The Effect of Intergovernmental Transfers on Income Per Capita in Turkey: A Panel Data Analysis
Zeynep Burcu Bulut ÇevikThis study tests whether intergovernmental transfers affect the income per capita of Turkey with panel data analysis for the period of 2010 to 2017 including the cities of Turkey. The Hausman test gives supporting results about using the fixed effects model. According to the fixed effects model results, when the transfers increase by 1%, income increases by 0.04% two periods later with the inclusion of control variables. The possible reasons for the significant but small impact of transfers on income are three-fold: i) the intergovernmental transfers may not be used efficiently, ii) municipalities may not update the resource allocation successfully with the increasing urbanization rate, iii) the increase in the number of immigrants may prevent the positive effect of transfers on income through labor market channels. Furthermore, the effects of transfers on income differentiate with respect to being a metropolitan municipality or not.
Türkiye’de Yönetimler Arası Transferlerin Kişi Başı Gelir Üzerindeki Etkisi: Panel Veri Analizi
Zeynep Burcu Bulut ÇevikBu çalışmada merkezi yönetimden yerel yönetimlere gönderilen yönetimler arası transferlerin, kişi başı gelir üzerine etkisi olup olmadığı test edilmiştir. 2010 ile 2017 yılları arasındaki şehir bazlı panel veri analizi yöntemi olarak sabit ekili model kullanılmış, Hausman testi sonuçları da söz konusu model yöntemini destekleyici sonuçlar vermiştir. Bu model sonuçlarına göre, Türkiye’de kişi başı geliri etkileyen diğer faktörler kontrol edildikten sonra, bu dönem transfer miktarındaki %1’lik artışın, iki dönem sonra kişi başı geliri %0.04 arttırdığı gözlenmiştir. Etki şiddetinin anlamlı fakat düşük olması üç şekilde açıklanabilir: i) Transferlerin harcama kanalında yeterince etkili ve etkin kullanılmaması, ii) Türkiye’de hızla artan kentleşmeye rağmen, il belediyelerin kaynak dağıtımını aynı hızda düzenleyememesi, iii) Son yıllarda artan göçmen sayısının transferlerin işgücü piyasasına yapabileceği olası pozitif etkiyi engellemesi. Ayrıca, büyükşehir belediyesi olan iller ile büyükşehir olmayan belediyelere verilen transferlerin gelirlere etkisinin farklılaştığı sonucuna ulaşılmıştır.
Intergovernmental transfers, which aim to reduce the fiscal imbalances between regions as well as to annihilate the local externalities, do not only affect the local economy but also influence the whole economy. They may affect the per capita income through different channels such as i) effect on labor market, ii) effect on human capital, iii) effect on growth and inequality. Firstly, intergovernmental transfers may increase local investment, which may influence the labor market positively. This positive influence may increase the per capita income level. Secondly, the transfers, spent on cultural and art activities, will have a positive impact on human capital, which will ultimately enhance the overall GDP level. Through this way, per capita income level will increase. Lastly, the inequality reducing and growth enhancing effects of transfers will affect the per capita income positively.
The literature about intergovernmental transfers is mainly focused on the law amendments and their possible effects by either discussions or by descriptive statistics. The econometric studies, which use municipality based data, are limited because of data availability issues. Especially since 2014, when a new law (Law No. 6360) was implemented, Turkish municipality-based data related to transfers has not been released publicly anymore. Due to this reason, the data related to municipality based transfers is calculated by utilizing Law No. 5779 and Law No. 6360. Other contribution of this study is that as far as I know, there is no other study which investigates the effect of transfers on income level for Turkey.
This study attempts to find out the effect of intergovernmental transfers on per capita income level between 2010 to 2017 for Turkey. The amount of municipality-based transfers is calculated via Law No. 5779 for the period of January 2010 to March 2014, whereas for the period of April 2014 till December 2017, Law No. 6360 is utilized. The amount of transfers is calculated not only city by city but also on a district basis since District Municipalities in Metropolitan areas distribute some of their portion to Metropolitan Municipalities.
In this study, the fixed effects model is preferred in panel data analysis, since the 81 cities of Turkey have different economic, spatial or sociological structures. The results of the Hausman test support this choice of model. In order to see the pure effect of transfers on income level, certain control variables are used. These control variables are trade share as a percent of GDP, inflation rate of CPI and population. They are chosen according to data availability as well as their possible significant effect on income level. Since the income level can be represented by GDP level, per capita GDP is used to represent the per capita income level as a dependent variable.
According to the fixed effect model results, the intergovernmental transfers affect the per capita income level significantly but with a small coefficient of 2-lag time. When the intergovernmental transfers as a percent of GDP in this period increases by 1%; two periods later, the per capita income level increases by 0.04%. This small and lag effect result shows the existence of obstruction in some channels of transfers to investments and labor market. Especially after 2011, the rise in the number of Syrian refugees may prevent the effect of transfers on the labor market. In addition, the resources of municipalities may not be distributed appropriately for the new local needs as the pace of urbanization rate increases in Turkey. Moreover, the low share of local expenditure over GDP may cause this small effect.
Furthermore, the effect of transfers on income level is distinguished according to the type of municipality: metropolitan or non-metropolitan. In non-metropolitan municipalities, the effect is significant but negative, whereas in metropolitan municipalities, it is small, significant but positive. This high differentiation shows the sharp distinction in local expenditures or resource allocation between metropolitan and non-metropolitan municipalities. Especially, the negative sign points out the possible existence of a moral hazard problem in non-metropolitan municipalities.