Capital Flows, Turbulences, and Distribution: The Case of Turkey

Onaran, Özlem Article Capital Flows, Turbulences, and Distribution: The Case of Turkey Intervention. Zeitschrift fuer Ökonomie / Journal of Economics Provided in Cooperation with: Edward Elgar Publishing Suggested Citation: Onaran, Özlem (2007) : Capital Flows, Turbulences, and Distribution: The Case of Turkey, Intervention. Zeitschrift fuer Ökonomie / Journal of Economics, ISSN 2195-3376, Metropolis- Verlag, Marburg, Vol. 04, Iss. 2, pp. 353-374, https://doi.org/10.4337/ejeep.2007.02.11 This Version is available at: https://hdl.handle.net/10419/277112 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Capital Flows, Turbulences, and Distribution: Th e Case of Turkey Özlem Onaran* Th is paper presents the mechanism of the boom-bust cycles in the context of domestic and international fi nancial liberalisation in the developing countries, and the eff ects of crises and exchange rate volatility on functional income distribution. It is based on the case of Turkey, which has experienced two severe crises in 1994 and 2001 after the liberalisation of capital fl ows, and which has also been hit the hardest during the May-June 2006 turbulences. Th e paper analyses the recent turbulences in the global economy and their consequences in the emerging markets as a case study to illustrate the endogenous formation of expectations. Th e recovery in Turkey after the turmoil is not based on a solution to the structural causes of the problem, since it has completely depended on the reversal of the capital outfl ows thanks to high interest rate, but the continuity of this game is far from clear. JEL classifi cations: E12, E22, E25, F32, G32 Keywords: fi nancial fragility, boom-bust cycles, post-Keynesian, distribution, Turkey 1. Introduction Many developing countries shared the common destiny of fi nancial crises in the 1990s and 2000s after the liberalisation of capital accounts in spite of the diff erences in the former * Vienna University of Economics and Business Administration. Th e paper received fi nancial support from the »Dr. Heinz Kienzl Prize« granted by the Austrian National Bank and the Vienna University of Economics and Business Administration. Th e author is grateful to two anonymous referees and Erinc Yeldan for helpful comments on an earlier version of the paper. Correspondence Address: PD Dr. Özlem Onaran, Vienna University of Economics and Business Administration, Dept. of Economics VWL 9, Nordbergstrasse 15, UZA4 3. Stock Kern D, A-1090 Vienna, Austria, e-mail: [email protected] Received 2 Feb 2007, accepted 10 May 2007 © Intervention 4 (2), 2007, 353 – 374 354 Intervention. Journal of Economics development policies as well as liberalisation processes. Five years after the latest crises in Turkey and Argentina in 2001, the emerging markets were aff ected by the global turbulences in the world economy in May-June 2006, when the US Federal Reserve Bank increased the interest rate, and masses of international investors fl ed out of the emerging markets. Th e turmoil calmed down after a few months, but the short-term memory of the investors, which has recorded the risks involved in the global fi nancial markets, may shape expectations in the future. Th e critical question is thus: »can it happen again?«, as Minsky formulated the question in his seminal paper on the US. Th e post-Keynesian theory as well as historical evidence unfortunately suggest that it is not a question of if, but of when and how deep. Th is paper addresses this question based on the case of Turkey, which has experienced two severe crises in 1994 and 2001 after the liberalisation of capital fl ows, and which has also been hit the hardest during the 2006 May-June turbulences. From Latin America to Asia, fi nancial capital fl ows have generated simultaneously phases of boom and systemic fragility, which then were typically followed by a bust. Th e bust phase has been an endogenous outcome of the boom phase in the sense that the fragility of the economy is a result of the »success« of the system. Th e length and depth of both the boom and bust phases may vary depending on the size of the vulnerability and the shock. But expectations, whose evolution is not easy to forecast, play an important role. Although the systemic fragility can be prevented by limiting the area of risk taking behaviour, thus regulating the fi nancial markets, fi nancial liberalisation creates interests that also prevent the regulation of these markets. In that sense the boom-bust cycles are not neutral with regard to distribution. Th e paper addresses the distributional consequences of the crises for Turkey. Th e paper is composed of seven sections including this introductory one. Section 2 describes the hypothesis of systemic fi nancial fragility and the boom-bust cycles in the developing countries, which have opened up their economies to international capital fl ows. Section 3 discusses the historical evidence from Turkey, as a case to illustrate the boombust cycles discussed in Section 2. Section 4 analyses the recent turbulences in the global economy and their consequences in the emerging markets as a case study to illustrate the endogenous formation of expectations. Section 5 discusses again the eff ects of the global turbulences on Turkey. Section 6 analyses the eff ects of crises and exchange rate volatility on functional income distribution in Turkey. Finally, the concluding section comprises the policy implications of the analysis. 2. Boom-bust Cycles in the Developing Countries Th is section presents the mechanism of the boom-bust cycles in the context of domestic and international fi nancial liberalisation. Th e underlying theory is an open economy extension of the post-Keynesian systemic fi nancial fragility and instability hypothesis of Mins ky (1982 and 1986). Based on the analysis of the currency crises since the 1997 Asian crisis, Arestis / Glickman (2002), Schroeder (2003), Foley (2003), Dymski (1999), Kregel Özlem Onaran: Capital Flows, Turbulences, and Distribution 355 (1998), Taylor (1998), and Isik (2004) have presented a Minskyan analysis of the fi nancial crises in the developing countries. Th e boom and bust cycles are based on the linkages between fi nancial and real variables, and develop endogenously out of the normal functioning of the economy. If good performance persists, lenders become more optimistic and are willing to hold more risky assets or fi rms, which plan investment in physical capital, accept higher debt levels. Th e debt / equity ratios increase and fi rms engage in speculative fi nancing patterns based on short-term fi nancing of long-term investment projects. Th e asset price booms during such episodes lead to an increase in the value of the collaterals and make it easier to borrow. However, this process makes the fi rms vulnerable to credit availability and interest rate shocks, which leads to fi nancial instability. In times, when there is a negative shock, and expectations evolve in a pessimistic direction, this fragility leads to a crisis through credit crunch, debt crisis, and bankruptcies. Th e fragility is latent, but a shock turns it into a crisis. Th e source of the shock, which causes the crash, is not important. It is the built-in vulnerability that leads to a signifi cant eff ect of the shock. Four properties of expectation formation play an important role in this process. First, expectations are formed under fundamental uncertainty and, therefore, agents are infl uenced by conventional wisdom, such that every investor – dealers as well as fi rms – in the economy is trying to guess what the other agents will guess. What is crucial is investor sentiment, not fundamentals. Second, competitive pressures among fi rms or fund managers push them to take similar risks, even when they would rather like to be more conservative. Th us conventional wisdom, i. e. expectations, is also competition-coerced (Crotty 1993). Th ird, expectations are self-fulfi lling. A phase of optimism leads to gradually more boomeuphoric expectations, increasing the risk appetite of the fi nancial investors as well as fi rms planning investment in physical capital. Fourth, expectations are endogenously evolving, and not static. Th us evaluations about what is reasonable change. Good times lead to a self-propelling adventurism and as expected profi ts are realised, investors become more self-confi dent in taking risks. But the opposite mechanism also works. Over-optimism increases fi nancial fragility, and fi nally, when an adverse shock comes, this fragility becomes visible to the investors. Th e shift to over-pessimism makes an expected crisis come true. After the crash and crisis, the investors will be cautious for a while, but eventually, after enough time has passed, competitive pressures and new search for profi table investment will start a new endogenous cycle of stability, to be followed by instability again. In the developing countries, the boom-bust cycles were triggered by both domestic and international fi nancial liberalisation. First the domestic fi nancial markets were liberalised. Th e increases in the real deposit and loan interest rates and the deregulation of fi nancial institutions set the initial conditions for the formation of fragility. Riskier credit supply by the banks, a shift to fi nancial investments at the expense of physical investments by the investors, short-termism, and an adverse-selection towards riskier projects with a higher expected return have been the outcome (Grabel 1995). When the international capital fl ows were liberalised at a later stage, high domestic interest rates attracted high capital infl ows, thanks to a high fi nancial arbitrage between 356 Intervention. Journal of Economics the interest rate and exchange rate due to initially low expected depreciation. Most of these capital fl ows to developing countries have been portfolio investments or short-term credit. As capital infl ows trigger growth in a country, boom-euphoric expectations and competitive international pressures lead to further capital infl ows. However, this leads to the appreciation of the local currency, which in turn results in an increasing foreign trade defi cit. In the meantime, in addition to the maturity imbalances of an economy without international capital fl ows, currency mismatches in the fi rms’ balance sheets, which borrow in foreign currency and invest in domestic currency, create new sources of fragility. Th e high domestic interest rates compared to the foreign currency interest rates and the low expected depreciation rate of the currency is the motivation behind this fi nancing pattern. Th e public sector may also be highly indebted as was the case in Turkey, but this has not been the situation in many other cases, like the Asian countries. As risks build up and, in particular, currency appreciation and the consequent current account defi cit increase beyond a critical point, international investors become slowly aware of the problems. However, this critical point also may change endogenously. Th e combination of some adverse shocks like the bankruptcy of a fi rm or a bank, or problems in the export markets, neighbour countries, world economy, or in the domestic political arena may turn this awareness into a speculation about a possible devaluation. Th e central bank may increase the interest rate in order to avoid capital outfl ow and to satisfy higher risk perceptions regarding expected depreciations. However, this intensifi es the debt problem in the meantime. Finally, the conventional wisdom starts to evolve towards pessimism and investors decide to leave the country before everybody else does. In the end, an expected depreciation becomes a self-fulfi lling prophecy. Imported input costs increase due to depreciation with a pass-through eff ect on infl ation. Th is cost shock and high interest rates lead to bankruptcies, credit crunch, and recession. Th e debt problem becomes magnifi ed by economic recession and depreciation. 3. Boom-bust Cycles in Turkey: 1989 – 2005 Turkey liberalised its capital account in 1989 as the second stage of its integration into the world economy, which was initiated in 1980 via an orthodox structural adjustment program. Th e earlier stages had included liberalisation in domestic fi nancial markets along with foreign trade liberalisation, goods and labour market de-regulation. Th e capital fl ows consisted mostly of volatile portfolio investments and short-term credit, with the share of Foreign Direct Investment (FDI) in total fi nancial account being limited to a range of 10 – 20 percent apart from a couple of exceptional years of FDI, like 1989, 2002, and fi nally 2006. Th e fi rst wave of capital infl ows1 reached an annual level of 3.7 percent of GNP in 1993, accompanied by an appreciation of the currency by a cumulative rate of 47.4 percent in real terms in fi ve years as of 1993 compared to 1988, and a current account defi cit 1 Financial account plus net errors and omissions, the latter represents unrecorded capital fl ows. Özlem Onaran: Capital Flows, Turbulences, and Distribution 357 of 3.5 percent of GNP. Figure 1 below portrays the boom-bust cycles in Turkey by illustrating the capital infl ows / GNP ratio, the current account balance / GNP ratio, and growth of GNP. Figure 2 (p. 358) shows the annual percent change in the real trade weighted effective exchange rate (defl ated by the Consumer Price Index [CPI]). Figure 1: Th e Boom-bust Cycles in Turkey, 1984 – 2006* -15 -10 -5 0 5 10 15 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006* Capital inflows/GNP (%) Growth (%) Current Account balance/GNP (%) * 2006: Th e cumulative values of the last 12 months where data is available, i. e. October 2005 – September 2006. Source: Own calculations based on data supplied by the Central Bank of Turkey, Electronic Data Distribution System. Th e accumulated risks associated with an appreciated currency, high current account defi - cit, combined with the mismanagement of the domestic borrowing policy by the government, who had the infeasible obsession to try to reduce the interest rates in the eve of the elections, ended up triggering a massive capital outfl ow in 1994.2 Th is fi rst currency crisis after the liberalisation of the capital account led to a depreciation of the currency by 23.9 percent in one year and a severe recession with GNP declining by 6.1 percent. 2 See Yenurk (1999) for a more detailed discussion of this period. 358 Intervention. Journal of Economics It did not take long until the international investors started to enjoy the defl ated asset prices in the stock and bond markets and the security that came with the already depreciated currency. Th e ratio of capital fl ows to GNP reached a level of 4.1 percent in 1995 and remained mostly high during the1995 – 2000 period. In the meantime, Turkey enjoyed high growth rates except for the year of the real (not fi nancial) earthquake of 1999. However, the hike in the infl ation rate (in CPI) to a level of 125 percent during the 1994 crisis had led price increases to stick to a new higher plateau of 78.7 percent average annual infl ation in the following years (1995 – 1999), compared with a previous average of 66.6 percent (1989 – 1993). At the end of 1999 the government decided to implement an anti-infl ation program within the context of a stand-by agreement with the International Monetary Fund (IMF). Th e program was based on a crawling peg exchange rate regime, using the exchange rate as a nominal anchor to curb infl ation.3 However, experience in 2000 proved that the exchange rate as a single nominal anchor was only partially successful to control infl ation, as had also been the case in many other countries, and the decline in infl ation was not enough to prevent a signifi cant real appreciation of the currency, 15.9 percent in one 3 See Yeldan (2002), Boratav / Yeldan (2006), Akyuz / Boratav (2003), Uygur (2001) for a more detailed discussion of the programme. Figure 2: Real Exchange Rate Index (Annual Change in , Trade Weighted Eff ective, Defl ated by CPI, 1989 – 2006*) -30 -20 -10 0 10 20 30 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 June-06* Dec-06* * To indicate the May-June shock the change in  both in June and December 2006 are shown with respect to December 2005. Note: 1995 = 100, a decline indicates depreciation. Source: Th e Central Bank of Turkey, Electronic Data Distribution System. Özlem Onaran: Capital Flows, Turbulences, and Distribution 359 year. At the same time the current account defi cit reached to 4.9 percent of GNP, which was higher than before the 1994 crisis. Th e guarantee of a low and controlled rate of depreciation coupled with high interest rates had attracted capital infl ows for the fi rst ten months of 2000, but the questions regarding the sustainability of the current account deficit accompanied by fi nancial risks in the private banking sector invited a series of pessimistic speculative expectations. Finally, an initial outfl ow of capital in November 2000 after a liquidity crisis related with a bank was followed by a more massive outfl ow in February 2001, the latter of which was also triggered by the political confl icts around the issue of banking reform and supervision. Th e political factor played the role of an exogenous catalyst in a fragile economy, where the investors were already waiting for a signal to move out. But even in the absence of a political confl ict, there could have been another triggering event, once the fragility is there. Th e overall capital outfl ow in 2001 amounted to 11.3 percent of GNP; currency depreciated by 21.2 percent in real terms in one year, and GNP decreased by a historically high rate of 9.5 percent. A brief balance sheet of the growth performance of Turkey during this period shows that the high volatility and crises have also led to a lower GNP growth rate (three percent annually) during the fi rst decade of international fi nancial liberalisation (1990 – 2001), compared to the 1980s (four percent per year). It must be also noted that the growth performance after the implementation of the export-oriented structural adjustment program is in general lower than in the previous decade of import substituting industrialisation (4.8 percent per year during 1970 – 1979). Th e dramatic fi nancial crisis of 2001 set the conditions for a long postponed restructuring process in the banking sector. Th is has also been in line with the preferences of the large scale fi nancial-industrial corporations, which were already competitive in the international markets and wanted to prevent the systemic fragility created by the weak elements in the banking sector, which were not able to cope up with the international standards of making business (Gultekin-Karakas 2006). Th e Independent Banking Supervision Institution took over the banks, which had operated without obeying the banking regulations, and restructured these banks using public funds to eventually sell them. Th is process also resulted in a signifi cant entry of international banks into the sector. Th rough the course of this reform process, the law for the independence of the central bank was also passed, and the monetary policy target gradually evolved towards infl ation targeting with a fl exible exchange rate system. In the period after 2001, the EU also turned into a more important anchor in partnership with the IMF to determine the direction of change as well as to signal the credibility of the programs to the international investors (Onis / Bakir 2005, Atac / Grünewald 2006). Th e targets of IMF programs and the steps to be taken to fulfi l the economic conditions of membership overlapped. In terms of the international institutions, which audited and supported the credibility of the economic programs that Turkey implemented until the 2000s, the IMF had been the only anchor. Even after Turkey started the Customs Union with the EU in 1996, or after the Helsinki Summit in 1999, where Turkey was accepted as a candidate country, the EU played a role more as a political anchor; and a sort of an implicit division of labour was made with the IMF for auditing economic restructuring. 360 Intervention. Journal of Economics Th e conditional green light to start accession negotiations in the 2002 Copenhagen summit has been eff ective in turning EU to a more extensive anchor. After the crisis of 2001, Turkey enjoyed an uninterrupted and high growth era, with a 7.5 percent average annual rate of growth in GNP during 2002 – 2005. High capital infl ows towards Turkey among other emerging markets have been the determining source of fi nance for achieving this growth rate. Th is is to some extent similar to what had happened after the 1994 crisis. Th anks to the defl ated prices in the asset markets, the depreciation of the currency lowered asset prices once again in terms of foreign currency and also decreased the likelihood of a depreciation in the coming period, creating the possibility of an appreciation after the over-shooting of the exchange rate. Additionally, the EU-anchor was a signifi cant factor in securing the capital fl ows in the period after 2001. Th e result was typically a continuous appreciation in currency; at the end of 2005 the Turkish lira (TL) was 47.4 percent appreciated compared to 2001; and the current account defi cit had reached a historically high level of 6.4 percent of GNP. Nevertheless, talking about the risks associated with such a high current account defi cit seemed to be a complete pessimism at that time. Th e market sentiments celebrated this period as a completely new era, where the EU-anchor was playing an important role in decreasing political risks, and creating the potential for a higher FDI infl ow. Th e optimists also emphasized that the current account defi cit was fi nancing new private investments, which would eventually improve competitiveness and exports. Th e government mostly cited the eff ect of increasing oil prices as an excuse for the increase in current account defi cit, and seemed to be particularly trusting the corrective capacity of the fl exible exchange rate system to tame speculative expectations. 4. 2006 and Global Turbulences Th e optimism about the start of a new era in the Turkish economy was disturbed by the global turbulences in the world economy in May-June 2006. Overall, between May 8 and June 13, emerging stock markets lost a quarter of their value. Within two weeks time after the initial international shock in May 11, the Turkish currency depreciated by 7.7 percent in real terms, and the trend continued in June with a cumulative real depreciation rate of 17.3 percent at the end of the month compared to April. Between May 10 and June 30, the Istanbul Stock Exchange Index fell by 32.3 percent in terms of US dollars. During the same period Hungary, Brazil, and South Africa were also among the emerging markets, which were hit most severely. But the outfl ow of capital was not particularly selective or related to the so-called »macroeconomic fundamentals«, with India, for example, also being one of the most aff ected countries in spite of its almost negligible current account defi cit. Among the new member states of Europe, in addition to Hungary with its seriously high current account defi cit, Poland and Slovakia were also aff ected.4 4 See Onaran (2006a) for an early article on the leading indicators of fragility in the Central and Eastern European new member states and Turkey. Özlem Onaran: Capital Flows, Turbulences, and Distribution 367 Figure 4: Foreign Debt of the Public and Private Sector as a Ratio to GNP () (left scale: total & public; right scale: private) 20 30 40 50 60 70 80 90 100 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006* 4 6 8 10 12 14 16 18 20 Total foreign debt/GNP Total public debt/GNP Private short-term debt/GNP Private long-term debt/GNP * 2006: Th e cumulative values of the last 12 months where data is available, i. e. October 2005 – September 2006. Source: Own calculations based on data supplied by the Central Bank of Turkey, Electronic Data Distribution System. tuted an important part of the capital infl ows with a share of 36.1 percent during the fi rst nine months of 2006, and the EU-anchor did have some eff ects. FDI infl ows from Europe in this period reached up to 92.9 percent. However, the continuity of the infl ows remain far from clear, since most of them were through mergers and acquisitions, particularly in the banking sector. Th e share of the manufacturing sector in total FDI inward stock also remained to be as low as 15.1 percent. It is true that the ratio of the stock of FDI to GNP remains to be quite low in Turkey (11.6 percent in 2005) compared to those in the Central and Eastern new member states of the EU (e. g. 55.9 percent in Hungary, 48.1 percent in Czech Republic, 31.1 percent in Poland, and even as high as 93.6 percent in Estonia, see UNCTAD 2006). Th is is often seen as a positive prospect about the possible trajectory of the FDI developments in Turkey. However, it is not clear whether such high rates can be achieved also in Turkey, which has a long tradition of large scale domestic corporations itself. For comparison, the same rate is 27.3 percent in Mexico, with a relation to US similar to that of Turkey to the EU, but a far closer geographical proximity. Finally, there are 368 Intervention. Journal of Economics doubts that FDI may create a higher level of import dependency due to lack of domestic backward linkages and supply of intermediate inputs from international subsidiaries, rather than contributing to an improvement in productivity and in the longer run in the current account balance.8 6. Functional Income Distribution Th rough the Boom-bust Cycles in Turkey Some domestic and foreign investors can make gains over the boom-bust cycles, buying and selling the domestic currency denominated assets at the right time. When the bust arrives, there are winners and losers of this process, but this is not necessarily a confl ict between the shares of fi nancial vs. non-fi nancial profi t income in total income. Th e labour share declines in all countries that have experienced currency crises (Onaran 2006b), and 8 See Görg / Greenaway (2003) for a review of the spill-over eff ects of FDI, and Mencinger (2003) for a discussion for the case of transition economies. Figure 5: Short Foreign Exchange Position of the Non-banking Sector 9,5 13,2 12,4 28,1 33,9 39,9 39,8 55 58,5 72,7 71,5 8 0 10 20 30 40 50 60 70 80 2005-Dec 2006-March 2006-June 2006-Sept Short position/GNP Short position/exports of goods and services Short position/international reserves Note: GNP and exports of goods & services are computed on a yearly basis. International reserves are stock values at the end of period. GNP value of June is used for September. International reserves are gross foreign exchange reserves of CBRT (including gold). Source: Central Bank of the Republic of Turkey 2006. Özlem Onaran: Capital Flows, Turbulences, and Distribution 369 this decline in the labour share then compensates for the increase in fi nancial costs for industrial fi rms. Evidence also suggests that industrial fi rms fi nd the chance to increase their returns from fi nancial activities (Istanbul Chamber of Industry 2003). Th e crises of both 1994 and 2001 have led to a clear and long lasting decline in the wage share in Turkey. Figure 6 shows the wage share in manufacturing industry.9 Th e percentage decrease in the wage share by far exceeds the rate of decline in production during the crises. After a crisis, employers push workers to accept dramatic wage cuts or compulsory unpaid leaves to avoid job losses. Eventually, profi ts are restored, and when the crisis is long past it is labour which has carried the burden of adjustment. Th e crisis also creates a negative eff ect on the bargaining power of labour for a long period afterwards. Diwan (2001: 1) defi nes crises as episodes of distributional fi ghts, which leave »distributional scars«. Although a strong economic recovery takes place after the crisis, with production returning to its pre-crisis level within a year, the fall in the wage share is much more persistent. Figure 6: Wages / Value Added, 1970 – 2005 10 15 20 25 30 35 40 1970 1975 1980 1985 1990 1995 2000 2005 Note: Th e data for the wage share in manufacturing in the national accounts exist for the period after 1987 at a sectoral level. Th is data is linked with the data in the Industrial Survey for the period after 2001 due to data availability. Source: Own calculations based on data supplied by the State Institute of Statistics. After the crises of 1994, the fall in the wage share continued also in 1995, with a cumulative decline of 24.8 percent compared to 1993. Th e shock in 2001 was more dramatic; the wage share has continued to decline throughout the next fi ve years including 2005. Th e 9 Due to lack of long time series data for wages, the analysis here is based on the manufacturing industry. Th e wage share data for the rest of the economy exists only from 1987 onwards, based on the national accounts. Th e data used here is reported in the Annual Survey of Employment, Pay- 370 Intervention. Journal of Economics initial decline of 13.7 percent reached fi nally to a cumulative fall of 26.8 percent in 2005 compared to 2000. Th e wage share in 2005 is as low as 1994. Strikingly, the whole era of Turkey’s liberalisation and integration into the world economy since 1980 has been a period of decline in the wage share. Indeed, the major negative shock to labour’s share took place in the early phase of neo-liberal structural adjustment and the recoveries in the later stages were minor and short-lived and were reversed by fi nancial crisis. Th e short period of increase in the wage share during 1989 – 1991 was interrupted by the 1994 crisis. Th e recovery after the 1994 crisis was rather slow, with the wage share in 2000 still below the previous peak of 1991. One important factor that has led to the deterioration in labour’s share during the crises is the exchange rate movements. Apart from the crisis episodes, the opening up of the economy was accompanied by signifi cant devaluations of the domestic currency with the aim of achieving higher international competitiveness. Be it due to the offi cial devaluations of the early stages of liberalisation or the market-made depreciations after the fi - nancial crises: there is a clear trade-off between the rate of depreciation and the wage share. Depreciation creates an increase in the price of the imported goods, and thus in overall input costs. During the fi nancial crisis depreciation of the local currency creates a signifi cant infl ationary shock. Th e magnitude of this shock is related to the import dependency of the economy, because the oligopolistic power of the fi rms to pass on import price changes to consumers. But the workers confronted with the threat of job loss during a crisis mostly fail to pass the consequent price shocks to their nominal wages. In the meantime, utilising the imbalance of power relations, the fi rms compensate the increase in input costs by a decline in labour costs. Th e reverse of this story has also been true during episodes of capital infl ow, and currency appreciation, when employers became more accommodated towards wage demands, e. g. during the episode of 1989 – 1993. However, this was soon disturbed by the currency crises. Table 1 demonstrates the eff ect of a nominal depreciation on the wage share in manufacturing industry, based on a regression analysis. Th e change in the wage share (in logs) is estimated as a function of the change in nominal exchange rate and the manufacturing value added (both in logs) and the fi rst lags of all the variables. Th e estimation results indicate that a ten percentage point increase in the ments, Production and Tendencies in Manufacturing Industry based on fi rm level surveys supplied by the Turkey Statistical Institute for all public sector fi rms and private fi rms with 10 or more persons for the period of 1950 – 2001. Th e wage and salary data in the survey include wages and salaries, overtime payments, bonuses, indemnities, payments in kind, before gross income tax, social security and pension fund premium deductions from the employees, but excludes the contributions to social security etc. by the employers. Th at is the reason why the wage share looks too low and is not comparable to the levels of the wage share based on national accounts methodology. Due to a change in the survey methodology, the data after 2001 is not announced, therefore the wage share based on the manufacturing industry surveys is extended using the percentage change in the wage share in manufacturing industry based on the national accounts for the years 2002 – 2003. Th e two series have a correlation coeffi cient of 0.87. Özlem Onaran: Capital Flows, Turbulences, and Distribution 371 depreciation rate (percent change in the exchange rate) leads to a 2.2 percentage point increase in the growth rate of the wage share. Th e persistence of a decline in the wage share is also signifi cant. Growth does not have a statistically signifi cant eff ect either in current or lagged form. Table 1: Estimation Results for the Wage Share in Manufacturing Variable Coeffi cient Prob. C 0.0008 0.9895 DLOG (manufacturing value added) 0.0037 0.9935 DLOG ( TL / $ ) - 0.2 209 0.0731 DLOG ( Wages / manuf ac turing v alue adde d)t-1 0.3293 0.0982 DLOG (manufacturing value added)t-1 0.3135 0.4594 DLOG ( TL / $ ) t-1 0.136 4 0. 2963 R-squared 0.2300 Durbin-Watson stat 2.1435 Note: Dependent Variable: DLOG (wages / manufacturing value added); method: Least Squares; sample: 1972 – 2005. Th e data necessary to analyse the eff ect of the recent turbulences on income distribution was not available at the time when this article was written. Both the annual industry surveys and the national accounts based on the income approach, which are the sources of the relevant data, only cover a period until December 2005. But the quarterly manufacturing industry surveys, which report real earnings, even if not value added, indicate only a minor decline of 0.2 percent in real earnings in the third quarter of 2006 compared to the same quarter of the previous year. Given the increases in productivity, this nevertheless corresponds to a decline in the wage share. However it is too early to say much on the further distributional eff ects, since the wage bargaining process also needs some time to adjust to the shock. 7. Conclusion Th e global turbulences of May-June 2006 and the massive, though temporary, capital outfl ows from the developing countries have once again raised doubts about the sustainability of a growth process dependent on capital infl ows. Th e recovery in Turkey after the turmoil is not based on a solution to the origin of the problem, since it has completely depended on the reversal of the capital outfl ows thanks to very high interest rates, but the continuity of this game is far from clear. A new wave of speculative fi nancial capital outfl ows from the emerging markets, which may be followed by further turbulences giv- en the global imbalances, remains to be a signifi cant risk factor, particularly for the most 372 Intervention. Journal of Economics fragile cases like Turkey. In May, neither the high fi nancial arbitrage nor the EU-anchor has protected Turkey against the capital outfl ows from the emerging markets. Th e EU- anchor has indeed failed to protect even Hungary, which is a member state. If the conventional wisdom of the markets shifts from optimism to pessimism, can the EU-anchor help Turkey at all, particularly when relations with the EU are getting tenser? Would the markets care whether the appreciation of the currency is a natural catching up phenomenon (Balassa-Samuelson eff ect), or due to the improved prospects for FDI infl ows, which increased investments fi nanced by imports and eventually help the country to cover the current account defi cit in the future? Th e evaluation of the fi nancial investors at critical turning points in the future will certainly depend on the recent history and how badly they were punished by volatility. Now that the boom has been underway for a long time and the recent turbulences have rather had a profi t-taking than punishing eff ect for the investors, a radical shift to over-pessimism can be postponed for another while, although investors are already quite cautious. However, the question is whether this eventuality can be ruled out completely. Simply ignoring the possibility of a massive outfl ow, which will trigger deeper real eff ects in the future, seems to be gambling in policy making. Th is behaviour is like ignoring a gas leakage in your house, and choosing a »wait and see« strategy, rather than trying to fi x the leakage. Sound policy requires taking the global turbulences and their consequences seriously and considering them as cases in defence of fi nancial regulation and international capital controls. Financial regulation along with industrial policy is the only long-run policy alternative to prevent fi nancial fragility and the potential causes of a future crisis. References Akyuz, Yilmaz. / Boratav, Korkut (2003): Th e Making of the Turkish Crisis, in: World Development, Vol. 31, No. 9, pp. 1549 – 1566 Arestis, Philip / Glickman, Murray (2002): Financial Crisis in Southeast Asia: Dispelling Illusion the Minskyan Way, in: Cambridge Journal of Economics, Vol. 26, No. 2, pp. 237 – 60 Atac, Ilker / Grünewald, Andreas (2006): Turkey’s Political Economy between Crisis and Strong Growth: Discussion of the Europeanisation Dynamics in Turkey, ESA Workshop, Vrije Universiteit Amsterdam, 31 August – 2 September Boland, Vincent (2006a): Turkey Feels Pressure of Plunging Lira; Ankara Faces Tough Rate Decisions If It Is to Maintain Economic Momentum and Keep Its EU Ambitions Alive, in: Financial Times, June 7, p. 7 Boland, Vincent (2006b): Country’s Business Elite and Rulers Drift Apart, in: Financial Times, May 23, p. 6 Boratav, Korkut / Yeldan, Erinc (2006): Turkey, 1980 – 2000: Financial Liberalization, Macroeconomic (In)-Stability, and Patterns of Distribution, in: Taylor, Lance (ed), External Liberalization in Asia, Post-Socialist Europe and Brazil, Oxford: Oxford University Press, pp. 417 – 455 Özlem Onaran: Capital Flows, Turbulences, and Distribution 373 Central Bank of the Republic of Turkey (2006): Financial Stability Report, URL: http://www. tcmb.gov.tr/yeni/eng/ Crotty, James R. (1993): Rethinking Marxian Investment Th eory: Keynes-Minsky Instability, Competitive Regime Shifts and Coerced Investment, in: Review of Radical Political Economics, Vol. 25, No. 1, pp. 1 – 26 Diwan, Ishac (2001): Debt as Sweat: Labor, Financial Crises, and the Globalization of Capital, mimeo, URL: http://info.worldbank.org/etools/docs/voddocs/150/332/diwan.pdf Dymski, Gary A. (1999): Asset Bubbles and Minsky Crises in East Asia: A Spatialized Minsky Approach, University of California, Riverside, Department of Economics, Research Paper, April Elliott, Larry (2006): Fair Exchange for Fairer World, in: Guardian Weekly, March 10 – 16, p. 8 Foley, Duncan K. (2003): Financial Fragility in Developing Economies, in: Dutt, Amitava K. / Ros, Jaime (eds.), Development Economics and Political Economy, Cheltenham / Northampton: Edward Elgar Goldstein, Morris (2005): What Might the Next Emerging-Market Financial Crisis Look Like?, Institute for International Economics Working Paper Series, No. 05-7 Görg, Holger / Greenaway, David (2003): Much Ado About Nothing? Do Domestic Firms Really Benefi t From Foreign Investment?, Discussion Paper Series IZA, No. 94 Grabel, Ilene (1995): Speculation-Led Economic Development: A Post-Keynesian Interpretation of Financial Liberalization Programmes in the Th ird World, in: International Review of Applied Economics, Vol. 9, No. 2, pp. 127 – 149 Gultekin-Karakas, Derya (2006): Rising Concentration and Centralization of Banking Capital in Turkey, Annual Conference of European Association for Evolutionary Political Economy, November 2 – 4, Istanbul Independent Social Scientists (2006): Turkey and the IMF: Macroeconomic Policy, Patterns of Growth and Persistent Fragilities, URL: http://www.Bagimsizsosyalbilimciler.org Isik, Sayim (2004): Application of the Post-Keynesian Financial Fragility Approach to the Financial Crises in Turkey (1991 – 2001) (in Turkish), in: Iktisat, Isletme / Finans, Vol. 19, No. 218, pp. 51 – 77 Istanbul Chamber of Industry (2003): Turkey’s Top 500 Industrial Enterprises, Istanbul: ICI Kregel, Jan A. (1998): Yes, It Did Happen Again – A Minsky Crisis Happened in Asia, Jerome Levy Economics Institute Working Papers, No. 234 Mencinger, Joze (2003): Does Foreign Direct Investment Always Enhance Economic Growth?, in: Kyklos, Vol. 56, No. 4, pp. 491 – 508 Minsky, Hyman P. (1982): Can »It« Happen Again? A Reprise, in: Challenge, Vol. 25, No. 3, pp. 5 – 13 Minsky, Hyman P. (1986): Stabilizing An Unstable Economy, New Haven, London: Yale University Press Onaran, Özlem (2006a): Speculation-led Growth and Fragility in Turkey: Does EU Make a Diff erence or »Can It Happen Again«?, in: Hein, Eckhard / Heise, Arne / Truger, Achim (eds.), Macroeconomics and Macroeconomic Policies – Alternatives Approaches to Euro pean Policies, Marburg: Metropolis, pp. 199 – 226 374 Intervention. Journal of Economics Onaran, Özlem (2006b): Labor’s Share in Developing Countries in the Era of Globalization, Annual conference of the Union for Radical Political Economy [URPE], Boston, held in January 2006 Onis, Ziya / Bakir, Caner (2005): Turkey’s Political Economy in the Age of Financial Globalization: Th e Signifi cance of the EU Anchor, Europeanization and Transformation: Turkey in the Post-Helsinki Era, Koc University, 2 – 3 December Schroeder, Susan K. (2003): A Minskian Analysis of Financial Crisis in Developing Countries, University of Bremen, Bremen Contributions to Institutional and Social Econom ics, No. 50 State Institute of Statistics [SIS] (1950 – 2001): Survey of Employment, Payments, Production and Tendencies in Manufacturing Industry, electronic data, Ankara State Planning Organisation [SPO] (2006): General Economic Targets and Investments, 2007, Ankara: SPO Taylor, Lance (1998): Lax Public Sector, Destabilizing Private Sector: Origins of Capital Market Crises, New School, Center for Economic Policy Analysis Working Paper Series, No. 3 Th e Central Bank of Turkey (2006): Financial Stability Report, Ankara: CBT Th e Economist (2006a): Shares in Emerging Economies: A Nasty Spillage, June 10, pp. 79 – 80 Th e Economist (2006b): Emerging Markets: Spot the Diff erence, July 8, p. 74 Th e Economist (2006c): Emerging Market Turbulence: A Turkish Bath? June 3, pp. 78 – 79 UNCTAD (2006): World Investment Report, Geneva: United Nations Uygur, Ercan (2001): Turkey From Crisis to Crisis: 2000 November and 2001 February Crises (in Turkish), Turkey Economy Institute, Working Paper, No. 2001 / 1 Voyvoda, Ebru / Yeldan, Erinc (2005): Turkish Macroeconomics Under the IMF Program: Strangulation of the Twin-Targets, Lopsided Growth and Persistent Fragilities, URL: http://www.Bagimsizsosyalbilimciler.Org / Yazilar_Uye / Vydec05.pdf Yapi Kredi Bank Economic Research Department (2007): Monthly Bulletin, Istanbul, mimeo Yeldan, Erinc (2002): On the IMF-directed Disinfl ation Program in Turkey: A Program for Stabilization and Austerity or a Recipe for Impoverishment and Financial Chaos?, in: Balkan, Nesecan / Savran, Sungur (eds.), Th e Ravages of Neo-Liberalism: Economy, Society and Gender in Turkey, New York: Nova Science Yeldan, Erinc (2007): Patterns of Adjustment Under the Age of Finance: Th e Case of Turkey As a Peripheral Agent of New-Imperialism, Th e Annual Meetings of URPE, Chicago, 5 – 7 Januar y Yenturk, Nurhan (1999): Short-Term Capital Infl ows and Th eir Impact on Macroeconomic Order: Turkey in the 1990s, in: Developing Economies, Vol. 37, No. 1, pp. 89 – 113