Publication: İstanbul ve ilçelerinde konut balonu oluşumu ve belirleyicileri
Abstract
Konut sektöründe görülen balonların hem ülke ekonomilerinde hem de küresel çapta önemli sonuçları olabilmektedir. 2008 krizi verilebilecek en güncel örneklerden biridir. Bu bağlamda İstanbul ve ilçelerinde, 2010-2019 döneminde konut balonu oluşumu incelenmiştir. Balon oluşumu incelenirken Phillips, Wu, Yu (2011) ve Phillips, Shi, Yu (2015) tarafından geliştirilen ve özyinelemeli test prosedürleri olan SADF ve GSADF testleri kullanılmıştır. Testler aynı zamanda balon içeren periyotların başlangıç ve bitiş tarihlerini tahmin eden prosedürleri içinde barındırmaktadır. Dolayısı ile il ve ilçe bazında balonların başlangıç ve bitiş tarihleri tahmin edilerek, bu periyotlarda ayrışma veya senkronizasyon olup olmadığı incelenmiştir. Sonuç olarak SADF testinde İstanbul’da ve otuz bir ilçede, GSADF testinde ise İstanbul’da ve otuz ilçede balon oluşumuna işaret eden sonuçlar elde edilmiştir. İl ve ilçe bazında tahmin edilen konut balonu periyotları incelendiğinde hem senkronize olan hem de ayrışan periyotlar tespit edilmiştir. Yapılan ikinci analizde İstanbul’da konut balonu oluşumuna sebep olan faktörler incelenmiştir. GSADF testinin sonuçları üzerinden yeni bir kukla değişken oluşturulmuştur. Bu değişken üzerinden dinamik probit ve logit modeller kullanılarak balon oluşumunda etkili faktörler incelenmiştir. Modele reel efektif döviz kuru, konut faizi gibi çeşitli değişkenler dahil edilmiştir. Ayrıca konut sektöründe etkili olabilecek politika değişiklikleri incelenerek, gerek görülmeleri halinde modele dahil edilmeleri sağlanmıştır. Dinamik probit ve logit model sonuçlarına göre konut faizinin konut balonu oluşumunda etkili faktör olduğu görülmüştür.
Housing bubbles may have direct implications for countries and the global economy. The 2008 global financial crisis can be given as a most recent example. In this study, the housing bubble existence in Istanbul and its districts were examined. This study covers the period 2010 - 2019. Recursive unit root tests proposed by Phillips, Wu, Yu (2011) and Phillips, Shi, Yu (2015) were used for housing bubble detection. These tests also offer a procedure to date-stamp the origination and termination of bubbles. Therefore synchronicity of bubble periods was investigated in Istanbul and its districts. SADF test predicts housing bubble in Istanbul and thirty-one districts. In the meantime, the GSADF test predicts housing bubbles in Istanbul and thirty districts. On the other hand, factors behind the housing bubble were investigated in Istanbul. For the analysis part, the output of the GSADF test was converted into a dummy variable. Dynamic logit and probit models were adopted to investigate factors. Corresponding independent variables were included in the analysis. Also, policy changes that may affect the housing sector were reviewed and effective policy changes were incorporated into the model. Dynamic logit and probit model results show that the home loan interest rate is the effective factor behind the housing bubble.
Housing bubbles may have direct implications for countries and the global economy. The 2008 global financial crisis can be given as a most recent example. In this study, the housing bubble existence in Istanbul and its districts were examined. This study covers the period 2010 - 2019. Recursive unit root tests proposed by Phillips, Wu, Yu (2011) and Phillips, Shi, Yu (2015) were used for housing bubble detection. These tests also offer a procedure to date-stamp the origination and termination of bubbles. Therefore synchronicity of bubble periods was investigated in Istanbul and its districts. SADF test predicts housing bubble in Istanbul and thirty-one districts. In the meantime, the GSADF test predicts housing bubbles in Istanbul and thirty districts. On the other hand, factors behind the housing bubble were investigated in Istanbul. For the analysis part, the output of the GSADF test was converted into a dummy variable. Dynamic logit and probit models were adopted to investigate factors. Corresponding independent variables were included in the analysis. Also, policy changes that may affect the housing sector were reviewed and effective policy changes were incorporated into the model. Dynamic logit and probit model results show that the home loan interest rate is the effective factor behind the housing bubble.