MOVING BEYOND CENTRAL PLANNING: CRAFTING A MODERN POLICY MANAGEMENT SYSTEM, LATVIA, 2000 - 2006 SYNOPSIS In 2000, Latvia's newly appointed state chancellor, Gunta Veismane, took on a daunting task. Since Latvia's independence from the Soviet Union in 1991, the new government had functioned without a clearly organized policy-planning process. Ministries produced policy papers that lacked input from stakeholders or essential information about costs and objectives, leaving decision makers in the dark when trying to set a course for Latvia's future. Veismane's job was to ensure that top officials had the information and analysis they needed to make informed policy decisions. She tapped Una Klapkalne, an experienced government official, to lead an elite unit in the State Chancellery to design and implement a new policy-making system. Between 2000 and 2006, Veismane and Klapkalne introduced rules and procedures that improved the quality of decision making and enhanced coordination across government. The World Bank lauded the system they created as a model for the region. Jonathan Friedman drafted this case study on the basis of interviews conducted in Riga, Latvia, during February 2012. Case published May 2012. INTRODUCTION "No one had ever done anything like this before in Latvia," recalled Una Klapkalne, who took over the nascent Policy Coordination Department in Latvia's State Chancellery in September 2000. Earlier that year, Sir Robin Mountfield, former head of the United Kingdom's Cabinet Office, had laid bare the sad state of Latvia's public administration. Ministries were "independent and defensive," setting their own policies according to their own priorities, Mountfield concluded in an assessment commissioned by the World Bank. The State Chancellery, designed to support the prime minister, acted like the government's "mailbox," where ministries deposited their proposals for Cabinet discussions, rather than an active manager and coordinator of the government's agenda. "There was a hole at the center of government," Mountfield concluded.1 Klapkalne was determined to fill it. Latvians had just weathered a turbulent decade. Following independence from the Soviet Union in 1991, the country had built a multiparty democracy and endorsed market-oriented economic policies. Exposed to new competitors, Latvia's industrial and agricultural sectors collapsed and GDP halved between 1990 and 1993, according to the Economist Intelligence Unit. A banking crisis caused renewed growth to stall in 1995, and Russia's default on its loans in 1998 caused a drop in exports that further undermined recovery. Latvia's politics were even less stable than its uneven economic performance. Governing coalitions fell apart with greater frequency than in any other European country-once a year on average-as the country's many small political parties struggled to build and maintain coalitions. Opaque privatization deals and allegations of fraud surrounding the banking crisis undermined Latvians' confidence in their political leaders. Despite the frequent turnover of governments, Latvia's major political parties managed to agree on the country's basic policy goals, which included membership in the European Union and NATO. Mountfield's report generated momentum to reform the policy planning system. His recommendations were broadly inspired by New Public Management (NPM), a set of principles that emphasized the merits of decentralization and managerial autonomy for inducing better performance within government institutions. Prime Minister Andris Šķēle, a wealthy entrepreneur in the agricultural sector, was an adherent of such principles and had supported piecemeal NPM-inspired reforms during his previous term as prime minister from 1995 to 1997. Additionally, the European Commission, the EU's executive body, informed Latvia in 2000 that although the country ranked near the top of candidate nations in terms of political and economic reforms, the pace of administrative reforms would have to quicken in order to satisfy accession criteria. Mountfield's report echoed the point. "The institutional framework at the heart of Latvia's government is seriously inadequate to plan the profound changes needed for EU accession and the other ambitious objectives shared by the government coalition parties and by the main opposition parties, and most especially to ensure their delivery," he wrote.2 In April 2000, Šķēle appointed a new state chancellor, Gunta Veismane, to lead the reform effort. Veismane, an economist by training, had served as the first director of Latvia's School of Public Administration, the government's training program for civil servants, where she had developed a reputation as a change agent. She had trained state secretaries (the ministries' top civil servants) and understood well the dynamics and problems at Latvia's center of government. Less than two weeks after her appointment, Šķēle resigned as prime minister amid a wave of scandals. Andris Bērziņš, the mayor of Riga and member of the conservative Latvian Way political party, succeeded Šķēle and endorsed the reform effort. Between 2000 and 2006, Veismane and her deputy Klapkalne transformed the State Chancellery into an active coordinator of government activities and created a policy planning system that the World Bank ranked among the region's best. THE CHALLENGE The Mountfield report revealed what Veismane, Klapkalne and other Latvian insiders already knew: Latvia had no effective policy-making system. During the Soviet era, Latvian line ministries had implemented directives from central planners in Moscow. After the fall of the Soviet Union, the Latvian government lacked procedures for developing and coordinating policies across line ministries. As a result, policy proposals proliferated, and Cabinet members spent their time plowing through documents that had little substance. There were few rules and procedures that stipulated when and how ministries could submit policy proposals and what information about costs and impacts they had to provide. Confusion abounded regarding the purpose and thrust of the proposals, which had no standard format. Many proposals were too long yet offered only general background on policy issues. There was no requirement for independent research, analysis or financial feasibility studies. Because few proposals specified the goals or results to be achieved, decision makers struggled to understand the merits and assess the impact and utility of government projects. Reflecting the chaotic process, the Cabinets of various coalition governments had passed 450 policy documents between 1993 and 2000. When Veismane and Klapkalne began their work, it was not clear which of the 450 documents were still active and which had expired, or what the relationship was between documents produced by successive governments within the same sector. Another problem was that there was no unit or mechanism to focus policy planning across government. Although the Cabinet had conducted a priority-setting exercise in an attempt to coordinate planning and focus ministries on key goals, the process was flawed. In the most recent exercise, the Cabinet had identified 19 policy areas and 300 related activities that supported priority goals. In turn, ministries identified a total of 1,700 activities they would perform. With governing coalitions lasting a year on average, there was little chance politicians would be around long enough to implement those ambitious plans. As a result, many within government found the process meaningless. Ministries, busy developing and implementing their own initiatives, frequently dragged their feet on Latvia's broader priorities-specifically laws and regulations necessary for EU accession. A survey produced for the World Bank found that in 1999 ministries submitted 46% of EU-related proposals to the Cabinet more than six months after their agreed-upon deadlines.3 Furthermore, line ministries did not coordinate plans with each other. Proposals submitted by one ministry were not rigorously reviewed by others before the Cabinet gave its imprimatur. As a result, policy documents often overlapped or contradicted each other, creating confusion about the government's aims in certain areas. Klapkalne described the perfunctory Soviet-era coordination process that characterized Latvia's government at the time: "Somebody walked around ministries to get signatures from state secretaries or deputy state secretaries. Nobody saw the last version before the Cabinet meeting." Critically, ministries developed policies and plans separately from the budget process, disconnecting intended activities from available resources. As a result, the Cabinet approved plans without a sense of financial feasibility, and the Ministry of Finance approved budgets without knowing what ministries intended to achieve. Overrun with low-priority policy documents that lacked essential information necessary for proper analysis, Cabinet meetings were tedious and unproductive. The World Bank survey found that in two Cabinet meetings in 2000, the Cabinet spent 40% of its discussion time on non-priority issues that represented 61% of the total items discussed. An internal Chancellery study found that the average Cabinet meeting in 2001 lasted 6.5 hours. An additional challenge-one that shaped all the reformers' efforts-was the need to navigate political and cultural sensitivities. Veismane knew she had to tread carefully in creating rules and procedures for policy planning because of Latvians' general aversion to any system that resembled Soviet-era centralized planning. In devising new rules and enforcement mechanisms, Veismane had to make it clear that the new system had no relation to the Soviet system. Inga Bite, an analyst who joined the department at a later stage, explained, "The meaning of planning was spoiled here. There was a perception of planning as Soviet, that plans should be many and long, detailed in situational descriptions but not in future activity. We had a thinking problem." FRAMING A RESPONSE From the outset, Veismane aimed to create a new unit, to be called the Policy Coordination Department, in the State Chancellery to design and manage a policy-making system. She said the need for a unit with this mandate became evident to her when she had trained state secretaries, and the idea was one of the central recommendations of the Mountfield report. Having won the endorsement of Prime Minister Bērziņš, she needed to present her proposal to the Cabinet and persuade ministers to support her efforts. She won their approval with an approach that was cooperative rather than confrontational, stressing the need to address shared challenges. "I didn't start with 'we will manage everything,' or 'we will look through your papers and tell you you're right or wrong,'" Veismane said. "Instead, I said 'we will help you, we will propose some challenges, propose some directions.' The point was to persuade ministers that the unit was necessary not only for the prime minister but for them as well." Building a team Although Veismane had political support from Bērziņš, she decided to find funds from within the State Chancellery to finance the new coordination department. As a matter of principle and best practice, she said, she wanted to reform the Chancellery without seeking additional funds. Additionally, seeking new funding might create political headaches and delays. Veismane conducted a functional audit of the State Chancellery in the summer of 2000 and spoke with every employee about his or her role to uncover duplicative and unnecessary functions. For example, in a holdover from the Soviet era, the Chancellery still employed a full-time dentist. Veismane fired 34 staffers. "It was not easy because it's dealing with people," she said. "But at the same time, they were doing very routine jobs, and people were staying at the State Chancellery for more than 10 or 20 years." In August, Veismane began reviewing the qualifications of 120 applicants for six positions in the coordination department. She said she looked for analytical skills among applicants, experience in public administration, and diverse and sector-specific expertise. To lead the unit, she selected Klapkalne, an adviser to the Ministry of Interior and former deputy state secretary in the Ministry of Defense. Klapkalne began her career in 1992 at the State Civil Service Administration, where she came to know the chancellor through her agency's cooperation with Veismane's training school. Veismane selected five others who were mostly young, Western-educated and fluent in English. Ugis Sics, former head of planning at Latvia's European Integration Office and one of the analysts selected, remarked that English language skills were highly sought-after at the time for government posts, and it was predominately the young that had those skills. "That was during the time when we joked that kids ran everything in government," he said. Defining a role Klapkalne was aware that her team would encounter political resistance when it began to implement a new planning system. Setting rules for policy planning and for analyzing ministerial plans meant the State Chancellery would take on a larger role than in the past. The coordination department could implement most changes through the Cabinet's Rules of Procedure but still needed political support from the Cabinet to approve the changes. "Creating something like that meant a shift in power, and the ministries understood that right away," Klapkalne said. "Formally, everyone accepted. In reality, those ministers who were weaker were happy about this, and stronger ministries didn't want to lose their power." Klapkalne's unit at the State Chancellery faced crosscurrents. Even though it served the prime minister, in practice the Chancellery was considered apolitical. Veismane and members of the coordination department were civil servants and were supposed to act without political bias. In order to set planning rules, analyze documents and make honest recommendations, her staff needed to remain objective, and they had to maintain their reputation for objectivity throughout government. However, they also understood that they needed the prime minister's support in order to make systemic changes. In introducing new initiatives that would push political limits, they could not take for granted continued support from Bērziņš or future prime ministers. They had to find ways to make themselves useful to the prime minister to sustain reforms without compromising the perception of their objectivity. Often, the coordination department would have to call on Veismane, who had a high profile and important connections, for backing. The coordination unit would have to make rules and procedures that decision makers found useful. Sics summed up the unit's objectives: "The idea was simple-that we want to improve the quality of information that the government as a principal gets from the administration [civil service structures in support of government]. And by improved information I mean better-written policy papers, explanations of some of the results that are expected from policies." Diagnosing problems With Veismane's blessing, Klapkalne and her staff began to seek more precise diagnoses of the problems in Latvia than what was contained in the Mountfield report and what Veismane had presented to the Cabinet. This required months of reviewing Latvia's past policy documents, studying models from other countries, and consulting with outside experts. "First we screened what is out there [in Latvia's past]," said Baiba Petērsone, another analyst on the team. "We found that planning does exist, but [the rules] were not written anywhere-no unified approach, criteria or system." Klapkalne said in reviewing Latvia's past policies, the group asked two questions: "When Cabinet makes decisions, are they well-substantiated or not? Second, just looking at the kinds of policy documents, what is in them, what do they look like, and should it be like that or not?" Klapkalne reached out to the embassies of the United Kingdom, United States and Canada for samples of policies to analyze, and to the United Nations Development Programme and World Bank for advice. "We tried to gather as much information about policy making as possible," she recalled. Department members met with Canadian consultant Gordon Evans, who worked primarily with the Lithuanian government, which was developing its own planning system. They also consulted New Zealand's former Treasury secretary, Graham Scott, on matters of linking policy planning and budget processes. The study lasted more than a year, and deliberations among department members were contentious at times, but the process yielded a reform plan all supported. The unit would set specific procedures and formats for introducing and coordinating policy documents. It would then pilot a planning system that linked policies to budgets in selected ministries, and it would monitor the new rules through reviewing all policy proposals to ensure they complied with standards. Klapkalne recalled the value of building team consensus for specific policy formats and reform options through vigorous-and sometimes intense-debate. "We had terrible quarrels between ourselves, extremely hard ones, because all of us are different, different backgrounds, different ideas of how things should be done, and very strong personalities," she said. "Doors were slammed and pens were flying around the office. But then we went out, and the quarrel was done, all of us were behind that idea, fully and totally." GETTING DOWN TO WORK Although Klapkalne intended to make major changes, she knew she had to implement reforms one step at a time so that her small unit would not be overwhelmed by the workload. "We were only six, so we could not do everything at once," she said. She and her staff would first unveil rules and procedures for policy making, then introduce mechanisms to better coordinate planning between ministries, and last take steps to link the planning and budgeting processes. Framing a policy planning system The first step in the reform process was to standardize the documentation procedure by creating four categories of policy documents-concepts, guidelines, programs and plans-each with its own format, content requirements and purpose. Any proposal ministries wanted to submit to the Cabinet would have to take the form of one of these four. Klapkalne and her staff unveiled the new classification system in late 2001 in a document titled "Policy Planning Guidelines." Aiming to lead by example, they wrote the paper in the format of one of the four types of policy documents. "We had to be role models," said Klapkalne. "So we created the Policy Planning Guidelines in the format of the new system, the format we wanted from others." The coordination department issued standard formats for each of the four categories of policy documents. Sics explained that for each document, ministries had to be aware of "who is the reader, what is the purpose, and what information should be provided." Ministries could present concept papers to the Cabinet to discuss multiple approaches to an issue if the government did not have a defined policy. Concept papers could lead to concrete actions-in which case ministries had to develop one of the three other policy documents-or they could stand alone as analytical documents. Guidelines, the category in which the coordination department introduced the document classification system, referred to policies that covered medium- or long-term spans, from five years to 10 and beyond. Ministries could develop guidelines to initiate new policies or clarify areas where existing policy was unclear. Ministries had to write guidelines in a format that required a formulation of the problem, basic principles of the new policy, objectives, results, proposed actions, financial impact, and reporting and evaluation procedures. Programs and plans had their own uses and formats. Ministries submitted programs as medium-term planning documents covering five-year periods to implement guidelines that had been approved by the Cabinet. In programs, ministries specified the document's relationship to its corresponding guidelines, and specified the objectives, indicators, main tasks, implementation schedule and source of funding in greater detail. Plans, the fourth category of documents, detailed the implementation of short-term tasks that would require one to three years. These tasks could be one of several envisioned in medium-term planning documents, or they could detail how ministries intended to overcome problems resulting from weak coordination with other ministries. Table 1 provides examples of policy documents created under the new classification system. The coordination department introduced the document-category system in late 2001, and the Cabinet incorporated the new rules in its procedures in March 2002. "Luckily, the legal department was working on internal rule of procedures [for the Cabinet], so we included these right away," Klapkalne said. "It became a legal document, which is key, because without that nobody would follow it." To build understanding and buy-in from other parts of the government, Klapkalne said the department had "selling events" for state secretaries and other civil servants. She said that political leaders were not too interested in the new system but that state secretaries and civil servants understood that the changes were necessary. Sics said the new system "showed what we thought was wrong and how we could fix it, and, surprisingly, most people agreed." In addition to setting new standards for policy documents, the coordination department reviewed each such document and legal act and submitted comments to line ministries. Klapkalne explained that the purpose of reviewing documents was to help train ministries and to ensure the quality of documents in the long term. "It was clear that at first the quality would be very poor because the system was new; nobody had been trained yet," she said. "We had to develop training courses so that people understood what they had to write. We were ourselves trying to explain to ministries and later on by examples-show them it's lacking substantiation, not written properly." The coordination unit worked with ministry staff to improve policy documents and organized training sessions through Latvia's School of Public Administration, the government's training program. "We were trying to think ahead," Klapkalne said. "We said we would follow every document and we would give comments. This was our way to safeguard the system." In addition to policy proposals, the coordination department required ministries to submit detailed impact assessments. Although policy papers discussed some situational factors behind selected policies, decision makers needed to know the expected financial, legal and practical ramifications of policies beyond the results and indicators included in policy papers. Klapkalne said ministries were especially weak in assessing financial impact. "There was a tendency at the time to say initiatives wouldn't cost anything, or would cost so much you can't do it-two options, both of them wrong. We had to persuade ministries how to calculate the cost of policies. This was a really tough part," she said. The Ministry of Finance played an important role in reviewing impact assessments along with the coordination department and the ministries of welfare, environment and justice. However, Sics said, the finance ministry failed to provide reliable guidance to ministries about standard costs for basic items, inflation rates, or changes in levels of benefits. Klapkalne's unit also worked to get ministries to perform evaluations, though she said these were often of poor quality. Ministries often claimed they lacked the resources to conduct evaluations or claimed that policies failed simply because resources were not available. Making policy planning more inclusive After creating clear formats for policy documents, Klapkalne and her unit looked to open up planning to incorporate input from other ministries and from stakeholders outside government. The goal was not only to make planning more inclusive and transparent but also to negotiate differences of opinion at lower levels so that high-level decision makers would have clear choices to consider. Klapkalne explained, "Our goal was to solve every issue at the lowest possible level ... to have as few unharmonized documents at government meetings as possible." The unit addressed the coordination problem by introducing new procedures, with the Cabinet's approval, that required ministries to share and openly debate proposals. Ministries would send documents to other ministries and to civil society groups, which then had two weeks to submit comments. Ministries would notify civil society groups about policy proposals through Latvia's Civil Alliance, an umbrella organization that relayed information to dozens of non-governmental organizations. Once the sponsoring ministry received comments, it had two weeks to update its proposals and resubmit them to all stakeholders. Disagreements ranged from legal wording and cost estimates to the coherency and direction of the proposal. The responsible line ministry then organized a meeting of all stakeholders including a representative of the coordination department to resolve any outstanding disputes. If the group failed to agree, disputes elevated to weekly meetings of state secretaries, the top civil servants in all line ministries. Outstanding issues were then addressed at weekly joint meetings of state secretaries and ministers, and finally, remaining items went to weekly meetings of the Cabinet. Klapkalne said the idea was to create multiple stages at which lower-level officials could resolve technical or minor issues so that ministers had to deal only with matters that involved clashes of legitimate opinion. She said, "This was a selling point for government that really 'you're working too much' and that 'you shouldn't be doing what civil servants should be doing; all those technical details should be solved at lower levels.' It was an effective selling point, and actually it turned out that it was true!" The new consultation procedures brought a level of transparency to policy planning that was unprecedented in Latvia and in the region. At every stage of the policy process, proposals, comments and discussion notes were posted to a government website open to the public. Cabinet meetings and joint meetings with ministers and state secretaries were open to public viewing, and civil society representatives were invited to participate. Sics said, "We required all ministries to publish their research; we published all drafts to a government website. Previously, ministries wasted their research budgets, but as their reports were made subject to public scrutiny they had to start spending that more wisely and take research more seriously." Linking planning and budgeting In 1997, Latvia's Ministry of Finance had introduced a system of program budgeting that was meant to provide the ministry with more information about how line ministries used funds. In standard program-budgeting systems used in some form by the U.S., U.K., Canada and New Zealand, ministries gave the detailed costs, objectives and expected outputs for every activity performed. Programs referred to groups of activities that had similar objectives. For example, the Ministry of Education could have a program to raise academic standards in universities, with several initiatives geared towards that goal. However, Latvia's finance ministry did not support ministries with detailed methodologies. Largely in the dark, ministries shaped budget programs around the activities already in place, and produced only some cost information without detailing activities or expected results. The finance ministry's inability to produce reliable three-year budget estimates, in part due to the quicker-than-expected rise in government revenue after 2000, further undermined the program budgeting system. In 2002, the coordination department set out to complement the program-budgeting system with what it lacked-a system of strategic planning. The purpose was to allow ministries to determine the administrative structures and resources, both financial and human, needed to achieve their aims and to make adjustments if they found their current structures and resources were inappropriate for the tasks. In this system, ministries would create three-year plans structured around budget programs, using the logic model of identifying the desired outcomes of programs, results, outputs, activities, inputs and costs, with a set of indicators to gauge progress. Klapkalne said her unit became interested after the finance ministry rejected the World Bank's proposal to develop strategic planning to complement program budgeting. "They [the World Bank] tried to sell the idea of strategic planning to the Ministry of Finance and it didn't work, so we took the idea and started to implement it," she said. Sics explained that the finance ministry simply did not have the resources or the mindset to support strategic planning reforms. "The issue was capacity," he said. "The people in charge just didn't have time to bother with more advanced stuff-didn't have the people to do it." Additionally, he said, the finance ministry still had a Soviet-era understanding of its role. "They were still accountants, not managers of public finance," he said. Veismane agreed: "The Ministry of Finance at that time was a routine ministry, without changes in attitude or way of doing things. There was no personal animosity, but we couldn't do anything with them. They had no interest in changing." Although the coordination department stressed the potential benefits of strategic planning, its members recognized that several ministries, especially finance, were skeptical about the usefulness of such planning. Without support from the finance ministry in setting reliable three-year budgets, the system set up by the department could not do everything it was designed to do. Although it could help ministries think about how to organize their units and allocate resources to achieve their goals, it would not serve as a medium-term planning system because it would not be linked to budgets. To establish the system's credibility, the coordination department decided to begin by piloting strategic planning in select ministries and letting the results speak for themselves. "The idea was right, we knew we should do it,' said Klapkalne. "... We had to go through a pilot stage to understand ourselves what are the benefits, how it should work, and how we really can sell this later on to the ministries." Klapkalne attempted to persuade ministry representatives in a meeting in late 2002 of the merits of strategic planning, and she succeeded in enlisting the Ministry of Agriculture for the pilot program. Sics said it was the perfect ministry, from the unit's perspective. "Agriculture had just undergone a functional analysis and fired 300. It had a history of transformation," he said. The agriculture ministry's state secretary, deputy state secretary and head of internal audit formed a working group that led the process of developing a ministerial strategic plan. The coordination unit helped on process issues but allowed the working group to develop a plan it thought would suit them best. "We gave the Ministry of Agriculture examples [of strategic plans], and they told us what worked for them and what didn't," said Sics. "They had sectoral expertise; we had methodological expertise." The agriculture ministry made important changes as a result of the strategic-planning exercise. The ministry reorganized its 30 budget programs into four program areas, each with a set of outputs and indicators. This change made it much easier to monitor progress and to reallocate funds as necessary. Sics said the ministry also reorganized several units. "They restructured after realizing their existing structure wouldn't allow them to achieve what they set out to achieve in their strategic plan," he said. "Drafting the strategic plan helped them identify internal structural flaws." Encouraged by the success of the agriculture ministry, other ministries joined in. During 2003 and 2004, the ministries of culture, finance, economy, justice and the state chancellery itself adopted strategic planning exercises. Again, the coordination unit offered general guidance and allowed the ministries to tailor the process to their liking. The culture ministry, for example, conducted the process using entirely internal staff, whereas the coordination unit brought in consultants to assist the justice ministry in developing its strategic plan. The culture ministry reported that the exercise of writing the plan helped identify gaps in policies and in mechanisms to collect information it would need to assess its own effectiveness. The coordination unit used the success of the pilot programs to persuade other ministries to make strategic planning standard across government. Based on the pilot programs, the coordination unit set government-wide guidelines, modified formats, and assigned staff to support specific ministries to develop their strategic plans. Sics said one of the lessons gleaned from the agriculture ministry's experience in particular was the need to involve management so that the process was a useful planning tool and not merely a compliance exercise. "Make sure you work with the management team," he said. "It's an important lesson. It's much better than setting up a specialized unit." OVERCOMING OBSTACLES From 2002, when the coordination unit began critiquing policy proposals, it faced political opposition that threatened its ability to continue to operate. "We started facing resistance when the system had to be implemented and we rejected proposals that didn't meet standards," Sics said. Such opposition often involved disagreements over spending. Veismane said, "We did a lot to create better strategies and structures for public administration institutions, but at the same time you find places where money's not spent in the right way. This is when ministries understood what we meant by policy coordination and what it meant as an instrument for the prime minister." Always serving under prime ministers who relied on fragile coalitions, the coordination department became caught up in disputes between coalition rivals. Leaders of some ministries bore grudges over what they perceived as excessively critical assessments of their proposals. Making matters worse, successive prime ministers called on the department to perform arguably political tasks in addition to its core work. In response, ministries leveled allegations of political partisanship and at various times sought the unit's demise. Although Veismane worked to shield the department from the political turmoil, she had to make difficult concessions. In November 2002, Einars Repše of Latvia's center-right New Era Party succeeded Bērziņš as prime minister. Bērziņš had helped initiate the reform effort in 2000, and while in power he had continued to support the coordination unit. Repše, who led Latvia's central bank from 1991 to 2001 and successfully introduced Latvia's new currency, had little experience in public administration. Although Veismane persuaded him to maintain the coordination department, Repše assigned the unit additional tasks that put it in the line of political fire. Specifically, Repše asked the unit to conduct audits of ministries' activities to find potential financial savings and to develop policy proposals on his behalf. "Everything changed at that time, when we started to do analytical work for him," Veismane said. "We found some not so nice cases where ministries were asking for more money than necessary, and when we brought this to his attention, ministries started to hate us." Repše also called on the unit to craft a pension reform plan, a highly sensitive policy issue, and to introduce a new salary structure in the civil service that would be consistent with a performance-based rather than a career-based model. Ostensibly, crafting these policies was the responsibility of other government ministries and agencies, but with coalition partners that had different perspectives on these issues, Repše wanted the coordination unit to develop proposals that included the policy options he endorsed. Repše's coalition partners argued that the coordination unit was serving the prime minister's political interests. From then on, Klapkalne said, "Every next government saw us as servants of previous governments. This created a lot of tension, the perception that we were working for political interests." The coordination department came under intense pressure in 2003 after unit members uncovered evidence of possible corruption in a government contract to install digital television. At Repše's request, Sics and fellow analyst Jānis Jankovskis analyzed a multi-million dollar contract that had been awarded to a company that was thought to be British-owned. Finding irregularities in procurement procedures, Sics and Jankovskis looked deeper and found that the company apparently had ties to numerous influential people in Latvian politics including Latvian media mogul Anderjs Ēķis and former Prime Minister Andris Šķēle. Repše suspended the deal and called for an investigation. The Prosecutor's Office launched a criminal investigation into more than a dozen individuals involved, and the case remained unresolved in 2012. The case was also taken to international arbitrage and settled in favor of the Latvian government. Šķēle's People's Party was in the political opposition when the scandal came out but joined a new coalition government in March 2004 under the Green Party's Indulis Emsis and led a coalition from December 2004 under Aigars Kalvītis. Veismane recalled, "Other parties in the coalition wanted to close it [the coordination unit] down. There was a big struggle at the political level to keep the department open. ... He [Emsis] was supportive, but couldn't be supportive publicly because another party, the People's Party, was pushing, saying, 'It's necessary to close it; this department is not necessary.'" The unit survived intact under Emsis but could not escape unscathed under the People's Party. Kalvītis, under pressure from within his party, offered Veismane an ultimatum: either fire Sics and Jankovskis or lose the whole department. In 2005, Sics and Jankovskis resigned. Klapkalne said this episode was the most difficult during her time at the coordination unit, but she said she appreciated that Sics and Jankovskis put the interests of the unit above themselves. "It was very hard not to be able to protect your people," she said. "But we all understood the idea and were behind it." Despite the turmoil, the unit continued to operate and, according to Veismane, had an effective working relationship with Kalvītis. "We had a good enough relationship when we talked about conceptual proposals and preparing him for Cabinet meetings," she said. ASSESSING RESULTS In 2006, the World Bank conducted a comparative study of administrative reforms in Latvia and seven other Central and Eastern European countries that joined the European Union in 2004. It ranked Latvia at or near the top of the group in terms of the quality of policy analysis and coordination. "Since 2000 Latvia has made significant and in some respects impressive efforts in institutionalizing a new policy making and strategic planning system voluntarily," the report concluded. "... Above all, the reforms have managed to challenge the dominating normative culture of public administration by pushing civil servants (and in many cases politicians) to be more policy and performance oriented. Policy has become a significant part of the language in public administration."4 The Policy Coordination Department achieved much of what its leaders had set out to do. With clear formats and content requirements for policy proposals, decision makers received documents that were easier to understand and contained more relevant information and meaningful analysis. Ministries that piloted strategic planning endorsed the exercise, and the Cabinet made strategic planning compulsory in all ministries beginning in 2006. Coordination requirements helped to make the policy process among the most transparent and inclusive in the region. "This broke down silos," Sics said. "We definitely see the benefits. The language of measuring impacts now is normal, [we are] getting good analysis." As new procedures forced civil servants to resolve technical issues without involving Cabinet ministers, Cabinet meetings became shorter and more focused on policy matters. An internal coordination department study found that the average duration of Cabinet meetings shrank to 2.3 hours in 2005 from 6.5 hours in 2001. With clearer policy documents and fewer technical issues to discuss, the Cabinet could act more quickly and effectively on a greater number of significant issues. Indeed, despite shorter meetings, the Cabinet discussed 2,796 items during 2005, up from 2,443 in 2001. However, there were still significant shortcomings. The coordination unit's internal review in 2006 found that although the quality of analysis of policy documents had improved, ministries did not always meet content requirements, particularly with respect to impact assessments. Although performance indicators were better understood than in 2001, the high rate of turnover of civil servants undermined steady progress. Policy documents were clearer and followed specific formats, but there were too many long-term documents, and the relationship between them was often uncertain. Five successive governing coalitions during a six-year span developed long-term planning documents with little consideration for the work of their predecessors. Without support from the Ministry of Finance, the coordination unit did not achieve its goal of synchronizing policy and budget planning. Strategic-planning exercises helped ministries properly organize resources, streamline budget programs, and ensure they had mechanisms in place to monitor results. However, without the foundation of a reliable three-year budget, and with mid-year budget adjustments common, ministries could not effectively use strategic planning to manage their budgets. Mārtiņš Krieviņš, an analyst with the coordination department who wrote the 2006 study and succeeded Klapkalne as head of the unit the next year, said the strategic planning exercises were a partial success. "A success from this system was not only the flexibility and trying to link it with the budget, but also non-governmental organizations evaluated [strategic plans] because they could clearly understand what are the priorities of the ministry, where they are going to invest money, what are the amounts of the money, what are the intended results," he said. "It opened up the black box of the ministry." REFLECTIONS The instability of governing coalitions and the adversarial relationships between coalition partners dented the Policy Coordination Department's reform efforts during its first six years. Although the unit consisted of civil servants, it designed and implemented reforms that were intended to shape policy making at the political level. It had to distinguish between the aspects of policy planning it could address and others that were too sensitive and politically infeasible. Political considerations forced the department to hold back in several areas. One of the unit's initial goals was to set procedures to identify government-wide priority policies around which ministries could organize their activities. However, these efforts never got off the ground. Una Klapkalne, who led the department until leaving for the private sector in 2007, recalled, "We tried to do something on this level, but we understood we couldn't do it because of our political culture." Political dynamics prevented the coordination unit from proceeding with its effort to invigorate the budget process as well. The unit had hoped to introduce negotiations between the prime minister, finance minister and line ministries over the appropriateness of policy goals and the use of funds. However, Ugis Sics, a department analyst, explained that the unit received feedback, in particular from the finance ministry, that such an approach was not politically feasible. With loose governing coalitions, a confrontational negotiating process could add fuel to the fire. "We wanted budget hearings with the Ministry of Finance and State Chancellery on one side of the table and the ministries on the other side, to have a discussion," he said. "We couldn't get to this point." In the areas where the coordination department did proceed, the unit had to play the role of an impartial quality-control team while fending off attempts by prime ministers and others to politicize its functions. "The hardest part was being able to balance being a civil servant and being in the most politically-influenced place where all the interests come together," Klapkalne said. "I never knew what kind of pressure I would be under." EPILOGUE As part of the Policy Coordination Department's 2006 internal review, Mārtiņš Krieviņš helped develop a second-generation policy planning system for the period 2007 to 2013. He developed the new planning system, outlined in a document approved by the Cabinet in 2006, following consultations with coordination unit's staff and representatives from line ministries, Parliament, and the president's office. The goal of the new system was to correct outstanding weaknesses, expand strategic planning, and address new challenges facing policy planning that had emerged since 2000. For example, Latvia's accession to the European Union in 2004 created new opportunities and obligations in policy planning. Latvia had to harmonize its numerous national planning documents and make its policy objectives consistent with EU membership, and the new system would give policy makers the tools to do that. Krieviņš said the coordination unit set the example, much as it had in 2001, by setting the time in force for the new planning system from 2007 to 2013 to match the EU's budgetary period. During 2007 and 2008, the coordination unit implemented its tasks as planned. Analysts worked with all 18 ministries to develop three-year strategic plans-still without active support from the Ministry of Finance-and initiated a review process to harmonize Latvia's many national planning documents. In late 2008, however, reforms were put on hold. With the onset of the global financial crisis caused by the failure of prominent Western banks, the Latvian government experienced a sharp drop in revenue and a 25% contraction in GDP. The crisis "brought the whole system to nothing," Krieviņš said. "The whole system started to work a little bit, and then it was dropped completely." Prime Minister Ivars Godmanis had to make painful decisions to keep Latvia solvent, including mass layoffs and wage cuts, and he had to negotiate a series of emergency loans from the International Monetary Fund (IMF). The task of producing the data and analyses necessary to manage the IMF loan was so enormous that the prime minister called on all arms of the government, including the coordination department, to set aside their core tasks to focus on more immediate needs. Krieviņš said that in that time of urgency, ministries disregarded their strategic plans and bypassed official planning procedures because there was a sense that the procedures were too cumbersome for the tasks at hand. "The government was preoccupied with the need of actually getting the money to keep the state running," Krieviņš said. "In those circumstance, nobody thinks in terms of budget plans." The coordination department contributed to the effort to guide Latvia through its fiscal crisis in several ways. At the request of the prime minister, the department conducted functional reviews of ministries to identify potential efficiency gains and took on the politically hazardous job of identifying agencies and programs whose functions were no longer required. Finance ministry officials became concerned that the department was performing certain tasks that arguably were their responsibility. Additionally, officials from other line ministries pointed a finger at the coordination department for suggesting what they deemed to be unwise or excessively harsh cuts. "It was a very active, but very sensitive time," recalled State Chancellor Gunta Veismane. "At that time, we were everyone's enemy." Pressure built on Prime Minister Godmanis to shut down the coordination department, but Veismane, friends with Godmanis from their time together as secondary-school students, convinced him otherwise. "I said, 'If you close this unit, you'll have no analysis, no information," she said. "... Here is the brain. These analysts prepare everything for you.'" Valdis Dombrovskis succeeded Godmanis as prime minister in March 2009 at the height of the crisis in Latvia and agreed to shut down the unit by 2010. "He (Dombrovskis) knew it was necessary to have the capacity of this unit," said Veismane, "but political pressure from other parties was too strong." Offsetting the loss of the Policy Coordination Department, Dombrovskis established the Cross-sectoral Coordination Centre in early 2012. He provided the unit with a similar mandate within the State Chancellery and appointed Krieviņš as its director. 1 Sir Robin Mountfield, Public Administration Reform in Latvia, Unpublished report, 2000. 2 Mountfield. 3 State Chancellery of the Government of Latvia, Study on Government's Policy Formulation, Decision-making and Implementation in Latvia, Unpublished report, November 2000. 4 Poverty Reduction and Economic Management unit, Europe and Central Asia, the World Bank, Public Sector Capacity in the EU 8: Report on Strategic Planning and Policy Management in Lithuania and Latvia, October 2006. 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. 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/ Jonathan Friedman Innovations for Successful Societies (c) 2012, Trustees of Princeton University © 2019, Trustees of Princeton University