Essays on econometrics of panel data and treatment models
Dissertation présentée en vue de l’obtention du titre de Docteur en Sciences économiques et de gestion
Le 13 Septembre 2013
par
Gianluca Papa
Sous la direction des Professeurs MATHIAS DEWATRIPONT and MICHELE CINCERA
Membres du jury :
Marco BECHT Michele CINCERA Dirk CZARNITZKI Mathias DEWATRIPONT Marjorie GASSNER
Bruno VAN POTTELSBERGHE
Université Libre de Bruxelles Université Libre de Bruxelles KU Leuven
Université Libre de Bruxelles Université Libre de Bruxelles Université Libre de Bruxelles
Année Académique 2012-2013
Introduction 5 Outside financing of investment and R&D in the 90's. A cross country
comparison between USA, UK and Japan 7
1.1 Introduction 7
1.2 An error correction model 11
1.3 Econometric specification 13
1.4 Data description 14
1.5 Estimation results and conclusions 17
1.6 References Chapter 1 24
1.7 Appendix 1.A: Data Definition 27
1.8 Appendix 1.B: GMM Estimates With Full Set of Instruments 28
1.9 Appendix 1.C: tests on cash flow coefficients 31
Financial leverage and managerial compensation: Evidence from the UK 33
2.1 Introduction 33
2.2 Empirical analysis of the effects of financial leverage on pay-for-performance
sensitivity 36
2.2.1 Data description 36
2.2.2 Estimation strategy and empirical results 38
2.2.3 The collapse of the stock market tech bubble as a source of exogenous
variation in firm's debt 41
2.2.4 Additional results on TSR and variance of TSR 46 2.3 Possible explanations for a non-monotonic relationship between financial leverage
and pay-for-performance sensitivity 46
2.3.1 Financial leverage and managerial compensation: a simple theoretical
illustration 47
2.4 Concluding remarks 51
2.5 References Chapter 2 52
Public funds additionality in R&D expenditure in presence of essential
heterogeneity: an empirical investigation using the Italian CIS3 data 57
3.1 Introduction 57
3.2 Literature review 59
3.3 Econometric methodology 60
3.3.1 Measuring additionality: ATE and ATT 64
3.4 Empirical model 65
3.5 Data 69
3.6 Estimation results 71
3.6.1 First-step results: determinants of expected subsidies 71 3.6.2 First-step results: additionality of public subsidies on R&D probability 72 3.6.3 Second-step results: additionality of public subsidies on R&D intensity 75
3.7 Conclusions 78
3.8 References Chapter 3 79
3.9 Appendix 3.A: Some computations of the econometric model 83
3.10 Appendix 3.B: Accounting Variables 86
3.11 Appendix 3.C: Categorical variables and Principal Component Analysis 87 3.12 Appendix 3.D: Main results of the model showing coefficients and p-values 91
The public health expenditure in Italy - the regional variability in per capita
spending: to which extent it is justified? 95
4.1 Introduction 95
4.2 Data description 98
4.3 Estimation strategy and first analysis of the results 102
4.4 Efficiency of the regional health systems - An analysis through the Efficiency
Frontier 110
4.4.1 A multi-dimensional indicator of data quality 110 4.4.2 A DEA approach to measuring the Efficiency Frontier 115 4.4.3 The DEA applied to the production of healthcare: a four-dimensional
Efficiency Frontier 119
4.4.4 Robustness check of the EF 123
4.4.4.1 Partial Frontier Approaches 123
4.4.4.2 Alternative modelling of expenditure determinants 125 4.5 The determinants of inefficiency – some considerations 125
4.6 Conclusions 127
4.7 References Chapter 4 128
4.8 Appendix 4.A - List of quality indicators of health care used in the study 131