Labour market effects of fixed-term employment contracts [Elektronische Ressource] : microeconometric analyses for West Germany / Tobias Hagen

-

English
266 Pages
Read an excerpt
Gain access to the library to view online
Learn more

Description

Labour Market Effects of Fixed-Term Employment Contracts – Microeconometric Analyses for West Germany Inaugural-Dissertation zur Erlangung des Doktorgrades des Fachbereichs Wirtschaftswissenschaften der Johann Wolfgang Goethe-Universität Frankfurt am Main Tobias Hagen Kelkheim 2004 2 Contents 1 Introduction 13 2 Fixed-Term Employment Contracts in Germany: Definition, Institutional Background, and Empirical Relevance 17 2.1 Definition of Fixed-Term versus Permanent Employment Contracts 17 2.2 Institutional Background in Germany 18 2.3 Empirical Relevance of Fixed-Term Contracts in West Germany 21 3 Methodological Background: Identification of the Effects of Institutions and Policy Interventions 29 3.1 Introduction: Estimation of Causal Effects of Binary Treatments Using Microeconometric Methods 29 3.2 Basics 30 3.2.1 The Evaluation Problem in General and the Parameters of Interest 30 3.2.2 Regression: Homogeneous Versus Heterogeneous Effects and Sources of Selection Bias 32 3.2.3 Social Experiments 35 3.2.4 Natural Experiments 35 3.2.5 SUTVA: Possible General Equilibrium Effects and Indirect Effects of Fixed-Term Contracts 36 3.2.6 Basic Approaches for the Generation of Control Groups 37 3.3 Evaluation Methods Requiring Cross Sectional Data 38 3.3.1 Propensity Score Matching 38 3.3.2 Parametric Regression Methods Versus Matching 45 3.3.

Subjects

Informations

Published by
Published 01 January 2005
Reads 10
Language English
Document size 12 MB
Report a problem

Labour Market Effects of Fixed-Term Employment
Contracts – Microeconometric Analyses
for West Germany


Inaugural-Dissertation
zur Erlangung des Doktorgrades
des Fachbereichs Wirtschaftswissenschaften
der Johann Wolfgang Goethe-Universität
Frankfurt am Main


Tobias Hagen


Kelkheim


2004






















2 Contents

1 Introduction 13

2 Fixed-Term Employment Contracts in Germany: Definition, Institutional
Background, and Empirical Relevance 17
2.1 Definition of Fixed-Term versus Permanent Employment Contracts 17
2.2 Institutional Background in Germany 18
2.3 Empirical Relevance of Fixed-Term Contracts in West Germany 21

3 Methodological Background: Identification of the Effects of Institutions and
Policy Interventions 29
3.1 Introduction: Estimation of Causal Effects of Binary Treatments Using
Microeconometric Methods 29
3.2 Basics 30
3.2.1 The Evaluation Problem in General and the Parameters of Interest 30
3.2.2 Regression: Homogeneous Versus Heterogeneous Effects and Sources of
Selection Bias 32
3.2.3 Social Experiments 35
3.2.4 Natural Experiments 35
3.2.5 SUTVA: Possible General Equilibrium Effects and Indirect Effects of
Fixed-Term Contracts 36
3.2.6 Basic Approaches for the Generation of Control Groups 37
3.3 Evaluation Methods Requiring Cross Sectional Data 38
3.3.1 Propensity Score Matching 38
3.3.2 Parametric Regression Methods Versus Matching 45
3.3.3 Instrumental Variable Approaches and Selection Models 46
3.4 Evaluation Methods Requiring Panel Data 47
3.4.1 Before-After Estimator 47
3.4.2 Difference-in-Differences Estimator 49
3.5 Pre-Program Test 50
3.6 Summary 51






3 4 The Role of Fixed-Term Contracts in Labour Demand 52
4.1 Overview: Why Do Firms Use Fixed-Term Contracts? 52
4.2 Theoretical Considerations 55
4.2.1 Introduction 55
4.2.2 Protection Against Dismissal in Economic Theory and Structure of
Adjustment Costs in Labour Demand 56
4.2.3 Dynamic Labour Demand with Firing Costs and Availability of Fixed-
Term Contracts 57
4.2.4 Matching and Equilibrium Labour Market Models with Fixed-Term
Contracts 61
4.2.5 Further Considerations on Institutional Reasons for Using Fixed-Term
Contracts 66
4.2.6 Summary and Conclusions from the Theoretical Considerations 68
4.3 Previous Empirical Results on the Effects of Dismissal Protection and Fixed-Term
Contracts 69
4.4 Why do Employers Use Fixed-Term Contracts? Evaluating the Effects of Firing
Costs of Permanent Work on the Use of Atypical Work 76
4.4.1 Introduction 76
4.4.2 Dataset, Model Specification, and Estimation Technique 77
4.4.3 Estimation Results 86
4.4.4 Summary of the Empirical Analysis of the Firms’ Use of Atypical Work 91
4.5 Empirical Analysis of the Role of Fixed-Term Contracts in Worker and Job Flows 92
4.5.1 Introduction 92
4.5.2 Methodology 94
4.5.3 Dataset and Weighting 99
4.5.4 Results 100
4.5.5 Summary of the Empirical Analysis of the Role of Fixed-Term Contracts
in Worker and Job Flows 110
4.6 Conclusions 111

5 Do Temporary Workers Receive Risk Premiums? – Effects of Fixed-Term
Contracts on Working Conditions and Wages in the Short-Run 112
5.1 Introduction 112
5.2 Theoretical Considerations 113
5.2.1 Compensating Wage Differential for Workers on Fixed-Term Contracts 113
5.2.2 Wage Penalty for Workers on Fixed-Term Contracts 114
5.2.3 Worker Preferences for Fixed-Term Contracts 119
5.2.4 Heterogeneity of Fixed-Term Contract Jobs 119
4 5.2.5 Use of Subjective Outcome Variables in Economic Analysis 120
5.3 Previous Empirical Results 121
5.4 Data Base, Estimation Sample, and Descriptive Statistics 124
5.5 Econometric Approach 137
5.5.1 Characterising the Selection Problem 137
5.5.2 Attempts to Account for Selection on Unobservables 137
5.5.3 Choice of Conditioning Variables 139
5.5.4 Further Specification Issues: Other Selection Problems 141
5.5.5 Implementation of the Propensity Score Matching Estimator 141
5.6 Determinants of the Type of Contract: Estimation of the Propensity Score 143
5.6.1 Model A: Using only Pre-Treatment Variables 143
5.6.2 Model C: Accounting for Job and Employer Attributes 149
5.7 Effects of Fixed-Term Contracts: Results of the Matching Estimator 156
5.7.1 Choice of the Matching Estimator and Checks on the Balancing Property 156
5.7.2 Effects of Fixed-Term Contracts for Men 161
5.7.3 Effects of Fixed-Term Contracts for Women 169
5.8 Summary and Conclusions 173

6 Do Fixed-Term Contracts Increase the Long-Term Employment
Opportunities of the Unemployed? 175
6.1 Overview: Are Fixed-Term Contracts Stepping Stones for the Unemployed or
Dead-Ends? 175
6.2 Theoretical Considerations 176
6.2.1 Basic Concepts: Job Search Theory and Determinants of Unemployment
Duration 176
6.2.2 Under Which Conditions Do Job Searchers Enter into Fixed-Term Contract
Jobs? 178
6.2.3 Why Should Fixed-Term Contracts be Stepping Stones Towards
Permanent Positions? 181
6.3 Fixed-Term Contracts and the Re-Employment Probabilities of the Unemployed 183
6.3.1 Introduction 183
6.3.2 Previous Results: Unemployment Duration Analyses Distinguishing
Between Employment Contracts 184
6.3.3 Modelling Framework: Discrete Competing Risks Hazard Rate Model 186
6.3.4 Data Base and Variables 188
6.3.5 Estimation Results of the Competing Risks Hazard Rate Model 193
6.3.6 Summary: Fixed-Term Contracts and the Re-Employment Probabilities of
the Unemployed 199
5 6.4 Effects of Entering into Fixed-Term Contract Jobs: Are Fixed-Term Contracts
Stepping Stones? 200
6.4.1 Introduction 200
6.4.2 Previous Studies: Causal Effects of Fixed-Term Contracts on Future
Employment Prospects 200
6.4.3 Econometric Approach 202
6.4.3.1 The Counterfactuals of Interest and the Policy Questions 202
6.4.3.2 Implementation of the Propensity Score Matching Estimator 207

6.4.4 Definition of the Outcome Variables 210
6.4.5 Estimation Results of the Propensity Score Equation (Discrete Hazard Rate
Model) 212
6.4.6 Estimation Results of the Matching Estimator 214
6.4.6.1 Matching Quality 215
6.4.6.2 Mean Effects of Fixed-Term Contracts 218
6.4.6.3 Heterogeneous Effects of Fixed-Term Contracts 224

6.4.7 Summary: Effects of Entering into Fixed-Term Contracts on the Long-
Term Employment Opportunities of the Unemployed 231
6.5 Conclusions 233

7 Summary and Conclusions 235

8 Appendix 239

9 References 253















6 List of Figures

Figure 1: Proportion of FTCs in Total Dependent Employment in West Germany.................... 23
Figure 2: Index of Permanent Contract and FTC Employment (1995=100%) ........................... 24
Figure 3: Kernel Density Estimation of the Hourly Wage (€) Distribution of Men by Type
of Contract – Unrestricted Tenure ................................................................................... 133
Figure 4: Kernel Density Estimation of the Hourly Wage (€) Distribution of Men by Type
of Contract – Tenure 2................................................................................................... 133
Figure 5: Kernel Density Estimation of the Hourly Wage (€) Distribution of Women by
Type of Contract – Unrestricted Tenure.......................................................................... 134
Figure 6: Kernel Density Estimation of the Hourly Wage (€) Distribution of Women by
Type of Contract – Tenure 2.......................................................................................... 134
Figure 7: Employment Effects – DEF.1.................................................................................... 219
Figure 8: Employment Effects – DEF.2.................................................................................... 219
Figure 9: Unemployment Effects – DEF.1................................................................................ 220
Figure 10: Unemployment Effects – DEF.2.............................................................................. 220
Figure 11: Out-of-labour-force Effect – DEF.1 ........................................................................ 221
Figure 12: Out-of-Labour-Force Effect – DEF.2 ...................................................................... 222
Figure 13: Employment Effects – Women (DEF.2).................................................................. 225
Figure 14: Unemployment Effects – Women (DEF.2) ............................................................. 226
Figure 15: Out-of-Labour-Force Effect – Women (DEF.2)...................................................... 226
Figure 16: Employment Effects – Age 32 (DEF.2) ............................................................... 228
Figure 17: Unemployment Effects – Age 32 (DEF.2)........................................................... 228
Figure 18: Out-of-Labour-Force Effects – Age 32 (DEF.2) ................................................... 229
Figure 19: Employment Effects – Workers With Formal Qualification (DEF.2)....................... 230
Figure 20: Unemployment Effects – Workers With Formal Qualification (DEF.2)................... 230
Figure 21: Out-of-Labour-Force Effects – Workers With Formal Qualification (DEF.2).......... 231

List of Tables

Table 1: Duration of FTCs in 1998 ............................................................................................. 25
Table 2: Proportion of FTCs in Total Dependent Employment in 1998/99................................ 27
Table 3: Overview of Matching and Equilibrium Models with FTCs ........................................ 64
Table 4: Share of Establishments Using Atypical Work............................................................. 83
Table 5: Share of Establishments Employing Atypical Workers by
Number of Employees and Industry .................................................................................. 84
Table 6: Means of Adjustment to Expected or Unexpected Demand Changes During the
Year in West Germany in 1996 ......................................................................................... 84
Table 7: Means of Adjustment to Increasing and Decreasing Demand in West Germany in
1993 ................................................................................................................................... 85
Table 8: Determinants of Employing FTC, FL, or TWA Workers............................................. 88
7
‡‡££‡Table 9: Effects of the Increase in the Minimum Employment Threshold Level
for the Protection Against Dismissal Law in October 1996 .............................................. 90
Table 10: Means of Job and Worker Flow Rates for Total Employment ................................. 101
Table 11: Means of Job and Worker Flow Rates by Employment Trend................................. 102
Table 12: Means of Job and Worker Flows by Type of Contract ............................................. 103
Table 13: Means of Job and Worker Flow Rates by Type of Contract and
Employment Trend .......................................................................................................... 105
Table 14: Decomposition of Worker Flows by Type of Contract............................................. 106
Table 15: Share of FTCs in Total Worker Flows...................................................................... 107
Table 16: Worker Flows by Contract Type and Industry.......................................................... 109
Table 17: Sample Means of Explanatory Variables.................................................................. 125
Table 18: Number and Proportion of FTCs by Ongoing Job Tenure........................................ 127
Table 19: Means of Usual Weekly Hours of Work by Type of Contract.................................. 128
Table 20: Two-Sample Wilcoxon Rank-Sum Test for Differences in the
Distribution of Usual Weekly Hours of Work by Type of Contract................................ 129
Table 21: Earnings Distribution by Type of Contract............................................................... 130
Table 22: Means of Hourly Wage by Type of Contract............................................................ 131
Table 23: Two-sample Wilcoxon Rank-Sum Test for Differences in the
Distribution of Hourly Wage in € by Type of Contract................................................. 132
Table 24: Subjective Assessment of Job Insecurity Distribution by Type of Contract............. 136
Table 25: Subjective Assessment of Career Opportunities by Type of Contract...................... 136
Table 26: Propensity Score – Men (Model A)........................................................................... 145
Table 27: Propensity Score – Women (Model A)...................................................................... 147
Table 28: Propensity Score – Men (Model C)........................................................................... 150
Table 29: Propensity Score – Women (Model C) ..................................................................... 153
Table 30: Matching Quality – Men (Model A).......................................................................... 158
Table 31: Matching Quality – Women (Model A) .................................................................... 158
Table 32: Means of Important Conditioning Variables (X) Before and
After Kernel-Based Matching – Men (Model A) ............................................................. 159
Table 33: Means of Important Conditioning Variables (X) Before and
After Kernel-Based Matching – Women (Model A)........................................................ 160
Table 34: Wage Effects of FTCs – Men (Model A) .................................................................. 162
Table 35: Wage Effects of FTCs – Men (Model B) .................................................................. 162
Table 36: Wage Effects of FTCs – Men (Model C).................................................................. 162
Table 37: Effects of FTCs on Job Insecurity – Men (Model A) ................................................ 164
Table 38: Effects of FTCs on Job Insecurity – Men (Model B) ................................................ 164
Table 39: Effects of FTCs on Job Insecurity – Men (Model C )............................................... 164
Table 40: Effects of FTCs on Career Opportunities – Men (Model A)..................................... 165
Table 41: Effects of FTCs on Career Opportunities – Men (Model B)..................................... 166
Table 42: Effects of FTCs on Career Opportunities – Men (Model C)..................................... 166
Table 43: Heterogeneous Wage Effects of FTCs – Men (Model A) ......................................... 168
Table 44: Wage Effects of FTCs – Women (Model A )............................................................ 169
Table 45: Wage Effects of FTCs – Women (Model B)............................................................. 170
8 Table 46: Wage Effects of FTCs – Women (Model C)............................................................. 170
Table 47: Effects of FTCs on Job Insecurity – Women (Model A)........................................... 171
Table 48: Effects of FTCs on Career Opportunities – Women (Model A)................................ 171
Table 49: Heterogeneous Wage Effects of FTCs – Women (Model A).................................... 172
Table 50: Duration of Completed Unemployment Spells by Kind of Transition ..................... 190
Table 51: Explanatory Variables............................................................................................... 192
Table 52: Model Choice on the Basis of Information Criteria.................................................. 193
Table 53: Estimation Results of the (Multinomial Logistic) Competing Risks
Hazard Rate Model With All Exit States......................................................................... 197
Table 54: Definitions of the ‘Non-Treatment’ State and the Counterfactual............................ 205
Table 55: Definition of Treated and Untreated Individuals ...................................................... 207
Table 56: Definition of the Outcome Variables ........................................................................ 211
Table 57: Logistic Hazard Rate Model (Propensity Score Estimation) .................................... 213
Table 58: Loss of Treated Observations due to Common Support Requirement and
Lack of Similar Untreated Within the Caliper (NN-Matching, Outcome 2)................... 216
Table 59: Means of Important Pre-Treatment Variables (X) Before and
After NN-Matching (DEF.1)........................................................................................... 216
Table 60: Means of Important Pre-Treatment Variables (X) Before and
After NN-Matching (DEF.2)........................................................................................... 217
Table 61: Probability of Being Employed on a Permanent Contract – DEF.1.......................... 223
Table 62: Probability of Being Employed on a Permanent Contract – DEF.2.......................... 223
Table 63: Probability of Being Employed on a FTC – DEF.2 .................................................. 224
Table 64: Probabilities of Being Employed on a Permanent Contract versus FTC –
Women (DEF.2) .............................................................................................................. 227
Table 65: Probabilities of Being Employed on a Permanent Contract versus FTC –
Age 32 (DEF.2)........................................................................................................... 229
Table 66: Probabilities of Being Employed on a Permanent Contract versus FTC –
Workers With Formal Qualification (DEF.2).................................................................... 231
Table 67: Descriptive Statistics for the Estimation Sample – Dependent Variables................. 240
Table 68: Descriptive Statistics for the Estimation Sample – Independent Variables ............. 240
Table 69: Probability of Being Employed on FTC (Omitting Federal State Dummies)........... 241
Table 70: Effects of FTCs on Mean Weekly Hours of Work – Men (Model A )...................... 242
Table 71: Effects of FTCs on Working Part-time – Men (Model A)......................................... 242
Table 72: Effects of FTCs on Mean Weekly Hours of Work – Women (Model A).................. 242
Table 73: Effects of FTCs on Working Part-time – Women (Model A) ................................... 242
Table 74: Effects of FTCs on Job Insecurity – Men (Model A) ................................................ 243
Table 75: Effects of FTCs on Career Opportunities – Men (Model A)..................................... 244
Table 76: Effects of FTCs on Job Insecurity – Women (Model A)........................................... 245
Table 77: Effects of FTCs on Career Opportunities – Women (Model A)................................ 246
Table 78: Duration of Continuous Employment Spells After the Transition from
Unemployment to FTCs and Permanent Contracts in Months ........................................ 247
Table 79: Means of Explanatory Variables by Kind of Transition ........................................... 247
9
‡Table 80: Estimation Results of the Multinomial Logistic Duration Model with Two
Destination States ............................................................................................................ 249
Table 81: Estimation Results of the Multinomial Logistic Duration Model with Four
Destination States ............................................................................................................ 251

List of Symbols
C {0,1} treatment dummy
CA adjustment costs
t
CR churning rate
E employment level in the current period t t
*
E desired level of employment
t
E (Y X ) expectation of Y conditional on X
R
F w t cumulated wage offer distribution ( ( ))i i
G ? kernel function ( )
HR hiring rate
IN linear index i
JCR job creation rate
JDR job destruction rate
N employment stock
N , N , N number of treated, number of untreated, number of controls
1 0 C
Pr (?) probability
e neighbourhood as a function of the propensity score ( )i
RR rotation rate
SR separation rate
TFR transformation (of fixed-term into permanent contracts) rate
U ,U , U unobserved error terms in case of treatment (1) or non-treatment (0) in the out-
it 1it 0it
come equations
V unobserved error term in the participation equation i
WTR worker turnover rate
X vector of explanatory (conditioning) variables
Y output t
Y observed outcome
Y outcome in the treatment state 1
Y outcome in the non-treatment state 0
Z vector of explanatory variables
*Z instrumental variable
a time-invariant individual-specific effect i
e X , e propensity score (participation probability conditional on X) ( )
g ( X ) , g ( X ) functions of X in the group of treated (1) and controls (0)
1 0
h bandwidth parameter
h t hazard rate of individual i after unemployment duration t ( )i
10
˛