Turkey's Graduation from the International Monetary Fund

The Articles of Agreement of the International Monetary Fund (IMF, or the Fund) underline the "temporary" nature of its financial assistance to its members. Nevertheless, repeat lending became a salient feature of IMF operations in the 1980s and 1990s. In 2001, loans to repeat users accounted for half of all IMF lending.

The IMF's Independent Evaluation Office (IEO) highlighted "prolonged" use of IMF support in a 2002 study where "prolonged user" is defined as a country that was under an IMF program for seven years out of any 10-year period. Forty-four countries met this definition of IMF addiction over the period 1971-2000. As of 2012, 25 of these had outstanding arrangements with the IMF. Turkey, one of the prolonged users identified in the IEO study, finally graduated from IMF support in 2008.

A growing concern among critics of the IMF is that long-term use of its resources may be particularly harmful because it creates patterns of dependency. The moral-hazard explanation of the prolonged use of IMF resources starts from the assumption that IMF support weakens the incentive to carry out economic reforms. IMF financing, in some instances, may reduce the incentive for governments to solve long-term structural problems that contribute to slow growth and underdevelopment.¹

The adverse-selection view, on the other hand, predicts that IMF programs have more successful growth performance among prolonged users than among short-term users. Over a longer time horizon, the IMF is able to screen countries and determine which are willing to commit to policy reform, "gradually mitigating the problem of asymmetric information that lies at the heart of the IMF's performance problem."²

The success of IMF programs is conditional on whether the economic reforms that the IMF prescribes are implemented. Empirical literature on IMF program compliance indicates that domestic political factors are important in influencing implementation. Factors such as regime type,³ special-interest groups,⁴ polarization⁵ and approaching elections⁶ affect the ability of governments to successfully implement IMF-supported programs.

A country that signed 19 programs with the Fund in 47 years certainly offers a variety of data for a single country (see table 1). Populist cycles and periodic fiscal crises have emerged as persistent features of the Turkish economy over the past few decades. As a result, the Fund's involvement in the Turkish economy has been almost continuous. In 2009, Turkey declared that it preferred to manage without IMF support and "move forward without a walking stick."⁷ The country managed to navigate through the global financial crisis without resorting to IMF support and is scheduled to clear its remaining $864 million debt in 2013, settling its long-standing account with the Fund.

This paper aims to provide insight into Turkey's evolving relationship with the IMF in the presence of external constraints as well as domestic-political-economy considerations. The purpose of the study is to examine the role of the IMF, not to undertake a comprehensive assessment of Turkey's development strategy.⁸ In their study of the IMF in Turkey, Arpac and Bird⁹ document the importance of special-interest groups, political cohesiveness and program ownership by the government. It is upon these findings that this paper builds.

TURKEY'S EARLY IMF PROGRAMS

Turkey began implementing liberal reforms in 1980. The program initiated that year was one of the first of its kind, based on close collaboration between the IMF and the World Bank and involving the application of "cross-conditionality." It was a far-reaching and gradualist program that envisaged stage-by-stage liberalization and integration into world markets.

The elimination of payment imbalances and the reduction of inflation were identified as the immediate objectives of the program. The principal instruments included monetary contraction, exchange-rate devaluation and the liberalization of key relative prices such as interest rates on bank deposits. The longer-term objective of structural transformation was to overcome the limits on economic growth imposed by the previous import-substitution strategy. The program sought to lay the foundations for sustained economic growth by way of financial reform and the liberalization of the foreign-trade regime.¹⁰

A three-year program — the first in the Fund's history — envisioned financial assistance of 1,250 million special drawing rights (SDRs), which were disbursed in full. This amount constituted the highest credit extended up to that point by the IMF, 6.5 times the Turkish quota with the Fund.¹¹

In June 1983, an additional one-year follow-on stand-by arrangement was approved for 225 million SDRs (75 percent of the quota). This was seen as a continuation of the earlier arrangement, and the conditionality was similar to that of the 1980 program. At the end of 1983, the newly elected government requested cancellation of the existing arrangements. In April 1984, a final one-year arrangement was approved for 225 million SDRs (52 percent of the quota) to replace the canceled arrangement.

Both the IMF and the World Bank highlighted this period as a model of successful adjustment.¹² An examination of the facts associated with post-1980 performance suggests that the policies were reasonably successful with respect to growth and balance-of-payments objectives with exports. Per capita growth averaged about 2.5 percent a year and social indicators improved; in the 1980s, Turkey became a star performer in terms of growth in manufactured exports.

After a decade of successful implementation, the momentum of the reform process was in decline towards the end of the 1980s. Despite the IMF's willingness to resume the reform process, the Turkish authorities declined to sign a stand-by arrangement on the grounds that this would be associated in the public mind with the failure of economic policies implemented almost a decade earlier. After all, the balance-of-payments problems were resolved, and access to international capital markets was restored; the IMF's role was officially over. However, the IMF and the international financial community, in general, continued to exercise an indirect influence during the post-1984 phase. For example, the practice of extra budgetary funds and export-tax rebates was subjected to strong criticism from the Fund, resulting in a transfer of 30 percent of revenues from these funds to the consolidated budget in 1986 and the eventual removal of the export-tax rebate system by the end of 1988. Yet, the direct conditionality that had characterized the early 1980s was to a large extent absent in the second half of the decade.

Nonetheless, during this period, domestic political pressures increasingly replaced external constraints, and this had significant ramifications for the pattern of economic performance during the 1990s. One significant turning point in this context was September 1987, which marked the return of unrestricted party competition. The distributional pressures that had been largely held in check in the early eighties manifested themselves in the form of larger fiscal deficits and higher rates of inflation. Finally, the August 1989 measures completed the last stage in the liberalization of the capital account and the establishment of the full convertibility of the Turkish lira.¹³ This resulted in a dramatic increase in inflows of short-term international capital. In worsening economic conditions, the Turkish government perceived large inflows of capital as a key mechanism to restore growth and chose to overlook basic structural deficiencies in the economy such as the large fiscal imbalances. In this way, Turkey entered into a highly fragile pattern of debt-led economic growth, resulting in successive financial crises in the post-1990 era.¹⁴

The financial liberalization process has also altered the process whereby public deficits have been financed. The establishment of domestic capital markets presented the government with the opportunity to borrow domestically. As a result, domestic debt started to increase. While Turkey had a domestic debt stock close to zero in 1987, this increased continuously, to 25-30 percent of GDP by 2000.

Another striking feature of the Turkish economy during the 1990s was high real interest rates as the growth in the government's financial requirements outstripped the increase in the size of the domestic financial system. As a result, Turkey has been caught in a vicious circle of increasing deficits and rising interest rates.

In the context of a highly fragmented party system, successive coalition governments in the 1990s lacked the capacity and incentives necessary for undertaking fiscal stabilization and regulatory reforms. There were a few attempts at reform well before the onset of the 2001 financial crisis, however. For example, the coalition government that had come into office in 1991 appeared to be strongly committed to undertaking a drastic stabilization program designed to overcome macroeconomic instability in the economy. Yet it was evident from even a superficial examination of the support base of the coalition government that powerful forces would try to prevent the government from implementing a far-reaching reform program.

The government represented a coalition of losers from the reform waves of the 1980s. The principal constituencies of the coalition included organized labor, the farming community and small businesses oriented towards the domestic market. Paradoxically, the government was committed to reform on paper, yet was elected on a populist agenda of redistribution. The inflows of capital provided a major escape route for policy makers confronted with acute distributional dilemmas. They allowed the economy to expand at a faster rate than would otherwise have been possible, a process that simultaneously helped to satisfy the distributional claims of the key groups involved. By the end of 1993, it was clear that the rapid growth of the Turkish economy was built on shaky foundations and was unsustainable on two counts. The budget was in disarray, given the rapid increase in expenditures without a corresponding increase in tax revenues. In addition, the balance of payments was very fragile; a growing current-account deficit was covered by inflows of primarily short-term capital. Hence, the traces of an impending crisis could be detected at the end of 1993 by observing the position of the budget and the balance of payments (figures 1 and 2).

When two major credit institutions had reduced the country's credit rating in response to the growing disequilibria, a crisis of investor confidence set in, resulting in a major outflow of short-term capital.¹⁵ Thus, the financial crisis manifested itself as a balance-of-payments crisis leading to massive depreciation of the exchange rate in the early months of 1994. The process culminated in the decision of the government to implement a major stabilization program in association with the IMF in April 1994.

A 14-month stand-by arrangement was built around a large, front-loaded, fiscal correction and a tighter stance on monetary policy. The program implementation was satisfactory to begin with. At the end of the second program review in April 1995, the program was extended by six months. However, in September 1995, the governing coalition was dissolved unexpectedly and early elections were called. This resulted in a sharp deviation from program understandings, and the final review was never completed.

The 1994 crisis and ensuing policy measures, however, were not serious enough to generate deep-seated structural changes in the Turkish economy. Following the smooth recovery process, many of the key elements of the disequilibrium of the early 1990s — huge budget deficits, chronic rates of high inflation and heavy dependence on short-term capital inflows — remained intact.

An extreme fragmentation of the political system, with one unstable coalition after another and the tenure of an average minister lasting just over one year, made implementation of significant economic reform extremely difficult. Chronically high inflation was averaging 60 percent, and debt rose from about 30 percent of GDP at the beginning of the decade to over 70 percent in 1999.

Consequently, by the end of the 1990s, policy makers had discovered that progressively fewer resources were available for populist redistribution once the domestic debt had been paid out. This realization forced politicians to seriously reconsider the feasibility of continuing on a populist path.

BACK FROM THE BRINK

In July 1998, the staff-monitoring program was agreed with the IMF, which initiated a number of targets and provisions related to the budget, monetary policies and various structural reforms. This was followed by a proper stand-by in December 1999, after the general elections. With the December 1999 program, Turkey took a serious step to deal with the structural causes of chronic inflation and budget deficits.

The new coalition government formed after the general elections of April 1999, incorporating the left-nationalist Democratic Left Party (DSP) under the leadership of Bulent Ecevit, the radical Nationalist Action Party (MHP) under Devlet Bahceli, and the right of center Motherland Party (ANAP) under Mesut Yilmaz, gave the impression of having a significant commitment to implementing a far-reaching program of disinflation and reform. The fact that an IMF program was agreed for the first time without the presence of a major crisis seemed to provide additional support for an optimistic assessment. The wide ideological spectrum of the coalition government did not seem to affect the prospects of the program at the beginning.

The stand-by program was ambitious, targeting an inflation reduction to 25 percent by the end of 2000, 12 percent by the end of 2001, and 7 percent by the end of 2002. Tight fiscal, monetary and exchange-rate policies formulated in line with decreasing inflation were the basis of the program. A pre-announced exchange-rate strategy constituted a novel element in an otherwise rather standard program of inflation stabilization. The new exchange-rate regime envisaged a gradual and smooth transition to a flexible exchange rate after the critical first 18-month period.

The program was based on targeting an exchange rate, pre-announcing the exchange-rate depreciation, and ensuring that the Central Bank operated in a fashion compatible with the pegging of the exchange rate — effectively operating as a currency board. The Turkish lira was allowed to depreciate gradually until 2003 against a currency basket made up of the U.S. dollar and the euro.

The stand-by also aimed to tackle fundamental structural problems in the key areas of taxation, privatization, banking regulation, and the reform of agricultural price-support schemes.

The program had a good start. There was a significant degree of optimism concerning the prospects for stabilization and reform in Turkey in early 2000. This optimism was the outcome of the EU's Helsinki summit, which took place during the same month as the signing of the IMF stand-by agreement. The endorsement of Turkey's candidacy for full EU membership at the summit provided a strong incentive for carrying out both political and economic reforms.

With time, the performance of the coalition government failed to match the sanguine mood of the early months of 2000. Although the coalition government displayed a unified front during its early months, it soon became apparent that the government lacked the necessary coherence and commitment to the reform process. Serious conflicts emerged during the year between the nationalist MHP and the two other partners over key aspects of economic policy. Essential parts of the reform agenda such as bank regulation, privatization, and the reform of the system of agricultural subsidies faced considerable resistance.

The formation of the main regulatory agency, the Bank Regulation and Supervision Authority (BRSA), constituted a key component of the program. The board of the BRSA had to be named by the end of March 2000 so that it could become fully operational by late August, a structural benchmark. However, by the end of August, the members of the board had just been approved. Furthermore, the ability of the agency to play a regulatory role was severely restricted by the presence of private banking lobbies that resisted any kind of regulation. Bank regulation also faced resistance from politicians who perceived public banks as a serious source of rent distribution for sustaining electoral support.

Another contentious structural-reform topic was agricultural subsidies. Reduction of price supports and their replacement by direct income-support schemes were crucial components of the program, since the price supports involved were substantially above EU norms. The MHP — the second-most dominant coalition member, which drew its support primarily from the rural poor — opposed the reduction of agricultural subsidies.

Approaching the autumn of 2000, the program failed to inspire enough confidence on the part of market participants, even in the presence of IMF support. The structural benchmarks were being missed, and structural-reform packages were not being put into effect. The absence of a leader who seemed truly committed to the program was also an important factor. The governor of the Central Bank was pushed to the front, having to defend the program in public; the coalition government seemed reluctant to display strong ownership of the program.

This weak commitment, in turn, undermined the credibility of the program. The rising current-account deficit created a further vulnerability regarding the sustainability of the program.

External factors also rendered the task of policy makers more complex. The international financial environment had become much more volatile, and investors were far more risk-averse following the Asian crisis of 1997 and Argentina's situation. Turkey suffered from higher energy prices, a higher euro, and higher interest rates on external borrowing in 2000.

When confidence in the reform program started to decline in fall 2000, the markets waited for a signal. It finally came from the banking sector.¹⁶ In November 2000, the BRSA asked the banks to reduce their open positions in order to ensure that they were within the legal limits by the end of the year. The deteriorating sentiment led to an increase in liquidity pressure and exposed the weaknesses of the few banks that had positioned themselves aggressively in the fixed-income market and relied heavily on overnight borrowing. The liquidity problem resulted in a crisis. In the days that followed, the overnight-market interest rate soared to over 2,000 percent. 17 The capital outflow stopped only on December 6, when an IMF package amounting to over $15 billion was announced.

The revised agreement with the IMF and the Supplemental Reserve Facility for $15 billion replenished reserves and tranquilized the markets for a while. In the aftermath of the November crisis, confidence in the stabilization program weakened, despite the strong support of the IMF. In December and January, this weakened confidence manifested itself in the very short-term basis (mostly overnight) of capital flows and demands for higher interest.

In February 2001, Turkey was still in the zone of vulnerability. Thus, an exogenous shock in the form of an unexpected event could easily trigger a crisis. This event came on February 19, when Prime Minister Ecevit disclosed that he had had a serious dispute with President Sezer during a meeting of the National Security Council and that the country was in the midst of a "political crisis." The dispute was related to the president's concerns about bank supervision and the accusation that the Ecevit government was not undertaking the necessary measures to fight corruption among public officials. The announcement was immediately perceived in the financial markets as a sign that the ruling coalition — and hence the stabilization program — could be falling apart. In the two subsequent days, a major speculative attack on the lira took place. The sudden and massive demand for foreign exchange caused a tremendous liquidity squeeze, as banks rushed to the interbank market to raise liras to buy foreign exchange from the Central Bank. Overnight interest rates skyrocketed to over 4,000 percent. When two public banks were unable to meet their lira obligations to other banks on February 21, the interbank payments system ceased to function altogether. Thus, on February 23, the government announced the flotation of the lira, bringing the exchange-rate-based stabilization program to an early end.

STRONG ECONOMIC PROGRAM

February 2001 marked the deepest economic crisis ever faced by Turkey. The GNP in real terms declined by 9.4 percent during the course of the year; per capita income dropped from $2,986 to $2,110 per annum; unemployment increased by 1 million people.

The government stayed in power, however. The Central Bank governor and the undersecretary of the Treasury resigned, but all members of the government remained in their seats. In an effort to increase its credibility, the government brought in an external actor.

That person was Kemal Dervis of the World Bank, who was appointed minister of economic affairs and assigned responsibility for defusing the crisis. Dervis was originally offered the role of the Central Bank governor but declined, arguing that "only a minister of economic affairs with strong coordinating powers would have any chance of succeeding."¹⁸ Dervis was offered the position of head of the Treasury, as well as coordinating responsibilities for the Central Bank and BRSA and negotiating with the IMF and World Bank. But no authority over the ministries of finance, planning, foreign trade or privatization was assigned to Dervis; each coalition party still wanted to retain some control over economic management.

Dervis formed a core team including the new governor of the Central Bank, Sureyya Serdengecti; the undersecretary of the Treasury, Faik Oztrak; and the new head of the Banking Regulation and Supervision Agency, Engin Akcakoca. The Dervis team prepared a new national program based on shock-and-awe fiscal policy and front-loaded structural reforms. On May 15, Dervis submitted a new letter of intent to the IMF and announced the "Strong Economy Program." Essentially, the 1999 stand-by agreement remained intact except for the fixed-exchange-rate component. However, the program clearly aimed at implementing a series of "structural reforms" and transforming "the old ways of policy making."¹⁹ Indeed, the reform process initiated by the 1999 program significantly accelerated during the post-crisis setting.

In January 2002, the eighteenth stand-by agreement was signed with the IMF, raising the previous $9.4 billion IMF credit to $19 billion, to cover the 2002-04 period.

With the intense efforts of Dervis's team, the same coalition government responsible for the outbreak of two crises played an instrumental role in the passing of a record number of laws through parliament. In less than a year, parliament passed 19 important structural-reform laws or regulations. The most important of these were a law granting full independence to the Central Bank, a banking law,²⁰ a complete reorganization of state banks (including substantial downscaling), an overhaul of agricultural policies (moving from distorting price supports to direct-income support), a civil-aviation law, a telecommunications law, a tobacco law, a law on sugar-industry regulation, a public-procurement law, and a law on public-debt management. More than half of the 19 new laws were passed in the four months following the February crisis.

Nevertheless, resistance to reform had not disappeared altogether in the aftermath of the February crisis. Major conflicts occurred between Dervis and the MHP over a number of key reform issues. One striking example of this kind of conflict occurred in the context of reforming the Telekom board, one of the main conditions of the stand-by program. The MHP was determined to maintain its control over the company and insisted on choosing four of the seven board members. The major disagreement between Dervis and the transportation minister, Enis Oksuz, led to a delay in the disbursement of the IMF loan. The Fund was decisive in resolving the issue, however. The board was reappointed two weeks later, and the episode ended with the resignation of Oksuz. The commitment of the IMF, on both financial assistance and program supervision, was instrumental in breaking the resistance to reform.

A factor that explains the ability of Turkey to implement a comprehensive adjustment program was the widely held belief that every other alternative had been tried and discredited. There was a close correspondence between the views of the Turkish government and its economic-policy makers and the IMF staff. The program was not seen as being imposed by external forces against the will of the government; on the contrary, the Turkish authorities were highly successful in presenting the program as having been formulated by them. As a result, the policy package was successfully internalized. Dervis resigned from the government in August 2002, due to increasing strains within the coalition, but implementation of the program stayed in place despite a looming general election in November 2002.

THE NEW GOVERNMENT

The process of reform implementation took a new turn following the victory of the Justice and the Development Party (AKP) in the general elections of November 2002. The political sea change reflected the deep impact of the economic crises. Members of the coalition government were penalized by the electorate for their poor economic performance. Consequently, the dominant political parties of 1990s were forced to quit the policy arena, paving the way for the emergence of a new political party, the AKP, a party with Islamist roots. The November 2002 elections were also important in putting an end to an era of political fragmentation and successive coalition governments, which had been associated with severe economic and political instability since 1991.

In the early days of the government, there were mixed feelings concerning the electoral success of the AKP. The Islamist roots of the new party and its lack of experience in office were causes for concern. There were also fears that, given its broad-based electoral support, the new government might give in to populist expansionism and deviate from the IMF-supported program. The AKP government, to the surprise of many, has adopted a strong reformist policy orientation. Capitalizing on its broad-based political support, the new government was able to accelerate the pace of economic and political reforms, with mutually reinforcing consequences. The "Strong Economy Program," put into practice in the aftermath of the 2001 crisis, has continued to be implemented without interruption. The outcome, judged by the principal macroeconomic indicators, has been impressive.

During the 2002-06 period, Turkey found itself in a kind of virtuous cycle, able to accomplish tight fiscal discipline. Inflation, for the first time in over three decades, has been stabilized at single digits. This has been accompanied by a strong surge in economic growth. The debt-to-GNP ratio has been falling steadily from the record level it reached during the crisis period, due to a combination of sharp fiscal adjustment and high rates of economic growth, averaging 7.2 percent during this period.

The government was particularly committed to fiscal discipline. The main policy pillar was achieving a 6 percent primary balance. This was supported by a decrease in current expenditures of the government, an increase in taxes, a restructuring of social-security institutions, structural changes in the government budget and a closure of extra-budgetary funds. In addition, the government has been able to continue key institutional reforms such as banking-sector regulation.

Having argued that macroeconomic performance during its early term of office has been broadly successful, it would be unjust to attribute this success exclusively to the AKP. The Turkish economy had been undergoing a series of reform initiatives. The government capitalized on some of the important ones introduced by the previous government. The AKP government, on the other hand, exhibited a stronger commitment to the program. The fact that a majority government was in power, as opposed to a coalition of different parties, certainly helped to maintain this commitment. The novelty of the AKP in government after the 2002 elections also contributed to program implementation. According to Hugh Bredenkamp, the IMF's resident representative to Turkey during the AKP's early years, the fact that the AKP government comprised a totally new cast of characters made a positive difference:

There were people who had never been in government before. The party itself had never been in government. Therefore, they brought a refreshing outlook. They had the willingness to take what was working and to stick with it. They did not assume that they knew better, and did not change everything around.²¹

Bredenkamp also argued that the previous government's legacy helped the AKP to continue with good program implementation. It also helped that, as they were implementing these policies, good performance was being recorded. So good implementation coincided with the bounce-back phase, which in turn reinforced further implementation. In Bredenkamp's view, the AKP government knew that a healthy economy was their biggest asset. They were not going to endanger it by taking risks: "The only thing they could do is to not to mess things up, and not allow the economy to deteriorate." So they left the economy on autopilot.

When the eighteenth stand-by agreement ended in 2005, the government requested a new IMF program on the grounds that the consecutive agreement would provide an anchor for policy, encourage capital flows and help dampen exchange-rate fluctuations. The nineteenth stand-by thus became effective for a three-year period from May 2005 to May 2008.

The 2001 crisis presented many opportunities for the AKP. Not only did it create the basis for economic reforms the party could then capitalize on; it also led to the demise of the established parties of the coalition era. With no serious opposition group left in the political arena, the AKP had the formidable position of continuing uninterrupted with its reform efforts.

The AKP commitment to both economic and political reforms paid off. While previous policies of center-right parties resulted in boom-bust cycles and high inflation, the AKP era was characterized by low inflation and high economic growth. A favorable external environment during the early years of AKP rule also contributed to this result. The global liquidity boom provided a major boost to growth in all emerging markets, from which Turkey has also benefited. It also helped to disguise some of the structural weaknesses of the Turkish economy, such as low domestic savings, large current-account deficits and high unemployment. The rate of employment did not rise throughout the high-growth period of 2002-07. The persistently high unemployment rates of around 10 percent led some critics to categorize this period as one of "jobless growth."

GRADUATION

By 2007-08, the fragility of growth based largely on external inflows became increasingly evident, even before the onset of the global financial crisis in the third quarter of 2008. Rendering these fragilities all the more significant was the decline in the government's reformist appetite. Content with the policies put in place in 2002, the party's attention from 2006 onward shifted toward consolidating its power. The policy makers remained committed to critical elements of the program, such as privatization and FDI initiatives, and did not interfere with monetary policy and banking regulation. However, there were reform reversals in other policy areas. The public-expenditure regime was diluted significantly with amendments to a 2003 law. The agricultural-subsidy reform of 2001-02 was overturned by 2007, and the long-due pension reform of 2008 only introduced minor changes to the existing overburdened system. Looking at the second phase of AKP rule, it seems that the single-party versus coalition-government distinction loses some of its strength in explaining the governments' reformist stance.

The AKP went on to win the 2007 elections, increasing its support from 34 percent in 2002 to 47 percent in 2007. The party's electoral success has been firmly rooted in the strong economic recovery of the post-2001 period.

Nevertheless, despite a rather positive economic outlook, rising domestic political instability and ideological polarization have put a dent in the optimism created in the previous episode. Severe political tensions sprang up in this period. During the row over the presidency in 2007, the party clashed with Turkey's military-backed secular establishment. Although the party increased its votes considerably in the July 2007 elections, by spring 2008 it faced a Constitutional Court case demanding its closure for anti-secular activities. When the economy entered recession in late 2008, the party had already been forced to walk a much finer political line than a few years before.

With this overloaded political agenda, the government started negotiations with the IMF. In the months following the expiration of the last stand-by, it was becoming clear that the global economic crisis was reaching Turkey's shores. The AKP government was expected to sign a new IMF stand-by loan agreement after the expiration of the nineteenth in May 2008. The argument was the same: the IMF program would be an anchor for policy and a safeguard against the global crisis. Turkish policy makers did hold extended negotiations with the IMF. The government on several occasions gave the impression that an agreement was in the works, only to retreat a few weeks later, stating that the disagreements with the Fund remained unsolved. Constant "talk" of an IMF deal certainly served as a pseudo anchor at a time of global uncertainty and calmed markets. However, negotiations were deadlocked in January and finally collapsed. This move was once again publicized as a sign of national strength and autonomy.

The rhetoric of sovereignty aside, there were significant disagreements between the IMF and the government. Fiscal allocations to municipalities were an issue of dispute. With looming local elections, cutbacks on municipal resources were not palatable to the AKP. The second issue was reform of the tax administration, which had been on the table since 2003. The IMF wanted the tax administration to be restructured as an autonomous agency to adequately address Turkey's narrow-tax-base problem. This was another disagreeable suggestion for the AKP, whose electoral base included a wide spectrum of small and medium-sized entrepreneurs, the self-employed and rural manufacturers — the prime beneficiaries of a lax revenue regime.

With no official assistance, the government weathered the global crisis alone. The crisis impacted Turkey through a fall in capital inflows as well as a dramatic loss in export revenues, the latter being the result mainly of a demand shock from the European Union. The fact that Turkey did not experience a single bank failure in the aftermath of the global financial crisis helped to boost the confidence of policy makers that the Turkish economy was in strong shape and would be marginally affected by the crisis.

GDP growth started to decline markedly in mid-2007. From an average of 7.2 percent during 2002-06, growth was down to 4.7 percent in 2007 and to 0.7 percent in 2008. In 2009, the growth rate plunged to -4.8 percent. Reversals of capital flows contributed largely to the record declines in growth during 2007-09.

On the fiscal front, the government managed to attain the primary-surplus target of 5 percent until 2007. The year 2007, however, was an election year; growth was on the decline, and non-interest expenditures increased in spite of stagnant revenues. The result was that a 5 percent primary surplus could not be attained. After some improvement in the first half of 2008, there was further deterioration through early 2009, as the government started a spending spree for the local elections of March 2009. Thus, fiscal policy was already loosened as the global crisis was reaching Turkey. In late 2008 and from March 2009 onwards, several fiscal-stimulus packages were announced that further degraded fiscal balances. Major reductions were made in consumption-tax rates and social-security premiums. A tax amnesty for all unrecorded assets and tax exemptions for incomes earned abroad were introduced. A comprehensive stimulus package was also announced to promote investment. It encompassed corporate and income-tax reductions, interest-rate subsidies, payment of workers' social-security premiums and allocation of investment sites. The estimated total burden on the budget of the fiscal measures designed to stimulate the economy amounted to 0.8 percent, 2.1 percent, and 1.6 percent of GDP, respectively, in 2008, 2009 and 2010.²² Government measures in response to the global crisis were distinctly market friendly and exceptionally weak on social protection and infrastructure investment. Turkey's fiscal plan had the lowest score for social protection in a sample of 35 economies, with only 1.5 percent of its stimulus funds assigned to social spending.²³

Nonetheless, with the economy recovering swiftly, registering growth rates of 9.2 percent and 8.5 percent in 2010 and 2011, respectively, the AKP entered the general elections of 2011 in much stronger shape than in July 2007. The 2011 general elections presented another victory to the AKP, increasing its electoral base to 50 percent. The party occupied a central position in Turkish politics, as well as strengthening its hold on key segments of the state and the bureaucracy.

CONCLUSIONS

Turkey's development path illuminates the difficulties of accomplishing large-scale economic transformation in an emerging democracy. Significant transformation has occurred over the past few decades under IMF-assisted programs, but this was a costly and crisis-ridden project. The country has experienced three costly financial crises along the way, in 1994, 2000 and 2001, and yet it has been the most devoted client of the IMF.

The 2001 crisis was particularly instrumental, paving the way for a series of regulatory reforms long-awaited by the IMF. The crisis served as a catalyst for change and created a broad-based coalition for reform.

IMF-backed reform programs have helped Turkey to become more robust and resistant to shocks. The Turkish economy appears to be in stronger shape than a decade ago, following an important wave of regulatory reforms. Turkey compares favorably to major emerging market economies such as India, and its growth in 2010 and 2011 approached that of China. At the same time, overly optimistic assessments of Turkey's economic performance during this period need to be qualified: the economy continues to display a number of structural weaknesses that naturally limit its ability to achieve stable and sustainable growth. The global crisis, like the earlier domestic ones, is a reminder that Turkey cannot achieve high rates of economic growth on a sustained basis by relying on capital inflows. In this regard, the policy makers and the IMF have not addressed the longer-term structural problems of the economy.