. Vodafone Policy Paper Series 3. 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, its board or its member countries. Available for download at the World Bank Publications, Documents & Reports site. July 2020 · Number 27 4