CEMENTING THE RIGHT OF OWNERSHIP: LAND REGISTRATION IN KYRGYZSTAN, 1999-2009 Maya Gainer drafted this case study based on interviews conducted in Bishkek and Kant, Kyrgyzstan, during November and December 2016. The British Academy-Department for International Development Anti-Corruption Evidence (ACE) Program funded the development of this case study. Case published February 2017. SYNOPSIS In 1999, eight years after emerging from decades of Soviet domination, Kyrgyzstan began an ambitious effort to officially recognize property ownership throughout the country and lay the groundwork for a vibrant real estate market. During five and a half decades of rule by the Soviet Union, citizens were not allowed to own land, and after Kyrgyzstani independence in 1991, the country began a nationwide program of privatization in a bid to stimulate economic development. The question was how to register and document property rights so that people could transact efficiently in a new land market. To meet the challenge, a new land agency, known as Gosregister, had to hire and train staff in completely new responsibilities, establish performance management and funding structures, improve efficiency by introducing new technologies, and ensure that staff did not engage in corruption. Despite political upheaval-including the overthrow of two governments in the space of five years-Gosregister steadily built its capacity and evolved into an effective land registry. By 2012, the agency had registered 92% of the country's privately held parcels, and in 2017, the World Bank's Doing Business rankings recognized its services as among the best in the world. INTRODUCTION When the State Agency for Registration of Immovable Property Rights opened its doors in 1999, the new land agency confronted many challenges, not the least of which was how to navigate a sea change in Kyrgyzstan's economy. "It was our first time dealing with private property," said Narynbek Isabekov, who joined the agency, known as Gosregister, at its establishment and in 2016 was its director. "We had no equipment, no proper experience," he said. The mountainous Central Asian country had been part of the Soviet Union until 1991 and was just starting to officially recognize private landownership. Under Soviet rule, all land had belonged to the state. Individual Kyrgyzstanis might be permitted to use a parcel but not to own or sell it. After independence, Kyrgyzstan embarked on a nationwide privatization and land reform program-a move that distinguished it from most other Central Asian countries, in which the state retained ownership of all land. Although the privatization process involved many institutions and various levels of government, most activity took place locally. Land committees at the rayon (a unit of local government equivalent to a district) and village levels drew up lists of plots and determined who would receive rights to use them. However, because of the decentralized process of assigning rights, implementation varied. Some rayons made decisions about allocation but did not issue official documentation certificates, and others simply failed to implement the process. The certificates that landholders received represented only a first step toward complete privatization. Although landholders could transfer their rights to others through sale or inheritance, the certificates conferred 49-year exclusive-use rights rather than full ownership, because the state was still the sole owner of land, according to the 1993 constitution. (In 1995, a presidential decree extended use rights to 99 years.) In addition, breaking up collective farms was the main focus. There was little effort to formalize property rights in urban areas or in the rural settlements where farmworkers lived. Privatization was a start, but Kyrgyzstan's leaders wanted to go further. "We hoped that if we quickly moved to a new system based on a market economy, we could achieve some positive results," said Tolobek Omuraliev, who became Gosregister's first director. The country's leaders planned to formally recognize rights to land and immovable property and create a transparent central register of ownership and boundary information in the hope it would lead to the establishment of credit markets that used property as collateral and create incentives to develop the land. The work of Peruvian economist Hernando de Soto, who posited that formal recognition of property rights provided a catalyst for economic development, influenced their views.1 In 1997-98, working with the US Agency for International Development, two rayons had piloted land registration systems by experimenting with procedures and communication strategies that could be applied around the country. Based on that experience, in 1998 Kyrgyzstan amended its constitution to allow private landownership and passed the Law on State Registration of Immovable Property. The law mandated the creation of a single agency charged with registering property rights: the State Agency for Registration of Immovable Property Rights, widely referred to as Gosregister. The registration law also said the government would not officially recognize or protect any ownership right until a valid claim had been registered with the agency. The 1990s saw dramatic changes in nearly every aspect of Kyrgyzstan's government and economy, and the new agency had to start almost from scratch. "We had ended all the Soviet rules of socialist property and began to abolish the institutions involved in real estate," Omuraliev said. Gosregister was responsible for building a new system. THE CHALLENGE To lay the foundation for an active real estate market, Gosregister had to recruit skilled staff, establish efficient procedures, and register individual parcels across the country-in an environment in which private landownership was largely an unfamiliar concept. In a country that had had no private ownership of land for 55 years as a Soviet republic and a history of nomadic herding among ethnic Kyrgyz before that, formalization of individual land rights was an unprecedented step. Gosregister had to introduce the idea of officially recorded land rights and educate landholders about the registration process. Even after the agency developed an effective message, communicating it posed a logistical challenge. The vast majority of Kyrgyzstan's approximately 5 million citizens lived in rural areas, and with 94% of the country's area more than 1,000 meters above sea level, some communities were difficult to access.2 The transition also would be challenging for government officials, who had to take on new and different roles. Many of Gosregister's staff had worked in Soviet-era institutions such as the Bureau of Technical Inventory (BTI), which kept records of residential properties, and the State Land Management and Land Resources Agency, whose primary responsibility was land management in rural areas. But despite experience with some aspects of land administration, "we weren't taught how to register private parcels," said Nuria Sooronova, who had been with the land management agency before joining Gosregister and later became the head of its cadastre division. Bakytbek Djusupbekov, who in 2016 headed Gosregister's Geographic Information System center, an internal group of technical experts, recalled that with a staff with no prior experience, one of the biggest challenges lay in "developing the legal base and procedures-all the steps from application to archiving." Gosregister also had to piece together fragmented land data. During privatization, while local officials had the primary responsibility for issuing certificates, the central Ministry of Agriculture had set land reform policy, and its rayon and oblast (province) agrarian reform centers had assisted local governments with land allocation. A separate State Land Management and Land Resources Agency, established in 1996, surveyed and mapped the boundaries of the newly divided parcels.3 Depending on the region and the land's use, records might be stored at the BTI, the land management agency, the state architect's office, the cartographic agency, local government offices, or some combination. Land records were scattered among various agencies across the country; all of them were on paper; and many lacked maps or other details. At the time of Gosregister's establishment, "There were simple records in books at the BTI, land management agency, and cartographic agency saying, 'Mr. X owns apartment Y,' or 'Mr. Z owns this land,'" Omuraliev said. Gosregister had to consolidate the existing records and build on them to develop a comprehensive database that would facilitate real estate transactions. In addition, some parcels-especially residential ones-had no records at all. Soviet officials had allocated plots for housing, but "people would lose the papers because it was just a sheet of paper saying someone gets to build on this piece of land," said Bolot Berikbaev, who coordinated registration work at the rayon level and later managed the national effort. During the land and agrarian reform process, the government had emphasized reorganizing state and collective farms and paid much less attention to residential areas. "There were no procedures to formalize houses," and few had documentation, said Myrzabek Shamshiev, head of the registry office in Issyk-Ata rayon outside the capital city of Bishkek. Because the 1998 registration law required that landholders record their claims at Gosregister to obtain government protection of their rights, the new agency had to establish mechanisms that would determine ownership and resolve disputes when documents were not available. Gosregister also had to contend with limited staff capacity and equipment. In addition to staff members' inexperience with administering private land, existing institutions did not have enough skilled cartographers and surveyors for a nationwide registration effort, and their equipment was often rudimentary. "We would get bruises on our palms" from the wooden sticks used for measuring boundaries during the land reform process, Isabekov recalled, "and we would use horses to measure the bigger land plots." Gosregister had to use its existing resources more effectively and train its workers in new responsibilities. As Gosregister's managers were developing procedures and registering properties, they had to contend with another common problem: corruption. Graft had flourished after the fall of the Soviet Union, and the privatization process had made it easy for well-connected people to obtain choice pieces of former collective farms for next to nothing.4 "After independence, people were given the rights to land shares and plots to build their own houses, but we are not talking only about people who didn't have their own properties then," said Jalalbek Baltagulov, director of Kyrgyzstan's association of municipal governments. There had been a number of violations of the land code and cases of unauthorized construction, he said, "and we should recognize that those violations were committed not just by ordinary people but also by people in charge of the process. Some used their positions to get land plots that were supposed to be for public use and not development. That is our reality." Although some damage had already been done, Gosregister had to be careful to avoid creating further opportunities for land grabbing during the registration process. In addition, administrative corruption such as bribery had become widespread.5 Gosregister had to design procedures and incentives to deter corruption among its staff. Dealing with all of the challenges required substantial resources, and to register the entire country, the agency had to fund salaries for enough skilled staff, acquire new equipment, and improve record storage facilities. The government had funded Gosregister's predecessor agencies from the central government budget, but continuing that model was risky. Kyrgyzstan's coffers were lean: In 1998, gross domestic product per capita was US$345, and government consumption expenditures totaled US$294 million.6 Development aid poured into Kyrgyzstan during the years after independence and could help Gosregister get started, but those funds were unsustainable. Gosregister had to develop revenue sources and financial structures that could keep the agency running for the long term. FRAMING A RESPONSE After parliament passed the registration law in December 1998, Kyrgyzstan's government had a broad legal mandate to establish the State Agency for Registration of Immovable Property Rights. However, the details of doing so were contentious, especially with regard to the agency's position in the government. The original proposal during the legislative process had been to place it under the Ministry of Justice; another possibility was to assign responsibility for registration to the State Land Management Agency. As the debate was going on, the World Bank reached out to Kyrgyzstan's government to discuss funding a project that would build the capacity of the soon-to-be-created land agency and register properties. Kyrgyzstan's president at the time, Askar Akayev, assigned Omuraliev, a close adviser who was serving as minister of local government and overseeing local land reform efforts, to negotiate with the World Bank. Omuraliev said that during their discussions, he raised an alternative idea for the agency's structure: merging the functions and workforces of the land management agency, the BTI, and the State Cartographic Agency into a single independent institution and having it move quickly to become self-financing. "We wanted to make Gosregister independent because it dealt with very important issues of property and land, and we didn't want any state body or local authority to have influence over it," he said. The new agency could work from the data collected by its predecessors. "Their archives contained very limited information, but it was at least something." Although the World Bank initially had favored placing the agency within the Ministry of Justice because it had experience with movable property registration and would ease the process of issuing and adjusting regulations, the land management agency "had the cadastral information and more capacity to deal with rural and agricultural land," said Asyl Undeland, a World Bank operations officer at the time who participated in the discussions and subsequently worked on the bank's registration project. But Omuraliev championed the idea of creating an autonomous agency, Undeland recalled, and she and her colleagues saw the value in it. "We agreed that it would be self-financing and autonomous but linked to the land management agency, and that turned out to be a strong factor for success," she said. Some other outsiders were skeptical, however. "The argument at the time was that it gave the land registry too much power, and there was a lot of worry about corruption," said Renee Giovarelli, a United States-based land law expert who helped draft the registration statute. A stand-alone agency might avoid political pressure to allow land grabbing or channel funds to officials, but with little direct oversight, it would have to rely on strong internal controls to prevent corruption. In early 1999, the president settled the debate with a decree that merged the three institutions and established Gosregister as an autonomous agency. Akayev also appointed Omuraliev as Gosregister's first director. When the agency began operations later that year, the government was concluding negotiations with the World Bank for the land administration project, and in June 2000, the bank approved US$9.4 million to support the land agency during the next five years. The project focused on two key areas: building Gosregister's capacity, mainly through training and technical advice, and registering a target of 600,000 properties around the country. Although the World Bank's land registration experts would oversee the project from Washington, D.C., a local team hired by the bank would lead implementation, working closely with Gosregister's managers. The World Bank's financial support was especially important because Gosregister had to cover its own costs. After a transitional period, government funding for the land agency-whose predecessors had been funded from the central budget-would cease. "There wasn't much money in the state budget," Omuraliev said, so the bank's long-term loan gave the agency time to develop before funding its own operations. Over time, the hope was that making the agency self-sufficient would force it to either generate its own revenue or attract funds from donors, and the national budget would not have to cover expensive endeavors such as mapping or nationwide registration. GETTING DOWN TO WORK During the next decade, Gosregister built itself into a fully functioning land agency. It first recruited and trained staff and developed its registration procedures, setting the stage for nationwide systematic registration. During and after the registration campaign, it made investments in long-term effectiveness, including information technology, performance management, and corruption prevention. Building the workforce The staff of the three institutions that made up Gosregister provided a starting point for the new agency's workforce. Although some employees of the BTI, the land management agency, and the cartographic agency had technical skills such as surveying, they had to apply them in completely new ways. Kyrgyzstan was just starting to introduce transactions like mortgages, for instance, a concept foreign to many workers. In addition, Omuraliev said, many staffers lacked adequate training for their new jobs. But because of high unemployment at the time, laying off existing staff would create its own problems. "I knew if I dismissed all those people, they'd protest," he said. Instead, Omuraliev and his senior managers decided to retrain staff members and keep them on provided they could pass an exam-especially because the agency needed employees to handle the planned nationwide registration of property rights. Staff members at the three predecessor institutions took a three-month course focused on the new registration law and the types of transactions and properties they would encounter in their work at Gosregister. Those who failed the exam at the end of the course could take the test once more, but a second poor result disqualified the person for work at the new agency. The agency was in the process of developing more-granular procedures, but the key elements had already been spelled out in the registration law; and as working groups ironed out the remaining details, employees received additional training. In addition to having to take the course and pass the exam before joining Gosregister, employees of the three combined institutions had the option of going through an additional round of training and exams to apply for positions under the World Bank project. The project competitively hired-mainly from the same pool of people with land administration experience-for its Project Implementation Unit, a team of Kyrgyzstani specialists who had the tasks of overseeing project activities and coordinating between the bank's technical advisers and Gosregister's management team. The project also hired staff to coordinate the systematic registration process at the oblast and rayon levels. A committee of bank staff and senior managers from Gosregister reviewed applications, conducted interviews, and oversaw the training and exams before deciding whom to hire. Although these people worked closely with Gosregister staff, they were on the World Bank's payroll and earned substantially higher salaries, so the positions were in high demand. "The World Bank project work started with a big competition . . . and there were about 25 applicants for every position," recalled Bolot Tashtanov, who obtained one of the positions and later transferred to Gosregister as its lead monitoring and evaluation specialist. Tashtanov acknowledged that the pay differential created some tensions, but he stressed that he and his colleagues worked hard to build relationships with their counterparts at Gosregister, and most joined the agency after the bank's projects ended. Gosregister assigned staff members to 49 local registry offices (LROs) around the country. The agency drew largely on the local offices of the former BTI to open branches in each rayon. Even though many areas were likely to have low transaction volumes and it would be difficult for an LRO to sustain itself using registration fees, it was important to have offices "in every rayon and every city," Djusupbekov said, because it was difficult for poor people to travel to register. Because Gosregister was in many ways a work in progress during its early years, the agency continued to devote substantial time and resources to staff training. As new procedures emerged and old ones evolved, Omuraliev said, staff "needed to learn them by heart, know how to apply them, and be able to explain to others why they are necessary." The agency opened two centers-one in the southern city of Osh and one in Kant, a town outside Bishkek-that provided general training on procedures, including annual refresher courses and more-specialized ones on skills for key positions. Staff members who were responsible for approving and recording land transactions had to complete another course and exam to be certified as registrars, with additional courses and certifications required for handling particularly complex transitions. The agency also trained staff members in technical areas such as advanced mapping and IT, especially as it adopted more-sophisticated technology. The Swedish International Development Cooperation Agency partnered with the World Bank to fund training and send mapping and registration experts to advise Gosregister, and several Gosregister managers and selected staff visited Sweden to learn about the Swedish registration system and to take advanced training courses. "We looked a lot at the Swedish experience," Sooronova said, and used their procedures as a model. In addition, staff who had been trained in Sweden "were given opportunities to introduce things they learned," she said, thereby enabling the agency to maximize the impact of the limited number of individuals who participated. Developing procedures Although the 1998 registration law stipulated the basic requirements for registering property ownership, Gosregister had to develop specific procedures for its staff. Gosregister's managers assembled working groups to establish standard procedures and regulations ranging from registration forms to mapping standards. The groups consisted mainly of representatives from registry offices around the country, members of Gosregister's central management team, World Bank project staff, and experienced lawyers from Kyrgyzstan and elsewhere enlisted for the project. The agency had to work with the Ministry of Justice to issue official regulations or revise legislation, Undeland said, but "Gosregister had some flexibility to adjust how the guidelines were implemented, and they could introduce new procedures." Based on the 1998 law and examples from around the world-notably Sweden, which served as a model of an advanced system and which had provided technical assistance for land agencies in developing countries since the 1980s, and Lithuania, which demonstrated how a post-Soviet country could make strides in land registration-the working groups drafted a registration manual. "The manual had chapters for different types of property, the types of codes associated with types of property, [and] step-by-step instructions for systematic registration," Tashtanov said, which provided the basis for every aspect of the registry offices' work. Over time, the agency adapted some of its procedures based on local registry offices' experiences. "We'd figure out better or faster ways of doing things," Berikbaev said. For instance, the team at Berikbaev's LRO suggested adding a field to the database for the year of a building's construction, because banks often asked for the information. Jakshylyk Toktosunov, who headed an LRO before leading Gosregister's registration department, said this bottom-up approach was typical. "The head office would suggest certain procedures, and we would test them," he recalled. Years later, he added, the agency maintained the same flexibility. "LROs still come to us with questions and suggestions if they run into a problem, and we consider it to see if we can amend the procedure," he said. Nationwide registration Because the 1998 law mandated that property rights had to be registered at the land agency before they would be recognized officially, Gosregister launched a nationwide property-registration campaign shortly after its establishment. The effort, paid for by the World Bank and coordinated by its project staff, launched in 2000 and lasted until 2007. Known as systematic registration, this approach was intended to secure and recognize tenure rights for all property owners, lower costs for the agency and allow it to offer free registration by creating economies of scale, and form a basis for an active land market nationwide. "If we made registration voluntary, it would take decades to finish," Omuraliev said. Especially because people were not yet persuaded of registration's importance and many could not afford it, allowing landholders to register only when they needed to make transactions would likely have resulted in only the most-valuable properties' being registered, with large amounts of land held informally. Gosregister's managers knew that registering an entire country's land rights would be a lengthy process. So, in order to quickly create a supply of registered properties needed to enable further real estate transactions and meet the needs of credit markets, they decided to focus initially on densely populated areas, where most transactions took place. The systematic registration process started just outside the major cities so that staff could gain experience. Then the process moved into the major cities of Bishkek and Osh, into large towns that served as district centers, and later, into rural and agricultural areas. "The idea was to give people an opportunity to use property as collateral, and there were more chances to do that in urban areas," Berikbaev said. To lay the groundwork for systematic registration, Gosregister engaged in nationwide and locally targeted information campaigns to explain how registration worked and why it was beneficial. Although the agency used mass media, the most effective mechanism involved public meetings. "We'd come with people they trust" such as community leaders, Tashtanov said, which helped people accept the process. Few people intuitively understood why registration was important, so giving people the opportunity to ask questions and see local leaders-whether the heads of neighborhood organizations, elected officials, or clan or religious leaders-endorse the process before a registration team came to their village helped smooth the way. After the information campaign, Gosregister's teams divided the area into small zones and sent teams of legal and mapping specialists house to house to confirm parcel boundaries and ownership. "If they had an ownership certificate, that was sufficient," Toktosunov said, and their information would be sent to the LRO with a recommendation that it be registered. However, not everyone had such a certificate. Tashtanov recalled that in rural areas, as few as 15% of households had documents demonstrating ownership. In some cases, people had lost their records, and others had never received them. When the residents were unable to provide certificates, Toktosunov said, "we had to dig through the archives to figure out who were the real owners." Hunting for records on specific properties was challenging. The Gosregister teams first looked to local-government archives, but the information might also be stored in several other places, such as the state architect's office or a notary's office. Sometimes they were able to find another copy of the ownership certificate; in other cases, Tashtanov said, municipalities had "household books" that listed who lived where and who was responsible for bills like electricity. That evidence, combined with local knowledge, was often enough to confirm ownership of a particular parcel, but if the documentation was spotty, registration teams could recommend the LRO issue a preliminary registration that could be used until the resident provided additional proof of his or her claim. To safeguard against legitimizing false claims of ownership or against entrenching disputes, once a registration team had compiled a preliminary map and list of owners, the team publicly posted the information for three months so that anyone who objected could dispute the claim at the land commission. LRO staff processed an area's registrations only after the notification period ended. "There was a strong transparency and awareness component. The community could see the maps of the land, who owns what, and whether they have the right establishing documents," Undeland said. The process helped expose specific instances of land grabbing because those who had taken properties during privatization could not provide evidence for their claims. As a result, she added, some land got reallocated to community members during registration. In about a fifth of cases, the registration teams encountered conflicting claims-often due to inheritance disputes-or neighbors who disagreed on a parcel's boundaries. Those cases went to specially established land commissions, which were typically set up at the rayon level and included Gosregister staff, local government officials, community representatives, and representatives of the state architect's office. The commission members weighed the evidence for each person's claim to a certain parcel and made a decision, after which the LRO could register the property. The commission's decisions were not final, however. Dissatisfied claimants could appeal to the regular court system. Neither did the commissions resolve every case. Although the systematic registration effort covered 2.5 million properties by the time it ended in 2007, approximately 200,000 other parcels remained unregistered because of conflicting information about ownership rights that required additional time to resolve.7 In addition, pockets of informal settlement, called novostroiki, emerged as Kyrgyzstanis migrated to urban areas, creating more complications for the registration process (Textbox 1). Preventing corruption As in many countries, land administration in Kyrgyzstan was seen as a fertile area for malfeasance in the forms of both petty administrative corruption such as bribery for faster services and larger-scale wrongdoing such as land grabbing. During its first few years of operation, Gosregister adopted several key measures to stop staff from engaging in bribery or fraud. One problem came to light during systematic registration. Groups of scammers impersonating government employees sometimes stole people's ownership certificates, likely with the intention of taking the land for themselves. "That information spread, and it made people nervous," Omuraliev said. Some villages even chased away the agency's teams because of concerns they were impostors. To solve the problem, the agency introduced distinctive uniforms for its registration teams and worked harder to publicize the date and time that legitimate Gosregister staff would be visiting the area. As clients started visiting LROs to register transactions, the risk grew that bribes might be solicited or offered for faster service. To reduce opportunities for corruption, Gosregister separated front-office functions from back-office functions, so that agents who processed transactions would never interact directly with clients. Managers assigned the case files to registrars randomly-later aided by the agency's registration software-so that clients would not know who was processing their transactions and registrars would have difficulty soliciting bribes from clients. The agency also attempted to enlist the public in corruption prevention by providing copious information regarding the registration process. LROs publicly posted fees and timelines for each service, Sooronova said, "so [clients] know what to expect." Offices also posted lists of phone numbers so citizens could report corruption, and they maintained separate complaint boxes. In addition, Gosregister adopted an official fee for expedited processing-a move that aimed not only to head off bribery but also to boost revenues. For some time, the change worked, preventing payments for faster service from being diverted to officers' pockets. However, that tactic eventually lost its effectiveness, said Sualir Himamov, head of Gosregister's financial and economic department. Clients would bargain with the officers who received their documents to mark them for expedited service in exchange for a bribe amount that was lower than the official fee. In November 2016, the agency dropped the expedited-service fee and resumed charging a single fee-but it also tightened procedures and monitoring to offer a shorter timeline for each type of service. "Delays create opportunities for corruption," Shamshiev said. Officers had to work within the law, he stressed-but the shorter timelines meant customers had no reason to look for workarounds. Building and monitoring performance Gosregister relied heavily on financial incentives to spur staff performance, and as a result, LRO managers had to closely track both quality and quantity metrics. "People need to be motivated. . . so we introduced performance-based salaries," Himamov said. "Salaries depend on the income a person brings to the agency," which for registrars was a percentage of each transaction they processed. The percentage varied based on the type of transaction-more-complex ones provided a greater percentage to compensate for the longer time required-and each registrar had monthly targets for the number and types of transactions they were expected to process. Support staff had fixed salaries but received bonuses based on overall revenues. Staff members hoping to boost their salaries could not cut corners on quality, however: each error resulted in an automatic deduction from the employee's paycheck. Disciplinary infractions, such as lateness, also resulted in financial penalties. Gosregister made performance monitoring a central part of its operations and culture. The central office tracked the number of transactions at each LRO, as well as revenues, through quarterly reports, and randomly inspected the offices to evaluate service quality. Monitors also used office visits to check for cleanliness and orderliness, make sure that the required service standards and anticorruption information were posted, observe processing to see whether all procedures were being followed, and randomly check records for errors. "We'd also do opinion research and interview people about what they needed and how they felt about the procedures," Djusupbekov said, and the agency would try to simplify the steps or reduce processing times accordingly. Much of the monitoring was carried out by specialists paid by the World Bank at the regional level, and after projects ended, monitoring responsibilities shifted to the central office. With fewer staff members dedicated to monitoring, the number of such visits to LROs fell significantly. Individual performance management, however, remained consistent even after the World Bank projects ended. Managing self-financing Financial self-sufficiency had been Gosregister's goal since its establishment, but the agency had time to make the transition. In addition to donor funding, the agency received several years of central government allocations. Systematic registration, a network of LROs, skilled staff, and information campaigns that educated citizens about property markets were all prerequisites for a market that was sufficiently active to generate fees covering Gosregister's operating costs. But by 2005, the pieces were largely in place, and each LRO was expected to cover its own operating expenses. The system had several key advantages over reliance on a budget set by the government. Because LROs were allowed to keep their revenues and distribute them in the form of salaries or bonuses, staff had strong incentives to work hard and run the LROs efficiently. Earning its own funds also gave Gosregister the flexibility to offer substantially higher salaries than the regular civil service did, which helped attract high-quality staff and reduce turnover. However, "we knew self-funding would have its challenges," Omuraliev said. In places like Bishkek, where the real estate market was active, LROs usually made more than enough to cover operating expenses and could offer bonuses or invest the funds in new equipment and other improvements. For LROs in other places, especially those in remote rural areas, transaction volumes were much lower, and some had difficulty covering their costs. Those LROs kept costs low by keeping less space and fewer staff, but even so, Himamov said, 9 of the 49 offices were unable to make ends meet. Ten others were able to cover their costs but had very little left over to invest in improvements. Because each LRO was a separate legal entity, Gosregister had no formal mechanism to compel high-income LROs to subsidize their weaker counterparts. The agency could do nothing more than ask offices with surplus funds to help others out of "solidarity," Himamov said. "So far, we've been able to convince them, because they don't want any tension," he said in 2016, adding that Gosregister was exploring changes to its legal structure to enable the pooling of certain funds that could be distributed by the central office. In addition to their uneven distribution around the country, revenues could be uneven over time. For instance, the real estate market slowed as a result of international sanctions placed on Russia in 2014, which reduced remittances from Kyrgyzstanis working in Russia-often a key source of real estate investment-and direct investment from Russians. However, Himamov said, "the Kyrgyzstani market isn't that integrated globally, so there's not much immediate effect." LROs-especially those in the key investment areas of Bishkek and Osh and in the resort town of Issyk-Kul-had time to evaluate global trends and adjust their staffing and operational plans accordingly. In rural areas, the market was typically less active and could be predicted to some extent based on more-limited factors such as ministries' plans for construction and activity during the same season during the previous several years. However, LROs had to be ready to adjust. In areas like Issyk-Ata, Shamshiev said, it was difficult to anticipate the number of transactions, and therefore the salaries of the LRO's workers. However, he added, staff became accustomed to the fluctuations and learned how to plan for variations in their pay. Introducing technology Gosregister adopted a strategy of gradual computerization, prioritizing the registration of all of the country's parcels on paper with simple mapping techniques and developing effective manual procedures before attempting to digitize. "We took a simple, step-by-step approach to IT," Djusupbekov said. In 2002, while systematic registration was in progress, a small IT team under the World Bank's implementation unit began designing software called the Automated Registration System, or ARS. During systematic registration, the LROs entered basic data on Microsoft Excel spreadsheets, which formed the basis for the records in the ARS. "As soon as you had data from a district, you'd put it in Excel and Access," a Microsoft database management program, "and then we had to link the databases," IT manager Jipar Davletova said. Davletova's department created an overall structure for the existing data and then worked on developing software to enable LRO staff to enter information directly into the database. Developing the ARS required staff from both the IT team and the LROs to conduct a thorough review of Gosregister's business processes. "We would track a file from beginning to end," Davletova said, and through that process, "we were able to get rid of certain duplicated functions and add others to optimize specialists' work." The reviews resulted in several procedural changes, such as enabling front-office clerks to input information directly into the system as they spoke with clients rather than having the registrars rely on application forms. People often needed assistance in filling out the forms, which led to long wait times, Berikbaev recalled. "Now they don't have to fill out anything manually; they get a printout and sign it," he said. The software also enabled managers to track information more accurately, such as the time it took to process a transaction or the number of transactions an individual had completed. Managers could also identify problems. "The ARS is a good tool for measuring performance. It tracks errors, omissions, and delays, and each case is linked to the person who handled it," Shamshiev said. Corruption prevention shaped several aspects of the design of the ARS. The system tracked the names of all of those who made changes to a record, and it created automatic backups of each title. The software also included different access and editing permissions depending on each person's role, and it barred users from making changes without supporting documentation. In the ARS, "a person has to submit an application and attach documents supporting the changes, and modifications are not allowed without them," Shamshiev said. The implementation unit coordinated the ARS's gradual rollout to the LROs, starting with pilots in the cities of Bishkek and Osh and the Issyk-Ata rayon in 2004. IT capacity at the rayon level was a challenge, Djusupbekov said, and it took time to introduce the system because of the need to train LRO staff. Gosregister also had to purchase computers and servers for the LROs before staff could begin using the system. But by 2007, the system was in use across the country. With completion of the ARS, which was solely for internal use, the IT team moved to make land information more accessible to the public. The agency piloted the Kyrgyz Land Information System in 2008 and launched it officially in 2011. The system integrated ownership data with digital maps, letting a user select a property on the map and get its registration details and vice versa. Information about the property, such as its size, was publicly available, but to get ownership details, a user had to pay for a subscription. As of 2016, most of the subscribers were banks, real estate agents, notaries, and government agencies. OVERCOMING OBSTACLES Gosregister's managers were keenly interested in preserving the agency's independence. The agency's position in relation to other parts of the government had been the subject of heated debate in the lead-up to the agency's creation in 1998-99, and it remained contentious in the following years. "The whole history of Gosregister is the history of small conflicts or big ones with other agencies," Isabekov said. Other institutions-such as the Ministry of Agriculture, which had led agrarian reform, and the Ministry of Justice, which oversaw the notaries who had previously administered property registration-occasionally attempted to take control of the agency or some of its functions, he recalled. Gosregister's proponents cited the agreement with the World Bank as the main reason for continuing the agency's autonomous status. The agreement referred specifically to Gosregister as the project's implementing agency. Maintaining Gosregister's independence was essential to the bank-funded project, they argued. "We would refer to that agreement every time there were attempts to control us," Isabekov said. For the bank, Undeland said, the primary issue was ensuring that Gosregister performed well, avoided corruption, and moved toward sustainability. The agency's placement was secondary, she said, but the team was wary of moves that might transfer its revenues or add new functions. In 2009, however, the agency's decade of autonomous operations ended. President Kurmanbek Bakiyev-an opposition leader who replaced Akayev as president after April 2005's "Tulip Revolution" and quickly came under fire for being even more corrupt and autocratic-decided to create an umbrella agency to oversee all types of registration, encompassing the passport agency, civil registry, vehicle registry, business registry, and state archives in addition to Gosregister. In October, Gosregister-renamed the Department of Cadastre and Registration, or DCR-became a part of the new State Registration Service (SRS). Gosregister's leaders stayed on during the transition, but they feared the move would hinder the agency's authority to make independent operational decisions or would cause its budget to be reallocated to other parts of the SRS. However, circumstances meant those concerns turned out to be largely unfounded. In April 2010, just a few months after establishing the SRS, Bakiyev was overthrown in another revolution. Unlike in 2005, when Bakiyev quickly consolidated power, Kyrgyzstan's politics remained in flux for some time. The country was focusing on drafting a new constitution, holding elections, and responding to political and ethnic violence in the southern city of Osh. The structure of the government's registration services was a low priority, and the DCR's work continued much as it had during its years as Gosregister. Whereas the DCR's operations remained intact, top-level leadership of the new SRS proved highly unstable. "They weren't around long enough to figure out how profitable the DCR could be," Isabekov said. Changes in government and corruption scandals saw the SRS have seven directors in the space of six years. From the point of view of DCR staff, the turnovers were positive developments because they prevented the SRS from meddling with the DCR's way of doing things. The DCR's financial independence also helped insulate it from the upheavals. "Gosregister was able to maintain semi-independence within the SRS because self-financing and autonomous status gives lots of flexibility," Undeland said. By 2014, the agency had generated enough revenue to finance not only the LROs but also its central office and a new Geographic Information Systems Center, created to retain members of the World Bank's implementation unit after the bank's projects ended in 2013. (Many World Bank project staff were hired for managerial positions at LROs or the central office, and the new center created positions for specialists in technical areas such as IT or monitoring and evaluation.) After several years of management turmoil, the SRS stabilized. The umbrella agency organized units that served as liaisons between overall management and the leaders of each department and began work on projects to reconcile the differing data each department held. For instance, because street names often changed-especially in the aftermath of Kyrgyzstan's two revolutions-the land registry, the civil registry, and other arms of the service sometimes used different names to refer to the same places. In 2014, the SRS created a cross-departmental team to build a unified address database and map. Although integration with the SRS was less problematic than expected, the transition was not always seamless. Occasional tensions arose about whether the SRS could use some funds from the DCR's budget to support its other subagencies, and the SRS sometimes asked the DCR to participate in projects that were outside its central mandate-for instance, Isabekov said, helping with the collection of biometric voter registration data when the government introduced a new identification system for the 2015 elections. "We now have to do some SRS work in addition to our own," Isabekov said, but he added that the tasks were manageable. ASSESSING RESULTS Despite a challenging environment, Kyrgyzstan created an effective and innovative land agency that served as a model for others in the region. As of 2017, a country that had no formal property registration until 1999, had a per-capita gross domestic product of US$1,100, and had experienced two revolutions in the space of five years had the eighth-best property registry in the world, according to the World Bank's Doing Business rankings.8 For each of the previous seven years, Kyrgyzstan had been in the top tier of countries worldwide, and in 2017, its distance-to-frontier score, which represented how close the country's performance was to the best performance in the data set, was 90.6 out of 100. Kyrgyzstan scored 24.5 out of 30 on the Quality of Land Administration index-above the average of 22.7 for member nations of the Organization for Economic Co-operation and Development-and received perfect scores for reliability of infrastructure and geographic coverage. The scope of registration in Kyrgyzstan was particularly impressive. During the 2000-2007 systematic registration campaign, Gosregister registered 2.5 million properties-more than four times the original target of 600,000.9 In the same period, approximately 1 million property owners registered transactions by coming to an LRO-mainly for sales, leases, or mortgages that took place after systematic registration had been completed in the area.10 The radical change in the country's property rights regime created an opportunity for Kyrgyzstan to largely avoid the informal tenure challenges that developing countries often encounter, and the systematic registration process reached most of the country's households. By 2012, the agency had registered 92% of the country's private parcels.11 The DCR, which many still called Gosregister after the renaming in 2009, helped lay the foundation for an increasingly active real estate market. The agency reported that sales more than doubled to 67,609 in 2012 from 25,901 in 2002. Mortgage transactions during the same period nearly tripled to 66,612 from 22,387. Although the gains also reflected broader economic performance, the numbers demonstrated that Gosregister had developed the capacity to handle an increasing number of transactions and that the agency's procedures had facilitated, rather than hindered, the growth of registered real estate transactions. After several years, the agency was able to generate its own resources, with all of its LROs' operating expenses covered by fee revenue-albeit with some revenue sharing to address the disparities in transaction volumes. In 2014, the agency established a self-sufficient central office as well. However, many of its investments depended on external funding-primarily by the World Bank. Technology was one area in which such investments paid off. By 2007, all LROs were using the Automated Registration System to record transactions and back up records, and in 2011, the agency launched an online platform for obtaining land information. Gosregister's achievements attracted much international attention. The agency served as an example for several developing countries-especially former Soviet republics. Most notably, Gosregister staff served as technical advisers for land registries in Kazakhstan and Tajikistan. The agency's record was not perfect, however. Corruption was an ongoing challenge, and the data were mixed. A 2007 survey of Gosregister clients conducted by World Bank consultants found that about 17% of respondents had experienced corruption at an LRO,12 although the rate had declined to 4% by 2012.13 The most common form of reported corruption involved bribery for faster service rather than the manipulation of records. However, Transparency International's 2013 Global Corruption Barometer found that 34% of respondents said they had paid bribes to land services within the previous year.14 Despite the agency's carefully designed procedures and incentives, Isabekov acknowledged, "we haven't been able to completely eliminate corruption." REFLECTIONS By the end of its first decade in operation, Gosregister had successfully introduced property registration in a country that had minimal experience with private land rights. It was a challenging task, but starting from scratch created opportunities to design institutions carefully. Although the agency's structure initially caused some concerns about insufficient accountability, Gosregister's staff and partners maintained that its establishment as an autonomous, self-funding agency contributed to its effectiveness and sustainability. The New Public Management-style approach gave Gosregister the flexibility to maintain a strong workforce and created financial pressures for efficiency. "When you have to earn your own money, you have to be smart," said Director Narynbek Isabekov. Exempt from civil service human resources policies, the agency could offer attractive salaries that helped reduce staff turnover and preserve institutional memory. Above-average pay coupled with tight monitoring also helped deter corruption because employees were unwilling to risk their jobs for onetime bribes. Gosregister also served as an example of effective cooperation with donors on institution building. "Donor assistance was key-especially for the equipment and training-but it's important to provide for targeted use of the donor funds," said Nuria Sooronova, head of the cadastre division. Sooronova stressed that capacity building enabled the World Bank's land registration projects to have a lasting impact. In addition, "The projects were very integrated with the head office," Isabekov said. During the two bank-funded projects, which ran from 2000 to 2013, the Project Implementation Unit and the agency's head office developed a close working relationship. When the projects ended, the agency was at risk of losing staff in such areas as monitoring and IT, where the implementation unit had taken the lead; but the agency was able offer the World Bank implementation team competitive salaries to stay on. Gosregister's experience offered lessons for other countries in the careful introduction of technology. The agency's systematic registration effort demonstrated the benefit of using simple, low-cost survey methods to secure tenure rights on a large scale. "Initially, our aims were just to provide registration services and get the data, so that's why we used simple survey procedures," said Bakytbek Djusupbekov, director of the agency's Geographic Information System Center. "If we went for more detail, it would take a lot of time and resources." Technology did eventually increase records security and tighten controls on access, but the agency focused on developing efficient manual processes first and avoided the pitfall of spending heavily for IT systems that would have required staff capacity it lacked at the time. Although systematic registration enabled Gosregister to formalize rights to millions of properties in a short time, certain tenure shortcomings lingered. Politicians and other influential people had been able to obtain land improperly during privatization, and the registration process did not investigate those issues. In some cases, public scrutiny of land allocation during systematic registration forced land grabbers to back off their claims, but in other cases, registration legitimized earlier land grabs. Gosregister also faced an ongoing challenge in dealing with pockets of informal settlement, called novostroiki. If municipal governments did not recognize the informal settlements, the agency could not register residents' property rights. Disputes over the status of several such settlements continued in 2017. Gosregister also was responsible only for privately held land. The government often did a poor job of managing state-owned land, and leases were vulnerable to corruption.15 It was important for Gosregister and other Kyrgyzstani institutions to broaden their perspectives on land issues, said World Bank consultant Asyl Undeland, who on worked the bank's land registration project as an operations officer. "Gosregister is doing a great job, but it's become very focused on providing a technical service. . . . There's a lot that needs to be done with getting banks to accept property as collateral, property valuation, taxes, zoning, and so on; and in a context like Kyrgyzstan's, you have to have a holistic approach." Although challenges remained in 2017, Kyrgyzstan had made great strides since 1999. Former Gosregister Director Tolobek Omuraliev said it took hard work to introduce an efficient and comprehensive property registration system in a country that had prohibited private landownership for decades. But also important, he said, were changes in knowledge and mind-set: "At the time, all of us were former Soviet citizens, and we had no idea about private property registration. Now the Kyrgyzstani people can understand and defend their property rights." 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 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). This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. 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 party's website requires the express written permission of the Princeton University Innovations for Successful Societies program or the Princeton University Library. d. 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. e. 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. f. 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. g. 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. Our status (and that of any identified contributors) as the authors of material must always be acknowledged and a full credit given as follows: Author(s) or Editor(s) if listed, Full title, Year of publication, Innovations for Successful Societies, Princeton University, http://successfulsocieties.princeton.edu/ (c) 2017, Trustees of Princeton University References 1 Hernando de Soto, The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else. Basic Books: New York, 2000, https://www.ild.org.pe/books/the-mystery-of-capital-versions/the-mystery-of-capital-all-versions/the-mystery-of-capital-usa. 2 Central Intelligence Agency, "Kyrgyzstan," World Factbook, 2017, https://www.cia.gov/library/publications/the-world-factbook/geos/kg.html. 3 Renee Giovarelli, "Land Reform and Farm Reorganization in the Kyrgyz Republic," Rural Development Institute Reports on Foreign Aid and Development No. 96, 1998. http://www.landesa.org/wp-content/uploads/2011/01/RDI_096.pdf. 4 Murat Cokgezen "Corruption in Kyrgyzstan: The facts, causes and consequences," Central Asian Survey 23(1)(2004): 79-94. 5 Saltanat Liebert and Medet Tiulegenov, "Public Administration in Kyrgyzstan," in Public Administration in Post-Communist Countries: Former Soviet Union, Central and Eastern Europe, and Mongolia, eds. Saltanat Liebert, Stephen E. Condrey, and Dmitry Goncharov. Boca Raton, FL: CRC Press, 2013, 92-93. 6 World Bank, World Development Indicators, "Kyrgyz Republic: General government final consumption expenditure" and "Kyrgyz Republic: GDP per capita, current US$," http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators#. 7 World Bank, Kyrgyzstan Land and Real Estate Registration Project Implementation Completion and Results Report, 10, http://documents.worldbank.org/curated/en/463911468047089384/pdf/ICR104300P04971C0disclosed071241091.pdf. 8 World Bank, Doing Business, Kyrgyz Republic," 2017, http://www.doingbusiness.org/data/exploreeconomies/kyrgyz-republic. 9 World Bank, Kyrgyzstan Land and Real Estate Registration Project Implementation Completion and Results Report, 10. 10 World Bank, Kyrgyzstan Land and Real Estate Registration Project Performance Assessment Report, 11, http://documents.worldbank.org/curated/en/303271468263988147/pdf/553820PPAR0p041IC0disclosed08131101.pdf 11 Klaus Deininger, Harris Selod, and Anthony Burns, "Kyrgyz Republic," in The Land Governance Assessment Framework: Identifying and Monitoring Good Practice in the Land Sector. World Bank: Washington, DC, 2012, 69. 12 Asyl Undeland, Assessing Gosregister's Performance: Building Monitoring Capacity and Public Opinion Survey Results, http://siteresources.worldbank.org/INTECAREGTOPRURDEV/Resources/FINAL_REPORT_ENG_Nov12_07.pdf. 13 World Bank, Second Land and Real Estate Registration Project Implementation Completion and Results Report, 14 Transparency International, Global Corruption Barometer 2013: Kyrgyzstan, http://www.transparency.org/gcb2013/country/?country=kyrgyzstan. 15 Klaus Deininger, Harris Selod, and Anthony Burns, "Kyrgyz Republic." --------------- GLOBAL CHALLENGES:Land Kyrgyzstan Innovations for Successful Societies (c) 2017, Trustees of Princeton University Terms of use and citation format appear at the end of this document and at successfulsocieties.princeton.edu/about/terms-conditions. 18 ISS is program of the Woodrow Wilson School of Public and International Affairs: successfulsocieties.princeton.edu. ISS invites readers to share feedback and information on how these cases are being used: iss@princeton.edu. (c) 2017, Trustees of Princeton University. This case study is made available under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.