engleski

FLEXIBILITY OF REAL AND NOMINAL WAGES: CASE OF SELECTED WESTERN BALKAN COUNTRIES

When studying and forecasting macroeconomic developments, wages play an indispensable part. In this paper flexibility of real and nominal wages is explored for the following Western Balkan countries: Bosnia and Herzegovina, Croatia, Macedonia, Montenegro, and Serbia and the following New Member States: Bulgaria, Romania and Slovenia. The main goal of this study was to identify whether selected Western Balkan countries show different behaviour in comparison to the selected New Member States with the respect to the response of nominal and real wages to shocks. The data used for the empirical analysis were the data on gross monthly wages for total economy expressed as nominal and real indexes with the previous period used as the base (previous period =100) taken from the WIIW Monthly Database. Since the data for all countries in the sample are available from January 2004 to June 2015, average monthly growth rates for that period were calculated. Namely, the real growth rates are below 0.5 per cent in all analysed countries but Montenegro, Romania and Serbia, where the real growth rates range from 0.51 per cent in the case of Montenegro to 0.55 per cent for Serbia. Since the main goal of this study is to access the relationship of real and nominal wages in selected Western Balkan countries and New Member States empirically, the vector autoregression (VAR) model is applied.Following the VAR procedure, Dickey-Fuller and Phillips- Perron unit root tests using various specifications were conducted and in almost all cases we cannot reject the null of a unit root. Similarly, we applied the Johansen cointegration test using different lag lengths. Also, in almost all cases, trace and max-eigen value statistics indicated no cointegration at 5% significance level. Accordingly, all series are first-differenced and seasonally adjusted using tramo-seats method and the stationary time series are obtained. The next step of the empirical analysis was to estimate eight bivariate VAR models (one for each country in the sample). Also, the model diagnostics tests were conducted. Furthermore, from obtained VAR models, impulse response function using orthogonalized shocks based on the Cholesky decomposition were derived. The results of the analysis have shown that the response of nominal wage shocks on nominal wages is initially positive in all countries in the sample. Similarly, the effect of a nominal shock on the real wage growth diminishes rather quickly (after 3 to 4 periods) in all countries. When comparing the group of NMS countries to the group of Western Balkan countries, the response remains slightly under the nought line for the first group, whereas fluctuating around zero for the Western Balkan states. Furthermore, a shock on real wage growth exhibits a positive effect on nominal wage growth in the first period for all countries in the sample but Romania, where that effect is small and negative. Lastly, a shock on real wage growth is initially positive in all analysed countries, but there is some fluctuation around the zero line which is more pronounced in the Western Balkan group of states and Bulgaria.

nominal wages ; real wages ; vector autoregression model ; Western Balkan countries

Ovaj rad je u potpunosti podržan od strane Hrvatske zaklade za znanost kao dio znanstvenog projekta „Statistical Modelling for Response to Crisis and Economic Growth in Western Balkan Countries“ (STRENGTHS) ; Broj projekta: HRZZ-IP-2013-11-9402, Voditeljica: prof. dr. sc. Ksenija Dumičić.