EX POST CONTROL AND STEERING OF GOVERNMENT AGENCIES: A COMPARATIVE ANALYSIS OF LITHUANIA AND ROMANIA.

• Author: Nakrosis, Vitalis

1. Introduction

   Over the past decades countries of Central and Eastern Europe (CEE) have undergone substantial political and administrative changes as part of their post-communist transformation and accession to the EU. The process also included numerous managerial reforms of public administrations (Mora and Ticlau, 2008; Nakrosis, 2008; Bouckaert et al., 2011). In an attempt to break from the communist past the CEE countries started from relatively similar positions and were subject to shared external pressures (specially economic and political transition, and accession to the EU). However, the outcomes of reforms are divergent in these countries. Therefore, this article seeks to explore how reform policies differed across countries and government organizations, as well as whether they produced different outcomes on result-based management.

   Although previous research provided single-country studies or broad few-country comparisons of national reforms (e.g. Bouckaert et al., 2011), there have been only a few comparative analyses based on multi-country statistical data. For instance, Overman, van Genugten and van Thiel (2015) included both Lithuania and Romania in the sample of six countries in their analysis of accountability arrangements in government agencies. They were found to fit the same general modernizing trajectory (with Lithuania having a somewhat higher degree of marketized accountability arrangements at agency level), but no specific mechanisms leading to these similarities and differences between these countries were investigated. Also, none of the previous analyses tested specific hypotheses of managerial reforms in the CEE region. Therefore, based on the existing theory this article pursues a more in-depth comparison of agency management (autonomy, control, and steering of government agencies) in the two selected CEE countries: Lithuania (1) and Romania (2).

   Our comparative analysis is limited to these two countries because they were the only CEE countries whose survey data on government agencies were available. Both countries were among the agencification leaders in Europe in the 1990s and early 2000s. Lithuania (together with Norway) charged most executive tasks (88%) to agencies from all Western and Eastern European countries included in the comparative 21-country CRIPO study, with Romania running closely behind (80% of these tasks were agencified) (van Thiel, 2011).

   The recent global financial crisis created pressures to rationalize the fragmented and inefficient agency structures. The number of agencies in Lithuania (central-level institutions, including territorial units) dropped from 1,023 at the end of 2008 to 707 at the end of 2011 as a result of de-agencification pursued by the Lithuanian authorities (Nakrosis and Budraitis, 2012). A similar process took place also in Romania: in 2009, the Romanian government restructured or abolished 119 agencies (Bouckaert et al., 2011). In addition, during the financial crisis the governments of both countries explored NPM ideas and introduced new managerial tools (Nakrosis and Martinaitis, 2011) in order to increase control over agencies' performance.

   The main purpose of the article is to describe and explain the level of ex post control and types of steering arrangements of government agencies in the two selected countries. More specifically, it answers the following research questions: (1) How do state-specific factors and (2) agency-level factors explain the similarities and differences of agency ex post control and steering in these countries?

   Agency control, which is our first dependent variable, is defined as a process that enables political principals to ensure that agencies pursue set objectives. Ex post control refers to the assessment of objectives' achievement and provision of performance related rewards/sanctions. Agency steering, which is our second dependent variable, refers to instruments used by political principals or agencies' CEOs to guide day-today activities of an agency.

   We focus on two groups of external (multi-annual planning and public reporting) and internal (quality standards and process restructuring) steering instruments.

   We define agencies as organizations based on public law, with a distinct legal personality, structurally disaggregated from the core government, but placed under the executive's control (Pollitt, 2004a; Verhoest et al., 2010). These agencies are engaged in service delivery, regulation and control, other policy execution, and advice to policy-making. Both formal (institutionalized) and informal mechanisms of agency control and (external and internal) steering, which are exercised by political principals, third parties, and agency management, are examined in this article.

   Our research is based on desk research and statistical analysis of the COBRA survey (3) data measuring the perceptions of agency managers from the joint COBRA database (COBRA (CRIPO) network 2011). We combine two levels of analysis (a country level and an organization level) in order to provide a balanced understanding of agency management in the CEE region.

   The article is divided into several sections. First, we outline our theoretical framework and hypotheses for the analysis of agency control and steering in the CEE region. The second section presents methodology and the main outputs of descriptive statistics and correlations, while the third section provides the empirical findings of our comparison. Finally, the article concludes by summarizing the main empirical results of our comparative analysis, as well as discussing their empirical implications and offering ideas for future research.

2. Theoretical framework

   In order to explain similarities and differences in ex post control and steering of agencies in the selected countries, we adopted two explanatory approaches: the transformative approach at country level and the NPM model at agency level.

   2.1. State-level factors

   We use the transformative approach (Christensen and Laegreid, 2001) as an overarching theoretical framework for state-level explanations. It argues that international pressures on agency management (like the NPM model or specific requirements of EU membership) are transformed by different state-specific factors or actor constellations (Verhoest et al., 2010, pp. 41-42).

   Two major trends affected agency management in the post-communist countries in the 1990s and 2000s. First, the transition from communism to democracy and a market economy shaped state institutions (including government agencies) and their relationships with other actors. Second, the accession process allowed EU institutions to influence public management reforms in the CEE countries through a combination of hard governance instruments (like the conditionality of the EU membership) and soft mechanisms (such as 'naming and shaming' or sharing of good practices).

   It is not surprising that state reforms and the EU accession were identified as the two main drivers of agency change in CEE, including Lithuania (Nakrosis and Budraitis, 2012). Nevertheless, the impact of EU accession and state reforms was not uniform. Although in the preaccession period all EU candidate countries implemented very similar requirements (e.g. professionalization of the civil services and the establishment of government agencies for the application of specific EU rules), previous studies of public management changes found somewhat divergent trends of de-politicization (Meyer-Sahling, 2011) and performance management (Hammerschmid and Loffler, 2015). Different political-administrative factors (types of administrative systems or different reform agendas) could have shaped varied domestic responses to the external pressures of Europeanization.

   During the processes of transition and EU accession, as well as the recent financial crisis, some CEE governments aimed to make their administrations more result- and customer-oriented through the introduction of various NPM solutions (such as business/strategic planning, quality management, or performance appraisal). Although the post-communist countries shared the main objectives of reforms, their trajectories and initiatives were quite different due to different reform priorities. Whereas the Baltic countries (including Lithuania) introduced more extensive systems of performance management, other CEE countries (including Romania) were more active in applying quality management tools, such as the Common Assessment Framework (CAF) (Bouckaert, Nakrosis and Nemec, 2011).

   According to Pollitt and Bouckaert (2011), results of public management reforms depend not only on the content of reform packages but also on their implementation process. Differences in reform implementation affect actual changes at agency level. Public management reforms, including the ones inspired by the NPM doctrine, are seldom carried out in a comprehensive way (Bouckaert, Nakrosis and Nemec, 2011). Lack of implementation capacity, which is one of the key factors affecting success of NPM reforms in post-communist countries, or resistance from various interest groups (including agency management, staff or civil service unions) can constrain achievement of the intended results.

   Furthermore, the style of reform implementation can matter to reform delivery at agency level. On the one hand, centralized or top-down reforms set mandatory requirements that should be met by all agencies in the public administration system. On the other hand, decentralized or bottom-up reforms define overall goals whose implementation remains largely at the discretion of agency managers.

   The level of centralization could be particularly important for the CEE countries characterized by the Rechtsstaat tradition and insufficient implementation capabilities (e.g. Neshkova and Kostadinova, 2012). Strict following of legislation and procedural regulations makes the execution of top-down reforms more likely. In contrast, delivery of more...