76389 Finance & PSD Impact FEBRUARY 2013 The Lessons from DECFP Impact Evaluations ISSUE 22 Our latest note shows how the rich administrative data already collected by many governments can be used to evaluate a policy reform, and provides an example where impact evaluation reveals that a policy didn’t work as well as intended. The impact of expanding simplified start-up procedures to more remote areas: The Minas Fácil Expresso program Miriam Bruhn and David McKenzie Reforms to make it easier to register In the absence of this program, firm owners a business are the most common type of would have had to travel to their municipal reform tracked by Doing Business, with over office to register at this level, and then 75 percent of countries adopting at least one separately into a large city to register at the reform in this area over the past decade. One state and federal levels. of the most popular types of reforms is to set Deploying an express unit costs up a “one-stop shop� by integrating different approximately 40,000 Reais (US$19,000) registration steps with different levels of for the first year, which includes the cost of government into a single streamlined training, computer equipment, furniture, and process. running costs. The goal was to increase the In larger municipalities this one-stop opening of formal businesses in these shop can be a physical office that firms can municipalities, and as a result, also increase come to. The State of Minas Gerais in Brazil tax receipts by bringing more firms into the introduced such offices into 31 of the state’s formal system. 853 municipalities covering the more populous municipalities with highest levels Evaluating the Impact of firm registrations. The simplifications The State Government set a target of resulted in a reduction from 8 to 4 opening 50 Minas Fácil Expresso offices by procedures required to register, and a the end of 2011, and then expanding further reduction in the time taken to register from to the rest of the state. An initial list of 50 28 to 9 days. municipalities was selected based on the However, such offices are expensive volume of registrations occurring in 2010, as to set-up and maintain, and so the challenge well as whether the municipalities had facing policymakers is whether they can expressed interest in having the program. In extend such services to less populous practice actual implementation occurred for municipalities, and if so, what the impact is? 32 of these 50 municipalities, beginning in September 2011, along with a further 23 Minas Fácil Expresso municipalities for a total of 55 In order to extend the one-stop shop municipalities by the end of August 2012. concept to more of the municipalities in a To evaluate the impact of this cost-effective way, the state began a program, we use administrative data on the program called Minas Fácil Expresso number of firms registering in each (Minas Easy Express). This involved setting municipality in each month, as well as up a system to mimic the one-stop shop monthly municipal tax collected. within existing municipal offices, but with The typical municipality in which documents received from entrepreneurs this program was introduced had a scanned and emailed into the central offices population of 56,000, and averaged 12 firms instead of processing them onsite. per month registering in the year before the Do you have a project you want evaluated? DECRG-FP researchers are always looking for opportunities to work with colleagues in the Bank and IFC. If you would like to ask our experts for advice or to collaborate on an evaluation, contact us care of the Impact editor, David McKenzie (dmckenzie@worldbank.org) new program was introduced. These Figure 1: Summary of Estimated numbers are much smaller than in the 31 Treatment Impacts on Firm Registrations municipalities with physical offices, but Average Impact in First Impact after larger than the averages for the remaining impact 2 months first 2 months 0 767 municipalities. Impact on Number of Firms Registering -0.2 We therefore combine propensity- per Municipality per Month -0.4 score matching with difference-in- -0.6 differences to obtain an appropriate -0.8 counterfactual for what the registration rates -1 would have looked like in the absence of the -1.2 -1.4 intervention. Nearest neighbor matching on -1.6 36 months of pre-intervention data enables -1.8 us to obtain control municipalities which -2 have very similar pre-intervention trends to the treated municipalities. As a second Implications control group, we also use the set of The results are consistent with other municipalities where the offices were recent experimental evidence that calls into planned, but had not yet been implemented. question the view that most informal firms would really like to formalize, but are just Results hampered from doing so by burdensome  We estimate that introducing the regulations (e.g. see Impact note 17). Instead Minas Fácil Expresso units actually most firms may see very little benefit from led to a drop in the number of formalizing. This is particularly likely to be registrations: 1.3 fewer firms true as one considers firms in more remote registered per month per treated municipalities. municipality, a 10.4 percent fall in As a result, policymakers need to re- registrations (figure 1). consider the cost-benefit to both private  This impact appears to be largely informal firms, and to the state, of trying to concentrated in the first two months bring these informal firms into the formal after the new units opened, but there sector. If they wish to bring such firms in, is also no evidence of subsequent increasing the benefits of being formal, and catch-up. increasing the costs of informality through  We find no significant increase in more enforcement are likely to be needed. total tax receipts in the municipality. More generally, this impact evaluation provides an illustration of how municipal administrative data can be used to quickly and cheaply evaluate the impact of municipal-level reforms. The data and Stata do files used to conduct this analysis are available in the World Bank’s microdata library to aid others who wish to follow this approach. For further reading see: Miriam Bruhn and David McKenzie (2013) “Using Administrative Data to Evaluate Municipal Reforms: An Evaluation of the Impact of Minas Fácil Expresso� World Bank Policy Research Working Paper no. 6358 Recent impact notes are available on our website: http://econ.worldbank.org/programs/finance/impact