NAIRU%20Estimation%20in%20Romania%20(including%20a%20comparison%20with%20other%20transition%20countries)

1
NAIRU Estimation in Romania (including a
comparison with other transition countries)
THE ACADEMY OF ECONOMIC STUDIES DOCTORAL SCHOOL
OF FINANCE AND BANKING

• Student Otilia Iulia Ciotau
• Supervisor Professor Moisa Altar

BUCHAREST,2004
2
Contents

• The papers incentives
• Features of unemployment rate in Romania
• Estimation methods
• Comparison of results
• Concluding remarks

3
Natural Rate and NAIRUIs there any difference?

• Natural rate of unemployment - Friedman (1968),
• Phelps (1968) the level of unemployment to
  which the
• economy would converge in the long run in the
  absence
• of structural changes to the labor market
• NAIRU (Non-Accelerating Inflation Rate of
  Unemployment) - Modigliani and Papademos (1975)
  the rate of unemployment at which there is no
  tendency for inflation to increase or decrease

4
Are NAIRU estimates useful?

• I have become convinced that the NAIRU is a
  useful analytic concept. It is useful as a theory
  to understand the causes of inflation. It is
  useful as an empirical basis for predicting
  changes in the inflation rate. And, it is useful
  as a general guideline for thinking about
  macroeconomic policy.
• Stiglitz, J. , Reflections on the Natural
  Rate Hypothesis

5
Features of Unemployment Rate in Romania

• The labor market have been strongly affected by
  the adjustment process from centrally planned to
  market-oriented economies
• Mass lay-offs
• Issues about underestimation of unemployment rate
  (masked unemployment, methodology)
• Labor force working in informal economy
• Active measures for unemployment mitigation (Law
  no76/2002).

6
Unemployment Rate in Romania (19941 20041)
7
Estimation methods

• Statistical methods
• Hodrick-Prescott Filter
• Univariate UC
• Bivariate UC (Okuns approach)
• Multivariate UC
• Reduced-form methods
• Phillips curve with constant NAIRU
• Elmeskov method
• Phillips curve with time-varying NAIRU

8
Hodrick-Prescott ( 1600)
9
Univariate UC for Romania

• Fitted model

10
Seasonal component and intervention variable

• The seasonal pattern is the sum of s/2 (two for
  quarterly data) cyclical components, with
  frequencies
• 
  
  
  

11
Back
12

• Period 25.9808 ( 6.49521 'years')
• Amplitude 0.0142053
• Rho 0.94072
• Variance 0.000111226

Estimated parameters for the cycle
13
Unemployment Rate Forecast
14
Univariate UC for Czech R.and Lithuania

• Fitted model
• Intervention variables Irr 2002. 1 Irr 2003. 4
  
• for Czech R.

15
NAIRU (UC-1 trend) Czech R.
16
UC-1 slope for Czech R.
17
Unemployment gap Czech R.
18
Bivariate UC unemployment rate and real GDP
(19941-20033)

• Okuns law
• SUTSE (Seemingly Unrelated Time Series
  Equations)

• Intervention variable
• For unemployment series irr 20021
• For GDP level 19971.

19
Common cycles
20
NAIRU (trend UC-2) and unemployment gap (cycle
UC-2)
UC-1 NAIRU
21
Potential Output (trend UC-2) and Output Gap
22
Unemployment Rates in Transition Economies
23
Multivariate framework

• SUTSE model for six countries

• Estimated parameters for the similar cycle
• Rho 0.96
• Period 21.56 (5.38987 years)

24
Correlation between cyclical components

• Czech R.
• Hungary 0.983
• Lithuania -0.244 -0.146
• Polonia 0.041 0.137 0.958
• Slovakia 0.176 0.104 0.459 0.523
• Romania 0.548 0.441 -0.004 0.155 0.848

25
Short-run commovements between unemployment rate
in Czech R. and Hungary
26
Correlation between seasonal components

• Czech R.
• Hungary 0.176
• Lithuania -0.151 0.669
• Polonia 0.218 0.799 0.504
• Slovakia -0.019 0.888 0.826 0.673
• Romania 0.049 0.806 0.655 0.942 0.791

27
Seasonal comovements between unemployment rate in
Poland and Romania
28
Seasonal components in unemployment rate Romania
29
Seasonal components in unemployment rate Poland
30
Seasonal comovements between unemployment rate in
Hungary and Slovakia
31
Seasonal components in unemployment rate Hungary
32
Seasonal components in unemployment rate Slovakia
33
NAIRU (UC-2 trend) and unemployment gap in Romania
Amplitude 0.5306
34
NAIRU (UC-2 trend) and unemployment gap in Czech
R.
Amplitude 0.94145
35
NAIRU (UC-2 trend) and unemployment gap in
Lithuania
Amplitude 0.74114
36
NAIRU (UC-2 trend) and unemployment gap in Poland
Amplitude 0.552935
37
NAIRU (UC-2 trend) and unemployment gap in
Slovakia
Amplitude 0.1882
38
NAIRU (UC-2 trend) and unemployment gap in Hungary
Amplitude 0.32301
39
Testing for hysteresis

• ADF, PP we cannot reject the unit root
  hypothesis for any of the six series
• Zivot and Andrews (1992) unit root test with
  structural break endogenously determined (prg.
  EViews)

40
Zivot, Andrews test results
Country AIC Model A AIC Model B AIC Model C AIC Model D Best model Estimated for best model Tmin Unit root test outcome
Czech R. 1.14232 1.17235 1.09873 1.16907 C 0.594029 3.8940 Not significant at 10
Hungary 0.29012 -0.02973 0.01730 0.43079 B -0.60811 -18231 Significant at 5
Poland 1.87665 1.32674 1.46456 1.79896 B -0.06758 4.7456 Not significant at 10
Slovakia 1.39449 1.85756 1.15908 1.33907 D 0.607344 7.3205 Significant at 1
Lithuania 2.98971 2.62469 1.87074 2.87342 C 0.188624 4.3976 Not significant at 10
Romania 2.78114 2.76273 2.13581 3.21371 C -0.32009 8.2090 Significant at 1
41
Breakpoints endogenously determined by ZA test
Country Breakpoint Significance
Czech Republic 1998 q2 Not significant at 10
Hungary 2001 q2 Significant at 5
Poland 1998 q2 Not significant at 10
Slovakia 1998 q4 Significant at 1
Lithuania 2003 q3 Not significant at 10
Romania 2001 q4 Significant at 1
42
Reduced-form methods

• Triangle model of inflation (Gordon)

where
43
Constant NAIRU (u 6.98)
Method Least Squares Method Least Squares Method Least Squares Method Least Squares Method Least Squares
Sample 19941 20041 Sample 19941 20041 Sample 19941 20041 Sample 19941 20041 Sample 19941 20041

Variable Coefficient Std. Error t-Statistic Prob.
C 9.932965 5.102861 1.946548 0.0599
DINF(-2) -0.404529 0.104494 -3.871320 0.0005
SOM(-1) -1.423750 0.535690 -2.657787 0.0119
DSOM -2.379647 1.320497 -1.802084 0.0804
CFE 0.220746 0.038290 5.765139 0.0000
OILM(-1) 0.166008 0.063716 2.605430 0.0135
REER -0.330547 0.128512 -2.572106 0.0146
R-squared 0.651039
Adjusted R-squared 0.589458
44
Elmeskov Method

• simplified accelerationist version of Phillips
  curve
• An estimate of is obtained for any two
  consecutive
• periods as which is
  substituted in (1) to give

45
Elmeskov Method
46
Time-varying NAIRU

• The basic inflation equation
• is supplemented by a second equation that
  explicitly allows the NAIRU to vary with time
• The method of estimation is Kalman filter with a
  standard deviation of 0.2 for the state variable
  as a smoothing prior (Gordon 1997).

47
Time-varying NAIRU
48
Comparison of results
HP Univar.UC Bivar.UC Multivar.UC Recursive Elmeskov Kalman1 Kalman2
2002.01 9.3396 9.1483 9.7034 9.0274 9.3396 9.5058 7.0909 3.6776
2002.02 9.192 9.4607 9.8096 9.0080 9.1919 9.3371 7.0898 3.6412
2002.03 9.0318 9.1727 9.8397 8.9093 9.0318 9.1498 7.0962 3.6078
2002.04 8.862 8.9239 9.9713 8.8573 8.8619 8.9064 7.0996 3.5797
2003.01 8.6849 8.7293 9.8537 8.6359 8.6848 8.6244 7.105 3.5547
2003.02 8.503 8.5727 9.8187 8.5837 8.5029 8.3205 7.1205 3.5241
2003.03 8.318 8.5915 9.8627 8.6024 8.3183 8.0061 7.1322 3.4987
2003.04 8.1327 8.5939 8.4646 8.1327 7.6892 7.1481 3.4794
2004.01 7.947 8.3691 8.4149 7.9469 7.3718 7.1481 3.4625
49
Conclusion

• The Romanian NAIRU is lower than in the other
  countries studied and also rather small comparing
  to Europe
• NAIRU in Romania is smooth comparing to the other
  five countries
• Uncertainty of the results

50
Further direction for research

• Estimating NAIRU based on unemployment rate
  calculated according to international accepted
  standard
• Using the series from claimant count just for
  improving the accuracy in a bivariate UC model
• Harvey and Chung(2000),
• Estimating the underlying change in unemplyment
  in the Uk

51
Predictive-testing (Romania UC-1)
52
Predictive-testing (Romania UC-1)
53

Auxiliary observation residuals (Romania UC-1)
54
CUSUM test UC-1
55
Bivariate UC better fit for unemployment than
in univariate case
56
Predictive testing bivariate GDP
57
Forecast for GDP and unemployment rate
58
Predictive testing for multivariate UC-1
59
Forecast multivariate UC