A SECOND LIFE FOR ONE-STOP SHOPS: CITIZEN SERVICES IN MINAS GERAIS, BRAZIL, 20032013 SYNOPSIS In 2003, the new governor of Minas Gerais, Brazil, pledged to improve government efficiency and serve citizens better. Residents of Minas Gerais, Brazils fourth-largest state by area and second largest by population, had long bemoaned the difficulty of obtaining such vital documents as work permits, passports, and drivers licenses, which are issued by a variety of federal, state, and local agencies. In 1996, the state government tried to solve the problem by experimenting with 26 one-stop shops that integrated related citizen services under a single roof, but the shops failed to reduce delay and confusion. From 2007 to 2010, the governor and his reform team restructured and expanded the one-stop shops. The reform team persuaded multiple levels of the government to cooperate more closely, revamped management practices, improved the physical appearance and organization of facilities, streamlined procedures, and installed an electronic monitoring system. Renamed integrated citizen assistance units (unidades de atendimento integrado), the new one-stop shops improved services, reduced delays, and sharply increased processing volume. In 2011, the team outsourced the management of six of the one-stop shops to a private company monitored by the state. The public-private experiment cut per-unit operating costs by 31%. By 2012, 30 one-stop shops were handling more than 6 million citizen transactions annuallymore than seven times the annual volume in 2009. By bringing together diverse agencies from multiple levels of government, Minas Gerais was able to greatly improve the reach and efficiency of its citizen services. Rushda Majeed drafted this case study based on interviews conducted in Belo Horizonte, Minas Gerais, Brazil, in May 2013. The case was prepared by ISS in partnership with the World Bank as part of the Banks Science of Delivery initiative. Case published January 2014. INTRODUCTION Residents of Minas Gerais, Brazils second-most-populous state, had long bemoaned the difficulty of obtaining work permits, passports, drivers licenses, and other documents. Many citizens arrive at dawn to receive a ticket number, and many times there are attempts to sell a place in line. There are citizens who wait five to six hours . . . in line, said one report on the states first generation of one-stop shops for citizen services.1 In parts of the state capital of Belo Horizonte, people complained that ticket numbers issued to those who lined up for services run out around 10 a.m. and many citizens are turned away.2 In 1996, the state had set up one-stop shops, called postos de servio integrado urbano (integrated urban service units), or PSIUs, to offer a variety of services in single locations around the state, but the shops failed to reduce delay and confusion. The failing one-stop shops were one of the challenges newly elected Governor Acio Neves da Cunha faced when he took office in 2003. With Brazil in the throes of a currency crisis in 1999, Minas Gerais had stopped repaying its US$13.5 billion of debt to the federal government because of a budget deficit.3 The states default scared off private enterprise, which plunged 29.1% in 1999 and an additional 19.3% in 2000.4 From 1999 to 2003, Minas Geraiss average annual rate of growth in gross domestic product was 0.7%even less than the sluggish, national average of 1.8%.5 The states economic underperformance reversed the achievements that had taken place from 1995 to 1999, when Minas Gerais had grown at an average annual rate of 2.9%. In 2002, the state governments personnel expenditure was 66% of net current revenue, and the states net consolidated debt to the federal government was 236% of state net current revenue.6 Neves pledged to pursue management shocks that would fix the insolvent and floundering Minas Gerais government and transform the state into the best place to live in Brazil.7 Humberto Falco Martins, who advised the government, summarized the conditions that had created a reform moment: The state was bankruptfiscally, managerially, and politically. It was the end of the road, the bottom of the well. People felt that no one had anything more to lose. That was the window of opportunity. The reforms that Neves implemented in his first gubernatorial term (200306) balanced the state budget and won him a second term in December 2006. Riding high on public approval after capturing 77% of the vote, he empowered the planning and management secretariat to move forward with an effort to improve the one-stop shops as part of a second phase of reform called The State for Results. In addition, starting in 2006, the World Bank supported the management shock with a loan of US$170 million, which it followed with a second loan of US$976 million in 2008. Determined to reduce state spending, improve the investment climate, and enhance the quality of citizen services, Neves mobilized a team whose members had strong management skills and experience. In 2003, he appointed lawyer Antnio Augusto Junho Anastasia to head a new department, the Secretariat for Planning and Management (SEPLAG). The creation of the new secretariat underscored the new governors priority for making the government work better and brought much-needed political clout to the task. Mauro Csar da Silveira, an adviser at the secretariat, said: The technical expertise and strong role of SEPLAG were very important in the beginning. . . . They had the keys to the vault. Renata Maria Paes de Vilhena, a career civil servant who had served in both the state and federal governments, became deputy secretary of state for planning and management under Anastasia. THE CHALLENGE The existing one-stop shops had been under the umbrella of the State Secretariat for Regional Development and Urban Policy, which had provided loose oversight. In 2007, Neves moved direct responsibility for the 26 one-stop shops to the planning and management secretariat. Under the supervision of Secretary Vilhena, the job of improving the shops went to Fernanda Valadares Couto Giro, assistant secretary of management, who had joined the secretariat shortly after its creation in 2003. Giros initial mandate was to coordinate the states e-government initiatives, as well as the governments telephone assistance line, in-person assistance services, and Web site. In 2007, her offices mission expanded to include review and reform of the PSIUs. Giro and her team identified several underlying problems that were contributing to the PSIUs poor performance. First and most important, the independence of the national, state, and local governments under Brazils federal system made it hard to design a unified approach. The disparate units and levels of government that provided services had little incentive to cooperate with each other. Second, staff training and vetting were poor; employees did not share an ethic of citizen service. Third, the shops lacked standard operating procedures, including hours of operation. Finally, in some instances, especially outside the main cities, managers cut deals with suppliers, thereby creating conflicts of interest. Winning agreement among agencies was a prime challenge. The PSIUs had no special authority to provide services themselves or to decide how the various agencies should staff their operations and manage their functions. The shops merely convened the local, state, and federal agencies responsible for providing specified services at single delivery points. Each agency controlled its own activities. There was no formal institutional partnership. And the respective managers and workers did not coordinate their work. Further, the PSIUs had no shared staffing plan and offered no guidance on the employment of civil servants, contractors, and interns. At the time, customer service was not a high-status job or a government priority. As a result, agencies often viewed the one-stop shops as places to send troublesome workers who usually lacked the right qualifications, Giro said. The shops were where the agencies put civil servants who wouldnt fit in any other place, she said. There were people sent to the PSIUs as a punishmentsent to the front line to go and serve the public. In the late 1990s, as government restructuring closed some state offices, agencies shifted employees to PSIUs instead of laying them off. Personnel policies at the one-stop shops also affected managerial capacity. Usually appointed based on political connections, most managers did not have the right skills to run a shop effectively. The PSIU coordinators, as the managers were known, came from diverse agencies and had varying salaries and varying years of experience. Staff accountability and responsibility were lax because managers had little authority over staff members, who reported back to their home agencies. Further, civil service rules gave the PSIUs no authority to schedule weekend shifts or evening hours. As a result, many shops were closed when many citizens wanted to visit. Further, employees received little training in working with the public. Partly as a result of that setup, the organizational culture failed to reward employeesboth managers and staffwho cared about citizens experiences. The mentality was one of obligation: I have to do it because I am a civil servant, but I dont have to worry about quality, Giro said. Measuring overall service efficiency and employee effectiveness was impossible because the Secretariat for Regional Development and Urban Policy, which had original responsibility for the PSIUs, had not created a central system to monitor performance. A 2007 diagnostic report noted, With this system, it is not possible to know how long each citizen waited in a line, how long each attendant took, nor the productivity of each attendant.8 Although sample wait times observed by the reports authors were as high as six hours,9 managers did not capture in a systematic manner the average PSIU wait time or the average wait time for each service, nor did managers track the number of canceled services or the level of demand for each service. The absence of shared procedures and standards meant that citizens had far different experiences depending on which PSIU they visited. Most PSIUs did not list the menu of offered services, which differed from unit to unit. There were no systemwide opening and closing hours or operating practices, and there were no common criteria to guide the physical layout of the shops or the choice of technology or other equipment. Many of the shops were uninviting to both the public and the employees who worked there. Rodrigo Gitirana Lyrio do Nascimento, charged with improving the shops infrastructure in 2007, said: The units were dark spaces. The walls were dirty with paint and ink marks because people wiped the fingerprinting ink on them. Cabling and electric systems were poor. Desktops and workstations had all been trashed. The furniture was bad. There was no architectural standard. The working conditions were very poor. Budget limits intensified the problems in one-stop shops outside the capital city of Belo Horizonte. Giro said: If a chair broke, the coordinators made deals with municipal government or local stores to get one through donation. The flip side, in some cases, was that they would provide expedited services for the people who agreed to the deal. Their function and focus were not on the technical management but on the political aspects. She added, These deals kept the unit functioning, but they were detrimental to the quality of service overall. In rural areas, coordinators wielded extraordinary influence because they could hold out the promise of services to buy votes for local politicians or to run for city council themselves. Giro said: The coordinators would use the provision of services as a means for getting votes. We used to joke that in the interior of the state, you have three important kinds of people: the police, the priest, and the coordinator of the PSIU because of the political currency involved in these politicized positions. As the quality of service at the one-stop shops slipped further, even the participating agencies began to have second thoughts about sending people to work there. Breno Eduardo Elias Dos Santos, in charge of managing business processes, recalled, A lot of the agencies started leaving the PSIUs because of their bad image. FRAMING A RESPONSE From the beginning, Giro said she believed that the one-stop shop was a sound idea that needed fixing, not replacement. People still approved of the service in the PSIU, she said. Citizens felt that even if they missed a workday, they could still get a number of services done in one day at a PSIU. They knew that they had to go to only one location to access the services they needed instead of running between different locations around the city. Eager to improve the performance of PSIUs, Giro and her team began to look at examples from other areas in 2007. The team studied one-stop shops in other parts of Brazil, including Bahia, the Federal District, Rio Grande do Sul, and So Paulo. Those site visits helped Giro and her team understand the difficulties, advances, technology, and quality of personnel, she said. One of the most compelling models was close by. When Neves first came to office, he launched a project called Programa Empressa Mineira Competitiva, or Competitive Minas, to attract business and reverse the collapse in private investment. The project, housed under the planning and management secretariat, created Minas Fcil, separate one-stop shops designed specifically for businesses to register and get licenses (see Textbox 1). Alex Francisco Barbosa, who was in charge of Minas Fcil units as director of information and modernization at the states Commercial Board, said: We had integrated services and also integrated work across different levels of government. So the governor came up with the following proposal: We have to do the same for the citizen, as well as for the state itself. The governor pushed the secretariat to consider the Minas Fcil model and proposed that improved PSIUs be renamed unidades de atendimento integrado (integrated citizen assistance units), or UAIs. The acronym, pronounced why, mimicked a local greeting. Giro said, Its a common term here, like Yo! or Aloha! . . . It was necessary to demonstrate to society that we were breaking with a previous culture. The governor had coined the term himself for that reason. The decision makers decided to expand the scope of Competitive Minas from facilitating business to include projects that would facilitate enable better citizen service delivery. They reconfigured Competitive Minas into an innovation think tank, Projeto Estruturador Descomplicar (Structuring Project Uncomplicate), which would work closely with Giro and the one-stop-shop coordinating office to facilitate the transformation of the original PSIU one-stop shops into the new UAIs (see Textbox 2). Giro and her team worked to trace the service delivery process at the PSIUs, aiming to identify gaps in service delivery before proposing measures to overhaul the system. We did actual fieldwork, she said. We went to the largest PSIUs in Belo Horizonte and took notes on what services the citizens were using, talked to attendants and citizens, and recorded procedures and the time taken. The outline of the new UAI model began to take shape. Based on its findings, the team proposed a complete reworking aimed at improving the citizen experience. Like the original shops, the UAIs would bring related services together in one place. Under the supervision of a UAI manager, uniformed clerks would accept and process applications at identical-looking booths or counters. Eventually, each clerk would be able to provide any service, once cross-trained. Efficient use of technology was another linchpin in the plan. Each UAI would have a standard, upgraded information technology system capable of supporting high volumes of work without interruption, as well as an automated service desk where citizens could access services online through a Web portal. The new one-stop shops would have longer hours of operation as well as hours on weekends and public holidays in order to meet public demand. Participating agencies would commit to little or no service disruption during the UAI business hours, which would be from 7 a.m. to 7 p.m. on weekdays and from 8 a.m. to 2 p.m. on Saturdays in Belo Horizonte. Units in the interior regions of the state would offer services from 8 a.m. to 5 p.m. on weekdays. The new hours would extend to new payment windows created within the units. Instead of having to wait for a bank to open in order to make a payment and then return to a service counter with a receipt, as had been required at the old shops, customers could conduct transactions within the UAI itself. Private banks operated these windows during the same hours as the rest of the services. Branding considerations were important to the planners. All units would have uniform architecture, with the same layout and color scheme. Service policies and procedures would also be made consistent. Nascimento said, In the past, citizens would go to one or the other unit depending on which ones offered better services. But now they would be able to access the same quality of services at any one-stop shop. That ambition extended to all parts of the state, including remote locations. Giro added, Our concept was that the citizen is the same, whether in Belo Horizonte or the interior. Additionally, the team sought ways to create more-flexible staffing, offer longer weekday hours, and extend weekend operations. In the past, the secretariat had contracted with a semi-autonomous public company called Minas Gerais Administration and Services (MGS) to provide support staff for functions like building maintenance. People hired through the company were governed by policies that were more flexible than the governments civil service rules, which made it possible to operate shifts in the evenings or on weekends. Vilhena and Giro proposed to extend that system to the UAIs. Instead of participating agencies staffing their own operations, as in the old PSIUs, the secretariat would now hire line employees through the company. Each agency responsible for a citizen service would train those employees in the particular skills needed to perform their duties in the one-stop shops. The shift would enable UAI managers and the secretariat to monitor performance, coach more effectively, and tell the company to remove or discipline those who failed to do their jobs. The secretariat planned to phase in those changes, starting with two one-stop shops in 2007, eight in 2008, and more in the following years. By 2013, only two PSIUs remained unchanged after the secretariat had transformed 24 PSIUs and added four new one-stop shops. (See Table 2 (a) and (b) for details on number of PSIUs converted to UAIs from 2007 to 2013.) GETTING DOWN TO WORK In 2007, the hard work required to restructure the old PSIUs into UAIs began. Giro assembled additional staff members to help her implement the new one-stop shops. Her office expanded from 6 people to 13 in 2007 and eventually to 17all of them doers: people who could get things done. She also relied on the Office of Projects, a unit in the planning secretariat that followed up on the strategic priorities of the governors management shock program. The involvement of that office meant there was a direct line to the governor, a signal of Table 2 (a): Minas Geraiss Three Generations of One-Stop Shops PSIUs UAIs UAIs on PPP model Total 2007 24 2 - 26 2008 18 10 (including 2 new UAIs) - 28 2009 10 18 - 28 2010 7 22 (including 1 new UAI) - 29 2011 2 22 6 (including 1 new PPP) 30 2012 2 22 6 30 2013 2 22 6 30 high-level political support that had been absent during the creation of the earlier one-stop shops. Giro said, This strategic link to government macroplanning was missing during the PSIUs, and this is what made the UAI project possible. Enlisting the agencies A one-stop shop was a convener, or Table 2 (b): Brief Explanation 2007 Of the 26 PSIUs, 2 converted to UAIs. 2008 Of the 24 remaining PSIUs, 6 converted to UAIs and 3 brand-new UAIs were added. 2009 Of the 18 remaining PSIUs, 8 converted to UAIs. 2010 Of the 10 remaining PSIUs, 3 converted to UAIs and 1 brand-new PSIU was added. 2011 Of the 7 remaining PSIUs, 5 converted to PPP UAIs and 1 brand-new PPP UAI was added. 2012 2 PSIUs remained. 2013 2 PSIUs remained. of the local, state, and federal governments. In Vilhenas words: The government of Minas Gerais understood that centralized service delivery through the UAI model should transcend the levels of government. And through the UAI, it took on the role of a catalyst and the integrator of that model. Giros office had to persuade the agencies to cooperate to streamline procedures, forge shared policies, and allow the Secretariat for Planning and Management to monitor performance. The secretariat gathered information on each service it would host at the UAIs and negotiated the obligations of each party before formalizing the relationship through a cooperation agreement. (See Textbox 3 for highlights of the agreement.) The reform team crafted incentives to induce the agencies to cooperate. Under the old system, participating agencies paid for the civil servants who provided their services, leaving the units maintenance costsestimated at 3.13 million reals (US$1.46 million) in 2007to the secretariat. Vilhena, who took over as planning secretary in 2007, agreed that the secretariat would build or upgrade the one-stop shops, provide equipment, and continue to cover maintenance expenses. In addition, it would hire personnel, pay their salaries, and manage them. The secretariat would also monitor the quality of service delivery. This arrangement would apply to all levels of government. In exchange, the agencies would agree to accept new standards of service and abide by UAI rules. Santos explained, The deal was that SEPLAG would bear the cost, and the agencies in return would adhere to standard quality and processes for providing the services. As an added incentive, the UAI coordinating team offered agencies expertise in improving business processes if they agreed to participate. For each service, the team worked with agencies to create flowcharts, hone procedures, and define relationships between the people who were part of each delivery chain. Oliveira Santiago Maciel, head of the State Department of Transit, said the decision to join in the UAI effort was easy for him because it did not add to the departments expenses. He said, On the contrary, the partnership allowed us to take our services to more locations, increasing our points of service. In addition, the measure allowed us to improve the quality of services delivered as well as citizen satisfaction. Because the secretariat would provide personnel, the outsourcing of services to UAIs allowed agencies to recall their own staff and assign them to other tasks at a time when budgets were tight. Letcia Alessi Machado, a Civil Police police chief and director of the states Institute of Identification, which issued citizen IDs, said she also viewed the UAI conversion as a win for her agency. The planning secretariat helped us with the renovation of the space and equipment, she said. They provide us with employees, and we provide the training. We make no financial contribution to our services at the UAIs. The only thing we offer to the UAI is the specific material to produce the ID: paper, ink. These are our requirements. The UAIs offered a lot of improvements. The location is much more appropriate. There is more space. There are enough seats. There were often legal and procedural obstacles to overcome, however, and Giros unit broached those issues during negotiations. In some instances, the secretariat merely had to win clearance from department heads to have the UAI workers perform some of the functions that department staff had performed before. Offering his office as an example, Maciel of the state transit department said: Our activities are regulated by federal lawspecifically, the Brazilian Transit Code and resolutions issued by Contran [National Transit Council]. They often say we are transferring competencies and services to a third party. We have to demonstrate to them that we are not doing so. For instance, the national department of transit raised objections when Maciels state office decided to make the electronic exam for a drivers license available at the UAIs instead of at department offices in the interior of the state. Maciel said he resolved the problem by stressing that his department would still control the content and quality of the exam and along with the planning secretariat, constantly supervise the administration of the exam and the workers involved. Several federal services agreed to locate services in the UAIs after Giro negotiated with them on rules that would allow contract workers to provide their services. The secretariat worked out an agreement with the Ministry of Labor and Employment to host work permit services and access to the national employment database at the UAIs. Eager to increase locations that offered passport services in Belo Horizonte, the Federal Police agreed to join UAI units in the city. Ultimately, the Federal Police expanded its partnership with the secretariat to allow UAI locations in other parts of Minas Gerais to issue passports as well. Although many agencies continued to offer services at their main offices in addition to the UAIs, they gradually began to view the UAIs as the best places to serve the citizenry. With time, the UAIs leverage increased. Giro said, Whenever an agency comes to us nowadays for being a partner, we make it clear that the UAI adds value for the agencys business and process. By 2013, 15 federal, state, and municipal agencies were providing citizen services at the UAIs.10 The Civil Police provided ID cards and criminal background checks. The Ministry of Labor and Employment helped citizens obtain work permits and job placement services through a national employment database. The Secretariat of Labor and Employment assisted with unemployment insurance. The Secretariat of Planning and Management helped state employees with payroll and benefits such as retirement and pension. The Housing Company of Minas Gerais worked with citizens to renegotiate debt agreements and make payments on loans. And an ombudsman counter logged public complaints. The Consumer Protection Agency also offered support to citizens. Other services included vehicle tax and fine payment, tax ID registration, and business registration. Minas Fcil counters were in place at 12 UAIs. (See Table 3 for a list of agencies and popular services.) Staffing and training Initially, the UAI coordinating team decided to staff the units fully with workers hired through MGS, the company that had provided some staff at PSIUs. MGS was a public corporation that supplied qualified employees to government agencies, managed supplies, and maintained government infrastructure. Created in 1994, when the state bought out the stock of a public-private joint venture that performed the same functions, the corporation reported to the planning secretariat as well as to a board of directors appointed by the governor. People who wanted to work in government as MGS contractors had to take a public exam to qualify. The rules that governed hiring, terms of service, and dismissal were different from those for civil servants, however, and slightly more flexible. When a UAI needed a worker, the coordinating office sent its request to MGS, which opened a selection process. The coordinator of a UAI would interview the MGS-assigned worker. If the worker did not have the right skills for the job, then MGS would send a replacement. For the 28 one-stop shops, the team hired more than 2,100 workers, split into two shifts. The largest one-stop shop, in the Praa Sete area of Belo Horizonte, provided more than 5,000 services per day and had about 300 workers per shift. Each unit had a coordinator or manager, an assistant coordinator, a supervisor Table 3: Agencies Popular Services Ministry of Labor National Employment System Work permit Secretariat of Labor and Employment Job placement services Unemployment insurance Unemployment insurance refund Civil Police * State Department of Transit * Institute of Identification (IIMG) Drivers license Vehicle registration Payment of transit insurance Payment of motor vehicle taxes and fees Payment of motor vehicle fines ID cards (IIMG) Criminal background check services (IIMG) Commercial Board Business registration Business closing (In one-shop shops with Minas Fcil services) Secretariat of Planning and Management Payroll services for state employees Pension and retirement services for state employees Benefit services for state employees Federal Police Passports Criminal records Consumer Protection Agency Consumer protection information Assistance on consumer grievances Federal Revenue Agency Tax ID number Ombudsman Receipt of and action on citizen complaints Housing Company Paying installments Property tax Renegotiation and debt agreements Issuance of payment delay documents Property registration Secretariat of Science, Technology and Higher Education - Tribunal of Justice Negotiation of reconciliation between parties Bank of Brazil Accepting the payment of taxes, bills, and fees for government services Institute of Social Welfare of Public Servants of Minas Gerais (IPSEMG) IPSEMG card IPSEMG refund Energy Company of Minas Gerais Issuance of payment delay document Payment of bills appointed by each agency to oversee its own services, service desk attendants, and cleaning and maintenance staff. Civil servants displaced by the MGS workers returned to their parent agencies. Some took voluntary retirement, and others stayed on in new positions. Giro said: For the actual agencies we were working with, they were able to get their agents back and we avoided the legal hassles of using them for tasks they were not originally hired to perform. Some nearing retirement left. Some lamented that they had worked for many years and now had to work in a different way [but] . . . there was no organized resistance from unions. Training for UAI attendants focused on customer service. Giro said: At the PSIU, the people didnt even have the proper training for dealing with the public. Now we make sure we provide training and courses on assisting citizens, on ethics, and on training in delivering services. We try to provide the workers with the best possible environment to motivate them and reflect the quality they need to have to provide the services. The coordinating unit also held leadership and management training for supervisors and coordinators. The team worked with agencies to develop task-specific knowledge, such as how to issue a work permit. It also brought employees from other parts of the state to Belo Horizonte for training specific to certain agencies, such as the Civil Police. Flexibility and adaptability were hallmarks of the UAIs. Giro and her colleagues came up with the idea of a balco unico (single desk), which let citizens receive any service at any window rather than having to go from window to window for various documents. Bruno Vidigal Coscarelli, a team member who focused on strategic priorities, said: We had cameras and saw a lot of people waiting for a specific service. For example, at a UAI, we would have 20 desks and attendants to provide services, divided by sections. You could see some workers with long lines, but two or three doing nothing because some services were more in demand than others. We did an analysis and changed the way of providing services at UAIs. For a time, the team decided that all UAI workers would be trained to provide every service. If one service faced high demand during a given day and another did not, the coordinator at a UAI could easily tap idle workers to help distribute demand and reduce wait time. However, most units could not sustain balco unico because of high turnover and training needs, and by 2013, very few one-stop shops operated on the principle. Employee motivation at UAIs remained an issue. Alexandre Lima Cabral, in charge of human resources, noted: The biggest challenge is to keep people motivated. They are working with sensitive information and performing important tasks, but the salary does not match the responsibility. And for many young workers, this is their first job. There is also a desire on the part of the workers to grow within the institution they work in[to have] a career plan. But unfortunately, there is not much opportunity for growth in the MGS model. Staffers were paid about 1,100 reals per month (US$540) under the MGS scale. Supervising the supervisors Not surprisingly, conflicts and other disagreements sometimes erupted between UAI coordinatorswho were employed by the secretariat to manage the shopsand the agency-employed supervisors, who were there to keep an eye on specific agency operations. Santos offered an example: When you tell a police officer he is no longer the authority in that place, it doesnt go down very well. Henrique Lage Tomich, a core UAI team member who handled operations, said: The biggest challenge we have today when installing new services is to make sure the supervisors [civil servants from the agencies] accept the UAI. The heads of the agencies accept the UAI. But these frontline people have to get used to seeing the UAI as a partner. We want to demonstrate that there is a gain here for them. There is always a challenge in the beginning because there is no vision of a partnership. The Civil Polices Institute of Identification, which issued identity cards, was an example. Police officers who wielded considerable authority at police stations did not have the same authority at a UAI. Moreover, they were not used to the UAI style of working with citizens. Chief of Civil Police Machado said: Sometimes the people there dont have the skills to deal with citizens and the workers. Some people might have the knowledge to occupy that position, but they lack the social skills needed to perform the task. We are there to please citizens. We have to do our jobs in the best way. Agency heads worked with supervisors to resolve issues. She continued: We invite the person for an open and honest conversation, asking him to embrace the new model. If he keeps committing the same mistakes, we transfer the person and replace him with someone else. In managing the agency supervisors, the UAI team would sometimes negotiate directly with agency heads. Erick Brazil de Vasconcelos, the teams communications head, said: In general, the coordinator identifies a problem with the supervisor from an agency. They talk to us. We talk to the head of an agency. It would place the coordinators in an awkward position if they had to talk to an agency directly and they dont have any direct authority over [the supervisor]. So Fernanda [Giro] and the agency head deal with these kinds of problems. Changing organizational culture Changing the culture within agencies that operated at a UAI was not always easy. Giro said: Change of culture is related to power and authority. Some agencies did not understand the agreement with us clearly and thought that they would lose power. But they had to follow our rules. One of the most difficult challenges was to ensure that even after the doors of a UAI closed for that day, everyone inside would be served. The agency supervisors, who were used to shorter business hours, wanted to go home, but the UAI rules said the workday ended when the last citizen who had entered the facility by the cutoff time had received service. Giro said, We had to work with the agencies to resolve such issues. Other procedures limited favoritism. The team worked with coordinators to enforce the UAIs ticketing system that provided citizens with numbers that indicated their places in line. Because of the policy of first come, first served, citizens could not jump a queue to be served out of turn. Machado commented, Through UAIs, we are offering equality and want to treat all citizens equally. The planning secretariat wanted the UAI coordinators to focus on personnel management and organizational culture. For that reason, they managed maintenance teams centrally. Building support teams ensured that UAI coordinators would not be caught up in maintaining equipment. Giro explained, We make sure that the materials are there, that the infrastructure is there, and that the units have the necessary materials to operate. The coordinators of the units now have to focus on the length of time and wait time for services. We want them to focus on improving the quality of services. Thats why we provide operational support: so that they dont have to worry about these things. To strengthen attention paid to citizens, the secretariat team monitored citizen complaints through an ombudsman desk at each UAI, a hotline, and the state government Web site. The turnaround time for resolving issues was quick. Team members had 2 working days to resolve complaints citizens submitted through the Web site, and 10 days for complaints citizens lodged with the ombudsman. Common complaints included unavailability of appointment or delays in service. Felipe Moreira de Oliveira, a business process analyst and manager of the UAIs citizen relationship department, said he typically received about 7 complaints a day but the number could go up to 16 or 17. Although the numbers were not high, some issues were time-consuming and difficult to resolve quickly. Oliveira said: Sometimes it is hard to resolve issues in two days. For instance, a citizen went to a UAI to renew an ID number she had had for 40 years, only to discover she would have to change it. She didnt want to do so. The complaint should have gone to the ID institute but instead came to us. Now we are trying to contact the institute. The team escalated complex issues to Giro or Vilhena, who would work with agency heads to find solutions. Apart from complaints, the team also passed on positive feedback to the individual one-stop shops.Oliveira added, When a member of the public voices praise of an employee at the ombudsmans desk or on the Talk to Us channel, we always forward it to the person who provided the assistance. The team also focused on making information on services and required documents easily available to citizens. Citizens could go online to the government portal; call a telephone hotline; call a UAI locally; e-mail a UAI unit; send a direct request to the UAI coordinating unit at Fale Conosco (Talk to Us), a channel for citizen feedback; or contact the ombudsman. Inside the units, clear signs and instructions helped orient visitors and enabled them to proceed to the right service area. Computer screens displayed information on the ticket numbers being processed, the type of service, and the status. Helping agencies participate Before moving a service to a UAI, the coordinating team signed an agreement with the responsible federal, state, or local agency. The document included a further agreement to evaluate and improve the service delivery process. Giro explained the importance of such reviews: Most agencies did not know what their processes were and did not even map them. We shared with them, and sometimes we even mapped it for them. They realize the importance of mapping and improving a process. Later, we can monitor and improve a process further. The coordinating unit provided the agencies with specific guidance on what to do and how to do it. Santos said, We understand the work flow. We identify the tools, the procedures, and the actors. After those activities, we present everything we see and recommend changes before agencies join a UAI. We also identify and register the structure we need, such as workstations and computers. Santos described one such detailed review involving the Civil Police. We identified two Civil Police services: IDs and criminal records, he said. For both, we mapped the process from beginning to end. We watched the process and observed it on-site. That was the only way we could map the process correctly and recommend changes. We interviewed police officers. We did the same thing for the other process. Depending on complexity, it usually took the team one to two months to finish mapping and implementing a process, Santos said. The team looked for innovative solutions to solve work flow problems. For instance, to meet the high demand, when the team transferred vehicle registration services to a UAI, the team members worked with the transport office to install a machine that would produce an electronic version of a Civil Police chiefs signature, which was required on all car registration documents. The paper-based test for drivers licenses was computerized when the service shifted to the UAIs, and Giros team worked with the agency to develop other new procedures. The work flow analysis extended even to building maintenance. To standardize management of the facilities, the team developed a manual for maintenance and cleaning. The manual listed the cleaning crews responsibilities, described the type of furniture at the UAIs, and explained how to clean each item. It also included special instructions for working in historic buildings, where some units were housed. Monitoring the UAIs Participating agencies relinquished the task of monitoring to the UAI central team, which developed management and monitoring tools to help ensure quality service delivery. Within a unit, agency supervisors were responsible for making sure that workers followed all the procedures for delivering their agencys service. But the agency supervisors had no authority to discipline workers; the UAI coordinator, who was employed by the secretariat, was the point person for approving vacation days and shift changes, working with employees to address customer complaints, and referring back to MGS those who performed poorly. Initially, the UAI coordinators were the primary monitors of service quality and employee performance, keeping an eye on interactions with customers and ensuring that employees followed rules and procedures and worked speedily. Most of that early monitoring was hands-on and interpersonalfor instance, making sure workers called out the next ticket number as soon as they were done providing service for a client. In 2008, the team developed a management database to replace the simple spreadsheets that the one-stop shop managers had used before. The database made it possible for the UAI team to track the number of customers served, the number waiting, and wait times both at individual units and across the whole system. Each UAI unit had a system that collected information on the total number of entry tickets, the number issued for each service, the number unaccounted for because a customer had left or had lost a ticket, and wait times. It also recorded customer satisfaction through a small keypad at each service window, thereby enabling citizens to log their level of satisfaction with their UAI interaction. The keypad contained four keys to indicate whether the customer was very satisfied with, satisfied with, neutral about, or dissatisfied with the service. All of that information from the individual units came together in a central hub, but there was neither the capacity nor the resources to effectively manage and analyze the data. In 2011 and 2012, Andrey Moraes Labanca and other members of the coordinating units technical team improved the tracking system so as to present the data in a more useful format and in nearly real time (with a 30-second delay). At the UAI coordination unit in the planning secretariat, a bank of computer screens showed the average wait at 28 units, the number of customers being served, the number waiting, and the number of attendants on duty and providing assistance. The system, called the situation room, also calculated the average of the total time for all services and the average service time per agency and type of service. The UAI coordinating unit purchased tablet computers so that team members could monitor units performance even when they were away from the central office. Labanca said he created the first version of the situation room by using only the basic version of Excel (the Microsoft Corp. spreadsheet program) and that as resources increased, he became able to fine-tune it and add functionality. The first version cost the team about 8,000 reals (US$4,770). The secretariat invested in the second version of the situation room, which cost nearly 1.8 million reals (US$965,122) and allocated an additional 1.5 million reals (US$804,268) for software licensing. By enabling the UAI team to track service delivery at each unit, the system allowed for identification of potential trouble areas and quick responses to incipient problems. Labanca said: With the situation room, we started acting preventively instead of just correcting issues we encountered. Another important point was that we wanted to provide the right information at the right time. In the past, we would receive phone calls from a UAI, and the problem had already passed. The coordinator would only let us know later by e-mail. But with the situation room, we could anticipate the problem even before the coordinator picked up the phone. The system also enabled the team to manage demand. Giro said: If we seethrough the situation roomthat a long line is developing, it helps us take actions to correct it by reconfiguring additional desks to attend to that demand. If we see that the line is diminishing, then we can switch new desks and act upon that demand. But the situation room did not resolve all bottlenecks. Labanca said: Some problems are very difficult to solve, like the high demand we have during the first two months of the year. The waiting time in the first two months is longer than we would like it to be. In the bigger units, we cant help it. So, we recommend that people go to other units around Belo Horizonte. We also try to find alternatives, like scheduling appointments. Although hard to measure because demand and the number of available services constantly increased, during vacation periodsDecember, January, July, and Augustdemand for documents could go up anywhere from 10 to 30%. To meet the high demand, UAI team members staggered employees lunch hours and helped schedule appointments. Vasconcelos noted: Making appointments distributes demands. If you dont get an appointment in one unit, you get it in another unit that same day. This is another area where we have had to raise awareness among agencies about the ease and benefits of using scheduled appointments. In the future, the team planned to upgrade the technology so that each unit would have display screens near the entrance, showing the wait time at both that unit and the next-nearest one. We want to make citizens aware of the time they will spend at the unit, and distribute demand, said Labanca. Managers also kept an eye on other aspects of service. Eliel Benetes Gonalves, coordinator of the Praa Sete UAI in Belo Horizonte, said, We monitor the idle time of each attendant. If the attendant finishes a service and takes too long to call the next ticket number or doesnt pause the system during a bathroom break, then it is counted as idle time. By law, they have a 15-minute break, and we keep track of the break. OVERCOMING OBSTACLES Although the new one-stop shops functioned well, there were two big problems. First, costs were high. The annual operating cost rose from 4.3 million reals (US$2 million) for 26 PSIUs in 2007 to about 68 million reals (US$29.1 million) for 29 one-stop shops in 2010.11 (As a comparison, the overall state budget stood at 30.5 billion reals [US$14.2 billion] in 2007 and 41.1 billion reals [US$23.4 billion] in 2010.12) In 2007, the average cost per service at a PSIU was 2.31 reals (US$1.08),13 but in 2010, the average cost per service at a UAI rose to 16.8 reals (US$9.23).14 (See Table 4 for the allocated state and one-stop shop budget from 2007 to 2012.) Table 4: State Budget15 Authorized Budget for PSIUs/UAIs16 2007 R$30,553,704,363 (US$14,287,400,000) R$3,129,207 (US$1,463,270) 2008 R$35,590,405,599 (US$19,609,000,000) R$16,450,000 (US$9,063,360) 2009 R$38,978,230,513 (US$16,662,100,000) R$39,410,000(US$16,846,600) 2010 R$41,113,937,207 (US$23,389,000,000) R$45,250,001 (US$25,742,000) 2011 R$44,998,615,907 (US$26,844,500,000) R$72,936,719(US$43,511,300) 2012 R$50,272,318,708 (US$26,954,900,000) R$71,000,000 (US$38,068,700) Second, the coordinating office found that MGS was often slow to furnish needed staff and supplies. Coscarelli said, I remember one case when we waited three months for MGS to change a lightbulb in a room. The centers required more agility than the MGS was able to muster. Drawing employees from MGS imposed other constraints, too. The MGS worker contracts did not allow the coordinating unit to award people who performed their jobs especially well or to sponsor programs such as a UAI of the Month. In 2009, the governor directed the planning secretariat to consider adopting a public-private-partnership model to address problems. Under the model, the state would run the one-stop shops through a management contract negotiated with a private company. The private company would manage the UAIs, and the state would reimburse expenses according to the number of services provided and the quality of service. Minas Gerais had succeeded in implementing such partnerships in other policy areas, such as infrastructure development and the management of state prisons. The question was whether this approach would work in the direct provision of citizen services. In 2009 and 2010, Vilhena worked with Giro and Bruno Vidigal Coscarelli, a team member who focused on strategic priorities, to analyze (1) the adaptations this approach would require and (2) the pros and cons of each new proposal. The team compared MGS costs and the market prices for roughly the same work to determine whether the private sector would offer more-efficient service at a lower cost. It also considered flexibility. Giro said: It is not necessarily a huge reduction in cost, but it is focused on better use of the money we spend. The private sector does a better job because of its flexibility. The private company could use its own recruiting procedures to find workers with the right skills and aptitudes, whereas MGS had to take people based on exam scores that might not bear much relationship to job performance. A partnership also would offer greater operational flexibility. A private company could quickly ask an employee to replace a broken lightbulb, whereas the existing system restricted UAI coordinators from easily fulfilling that kind of request because of the states strict procurement guidelines. When first presented with the idea, private companies were skeptical. An initial 2009 call for bids for the management contract failed because of company complaints about the level of remuneration and the lack of guarantees regarding service demand, which the firms thought would likely fluctuate. After further discussion, a new, 2010 tender guaranteed that the government would pay 70% of the estimated demand to the winning bidder, thereby establishing a floor level of compensation. (The estimate included only in-person services and did not count those delivered online or via the automated service desk.) Any additional payment would be based on performance, measured by the coefficient of efficiency. In practice, the companies concerns did not play out; demand was always higher than projected, so the government did not have to pay out funds for services that werent provided. The 20-year contract also introduced technical requirements related to the bidding companies experience in managing infrastructures and services. When the secretariat put management of additional units out for a second bid, in 2013, it reduced the contract term to 17 years. It also agreed to review the reliability and validity of the coefficient of efficiencythe main metric of performanceevery five years. The team focused on the price and quality of service when selecting the winning bid. Under the MGS model, the average cost of a service was 16.80 reals (US$9.23). In the bidding process, the team stipulated that the cost per service could not exceed 13 reals (US$7.14). Minas Cidade, the company that won the bid to run the first six units, offered to process a service for 11.60 reals (US$6.37), thereby reducing costs by 31% compared with the other UAIs. To get the new units up and running, Minas Cidade staffed them with private sector workers. The PSIU civil servants returned to their parent agencies. (For the second bid, the secretariat planned to encourage the winning company to retain MGS workers. The team was still working out the final details of the bid in 2013.) The UAI coordinating team adapted its systems to monitor service delivery under this new arrangement. The situation room would continue to operate, and the coordinating unit would continue to calculate coefficients of efficiency. Based on experience, the UAI team closely watched certain aspects of the operations. For example, the team paid special attention to the ticketing systems because the number of tickets given out affected the coefficient of efficiency, which in turn determined the amount that the private company would be paid. In some cases, a citizen could receive a ticket but leave without obtaining a service or could simply lose the ticket. Because the company was paid on the basis of coefficient, which put a 40% weight on average wait time for service, a long wait time because of a lost or unused ticket counted against the company and it lost revenues. Getting the system right was still on the agenda in 2013. The UAI coordinating unit had to clear the new partnership procedures with the participating federal, state, and local agencies; and a few resisted the move. Letcia Alessi Machado, a Civil Police police chief and director of the states Institute of Identification, said: The greatest challenge is the high turnover, and we have no control over the people who get the training. I think this is the biggest challenge. The private sector controls the qualifications of the people. The only thing we receive is a list of names. And we do background checks. But we dont have any other control. Giro acknowledged occasional difficulties in navigating the public-private partnership. The main objective of the state is to provide quality services for citizens, she said. But for the private company, it is profit. Our effort is to maintain profit but also to provide quality service for citizens. We want to control a private companys operations. It has to keep in mind our objective: To provide quality service, not just provide any service. Some worried that the lower salaries the private company paid might influence service quality. In 2013, most workers supplied by MGS received about 1,100 reals per month (US$536) compared with the public-private employee wage of about 670 reals per month (US$327) at public-private units. (Employee wages at public-private units were about four times Brazils poverty line, which stood at 140 reals (US$68) per month in 2013.) The UAI team members noted, however, that public-private units provided upgraded working conditions, such as larger common areas for staff; and they motivated workers through awards and recognition. Further, the UAI coordinating team had to teach the private partner and assist more than it had anticipatedat least in the initial start-up period. For its part, Minas Cidade continued to complain that the coefficient of efficiency was not the best measure for basing payments. In particular, it complained that the UAI ticketing system yielding inaccurate data. It also complained that the delays due to agency procedures, such as waiting periods to access secure systems for IDs, should not be used against it when calculating its efficiency. Overall, however, the secretariat viewed the pilot partnership as a success. The team worked with Minas Cidade to convert five of the seven remaining PSIUs into public-private units and added a new one. All six units had become fully operational by the end of 2011. Vilhena compared the pilot units with the others: We have 95% approval among citizens, and it cost 31% less [than the others]. The secretariat planned to transfer the remaining one-stop shops to a public-private partnership in 2014. The public-private model for the UAIs was not ideal for all, however. Employees at the units that would be transferred to a public-private model did not know whether they would keep their jobs, and they expressed concern. In response, the UAI team opened a special hotline to answer questions and provide information. The secretariat also planned to urge whichever private company won the second bid to retain workers. Breno Eduardo Elias Dos Santos, in charge of managing business processes, said: With the new bid, we are encouraging the private partner financially to absorb the workforce. It is also in the interest of the company, which will not have to spend on training. But we know it is not possible to save everyones job. In 2013, newspaper reports indicated issues at some of the public-private units in operation. For instance, citizens in Uberlndia, the second-largest city in the state, complained about the long wait time at their public-private unit, with some of them reporting they had waited for ID cards for more than eight hours. The coordinator of the unit reported that the managing company had taken measures to meet demand, including hiring 10 new employees.17 ASSESSING RESULTS In 2013, Minas Gerais had 30 one-stop shops in operation, with 1,810 MGS state employees and 273 workers employed under the public-private partnership. The units provided 727 counters for services offered by 15 government agencies. In 2012, the one-stop shops handled 6,245,739 citizen transactionsmore than seven times the 2009 total of 881,796 and up from 3,395,480 in 2010 and 4,819,341 in 2011.18 An increasing share of transactions took place at the UAIs compared with agency home offices. For example, the one-stop shops provided 52% of the Civil Polices ID cards in Minas Gerais in 2011, 59% in 2012, and 64% during the first five months of 2013. For criminal background check documents, which citizens had to present as part of job applications for all public and most private jobs, it issued 55% (2011), 63% (2012), and 62% (January to May 2013) of the total through the one-stop shops. Citizen surveys Citizens gave the UAIs high marks. In 2008, the UAI coordinating unit carried out a citizen satisfaction survey at two UAI units: one in the Barreiro municipality of Belo Horizonte and the other in the smaller city of So Joo del Rei, about 200 kilometers south. Both units ranked high in citizen satisfaction, with 97% of respondents satisfied with services at Barreiro and 95% at So Joo del Rei. (For each unit, 3% of respondents were neutral.) At Barreiro, 64% said they were served within 15 minutes, and 23% said it took 15 to 30 minutes. It took 8% of the people surveyed 30 to 60 minutes and the remaining 5% one to two hours to receive a service. At So Joo del Rei, 87% received service within 15 minutes, 11% from 15 to 30 minutes, and 1% from 30 to 60 minutes and 1 to 2 hours each. In 2009, the team contracted research firm Instituto Olhar Pesquisa e Informao Estratgica to conduct citizen perception surveys. The institute conducted 4,349 interviews across 12 UAI units over a period of two weeks, targeting people being served at the units. The survey found that of those citizens who used the one-stop-shops, a higher proportion came from lower-income groups than from higher-income groups. For instance, of those surveyed, 20% had monthly incomes less than or equal to 465 reals; 32% had incomes in the range of 466 to 930 reals; 30% had incomes of 931 to 1,860 reals; 13% had incomes of 1,861 to 3,255 reals; and 5% had incomes greater than or equal to 5,116 reals.19 Forty-two percent had an elementary school education; 41% had completed secondary school; 9% had enrolled in or completed undergraduate education; 3% had technical degrees; and 2% had graduate degrees. (One percent said they were literate, and 2% said they were illiterate.) Although 60% used computers, 63% percent did not have access to a computer at home. Most citizens told the surveyors that they had visited the one-stop shops to use employment or ID services. They ranked agencies and services, indicating which one they deemed most important: national employment system of the Ministry of Labor and Employment (42.3%), IDs at the Civil Polices Institute of Identification (40%), drivers licenses and other services at the Civil Polices State Department of Transit (5.5%), Other agencies and services (4.3%), Minas Fcil for business registration (2.6%), Secretariat of Planning and Managements payroll and benefit services for state employees (2.5%), military police services (1.8%), and Consumer Protection Agency (1%). In assessing the effectiveness of service, 60% of the respondents were able to receive and complete a service on their first visit; 8% and 6% had to make two visits and more than two visits, respectively. The survey identified several reasons citizens did not receive service on their first visit: they provided incomplete or wrong documents; document issuance required a waiting period; state job placement service lacked job openings; applicant did not meet job requirements; or applicant had to wait for response from employer. Twenty-six percent of respondents indicated they had not yet received the service they came for. The survey did not elaborate on those instances. The survey measured wait time as well. Seventy-three percent waited less than 10 minutes to receive a service; 8% waited 11 to 15 minutes; 9% waited 16 to 30 minutes; 4% waited 31 to 60 minutes; 5% waited more than one hour; and 1% said they did not know (presumably because the service was in progress). In measuring the time it took for a requested service to be completed, 15% of the respondents said their service request was completed in a very short amount of time; 45% said it was completed in a short time. Twenty-two percent said it was completed in a regular amount of time; 5% and 2% said it took long or very long, respectively. Ten percent said they did not know. On a scale of 1 to 5, with 5 being the highest score, citizens rated ease of assistance highly, at 4.9. Clarity of information received a 4.2. And politeness and respect accorded by attendants received 4.4 and 4.5, respectively. On a scale of 1 to 5, with 5 being the highest, citizens ranked cleanliness of UAIs highly, at 4.9; comfort, location of desk, and attendants who helped new arrivals at 4.8; and organization at 4.7. On a scale of 1 to 10, with 10 being the highest, citizens ratings of the 12 UAIs sampled ranged from 8.6 to 9.6. Overall rating for the units stood at 9.2. The UAI brand became stronger over time. Giro said: In 2007, when we opened the UAI, some agencies had left the PSIU. We had to knock on their doors. Now it is them knocking on our door to join the UAI. Service efficiency measures A specially designed measurecalled a coefficient of efficiencyprovided useful data about efficiency across the units, compared the performance of the traditional UAIs with that of the public-private partnerships, and helped calibrate the performance-based payments to the private firms operating some of the one-stop shops. The coefficient of efficiency measured the performance for each UAI on a daily basis. Ranging from 0 to 100, the number was calculated by means of a formula that took into account citizens evaluations of the service they received (captured through the keypad and assigned a 50% weight), average wait times for service (preferably eight minutes or less, assigned a 40% weight), and the degree of assistance (the number of tickets issued per day minus the number of canceled tickets, assigned a 10% weight). The weighting assigned to each component was based on the assumed degree of importance to citizen-customers. The UAI coordinating unit set a target for each unit to achieve a coefficient of 95% or above. The coordinating unit monitored the coefficient on a daily basis and took action when the coefficient fell below 95%. Oliveira, in charge of citizen complaints, said: We monitor day-to-day operations to be aware of problems like lack of training, offline systems, lack of personnel, or high waiting times. So if the coefficient falls below 95%, we often know or can anticipate the reason. We escalate it to Fernanda [Giro] or, depending on the problem, to the [planning] secretary. If a UAIs coefficient fell below 95%, coordinators worked with attendants to reduce wait times. For instance, special staff reminded people to hold on to their tickets and assisted them to the right counters when a number was called. Gonalves said: People may lose their tickets or forget their numbers. This change let us bring up our coefficient . . . especially on Mondays and Tuesdays, when we have the greatest demand. In 2012, of the 22 UAIs, 14 had coefficients of efficiency of more than 95%, including 4 that achieved 100%. The efficiency of two units stood at 92%, two at 91%, and one at 90%. The remaining three had rates of 88%, 86%, and 82%. Of the six privately managed units, one had a coefficient of efficiency of 100%. Two operated at 99%, one at 97%, and one at 91%. The sixth unit, in Uberlndia, the states second-largest city, with a population of over 650,000, performed at 80% efficiency. During the first five months of 2013, two public-private units performed at 100% efficiency; two were at 90% and 92%; and two fell to 86% and 79%. Management problems were one reason for low efficiency, and the private company replaced the manager in one of the low-performing units. By comparison, seven UAIs performed at 100% efficiency, four at 98%, two at 95%, two at 92% and 93%, and three at 90%. The four units that fell below 90% had efficiency levels of 88%, 85%, 84%, and 80%. Continuing challenges Other problems remained. The UAI page on the state governments Web site was not user-friendly, and the UAIs did not have a separate site. Santos said: Publicity is a weakness for the government. We publicize on radio and in newspapers with wide circulation. We also go to the TV channels to inform citizens. We also put flyers around the city. But it doesnt work as well as we would like it to. Citizens are usually surprised when they find out about the UAIs. Citizens surveyed in 2009 said they learned about the UAI through friends, family members, or work colleagues. Others said they were familiar with the old PSIU and had merely returned to the revamped UAI. The Web, radio, and local newspapers had also helped inform some of them. Finally, some had learned of the UAI by walking past a unit. Eighty-nine percent of the people surveyed did not know of Minas Online, the go-to government portal for citizens; 8% knew of it but had not used it; and 2% knew about the portal and had used it. (One percent did not answer.) To better provide information through technology, team members planned to create mobile applications that would enable citizens to access information about the nearest unit and wait times and to schedule appointments. Labanca noted the benefits: Scheduling is one of the best solutions because you distribute demand and use full capacity [of a unit]. The downside is that there might be a longer wait. For example, it could take several days to get an appointment for a service in high demand. REFLECTIONS Brazils Minas Gerais one-stop shops had distinctive characteristics that shaped the strategies public servants could use to manage performance. Service delivery at the one-stop shops was easily measurable and required little discretion on the part of employees. The number of delivery pointsthe number of one-stop shops themselveswas relatively small. Both of those conditions theoretically made monitoring and supervision easier than for other high-demand services, such as healthcare or education. But the reformers faced complex challenges, nonetheless. The Minas Gerais one-stop shops required cross-agency collaboration, which increased the level of difficulty. Further, the reform team had to contend with work rules and norms that created a challenging organizational culture. And the team had to find ways to offer incentives that made it attractive for the agencies to join, that improved processes, and that continued to house services at a one-stop shop. The reforms arose from a political commitment to serve citizens better. Improvement of services was an important plank in Gov. Acio Neves da Cunhas second phase of management shocks. It was also a way to enhance citizen outreach at a time when budget cuts and retrenchment had long dominated the news. The governor and his top-level aides provided the reform team with various types of support, from the use of in intergovernmental negotiations to the creation of advisory teams, which offered ideas, eliminated bottlenecks, and helped track implementation. Fernanda Valadares Couto Giro, who took over as head of the unidades de atendimento integrado (integrated citizen assistance units), or UAIs, coordinating unit, headed a skilled reform team, making it possible to plan and implement the new one-stop shops. She expanded the team to include architects and engineers, business process managers, and technology specialists to revamp the postos de servio integrado urbano (integrated urban service units), or PSIUs; improve the delivery chain; and monitor the units effectively. For the reform team, one of the first steps was to get out of the office and trace the delivery chain to find out what was not working in the original one-stop shopsthe PSIUs. Reform team members observed and timed activities firsthand. They also diagnosed problems. Then, to try to streamline business processes, the reformers met with the different agencies housed in the one-stop shops. They created standard operating procedures so that employees, managers, and citizens knew what to do and what to expect. They also developed metrics and methods for assessing results continuously, so that they could spot and solve problems. The reform team placed special emphasis on training employees at the one-stop shops. Institutional training emphasized high-quality customer service. The program leaders also ensured that UAI coordinators, too, were properly trained in managing attendants and catering to citizens. And respective agencies provided their own training at regular intervals. Top decision makers provided a supportive environment so that team members could develop solutions to the problems they observed. For example, Renata Maria Paes de Vilhena, a career civil servant who had served in both the state and federal governments, and Giro backed the technical team when that team came up with a technological solution to the monitoring of service delivery. The teams innovations saved time and money. For instance, the situation room screens displayed wait times at UAI headquarters, making it easy for the team to observe delays or other issues. Andrey Moraes Labanca, a UAI technical team member, said: Everything became more efficient. We didnt have to call a manager to ask about problems or wait for them to report. And with the time available, we could focus on other areas and improve service. Team members were especially skilled at finding simple solutions to recurrent problems. In one example, Erick Brazil de Vasconcelos, in charge of communications, said: One of our greatest successes was the scheduling of appointments for two services: unemployment insurance and work permit cards. All over Brazil you would see people forming lines in the early morning every day. We made these serviceswhich are more complexrequire more documents, take longer to complete, and by appointment only. And since we have installed these, we no longer see people sleeping at the door and waiting for the door to open. Gov. Antnio Augusto Junho Anastasias 2011 citizen assistance policy further helped institutionalize citizen rights and gave the team additional tools to enhance the services due the public. Giro said: The decree deals with the quality of citizen assistance for services that the state provides. It also details standards, such as wait times. The procedures are so detailed and exacting that the agencies find it better to join a UAI rather than invest in their own citizen centers. One of the most interesting aspects of the reforms was the phased manner in which the team remodeled the PSIUs into UAIs, experimenting with changes along the way. For instance, the team piloted two new units in 2007 before ramping up restructuring of the remaining PSIUs. In addition, the team experimented with a new public-private-partnership model for cost savings before applying it to other units. Although it was difficult to evaluate the success of the private-public model in 2013, the team learned from previous generations of one-stop shops to inform service delivery at the new units. In 2013, the UAI team remained committed to improving services further. We spend a lot of time discussing how to innovate and improve the delivery of services, said Secretary Vilhena. The UAIs have great positive feedback because we are providing an important service for society. Innovations for Successful Societies makes its case studies and other publications available to all at no cost, under the guidelines of the Terms of Use listed below. The ISS Web repository is intended to serve as an idea bank, enabling practitioners and scholars to evaluate the pros and cons of different reform strategies and weigh the effects of context. ISS welcomes readers feedback, including suggestions of additional topics and questions to be considered, corrections, and how case studies are being used: iss@princeton.edu. Terms of Use Before using any materials downloaded from the Innovations for Successful Societies website, users must read and accept the terms on which we make these items available. The terms constitute a legal agreement between any person who seeks to use information available at www.princeton.edu/successfulsocieties and from Princeton University. In downloading or otherwise employing this information, users indicate that a. They understand that the materials downloaded from the website are protected under United States Copyright Law (Title 17, United States Code). b. They will use the material only for educational, scholarly, and other noncommercial purposes. c. They will not sell, transfer, assign, license, lease, or otherwise convey any portion of this information to any third party. (Republication or display on a third partys website requires the express written permission of the Princeton University Innovations for Successful Societies program or the Princeton University Library.) d. In all publications, presentations or other communications that incorporate or otherwise rely on information from this archive, they will acknowledge that such information was obtained through the Innovations for Successful Societies website. A suggested citation format is as follows: [Document author if listed], [Document title], Innovations for Successful Societies, Princeton University, accessed at http://www.princeton.edu/successfulsocieties on [date accessed on Web] e. They understand that the quotes used in the case study reflect the interviewees personal points of view. Although all efforts have been made to ensure the accuracy of the information collected, Princeton University does not warrant the accuracy, completeness, timeliness, or other characteristics of any material available online. f. They acknowledge that the content and/or format of the archive and the site may be revised, updated or otherwise modified from time to time. g. They accept that access to and use of the archive are at their own risk. They shall not hold Princeton University liable for any loss or damages resulting from the use of information in the archive. Princeton University assumes no liability for any errors or omissions with respect to the functioning of the archive. Innovations for Successful Societies (ISS) is a joint program of Princeton Universitys Woodrow Wilson School of Public & International Affairs and the Bobst Center for Peace & Justice.The Woodrow Wilson School prepares students for careers in public service and supports scholarly research on policy and governance.The mission of the Bobst Center for Peace & Justice is to advance the cause of peace and justice through mutual understanding and respect for all ethnic traditions and religious faiths, both within countries and across national borders. References 1 Res Pblica, Consultoria para diagnstico e propostas de melhoria do PSIUPrograma de Servios Integrados Urbanos, Governo de Minas Gerais, Contrato N 527/07, Processo N 1501122000167/2007, 19, Setembro de 2007. 2 Ibid. 3 Diana Jean Schemo, World Bank Cuts Off Loans to 2 Brazilian States, New York Times, 29 January 1999. 4 Managing for Development Results: Improving Public Services Delivery in Minas Gerais, Brazil, World Bank, 1, 2010. 5 BrazilMinas GeraisWorld Bank Partnership: Building on a Strong Foundation and Leading to Next Steps, Brazil Country Management Unit, World Bank, 18, 2007. 6 Managing for Development Results, 1. 7 BrazilMinas GeraisWorld Bank Partnership, 6. 8 Res Pblica, 11. 9 Ibid, 19. 10 Agencies at the UAIs included the Civil Polices State Department of Transit and the Institute of Identification; Federal Police; Federal Revenue Agency; Commercial Board; Secretariat of Labor and Employment; Ombudsman; Housing Company of Minas Gerais; Ministry of Labor and Employment; Secretariat of Science, Technology, and Higher Education; Tribunal of Justice; Bank of Brazil; Institute of Social Welfare of the Public Servants of Minas Gerais; Energy Company of Minas Gerais; and the Secretariat of Planning and Management. 11 The 2007 budget for the 26 units ran over the allocated 3.13 million-real budget by more than 1 million reals. 12 Governo de Minas Gerais; http://www.planejamento.mg.gov.br/planejamento-e-orcamento/orcamento-do-estado-de-minas-gerais. 13 Posto de Servio Integrado Urbano: Diagnstico e perspectivas, Governo de Minas Gerais, 59, July 2007. 14 Brazils annual average inflation rate was 5.678 in 2008, 4.888 in 2009, and 5.039 in 2010. International Monetary Fund World Economic Outlook Database, April 2013. 15 Governo de Minas Gerais; http://www.planejamento.mg.gov.br/planejamento-e-orcamento/orcamento-do-estado-de-minas-gerais. 16 Numbers represent the authorized annual budget for PSIUs/UAIs. The actual amount spent may vary. 17 Ivan Santos, More Than 7,000 Documents Left in UAI-MG, 5 March 2013; http://www.correiodeuberlandia.com.br/atadiaria/2013/03/05/mais-de-7-mil-documentos-esquecidos-na-uai-mg/. 18 Institutional Presentation: Atendimento ao cidado: 10 anos choque de gesto, SEPLAG, Governo de Minas Gerais, April 2013. 19 Equivalent 2007 dollar amounts: 465 reals (US$198.77); 466930 reals (US$199.20US$397.55); 9311,860 reals (US$397.98US$795.1); 1,8613,255 reals (US$795.50US$1,391.40); and 5,116 reals (US$3,052.01). Table 5: Relevant Terms Descriptions PSIUs Postos de servio integrado urbano (integrated urban service units), one-stop shops for citizen services Programa Empressa Mineira Competitiva Competitive Minas, a program for simplification of business procedures Projeto Estruturador Descomplicar Structuring Project Uncomplicate, a project to strengthen state-business, state-society, and state-state relations Minas Fcil One-stop shops for opening and closing businesses UAIs Unidades de atendimento integrado, integrated citizen assistance units, one-stop shops for citizen services First management shock (2003-06) Gove. Acio Nevess set of reforms to reduce state expenditure and improve finances Second management shock (2007-10) Gov. Acio Nevess set of reforms to reduce state expenditure and improve finances Third management shock (2011-present) Gov. Antnio Anastasias reforms focused on citizen services Table 6: Key Decision Makers and Team Members (mentioned in the case study) Titles Acio Neves da Cunha Governor of Minas Gerais (200310) Antnio Augusto Junho Anastasia Secretary of state for planning and management (200306); Deputy governor (200710) Renata Maria Paes de Vilhena Deputy secretary of state for planning and management (200306); Secretary of state for planning and management (2007present) Fernanda Valadares Couto Giro Assistant secretary of management Breno Eduardo Elias Dos Santos UAI Coordinating Team member (business process manager) Rodrigo Gitirana Lyrio do Nascimento UAI Coordinating Team member (architecture, infrastructure) Felipe Moreira de Oliveira UAI Coordinating Team member (business processes, citizen relationship) Andrey Moraes Labanca UAI Coordinating Team member (information technology infrastructure and systems) Henrique Lage Tomich UAI Coordinating Team member (operations, UAI services) Alexandre Lima Cabral UAI Coordinating Team member (human resources, operations) Erick Brazil de Vasconcelos UAI Coordinating Team member (communications) --------------- ---------------------------------------- --------------- ---------------------------------------- Michael Scharff Innovations for Successful Societies 1 2012, Trustees of Princeton University Terms of use and citation format appear at the end of this document and at www.princeton.edu/successfulsocieties ISS is a joint program of the Woodrow Wilson School of Public and International Affairs and the Bobst Center for Peace and Justice: www.princeton.edu/successfulsocieties. ISS invites readers to share feedback and information on how these cases are being used: iss@princeton.edu. Rushda Majeed Innovations for Successful Societies 2 2014, Trustees of Princeton University Terms of use and citation format appear at the end of this document and at www.princeton.edu/successfulsocieties ISS is a joint program of the Woodrow Wilson School of Public and International Affairs and the Bobst Center for Peace and Justice: www.princeton.edu/successfulsocieties Rushda Majeed Innovations for Successful Societies Innovations for Successful Societies 37 2011, Trustees of Princeton University