Appendix I: Case Discussions

Worries of corruption were much more pronounced than in Estonia. Whether this inhibited effective implementation of opposing policies is a closely related but fundamentally different question. It is distinctive that corruption and the problems of a lack of rule of law pointed out mostly stem from ‘political corruption’ – bribes and misusage of public resources by politicians and political parties – rather than ‘administrative corruption’ (see e.g. BTI 2006: 3). Tellingly, Nørgaard and Hansen with Ostrovska and Møller (2000: 39-40) report that ministers generally strongly trusted their subordinate officials in performing their duties impartially and effectively while trust upwards – from lower-levels servants to their ministers – was equally weak. This testifies to the possible prevalence of political but not administrative corruption. Implementation was deemed weak in one BTI-report (BTI 2003: 3-4) but as a general trait, I maintain the existence of basic responsiveness throughout the period.
LESOTHO (2002-)

The administration was ineffective throughout the period. In 1944, a treasury was established in Lesotho marking the first administrative institution. After independence, governance was centralized. Under democracy, a two-tier governance system existed with a central administrative level and a local one led by chiefs (Daemane 2011: 134, 136, 168). Despite a contested authority, the state still extended throughout the territory via the chieftainships (Coplan and Quinlan 1997: 47; BTI 2010: 5).

The ruling party controlled access to the state, including state jobs. The state had virtually no autonomy (Kabemba 2003: 16). Similarly, judicial independence was weak (African Development Bank 2006: 27). At the most central levels of governance, politicization may have caused a rudimentary sort of responsiveness but below this level, lack of training, resources, and incentives caused widespread divergence from political goals (BTI 2010: 5; Daemane 2011: iii; Mwangi 2016: 209). Local public service delivery suffered under a lack of institutional capacity and budget controls (African Development Bank 2006: 27; BTI 2010: 5).
LIBERIA (2006-)

The administration was ineffective throughout the period. The state infrastructure was completely destroyed after the civil war. Through the 2000s, there were no functioning public utilities beyond Monrovia (BTI 2003: 4; McFate 2013: 25-26). This was highly surprising given the early and independent process of state-building in the 19^th century, resembling many European countries, but the country remained a bunch of autonomous communities (Gerdes 2013: 14, 19).

The legacy of non-meritocracy was strong. Samuel Doe institutionalized an ethno-exclusive state apparatus in which the 5 % large Krahns were given most state jobs (Adebajo 2002: 45-46; Gerdes 2013: 34, 39). Charles Taylor continued this ethnic distribution of state jobs from 1997 to 2003 (McDonough 2008: 365). More generally, elections before and under democracy produced winners among former officials (Reno 2008), indicating the continuance of politicization in recruitment.

With the failure of the state came a widespread deinstitutionalizing and a breach of all morale and legal bonds on civil servants (see Pham 2004: 1). Corruption levels were high, public service delivery quality low, and fiscal controls weak (Reno 2008). Officials sought personal gains through public means (BTI 2006: 5). No responsible or accountable judicial sector existed (Gompert et al. 2007: 71).

LITHUANIA (1992-)

The administration was effective from 2000. Despite occupying some areas of the Lithuanian soil until 1994, there are no indications in the literature that Russian security forces hindered the penetration of administrative decrees from Vilnius or occupied public buildings although there were a few ransacks of ministries in 1990 (see Kasekamp 2010: 169). Otherwise, the same qualification of Soviet penetrative legacy applies to Lithuania as to Estonia and Latvia (see also BTI 2003: 3).

Establishing meritocracy was a struggle throughout the 1990s. In the judiciary, reforms until 1991 only focused on issues of improving processing but not the hiring and status of judges. From 1992, the Constitutional Court, however, ensured the independence of the courts despite attempts to politicize (Raulickyte 2001: 182-188). The rest of the civil service was more problematic. A Civil Service Law was passed in April 1995 and since then Lithuania was actually a frontrunner in terms of formalizing civil service regulations in numerous reforms and legislative amendments in the late 1990s (Vanagunas 1999: 213; Meyer-Sahling 2009: 7). Urbanovic and Garcia-Zamor (2011: 183) in fact believes that the 1995 law was sufficient for creating meritocracy also at the behavioral level but I stick to the interpretation of Palidauskaite, Pevkur, and Reinholde (2010: 47-49) who assert us that meritocracy was gradually implemented from 1995 to 1999. In support of this, Pivoras (2008: 123) contends that a radical depoliticization only occurred in the 21^st century, and Meyer-Sahling (2009: 19) points to the civil service act in 1999 as the consolidating law.

Regarding responsiveness, Lithuania was very much like Latvia. Corruption was mostly political, as indicated by a lack of trust from servant to minister but strong trust from minister to servant (Drengsgaard and Hansen with Brøgger 2004: 27; Velykis 2010: 2). Implementation problems were arguably more frequent than in Latvia but this could also be attributed to greater public openness and attention to governance problems (Drengsgaard and Hansen with Brøgger 2004: 33-34). In general, implementation problems were gradually reduced during the period, and given that the outset was a legacy of relatively law-governed administration (see Møller and Skaaning 2010: 338), I conclude that responsiveness existed from independence.

MACEDONIA (1991-)

The administration was ineffective throughout the period. Territorial penetration was problematic at least as long as there was successful armed resistance in the Albanian areas of the country. For instance, still in 2002 the western part of Macedonia was managed primarily by Albanians administrative structures, functioning along patron-client networks with the Albanian Democratic Party on top (BTI 2003: 4). The basic administrative structures that did exist only came to function on Macedonian premises from late 2003 when, as indicated, international police forces cooperating with newly strengthened Macedonian ones took control of the Albanian insurgency areas (see Ramet 2006: 582-583; Atanasov 2005: 169). This continued through the 2000s (Daskalovski 2009: 264).

Macedonian administration is conceptualized as Eastern Oriental by Møller and Skaaning (2010: 334) – that is, on a par with Bosnia and Albanian and some of the most malfunctioning administrative systems in Europe. The literature is unambiguous in describing the administration as highly politicized, marked by the power politics between parties over state control (e.g. Ramet 2006: 571-572; Meyer-Sahling 2012: 7; Cierco 2013: 483-484; Verheijen and Rabrenovic 2015: 19). The constitution of 1991 was explicit on the power of deputies to elect judges and servants (Rossos 2008: 267). Judges in the 1990s were far from independent of political selection in terms of appointment and promotion, functioning to the benefit of parties or, most often, presidents and government parties (Ramet 2006: 571-572). This continued in the 2000s (Boduczynski 2010: 143). This trend was only slightly mediated by reform attempts in 2000 setting the formal requirements of meritocracy (Cierco 2013: 484), some more notable improvements in 2005 (Daskalovski 2009: 266), and the otherwise very ambitious program for administrative overhaul and reform in 2010 installing a civil servants agency and an administrative agency for monitoring hiring processes (Meyer-Sahling 2012: 19; Cierco 2013: 484).

Just as with politicization, the Eastern Oriental tradition also left a legacy of excessive corruption (Møller and Skaaning 2010: 334; see also Cierco 2013: 488). As Boduczynski (2010: 143), “The judiciary was barely reformed in the 1990s. It was slow and inefficient, and in 1997 research found that eighteen of thirty courts were not performing their duties.” Such painstaking assessments are common among studies of corruption in Macedonia and continue to signify the administration through the 2000s. Cierco (2013: 485) reported on bureaucratic inertia and slack at very high levels as shown in numerous complaints of the Ombudsman. But such state administration corruption also was prevalent at local and central levels of the civil service, including notably the ministry of the interior, and has been described as a ‘huge problem’ (Ramet 2006: 571; see also Daskalovski 2009: 267; Irwin 2010: 345). As in other Balkan countries, we thus see the mix of corruption and unresponsiveness.

MADAGASCAR (1993-2009)

The administration was ineffective throughout the period. The state never penetrated the vast island territory fully. The lack of infrastructural presence and power was daunting, even deteriorating in the 2000s hampering public service delivery as well as separating the capital from the countryside (Moser 2008: 140; Bellemare and Moser 2012).

Throughout the history of the country, neopatrimonialism had been a firm trait. The President’s office at the same time distributing resources via state offices in arbitrary fashion while it was abused by powerful elites in and outside formal politics via corrupt officials (Marcus 2004: 1). Thus, the neopatrimonialism of the system was truly detrimental for both meritocracy and responsiveness. Additionally, anti-corruption campaigns were ineffective while the executive continued exploiting offices to bias policies against the opposition through the 2000s (Ratsimbaharison 2016: 8).
MALAWI (1994-)

The administration was ineffective throughout the period. Local governance structures were in disarray because of waning resources and neopatrimonialism but they did exist throughout the territory (BTI 2006: 4; BTI 2008: 7).

Originally under Banda, the logic of hiring was based on distribution of state resources, patronage and clientelism. The Public Service Commission was exploited by top-level civil servants to hire their own kin. Attempted reforms from 1994 did not succeed because the multiple parties could not agree on terms for the installation of meaningful meritocracy (Adamolekun, Kulemeka, and Laleye 1997: 214; Anders 2002: 334-336). Patriominialism in hiring and firing continued after Banda (BTI 2006: 4; Hall-Matthews 2007: 78). In the late 2000s, all senior level positions continued to be filled at the will of the president (BTI 2012: 6). Despite some centralization, there were still clear political skirmishes in the hiring processes and cliques across government agencies because of separately hired cohords (Cammack and Kelsall 2011: 88). The only strong administrative organ ruled by meritocracy was the judiciary which remained fundamentally independent and capable of circumventing government decrees (Vondoepp 2005: 276-278).

The culture of fear of Banda and subsequent absolute loyalty to his will prevailed until 1994 but naturally disappeared after the introduction of multiparty elections (Anders 2002: 331). From then on, the Malawi bureaucracy looked much like the typical African one with inefficient and corrupt administrative practices (Anders 2002: 330). Segments of conflicting ethnicities among civil servants caused deficits in responsiveness. The presidency under Mutharika gave the most concerted effort at fighting corruption by purging old bureaucrats hired under Banda and Muluzi. Yet, the culture of corruption and organizational deficits in the coordination between agencies and local and central authorities continued (Hall-Matthews 2007: 79; Cammack and Kelsall 2011: 92; Eggen 2011: 313).

MALI (1992-)

The administration was ineffective throughout the period. Territorial penetration suffered the same fate as resource supremacy given that the weakening of the state implied a territorial delimitation. The north was cut loose and more generally, large parts of the population were not affected by any state infrastructural unit (BTI 2006: 3-4).

Regarding meritocracy, the ministries of finance and economics achieved relatively high degrees of autonomy from political management. However, they were islands of excellence in a sea of generally low autonomy with frequent political interferences (Bergamaschi 2007: 276). Since 1992 at least, the president distributed state jobs and thus circumvented meritocracy. State jobs went to opposition persons in rebel areas to quell resistance (Baldwin 2005).

Responsiveness was likely weak. Ministries not just lacked autonomy against politicians but moer generally against societal interests and interference (Bergamaschi 2007: 276). Tellingly, initiatives by donors of building local capacity and ownership for achieving better development results had limited impact (Bergamaschi 2007: 271). Also, regional big men were notoriously known for interfering in the command lines of the central administration (Bøås and Torheim 2013). The institutional capacity of the state, including the judiciary, was described as weak, covering a pervasive culture of corruption, ineffectiveness, and inefficiency (Smith 2001: 77; BTI 2003: 4).

MAURITIUS (1968-)

The administration was effective throughout the period. According to BTI (2006: 4), there was a functioning legal and administrative throughout the islands. The island of Rodrigues had a particular regional autonomy politically and economically (BTI 2010: 5). From independence, the island politicians overtook a small but capable and strong civil service modelled by the British and Creoles (Meisenhelder 1997: 281; Carroll and Carroll 1997: 468). This indicates that territorial penetration existed from independence.

From independence, Mauritius chose the Westminster system with a strong separation of the judicial, legislative, and executive powers (Kasenally 2011: 163). A group of educated British and Creole bureaucrats comprised the civil service from early on and managed the consolidation of a meritocratic system. They would not suffer competence for nativization of the bureaucracy and thus embarced the British administrative institutions including the building of a strong public administration school that ensured the creation of a professional civil service in the decades after (Carroll and Carroll 1997: 474-476). The state administration was in subsequent deacdes known for its developmentalism forged by state autonomy and technocratic nature (see Bräutigam 1997: 54) – a strong indication of meritocracy.

Responsiveness was also strong. The state was autonomous in recruitment matters and some agencies such as in the finance sector had considerable freedoms in policy-making. However, it was party alignments that determined economic policy (Bräutigam 1997: 54). The civil service was not only in tune with civil society and the complexity of ethnic groups and tribal societies. It also balanced a concern for the public interest with loyalty to the minister at hand (Carroll and Carroll 1997: 470-477). The country had a long tradition of local capacity building by hiring of civil servants locally (Bräutigam 2008: 146-147). While this strengthened decentralized management, it did not hamper control from the center as indicated by the overall success of the developmental project. The politicians did not fear the bureaucrats or local officials since they trusted them to act according to the nation’s interest as they would themselves (Meisenhelder 1997: 281).

MEXICO (2000-)

The administration was ineffective throughout the period. Although criminal organizations made some urban areas almost impossible for state officials to enter, it was not yet a fundamental problem of penetration in the 2000s (BTI 2003: 4-5). Administrative structures existed throughout (BTI 2012: 6).

Before democratization in 2000, civil service appointments were dominated by the Revolutionary Party. Just as Presidents were rotated, so were bureaucrats through patron-client ties (Grindle 1996: 49). Also, the judiciary was effectively subordinated to the party in all aspects (Domingo 2000). Politicization and nepotism continued after 2000. Judicial independence was only ensured at the federal level whereas regionally, judges were appointed by governors and by political and personal concerns (Beer 2006: 43; Sousa 2007: 92-93,97). In the remaining civil service of the 2000s, only a few agencies such as the tax administration had firm traditions of meritocracy (Echebarria and Cortazar 2007: 151). Otherwise, politicization was still extensive so as to make formal meritocracy virtually ineffectual (BTI 2010: 7; Espinosa and Moreno 2015). A reform in 2003 was supposed to install meritocracy at all three government levels but this never succeeded due to continuance of patrimonial culture and party influence. In 2006, only a few thousand of the total of 45000 civil servants were career civil servants as set out in the 2003 reform (Morris 2010: 153-155).

Family ties and capture by criminal gangs were widespread disturbances of civil service performance and responsiveness. Judicial efficiency was also hampered and law enforcement was described as inept and corrupted (Beer 2006: 44). Reforms in 2002 and 2003 sought to curb corruption through performance payment and other NPM measures but, despite being well-received among the press and population, corruption continued (Morris 2010: 137, 152).

MOLDOVA (1991-)

The administration was ineffective throughout the period. Territorial penetration was highly constrained throughout the period. The Transnistria region was basically completely outside the control of any administrative bodies of the Moldovan state. Transnistria was a coherent and strong political unit resisting the unitary state system of Moldova (Heintz 2008: 2-3). Even the region of Gagauz remained ambiguously governed by Moldova after the provision of special status in 1994. Here, local strongmen more than central Moldovan authorities conducted daily businesses (Bollerup and Christensen 1997: 216; see also BTI 2003: 4; BTI 2006: 4; BTI 2008: 6-7).

The Moldovan administration was highly inspired by Tsarist Russian traditions as it took up independence. Or rather, business as usual merely continued (Møller and Skaaning 2010: 335). The only two major reforms, in 1994, 1998, and 2003, did not alter the game of politics around hiring and firing civil servants but merely changed the organization of ministries and the balance of centralization-decentralization. EU and independence reports pointed extensive politicization from the top down through the bureaucracies in the ministries and in the judiciary through the 1990s and 2000s (Johannsen with Jensen and Møller 2004: 28, 31; Tudoroiu 2015).

Much the same story goes for the persistence of unresponsiveness. Stent (2007: 2) documented that the administration was rife with corruption and uncontrolled ties between political-administrative and business elites. This clientelist was deemed weak in the ability to implement even simple initiatives as policies were constantly successfully captured by cronies, particularly farm and factory managers (Johannsen with Jensen and Møller 2004: 15-16). Poor staff coordination and motivation undermined implementation, including of basic public services and tax extraction (Way 2002: 127; Johannsen with Jensen and Møller 2004: 33; BTI 2006: 4). Reforms in 2008 (see BTI 2010: 6) and changed political signals in 2009 (see Tudoroiu 2015: 660) were insufficient to change the basic character within the period of study.

MONGOLIA (1990-)

The administration was ineffective throughout the period. During communism, a modern administration was established at the central level. However, after communism the central authority weakened and reforms to further decentralization and thereby indirectly increase central control largely failed (Enkhbat and Odgaard 1996: 165; BTI 2003: 3). Although a basic infrastructure existed throughout the territory, it did not function everywhere (BTI 2006: 5). Roads and communications line were poorly maintained (Enkhbat and Odgaard 1996: 172), and many livelihoods still depended on pastoral animal husbandry, agriculture managed by nomads (BTI 2006: 5).

Although technocratic elites were mostly in charge of running the administration in the transition years, there were initially no reforms of the Nomenklatura system and thus politicization continued (Ginsburg 1999: 264-265). Reforms in 1991 was the first fragile attempt at defining the role and status of civil servants but high turnover rates at the election in 1992 proved that politicization was still dominating (Enkhbat and Odgaard 1996: 168-170; Damiran and Pratt 2008: 193). From 1997, reforms sponsored by the Asian Development Bank were more successful. Parts of the ministerial bureaucracy implemented systems of competitive exams. Judicial independence was in fact achieved from 1998 on the basis of the same establishment of human resource management offices (Damiran and Pratt 2008: 193-194; Gramckow and Allen 2011: 1-2). Although some problems of representation of the executive powers among judges and politicization persisted (Brøgger, Nørgaard, and Cummings 2004: 30), meritocracy via competitive examinations were basically ingrained in the bureaucracy and judiciary from 2003 when examinations became organized and made publically available (Damiran and Pratt 2008: 197). Here, I settle with the subjective accounts of Brøgger, Nørgaard, and Cummings (2004: 30) and Damiran and Pratt (2008: 199) that note meritocracy as supported by a majority of ministers and ingrained in the culture of governance of Mongols.

Basic responsiveness was ruined as decentralization failed in the 1990s. In addition, professional and able servants increasingly shifted to the private sector due to insufficient salary levels. Generally, there was a lack of coordination and resources for effective implementation and extraction of resources from local communities (Enkhbat and Odgaard 1996: 171; BTI 2008: 6; Damiran and Pratt 2008: 197). The senior civil service and judiciary also remained highly corrupt. Anti-corruption laws in 1996 and 2006 were more cosmetic than effective (Fritz 2007: 200).

MONTENEGRO (2006-)

The administration was ineffective throughout the period. Montenegro had a basic administrative structure throughout the territory as it was free from the territorial disputes of Serbia (BTI 2008: 6). It is noted that when Milosevic stepped down in 2000, Montenegro already had much of the infrastructure of a functioning state (Morrison 2009: 182). As in Serbia, politicization, local-level patronage, and corruption plagued the administration and judiciary despite a few attempted reforms (Trivunovic, Devine, and Mathisen 2007: 5-8; Meyer-Sahling 2012: 34-35; Verheijen and Rabrenovic 2015: 19). Not least, this has been noted in EU reports of 2005 and 2008 upon assessing the potential for future EU membership (Boduczynski 2010: 232). Thus, both the degree of responsiveness and meritocracy was weak.

MOZAMBIQUE (1994-2004)

The administration was ineffective throughout the period. The Portuguese colonial administrative structures were upheld at independence and further strengthened by communist centralism under Frelimo (BTI 2003: 5). Apart from territorial penetration, however, the state apparatus under democracy was in deep crisis. Meritocracy was never close to being implemented. The Portuguese legacy of patrimonialism continued but took a turn towards excessive politicization to align the state with the socialist ambitions of Frelimo. The state equaled the party and vice versa (Lalá and Ostheimer 2003: 49; BTI 2003: 5). This lack of separation between party and state merely continued under democracy, although in a less systematic fashion (Braathen and Orre 2001; BTI 2006: 4). Not least, this was because the Frelimo party remained the most powerful party in Mozambiquan politics (Meneses and Santos 2009: 130). Clientelism and personalism in hiring and firing grew and, despite improvements from 1986, judicial independence was still not ensured (Braathen and Orre 2001: 203-204).

Unresponsiveness was widespread. The system of district administration functioned poorly. As the central state apparatus had little authority over local chiefs, they ruled for themselves and the district administrations were merely symbolic (BTI 2003: 5). Also, the administration was highly heterogenous as different cultures, notably a Muslim and an African traditional one, existed and interfered with one another (Meneses and Santos 2009: 130-131). The streamlining of the state under Frelimo hegemony was lost to a younger generation which lacked absolute loyalty and was unwilling to take much responsibility in the face of low salaries and harsh disciplinary measures. Efficiency and effectiveness suffered because of this (Lalá and Ostheimer 2003: 49). Most generally, the typical lack of separation between private and public spheres and thus excessive corruption existed in the administration and the judicial system (Braathen and Orre 2001: 203).
NEPAL (1991-)

The administration was ineffective throughout the period. Territorial penetration suffered because of the Maoist insurgency and a general lack of state infrastructure beyond the valley area (BTI 2006: 4-5). As indicated, even after the extension of state authority in 2006-2007 the southern flatlands close to the Indian border were not covered by the state (BTI 2010: 5).

Throughout the democratic period, judicial independence was not the norm. Despite some notable examples of autonomy in the Supreme Court, the 1990 constitution continued to give much power to the Prime Minister and King in appointments to the courts (Pimentel 2010: 290). This was the case for the rest of the bureaucracy was as well. As a consequence, the Hindu caste dominated the state offices infiltrated with party patronage appointments and some family nepotism (Jamil and Dangal 2009; Dahal 2010: 17; BTI 2010: 5).

According to BTI (2003: 4), the administration in the arly 2000s is “is weak, riddled with corruption and urgently in need of reform.” Sharma (2012) similarly documents excessive administrative corruption and fundamental unresponsiveness of the civil service to the civilian government at several occasions due to the continued loyalty to the king. The judiciary was also marked by widespread corruption (Pimentel 2010: 290). Personal connections and bribery still dominated the administration in the end of the 2000s making it highly ineffective (BTI 2012: 7).

NICARAGUA (1984-)

The administration was ineffective throughout the period. It is well-established in the literature that the state never reached all stretches of the territory. Despite notable investments in the 1980s (Ruccio 1988), infrastructure and roads remained limited in many areas (BTI 2012: 6). Two Atlantic coastal regions were autonomous from 1987 with little state presence (BTI 2003: 3).

Meritocracy was never achieved. Judicial independence was achieved de jure in the constitution of 1987. However, the law left small loopholes for the president to pack the Supreme Court. In practice, the norm was that the parliament and president each appointed their own preferred judges to secure a balance of power (Wise 1994; see also Sousa 2007: 107). In terms of the rest of the civil service, there was never in the period under study a professional civil service (see Miranda and Ratliff 1994: 290). Aleman and subsequent presidents practiced appointment of family members and party loyalties to public offices in a widespread fashion (Brown, Cloke, and Rocha 2007: 193). Echebarria and Cortazar (2007: 139) ranked the country at the bottom of the Latin American list of degree of meritocracy.

Responsiveness was similarly detrimental. Slow and inefficient implementation was the rule rather than the exception (see Sousa 2007: 198). The bureaucracy was described as dysfunctional and clientelist (Echebarria and Cortazar 2007: 141, 148). Not only was political corruption widespread, as exemplified by the massively personalist regime of Aleman, but the lack of professionalism and morale guidance of the bureaucracy nurtured corruption in all aspects of government. President Bolanos from 2001 to 2007 sought to purge corrupt officials but the civil service system as such remained highly corrupt due to its vested interests (Brown, Cloke, and Rocha 2007: 191-192).

NIGER (1993-1996; 1999-2009)

The administration was ineffective throughout both periods. The state, as indicated, lacked full extension of its authority to all peripheral regions and completely lost control over the north towards the end fo the 2000s (BTI 2003: 4; BTI 2010: 7).

Regarding meritocracy, Alou (2012) describes Niger as a neopatrimonial state. Given that the concept of neopatrimonialism is well-established as a description of patronage and clientelist appointments in the majority of African states, I find this sufficient to code Niger as non-meritocratic. Similarly, neopatrimonialism points towards unresponsiveness. Indeed, excessive corruption was the norm for the majority of the technocratic elite, according to Ibrahim (1994: 30). The administration was generally corrupt and inefficient (BTI 2003: 4; BTI 2006: 5).
PAKISTAN (1988-1999; 2008-)

The administration was ineffective throughout the period. Relying on all structures from the British colonial administration, territorial penetration was basically obtained (see BTI 2010: 5-6). The bureaucracy in the 1990s was “relentlessly politicized” (Diamond 2000: 93). Similarly, the judiciary was not independent but politicized. Despite the existence of a Public Service Commission from the 1930s, it lost all autonomy in the 1970s and caste and class never lost their dominance as conditions of recruitment (Jones 1997: 325, 331). This was contrasted by the development of the judiciary which attained greater independence from 2008 (Shah 2016: 89). On balance, meritocracy was thus never established.

Responsiveness was also weak. Pervasive corruption continuously hindered effective delivery of public goods. Despite the original promise of high morale as inspired by the British, bureaucratic corruption quickly increased and was said to have become endemic (Jones 1997: 337). Coordination between agencies was also weak which limited the effectiveness of government policies. The judiciary remained dishonest and usually deferred the wishes of the executive (McCartney 2011: 187).
PANAMA (1991-)

The administration was ineffective throughout the period. The drug war and related problems of border control with Colombia weakened territorial penetration. More generally, some state institutions did not reach all segments of the population (BTI 2006: 3).

Meritocracy was never achieved due to a legacy of clientelism and personalism which continued after 1991 as extensive party politicization. Bureaucrats were fired for demonstrating against the government, nepotism, and patronage appointments were widespread (Furlong 1993: 38). In comparative analyses of Latin American bureaucracies, judicial independence was assessed as low (Sousa 2007: 107), and the degree of meritocracy in the rest of the civil service was very low, close to 0 on a scale to 100 (Echebarria and Cortazar 2007: 139).

The nepotism and patronage also regarded an interference of societal forces in the affairs and job security of the bureaucrats, outside the influence of the executive power (Furlong 1993: 38). Additionally, low salaries contributed to a lack of functional capacity, inefficiency, and corruption in the civil service and the judiciary (BTI 2006: 3; Echebarria and Cortazar 2007: 141, 147).

PAPUA NEW GUINEA (1975-)

The administration was ineffective throughout the period. As indicated, the Papuan state only reached 10 % of the arable land and 40 % of the population (Dinnen 1999: 282). Coverage was poorest in inland areas of the big island which colonial settlers had left untouched while residing primarily in coastal areas and on small islands (Dinnen 2001: 17; BTI 2003: 3). Qualitatively speaking, infrastructure, roads and communications, was either very poorly built or wholy absent in rural sites. Modern bureaucratization only began by a UN mission in 1962. Thus, the presence of administrative structures remained scarce and unevenly spread across the territory (Dinnen 2001: 30, 185).

Politicization of the bureaucracy was widespread from early on and pervaded all levels. There was even an expression, Wankotism, marking the widespread use of appointing servants due to family relations (Payani 2000: 135-140). Similarly, Dinnen (1999: 14, 191) asserts us that patrimonialism dominated the bureaucracy with traditional Big Man politics causing highly personalized relations between politicians and bureaucrats.

Responsiveness similarly suffered under the aegis of patrimonialism and Big Man politics. The state was an instrument of distributing public wealth for private, parochial, or tribal purposes but in ways that more than not pervaded the formal government while favoring ‘big men’ inside and outside the political system. From this stemmed widespread corruption, illegal practices, inefficiency in public goods provision, and the generally poor implementation of government policies (Payani 2000: 135, 147; Dinnen 2001: 184; BTI 2003: 3).

PARAGUAY (2003-)

The administration was ineffective throughout the period. Infrastructure was not particularly sophisticated but the state structures did exist throughout the territory (BTI 2003: 3; O’Shaughnessy and Diaz 2009: 8).

Meritocracy was never established. In the judiciary, independence in personnel management was intended at reforms in the early 1990s and similar initiatives in the 2000s (Basabe-Serrano 2015: 350). The political system did in fact have some inclination to reform the judicial system (Sousa 2007: 92-93). However, judicial independence remained at low levels (Sousa 2007: 107) due to continuance of strong party and presidential influence over Supreme Court appointments. The rules that did exist were only enforced selectively and had many loopholes (Richards 2008: 110; Lambert 2011: 1-2). Similarly, the civil service was dominated by nepotism, politicization, and clientelism in appointments and promotions with no systematic rules protecting the civil servants’ jobs (Lambert 2011: 9; Basabe-Serrano 2015: 350-351). The degree of meritocracy was scored as low by Echebarria and Cortazar (2007: 139). Only a few ministries, the finance ministry, and the central bank were dominated by merit-recruited servants (Richards 2008: 105; O’Shaughnessy and Diaz 2009: 8).

The legacy of the predatory state under Stroessner left nepotism, clientelism, and corruption at the core of what civil servants spent their time on (Richards 2008: 101; O’Shaughnessy and Diaz 2009: 7; Auriol, Flochel, and Straub 2009: 10). The capacity to turn political wishes into firm results was extremely low (Echebarria and Cortazar 2007: 141). Reforms to improve administrative effectiveness were never conducted because of resistance in the Colorado Party (Nickson and Lambert 2002: 172; Richard 2008: 107). In short, responsiveness was severely undercut.

PERU (2001-)

The administration was ineffective throughout the period. Since the early 20^th century at least, administrative structures were established throughout the territory (BTI 2003: 4). One possible exception was the territorial conflict with neighboring Ecuador which was, however, settled by Fujimori in 1998 as part of his state-building policy (Burt 2004: 247).

In the 2000s, some reforms were made with relative success in the judiciary and civil service to obtain autonomy in hiring and firing (Sousa 2007: 92-93; Echebarria and Cortazar 2007: 131). However, de facto independence of judges remained very low and merit recruitment was only partially implemented in some sectors (Sousa 2007: 107; Echebarria and Cortazar 2007: 131, 139). A legacy of politicization from the Fujimori regime survived the transition to democracy (Landa 2001: 2-6). The judiciary similarly remained politicized at the lower courts and the Supreme Court in contrast to the more independent Constitutional Court (Dargent 2009: 252).

Corruption was widespread and penetrated the administration to such an extent that it worked as a regressive tax against the poor (see also Landa 2001: 2-5; BTI 2006: 5; Sousa 2007: 98). The administration was described as dysfunctional and weak in capacity of implementation and policy-formulation (Echebarria and Cortazar 2007: 147-148; see also BTI 2006: 5). In turn, there is no doubt that responsiveness was undermined severely.

PHILIPPINES (1986-)

The administration was ineffective throughout the period. Throughout the democratic years, the state was cut off from any meaningful control in Mindanao and the Cordilleras as well as other areas where armed insurgency groups dominated (BTI 2003: 4; BTI 2006: 4; BTI 2008: 5; BTI 2012: 8).

The constitution of 1987 was not able to constrain the president in undermining the autonomy of the executive administration and the judiciary. The appointment of judges remained a political exercise under which the president and parliament tried to balance national interests (Rose-Ackerman, Desierto, and Volosin 2011: 247-248; for a contrasting assessment of the judiciary, see Bakker 1997: 87). Likewise, the rest of the bureaucracy lacked the Weberian traits assumed in laws since family and friends were still hired by nepotistic principle and party patronage fared well (Kang 2002: 75-77; Hodder 2010). These traits together constituted the image of the Philippine state as predatory, patrimonial, and unable to separate oligarchic and big business interests from state administration (Ferrer 2005: 112). This was the case against the backdrop of some more well-regulated and autonomous agencies (see Gonzalez 2013: 139-140).

Favors to family and friends naturally dominated the administration of state resources given the nepotism of the hiring system (Hodder 2010; see also Goh 2015: 165). Corruption and nepotism was the facilitators of the crony capitalism that came to dominate politics under democracy (Kang 2002: 75-77). As put by Bakker (1997: 113), the sociocultural pressure on the bureaucracy from rich elites and business interests was immense. It hampered government control of policies and implementation and created a fragmented state apparatus (Karadag 2011: 26).

POLAND (1989-)

The administration was effective from 1999 to 2005 but became ineffective from 2006. Inheriting the structures of independent statehood in the interwar period and the extensive communist system territorial penetration was in place in 1989. Reforms of the administration updated this and continued throughout the period (Ekiert 2003: 305-306; BTI 2003: 3).

In terms of meritocracy, Poland was a borderline case in 1989. Its western regions, historically placed in East Germany, inhabited Prussian administrative traditions where the Eastern parts inhabited Russian ones (Møller and Skaaning 2010: 332-333). Changes were thus needed, but in the 1990s politicization and the peculiar Soviet legacy of nepotism at the local levels continued (O’Dwyer 2006: 65-66). Judicial independence was largely sanctified and implemented by 1997 after years of similar patterns of politicization (Bodnar and Bojarski 2012: 667). In 1998, a civil service law established a clear division between administrative and political levels and positions with meritocratic procedures for appointments established from the Director-general down (Gwiazda 2008: 808). Processes of hiring and firing further down in the civil service could be manipulated by the politically selected directors (Gwiazda 2008: 808), but most accounts (e.g. Heywood and Meyer-Sahling 2008: 29, 43) assert us the year 1998 as transitory for bringing meritocracy to the Polish administration.

Meritocracy took a negative turn in 2006. As Gwiazda (2008: 802, 818), party patronage appointments were few under the Democratic Left Alliance government from 2001 but increased again under the Law and Justice government from October 2005. In March 2006, a new law abolished the 1998 law by excluding higher positions from the class of civil servants and abolishing the office of the civil service and council of the civil Service which had hitherto guaranteed a minimum level of meritocracy. The courts managed to stay independent under increased pressure from this new government but the 2006 changes substantially increased the number of politicized positions to a level close to that of the 1990s with 50 % turnover or more (Heywood and Meyer-Sahling 2008: 28; BTI 2010: 7; BTI 2012: 7).

Regarding responsiveness, the 1990s saw very ineffective and corrupt systems of governance, exacerbated by the failures of the patronage system (O’Dwyer 2006: 82-83). However, most accounts (e.g. Ekiert 2003: 306, 312-313) contend that governance was quite effective relative to the task of transforming the socialist economy to EU standards. The administration ‘worked to its best’, as one account asserts us. The decisive moment according to Ekiert (2003: 312-313) was the reform of the tax administration in 1994 which was successful and served to increase the taxation capacity of the state to create a basically efficient and effective state. This also marked the dominance of the ministry of finance over the other ministries thus mitigating ministerial conflicts and divergences. Corruption scandals occurred as in all post-communist countries but corruption was generally controlled (see e.g. Heywood and Meyer-Sahling 2008: 24).
ROMANIA (1991-)

The administration was ineffective throughout the period. The communist legacy of a penetrating state apparatus was particularly strong in Romania (BTI 2003: 3). As an indication of this, it is noted by Boia (2001: 156) that forces of law and order could easily have dissolved demonstrations peacefully rather than through repression in 1989.

In terms of meritocracy, Iliescu the strong patrimonial character of the Ceasescu regime in his first presidency until 1996. Under Constantinescu, some elements of patrimonialism were lifted but in general, patrimonialism was merely decentralized in the sense that the number of politicized offices controlled by the government party decreased while the control of opposition parties increased (Seelin 2004: 120, 125). The judiciary was similarly dependent on part patronage throughout the 1990s (Dallara 2014: 62). To Møller and Skaaning (2010: 334), this is not surprising as Romania was one of the most backward administrative systems already in the interwar period, relying on a mix of formal institutions from Vienna and practices from Russia.

The first piece of legislation on civil service came in 1999. Subsequent reforms in the 2000s brought Romania closer to the Eastern European mainstream (Lee 2009: 279-280). For instance, new requirements for examinations before appointment to the judiciary were installed in 2003 (Dallara 2014: 62). And human resource management and controls increasingly were institutionalized in ministerial hiring and promotion (Cepiku and Mititelu 2010: 71-72). Despite some improvements in practice, the majority of scholars point to meritocracy as a weak fact of life in ministries exemplified by the routinized interference of the ministry of justice in selecting judges (Mendelski 2012: 31-32), the generally high turnover rates among judges (Dallara 2014: 62) and in other ministries, including the continued relevance of patronage networks (Lee 2009: 276; Michalak 2012: 396).

Romanian administration after communism was infamous for rampant corruption which included the inefficient management of public resources and a dependence on local client networks (see e.g. Sellin 2004: 118; Michalak 2012: 393). Nepotism also flourished at the local level hindering any strong transformation of the economy: An anti-reform mentality dominated among the ranks of the senior and lower civil service (Sellin 2004: 125). The most recent waves of administrative and economic reforms in the late 2000s revealed the continuance of weakly coordinated administrative activities and administrative corruption (Michalak 2012: 404-405; Dallara 2014: 68-69).
RUSSIA (1992-1999)

The administration was ineffective throughout the period. In the 1990s, bureaucratic chaos existed to a significant extent (Levitsky and Way 2010: 188-189). And there was a strong favoritism of the local over the center affecting actual governance penetration (Bunce 2003: 181). But from Soviet times the administrative structures did certainly exist.

Russia was the motherland of personalist-patrimonial administrations in many Eastern European countries (see Møller and Skaaning 2010: 325). While this was fundamentally transformed in the Soviet Union, it only introduced the highly politicized civil service system of Nomenklatura in which civil servants were hired as communist party loyals (Kotchegura 1999: 20). An evaluation report covering the years 1991-2006 strongly suggested that the communist legacy had been preserved in these years (Barabashev and Straussman 2007: 373). The works of Roskadri (an agency overseeing civil service appointments) were circumvented throughout the 1990s and similar reform attempts in 1997-1998 failed (Barabashev and Straussman 2007: 377-378). Notably, presidential powers were decreased not just by parliamentary controls but also by the continued existence and functioning of the Supreme Soviet to which former communist civil servants still served (Colton and Skach 2005: 118; Barabashev and Straussman 2007: 373). The other half of the administration was recruited as personal loyals of the president through nepotistic or oligarchic networks of power (Colton and Skach 2005: 118). The same pattern could be seen in the judiciary (Boylan 1998: 1333). Meritocracy was thus bloated at virtually all levels of the administration.

Responsiveness suffered from much the same deficiencies as with the subordination of the security apparatus. Since purges of Soviet civil servants did not occur in the early 1990s, the problem of unmotivated officials longing for the past and engaging in corruption against government policies was highly frequent (Barabashev and Straussman 2007: 374-376). That is, the basic discrepancy became the officials hired by the president and those serving the Supreme Soviet produced huge implementation deficiencies (Kotchegura 1999: 22). Small oligarchic elites and regional elites managed to capture the state partly by bribing officials (Åslund 1999: 64; Barabashev and Straussman 2007: 377; Mendras 2012: 110).

SENEGAL (2000-)

The administration was ineffective throughout the period. The Senegalese state structures were quite old stemming from the European model in the 19^th century. Only Casamance was included late in the colonial period. That state was, however, present in all areas even though less so in remote areas (BTI 2006: 3). The state was even described as omnipresent (BTI 2003: 3-4).

The collusion of state and party was very strong in terms of jobs. Senegal was thus described as a clientelist democracy (Johnson 2009: 40-41). This was a legacy of decades of personalism and politicization of the administration from independence as established by the Socialist Party under Senghor. He basically established a patrimonial state (Diouf 1992: 119; Johnson 2009: 49). The ruling bureaucratic stratum was forced to rely on support from religious leaders and patrons to survive in office (Diouf 1992: 119-120). This continued under the Wade government from 2000 (Johnson 2009: 55).

In terms of responsiveness, haplessness dominated in the administration and corruption was widespread. This was also a legacy of the past (Diouf 1992: 119). Decentralization reforms were highly ineffective because of disloyal bureaucrats at the local level (BTI 2003: 3-4; BTI 2006: 3; Mbow 2008: 158).

SERBIA (2006-)

The administration was ineffective throughout the period. The same argument as for resource supremacy applies to the weak territorial penetration in Kosovo. In a 2015 assessment of the Serbian judiciary, it is deemed to be “inefficient, politically dependent, expensive, and placed in a perpetual reform cycle” (Kmezic 2015: 509). This is the typical view of how Serb reforms of the administration and judiciary to become less politicized and more effective in service delivery have either been absent, insufficient, or incompletely implemented (see e.g. Edmunds 2009: 135; Janicijevic and Milikic 2011: 434). Merit rules are only rarely respected, typically through the peculiar version of politicization in which demands for official exams are lowered so as to place actual selection of recruits by the oral exams (Meyer-Sahling 2012: 34-35). Even after 2006, key positions in the state bureaucracy and judiciary are still determined by government connections (Edmunds 2009: 135; Verheijen and Rabrenovic 2015: 19). Moreover, corruption and patronage tend to dominate at all levels of government and was deemed pervasive in a series of reports of international organizations and the EU during the 2000s (for an overall assessment, see Ramet 2010b: 286, 298).

SERBIA & MONTENEGRO (2000-2005)

The administration was ineffective throughout the period. Territorial penetration followed the ebbs and flows of resource supremacy in the 1990s and early 2000s. It was thus very weak, if not outright absent, in the Kosovo region (BTI 2003: 4).

The post-communist Serbian administration, normally subsuming Montenegro, can be conceptualized as Eastern Oriental, closely attached to Ottoman rule and thus among the most politicized and corrupt administrative models of Eastern Europe (Møller and Skaaning 2010: 334; see also Sevic and Rabrenovic 1999: 49). Serbia developed its first independent state administration when the country rebelled against the Turks in the 19^th century but the Ottoman principles and informal institutional legacies stuck. After WWII in Tito’s socialist Yugoslavia, administration was greatly centralized in Belgrade, a class of civil servants was legalized, and the peculiar communist bureaucratization took hold in which only ideologically appropriate candidates achieved ministerial jobs (Sevic and Rabrenovic 1999: 54-56, 67). As a counter-reaction, Milosevic deliberately led an anti-bureaucratic revolution in which communist officials were purged and a wave of politicized positions installed with the purpose of furthering Serbian nationalist policies (Haug 2012: 336-337). This applied to the civil servants and the judges alike (Boduczynski 2010: 172). Judicial reform to increase professionalization and political independence stalled after Djindjic’ death and the return of Kostunica to power (Ramet 2006: 532).

The Ottoman Empire also left a legacy of deeply penetrating and widespread corruption at all levels of government (Møller and Skaaning 2010: 334). Many of the crimes, corruption, and lawlessness conducted by judges and officials in the 1990s continued after 2000 under the Kostunica-Djindjic leadership (Ramet 2006: 527). Here, unresponsiveness was described as widespread state capture from below of the ministry of finance and ministry of education, notably, where officials worked to appropriate state funds for party purposes or extra-parliamentary sectors (Pesic 2007). But if attentive to political signals, it was more likely the majority party than the government as such (Sevic and Rabrenovic 1999: 73).

SIERRA LEONE (2002-)

The administration was ineffective throughout the period. Territorial control was undermined by the civil war but reestablished, like resource supremacy, with the deployment and department of the UN mission and the successful recruitment of a new police force (BTI 2003: 4-5; BTI 2008: 7; Harris 2013: 19).

Institutional decay and politicail interference came to dominate in the 1970s. Favoritism and nepotism emerged as principles of politico-administrative relations (Ndumbe 2001: 90-91). There are no indications that the administration was reformed after the civil war ended and thus, politicization, favoritism, and nepotism likely continued. Quite the contrary: Neopatrimonialism continued in the 1980s and through the civil war (Bøås 2001).

The state was a predatory, patrimonial one with a lack of central control despite excessive attempts to do this. Chiefs upheld administrative autonomy. Generally, the ministries were hampered by ineffectiveness. Only the finance ministry had a small, effective bureaucratic elite (Harris 2013: 153-154). Administration remained unreliable and corrupted (BTI 2006: 5).

SLOVAK REPUBLIC (1993-)

The administration was ineffective throughout the period. Territorial penetration was equally strong as in the Czech Republic when Slovakia was given formal and de facto control during late 1992. But the more Eastern European (Hungarian) patrimonial-administrative legacy soon showed (see Malikova and Staronova 2001: 8; Møller and Skaaning 2010: 334). In the mid- and late 1990s, bureaucracy was heavily politicized as President Meciar ensured the hiring of political loyals at all levels, including among judges which inhibited a stable rule of law. His party, HZDS, dominated administrative positions and decisions (Goldman 1999: 62; Kopecky and Mudde 2000: 66-67). With the end of the Meciar presidency in 1998, more regulation of the civil service was introduced. A civil service act was adopted in 2001 and a civil service office was installed in 2002 to oversee the hiring and firing processes. However, rather than a deep-felt wish of politicians the meritocratic reforms were more paper than practice and were only wholeheartedly implemented in ministries and agencies conducting EU affairs (Gajduschek 2007). The judiciary’s independence was secured from the mid-2000s but the ministry of the interior which managed civil service affairs was excluded from the reform processes, and the overall personnel management system was thus more politicized than meritocratic (O’Dwyer 2006: 66-68, 82-83; Nemec 2009: 291; Kosar 2013: 21-22). From 2006, civil service affairs were actually deregulated and thus even more politicized (Meyer-Sahling 2009: 15).

Corruption is widespread in Slovakia and in most surveys more so than in comparable countries such as the Czech Republic, Hungary, and Poland. Efforts to fight corruption have largely been unsuccessful (Malikova and Staronova 2001: 1; Møller and Skaaning 2010: 341). It was, however, difficult to assess whether corruption generally inhibited implementation. Nevertheless, implementation of public services was generally problematic in the 1990s and early 2000s as politics was based on patrimonial, clientelist logics which weakened bureaucratic autonomy, center-periphery relations, and control. The majority of scholars points to this as the reason for a generally ineffective system of governance (see e.g. BTI 2003: 5; Adam 2004: 8; Nemec 2009: 296). Yet, the decentralization of fiscal management from 2001 to 2004 served to mitigate the most significant features of this problem (BTI 2008: 5; Nemec 2009: 299). As I take the center-periphery relations as most detrimental to responsiveness, I code responsiveness from 2005.
SLOVENIA (1991-)

Administrative effectiveness was established throughout the period. Territorial penetration was already in place at the time of independence in 1991. A rather sophisticated system of public finance already existed (Borak and Borak 2004: 62-63). This relied on the immediate transfer of federal Yugoslav officials to the Slovene administration in 1991 (Luthar et al. 2008: 510), but it also hinged on decades old infrastructures (telecommunications, railways, and roads) established in grand projects after WWII (Luthar et al. 2008: 461-463; for similar assessment of later years, see BTI 2003: 4).

In contrast to Croatia, the administrative legacy is mostly associated with the Austrian, Weberian tradition rather than the Hungarian patrimonial one (Møller and Skaaning 2010: 334; see also Luthar et al. 2008: 262-263). Some less fortunate legacies of communism plagued Slovenia in the first years of independence, and still partly did so in 2010. Most notably, political parties still tended to treat ministries as their feudal territory. As noted by Fink-Hafner (2010: 246), this did not leave the administration immune to clientelistic relations with politicians. However, the communist period also saw the development of strong judicial autonomy (Dallara 2014: 37), and the most problematic feature of communism was the rigidity of administrative rules which in itself poses no problem to meritocracy. Thus, it seems fair to say that meritocracy was already reasonably in place in 1991. Judicial autonomy was reassured in the constitution of 1991 (Dallara 2014: 37), and further strengthening of the professionalism of civil servants was conducted in reforms of the 1992-2004 era – a period much praised by the EU and other international agencies as models for post-communist countries (BTI 2003: 4; Meyer-Sahling 2009: 17; Dallara 2014: 37-39). In later years, meritocracy has been even more firmly established with a ratio of 1-870 of officials hired on personal trust (Cehovin and Hacek 2015: 146) and very few positions filled by party loyalties (Meyer-Sahling 2009: 37).

Responsiveness is equally strongly founded in the Austrian tradition (see Møller and Skaaning 2010: 334). Slovenia has been one of the least corrupt countries in the world. Where corruption persists, it is due to individual rather than systemic traits (Ramet 2006: 558; Fink-Hafner 2010: 246). EU-driven reforms of the judicial sector have been effectively implemented, improving efficiency and the fight against corruption. As stated by Dallara (2014: 36), “The fact that a very efficient structure for the negotiation was rapidly organized is evidence of this high administrative capacity.”

SOUTH AFRICA (1994-)

The administration was ineffective throughout the period. Territorial integrity was fully appreciated by 1994 and beyond as the ANC movement was never geographically concentrated and because South Africa had an old bureaucratic system at its disposal (BTI 2006: 4). The British had built roads and telecommunications from late 19^th century to replace the rudimentary Boer Republics (Robinson 1996: 37-38).

Meritocracy initially existed but was dismantled from around 1999. Throughout, judicial independence was secured as the president followed the recommendations of an advisory board to the hiring of judges (Madhuku 2002: 235; Butler 2009: 124-125). According to Kopecky (2011: 722), the South African bureaucracy scored 0.39 on a scale from 0 to 1 with 0 being fully Weberian. This placed among South Africa among probably the most meritocratic bureaucracies in Africa. However, the number also hided some information as merit recruitment is only one component of Weberian bureaucracy. South Africa inherited a British-style administration with a powerful Public Service Commission that oversaw appointments to avoid politicization. However, from the Mbeki presidency the state and the ANC increasingly colluded with ANC achieving an ever stronger position in the party system (Booysen 2015: 28). Decisive changes occurred in early 2000 with a bunch of restructurings in the public sector: Notable agencies were brought under direct political control, including its staff, two top levels of the administration became politicized and the next two were hybrid with some political and some administrative considerations while the levels below remained empowered by the administration itself (Cameron 2010: 686). Most career promotions further became dominated by political considerations and party patronage decisions (Cameron 2010: 687). These traits were only exacerbated under Zuma from 2007 who consolidated huge access for the President and the ANC to fuse the state with the party by appointing high-level servants and thousands of workers in the public sector (Butler 2009: 120-121; Booysen 2015: 27-29).

It was hard for the ANC to implement their projets because the civil servants kept implementing their own projects and taking their share of the public funds (Booysen 2015: 31). Corruption was widespread and breakdown of discipline was common. Old authoritarian values persisted among the bureaucrats some of whom were hired under apartheid and outright disloyal of the ANC leadership (Cameron 2010: 686; see also von Holdt 2010). In turn, overall responsiveness was severely undermined.

SOUTH KOREA (1988-)

The administration was effective throughout the period. The early settlement of the territory and the lack border changes later on (except the partition of Korea after WWII) provided for a highly functional administrative structure throughout the territory (BTI 2003: 3; Kim 2006: 64). Under autocracy, Park nurtured the bureaucratic organization further (Clifford 1994: 47).

Despite some extent of capture by big business in the 1980s, meritocracy was maintained and further refined under democracy by reforms after the financial crisis in 1997 (Cheng, Haggard, and Kang 1998). Responsiveness similarly remained intact. Anti-corruption efforts were conducted after 1997 and although not quite effective, basic responsiveness was still in place under democracy. Policy-coordination was still highly effective (Cheng, Haggard, and Kang 1998; BTI 2003: 3).
SRI LANKA (1991-2010)

The administration was ineffective throughout the period. The civil war excluded the state from being present in the vast areas control by the Tamil insurgents (Stokke 2006). However, with the end of fighting in May 2009 the state achieved control of the whole island (BTI 2010: 5; Thiranagama 2013: 92-93). Therefore, territorial penetration was established from 2009.

Since independence, interferences in the judiciary from inside and outside the political system were common. A constitutional amendment in 2001 was supposed to install judicial independence but the law was abolished in 2010 without it having ever had any substantial effect (DeVotta 2014: 150). Despite the fact that meritocracy was part of the large reform move from the late 1970s and several times afterwards (Kumara and Handapangoda 2008: 83), party patronage remained a dominating principle in recruitment patterns to the state administration (Root, Hodgson, and Vaughan-Jones 2001: 1358; McCourt 2007: 431). Some (e.g. McCourt 2007: 431) believe that the reform in 2001 basically installed meritocracy in the civil service but others (e.g. BTI 2003 4) dispute this and the lack of meritocracy in the judiciary in any case is sufficient for coding meritocracy as absent.

Over the years, substantial inter-agency conflicts emerged in the bureaucracy and politicization hampered the competence of civil servants and led to massive demoralization while corruption remained relatively widespread. Officers felt no sense of responsibility and were not monitored. There was no functioning unitary command in the civil service which hampered efficiency (Root, Hodgson, and Vaughan-Jones 2001: 1358-1360; BTI 2003: 4). Thus, responsiveness was, on balance, weak (see BTI 2008: 5 for a different assessment).

TAIWAN (1996-)

The administration was effective throughout the period. Under Japanese colonialism, a strong bureaucratic organization and infrastructure had been built (Gold 1986: 34-35). The existence of functioning administrative structure extending across the territory continued under KMT leadership (Fields 1997: 122-125) and democracy (BTI 2008: 5).

Taiwan, like South Korea and Japan, became one of the classic developmental states in the Cold War with highly developed and autonomous bureaucratic agencies (Gold 1986: 34-35; Fields 1997: 122-125; Wu 2007: 977). The bureaucracy remained firmly meritocracy through the democratic transition and beyond with the exam-centered recruitment process inspired by the Chinese (So 2015). Also, judicial independence was secured early on in Taiwan’s independent life by inspiration from the German and American traditions (Chisholm 2014: 895-896).

Responsiveness was strong throughout the period. Policy-making was effective by cooperation between politicians and a highly skilled bureaucracy which remained small and efficient (Clark 2000). Policy-coordinating agencies, as in South Korea and Japan, helped controlling bureaucracy when needed (Cheng, Haggard, and Kang 1998). In any case, this need was scarce since the administration had a firm culture of discipline, regularity, authority, and competence as guiding principles ensuring low levels of corruption (Gold 1986: 34-35; Fields 1997: 129; BTI 2008: 5).

THAILAND (1992-2006)

The administration was ineffective throughout the period. In the 1990s, old territorial structures of the state in terms of resource control were strengthened further (Vandergeest and Peluso 1995: 391-392). In the 2000s, the administrative presence throughout the territory was secured (BTI 2003: 3-4).

Meritocracy was in place throughout the period. Originally, Thai bureaucracy was known for its autonomy on a par with South Korea and other developmental states in East Asia. However, this slowly weakened under democracy (Bowornwathana 2001: 297, 306; Ockey 2004: 147). Some accounts point out that meritocracy was even dismantled by Thaksin after the Asian Financial Crisis due to his use of politicization of state agencies (Painter 2006: 27; Prasirtsuk 2007: 887-888). Also, a critical account points out that the king had access to interfere in judge appointments and promotions (Coorey 2010: 3-4). Nevertheless, the majority of studies point to the continued existence of basic meritocracy. Many studies even point out that Thaksin’s attempts of politicizing the bureaucracy, as they went beyond a few central state agencies, stranded against the rules of conduct in the administration and the autonomous powers of the high-level civil service. Politicization never reached below agency directors and some permanent secretaries (see e.g. Bowornwathana 2001: 297-306; Ockey 2004: 147-149; Painter 2006: 28; Prasirtsuk 2007: 887-888). As Unger (2003: 181) notes, writing after Thaksin’s entry, merit still largely determined promotion and recruitment to the public sector.

Thaksin’s politicization attempt was not just a matter of getting absolute political control but a result of existing problems in getting the bureaucracy to comply with government policies. Its autonomy was to be dismantled by a range of NPM reforms (Painter 2006: 27). Bowornwathana (2001: 296-297) strongly suggests that the general relationship between politicians and bureaucrats was smooth with the bureaucrat functioning as expert while demanding promotions in return. Yet, he also acknowledges the power of the senior bureaucrats in setting the pace of policy-making and eventually deciding between policy alternatives. Thus, responsiveness would in any case be ambiguous but when taking account of policy implementation directly, it seems clearer that unresponsiveness was the typical outcome. As Painter (2006: 28) notes, the strength of the bureaucracy was neither implementation nor deliberation with the political system but the maintenance of privileges. The relationship between the bureaucracy and the private sector was particularistic, clientelist, and personalized (Doner and Ramsay 1997: 238). More generally, the bureaucracy had been highly influential in policy-making for deacdes but no party was able to circumvent this or establish any effective accountability institutions (Unger 2003: 181). Judges and civil servants often delayed projects to achieve privileges or serve private clients (Ockey 2004: 147-149).

TRINIDAD & TOBAGO (1962-)

The administration was effective throughout the period. Sources are scarce. Yet, there are no indications that the state lacked penetration of the small island territory. By contrast, decolonization meant a quick and strong build-up of a bureaucratic organization (Hope 1983: 50). Meritocracy was established by the same token. Politicians worked with permanent ministerial officials (Hope 1983: 50). A Public Service Commission was established in 1950 and its independence reassured in 1962. Subsequent decades saw allegations of corruption and politicization in and of the commission but there was no serious evidence of major breaches to the principle of meritocracy (Bissessar 2001: 75-80). This was consolidated by the transition in 1962 to a British system in which ministers, not the prime minister, controlled the civil servants (Ryan 1972: 345).

The standards of effective cooperation between the government and the ministries were not severely diminished in decolonized Trinidad & Tobago (Ryan 1972: 358). There were some conflicts between the politicians and the servants in the decades after independence and some mistrust between them (Ryan 1972: 354; Hope 1983: 51). However, on balance, responsiveness seems to be in place. The colonial legacy from the British served the administration towards responsiveness, and there was no radical break from this as in Jamaica. Indeed, career paths were streamlined alongside measures to improve implementation (Riley and Nunes 2004: 5). The lack of confidence that did exist between bureaucrats and politicians was more a result of mistrust of politicians than the other way around, and bureaucracy was firmly autonomous of society (Ryan 1972: 354, 358).
TURKEY (1983-)

The administration was ineffective throughout the period. Turkey was a deviant case among Muslim countries due to its continued strength of state penetration and institutionalization under democracy. This was a legacy of having been the center of Ottoman rule (Kazancigil 1994: 216). Despite strong deficiencies in organizational coherence and the ability to extract resources from society, the bureaucracy was vividly present throughout the territory (BTI 2003: 4; BTI 2006: 7).

During the 1980s, a return to Ottoman traditions of personal loyalty of the government was conducted. Merit mattered little in hiring and firing civil servants who were instead expected to show absolute loyalty to the Atatürk thought (Heper 1989: 461). In the 1990s and 2000s, meritocracy continued to exist mostly on paper and only effectuated in a few sectors. Typically, a combination of merits and political patronage guided recruitment. Training of senior officials in particular was largely seen as redundant (Acar and Özgür 2004: 203-205). Judicial independence was also never a strong trait as the military and executive could constantly interfere in the appointment and promotion of judges according to their attitudes toward Atatürk ideas, particularly with regards to upholding a secular order. Only in September 2010 did a constitutional amendment provide better guarantees for judicial independence (Asik 2012: 146-148).

The judicial system was even more susceptible to supporting Atatürk ideas, if needed at the expense of government intentions. Since the inauguration of the bureaucracy and judiciary in the Kemalist state, they were highly politicized – if not by employment then by education and training which served to deeply ingrain a concern for the well-being of the nation (Baran 2000: 130, 134). The Turkish judiciary, for instance, was educated to “avoid the spirits of the times” – if necessary, against elected representatives (Coskun 2010: 44). Thus, on the one hand politicization could ensure responsiveness but it came with a risk of future unresponsiveness in case of a significant change in policy substance. Asik (2012: 146) shows that the judiciary conducted ‘military tutelage’ in the 1980s and 1990s sometimes at the expense of the wishes of the elected governments. However, the balance changed in the 2000s as the military position in politics was weakened. From that moment, it conducted ‘civilian tutelage’ in which its decisions were more and more controlled by the executive. In the absence of other analyses focused explicitly on the question of responsiveness, I take this to mean that the administration was unresponsive until 2002 and the appointment of General Özkök and thus responsive from 2003.

Corruption indeed existed throughout the period in the form of bribes regarding land and real estate, money laundering, and public contracts. However, reforms in the early 2000s comprehensively and effectively ensured a significant improvement in public service delivery, transparency, and responsiveness including the local level (Köseoglu and Morcöl 2014: 137, 150). The corruption that continued generally did not hinder law enforcement (BTI 2008: 5).
VENEZUELA (1959-2005)

The administration was ineffective throughout the period. Just as for resource supremacy, the only substantial problem of territorial penetration existed by the Colombian border and in some rain forest areas. However, these problems were insignificant and thus, territorial penetration existed.

The elite settlement in 1958 established a highly politicized administration similar to the spoils systems of Southern Europe. The executive controlled the appointments to the bureaucracy at an overall level while he led the other political parties share the fruits too. This system continued in relative stability until 1998 (Cornell and Lapuente 2014: 1292-1293). Similarly, the judges were completely controlled by the parties in terms of tenure and promotion, hiring and firing (Burns 2009: 97). From 1988 to 1998, the judiciary underwent a transition period with less systematic patronage appointments but it was still highly politicized (Burns 2009: 108-109). The politicization continued under Chavez from 1998 into the 2000s. It only grew less predictable and more personalist with Chavez and his party controlling almost all appointments to the civil service and the judiciary (BTI 2003: 4; Taylor 2014).

Responsiveness suffered under the spoils system because the politicization was never perfect and the bureaucracy was vulnerable to interference from outside forces. Against its official intentions, the two-party system functioned on terms of favoritism, clientelism, and nepotism (Peeler 1985: 77-78; Baburkin et al. 1999: 143). The politicization tended to create a lack of morale and pecuniary relationship between the president and the servants which caused a lack of responsiveness to his wishes if monitoring was weak at a specific time (Stewart 1978: 123). Chavez’new system from 1998 only exacerbated these problems of corruption and clientelism (BTI 2003: 4; Coronel 2006: 1-3).

UKRAINE (1991-)

The administration was ineffective throughout the period. Although the huge territorial size of Ukraine nurtured patron-client relations and a strengthening of decentralized rule in the transition years, Ukraine never was near state collapse (Kudelia 2012: 418). As pointed out by Gallina (2008), to the extent that state capacity was weak this was because of political dispute over the appropriate degree of federalism rather than administrative insufficiencies. The Soviet legacy instead had left an extensive ministerial system and apparatus for extraction of resources (Nordberg 1998: 43; Fritz 2007: 109).

Ukraine’s administrative personnel system was inherited from Tsarist and communist Russia for the most part, although the western parts had mixed traditions of Austrian and Hungarian rule (Møller and Skaaning 2010: 336). Under Krachuk, the administration was in a state of disarray and thus did not produce any reforms. The National Bank and numerous other key public agencies and the courts were politicized while officials hired under the norm of Nomenklatura continued in many positions (Nordberg 1998: 42; Condrey 1998: 30-31; van Zon 2000: 39; Kudelia 2012: 419). Under Kuchma from 1994 to 2004, patronage continued to drive recruitment throughout the bureaucratic apparatus. There were frequent reshufflings of high-level offices to thwart the growth of political competitors and cultivate regional networks (BTI 2003: 2; Pleines 2012: 127; Kudelia 2012: 420). Much changed politically with the Orange Revolution in 2004 and the inauguration of Yushchenko (sitting until 2010). However, he continued Kuchma’s norm of appointing all agency leaders and delegate his power to his own and other parties on the agreement that he was supported while they gained control of certain offices regionally (Kudelia 2012: 423; Pleines 2012: 127-133).

Ukraine was a prime example of a captured state throughout the period. Oligarchs, mostly local business elites, were successful in constantly holding the state apparatus at arm’s length to their benefit by bribing judges, ministerial heads, and local tax officials (Kudelia 2012: 418; Pleines 2012: 128). Political power struggles between parties and between central and regional levels embodied struggles at the administrative levels inhibiting any centrally coordinated policies to be fulfilled (Gallina 2008: 191, 197). This characterized the 1990s (van Zon 2000: 35-36). Attempts to mitigate this and control corruption in the 2000s were few and mostly ineffective as they were met with regional and political barriers. The latest efforts after the 2004 revolution neither effectuated a well-governed polity (Fritz 2007: 201). On top of this, the whole period saw a conflict between the Donbas and Crimea regions on the one side claiming autonomy and the Kiev administration attempting to govern them on the other side. The effect was that state capture regionally and centrally continued unabatedly (BTI 2010: 7; BTI 2012: 8).

URUGUAY (1985-)

The administration was ineffective throughout the period. As for resource supremacy, the state did not suffer any territorial or security breaches in connection with the transition to democracy. Decisions could be implemented from center to periphery without obstacles (BTI 2003: 4).

Meritocracy was still undermined by the legacy of the coparticipación between the Colorado and Blanco parties. The degree of meritocracy was relatively in Latin America but below that of Brazil (Echebarria and Cortazar 2007: 139). Despite a forceful movement since the 1960s towards meritocracy and an undermining of the two parties (Buquet and Pineiro 2014: 7-9), political rationale still strongly dominated appointments to the administration (Bergara with Pereyra et al. 2004: 24). A large part of the bureaucracy was still occupied on the basis of patronage and clientelism of party officials (Bergara with Pereyra et al. 2004: 25). While the judiciary was and remained independent in all relevant aspects of personnel management (BTI 2003: 4; Bergara wth Pereyra et al. 2004: 29; Sousa 2007: 98), the rest of the civil service remained mired in the competition between Colorados and Blancos (Lanzaro 2006: 33). Numerous reforms and formal assurances in the constitution did not change this (Panizza 2004: 23; Lanzaro 2006: 8).

The new constitution of 1966 did not change unresponsiveness immediately (Finch 1991: 207-211) but it seems that responsiveness was achieved at some point during the years of dictatorship. In the post-1985 era, the administration was assessed as having high levels of inefficiency measured as expenditure relative to size of the public sector (Bergara with Pereyra et al. 2004: 24-27). However, reports of low motivation and inefficient behavior do not agree with the rest of the accounts which point to the high levels of effectiveness in implementation, the relatively impressive functionality of the public sector, and the effective controls on public spending and streamlining of state agencies (BTI 2003: 4; Panizza 2004: 6-7; Echebarria and Cortazar 2007: 147; BTI 2012). With these overall assessments of the democratic period, I code responsiveness as present from 1985.

ZAMBIA (2008-)

The administration was ineffective throughout the period. Some ineffieicnies in the public sector existed but the state did penetrate the territory fully (see BTI 2006: 4; BTI 2012: 5). In the constitution, judicial independence was firmly assured but in reality, political interferences were frequently reported (Alemika 2009: 3-4). More seriously, the remaining civil service was affected by reforms in 1968 which introduced great party control over state jobs and installing pparty control over the appointment of local governors and officials as part of decentralization. Such a spoils system and alongside the general traits of neopatrimonialism of nepotism and clientelism dominated the bureaucracy in the 1990s and 2000s still (Szeftel 2000; Stevens and Teggemann 2004: 50-51; von Soest 2007).

Efficiency in the public sector was known to be very limited in the early 2000s (BTI 2003: 3). With regime change in 1991, corruption virtually became institutionalized at all levels of administration and in the civil service and judiciary as well (Momba 2007: 116). A reform in 2000 sought to build capacities for better implementation and fight corruption had mixed results. Some improvements were made but the assessment pointed out that results were lagging overall (Stevens and Teggemann 2004: 48; BTI 2006: 4). Particularly from 2005, the improvements were reversed (BTI 2012: 5).