Ekonometride Güncel Yöntemler ve Uygulamalar
1923’den Günümüze İşsizlik Histerisinin Geçerliliğinin Analizi
Ebru Çağlayan Akay, Hoşeng Bülbülİşsizlik, ülkelerin ekonomik durumu hakkında bilgi veren önemli göstergelerden biridir. Histeri etkisine göre ekonomik şoklar işsizlik serisi üzerinde kalıcı etkiler bırakmakta ve işsizlik serisi birim kök süreci izlemektedir. Türkiye işsizlik sorununun ciddi olduğu ülkelerden biridir. Bu nedenle işsizlik histerisi teorisinin geçerli olup olmadığı araştırmacılar tarafından oldukça ilgi çeken bir konu olmuştur. Söz konusu bu ilgi ekonometri literatüründe işsizlik oranlarının çeşitli birim kök testleri ile incelenmesine neden olmuştur. Bu çalışmanın amacı, 1923 yılından günümüze Türkiye için işsizlik histerisi hipotezinin geçerliliğini incelemektir. Bu amaç için, Güriş (2018) tarafından önerilen ve hem doğrusal olmamayı hem de yapısal kırılmayı birlikte ele alan Fourier Kruse Testi kullanılmıştır. Çalışmadan elde edilen bulgular, Türkiye’de Cumhuriyet dönemi itibariyle işsizlik histerisi hipotezinin geçerli olduğunu ve yaşanan krizlerin, şokların ve kırılmaların işsizlik üzerinde kalıcı etkiler meydana getirdiğini göstermektedir. Bu yapısal değişimlerin kalıcı bir yapıya sahip olması nedeniyle emek piyasası için uygulanacak olan politikalarla işsizlik üzerinde kalıcı etkiler söz konusu olacaktır.
An Analysis of the Validity of Unemployment Hysteresis from 1923 to the Present
Ebru Çağlayan Akay, Hoşeng BülbülUnemployment is one of the most important indicators of a country’s economic condition. According to the hysteria effect, economic shocks can permanently affect a country’s unemployment series. In this case, the unemployment series follows the unit root process. The validity of the theory of unemployment hysteresis is a subject of considerable debate among researchers. This interest has led researchers to examine unemployment rates using unit root tests with differing properties. According to Phelps’ (1967) preliminary natural unemployment rate hypothesis, the unemployment rate tends to return to a previous balance. The natural unemployment rate hypothesis was revised in lieu of the continuously high unemployment rates experienced by European countries in the mid-1970s. For the first time, Tobin (1972) stated that real changes in unemployment rates had a damaging and lasting effect on the natural unemployment rate hypothesis. Blanchard and Summers (1986) developed the unemployment hysteresis hypothesis to counter the natural unemployment rate hypothesis. Unemployment hysteria is defined as the high dependence of current unemployment on previous unemployment. Moreover, unemployment hysteria can be defined as the permanent effects of temporary shocks on the unemployment rate. If the current unemployment rate is dependent on past values with coefficients close to 1, it is considered to indicate a hysteria effect. This means that the unemployment rate series has a unit root. Thus, the term hysteria can also be termed as unit root hysteresis. If the unemployment hysteria hypothesis is valid, a shock in the economy will exert a lasting impact on unemployment rates, and unemployment rates will not return to the previous level. In this case, the unemployment rate series follows a nonstationary process. Another approach, known as the structuralist approach, was developed by Phelps in 1994. According to this approach, the natural unemployment rate is affected by structural factors. Unit root tests are widely used to test these theories.
Unemployment arose in Turkey in 1970 and remains the nation’s most significant challenge. Since the end of 2010, financial stability has been one of the fundamental objectives in Turkey. Related policies aim to achieve macro-financial balance without compromising price stability; however, the failure of the Central Bank to reach the targeted inflation level leads to increased unemployment. Despite economic growth in Turkey in recent years, the unemployment rate has increasingly persisted, reaching double digits at present.
This study examines the validity of the unemployment hysteria theory in the case of Turkey, using data for the period 1923–2019 and unit root tests. Several studies have investigated unemployment hysteria in Turkey applying different periods and unit root tests.
There appears to be no consensus on the validity of unemployment hysteria upon examining the study findings. It is apparent that unit root tests used in these studies do not simultaneously consider both nonlinearity and structural breaks. Failure to select a unit root test that is appropriate to the structure of the series may result in inaccurate estimation results. In this study, unlike previous investigations, the validity of unemployment hysteresis was examined considering both nonlinearity and structural breaks from 1923 to the present using the Fourier Kruse test proposed by Güriş (2018). This test examines both nonlinearity and structural breaks simultaneously. The notable feature of this test as opposed to KSS (2003) and Kruse (2011) tests is its higher test power.
The result of the Fourier Kruse test revealed that the unemployment hysteria hypothesis is valid for Turkey. This finding indicates that Turkey’s unemployment rate is permanently affected by shocks. Therefore, enduring and deep-rooted solutions are required to reduce the unemployment rate. As a result of shocks in the economy, unemployment rates will not be able to return to previous levels and will tend to remain at a newly established equilibrium level. The negative effects of these structural changes must be reduced, and to do so, new structural reforms to increase the efficiency of the labor market are necessary. Such policies could also have lasting effects on unemployment rates.