Türkiye'de dış ticaret ve döviz kuru (1981-1995 vektör otoregresif yaklaşımı
Başlık çevirisi mevcut değil.
- Tez No: 55736
- Danışmanlar: DOÇ. DR. BURÇ ÜLENGİN
- Tez Türü: Yüksek Lisans
- Konular: Mühendislik Bilimleri, Engineering Sciences
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1996
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 65
Özet
Fiyat özellikle az gelişmiş ülkelerin dış pazarlarda rekabet edebilmeleri için en önemli araçlarından biridir. Bu ülkelerin ucuz emek kullanarak üretim yapmalarından kaynaklanan fiyat avantajlarının yanında ve maliyet avantajından daha çok döviz kuru politikalarıyla göreli olarak daha ucuz fiyatlarla dış pazarlara çıktıkları görülmektedir. Ancak devalüasyonun ne denli iyi ve başarılı bir politika olduğu günümüzde sıkça tartışılmaktadır. Pek çok istikrar programında devalüasyon önemli bir bileşke olarak görülmektedir. Ancak devalüasyonların göreli fiyatları tekrar ayarlayarak amacına ulaşmalarına karşın bunu çok pahalıya mal ettikleri tartışılmaktadır. Devalüasyon karan pek çok karar verici için en son ve en zor seçeneklerden biri bulunmaktadır. Bunun en önemli sebeplerinden birisi devalüasyon - enflasyon kısır döngüsüne girilme korkusudur. Devalüasyonun ödemeler dengesi üzerinde etkili olabilmesi nominal devalüasyonun reel döviz kurunu değiştirmesine, daha doğrusu arttırmasına bağlıdır. Devalüasyonun etkinliğinin sorgulanmasının temel nedeni, devalüasyonun etkisinin çoğunlukla enflasyon oluşumuyla nötr hale geldiği endişesidir. Ayrıca devalüasyon, ithal girdi fiyatlarında artış, kredi stoğunda azalma gibi üreticiyi caydırıcı sonuçlar doğurabildiğinden üretimde daralma görülmektedir. Kısa dönem esnekliklerinin Marshall-Lemer Koşulunu karşılamaması J-Eğrisi etkisine yol açar. Bu etki, üretici ve tüketicilerin devalüasyon sonucu oluşan yeni koşullara uyum süreci içersinde ortaya çıkmaktadır. Tüm bunları incelemek üzere Türkiye'de 1981-1995 dönemi ele alınarak her değişkenin içsel olduğu ve her içsel değişkenin kendisinin ve diğer değişkenlerin gecikmeli değerleriyle açıklandığı bir eşanlı denklemler sistemi olan vektör otoregresif dört model kurulmuştur. Modellerde ithalat, ihracat, döviz, enflasyon ve üretim değişkenleri bulunmaktadır. Kurulan modellerin üzerinden etki-tepki fonksiyonları incelenmiştir. Yani her modelde dövizdeki bir standart sapmalık ani değişimin tüm değişkenler üzerinde 36 aylık bir süreçte nasıl bir değişime yol açtığı incelenmiştir. Tüm modellerde dövizdeki artış sadece kısa dönemde enflasyonun üzerinde kalabilmiştir. Süreç içersinde fiyat artışları dövizden daha hızlı artmakta ve sonuç olarak reel devalüasyonu sıfırlamakta ve hatta üzerine çıkmaktadır. İthalat, dövizdeki artışın enflasyonun üzerinde olduğu dönemde çok azalmakta ama daha sonra yine artmaktadır. İhracatta kısa sürede toparlana j-eğrisi etkisi görülmekte ve önemli düzeyde bir artış sağlanabilmektedir. Ancak bu da kısa süreli bir iyileşme olmaktadır. Reel devalüasyonun sürdürülememesi ile ihracatta da çok fazla düşüş olmaktadır. Üretim daralmaktadır. Sonuç olarak denilebilir ki Türkiye'de devalüasyon, dış ticaret ve ekonomik büyüme bazında hedeflenen iyileşmelere yol açamamaktadır. Devalüasyon kendisini oluşturan sebepleri tekrar doğurmaktadır. VI
Özet (Çeviri)
Foreign trade has always been done in the human history and has developed day by day as our world becomes smaller by economical communities. Even though the rules has changed we always have the 'price' as a very important criteria in the highly competitive market. Less developed countries besides their holding the cost advantage because they use very cheap labor, handle their competitive position by foreign exchange rate policies- that is they make their export products cheaper by devaluation. But it is being immensely discussed if devaluation is a successful or good policy. Devaluation is a policy which changes the demand structure. That is it makes the import products more expensive for the local people so that the demad for these products will shift towards the substitute products produced in the homeland and on the other hand the export products will become cheaper for the people outside the country so they will demand more of those products. This is true only when the price of the product does not change in the countries own currency and when sum of the demand and supply elasticities are more then 1. There are some findings showing that devaluation can have a shrinking effect on the supply side if imported raw materials are highly used in the production, if wages are determined according to the consumer price index and if credit supply declines. After devaluation the cost of the products will increase because the price of the imported raw materials will increase. So then overall prices will increase. Then the inflation will decrease and even neutralize the planned effects of devaluation. Although there are many negative effects of devaluation, it is still an important policy for export-led growth. In many Latin-American economies we can see higher foreign exchange rates in 1980s compared to 1970s and this is one of the key effects to explain their export growth. Statistical findings regarding the effects of devaluation are so many and some are actually contradictory. Some statistical anlysis say that devaluation causes inflation and decreases the aggregate production. On the other hand some claim that there is no direct causality between foreign exchange rate and price index. In another study it is said that devaluation has a shrinking effect for less developed countries but has an expanding effect for developed countries. Also it is said that the shrinking effect of devaluation is a short term phenomenon which turns to be the contrary in the long Ifrm. VIIThe foreign exchange rate policy in the Turkish economical history changed a lot. In the first years of the Republic value of the Turkish Lira was always high compared to the other currencies and the dominant policy was tight money policy, during the 2nd World War imports fell behind exports. Therefore rezerves of TCMB increased too much. Prices were so high. In 1946 the first devaluation of tine Republic history was made. This devaluation is unique in our history because it was made in a period whwn there is no shortage of foreign currency. Its aim was to limit the imports which was supposed to increase too much and also to increase exports. But unfortunately import growth rate became to be higher than the export growth rate. In 1950s foreign trade gap begun to be a big problem and at the end in 1958 the highest devaluation (%236) of our economic history was made. This time it was because of an economical depression. But this application was not successful and there came the time of Growth Plans. During this period Turkish Lira was allowed to be overvalued because they aimed to provide the production of the imported products in the homeland. In this way the investments required to build up our industry would be cheaper. The main purpose was to prevent the baby industries untill they can grow to compete in the world market. But this never happened. Imports increased too much compared to exports. This could be financed untill 1970s when the oil crisis burst out. This system could be carried out 1977 when the government could not help doing high devaluations one after the other. Foreign currency shortage came to its last limits. Then came the 24 January Decisions after which opened the new era of liberalism. In the first years of the 1980s exports increased while the inflation was decreased. This was possible because the social and political circumstances was available to suppress the demand, the wages etc. This did not last long. In the last years of 1980s we see the inflation and imports rising too much and so the foreign trade gap growing. In 1994 the government came to a point that they can not afford the increasing borrowing interests. They tried to suppress the interest rates when the Turkish Lira was overvalued. So came the crisis and 5 April Decisions which provided a very high devaluation. So we see that devaluation is very immensely used. Turkey as one of the less developed countries trying to establish its development strategy on an export-led growth basis uses the foreign exchange rate policy as part of its development strategy. It has not been long since we see higher exhange rates inTurkey. Only after the big crisis in the last part of the1970s exchange rate policy has turned out to be a very important part of a new era. In the beginning devaluation was very successful in the export growth. But Turkish industry is very much dependant on imported raw material. Therefore import costs increased. In 1988 import growth rate exceeded export growth rate. In Turkey we can not see a struggle to establish a lasting competitive advantage other than price and devaluation does not guarantee an everlasting success in foreign trade. It must be used only as a temporary part of a bigger program for the industrial growth effort. But in Turkey we can not see increase in investments or establisment of new and more eloborate industries to be prepared for coping in the global market. In these circumtances each devaluation became the cause of the preceding devaluation. This is because price increases after the devaluation can not be prevented, production can not be increased, other industries that can be successful in foreign trade are not established e.t.c. VISo in this thesis the relationship between foreign exchange rate and foreign trade is analyzed statistically in a vektör autoregressif approach. Vektör auotoregressif function is chosen because when we are dealing with economical variables we know that these are interrelated that is we can not tell apart which one causes the other. In this method (VAR) each variable is considered endogeneous that is each variable is determined within the model. In this analysis first of all we must be sure that any variable that is important to explain the others is included in the model. Second all the variables should be stationary. For this Dickey-Fuller's unit root test is applied. In the analysis the variables are import (ITH) (aggregate import in million US Dollars), export (IHR) (excuding sectors other than production, in million USD), foreign exchange rate (DOV) (sum of 1 USD and 1,5 German Marks), inflation (TEF) (wholesale price index), production (SUE) (industry production index). All the variables are monthly data taken from The State Institute of Statistics. Also two other variables are derived from these ones - real exchange rate (RDOV) which is equal to DOV divided by TEF and change in real exchange rate (RKD) which is equal to first difference of DOV substracted by first difference of TEF. Then all the variables are diverted to their natural logaritm form. All the variables are found to be stationary in their first differences except RKD which is stationary in its original form. Also it was necessary to use a dummy variable for TEF in this stationarity analysis because of the big devaluation in April 1994. This dummy variable D1 is valued 1 for this month and 0 for the others. So then the names of the variables are now DLITH, DLIHR, DLDOV, DLTEF, DLSUE, DLRDOV. (Table 5.1). In order to test the existence of a long term relationship between these variables. This is tested for each double. Here the error term for linear combination of two variables should be integrated of order zero. According to this : a) Imports Granger cause production. b) Exchange rate Granger cause imports. c) Imports Granger cause exports. d) Production Granger cause exports. These error terms which are found to be stationary are included to the model as the error correction terms in order to introduce into the analysis the low-frequency information which is filtered out in first differencing. This is called the error-correction modelling. The following step was to determine the optimum number of lags on each variable. For this four different models are built Model 1 : DLITH, DLIHR, DLTEF, DLSUE, DLDOV /C, D1 Model 2 : DLTIH, DLIHR, DLTEF, DLSUE, DLRDOV /C, D1 Model 3 : DLITH, DLIHR, DLTEF, DLSUE, DLDOV /C, D1, S,...S(10) Model 4 : DLITH, DLIHR, DLTEF, DLSUE, DLRDOV /C, D1, S,...S(10) S is the dummy variable for seasonal adjustment. It is 1 for January and 0 for the other months. For the other months its suitable differences are used. Each model is estimated with 3,6,9 and 12 months lag lengths by use of Vektör Auto Regression function. Covariance matrics determinants of their error terms are calculated. Then the optimum number of lags are determined by use of Likelihood Ratio test VIIstatistic. (Table5.2). According to this for model 1 and 2 the optimum lag length is 12 while it is 9 for models 3 and 4. In the last part of the analysis impulse-response function analysis is performed. That is responses of the variables in the model to one standard deviation change in one of the variables are observed within a 36-month period. Model 1 : It is composed of imports, exports, wholesale price index, industry production index, exchange rate, fixed variable and the dummy variable for inflation. According to Johnnsen Cointegration Test there are two cointegration vectors. The responses of the variables to one standard deviation change in exchange rate (Figure 5.1, Page 36): a) In the first two months exchange rate increases with a very high acceleration. After that till the 28th month the rate of increase slows down and finishes after this month. It is almost constant for the remaining months. b) Wholesale price index increases like the exchange rate but the rate of increase is lower than that of the exchange rate for the first months. But it is observed that inflation increases with a higher slope which exceeds that of the exchange rate. So we can see that after the devaluation the price index changes cannot be controlled and so the effects of devaluation are nulled. c) Imports decrease in the first months when the devalution is still higher then the inflation, but increases almost continuously afterwards. d) While the response of imports is immediate it takes time till exports began to increase due to devaluation. This is consistent with the J-curve theory. After the first months' deterioration a rapid improvement which does not take long comes. Because the reel devalution can not be continued. e) Industry production index falls after the first month going after the fall in imports. That's because production in Turkey is very much dependant on imported raw material. The index increases gradually as the imports began to increase but this is a temporary improvement. For the rest of the period it even falls under its initial level. So we can say that devaluation causes the economy to shrink. Responses to a one standart deviation change in exports (Figure 5.2, Page 38): Exchange rate in the first 4 months decreases. Then it increases and then fall again very rapidly. In the end its level is much more lower than the initial level. Inflation almost for the whole period is in a increasing structure. It is only lower than the exchange rate for 1-2 months in the mids of the analysed period. Imports also has an increasing inclination. Exports for the whole period decreases. Industry production index increases and decreases with little movements but it is again lower than the initial level in the end. Responses to one standard deviation change in inflation (Figure 5.3,Page 39): Exchange /ate falls. Inflation decreases at first but increases for the rest of the period. Imports increases too much. Exports except the increase in the 3rd and 8th months keeps at a very low level. Production increases. vinModel 2 : Model 2 is composed of imports, exports, inflation, real exchange rate, constant term and the dummy. There is one cointegration vector. Responses to one standard deviation change in the real exchange rate (Figure5.4,Page 42): Real exchange rate increases in the first 3 months but decreases very much after that. Inflation increases consistently and much more than the real exchange rate. Imports decrease at first but increase as the real exchange rate decreases. Exports shows a very good improvement but deterioriates as the real exchange rate falls. Productiort shrinks. Model 3 : Model 3 is the seasonally adjusted form of Model 1. This is made by including seasonal adjustment dummies. There are 2 cointegration vectors. Responses to one standard deviation change in the exchange rate (Figure 5.5, Page 44): Exchange rate increases with a very high slope at first then stays almost constant after a fall. Here inflation stays behind the exchange rate. But this is not true regarding the accelaration of the increase. So again the real devaluation could not be preserved. Import declines too much at first. This is the period when the exchange rate is very high. After that imports increases with a very high slope. Export makes the J-Curve and then declines very rapidly to a very low level. Production also declines as the import declines then an improvement comes. Responses to one standard deviation change in exchange rate when the lag length is 12 instead of 9 (Figure 5.6, Page 46): Exchange rate increases. Inflation also increases but catches the exchange rate after the 10th month and exceeds it. Imports after the decline in the first 10 months exceeds initial level. Exports again makes the j- curve then a decline proceeds. Production shrinks. Model 4 : Model 4 is composed of export-import ratio, exchange rate, inflation, production, constant term and the inflation dummy. Responses to one standard deviation change in real exchange rate (Figure 5.7, Page 48): Exchange rate and inflation always increases. But inflation has a much more higher accelaration. Export- import ratio improves at first that is it increases but decreases afterwards. This fall is not that much dramatical though. Production declines. For the same model if we put real exchange rate instead of the exchange rate the responses (Figure 5.8, Page 49): Real exchange rate comes to a very high level at first but deterioration proceeds. Inflation increases all the time. Export -import ratio shows a very good improvement at first but declines afterwards. Again this decline is not very radical. Production decreases. As a conclusion we can see that devaluation causes the imports to decline while this improvement does not last long. On the other hand exports shows a characteristic j-curve in which improvement comes very rapidly. Again this is not a long lasting good result. These conclusion is also supported with the expot-import ratio analysis. The reason for this short-period improvent followed by a deterioration is inflation. Since this increase can not be stopped real devaluation can not be preserved. Also production shrinks. And so the same situtations that forced a devaluation at first occurs again. As a result in Turkey devaluation does not produce the aimed targets like improvement in foreign trade and economical growth. IX
Benzer Tezler
- Türkiye'nin dış ticaretinde 1980 sonrası dönemde meydana gelen yapısal değişiminin ekonometrik analizi (1973-1992)
Başlık çevirisi yok
ZEKERİYA MIZIRAK
- Türkiye iktisat politikalarının belirlenmesinde iktisadi kurum-kural ve kuruluşların rolleri
Başlık çevirisi yok
İBRAHİM GÜRAN YUMUŞAK
- Türk tütün politikasının Avrupa Topluluğu ortak tütün politikasıyla bütünleşmesi
Başlık çevirisi yok
HALUK TANRIVERDİ
Doktora
Türkçe
1990
Ekonomiİstanbul ÜniversitesiOrganizasyon ve İşletme Politikaları Ana Bilim Dalı
PROF. DR. YILDIRIM ÖNER