162 Pages
English

Institutions and Public Sector Performance: Empirical Analyses of Revenue Forecasting and Spatial Administrative Structures [Elektronische Ressource] / Björn Kauder. Betreuer: Thiess Büttner

-

Gain access to the library to view online
Learn more

Description

Institutions and Public Sector Performance:Empirical Analyses of Revenue Forecastingand Spatial Administrative StructuresInaugural-Dissertationzur Erlangung des GradesDoctor oeconomiae publicae (Dr. oec. publ.)an der Ludwig-Maximilians-Universit at Munc hen2011vorgelegt vonBj orn KauderReferent: Prof. Dr. Thiess ButtnerKorreferent: Prof. Dr. Gebhard FlaigMundlic he Prufung: 4. November 2011Promotionsabschlussberatung: 16. November 2011AcknowledgementsFirst of all I would like to thank Thiess Buttner and Gebhard Flaig for their willingness tosupervise my thesis. I am especially indebted to Thiess Buttner for his ongoing support,encouragement, and inspiration. This fruitful collaboration is re ected particularly in hisco-authorship of Chapters 2 and 3. Moreover, I gratefully acknowledge the Ifo institutefor providing me with an optimal research environment and nancial support. I especiallythank my colleagues at the Department of Public Finance and my fellow doctoral studentsfor contributing to a pleasant working atmosphere. Special thanks go to Nadine Fabritzfor her willingness to help and her kind encouragement. I am also grateful to ChristianBreuer for extensive discussions regarding the political background of all the chapters andto Alexander Ebertz for his support particularly during the nal stage of this dissertation.Editorial support by Daniel Rees and Deborah Willow is greatly appreciated.

Subjects

Informations

Published by
Published 01 January 2011
Reads 6
Language English
Document size 1 MB

Institutions and Public Sector Performance:
Empirical Analyses of Revenue Forecasting
and Spatial Administrative Structures
Inaugural-Dissertation
zur Erlangung des Grades
Doctor oeconomiae publicae (Dr. oec. publ.)
an der Ludwig-Maximilians-Universit at Munc hen
2011
vorgelegt von
Bj orn Kauder
Referent: Prof. Dr. Thiess Buttner
Korreferent: Prof. Dr. Gebhard Flaig
Mundlic he Prufung: 4. November 2011
Promotionsabschlussberatung: 16. November 2011Acknowledgements
First of all I would like to thank Thiess Buttner and Gebhard Flaig for their willingness to
supervise my thesis. I am especially indebted to Thiess Buttner for his ongoing support,
encouragement, and inspiration. This fruitful collaboration is re ected particularly in his
co-authorship of Chapters 2 and 3. Moreover, I gratefully acknowledge the Ifo institute
for providing me with an optimal research environment and nancial support. I especially
thank my colleagues at the Department of Public Finance and my fellow doctoral students
for contributing to a pleasant working atmosphere. Special thanks go to Nadine Fabritz
for her willingness to help and her kind encouragement. I am also grateful to Christian
Breuer for extensive discussions regarding the political background of all the chapters and
to Alexander Ebertz for his support particularly during the nal stage of this dissertation.
Editorial support by Daniel Rees and Deborah Willow is greatly appreciated. I bene ted
a lot from comments and suggestions of participants at conferences and workshops in Kiel,
Munich, Nuremberg, and Uppsala, and at seminars at CES and Ifo. I would especially like
to thank S ren Bo Nielsen for insightful discussions and comments on Chapter 4. Last but
certainly not least I would like to thank my parents for their encouragement and support
throughout the entire project.
iContents
1 Introduction 1
2 Revenue Forecasting Practices: Di erences across Countries and Conse-
quences for F Performance 9
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Forecasting Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3 Conditions Faced by Forecasters . . . . . . . . . . . . . . . . . . . . . . . . 19
2.4 Institutions and Independence . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.5 Determinants of Forecasting Performance . . . . . . . . . . . . . . . . . . . 27
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3 Revenue Forecasting in Germany: On Unbiasedness, E ciency, and Pol-
itics 50
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.2 Investigation Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.3 Data and Descriptive Statistics . . . . . . . . . . . . . . . . . . . . . . . . 56
ii3.4 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4 Spatial Administrative Structure and Inner-Metropolitan Tax Competi-
tion 80
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.2 Theoretical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.3 German Municipalities and Institutions . . . . . . . . . . . . . . . . . . . . 88
4.4 Agglomerations and Empirical Strategy . . . . . . . . . . . . . . . . . . . . 91
4.5 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
4.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5 Consolidation of Municipalities and Impact on Population Growth { A
Propensity Score Matching Approach 117
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.2 Institutional Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
5.3 Identi cation Strategy and Data . . . . . . . . . . . . . . . . . . . . . . . . 123
5.4 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
5.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6 Concluding Remarks 151
iiiList of Tables
1.1 Cities with the Highest Intensity of Incorporation . . . . . . . . . . . . . . 4
2.1 Descriptive Statistics of Forecast Errors . . . . . . . . . . . . . . . . . . . . 18
2.2 Forecasting Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 Institutional Characteristics and Independence . . . . . . . . . . . . . . . . 25
2.4 Determinants of Forecast Error . . . . . . . . . . . . . . . . . . . . . . . . 28
2.5ts of Forecasting Precision and Accuracy: Total Revenues . . . 30
2.6 Determinants of Forecasting Precision and Accuracy: Disaggregated Revenues 33
2.7 Timing of Forecasts and Time Span . . . . . . . . . . . . . . . . . . . . . . 43
3.1 Descriptive Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.2 Results on the Forecast Bias: May current / May next / November next Year 63
3.3 Results on Forecast E ciency I: May current / May next / November next
Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.4 Results on Forecast E ciency II: November next Year . . . . . . . . . . . . 66
3.5 Results on Forecast E ciency III: May current / May next / November next
Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
iv3.6 Results on Politics: May current / May next / November next Year . . . . 70
3.7 Results on the Forecast Bias: November Forecast for the next Year (GDP
Forecast Error based on Research Institutes) . . . . . . . . . . . . . . . . . 73
3.8 Results on Forecast E ciency I: November Forecast for the next Year (GDP
Forecast Error based on Research Institutes) . . . . . . . . . . . . . . . . . 74
3.9 Results on Forecast E ciency II: November Forecast for the next Year (GDP
Forecast Error based on Research Institutes) . . . . . . . . . . . . . . . . . 75
3.10 Results on Forecast E ciency III: November Forecast for the next Year
(GDP Forecast Error based on Research Institutes) . . . . . . . . . . . . . 76
3.11 Results on Politics: November Forecast for the next Year (GDP Forecast
Error based on Research Institutes) . . . . . . . . . . . . . . . . . . . . . . 77
4.1 The Largest German Cities and their Regions (of 50 km) . . . . . . . . . . 95
4.2 Local Business Taxation in Metropolitan Areas of 15 km . . . . . . . . . . 100
4.3 Local Business Taxation in Metropolitan Areas of 25 km . . . . . . . . . . 101
4.4 Local Business Taxation in Metropolitan Areas of 50 km . . . . . . . . . . 103
4.5 Local Business Taxation in Metropolitan Areas with Parameter of 10 % . . 104
4.6 Local Business Taxation in Metropolitan Areas with Parameter of 1 % . . 105
4.7 Descriptive Statistics (Regions of 15 km) . . . . . . . . . . . . . . . . . . . 109
4.8 Descriptive Statistics (Regions of 25 km) . . . . . . . . . . . . . . . . . . . 110
4.9 Descriptive Statistics (Regions of 50 km) . . . . . . . . . . . . . . . . . . . 110
4.10 Descriptive Statistics (Regions 10 %) . . . . . . . . . . . . . . . . . . . . . 111
4.11 Descriptive Statistics (Regions 1 %) . . . . . . . . . . . . . . . . . . . . . . 111
v4.12 Correlation of the Number of Surrounding Municipalities . . . . . . . . . . 112
4.13 Correlation of the Population per Surrounding Municipality . . . . . . . . 112
4.14 Correlation of the Population Share of the Core City . . . . . . . . . . . . 112
4.15 Correlation of the Population of the Region . . . . . . . . . . . . . . . . . 113
4.16 Local Business Taxation in Metropolitan Areas of 50 km (extended) . . . . 114
5.1 Reforms of the Administrative Structure . . . . . . . . . . . . . . . . . . . 122
5.2 Descriptive Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.3 Treatment Probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.4 Average Treatment E ect on the Treated . . . . . . . . . . . . . . . . . . . 134
5.5 Average Treatment E ect on the Treated (Di erent Time Periods) . . . . . 135
5.6 Average Treatment E ect on the Treated (Di erent Sizes) . . . . . . . . . 138
5.7 Balancing Property (Radius Matching) . . . . . . . . . . . . . . . . . . . . 139
5.8 Descriptive Statistics (Incorporation <= 1972) . . . . . . . . . . . . . . . . 142
5.9 Descriptive Statistics (Incorporation > 1972) . . . . . . . . . . . . . . . . . 143
5.10 Descriptive Statistics (Population <= 1500) . . . . . . . . . . . . . . . . . 144
5.11 Descriptive Statistics (Population > 1500) . . . . . . . . . . . . . . . . . . 145
5.12 Balancing Property (One-to-one Matching) . . . . . . . . . . . . . . . . . . 146
5.13 Balancing Property (10-Nearest-Neighbors Matching) . . . . . . . . . . . . 147
5.14 Balancing Property (Kernel Matching) . . . . . . . . . . . . . . . . . . . . 148
viList of Figures
1.1 Standard Deviation of Next-Fiscal-Year Forecast Error (in %) . . . . . . . 2
1.2 Relevance of Local Business Tax . . . . . . . . . . . . . . . . . . . . . . . . 3
2.1 Forecast Errors by Country/Institution . . . . . . . . . . . . . . . . . . . . 15
2.2 Forecast Errors by Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.1 Relative Forecast Error (in %): Current-Year Forecast May . . . . . . . . . 59
3.2 Relative Forecast Error (in %): Next-Year Forecast May . . . . . . . . . . 59
3.3 Relative Forecast Error (in %): Next-Year Forecast November . . . . . . . 60
4.1 Number of Municipalities per 100 Square Kilometers . . . . . . . . . . . . 90
4.2 The De nition of Agglomerations (First Approach) . . . . . . . . . . . . . 93
4.3 The De nition ofs (Second Approach) . . . . . . . . . . . . 94
4.4 Local Business Taxation in Germany (\Rate of Assessment") . . . . . . . . 97
viiChapter 1
Introduction
\Don’t make the tax gures seem better than they are", the president of the German Court
of Auditors remarked in apprehension of budget imbalances { he was concerned about too
optimistic revenue forecasts. The performance of agents in the public sector, such as
revenue forecasters, depends on the design of institutions. Also local politicians react to
incentives originated in institutions: \Certainly, this is local-business-tax cannibalism"
claimed the head of the economics department of the city of Frankfurt. He nds his city
exposed to increased tax competition, induced by tax cuts of surrounding municipalities.
The question arises which institutions cause these statements. Since the institutional
framework determines the decision making of politicians and bureaucrats, the performance
of the public sector hinges crucially on the respective environment.
This book looks at two aspects where the e ects of institutions on the performance of
political agents are particularly relevant { one on the country and the other on the local
level. The rst concerns the environment in which revenue forecasters work. Since revenue
forecasts are the basis of every budgetary process, they feature prominently in decisionsUS: CBO 10.175
Japan 10.003
US: OMB 9.031
Ireland 7.608
Netherlands 6.203
Germany 5.419
Canada 5.044
Italy 4.626
New Zealand 3.939
CHAPTER 1. INTRODUCTION 2
Belgium 2.611
France 2.542
Austria 2.279
Figure 1.1: Standard Deviation of Next-Fiscal-Year Forecast Error (in %)
United Kingdom 1.977
12
10
8
6
4
2
0
Figure covers the period from 1995 to 2009 (fewer observations for some countries). CBO { Congres-
sional Budget O ce. OMB { O ce of Management and Budget.
regarding economic policy. But as Figure 1.1 shows, their quality di ers notably across
countries and institutions. This gives rise to the question why this is, and calls for an in-
vestigation into the determinants of forecasting precision. Among these determinants may
be the assignment of the task: While in some countries revenue forecasting is performed
by independent institutions, others produce the gures in their ministries. But also fur-
ther conditions di er that can a ect the performance of forecasters and hence the policies
implemented, such as the structure of the tax system or the timing of forecasts. Exploit-
ing variation in these di erences allows for identifying institutions that lead to superior
forecasts. Also the information that is provided by the government has to be mentioned
in this context. It enables (superior members of) a government to \optimize" the gures
strategically in order to in uence the forecasts in the preferred direction. The question of
whether such manipulation or other biases exist and whether forecasters use all the relevant
information available at the time of the forecast calls for further empirical assessment.
US: CBO
Japan
US: OMB
Ireland
Netherlands
Germany
Canada
Italy
New Zealand
Belgium
France
Austria
United Kingdom