The Economic Role of Public Administration in Central Asia (1)
The Economic Role of Public Administration in Central Asia: Decentralization and Hybrid Political Regime
（The paper was written within the framework of the project “Emerging Market Economies in Central Asia: The Role of Institutional Complementarities in Reform Process”.）
By Alexander Libman, CDSEMannheim, IE RASMoscowand ECNU Shanghai, July 2008
The aim of the paper is to understand how the organization of public administration inCentral Asiashapes the results of economic development in the region. It discusses the main factors of bad quality of public administration in the region, paying particular attention to the link between political regimes and public administration. Moreover, it provides an overview of decentralization and devolution of power in Central Asian countries as one of the main channels of transformation of administration. The paper covers both formal decentralization and informal distribution of power between levels of government.
1. Introduction
While the first generation of economic reforms focused mainly on development of the framework for markets, including privatization and liberalization, the current debated is centered around the “second transition” reforms, including the organization of governance at different level: within individual businesses (corporate governance), on markets (competition policy) and in the state apparatus (public administration), able to make markets work. However, reforms in this sphere turned out even more problematic, than the establishment of basic market institutions. This paper focuses on the quality of public administration in the countries ofCentral Asia. In particular, it looks at four post-Soviet republics:Kazakhstan,KyrgyzRepublic,TajikistanandUzbekistan. Most of them carried out at least certain forms of first-generation reforms (though in Uzbekistan and to a certain extend in Tajikistan the deficits even in this field are severe), and all of them face the problem of development of efficient public administration. Moreover, all four countries are to a certain extend similar to each other from the point of view of political regime. The fifth country of the region – Turkmenistan – differs substantially from the rest of the region, both because of lack of even elementary market reforms and strict authoritarian political system and is therefore considered only briefly.
The paper starts with surveying the main indicators of economic effects of public administration. Moreover, it focuses on different approaches to explain the problems of public administration. It pays particular attention to the connection between public administration and political regimes. Most countries of Central Asia provide good examples of negative effects of hybrid regimes on public administration. Finally, a significant part of the paper is devoted to the problem of decentralization. One of the often neglected problems of the transition studies is to correctly define the object of studies: it is hardly sufficient to consider the national level of political decision making and economic reforms. Both international economic and political ties (Ananyin, 2005: 129-131) and subnational administrative and market structures are transformed. Decentralization and devolution of power to selfgovernance institutions is often considered to be one of the key elements of successful transformation of public administration in transition countries. The problem is, however, that the de-jure devolution covers only the top of the iceberg: an even more important feature is the development of informal relations between governments and regional elites. Informality of decentralization and specifics of political system make its effects for the development of public administration (and thus economic institutions) problematic.
2. Quality of public administration in central Asia
The countries of the region have been known for a notoriously low quality of public administration. The enforcement of the rule of law in Central Asia is bad; moreover, the bureaucracy in Central Asian countries is not only unable to protect the property rights, but is itself a source of major disturbance for the private enterprises. The Doing Business 2008 survey of the World Bank demonstrates the persistence of administrative barriers in Central Asia.The performance of the countries summarized in Figure 1 demonstrates significant problems of state administration; in particular, the excessive licensing, barriers to international trade, and, in most countries, excessive and arbitrary taxation, seem to be an important obstacle for business in Central Asia. Although Turkmenistan is not covered by the Doing Business survey, it is safe to say that in this country the absence of elementary human rights implies the absence of the rule of law (BenYishay and Betancourt, 2008). Moreover, in spite of several governmental reforms in the public administration described in the next section, business community still perceives significant lack of commitment of the governments to remove administrative barriers (Suhir and Kovach, 2003).
One of the main problems for regional economies from the point of view of protective state is a very high level of corruption. Corruption is endemic in all countries of Central Asia, covering all spheres of economic activity. Currently it seems to be routinized in the governmental hierarchies (Osipian, 2007). It is claimed that in several cases public agencies are taken over by private interests and criminal activities (Karklins, 2002); in the majority of cases corruption is the “connecting material” for the nexus of political, bureaucratic and business actors playing the crucial role in the economies of the region, where public officials seem to play the leading role. The “control of corruption” indicator of the World Bank governance studies indeed indicates an extremely high perceived level of corruption in Central Asian economies (see Figure 2). The only country with positive dynamics is Tajikistan recovering from the civil war; other countries either stagnate or worsen their corruption environment.
Swartz et al. (2008) apply the data of the BEEPS survey (2002) of the World Bank to compare the level of perceived corruption in the countries of the region and conclude that large countries (Uzbekistan and Kazakhstan) are perceived to be less corrupt, than smaller ones (Kyrgyz Republic and Tajikistan) and that there is no clear distinction between Uzbekistan and Kazakhstan: both countries are perceived to be less corrupt in different dimensions. Tajikistan seems to be the most corrupt country in the region. One should, however, notice that the firm-level corruption survey of BEEPS provides a much better evaluation of corruption in Uzbekistan (as well as Belarus) than other indicators, mostly because BEEPS provides a much better evidence of administrative corruption (i.e. bribes paid by businesses to public officials), which are obviously less important in Uzbekistan due to predominance of the public sector (Knack, 2006). Moreover, repressive political system limits the access to information. To provide just one example of administrative interventions, in 2002 the presidential decree made re-nationalization of any enterprise possible, if it has changed the “main direction of its activity” (Olcott, 2005: 156-157). The new round of the BEEPS survey (2005) indicates negative dynamics of Kyrgyz Republic from the point of view of corruption.
3. Determinants of quality of the public administration
3.1. Incentives and resources
The deficits of the public administration in the countries of Central Asia could be related to two parameters. On the one hand, they may be caused by the structure of incentives governing the relations in the public sector, causing inefficient patterns of behavior. On the other hand, the main factor of bureaucratic inefficiencies may be the deficit of resources available for the public administration, which is therefore unable to execute its functions. Both factors are interrelated; accumulation of resources for public administration depends upon the size of the shadow economy, which, in turn, is crucially determined by the costbenefit relations of the legalization decision; and bureaucrats adjust their behavior to the available resources. From the “incentives” point of view, the poor quality of bureaucracy in Central Asia may be related to informal institutions of the society and political regimes. The “resources” point of view suggests that the low quality of public administration depends upon the institutional capacity of the state, which, in turn, may be influenced by the political instability and economic development.
The informal institutions point of view on bureaucratic inefficiency (as well as on failures of transition in general) has been quite popular in the literature so far (Leipold, 2006). It suggests that the behavior of bureaucrats and their “customers” in the private sector is governed by the long-lasting informal rules, inherited from the Soviet or even pre-Soviet past. Basically, these rules could refer to two areas of activities. First, they determine the omnipresence of corruption in the countries of Central Asia as a traditional instrument of “problem-solving” accepted by both public officials and population (Taskanov, 2001). Unlike the “European” corruption, i.e. informal payments for “excessive” services or “privileged” treatment, the “Asian” corruption comprises informal payments for any services of the government agencies. Central Asia seems to have been particularly corrupt even in the Soviet Union. Second, the relations within the governmental apparatus are governed not by formal rules and hierarchies, but by informal power balance between individual clans. It is necessary to note that informal “gift exchange” in bureaucracy exists in various countries and may increase the efficiency of public sector as well (Dodlova, Yudkevich, 2006); Chavance (2008:67) points out the problems related with the concept of transparency which is key for criticism against informal relations in bureaucracies. In heterogeneous countries informal structures, can however become an important source of unequal treatment in the law implementation, which is sometimes considered the cornerstone of an efficient governance system (Rothstein, Teorell, 2005). It seems to be true for the Central Asian region, where informal relations connect particular groups of bureaucrats seem to have strong connections in both legal and illegal sectors of economy (Marat, 2006).
The origin of these interests and identities separating the bureaucratic corps can differ: the literature refers to the traditional tribal interests, as well as the Soviet legacy (Starr, 2006; Jones Luong, 2001) and can broadly be related as the system of “clans” dominating the Central Asian bureaucracies, as well as politics and economies. The main question is, however, the degree of persistence of these institutions, especially in the bureaucracy. First, the informal practices are not inflexible: Moskovskaya (2005) provides an overview of changes occurring on the microlevel in Central Asian societies. Second, Central Asian governments (in particular, in Kazakhstan and Uzbekistan) invested heavily in the international training of high-ranked bureaucrats (they paid at least more attention to it than Russia, still heavily influenced by the “delusions of grandeur” of its own educational system); in fact, the absence of traditional Soviet schools of thought is sometimes considered one of the reasons for successful reforms in some areas in Kazakhstan – in particular, in the banking regulation (Marchenko, 2005). It can also have at least a certain impact on the functioning of bureaucracies in these countries. Perlman and Gleason (2007) show that divergence of reform paths in Kazakhstan and Uzbekistan is influenced rather by policy choice than by institutional values. Although one can hardly claim that traditional clans play no role in the administrations of Central Asian countries, it is probably necessary not to overestimate their importance as opposed to the alternative “incentive-based” and also “resource-based” explanations.
The second source of incentives for bureaucrats can therefore be related to the current organization of the public administration (including both formal and informal incentives). It is also obviously a result of a path-dependent process, but at least may be more flexible than the rigid informal institutions and practices. First, there are still important deficits in the organization of bureaucracies in Central Asia, which have been developed for a relatively short period of time. Alongside with deficits in the legal system (Kangas, 2004), the deficits in the formal structure of public management may have detrimental effect on incentives for officials. Some countries of the region invest heavily in the development of bureaucracy; it is true particularly for Kazakhstan, which implemented a series of administrative reforms prior to other CIS states (with the only exception of Russia). In 1998 it established a special Public Service Agency to co-ordinate the transformation of the public administration. In 2000 the political positions were separated from administrative ones, which are protected by the irremovability rule included in the law (Mahmutova, 2001). The most recent initiatives include: (1) administrative directives (reglament) for each public service, specifying the exact list of actions to be carried out by a public official in a particular situation; the basic standards were passed in Summer 2007 (the first country in the CIS) and are monitored by the Disciplinary Councils of the Public Service Agency; (2) outcome-oriented budgeting in public service; the Concept of Public Planning, passed by the government of Kazakhstan in December 2007, suggests that the outcome-oriented budgeting and planning should be implemented at all levels of public service and for all public programs; (3) strategic long-term planning for all public agencies; (4) appointment of special representatives of the president in all public agencies. Nevertheless, the ambitious plans of administrative reform in Kazakhstan face the traditional problems of post-Soviet countries: they fail to have a real impact on the behavior of public officials, are often inconsequent and hardly implemented (Alternativa, 2007). The deficit of real public control over bureaucracies either makes the implementation of any reform problematic, or creates an additional source of revenue-extraction for public officials.
However, at least from the point of view of formal reforms Kazakhstan outperforms other countries, which failed to develop even the basic rules for public service. It is true especially for Turkmenistan, but also to a certain extend for Uzbekistan. Perlman and Gleason (2005) claim that administrative reform in Central Asia is to a great extend an outcome of external pressure from international organizations, than of internal political shifts. Their analysis of administrative reform in Kazakhstan and Uzbekistan suggests the existence of two main strategies: while the former focuses on the “stationary bandit” strategy, investing into improving the bureaucracy and thus enhancing economic growth to extract more rents, while Uzbekistan carried out reforms aiming to direct resources to the ruling elite regardless of economic consequences.
The other side of the discussion is represented by the “resource” arguments. The question is whether relatively poor Central Asian countries are able to create an efficient public sector at all. In fact, there is a certain relation between resource endowment and quality of public services. For example, in Kazakhstan the main reason for popular dissatisfaction with the services of Ministry of Justice is the too long waiting time; it, however, seems to crucially depend on the number of requests to be served by public officials and hence by the size of bureaucratic staff (Turisbekov et al., 2007). The problem of resource constraints is particularly acute for smaller countries like Kyrgyz Republic and Tajikistan, which indeed seem to face a resource constraint in their economic development, rather than for rich oil exporting countries like Kazakhstan, Uzbekistan and Turkmenistan. Treisman (2003) shows, that in the post-Communist world corruption is well predicted by the initial level of economic development at the start of the transition. These factors may also be at work in Central Asia.
However, resource constraint is not limited to purely economic component: it may also reflect the absence of qualified personnel for public agencies, and in this case influence all countries of the region regardless of their oil and gas wealth. The problem is that partial replacement of bureaucracies with newly appointed (and even internationally trained) staff is often insufficient to change the behavioral patterns; new bureaucrats, in spite of their international experience, are likely to quickly adapt to existing structures. A solution implemented in some countries of the CIS is to dissolve the complete agency and to construct a new one from scratch: however, the only case of relative success is Georgia, which was able to replace the old highly corrupt traffic police (it is may be one of the most corrupt institutions in the post-Soviet world; and it is perceived as the most corrupt structure by the population of Kazakhstan, see Turisbekov et al., 2007) with a new efficient service, but in this case international help and small size of the country seem to be very important. Large countries like Kazakhstan and Uzbekistan are hardly able to pursue this strategy.
On the other hand, natural resource wealth of Uzbekistan, Kazakhstan and Turkmenistan may however have even a negative impact on the development of public administration. One of the effects of the “resource curse” documented in the literature is the development of the so-called “petro-state”, when the revenue from natural resources allows the governments to delay the development of public bureaucracies and reduces pressure from their inefficiencies (Alayli, 2005). The problem is indeed present in Central Asia (Sabonis- Helf, 2004). The point is, however, that the resource curse seems to be linked with the specifics of the political system. The impact of politics on the development of public bureaucracy is to be examined in the next section.
An additional resource constraint may be the weak state capacity. It has been demonstrated empirically, that the state weakness has a clear negative effect on corruption control (Shen and Williamson, 2005). In Tajikistan in the early 1990s the civil war environment made any development of efficient (or even somehow functioning) public administration impossible. Current political turbulences in Kyrgyz Republic also have a negative impact on development of public administration.
3.2. Political regimes and public administration
One of the most interesting questions from the point of view of success of administrative reform is probably how it is related to political regime. The relation between bureaucrats and politicians is crucial for understanding the existence of predatory state or market enhancing policies in emerging economies (Dixit, 2006). Basically, there are two main approaches in the development studies. The first tradition claims that an efficient bureaucracy is an outcome of democratization (though, probably, not an immediate and direct one), and therefore, a non-democratic country is hardly able to create a well-functioning public sector. The most studied aspect of the interaction between political regimes and quality of administration is the relation between democratization and corruption. The argumentation in the literature, however, mostly deals with corrupt politicians; and does not explicitly consider the link between politicians and bureaucrats. The assumption is that bureaucrats immediately respond to changes of incentives from (honest or dishonest) politicians; it is obviously a questionable statement. One of the key arguments in favor of lower corruption in democracies – higher competition and system of checks and balances (Johnson, 2000) – does not work for bureaucracies. Nevertheless, democracy still may reduce corruption of bureaucrats by empowering the population, which may than take direct actions forcing the governments to change regulations to lower corruption. Thus democracy actually becomes a system of mutual learning of population, politicians (and, unfortunately, corrupt bureaucrats searching for new options for rent-seeking as well), may be resulting in reduction of corruption. Moreover, free press acts as an additional instrument able to shed light on corrupt behavior and therefore increase accountability (Adsera et al., 2004). The problem is, of course, that democracy may actually fail to ensure higher accountability as well. For example, even if citizens are empowered with the right to fight corruption, they are likely to face a collective action problem while using it (Philp, 2001). Moreover, in a democracy corrupt bureaucrats may be better protected from the pressure of the politicians, unable to resolve to direct repressions. Nevertheless, there is a wide literature showing the negative relation between democracy and corruption (for a survey see Seldadyo and de Haan, 2006; Seldadyo, 2008); although in the post-Communist world there seems to be no significant relation between democracy and corruption (Treisman, 2003), the discussed specifics of corruption indicators for Uzbekistan and Turkmenistan may influence the results.
The effects of the petro-state may also depend on the level of democratic development. Damania and Bulte (2003) and Robinson et al. (2006) show that the resource curse is present particularly in regimes with low political accountability. Durnev and Guriev (2007) examine one of the possible channels of this interaction, looking at the corporate transparency developing as consequence of potential expropriation threat. Another channel is discussed by Egorov et al. (2007): non-democracies in resource-abundant countries tend to repress media, since they face lower need of good information, and therefore indirectly reduce the accountability of bureaucracy. That is why democracy may become a necessary precondition for exploiting the resource wealth to remove the resource-driven limitations for public administration reforms. Otherwise resources corrupt politics, as well as public sector and bureaucracy, leading to a self-sustaining inefficient equilibrium.
From this point of view the non-democratic nature of political systems of Central Asian countries may at least contribute to the deficits of public administration. The alternative tradition, however, suggests that exactly the ability to establish good bureaucracy is one of the key factors allowing some non-democracies to demonstrate extraordinary growth rates. Jamali et al. (2007) distinguish between autocracies and bureaucracies as two types of nondemocratic regimes. Bureaucracies, or dictatorships which codify and announce law, do not differ from democracies in terms of economic performance. It makes a “roundabout” way to economic prosperity possible: instead of developing democratic institutions and trying to achieve economic growth, the country can focus on economic and administrative governance (e.g. good administration) remaining non-democratic, and thus achieve high growth rates (Ahrens, 2008). Indeed, there have been numerous examples of non-democracies able to create large and relatively efficient bureaucratic machinery, sustaining their existence for centuries. The main advantage of non-democracy is the ability of politics to control bureaucracy by force and execute direct pressure over public agencies. This argument is as questionable as the “competition” and “social control” arguments for democracy: there may be significant information asymmetries, influencing both control of corruption and selection of bureaucrats, and non-democracies may experience problems with information transmission through the levels of hierarchy (the “yes man” problem described by Prendergast, 1993). Moreover, the power logic may also influence the quality of bureaucracy through appointment mechanisms. Egorov and Sonin (2006) show that autocrats appoint incompetent officials, and Dodlova (2006) claims that low political accountability leads to expansion of de-facto authority towards appointed bureaucrats from politicians. Historical experience demonstrates large differences in economic performance and quality of governance among non-democracies (Besley and Kudamatsu, 2007), hence, it is reasonable to assume, that the ability to develop good public administration and economic governance as a substitute for democracy is characteristic for particular types of non- and semi-democratic regimes. Therefore the main question for this paper is whether the specific type of semi-democracy existing in Central Asia is capable of developing good public management institutions. Although it failed to manage it in the past, it may still be possible in the future.
The problem for the Central Asian countries, however, is that most of them (especially Kazakhstan, Tajikistan and Kyrgyz Republic, but also to a certain extend in Uzbekistan) belong to the group of hybrid regimes. The non-democratic governments enforce their rule, being limited in their application of direct force. It is possible to claim, that the hybrids combine disadvantages of both democracies and strong non-democracies from the point of view of bureaucratic development. First, being unable to use force, or coercive power, the politics (or, more specifically, dominating clans) are likely to use corruption as one of the instruments of control. Corruption, not repressions, becomes “the blood of the regime” (Mau et al., 2007:100). Nevertheless, hybrids are often characterized by poor control over individual agencies, able to extract rents, and extraordinary shift of real authority to bureaucrats: both instruments – coercion and public control – do not function. Second, inefficient bureaucracy may be used as an additional “instrument of control” over private actors. If the government uses force as instrument of dominance, it needs an efficient bureaucracy to implement the repressions; if it is, however, unable to use force, the very inefficiency of bureaucracy becomes a mighty weapon. Individuals and firms have to pay bribes and disobey law in order to exist (and cannot survive otherwise); however, that is why they can be “taken hostage” by the government, able to legally prosecute them in case of any conflict (Libman, 2008). Kazakhstan indeed provided examples of this policy of “hostagetaking” in the first half of the 2000s (Libman, 2006). There exists a growing literature demonstrating a non-linear quadratic or even cubicrelation between corruption and democracy: in this case low level of democracy leads to a eruption of corruption, going down in higher developed democracies (Montinola and Jackman, 2002; Sung, 2004; Rock, 2007). The relation may, however, be even more sophisticated (Treisman, 2007a). Finally, the state failure, which has been identified as one of the reasons for the low quality of autocracies, is far more often in hybrids than in pure democracies and autocracies (Bates, 2008).
The problem of hybrid political regimes is obviously absent in Turkmenistan. However, until recently the repressive regime seemed to be ideologically directed towards reducing public services. Under president Niyazov public educational system was reduced, pensions were abolished and public libraries and hospitals were partly liquidated. The new president seems to move away from extremes of the old regime, but any assessment is still premature.