This paper examines the synchronization of business cycles across the G7 countries during US recessions since the 1870's. Using a dynamic measure of business cycle synchronization, results depend on the globalisation period under consideration. On average, US recessions have significantly positive effects on business cycle co-movements only in the period following the breakdown of the Bretton Woods system of fixed exchange rates, while strongly decoupling effects among the G7 economies are documented during recessions that occurred under the classical Gold Standard. During the 2007-2009 recession, business cycles co-movements increased to unprecedented levels. (author's abstract) / Series: Department of Economics Working Paper Series

The study is intended to identify relevant channels and possibly biasing factors with respect to fiscal multipliers, and thus to contribute to improving the precision of multiplier forecasts. This is done by, first, defining the concept of the multiplier used in the present study, presenting the main theoretical channels of influence as discussed in the literature and the problems of empirical identification. Second, by conducting a meta-regression analysis on the reported multipliers from a unique data set of 1069 multiplier observations and the respective study characteristics in order to derive quantitative stylzed facts. Third, by developing a simple multiplier model that explicitly takes into account the time elapse of the multiplier process as an explanatory factor that has been largely overlooked by the relevant theoretical literature. Fourth, by identifying, for US macroeconomic time series data, the extent to which fiscal multiplier estimates could be biased in the presence of financial cycles that have not been taken into account by the relevant empirical literature.:List of Figures IV List of Tables VI List of Acronyms VII List of Symbols IX 1 General Introduction, Aim and Scope 2 Principles of the Measurement, Theory and Estimation of Fiscal Multipliers 2.1 Introduction 7 2.2 Definition and Measurement of the Fiscal Multiplier 7 2.3 Determinants of the Fiscal Multiplier 14 2.4 Principles of Estimating Fiscal Multipliers 29 2.5 Conclusions 38 3 A Meta-Regression Analysis of Fiscal Multipliers 43 3.1 Introduction 43 3.2 Literature Review 45 3.3 Data Set and Descriptive Statistics 49 3.4 Meta Regression—Method 54 3.5 Meta Regression—Moderator Variables 56 3.6 Meta Regression—Results 60 3.7 Conclusions 74 4 The Multiplier Principle, Credit-Money and Time 82 4.1 Introduction 82 4.2 Literature Review 85 4.3 Developing an Augmented Multiplier Model 89 4.4 Dynamic Stability of the Multiplier Process 106 4.5 Identifying the Lag-length 109 4.6 Conclusions 111 5 Financial Cycles and Fiscal Multiplier Estimations 114 5.1 Introduction 114 5.2 Literature Review 116 5.3 Asset and Credit Markets and Fiscal Multiplier Estimations 118 5.4 A Formal Framework 120 5.5 Empirical Strategy 124 5.6 Data 125 5.7 Structure and Identification 126 5.8 Effects of Fiscal Policy Changes—Baseline vs. Augmented Models 132 5.9 Robustness 140 5.10 Conclusions 142 6 General Conclusions and Research Prospects 148 Bibliography 153

This paper considers the linkages between output growth and output volatility for the sample of G7 countries over the period 1958M2-2011M7, thereby paying particular attention to spillovers within and between countries. Using the VAR-based spillover index approach by Diebold and Yilmaz (2012), we identify several empirical regularities: i) output growth and volatility are highly intertwined, with spillovers taking place into all four directions; ii) the importance of spillovers has increased after the mid 1980s and reached unprecedented levels during the recent financial and economic crisis; iii) the US has been the largest transmitter of output and volatility shocks to other countries. Generalized impulse response analyses point to moderate growth-growth spillovers and sizable volatility-volatility spillovers across countries, suggesting that volatility shocks quintuplicate in the long run. The cross-variable effects turn out negative: volatilty shocks lead to lower economic growth, growth shocks tend to reduce output volatility. Our findings underline the increased vulnerability of the G7 countries to destabilizing shocks and their detrimental effects on economic growth, which are sizeably amplified through international spillover effects and the associated repercussions. / Series: Department of Economics Working Paper Series

Diese Dissertation besteht aus drei Essays, welche den monetären Transmissionsmechanismus via das Kreditangebot von Banken in einem Umfeld mit Überschussreserven analysieren. Im ersten Aufsatz wird die Effekten der 2008 Handlungsrahmens Änderungen der Fed auf den Transmissionsmechanismus untersucht. Ich schätze die Reaktionen in der Periode vor 2008 und zeige, dass den Bankkreditkanal aktiv ist. In der Periode nach 2008 steigen Bankkredite nach eine Geldpolitikkontraktion. Ich habe ein Regimewechsel-TANK Modell entwickelt, um den Transmissionsmechanismus über beide Systeme zu vergleichen. Das Modell zeigt, dass nach einem kontraktiven Schock unter dem alten System die Produktion sinkt, und, dass unter einem neuartigen System das Kreditangebot stimuliert. Dies ist aufgrund einer Friktion der Fall, die durch die Liquiditätsmanagementkosten der Banken verursacht wird. Im zweiten Aufsatz analysiere ich ob der Bankkreditkanal in den USA nach der Finanzkrise vorhanden ist. Dieser Kanal stützt sich auf die Annahme verbindlicher Reserveanforderungen. Ich finde Belege für den Kreditvergabekanal vor der Finanzkrise. Seit der Krise ist der Bankkreditkanal nicht mehr vorhanden. Stattdessen ist eine kontraktive Geldpolitik jetzt mit lockereren Liquiditätsbeschränkungen verbunden, und somit mit einem Anstieg der Bankkredite. Im dritten Aufsatz, D. Zander und ich identifizieren heterogene Bankenreaktionen auf geldpolitische Schocks in den USA. Unter Verwendung von einem informationsrobusten Instrument, zeigen wir, dass der Grad der Bargeld-Liquidität systematisch beeinflusst, wie Banken infolge eines geldpolitischen Schocks ihr Kreditvergabeverhalten ändern. Wir finden, dass nach einem kontraktiven Schock hochliquide Banken mit einer Ausweitung der Kreditvergabe reagieren, während weniger liquide Banken eine gedämpfte Antwort zeigen. Wir zeigen, dass das Vernachlässigen von Informationseffekten zu qualitativ unterschiedlichen Ergebnissen für liquide Banken führt. / This dissertation consists of three essays that analyze the monetary transmission mechanism via banks’ credit supply to the real economy under an environment of excess reserves. The first essay, examines the effects of the Fed’s 2008 operational system switch on the transmission mechanism of monetary policy. In the pre-switch sample the bank-lending channel is shown to be active, while in the latter sample, bank loans increase after a monetary contraction. Additionally, a regime-switching TANK model is used to compare the transmission mechanism across both systems. Under the old-style system real activity declines after a monetary contraction, while under a new-style system, monetary tightening stimulates credit supply, due to the presence of a friction introduced by banks’ liquidity management costs. The second essay analyzes whether the bank-lending channel is still present in post financial crisis U.S. data. This channel relies on the key assumption of binding reserve requirements, which is at odds with post-crisis data. Using a two-step regression approach, I find evidence supporting the lending channel in the subsample prior to the crisis. Moreover, since the crisis the lending channel is no longer active. Instead, monetary tightening is associated with looser liquidity constraints, and thus, with bank lending growth. The third essay is joint work with D. Zander. We identify heterogeneous bank reactions to monetary policy shocks in the U.S. using macro-econometric techniques and micro-level data. Using an informationally-robust instrument we show that the degree of cash-liquidity systematically influences banks’ lending behavior. Concretely, after a contractionary shock, liquid banks (those with excess reserves above 1% of assets) react by expanding lending, whereas less liquid banks have a muted response. We show that neglecting to control for the information effects of monetary policy, yield qualitatively different results that are at odds with economic theory.