Türkiye’de Toplam Kamu Harcamaları ve Üretim İlişkisi

Çalışmanın amacıTürkiye’de kamu harcamalarıve toplam üretim düzeyi arasındaki ilişkiyi incelemektir. Bu amaçla 1924–2007 Türkiye yıllık kamu giderleri ve Gayri Safi Yurtiçi Hasıla veri setinden yararlanılmıştır. Değişkenler arasıilişki En Küçük Kareler Yöntemi EKKY , Johansen Eşbütünleştirme cointegrasyon Testi, Granger Nedensellik Testi ve Vektör Hata Düzeltme Modelinden Vector Error Correction Model elde edilen genelleştirilmişetki-tepki fonksiyonlarıile tahmin edilmiştir Johansen Eşbütünleşme Testi sonuçlarına göre iki değişken arasında uzun dönemli bir ilişki söz konusudur. Granger Nedensellik Testi değişkenler arasında iki yönlü bir nedensellik ilişkisine işaret etmektedir. Elde edilen etki-tepki impulse-response fonksiyonlarına göre Türkiye açısından Keynes hipotezi Wagner hipotezine göre daha baskındır. Çalışmada öncelikleri göz önüne alan, proaktif, sinerji yaratan, yeniliğe önem veren ve sonuçlarıhesaba katılmış, öncelikleri olankamu harcamalarının üretim sürecine pozitif etkileri olacağıvurgulanmaktadır.

The balancing role of the government to close the production and consumption gap is always in question during the depression economics. Together with the major developed countries, Turkish real sector also experienced an economic crisis during last years. Reevaluating this reality this paper aims to analyze the relationship between government spending and aggregate production level in Turkey. The role of the government has been discussed more by the globalization when the consumption and the labour force are considered. Especially after 2007 and 2008, there was an economic crisis in real sector. This signaled us that the loss functions should be strengthened by the productive government policies besides the monetary policy tools. The fiscal policy in this sense may minimize the social loss boomed by the recession and help for the stabilized economic environment. The fiscal policies which are assuring confidence, productive, sustainable may stabilize the cyclical production collapse. So we analyzed the role of fiscal policy on the adjustment mechanism both considering long and short run. The paper briefly gives literature review which considers this relationship. We also considered the role of government spending in economic growth theories, emphasized the theoretical background of the empirical evidence. In the third section, we presented the data and the methodology of the paper. The fourth section gives the empirical results. The fifth section of the paper evaluates the empirical evidence and the last section is the conclusion. To achieve the objectives of the paper, yearly Turkish government spending and Gross Domestic Product data covering the years from 1924 to 2007 are employed. Using the yearly data had given us an opportunity to regard the seasonal effects. The relationship among variables is estimated by Ordinary Least Squares Method OLS , Johansen Cointegration Test, Granger Causality Test and generalized impulse-response functions obtained from Vector Error Correction Model VECM . The variables are integrated in the same order which increased the robustness of the results. According to the Johansen Cointegration Test results, there is a long-run relationship among the two variables. Granger causality test results mention a two side causality relation among the variables. The result obtained from VECM indicates that the response of production to government spending is persistent at 0.8 within three years of time being. However the response of government spending to Gross Domestic Product GDP is persistent after the three years at 0.3. So according to the impulse-response functions, Keynes hypothesis is dominant compared with Wagner hypothesis and this result is consistent with the OLS estimates. The theoretical part of the article mostly lies on the Keynesian economic models. Keynesian models claim that the government spending policies may be used to affect the economic growth and adjust the cyclical movements. The endogenous growth models also emphasize the role of government for to increase the aggregate production level. The endogenous growth models such as Romer 1986 , Lucas 1988 , Barro 1990 and Rebelo 1991 consider the role of government for increasing the economic production. This paper also considers the conflict or debate about the Keynes and Wagner hypothesis. Wagner claimed that when the economy booms, the government spending will increase. However, according to Keynes, the government spending increases economic growth and Keynes does not take this variable exogenous. The long run positive effects of the government spending on production are consistent with the Barro 1990 , Barro ve Sala-i Martin 1990 endogenous growth models and Ravn et al. 2007 . So the results support the three theoretical hypotheses. First, supporting the endogenous growth models by its role of adjustment, government spending also has effects on the steady state growth level. Second, as mentioned before, Keynes and Wagner hypothesis are both valid, but the first one is dominant then the latter. Third the empirical results also support the hypothesis of Ram and Army curves. These curves claim that when the government spending increases more then the optimum point, the efficiency of the government spending on economic growth will be diminished. The results also indicates that on the time varying production function, the effects of government spending has a response of diminishing acceleration. As most of the developed countries, also in Turkey the expansionary role of the government spending had been arise after the economic crisis. In this frame, besides the discounts, it has been mentioned that the government should also support the diminishing demand. We obtained results that the government may take a role to adjust the transitory fluctuations in the demand side of the economy. As an emerging market in Turkey, especially in the periods when the uncertainty increases and the trust environment diminish, the interest rate policy may not affect the expected inflation rates. Consequently, the real interest rates may not stimulate the investment and consumption by the credit channel. In the periods like this, the target oriented fiscal policies may support the monetary policy. So the fiscal policies should be implemented for to increase the financial strength and help for the credit channel. The production and the employment level will be affected positively. In this paper, it is claimed that the government spending which are considering priorities, proactive, bringing synergy, taking account of the results, giving weight to innovation will affect production process positively. However there are two constraints concerning the Turkish data which prevents further analysis and interpretations. This is mostly because of deficiency of data. The first is in Turkey for the period we considered, there is no sectoral government spending data. Consequently we could not evaluate the type of the government spending whether it is productive or not. Second, it is not possible to confirm the dimension of government spending by region and province basis. The income elasticity of the government spending will change according to the population and geographical coordinates, so functional structure and intensity could not be confirmed. By the derived database, the efficiency and the contribution of the government spending in East Anatolia will be able to be determined. This will also contribute to the efficiency of the government investment in the future. The productive and unproductive government spending have been realized in the past will be evaluated in broader spectrum in this sense. If the database is prepared by data mining in the future, investment, health, communication spending will be analyzed deeper. We think that the fiscal data will be prepared by data mining will enlighten the retroactive problems