The Era of Speculation-Led Growth and the 2001 Crisis, 1990–2001

Abstract

The 1990s can be termed as the ‘lost decade’ for the Turkish economy. It opened up by the imposition of full capital account deregulation and completion of financial liberalization; and had seen two crises in 1994 and 2001. It was one of extreme volatility, characterized by narrowing of the time horizon. The economy was trapped in a vicious cycle wherein volatility of the flow of foreign finance lured by high domestic interest rates led to deeper adjustments in the exchange rate, which, in turn, resulted in severer turbulences in the traded sectors, and warranted ever more inflows of foreign capital, leading to higher volatilities in external finances. Consequently, mechanisms of income distribution had to be readjusted, new forms of rent-seeking were enacted, and new coalitions had been formed across the bureaucracy, industrial conglomerates, and the banking sector. The banking sector came to the forefront of financial speculation; and the main macroeconomic prices –rates of interest and foreign exchange– were re-structured under a new set of equilibrium relationships. All this processes generated severe repercussions for the real economy, as wage-labor bore the brunt of adjustments.

Introduction: Setting the Stage

The 1990s can be termed as the “lost decade” for the Turkish economy. The stage was set by the completion of “external liberalization” in August 1989 with the announcement of the Decree No. 32, which opened up the capital account of the balance of payments and gave rise to a whole set of new modes of macroeconomic adjustments for the domestic economy. Perhaps in its entire history, Turkey suddenly confronted a new era in which the “constraints” of the “external gap” was eliminated and the domestic economy met with a new instrument: the real interest rate. Short-term financial flows, lured by the arbitrage opportunities of a new emerging market, seemed to have alleviated the external imbalances once and for all. This new process meant significant reallocation of investments, along with realignment of the main macroeconomic prices, namely, the rate of interest and the exchange rate. There was also significant reorientation of the parameters of distribution, as the sphere of finance commenced to ascend against the “real sectors” in general, and industrial labor in particular.

The decade was one of extreme volatility, characterized by narrowing of the time horizon. Volatility of the flow of foreign finance led to deeper adjustments in the exchange rate, which, in turn, resulted in severer turbulences in the traded sectors, and warranted ever more inflows of foreign capital, leading to higher volatilities in the external finances. Consequently, mechanisms of income distribution had to be readjusted, new forms of rent-seeking were enacted, and new coalitions had been formed across the bureaucracy, industrial conglomerates, and the banking sector.

This chapter accounts for these observations. It is organized under four additional sections. The first section following this Introduction analyzes the main elements of change in the global commodity and finance markets, elaborating the rise of speculative growth and the new nature of the business cycle. Then, the following section discusses Turkey’s mode of adjustment to the new global order and documents the context of capital account liberalization in 1989, studying also the effect of this move on the “real economy.” The section before the Concluding Comments focuses on the elements of the IMF-induced, exchange-rate-based disinflation program that was initiated at a time when the decade came to a close by setting the stage for the eruption of one of the most severe crises of Turkish economic history in November 2000 and February 2001. The last section concludes with an overview and discussion over new forms of dependency across the developing world.

The Changing Global Context: “Financialization” of the Third World

The 1990s can be understood as a case of “financialization redux” at the level of the global economy. The decade opened up with trumpets echoing “the end of history” à la Francis Fukuyama. The Soviet system of “real socialism” collapsed, and all the global markets, with the exception of labor, were started to be integrated under one logic: free mobility of capital—especially “finance capital.” Removal of barriers over international finance has granted it with extensive deregulation, while labor was trapped within national borders.

Deregulation of finance capital was in the making since the early 1970s. Limits of the “golden age” of Bretton-Woods era were already reached with the decline in the rate of profits in the manufacturing industries of the developed countries. Technologies matured, capital intensities increased, and easy gains in productivity along the assembly line and the scale economies were exhausted. These developments led to the collapse of the gold-exchange standard in 1971, and along with excess accumulation of petrodollars and pension funds in the hands of few western banks searching for lucrative speculative deals, regulations on the mobility of financial capital could not have been sustained any more. Financial deregulation meant dismantling the rules and interventions of the nation states. Any regulation inhibiting the quest for financial profit across the globe was cursed as backwardness. Furthermore, along with a severe and sudden reorientation of priorities of capital accumulation away from industry to finance and banking, the process of deindustrialization has intensified and industrial labor has been caught within the confines of short-term speculative capital flows under national constraints.

All of this was referred to as globalization, which was hailed as an unstoppable planetary motion toward global “civilization.” Thereby we can deduce three interlinked aspects of global capitalism in the juncture of the 1990s: neoliberal restructuring, neoliberal globalization, and financialization. Neoliberal restructuring had been propagated with the counter attacks of monetarism and supply-side economics during the 1980s in the hands of Ronald Reagan in the USA, Helmut Kohl in Germany, Margaret Thatcher in the UK, and Turgut Özal in Turkey. The assault reached its zenith in the 1990s with the rhetoric of “the end of history,” when all political-economic questions were declared to be resolved, all unknowns were behind, and the world was on a sustained path toward global bliss. The states would now allegedly assume the role of a bystander, a referee, setting the rules of the game and ensuring that the rules were obeyed.

This idea of a neutral state, standing at an equal distance from all participants in the workplace, however, was far from reality. The state apparatus has, in fact, was reorganized to ensure the supremacy of capital over labor; and any dissent was brutally suppressed with accusations of backwardness and/or outright military force against labor organizations. The neoliberal state was actually a stronger state, given its new instruments of control over society, such as the newly formed regulatory bodies, committees, task forces that typically consisted of a handful of “technicians” and were most often enacted with powers outside the parliamentary jurisdiction.

What lied at the heart of this restructuring was the ascendancy of finance over industry, characterized by a global process of financialization imposing its logic of short-termism, liquidity, flexibility and immense mobility over the objectives of long-term industrialization, sustainable development and poverty alleviation with social welfare states. Financialization, as it stands, is a loose term and no consensus yet exists among economists on its definition. However, starting from David Harvey’s seminal observation that “something significant has changed in the way capitalism has been working since about 1970” (Harvey 1989: 192), a set of distinguishing characteristics of the concept can be unveiled. Krippner (2006: 174), in line with Giovanni Arrighi’s The Long Twentieth Century, defines it as a pattern of accumulation in which profits accrue primarily through financial channels rather than through trade and commodity production. According to Epstein (2005: 3), “financialization means the increasing role of financial motives, financial markets, financial actors and financial institutions in the operation of domestic and international economies.” In a broader way, we can consider “financialization” as a phenomenon that can be described as increasing financial motives, and increasing volume and impact of financial activities within and among countries. As Duménil and Lévy underline:

What is at issue here, are not markets and states per se, but the stricter subjection of these institutions to capital: on the one hand, the freedom of capital to act along its own interests with little consideration for salaried workers and the large masses of the world population, and, on the other hand, a state dedicated to the enforcement of this new social order and the confrontation to other states. (Duménil & Lévy, 2004: 3)

Over the decade, waves of financial crises were witnessed. The first of these waves typically erupted in the “emerging market” economies of Mexico in 1994, Turkey in 1994 and then again in 2001, Brazil and Russia in 1998, Argentina in 2001, and of course, the “Asian Flu” of 1997. Almost all of these financial crises were explained, one way or another, by a form of moral hazard—lack of “prudential” regulation and biased incentives emanating from the assumption that the risk-takers were too large to fail.

Under this new episode, crises erupted mainly due to premature financial liberalization, lack of governance, lack of rule of law, etc. Typically, countries, which were lured into the trumpets of “end the financial repression; hail to the free financial markets” (à la McKinnon, 1973; Shaw, 1973), liberalized their financial sectors too prematurely and too hastily without paying attention to their macroeconomic fundamentals. In these economies, capital-account deregulation often led to increased interest rates. Based on the motive to combat the “fear of capital flight,” this commitment stimulated further foreign inflows, and the domestic currencies appreciated inviting an even higher level of short-term capital and “hot money” inflows into the often shallow domestic financial markets.

The experience of the 1990s was thus a new global order of instability. One can trace out the main mechanics of this instability as follows: with (prematurely) opening up of the capital account, short-term foreign finance (hot money inflows) pour in with an attempt to take advantage of the speculative financial arbitrage opportunities. Currency appreciates, import costs fall and the domestic agents enjoy a sudden relaxation of their budget constraints. The initial bonanza of debt-financed public spending (e.g. Turkey) or private spending (e.g. Mexico, South Korea) escalates rapidly and worsens the fragility of the shallow domestic financial markets. Eventually, the bubble bursts out and a series of severe and onerous macroeconomic adjustments are enacted through very high real interest rates, sizable devaluations, and a harsh entrenchment of aggregate demand accompanied by the short-term “hot money” outflows. Elements of this vicious cycle are further studied by Adelman and Yeldan (2000), Calvo and Vegh (1999), Dornbsuch et al. (1995), and Diaz-Alejandro (1985). This cycle is more recently referred to as the Diaz-Alejandro-Taylor cycle in Köse et al. (2007), following Diaz-Alejandro (1985) and Taylor (1998). A schema of this cycle is portrayed in Fig. 8.1.

Fig. 8.1

The Alejandro-Taylor cycle: vicious cycle of capital flows & macroeconomic disequilibria

Figure 8.1 discloses main features of what I will term as the Alejandro-Taylor Cycle. As this is a closed system, one can initially start from any point of this cycle. Suppose that given the threat of capital flight, or any lucrative expectation of short-term gain from capital inflows, the domestic rate of interest is increased. Speculative arbitrageurs storm in, given the absence of any regulation and the currency appreciates due to the bonanza of foreign exchange. Imports expand, current account widens, and most probably foreign indebtedness rise as well. All these mean increased external fragility and thereupon a more intensified interest hike is warranted. The cycle recommences, as the country is set into a trap of high interest rates, appreciating currency and collapse of the domestic import-competing industries. The end result is an unsustainable path of speculation-led growth.

Historical evidence suggests that the main characteristics of this variety of crises typically involved the following:

• i. International capital market has been the major source of shocks;

• ii. Flows have largely originated from and been received by the private sector;

• iii. The financial crises have mostly hit emerging market economies that were considered to be highly credible and successful;

• iv. The rise of capital inflows has been characterized by a lack of regulation, on both the supply and the demand sides.

Under these conditions, many developing countries have suffered from premature de-industrialization, serious informalization, and consequent worsening of the position of wage-labor, resulting in a deterioration of income distribution and increased poverty. Many of these phenomena have occurred under the neoliberal “conditionalities,” imposing rapid liberalization of trade, privatization of public enterprises, and premature deregulation of the indigenous financial markets. Thus, across all economies, industrialized or peripheral, wage incomes collapsed; income share of wage labor in aggregate domestic product fell; and the appropriated surpluses fed the rising corporate profits. Turkey’s experience with capital account liberalization and financial deregulation over the 1990s disclosed almost all of these key attributes. It is to these issues that I now turn attention.

The Decade of Speculation-Led Growth

The decade of the 1980s was marked by the reorientation of the Turkish economy to integrate with the global markets. A series of reforms and structural adjustment conditionalities were enacted, resulting in tariff liberalization, export promotion and a severe wage repression for labor incomes. The economy, however, entered a period of reform fatigue by 1988 and slowed down significantly in 1988. Realizing that the “fruits” of export promotion and globalization were “delayed,” Özal government initiated the liberalization of the capital account in order to access international finance capital.

The Decree No. 32 was the main policy document leading to the full liberalization of the capital account. In a nutshell, it covered the following:

• All the residents of Turkey, including private persons, corporations and banks can bring and take out foreign exchange in any magnitude to and from Turkey without any restrictions.

• Nonresidents can purchase any form and quantity of assets from Turkey, bring in and take out their yields in any form of denomination, Turkish Lira or foreign exchange.

• Residents are free to introduce any type of assets to be sold domestically as well as abroad, and free to transfer the returns in and out freely.

• Nonresidents are free to bring in any foreign credit, or purchase domestic credit from within and transfer monies in any denomination to and from Turkey.

Historically speaking, the elements of this maneuver were quite liberal, even more so than the advanced economies of Europe at the time, and Turkey was severely criticized for not creating the necessary institutional infrastructure to oversee the flow of funds in a very narrow domestic financial system. Main indicators of this episode involved mostly a new division of responsibilities and rent-seeking opportunities among the banking sector, the industrial bourgeoisie and the state. First of all, it was clear that, as a ratio of GDP, Turkey indeed experienced significant financial deepening. Securities issues in total rose from 6.5% of the GDP in 1990 to 40% by the end of the decade (Yeldan, 2001). Time deposits also rose by almost two-folds as a ratio, taking advantage of the increased rate of return, that is, the real rate of interest. Banking sector credits and the volume of transactions both in the primary and secondary financial markets expanded feverishly.

Nevertheless, much of this transition relied on mainly two factors: issuance of Government Debt Instruments (GDIs) and significant dollarization of domestic deposits. In fact, financial deepening can be argued to have led the residents to switch to foreign currencies (dollarization), paving the way for a new form of deficit financing by the government.

With the advent of financial liberalization, Turkey experienced, perhaps for the first time in its entire republican history, a substantial alleviation of the foreign-exchange constraint. The foreign-exchange scarcity disappeared, and Turkish credit and money markets experienced a sizeable inflow of foreign exchange within a few months’ time, releasing all concerns about the external deficit.

A direct effect of this process was the onerous adjustments forced by the so-called open-economy trilemma, according to which, in an open economy, only two out of the following three can be chosen and implemented by the authorities: independent monetary policy (conduct of money supply), the foreign exchange regime (free float or fixed exchange rates) and the capital account regime (open or closed). Yet, in a developing, emerging market economy such as Turkey, what actually happened with an open capital account was that Turkey could have control over neither monetary policy, nor the foreign exchange regime. Simply put, the rate of interest and the exchange rate collapsed into a single price and constituted the main operational indicator for the inflows and outflows of foreign exchange—the “hot” component of foreign finance. Thus, the impossible trilemma had been observed to work even under more stringent conditions where the emerging market economies that had opened up their capital account to international flows of finance, had actually lost their control over both the independent monetary policy and the foreign capital mobility. The implications were a structurally constrained economy to yield ever increasing real rates of interest and monetary deflation under the threatening conditionalities of international speculative finance.

Table 8.1 introduces the critical economic indicators of this transition. Here, along with the pre-1989 period, the “lost decade” of the 1990s is divided into four main episodes: uncontrolled financial liberalization (1989–1993); the 1994 crisis; return to “hot money” driven growth (1995–1997); and the contagion of the Asian crisis (1998 and 1999). All of these episodes can be contrasted against the export-orientation era of the 1980s. The first of these episodes (1989–1993) is characterized by ad hoc and often politically-motivated interventions aiming at deregulation of the financial asset markets. As noted above, the Decree No. 32 had been introduced with an eye on foreign “hot money” inflows. The unavoidable home-currency appreciation under this foreign-exchange boom led to the 1994 crisis, the first full-fledged financial-cum-real crisis in Turkey. After 1994, domestic economic policies were realigned for hot-money driven, speculation-led growth. Yet, the contagion of the Asian crisis hit the Turkish economy under these prolonged structural imbalances. This episode was finally cut by the introduction of the IMF-led disinflation program under the exchange rate-based tablita.¹ The end result would be the November 2000 and February 2001 crises, which are narrated in detail below.

Table 8.1 Macroeconomic adjustment processes: Turkey, 1983–1999

Reading from Table 8.1, recovery of GDP growth from the 1988 deceleration is clearly visible. Fueled by inflows of short-term foreign finance, investment expenditures continued on their expansionary path; yet, as discussed in detail by Yeldan (2001), their share as a ratio to the GDP has not revealed a structural shift. One of the main reasons of this was the switching of destined investments away from industry, to one-time expenditures such as construction and housing (Boratav and Yeldan, 2006; Yeldan, 2001). Both of these were non-traded sectors and led to the widening of the current-account deficit by 1993 and reaching its climax in 1994 as the balance of payments crisis exploded.

The adjustment experience of the real sector to financial liberalization had been one of boom-and-bust cycles. As documented in Table 8.1, the post-1988 performance of GDP revealed intensified short-term business cycles, along with rates of annual growth ranging between 8% (1993) and −5.5% (1994). Following the production cycle, both consumption and investment demand fluctuated sharply over the same period. Similarly, the external economy was in turbulence with the balance on current account suffering from severe fluctuations between US $ −6.4 billion in 1993 and US $ 2.6 billion in 1994, and again US $ −2.3 billion in 1995. Domestic rate of inflation reached the plateau of 70–80% per annum and displayed strong resistance at this threshold.

In fact, inflation could never be kept under control. Hovering around the plateau of 70–80% per annum caused serious appreciation of the real exchange rate (see Fig. 8.2). Indeed, in the early decade, Turkish Lira appreciated significantly, by as much as 15% per annum. Such appreciation meant worsening of the current account balance together with the deceleration of the export revenues. Exports as a share of GDP dwindled to less than 10% from its peak of 12.8% in the late 1980s. Import expansion continued in an intensified manner over the whole decade.

Fig. 8.2

(Data source Annual reports of the Central Bank of the Republic of Turkey [Author’s calculation])

Real exchange rate index (TL/USD), purchasing power parity (PPP) in consumer prices

Early years of the “lost decade” also witnessed real increases in manufacturing wages. This was the end of a period of secular decline over the 1980s. Led by the “spring uprisings” of the late 1980s, wage remunerations of industrial labor increased at an annual rate of 10% in real terms, halting back the losses of the Özal decade. Yet, all of this increase in real wages would be taken back in 1994 with the eruption of the financial crisis, along with the rapid escalation of inflation and depreciation of the Turkish Lira. Wage data are scarce and often not reliable in Turkey, especially for the period considered. In what follows, wage data in Table 8.1 is limited only to the employees in the manufacturing sector through the official data of the Turkstat (then the State Institute of Statistics). The relative position of wage labor in the global realm can further be examined through the unit wage costs denominated in foreign currency. Such costs are calculated by taking into account the (average) productivity of labor, deflated by the exchange rate. In this way, unit wage costs account for the degree of competitiveness of the industrial sector across the global economy. Adjusting for the competitive devaluations of 1994 and 2001, the unit wage costs remained roughly 25% lower than their value in 1993. As Table 8.1 attests, manufacturing industry wage share in value added receded to 16.1% as a result of the wage suppression led by the devaluation of 1994. From 1994 onward, this share stayed more or less stable around this rate up to the end of the decade.

The impact of financial liberalization had been sudden and deep. Theoretical expectations of this maneuver were deepening of the financial system and thus to achieve a higher savings ratio supporting fixed investments. Financial deepening, as measured by the ratio of financial assets to the gross domestic product, would be the key element of this transition.

Data reveals that such a deepening did in fact occur. As a ratio to GDP, total financial securities expanded, for instance, from a ratio (to the GDP) of 7.8% in 1988 to 24.8% in 1994 (Balkan & Yeldan, 2002; Boratav and Yeldan, 2006; Yeldan, 2001). However, this increase was predominantly explained by securities issued by the public sector to cover its expanding fiscal deficits. Public securities issued rose from 6.9% in 1988 to 22.7% in 1994 to reach 38.7% in 1999; whereas private sector securities issued stayed at 2.1% of the GDP in 1994, rising only marginally from its miniscule level of 0.9% in 1988. By the end of the decade the share of private securities had fallen to 1.1%.

Total deposits, likewise, expanded. The ratio of total deposits was 15.7% in 1988 and reached 39.5% by 1999. Again, this was problematic since the major expansion came from foreign-exchange deposits, as their ratio rose from 4.2% in 1988 to 22.4%. This was mainly due to agents’ preferences for dollarization, in an attempt to protect against the inflationary losses. As credibility of the Turkish Lira was lost, economic agents tried to protect their assets by shifting into dollar-denominated deposits. In fact, banking sector credits to the enterprise sector fell, as a ratio to the GDP. These stood at 17.6% in 1988; and fell to 13.3% in 1994, averaging around 18% over the remaining years of the decade (Boratav and Yeldan, 2006).

All of these were contrary to the expectations of financial deepening. The realization of financial deepening meant a new round of formation of coalitions in the Turkish socio-economic structure. The government continued to run fiscal deficits as was dictated by high interest costs under the high interest rate trap enforced by the threat of capital out-flight. The banking sector, on the other used this opportunity to borrow cheap abroad and extend these foreign monies as domestic credit to the government sector. The high interest burden unavoidably led to expansion of the “public sector borrowing requirement” (PSBR) as a result of the high interest burden. The government debt instruments (GDIs) were critical in financing the budget deficit of the public sector (the central budget as well as the state economic enterprises, including the social security administration deficits). The stock of securitized domestic debt grew rapidly and the stock of GDIs reached 22% in 1994 from 6% of the GDP in 1989. Interest costs on domestic debt grew to 10.6% of the GDP by 1994, and then continued viciously to increase by almost ten-fold in real terms over the decade. As a further comparison, interest costs on servicing the debt reached 1,010% of public investments, and 481% of the transfers accruing to social security institutions in 1998 (Balkan & Yeldan, 2002).

As these were being realized in the government accounts, the banking sector was lured by the real interest rate exceeding 30% per annum, and evolved into arbitrageurs of “hot money” finance. The banks continued their borrowing from abroad and channeled “hot money” flows to the public sector to cover the PSBR. In so doing, there were significant pressures to run open positions in the banking sector balance sheets (see Fig. 8.3). With the bonanza of foreign exchange, Turkish Lira appreciated (see Table 8.1) giving rise to current account deficits. These twin deficits (fiscal and external) were financed by external borrowing of the banking sector and the debt instruments of the public sector. This was a fragile environment and the bubble burst in 1994; when the interest rate rose to unprecedented levels and yet could not sustain the inflows of foreign capital as desired.

Fig. 8.3

(Data source Annual reports of the Central Bank of the Republic of Turkey)

Assets, liabilities and open position of the banking sector (%)

Thus, the episode was set with the completion of the triumvirate; the foreign capital centers would be bringing in “hot money” to the domestic banking sector, which, in turn, was channeling these to the public sector. In the meantime, the rate of return on speculative financial arbitrage would reach to almost 30% just before and immediately after the 1994 crisis (see Table 8.2).

Table 8.2 Speculative short-term foreign capital (hot money) flows and selected financial indicators (Million US$)

The instruments and consequences of this process are portrayed in Table 8.2. Measured as the ratio of the domestic interest rate (approximated by the rate of return on GDIs) to the rate of depreciation, the domestic rate of return offered to the “hot money” transactors was generally around 30% especially after the 1994 crisis. Despite its fluctuations, the size of the financial arbitrage was instrumental in the expansion of the banking sector’s short-term borrowing. The volume of inflows and outflows of banking sector foreign credits reached 122 billion US$ and 118 billion US$, respectively, in 1993, exceeding the size of the overall GDP. Given the shallowness of the domestic financial sector, this magnitude, no doubt, meant severe fluctuations for the financial transactors, creating uncertainty and high risk.

It should to be noted that one must consider the gross magnitudes of such flows rather than net amounts, because that is where the destabilizing consequences of speculative short-term capital movements prevail. In columns 2 and 3 of Table 8.2, the gross inflows and outflows of “hot money” to the domestic financial markets for the post-1990 period are reported. For the purposes of this chapter, “hot money” is identified as the foreign exchange credits brought by the banking system, so as to distinguish it from the net errors and emissions in the balance of payments statistics (which mostly account for the so-called “unrecorded” transactions).

The gross inflows grew rapidly from US $43 billion in 1991 to reach US $122 billion in 1993. After a brief deceleration during 1996 and 1998, they again reached US $122 billion in 1999. This magnitude was almost two-thirds of the size of the overall Turkish GNP. Clearly, the domestic financial system, under a severe pressure exerted by international speculative centers, was no longer in a position to conduct an independent monetary and foreign exchange policy. Furthermore, those centers constituted the major reason behind short-termism and volatility of the real business cycles, leading to increased fragility of the financial and the external position of the domestic economy and worsening of the distribution of income (Balkan & Yeldan, 1998, 2002; Yeldan, 2001). These issues are examined in more detail in the following section.

Thereby emerged a vicious circle: as the budget deficit expanded, the government had to issue GDIs with substantial returns on its securities, propelling the banking sector to bring in higher volumes of foreign credit so as to augment its indebtedness. The risk and uncertainty involved at the background, coupled with the widening trade deficit, resulted in even higher rates of interest for the government during the next round of the cycle.

The cycle was abruptly broken in 1994, when the sources of foreign finance dried up and Turkey experienced a sudden stop, perhaps the first of its kind in retrospect. The behavior of the real exchange rate turned out to be the dominant driver of macroeconomic adjustment (See Fig. 8.2).

The exchange rate was on a real depreciation trend over the 1980s. The strategy of export promotion necessitated a depreciating Turkish Lira. The success or failure of this choice left aside, the adjustments entailed by real depreciation implied contraction of wage incomes, as explained above. As also shared within the common history of late industrializers attempting to pursue an export-led industrialization strategy amidst a darkening external environment, Turkey’s strategy of export promotion over the 1980s based on intensive currency depreciation failed. Lira depreciated in real terms by as much as 45% by mid-decade; and real wages were severely suppressed to generate a domestic surplus to be exported abroad (Yeldan, 1995). Suppression of wages were also the end result of the repressive conditions of the military regime through its dismantling of the trade unions and the changes imposed in the Labor Law, banning of the right to strike and restricting collective bargaining. Yet against all this, Turkish industry failed to pick up as a structural leader of expert-led growth, and Turkey entered the 1988 deceleration referred to as the reform fatigue (Yeldan, 2001).

By 1990, however, these dynamics changed. Ensuing capital liberalization caused the Lira to start to appreciate strongly. Given high gains of arbitrage, speculative foreign hot money flew in and gave rise to the trap of high interest rates and appreciating domestic currency (cheap foreign exchange) whose dynamics were discussed above. Figure 8.2 attests that, compared to 1988, the Lira enjoyed real appreciation by as much as 40%. The 1994 devaluation reversed the trend. After then, there emerged a brief episode of stable real exchange rate. At a time of very high inflation, the central bank was successful in maintaining the competitiveness of the Lira. It is clear from the Fig. 8.2 that, over 1995–1999, the real exchange rate was almost stable. This could be mentioned as a “successful” strategy on the part of the Central Bank, which at the time of significantly high rates of inflation, could nevertheless maintain a “competitive” exchange rate by aligning the nominal value of the spot exchange rate through a series of mini-devaluations and monetary accommodation.

The high risk element of these operations was, nevertheless, the banking sector. At a time of significant appreciation in early part of the decade, the banks’ foreign exchange liabilities exceeded their foreign exchange assets, creating substantial “open positions.” The risk contained in maintaining such high rates of open positions, reaching as much as 10% of total assets by the end of the period, began to take its toll from 1994 onward. Figure 8.3 portrays the magnitudes involved.

The distribution of the open positions across the banking sector revealed that, not surprisingly, the private deposit banks were the key actors of the operations. In 1993, just before the eruption of the financial crisis, the tensions were already setting in. As a ratio to “paid capital,” banking sector’s cumulative open position was already at its peak at 178%. After a brief fall due to the 1994 crisis, the open positions were sustained at ratios approaching to 80% of paid capital, or to almost 20% of the GDP.

As such, new coalitions had been re-grouped throughout the period; the banking sector came to the forefront of financial speculation; and the main macroeconomic prices—rates of interest and foreign exchange—were restructured under a new set of equilibrium relationships. All this process generated severe repercussions for the real economy, and wage-labor bore the brunt of adjustments.

IMF’s Exchange-Rate-Based Disinflation Program and the 2001 Crisis

1998 turned out to be a crucial point in Turkey’s recent macroeconomic history. By then, it was clear to the Turkish bourgeoisie and the state that the ongoing episode of speculation-led patterns of growth driven by hot money finances was on thin ice and was too risky. Maintained over conditions of almost hyperinflation at rates of 60–80%, and the ever deepening fiscal deficits of the government against the backdrop of GDI issuances carrying a real rate of interest exceeding 20%, it was clear that the Turkish macroeconomic structure was unstable and too risky. The public sector used to crowd out almost half of the private savings funds and the domestic economy turned into a bastion of financial speculation and arbitrage-led rent seeking.

In the meantime, the IMF itself was on the loss of severe credibility loss due to its “mis”-handling of the East Asian crisis that erupted in 1997. IMF’s dogma on austerity at all expense, everywhere and under every condition, resulted in deepening of the 1997 crisis and meant severe deflation for the once-tigers of Asia. Thus, the IMF was in need of a “showcase” of successful stabilizer, and Turkey was a welcome agent to pursue an old idea about disinflation under an exchange-rate-based schedule, which had been pursued in Latin America and had failed. But this time, it was alleged, lessons were learned and Turkey’s would be a totally new and indigenous strategy. The Staff Monitoring Programme (SMP) was initiated in 1998 to this end, and the IMF opened up a station in Ankara to follow the economy (in particular the government fiscal operations) more closely.

Thereby, a comprehensive disinflation program was enacted in July 1998 under the guidance of the IMF. The program administered under close supervision of the SMP aimed at improving the fiscal balances and reducing the long-lasting price inflation. However, the program could not have been put in full action due to the continued political uncertainty surrounding the general elections and two unfortunate earthquakes in 1999. As public expenditures continued to expand, fiscal balances deteriorated even further. Deficit-financing requirements exerted heavy pressures on the fragile domestic financial markets, giving rise to substantially high real interest rates. Finally, in December 1999, the government adopted another disinflation program, aiming at decreasing the inflation rate to single digits by the end of 2002. Aided by the supervision and technical support of the IMF, the new program relied on an exchange-rate-based disinflation program, coupled with monetary control through setting upper limits to the net domestic asset position of the Central Bank (CB). Accordingly, the CB committed itself to a policy of no sterilization, whereby changes in the monetary base would directly reflect changes in the net foreign assets of its balance sheet. The program further entailed a series of austerity measures on fiscal expenditures and set specific targets for the balance on the primary budget, that is, budget balance net of interest payments.

Main elements of this program is narrated extensively in the Turkish crisis literature. It was finally initiated in December 1999, by announcing a Letter of Intent. It was understood that it would cover a time horizon of three years, 2000 through the end of 2002. For the technical aspects of the program, the following paragraphs draw heavily on Yeldan (2002) and Ertuğrul and Yeldan (2003).

The program was based on three main components:

• i. austerity in public expenditures subject to specific targets for non-interest fiscal surpluses;

• ii. a pre-announced calendar for the rate of currency depreciation in line with the targeted rate of inflation; and,

• iii. a monetary rule which subjected the liquidity generation mechanism to the net foreign asset position of the Central Bank (CB), thereby forcing the CB to act as a semi-currency board.

The program announced that the rate of currency depreciation would be set according to a pre-announced calendar, thereby fixing the nominal values of an exchange rate basket on a daily basis throughout the year. For this purpose, the CB declared an exchange rate basket consisting of 1 US$ + 0.77 Euro, and announced a daily calendar of depreciation rate which added up to a cumulative 20% by the end of 2000. The pre-announcement of exchange rate depreciation in accordance with such a tablita was regarded to be the backbone of the program in its attempt to break the inflationary inertia of three decades.

The idea of declaring an exchange rate basket and fixing its daily values throughout the year was not a new experiment, as indicated above. Similar programs were administered in Latin America in the 1980s under the name of exchange rate-based disinflation. The primary example was its implementation in Chile in 1981 through 1983. It generated a very high external deficit and collapsed with a series of onerous adjustments. Then, starting from 1991, it was also implemented in Argentina, under its convertibility programme. Argentina had initial success in bringing its inflation to an end, but after the second half of the 1990s, especially with Brazilian devaluation in 1998, it lost competitiveness very quickly due to its fixity of the exchange rate (at 1 US dollar exchanging for 1 Argentinian peso). The Argentinian economy collapsed along with Turkey in 2001.

What was allegedly unique in the Turkish program was the argument that it entailed an exit strategy. Accordingly, the exchange rate basket (the daily tablita) would be fixed only in the first 18 months of its initiation; and thereafter it would gradually be allowed to float within limits. The limits would be expanded at 6-month intervals to leave it to free float at the end of the stabilization plan horizon—31 December 2002. The details of this “exit strategy” is portrayed in Fig. 8.4.

Fig. 8.4

(Data source Central Bank of the Republic of Turkey)

Path of the nominal exchange rate basket under the stabilization program, January 2000–December 2002

As can be observed the exchange rate basket of “1$ + 0.77 Euro” was announced on a daily basis to generate a cumulative “depreciation” over 2000; and then would be granted partial floating within a band of 7.5% starting from June of 2001. This band would then be expanded at rates of additional 7.5% from end to end at every six months until 31 December 2002, after when the Lira would be under free float along with an inflation targeting central bank proper.

In order to sustain the tablita on exchange rate depreciation, the program further limited the CB’s rule of monetary expansion only to changes in its net foreign asset position in its balance sheet. For this purpose, specific upper ceilings were set on the net domestic assets of the CB. More specifically, the CB’s stock of net domestic assets was fixed at its December 1999 level. It was further announced that the CB would be allowed to change its net domestic asset position within a band of ±5% of the monetary base, to be revised at three-month intervals. To be able to meet the liquidity needs of the banking sector, the reserve requirement ratios were significantly lowered.

In order to evaluate the implications of this rule more clearly, one should observe that the CB balance sheet has the following operational identity:

$$Monetary\;Base \, = \, Net\;Foreign\;Assets \, + \, Net\;Domestic\;Assets$$

As a result of restrictions set on the upper ceiling of net domestic assets, the program limited monetary expansion only to increases in the stock of net foreign assets. This means that the CB would not be able to increase the stock of money supply by, for example, borrowing foreign exchange from the banking system or by using IMF’s credit facility. Furthermore, since the CB was constrained in not to increase its domestic assets, this meant that it could not open any domestic credit neither to the public sector, nor to the private banks who were failing as a result of any liquidity shortage. The CB would be able to issue Turkish Lira and expand its monetary base only by purchasing foreign exchange from the banking sector in a manner where its foreign liabilities would not be increased.

Thus, according to this rule, the liquidity generation mechanism available to the CB practically meant a regime of semi-currency board in monetary operations. Within this mechanism monetary policy was restricted to the direction of foreign exchange flows, and as such, the most important element to sustain the liquidity needs of the economy depended upon the continuation of foreign credit available to the system.

These technical aspects of the program relied on the monetary approach to the balance of payments in its theoretical foundations. This approach was used by the IMF researchers in their country program modeling exercises for the determination of the liquidity generation mechanism and the resolution of the balance of payments equilibrium. This approach, which provides the underlying frame of reference in almost all IMF-style austerity programs, expects the real exchange rate to be in long-run equilibrium at its purchasing power parity level, and maintains that the domestic supply of money will be “endogenized” in a regime of open capital account. A simple portrayal of this theoretical apparatus is narrated in Fig. 8.5.

Fig. 8.5

Theoretical expectations of the currency board operative

Accordingly, suppose that an initial equilibrium money supply is being generated in the money market at some “equilibrium” rate of interest, R₀. Suppose that (due to most probably attracted by the perfect foresight of the exchange rate values ahead, which eliminated all the depreciation risk) there is an inflow of foreign financial capital. Then the CB is not allowed to sterilize, and passes all this new inflow as monetary expansion. The Money supply shifts out to MS’. This means a reduction of the equilibrium interest rate to R’. If, on the other hand, there is an outflow of foreign capital, then the CB allows the money supply to shrink, the money supply shifts to the left MS’’ to yield a higher rate of interest at R’’.

Thus, it was expected that the liquidity available in the domestic economy would be managed directly by interest rate signals in smoothly operating financial markets: rising domestic interest rates would invite foreign inflows allowing for monetary expansion. Excess liquidity, in turn, would be signaled through lower rates of interest, letting foreign capital outflows to balance once again the equilibrium level of liquidity in the domestic money market. The market, through its free operations is expected to deliver an “optimal” interest rate domestically. This optimal interest rate was to be conditioned by the movement of the exchange rate basket under the daily scheme and would converge to the depreciation envisaged −20% for the first year. The theoretical expectation was that when both the exchange rate and the rate of interest would be falling in a controlled manner, this process would force the domestic inflation on prices to stabilize. After three years of experience, the program would end, given its exit strategy.

The Turkish bureaucracy was, in dramatic words, bewildered. These models of imaginary capitalism, narrated in the seminar rooms of the IMF, however, were far from reality. First and foremost, by fixing the rate of exchange basket under a fully liberalized capital account that granted full mobility to financial capital, meant a heavy inflow of foreign finance. Turkish Lira appreciated almost instantly in real terms. Monetary expansion and the optimistic credibility gained under the IMF’s protégé led the interest rate to fall very strongly and almost instantly. The stability aspect of the monetary approach to the balance of payments proved to be only one-sided: as flows were coming in, the economy expanded and everything has been optimistic; yet at the slightest sign of fragility, the direction of foreign flows was reversed and there could have been no mechanism to reach a new equilibrium. The economy simply suffered from severe illiquidity, as the domestic asset markets could not reach any equilibrium and collapsed. This asymmetrical mechanism of the domestic asset markets was clearly the result of shallow and fragile nature of the asset markets, and the deregulated financial deepening.

The fall in the interest rate and the real appreciation of the Turkish Lira were welcomed vehemently. Consumption and investment expenditures, led by cheapening imports, exploded. Current-account deficit widened to an unprecedented 4.8% against the GDP, and there occurred a heavy short-term foreign indebtedness. Against all this, Ertuğrul and Yeldan (2003: 8) vividly comment that:

Given these structural conditions, the program should have envisaged the destructive effects of such a possible liquidity squeeze on the interest rates and on the fiscal balance. The Central Bank was deprived of all of its traditional tools of austerity and crisis management and was left defenseless against possible “speculative attacks” and “sudden stops.” Under these conditions, it is no surprise that the viability of the program would finally suffer when the “uneasy speculators” shift focus and decide to reverse their flows, leaving the incipient country illiquid and dried out.

It has to be underlined that the CB had, in fact, successfully administered its role as the “currency board,” supplying domestic money in response to changes in its foreign-asset position. Figure 8.6 portrays the evolution of this mechanism during the first 10 months, just before the eruption of the first turbulence in late November 2000. The figure discloses the paths of the monetary base, open market operations (OMOs), net foreign assets (NFA), and net domestic assets (NDA) of the CB, as measured by the end-of-week observations, between January 7 and December 1, 2000. As seen in the figure, the CB successfully expanded its monetary base mostly due to the rise in foreign inflows over the course of the program.

Fig. 8.6

(Data source Balance sheet reports of the Central Bank of the Republic of Turkey)

Monetary base, net domestic assets, net foreign assets and net open market operations, 7 January 2000–1 December 2000, end-of-week observations, million Turkish Liras

Thus, the basic message that emerges from the data disclosed in Fig. 8.6 is clear: Turkish monetary authorities successfully implemented the monetary program within the given targets, conditioning the CB operations to net foreign inflows. In this sense, the outbreak of the November 2000 crisis and the ultimate collapse of the program in February 2001 cannot be attributed to any divergence from monetary targets. Quite the contrary, the culminating financial chaos can only be understood within the realm of the successful implementation of both the exchange rate (basket) depreciation targets and the liquidity generation mechanism as followed by the CB—mimicking a currency board.

In fact, the unavoidable appreciation of the domestic currency, accompanied by the explosion of foreign capital inflows, was already in progress, deepening the financial fragility of the domestic economy. A very strong upturn in domestic absorption (accompanied by the appreciation of the Turkish Lira) and the impact of the Customs Union with the EU were the two major reasons behind the rapid expansion of the current-account deficit that reached 9.5 billion US dollars by the end of 2000. This outcome was solely due to the deterioration of the trade balance.

Under these conditions, the economy suffered from yet another financial crisis in February of 2001. These events led to an acute liquidity crisis and the consequent demise of the disinflation program. Turkish Lira was forced to get off the “fixed anchor” and started to free float on 22 February. The exchange rate, as measured by TL/US$, depreciated by 47.7% in six weeks. The crisis conditions spread to the real economy with massive lay-offs and increased social unrest. Once again, the bust phase of the financial cycle struck the Turkish economy after an interval of only two years.

The reform saga of Turkey would continue with the continuation of the IMF directives, then to be pursued with the newly appointed Minister Mr. Kemal Derviş from the World Bank, along with the introduction of a “new” Transition to Strong Economy Program in the hasty rhetoric of “fifteen laws to be enacted in fifteen days.” This program is examined and discussed in the next chapter of this volume.

Concluding Comments

The 1990s were a period of “lost decade”—yet for whom? Clearly, the unregulated and under-supervised banking system was unleashed to gain “speculative rents,” while the brunt of adjustments fell on the wage-labor. The rise of “financial rent” took a significant toll of the distribution of aggregate income. Financialization was carried out through the massive borrowing requirements of the public sector, which was strapped into a vicious cycle of “borrowing – high interest costs – re-borrowing.” This cycle could sustain itself until the contagion of the Asian crisis and would lead to one of the most peculiar experiments in the history of monetary economics—the IMF-led exchange-rate-based disinflation program that was initiated in December 1999. In the words of Balkan and Yeldan (2002: 51):

The post-1989 experience shows the serious problems confronting a developing economy that moves into full external and internal deregulation of its financial system under conditions of high inflation. The specter of capital flight became the dominant motive in policymaking and created unsustainable commitment to high real interest rates and expectations for cheap foreign exchange. Meanwhile links between the financial sector and the real sector have been severed. Instability in the rates of interest and foreign exchange created feedbacks which led the economy further into instability.

Turkey’s post-1989 experience also shows how a “peripheral economy,” trapped within conditionalities of neoliberal restructuring, lost instruments of an indigenous development strategy and was strangulated under the caprices of global finance capital, dictating a speculation-led growth with premature deindustrialization. This new form of conditionality meant the restructuring of traditional forms of dependency, based on an international division of labor pushing the underdeveloped world toward becoming producers of primary/agricultural goods and raw materials, and consumers of manufacturing durables under a regime of “embedded liberalism.” This global division of labor had reached its limits in the 1970s, as technologies in the developed economies matured and industrial profits started to fall (given the unwarranted rise of the organic composition of capital).

“End the financial repression” was the battle cry of global capitalism. Financial flows were liberalized, new instruments of finance were created globally within, what Susan Strange termed, casino capitalism, and manufacturing industries moved off to the new sweatshops of the globe. An unstoppable race to the bottom was started as the underdeveloped nations were one by one stripped off their domestic savings and were pushed into an ever-expanding list by way of which globalization dictated them to privatize, liberalize, deregulate, and adopt flexible norms of labor employment.

As part of an ideological brainwash, the less-developed countries began to be termed as “new emerging markets” or “emerging economies,” and concepts such as “development,” “industrialization,” “working classes” or “bourgeoisie” came to be replaced with a new jargon comprising terms like “austerity,” “financialization” and “market players.” The “new emerging markets” were, in turn, conditioned to a deflationary path where their macroeconomic policies were restricted to a balanced budget, entrenched fiscal expenditures, and a relatively contractionary monetary policy with an ex ante commitment to high real interest rates. While this new episode of financial dependency replaced the traditional forms of industrial/agrarian duality, dynamics of capitalism were in operation and the global economy was making headway onto the 2008–2009 crisis.

Note

• 1. The tablita, meaning “little table,” was the term coined to refer to the schedule of exchange rate fixity over the calendar year, given the exchange-rate based disinflation programs that were administered in Chile (1981–1983) and Argentina (1990–2001). The schedule gave a perfect foresight for the path of the nominal spot value of the market exchange rate and gave clear incentives for tradables to expand. The expectation was that, through “fixing” the market value of the rate of exchange, inertial dynamics of inflation would also be put under control.