Heinemann et al (eds ) monetary policy, financial crises, and the macroeconomy (2017)

Watzka Eds., Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing pp.. Watzka Eds., Monetary policy, financial crises, and the macroeconomy: Festschrif

Frank Heinemann · Ulrich Klüh

Sebastian Watzka Editors

Monetary

Policy, Financial Crises, and the MacroeconomyFestschrift for Gerhard Illing

and the Macroeconomy

Monetary Policy,

Financial Crises,

and the Macroeconomy

Festschrift for Gerhard Illing

123

IMK - Macroeconomic Policy Institute

Hans-Böckler-Foundation

Düsseldorf, Germany

ISBN 978-3-319-56260-5 ISBN 978-3-319-56261-2 (eBook)

DOI 10.1007/978-3-319-56261-2

Library of Congress Control Number: 2017951194

© Springer International Publishing AG 2017

This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG

The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

This volume contains invited contributions by (former) students, colleagues, andfriends of Gerhard Illing, whose 60th birthday served as an occasion for collect-ing these articles Nearly all contributions were presented in a special birthdaysymposium.

Gerhard Illing’s research focuses on the relation between monetary policy,financial crises, and the macroeconomy He has often argued that financial andmacroeconomic instabilities are a key issue for our societies, an important researchtopic, and a challenge for macroeconomic policy He encouraged students andcolleagues alike to take the issues of financial crisis prevention and resolutionseriously, even at a time when most macroeconomists believed that the greatmoderation had made crises in mature economies a thing of the past His pioneeringapproach combines strong theory to explain causal relationships with a clear view

on data and general macroeconomic developments His proficiency with gametheoretic and microeconomic methods has helped him (and others) to advancemacroeconomics in novel and very fruitful directions In particular, he contributed

to making mechanism design an important tool for macroeconomic policy analysis.The editors owe Gerhard many thanks for his inspiring views His open, curious,and analytical mind often pointed us to upcoming research topics, policy debates,and methodological innovations

Many chapters in this volume follow the approach of applying microeconomicand game theoretic methods to monetary policy and financial crises They also con-tain interesting empirical results, reflecting Gerhard’s view that evidence antecedesany application of models They discuss recently suggested measures for centralbanks’ responses to liquidity shortages and to the liquidity trap They developmethods for assessing the potential of contagion via the interbank network andfor capturing the interaction between micro- and macroprudential regulation Inaddition, they contain empirical analyses of macroeconomic effects of Germanunification and current developments in the German housing market

A wider audience might be especially interested in the chapters that point

to avenues for re-conceptualizing and renovating macroeconomics One potentialstarting point for such renovation is the application of new microeconomic methods

v

to macro problems This is reflected in an insurance-based approach to evaluateproposals for solving the sovereign debt problem in the Euro Area It is alsoclearly visible in a new explanation for rising income inequality that is based oncontract theory and advances in IT technology Re-conceptualization, however, willalso require a more fundamental, transdisciplinary critique of the current state ofmacroeconomics Such critique is provided in a detailed analysis of the dogmaticsuperstructure of the process of financialization, which many believe has been animportant driver of the developments in recent decades.

The symposium on which this volume is based took place at Maximilians-University (LMU) in Munich from March 4 to 5, 2016 The conferencewas characterized by an extremely lively exchange between academics andpractitioners, very much in the spirit of Gerhard’s approach to economics Wewould like to thank all participants for their participation in the conference and theircontributions to this volume

Ludwig-The atmosphere, depth, and policy relevance of the symposium greatly benefitedfrom two policy panels The panelists (Peter Bofinger, Charles Goodhart, Hans-Helmut Kotz, Bernhard Scholz, and Hans-Werner Sinn) have done a great job intranslating research results into policy advice and to enliven the discussions duringsessions and afterward We thank them for their presence and their inputs

One secret of a successful conference is a generous host providing the necessaryinfrastructure and a committed team doing the background work Many thanks go

to the Ludwig-Maximilians-University (LMU) for its support and hospitality Itallowed all participants, many of who had spent an important part of their career

at LMU, to feel very much at home and at ease Our special thanks go to Mrs.Agnes Bierprigl and to the other team members at the Seminar for Macroeconomics.Their dedication and effort were crucial to make this event happen and to ensure itssuccess

We also express our thanks to Mr Alen Bosankic, Ms Jasmina Ude, and Mr.Moritz Hütten for reading proofs and preparing chapter drafts The team at SpringerPublishing has not only been very patient but also very forthcoming with supportand assistance

Finally, it is our pleasant duty to acknowledge financial support from DeutschePfandbriefbank and Cesifo

Monetary Policy, Financial Crises, and the Macroeconomy:

Introduction 1

Frank Heinemann, Ulrich Klüh, and Sebastian Watzka

Part I Liquidity From a Macroeconomic Perspective

Balancing Lender of Last Resort Assistance with Avoidance

of Moral Hazard 19

Charles Goodhart

Optimal Lender of Last Resort Policy in Different

Financial Systems 27

Falko Fecht and Marcel Tyrell

Network Effects and Systemic Risk in the Banking Sector 59

Thomas Lux

Contagion Risk During the Euro Area Sovereign Debt Crisis:

Greece, Convertibility Risk, and the ECB as Lender of Last Resort 79

Sebastian Watzka

The Case for the Separation of Money and Credit 105

Romain Baeriswyl

Part II Putting Theory to Work: Macro-Financial Economics

from a Policy Perspective

(Monetary) Policy Options for the Euro Area: A Compendium

to the Crisis 125

Sascha Bützer

On Inflation Targeting and Foreign Exchange Interventions

in a Dual Currency Economy 163

Ivana Rajkovi´c and Branko Uroševi´c

vii

Macroprudential Analysis and Policy: Interactions

and Operationalisation 177

Katri Mikkonen

Are Through-the-Cycle Credit Risk Models a Beneficial

Macro-Prudential Policy Tool? 201

Manuel Mayer and Stephan Sauer

Assessing Recent House Price Developments in Germany:

An Overview 225

Florian Kajuth

Part III Re-Conceptualizing Macroeconomics:

An Interdisciplinary Perspective

German Unification: Macroeconomic Consequences

for the Country 239

Axel Lindner

Approaches to Solving the Eurozone Sovereign Debt Default Problem 265

Ray Rees and Nadjeschda Arnold

Appraising Sticky Prices, Sticky Information and Limited Higher

Order Beliefs in Light of Experimental Data 297

Camille Cornand

Rising Income Inequality: An Incentive Contract Explanation 307

Dominique Demougin

No More Cakes and Ale: Banks and Banking Regulation

in the Post-Bretton Woods Macro-regime 325

Moritz Hütten and Ulrich Klüh

Greetings from Bob Solow 351

and the Macroeconomy: Introduction

Frank Heinemann, Ulrich Klüh, and Sebastian Watzka

Since the early 1970s, financial instability has been on the rise For some timethis trend had been mainly associated with emerging markets, even though therewere occasional crises in some high-income countries as well In the industrializedworld, the increasing instability of economic systems had been masked by thefact that macroeconomic aggregates appeared to become more stable The subduedfluctuations of the Great Moderation seemed to validate the view that crises anddepressions were a thing of the past

This changed in 2007/2008, when a global financial crisis of yet unknownmagnitude and character hit the U.S., Europe, and, through spillovers, the wholeworld This crisis validated all those who had warned that depressions were stillone of the main problems with which economics had to cope It brought up manynew and controversial policy topics that still are not resolved satisfactorily Also, ithas put into question many of the dogmas that had characterized macroeconomicthinking since the late 1970s

Gerhard Illing is at the forefront of those who have constantly argued thatfinancial and macroeconomic instabilities are a key issue for our societies, animportant research topic and a challenge for macroeconomic policy Thus, he is one

of those whose views have been validated by the crisis This volume is a collection

of contributions to a symposium held to celebrate Gerhard’s sixtieth birthday

© Springer International Publishing AG 2017

F Heinemann et al (eds.), Monetary Policy, Financial Crises,

and the Macroeconomy, DOI 10.1007/978-3-319-56261-2_1

1

Gerhard’s approach to macroeconomic analysis is unique in the way it balancesdifferent perspectives He is one of the few German economists with an eyefor the demand side of the economy But he also looks at the supply side.

He is a skillful microeconomist and he has used his microeconomic expertisefrequently to illuminate macroeconomic puzzles In spite of this ability, Gerhard

is a macroeconomist by heart who does not force micro-foundations upon anymacroeconomic problems Finally, he is an economist with a strong preference for

academic rigor and policy relevance, and wants to achieve both at the same time.

Gerhard’s research interests are multifaceted He has published and edited booksand papers on diverse topics, such as game theory (Holler and Illing 2009), thedigital economy (Illing and Peitz 2006), and spectrum auctions (Illing and Klüh

2004) But his main interest in recent years has been (i) the nature and role ofliquidity for macroeconomic and financial policies; (ii) the design of policies,instruments, and strategies to cope with the macro-financial problems characterizingmodern capitalist societies; (iii) the integration of new methods and views intomacroeconomic thinking

This volume is organized along the above three lines of research Part I dealswith liquidity and contagion of liquidity crises Liquidity becomes a relevant issuethrough frictions, in particular those analyzed by information economics (Illing

1985) It has many facets, ranging from market and funding liquidity to monetaryforms of liquidity And it has been at the heart of the analysis of financial crises andthe optimal response to their occurrence (Illing2007)

Part II looks at policies, in particular those at the nexus between macroeconomicsand finance The crisis has brought about a revival of aggregate demand policies, atrend already foreseen in Illing (1992) and Beetsma and Illing (2005) It has putmonetary policy in a very difficult position, caught between macroeconomic andfinancial stability (Cao and Illing2015) and faced with the manifold challenges ofthe zero lower bound (Illing and Siemsen2016) The crisis has made necessary are-assessment of fiscal policy (Illing and Watzka2014) and public debt (Holtfrerich

et al.2015), and it has raised the question of how to complete the re-regulation ofthe financial sector, with a view to strengthen its macroprudential dimension (Illing

2012)

Part III presents approaches for a re-conceptualization and renovation of conomics The failure of large parts of the economics profession before and duringthe crisis has made such a re-conceptualization necessary Economists have trustedtoo much in efficient markets As a consequence, they did not warn sufficientlyabout the imbalances that were building up During the crisis, they were not able

macroe-or not willing to prevent the austerity backlash that has kept so many economies indepression mode

Looking for new approaches in macro-financial economics does not mean,however, that everything that has been done before should be disposed of Thoselike Gerhard who have studied financial instability before the crisis have come upwith important and often surprising insights (see, e.g Heinemann and Illing2002;Goodhart and Illing2001) The problem has not been a lack of good theory, nor ofgood empirics, but a missing focus on relevant questions

1 Liquidity and Contagion of Financial Crises

It is difficult to overestimate the role of liquidity as a key or possibly evenparamount concept in macroeconomic thinking Monetary macroeconomics as adiscipline could not be constituted without the notion of liquidity In the history ofeconomic thought, liquidity has been central in constituting different paradigms ofmacroeconomics It has informed early discussions of macroeconomic issues, such

as in Gresham’s law It has been central to physiocratic views of the economy, inwhich some see the beginning of economic thinking in circular flows The concept

of liquidity is closely related to Say’s law (Klüh2014), and it is one of the mainfeatures of Keynesian economics and all “modern macroeconomic” DSGE models.The view on the role of monetary aggregates divides different schools and is adefining element of many controversies regarding monetary policy and financialmarket regulation

Proponents of real business cycle theory and perhaps growth economics mightargue that liquidity and monetary effects are only temporary and the welfare lossesarising from fluctuations are small in comparison to the long-run gains of realeconomic growth Indeed, if one assumes complete markets and perfect rationality,liquidity is of no major concern This view, however, has been largely knockeddown by recent experience As soon as one starts to look at the pathologies ofcapitalist societies, focusing on liquidity becomes inevitable (Goodhart and Illing

2001) because the run effects of misdirected investment activities, run unemployment, and high youth unemployment rates that are associated withfinancial crises are estimated to protract growth for several years with no chance ofreturning to the old growth path

long-In spite of its overwhelming importance, many economists perceive liquidity

as a riddle within an enigma Trained to think in models in which real exchangedominates, the importance of the nominal dimension of economics that directlyfollows from the notion of liquidity is often difficult to accept More importantly,the frictionless or friction-poor world of many models provides only little space for

a concept that is largely a consequence of frictions These frictions are many andmost can be traced back to incomplete information

But what is liquidity? And when does it (or a shortage of it) constitute aproblem? Charles Goodhart (2017), in the first chapter of this volume, sets outhis analysis by asking these fundamental questions He contextualizes his analysis

of lender-of-last-resort (LOLR) policies by first looking at the nature of liquidityproblems Liquidity shortages have a dual nature On the one hand, a lack ofliquidity in most cases reflects some kind of solvency concern: if payments andrepayments are certain, both with respect to their incidence and with respect to thedetails of their occurrence, the ability to borrow ensures liquidity On the otherhand, illiquidity does not necessarily reflect actual solvency problems, becausefundamentally solvent banks can become illiquid due to the network effects infinancial markets Goodhart argues that there is no clear cut distinction betweensolvent but illiquid and insolvent banks

Thus, the provision of liquidity during banking crises must compromise twogoals: on one hand, systemic crises should be avoided because of the huge losses tosociety, on the other hand, any implicit guarantee for providing liquidity to banks

in distress raises concerns that banks may game the rules and exploit tax payers.Moral-hazard should be avoided

Against this background, determining optimal last resort policies involves ficult judgements Depending on which of the two views of liquidity shortages isemphasized, very different policy recommendations follow If liquidity problemsare mainly a reflection of solvency problems, policy should be more restrictive Ifliquidity problems are a reflection of the inherent fuzziness and non-linearity of theliquidity-solvency nexus, central banks should have maximum flexibility to preventunnecessary harm to the economy

dif-The standard advice in the literature has been influenced strongly by the firstview To prevent lending to insolvent and thus likely irresponsible players, thecentral bank should mostly lend to the open market and not to individual banksvia LOLR measures The fear of unwarranted support to failed institutions hasalso dominated changes in crisis-management arrangements after the crisis, such

as the Dodd-Frank act As a consequence, there is a risk that central banks will haveinsufficient flexibility when the next crisis comes

Goodhart argues that this underestimates the importance of the second view, and

in particular the dynamics of contagion Provision of liquidity to the market is nothelpful to stave off contagious banking crises, because the market allocates extraliquidity to those institutions who are not directly affected by the crisis While open-market operations may prevent a complete meltdown, they may leave us with apartial meltdown and severe macroeconomic consequences

Instead, Goodhart recommends that a central bank should treat the first bank torun out of liquidity most toughly up to letting the bank fail, but provide liquidity atmore favorable conditions to other banks in distress that may have been affected bycontagion This mechanism raises incentives for banks to avoid illiquidity but savesthem from the network effects and, thereby, avoids systemic crises Nevertheless,any LOLR policy creates moral-hazard incentives For Goodhart, the only way

to properly take this into account would be a much more ambitious approach tochange incentives The rules should be such that they come as close as possible to

an unlimited liability arrangement, for example through multiple liability schemesand a much stronger emphasis on bail-in-able debt

The question, whether central banks should provide liquidity to the market or toindividual institutions in distress, is also analyzed by Falko Fecht and Marcel Tyrell(2017) in the second chapter of this volume Building up on a model by Diamondand Rajan (2001), they ask whether the answer may also depend on the nature ofthe financial system A key ingredient are the losses that arise if a bank needs toliquidate or sell projects that it cannot continue to finance Fecht and Tyrell assumethat in bank-based financial systems, such as continental Europe, intermediarieshave more information about the profitability of projects that they are financing than

in a market-based system such as the United States Bank-based financing allowsbanks to extract a larger share of the liquidation value of a project, while the market

value to outside investors is higher in market-based systems, where information isless asymmetric.

From these assumptions, Fecht and Tyrell derive a number of results that inform

us about differences in LOLR policies between the two systems They show thatthe provision of liquidity by open-market operations leads to inefficiencies that aremore severe for a bank-based than for market-based system Providing liquidity toindividual institutions is more preferable in a bank-dominated system

The employed model does not account for moral hazard effects that may provide

a general argument for open-market operations LOLR assistance to individualinstitutions may also be more costly for the central bank Assuming that thesecosts are comparable in both systems, Fecht and Tyrell conclude that in bank-basedfinancial systems with their rather illiquid assets, LOLR assistance to individualinstitutions may be a more favorable instrument than providing liquidity to themarket via open-market operations, while the opposite may be true in a market-oriented financial system

The model by Fecht and Tyrell considers contagion via the relative prices ofassets in terms of liquidity, but it does not account for contagion arising from directlinks between banks These contagion effects are the reason why Goodhart rejectsthe clear distinction between insolvency and illiquidity The dynamics of contagionthat are at the heart of Goodhart’s analysis are largely a consequence of the fact thatfinancial systems are complex networks Should the central bank or supervisor have

a very good grasp of the systemic consequences of a specific support measure orpunishment, official responses to liquidity problems could be much more targeted.The degree of moral hazard would be reduced and the flexibility of the centralbank increased Moreover, one could start devising incentives to reduce systemicallyrelevant network effects, for example through special rules for money-center banks

In his contribution, Thomas Lux (2017) argues that the pre-2008 mainstreamapproach to macroeconomic research had “deliberately blinded out” these issues,mainly because of the purported efficiency of financial markets The post-crisisresearch on interbank networks and contagion dynamics is becoming more receptive

to the alternative view, which emphasizes market inefficiencies, behavioral aspects,non-linearity, and non-standard probability distributions

Lux shows that this literature has yielded a set of important stylized facts rangingfrom topological features such as core-periphery structures to stability characteris-tics (such as the surprising persistence of certain linkages) He also recognizes firstsuccesses in explaining the self-organization of the system However, attempts totheoretically measure and then internalize network externalities are in the fledglingstages, at least academically Thus, the potential for informing policies to changethe system’s structure in an attempt to contain contagion remains limited

Lux presents simulations of a stochastic model of link formation and spillovers

An individual default of one bank affects in most cases only few other institutions.But for a small number of banks, their default triggers a system-wide collapse Moststress tests by monetary authorities have only considered the financial stability ofindividual institutions and neglected the propagation of liquidity shortages throughthe banking system One reason is data limitations Moreover, as contagion happens

through a multitude of channels and because balance sheets change quickly, policiesgrounded in theory may quickly be outdated What, then, is the role of the newgeneration of models described in the chapter? According to Lux, network modelsmay help to get a better grasp of the capital cushions needed to prevent shocks andshock transmission in an otherwise fragile system By focusing on capital buffers,Lux picks up an argument that has been crucial for the crisis response so far: moretargeted measures focusing more explicitly on the structural problems would bedesirable However, a lack of knowledge about the impact of these policies precludestheir implementation The second-best method might be to focus on capital, anargument implicit in Illing (2012, p 17).

Sebastian Watzka (2017), in his chapter, considers the liquidity risk again from adifferent angle He discusses the euro area debt crisis—and in particular the Greektragedy—under the assumption that some of the risk premia in Greek governmentbond yields were due to what the ECB referred to as “convertibility” risk, i.e.the break-up risk of the euro area This idea has forcefully been demonstrated by

De Grauwe and Ji (2013) arguing that an individual euro area member country isnaturally lacking a LOLR and this by itself would generate multiple equilibria withunduly high liquidity risk premia for countries of which investors believe that publicdebt is too high To test for such effects, Watzka empirically assesses how importantnon-fundamental contagion was during the early phase of the Greek debt crisis Heconcludes that Mario Draghi in his famous 2012 London speech reassured marketsthat the ECB was in fact acting as LOLR for euro area countries, if certain criteriawere met

A crucial and usually innocuous assumption in most papers on banking crises

is that money and credit are intrinsically conjoined Does this need to be the case?The crisis shows that the pursuit of price stability (which had been achieved almostuniversally before 2007) does not imply financial stability In contrast, there areimportant ways in which policies to achieve one can be detrimental to the other

An important reason for this perceived antagonism is the nature of moneycreation through credit markets It is therefore not surprising that a radical departurefrom this approach has been envisioned by some In his contribution to this volume,Romain Baeriswyl (2017) argues that the close connection between money andcredit is a relic of the Gold Standard With unredeemable fiat money, there are fewreasons to stick with it But what would be the inter-sectoral and inter-temporalimplications of such a departure? Baeriswyl argues that the provision of liquidityvia the credit market has the largest effects on private credit volume and primarilystimulates demand for goods that are bought on credit such as real estate Hence,expansionary monetary policy fuels asset prices and may cause price bubbles, alongwith its stimulus effects for aggregate demand

For targeting consumer price inflation, lump-sum transfers of money to sumers are likely to be more effective Lump-sum transfers from the central bank tothe citizens sound radical at first, but has some important advantages In Baeriswyl’sview, these advantages strongly outweigh the disadvantages In particular, thepursuit of price stability would no longer require destabilizing the financial systemthrough credit creation or contraction Finally, there would be less interference with

con-inter-temporal decisions, because interest-rate policies prevent the free adjustment

of credit markets to supply and demand of real resources as savings and investment

By contrast, lump-sum transfers of money stimulate demand without directlyaffecting interest rates

Baeriswyl’s analysis does not stop here Separating money from credit wouldhave far-reaching implications that go beyond monetary policy For example, itseems to require a departure from fractional-reserve banking Lump-sum transfersalso require a re-assessment of the way central banks absorb liquidity Proponents

of credit-based money creation often raise three interrelated arguments against itsabolition First, they argue that lump-sum transfers constitute fiscal policy Because

of their distributional consequences, transfers need to be decided upon by electedofficials, not technocrats Second, they believe that the current system is better thanoften assumed in bringing investment and savings in balance Has it not allowedeconomic growth for large spells of the last two centuries? Third, they questionthe need to focus so much attention on central bank policy If fiscal policy is pro-active, a credit-based monetary system can work smoothly Fiscal policy takes centerstage in absorbing excess liquidity and savings and in making sure that investmentexpenditures are sufficient It can also take the necessary steps to prevent or escape

a liquidity trap

Unfortunately, European fiscal policy currently appears rather dysfunctional: itneither uses the opportunity of a huge excess supply of savings and demand for safeassets to boost public investment, nor does it exploit the large multiplier effects offiscal policy in a liquidity trap for stimulating demand This has raised a discussionfor helicopter money as an additional instrument for central banks Baeriswyl justgoes one step beyond and suggests to replace the credit channel completely by ahelicopter

Macroeconomics is a policy-orientated science A main challenge is to take theoryand empirical scrutiny as far as possible while always having policy in mind.Bringing cognitive interest and policy relevance together has always been a hallmark

of Gerhard Illing’s thinking This has been most visible during the symposium thathas given rise to this volume A frequent comment of participating central bankerswas that if academic conferences would always be so interesting, they would haverather remained in academia While all three parts of this book reflect this practicalside of macroeconomics, this section puts special emphasis on it

Financial markets and institutions are not just playing a dominant role intransmitting monetary policy to the real sector In recent years, they have oftenabsorbed policy impulses Macroeconomic policy feeds into the peculiar logic ofexpansion and contraction that increasingly characterized the financial sector From

a certain point on, however, periods of financial contraction become a source offiscal and growth risk Finance, thus, simultaneously charges and discharges policy

As fiscal policy has taken a backseat since the beginning of the 1970s, monetarypolicy has found itself in the center of this double role It faces a difficult conflict.

On the one hand, it tries to fulfill its role as a levee against the negative realconsequences of financial contraction On the other, it tries to enclose the dangers

of excessive financial expansion As the instruments to achieve the first may inhibit

or even foil the instruments available to achieve the second, a conflict emerges Anintriguing analysis of this conflict and its relation to liquidity issues is provided inCao and Illing (2010,2011)

The challenges for monetary policy are all the more acute when fiscal policybecomes increasingly passive This is most obvious in the case of the Euro crisis,which is surveyed and analyzed in the first chapter of the second part Here, SaschaBützer (2017) first illustrates the dramatic failure of fiscal policy The institutions

of the European Monetary Union lack mechanisms to pool risks across its memberstates and put the burden of adjustment on these national states while stripping them

of some of the most effective instruments to achieve these adjustments, like nationalinterest and exchange rates Integrated financial markets would be an alternative tofiscal risk pooling, but financial integration stopped short of the standards achieved

in other currency areas Apparently, several member states have been overcharged

by these demands An almost religious belief in austerity and structural reform hasprolonged the recession It has led to an increase in indebtedness and thus defeateditself Finally, it has pushed monetary policy in a situation that is perceived as anoverburdening of its possibilities and mandate

In Bützer’s view, monetary policy has been the victim of a cure that has nearlyproven fatal While the detrimental effects of fiscal contraction were recognized by

many monetary policymakers, structural reforms have been viewed at as “a panacea

to jump-start growth and generate employment” (p 143) Against the backdrop of

hysteresis, the combination of procyclical fiscal, impotent structural and insufficientmonetary policy is now yielding medium- to long-term effects

After describing the current situation, Bützer looks at options available now

He analyses their potential in keeping the Euro area together and leading the wayout of depression Simultaneously, he asks whether the expansionary effects ofthese policies are outweighed by their disadvantages in terms of financial stabilityand redistributive effects He concludes that conventional monetary policy and

quantitative easing “have run out of steam at the zero lower bound and increasingly

pose risks to financial stability, the outright creation of broad money through lump-sum transfers from the central bank to private households may well be the most effective measure to achieve the Eurosystem’s primary objective and lift the economy out of its slump” (p 155).

He recognizes that there are dangers associated with putting the central bank

in such an exposed position In the end, however, he favors managing credibility,independence, and financial stability risk to letting the Eurozone unravel

Bützer’s analysis illustrates the ever expanding universe of central bank ments This points to a policy challenge that emerging-market central banks havealready faced long before the crisis In these countries, monetary policy has often

instru-been characterized by the use of multiple instruments Sometimes this has instru-been due

to multiple objectives In other cases, central banks have felt that a combination ofinstruments might be preferable to achieve a single goal

Using the example of foreign-exchange-market interventions, Ivana Rajkovi´cand Branko Uroševi´c (2017) develop a framework to analyze this multiplicity.The context is a small open economy with pronounced euroization It follows aninflation-targeting strategy In such a dual currency setup, the degree to whichforeign currency is employed to store value or extend credit affects how the policyrate is set If interest rates are the only instrument, monetary policy faces constraintsthat can be relaxed by foreign-exchange interventions The responses to domesticand international shocks become less extreme and policy is less distortionary.However, to successfully operate with different instruments requires pre-conditions

In particular, central-bank risk management needs to be developed further to takeinto account the cost of foreign exchange interventions Furthermore, monetary andmacroprudential polices have to be calibrated jointly

This important take-away from the chapter of Rajkovi´c and Uroševi´c is furtherrefined in the next three contributions of this volume that deal with the conceptualbasis, measurement, and data requirements of macroprudential regulation

Katri Mikkonen (2017) reviews recent contributions to macroprudential policyanalysis She first looks at the relationship among macroprudential, monetaryand microprudential policies, emphasizing synergies and the need to focus oncomparative advantages In a second part, she presents an operationalization ofmacroprudential policy Recent work at central banks has come up with new ways

of risk identification and assessment With a view to get a holistic picture ofmacro-financial risks, qualitative and quantitative techniques have been married

in innovative ways, for example in novel early warning systems Recent workhas also come up with new views on macroprudential instruments, for examplecountercyclical capital buffers, loan-to-income ratios, or a time-varying net stablefunding ratio

Mikkonen concludes that much has been done to improve macroprudentialpolicy However, policies so far cannot be based on a stable set of stylized factsand instruments The financial cycle has received less attention than the businesscycles Missing data and tools to model complexity in quickly changing systems

limit the applicability of many models “There is no universally accepted dogma for

macroprudential policy” (p 196) Trial and error will remain important elements

of existing policy approaches Much more empirical research needs to be carriedout

Manuel Mayer and Stephan Sauer (2017), in their contribution, study prudential aspects of measuring credit risk Though the practice is currentlycontested, banks use their own estimates for the probability of default and the lossgiven default The respective models follow different approaches Accordingly, animportant distinction with macro-financial relevance is the one between point-in-time (PIT) models (using all currently available information) and through-the-cycle(TTC) models (canceling out information that depends on the current position

macro-in the macro-fmacro-inancial cycle) TTC models are often viewed as favorable for

macroprudential regulation, because credit risk estimates do not improve riorate) in a boom (recession) Thereby, constant equity requirements are lessprocyclical than if risk weights need to be adjusted when risk is measured by PITmodels.

(dete-Mayer and Sauer question the perceived superiority of TTC, performing a range

of empirical tests on the relative reliability of the two methods They show thatTTC are more difficult to validate Having a theoretically good but empiricallyquestionable method might do more harm than good It also opens the door formisunderstandings between the supervisors and the supervised Taken together,their arguments favor PIT models for measurement purposes To compensate forthe pro-cyclical nature of these models, the authors argue for a more extensive use

of counter-cyclical capital buffers

Florian Kajuth (2017) concludes Part II with a discussion of a current dential topic, the rise in house prices, in particular in German urban agglomerations.House price developments are crucial to understand macro-financial dynamics(Illing and Klüh2005) The analysis looks at German house prices from at leasttwo different angles One the one hand, it discusses issues of data availability andquality, comparing parametric and non-parametric approaches In this way, it raisesawareness for an often neglected but extremely important issue: the availability(or lack thereof) of data for macroprudential and other policy purposes On theother hand, the chapter asks whether there is reason for concern Did expansionarymonetary policy result in substantial overvaluations, thus giving rise to prudentialconcerns?

macropru-Kajuth provides extensive evidence for the deplorable state of property pricestatistics in Germany In particular, there is a lack of time series that go backsufficiently in time Moreover, existing statistics lack comprehensiveness It istherefore necessary to rely on cross-sectional variations of housing markets inGermany Using this information and a range of other sources confirms that someurban areas do indeed seem to be overvalued For Germany as a whole, however,there is no indication of a bubble, at least not yet

The financial crisis has left a deep mark on the kind of topics that are on conomists minds New methods have evolved, and macroeconomic issues havebecome more interesting to those who were previously focused on microeconomics.Macroeconomics has changed quite a deal since Lucas’s now infamous quote

macroe-that “depression prevention has been solved” (Lucas 2003, p 1) The chapters

in this volume reflect some of these developments Macroeconomics is currentlyundergoing a period of re-conceptualization (Blanchard et al 2010) This periodstarted already before the crisis, but went largely unnoted, with few exceptions,such as the ones discussed in Beetsma and Illing (2005) The final section of this

volume looks at five elements of this trend:

– A renewed focus on stylized facts, economic history and path dependence,– The application of established methods to new problems, such as the institutionalstructure of the Euro area,

– The application of new methods to old topics, building in particular on insightsfrom behavioral and experimental economics,

– The resurgence of distributional issues as a topic of macroeconomic research,and

– The emergence of inter-disciplinary work to re-embed economics in the socialsciences and contextualize its findings

Axel Lindner (2017), in the first chapter of Part III, shows that going back alittle further can yield important insights about the present situation He looks at themacroeconomic effects of German unification and argues that the German economyhad been off steady state already before unification At the same time, Germanyseems to have been on a trajectory that very much resembles the dynamics that

we now associate with the anamnesis of the Euro crisis In particular, investmentwas trending down already before unification, and continued to do so after a briefjump in the beginning of the 1990s Moreover, the financial balance had been

on an increasing trend during the eighties, a trend it returned to around 10 yearsafter unification The wage share in national income follows a similar pattern, yetwith the opposite sign These observations cast some doubts on the view that thesedevelopments were a consequence of introducing the Euro

The problems of the Euro area are at the core of the chapter by Ray Rees andNadjeschda Arnold (2017) They ask whether insurance-based approaches can helpsolving the sovereign default problem and argue that the economics of insurancemarkets can guide a redesign of the common currency area This redesign seeks

to preserve decentralized fiscal policy Its main idea is to use risk-based insurancepremia as an instrument to increase fiscal discipline Rees and Arnold encouragethe creation of an independent insurance agency This agency ensures incentivecompatibility by promising to remove the threat of sovereign default if certain

conditions are fulfilled Its main instrument are risk-based premia “payable ex

ante into a mutual fund that must at least break even in expectation” (p 267) In

case of a fiscal emergency, the mutual fund arranges automatic payouts Regularreviews of fiscal plans, minimum insurance reserves, and reinsurance arrangementscomplement the set-up

Rees and Arnold compare this insurance-based approach with the existingEuropean Stability Mechanism and different suggestions for Eurobonds Theyconclude that none of these alternatives is incentive compatible, because they fail

to make the costs of default risk accountable for governments ex ante

Camille Cornand (2017) shows in her contribution that new empirical approachescan yield important insights about macroeconomic phenomena In an attempt toprovide additional foundations for the non-neutrality of money, she compares therole of three potential explanations for nominal rigidities: sticky prices à la Calvo(1983), sticky information à la Mankiw and Reis (2002), and limits to the level

of reasoning that price setters achieve The latter is based on the observationthat subjects in laboratory experiments fail to reach common knowledge wheninformation abounds.

Cornand uses the data from an experiment by Davis and Korenok (2011), inwhich subjects play the role of price-setting firms in a macro-environment withstochastic demand shocks The data reveal a sluggish adjustment to shocks, even

if these shocks are publicly revealed Cornand investigates which model yields thebest fit of these price adjustments and finds that the sticky-information model fitsbest

Selecting models on the basis of laboratory experiments provides an alternative

to assuming artificial frictions in macroeconomic models Experimental data alsoallow estimating behavioral parameters independently from other model parameters,while empirical tests with macroeconomic field data allow only a joint estimation

of all model parameters The estimated behavioral parameters may then be used forcalibrations and as restrictions in the joint estimates of other model parameters withmacroeconomic field data

One should not underestimate the significance of these and other behavioralinsights into wage and price stickiness The rejection of Keynesianism by Lucas andothers was largely justified with the argument that Keynesians were unable to derivesuch stickiness from micro-founded models with optimizing agents The data fromexperiments and the results from behavioral economics more generally show thatnominal rigidities and non-rational expectations are just a fact of life This makespragmatic reasoning much easier, as it is not hampered anymore by the requirementthat all macroeconomic variables need to be derived from rational choices

In the penultimate chapter, Dominique Demougin (2017) analyses an issue thatmore and more dominates the policy debate After having long been relegated tothe fringes of macroeconomics, the rising inequality of income and wealth nowtakes center stage Using an incentive contract approach, Demougin provides anovel explanation for this trend Information and communication technology allowsmanagers to better monitor worker behavior This redistributes informational rentsfrom the bottom to the top of the income distribution While middle managementwins, firm owners win big They do not just gain from a redistribution of rents from

a given output: they also benefit from increased worker effort and productivity Themirror image of this effect is that workers are penalized twice They lose the rentsthat they had enjoyed before, and they suffer from a work environment that requireshigher effort

Demougin uses a standard hidden action problem to explain increasing incomeinequality The argument is solely based on the organizational structure of the firmand, thus, provides an alternative to standard explanations based on globalization

or skill-biased technological progress Demougin’s numerical exercise replicates

a sizeable number of crucial features of the macroeconomic environment since theearly 1970s While technology advances, wages dynamics are at best subdued, if notstagnant The wage share in income declines Working conditions are increasinglyresembling a treadmill with little space for discretionary decisions by workers.Certain groups of the society are able to keep up as middle managers, so the wage

distribution starts to become more uneven But the strongest implication of the rise

in information and communication technology is that the very top of the incomeand wealth distribution experiences large gains, a feature that cannot be explained

by skill-biased technological progress

Moritz Hütten and Ulrich Klüh (2017), in the final chapter of this volume, pick

up the fact that macroeconomic developments since the end of the Bretton-Woodsregime display peculiar characteristics Not only has there been a redistributionfrom the bottom to the top and from labor to capital, but in parallel, inflationhas come down and is too often close to deflationary levels Unemployment hasbecome a constant feature of capitalist societies, while it was largely absent in thedecades before Public debt as a share of GDP has trended up, in part because theincidence of financial crisis has increased continuously Exchange rates and otherprices on financial markets have exhibited a degree of volatility seemingly unrelated

to fluctuations in fundamental variables

All this has taken place in a context in which the task of stabilizing conomic and financial fluctuations has been concentrated in the hands of centralbanks These, in turn, have largely bought into the notion that some degree ofunemployment is necessary to keep inflation in check, in particular the very lowinflation targets that have become standard Fiscal policy has been confined toimplement a regime of institutionalized austerity (Streeck and Mertens2010) Andstructural policies have often followed the prescriptions of the so called Washingtonconsensus

macroe-Hütten and Klüh argue that the beginning of the 1970s is a watershed betweentwo ways of organizing economic activity in capitalist societies The end of theBretton-Woods system did not only change the way exchange rate movements andinternational capital flows are organized A “regime change” occurred that led to adynamic adjustment of capitalism, in which finance becomes increasingly important(financialization) Regrettably, there have been only few attempts to characterizethese two phases of economic history holistically

The chapter first introduces the concept of “macro regimes” as a framework for

analyzing macroeconomic aspects during periods of large social transformations.Building on approaches from political science and sociology, macro regimes aredefined as arrays of implicit or explicit principles, norms, rules and decision-formation procedures that lead to a convergence of actor expectations Both theconvergence of expectations (i.e the emergence of regimes) and the divergence ofexpectations (which usually marks the beginning of a regime change) are reflected

in specific characteristics of time series

In the view of many observers from other social sciences, a characteristic feature

of the macro regime in the last four decades is the increasing role of finance insociety This element of the current macro regime, often coined financialization, isthe focus of the chapter Can the macro regime approach itself explain financializa-tion? What does financial sociology contribute to its understanding? And how couldfinancialization happen on the watch of economic experts that now frequently rejectit?

Thereby, this volume ends with a reflection on the roles that economics in generaland macroeconomics in particular play in our society This question has also been

characteristic for the symposium held in honor of Gerhard Illing and for Bob

Solow’s letter at the very end of this book Solow asks: “Why is it so difficult?”

referring to the combination of expert technique with common sense in economics.One explanation might be that economics is faced with a difficult double role Onthe one hand, it can be considered a science (the objective of which is to distinguishtrue and false) On the other hand, it is a toolbox for policy Put differently, it is

a language that is employed within the economy to organize discourse about theeconomy In this second role, it is highly political, applied, sometimes useful, andsometimes counterproductive

Gerhard Illing has taught many people how to walk on the fine line betweenacademic scrutiny and policy relevance that emerges from this double role

References

Baeriswyl, R (2017) The case for the separation of money and credit In F Heinemann, U Klüh,

& S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 105–121) Cham: Springer.

Beetsma, R., & Illing, G (2005) Revival of aggregate demand policies – Introduction CESifo Economic Studies, 51, 497–509.

Blanchard, O., Dell’Ariccia, G., & Mauro, P (2010) Rethinking macroeconomic policy Journal

of Money, Credit and Banking, 42(s1), 199–215.

Bützer, S (2017) (Monetary) Policy options for the euro area: A compendium to the crisis.

In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 125–162) Cham: Springer.

Calvo, G (1983) Staggered prices in a utility maximizing framework Journal of Monetary Economics, 12, 383–398.

Cao, J., & Illing, G (2010) Regulation of systemic liquidity risk Financial Markets and Portfolio Management, 24(1), 31–48.

Cao, J., & Illing, G (2011) Endogenous exposure to systemic liquidity risk International Journal

of Central Banking, 7, 173–216.

Cao, J., & Illing, G (2015) ‘Interest rate trap’, or why does the central bank keep the policy rate

too low for too long? The Scandinavian Journal of Economics, 117(4), 1256–1280.

Cornand, C (2017) Appraising sticky prices, sticky information and limited higher order beliefs

in light of experimental data In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 297–306) Cham:

Springer.

Davis, D., & Korenok, O (2011) Nominal price shocks in monopolistically competitive markets:

An experimental analysis Journal of Monetary Economics, 58, 578–589.

De Grauwe, P., & Ji, Y (2013) Self-fulfilling crises in the Eurozone: An empirical test Journal of International Money and Finance, 34, 15–36.

Demougin, D (2017) Rising income inequality: An incentive contract explanation In F

Heine-mann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 307–323) Cham: Springer.

Diamond, D W., & Rajan, R (2001) Liquidity risk, liquidity creation, and financial fragility: A

theory of banking Journal of Political Economy, 109, 287–327.

Fecht, F., & Tyrell, M (2017) Optimal central bank policy in different financial systems In

F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 27–58) Cham: Springer.

Goodhart, C (2017) Balancing lender of last resort assistance with avoidance of moral hazard.

In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 19–26) Cham: Springer.

Goodhart, C., & Illing, G (2001) Financial crises, contagion and the lender of last resort: A reader Oxford: Oxford University Press.

Heinemann, F., & Illing, G (2002) Speculative attacks: Unique sunspot equilibrium and

trans-parency Journal of International Economics, 58(2), 429–450.

Holler, M., & Illing, G (2009) Einführung in die Spieltheorie (7 Auflage) Berlin: Springer.

Holtfrerich, C., Feld, L., Heun, W., Illing, G., Kirchgässner, G., Kocka, J., Schularick, M.,

Streeck, W., Wagschal, W., Walter, S., & Weizsäcker, C (2015) Staatsschulden: Ursachen, Wirkungen und Grenzen (Bericht einer Arbeitsgruppe im Auftrag der Nationalen Akademie

der Wissenschaften Leopoldina) Berlin: Union der deutschen Akademien der Wissenschaften

e V.

Hütten, M., & Klüh, U (2017) No more cakes and ale: Banks and banking regulation in the

post-bretton woods macro-regime In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 325–349) Cham:

Springer.

Illing, G (1985) Geld und asymmetrische Information Studies in Contemporary Economics 13.

Berlin: Springer.

Illing, G (1992) Neue Keynesianische Makroökonomie Tübingen: Mohr-Siebeck.

Illing, G (2007) Financial stability and monetary policy – A framework (CESifo Working Paper

No 1971) April 2007.

Illing, G (2012) Finanzmarktstabilität – die Notwendigkeit eines effizienten Regulierungsdesigns.

In M Held, G Kubon-Gilke, & R Sturn (Hg.), Lehren aus der Krise für die Makroökonomik

“Jahrbuch Normative und institutionelle Grundfragen der Ökonomik” Band 11, 2012:283-306.

Illing, G., & Klüh, U (Eds.) (2004) Spectrum auctions and competition in telecommunications.

Boston, MA: The MIT Press.

Illing, G., & Klüh, U (2005) Vermögenspreise und Konsum: Neue Erkenntnisse, amerikanische

Erfahrungen und europäische Herausforderungen Perspektiven der Wirtschaftspolitik, 6(1),

1–22.

Illing, G., & Peitz, M (2006) Industrial organization and the digital economy Cambridge, MA:

The MIT Press.

Illing, G., & Siemsen, T (2016) Forward guidance in a model with price-level targeting CESifo Economic Studies, 62(1), 47–67.

Illing, G., & Watzka, S (2014) Fiscal multipliers and their relevance in a currency union – A

survey German Economic Review, 15(2), 259–271.

Kajuth, F (2017) Assessing recent house price developments in Germany – an overview In

F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 225–235) Cham: Springer.

Klüh, U (2014) Sismondis Spur: Krisen- und Selbstverständnis der Ökonomik In M Held, G.

Kubon-Gilke, & R Sturn Normative und institutionelle Grundfragen der Ökonomik Marburg:

Metropolis Verlag.

Lindner, A (2017) German unification: Macroeconomic consequences for the country In F.

Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the conomy: Festschrift for Gerhard Illing (pp 239–263) Cham: Springer.

one-hundred fifteenth meeting of the American Economic Association, January 4,

Lucas%20priorities%20AER%2003.pdf

Lux, T (2017) Network effects and systemic risk in the banking sector In F Heinemann, U Klüh,

& S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 59–78) Cham: Springer.

Mankiw, G., & Reis, R (2002) Sticky information versus sticky prices: A proposal to replace the

new Keynesian Phillips curve Quarterly Journal of Economics, 117, 1295–1328.

Mayer, M., Sauer, S (2017) Are through-the-cycle credit risk models a beneficial

macro-prudential policy tool? In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 201–224) Cham:

Springer.

Mikkonen, K (2017) Macroprudential analysis and policy – Interactions and operationalization.

In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 177–200) Cham: Springer.

Rajkovi´c, I., & Uroševi´c, B (2017) On inflation targeting and foreign exchange interventions in

a dual currency economy In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 163–176) Cham:

Springer.

Rees, R., & Arnold, N (2017) Approaches to solving the eurozone sovereign debt default problem.

In F Heinemann, U Klüh, & S Watzka (Eds.), Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing (pp 265–295) Cham: Springer.

Streeck, W., & Mertens, M (2010) Politik im Defizit: Austerität als fiskalpolitisches Regime

(MPIfG Discussion Paper 10/5).

Watzka, S (2017) Contagion risk during the euro area sovereign debt crisis: Greece, convertibility risk, and the ECB as lender of last resort In F Heinemann, U Klüh, & S Watzka (Eds.),

Monetary policy, financial crises, and the macroeconomy: Festschrift for Gerhard Illing

(pp 79–104) Cham: Springer.

Frank Heinemann is professor of macroeconomics at the Berlin University of Technology.

His main research interests are monetary macroeconomics, financial crises, and experimental economics.

Ulrich Klüh is professor of economics at Hochschule Darmstadt His main research interests are

macroeconomic theory and policy, central banking, financial markets and institutions, and history and theory of economic thought.

Sebastian Watzka is senior economist at the Macroeconomic Policy Institute (IMK) at the

Hans-Böckler-Foundation Before joining the IMK he was assistant professor at the Seminar for Macroeconomics of the University of Munich, LMU His research interests are monetary policy and financial markets, financial crises, inequality and unemployment.

Liquidity From a Macroeconomic

Perspective

with Avoidance of Moral Hazard

Charles Goodhart

Abstract Solvency is rarely clearly defined, since it depends on valuations relating

to future outcomes, which are themselves affected by policy decisions, includingCentral Bank Lending of Last Resort (LOLR) Positive LOLR may cause losses andmoral hazard, whereas refusal could trigger a contagious panic Measures to limitmoral hazard, and hence allow more systemic protection include: (i) treating thefirst failure more strictly; (ii) involving other banks in any rescue; (iii) tougheningthe incentive structure for bank borrowers

If an agent is certain to repay her debts, on time and meeting all the required termsand covenants, she can always borrow at current riskless market interest rates So aliquidity problem1almost always indicates deeper-lying solvency concerns.The solvency concerns that lenders may have about borrowers may, or may not,however, be well founded I start in Sect.2by noting that the definition of solvency

is fuzzy The future likelihood of a borrower defaulting is probabilistic, and so theterms (the risk premia) and conditions on which a borrower can raise cash, heraccess to liquidity, are stochastic and time varying, Sect.3

There is a common view that a Central Bank should restrict its activities

in support of financial market stability to lending into the general market viaopen market activities, rather than lending to individual banks via Lender of LastResort (LOLR) measures I explain why I disagree with that argument in Sect.4.Nevertheless the banks most in need of LOLR will generally be those that have been

least prudent Even though the Central Bank will choose not to support the most

egregiously badly-behaved (and/or those whose failure is least likely to generate

enhanced premia that reveal existing solvency concerns to a wider public.

C Goodhart (  )

London School of Economics, Financial Markets Group, London, UK

© Springer International Publishing AG 2017

F Heinemann et al (eds.), Monetary Policy, Financial Crises,

and the Macroeconomy, DOI 10.1007/978-3-319-56261-2_2

19

secondary contagious-failures), the use of LOLR does entail a degree of insurance

(against failure) and hence generates moral hazard I discuss in Sect.5various ways

of mitigating such moral hazard

The use of language in macro-economics is slipshod2; (perhaps this helps to explainour penchant for arid mathematical models) Solvency is just such a slippery term

We think that we know what it means, i.e that the value of assets is greater thanthe valuation of the liabilities But in practice we do not, because it all depends onhow the assets (and liabilities) are valued, and that depends on the viewpoint of thevaluer, and also on the (changing) conventions and practices of the accountant.Consider, for example, the British mortgage bank (Northern Rock) in September

2007, at the time when it asked the Bank of England for liquidity assistance.Looking backwards, to the prior bubble phase, it had very few non-performingloans, and was undoubtedly solvent (historic cost accounting) Looking forwards,

to the likely future bust phase in housing, it was most probably insolvent (since ithad expanded aggressively), as turned out later to be the case

Moreover, the assessment of the solvency of an institution, especially one seeking

LOLR assistance from a Central Bank (CB), is not independent of the CB’s own

actions and of the wider public’s (the market’s) interpretation of those same actions,(as in the case of Northern Rock).3 The valuation of a going concern (where anyhelp has been covert) is much greater than that of a concern, which is either gone or

2 Examples are:

1 ‘Real’, as in real interest rates: Really means ‘adjusted for (expected) price changes’, but whose expectations and what prices? Not much ‘real’ about it; at best ‘uncertainly measured adjustment for future price changes’.

2 ‘Natural’, as in the natural rate of unemployment Really means the level at which some other variable, e.g inflation, would remain stable Often treated as being synonymous with

‘equilibrium’, but equilibrium carries a connotation that there are forces restoring such an equilibrium, once disturbed This latter remains contentious.

Central Bank will now restore order, or of greater concern, i.e I did not know things were so bad.

In the case of Northern Rock, Robert Peston of the BBC leaked that LOLR and had no incentive

to calm the public Moreover, Northern Rock had many depositors who interacted electronically When a large number of these sought to withdraw simultaneously, the Northern Rock website crashed The depositors interpreted this as a refusal of Northern Rock to allow withdrawals, and physically ran to do so from their nearest branch Similarly when the authorities, e.g the Treasury, guarantees the withdrawal value of an asset, as in the Irish bank deposits or US Money Market Mutual Funds, in 2008, this may calm the situation so that no further supporting action is needed But alternatively, if the potential financial losses are feared to be large and the solvency of the guarantor is itself questionable, as it was in the Irish case, this can lead to both entities, guarantor and guarantee, dragging each other down, in a ‘doom loop’.

needed patent public help to continue; hence there is a serious stigma effect of beingobserved to need LOLR assistance from the Central Bank, with potentially severeeffect in delaying and distorting recovery processes.

Accountants have their own incentives Although it is a crime to continue tradingwhen knowingly insolvent, I am unaware of any bank having closed its doorsbecause its accountant told them to do so But, once a bank does close, most oftenbecause its liquidity problems become insuperable, the incentive of an incomingforensic accountant will be to exaggerate the potential scale of problems, therebyfuelling potential panic and risk aversion, because such an accountant will not want

to have to claw back money from creditors on a future occasion, to meet furtherlosses, if, indeed, such future claw back can be done at all Too often creditors areoriginally told to expect large losses, whereas, after several years and a recoveryfrom the crisis, they get paid back practically in full, (e.g as in Lehman BrosLondon)

So (forward-looking) solvency problems exhibit themselves as liquidity problems.Borrowers, including banks, would, as a generality, not have a problem in gaining(funding) liquidity from markets if they were perceived as absolutely certain to payback in full as contracted There are a very few technical exceptions, where timing,terrorism, IT breakdowns, as with Bank of New York in the 1980s, or some otherexogenous event prevents access to markets; but the common, heuristic rule is that

a shortage of liquidity presages (market) concerns about solvency

The CB then has to balance concerns both about potential loss from LOLRlending and the implications of being seen to support risk-loving, even reckless,management on the one hand (if it does support), against concerns about fuelling thepanic, amplifying downwards pressures on asset prices and contagion on the other,(if it does not support) It is a difficult act of judgment, and there are no absoluteclear rules

In particular, the ‘solvency’ of any potential borrower is not a deterministic,exogenous, knowable datum, but depends on many time-varying future develop-ments, not least how the CB itself responds to requests for LOLR assistance, andwhether (and how) that becomes known There is a most unhelpful misinterpretation

of Bagehot (1873) that contends that he claimed that the Bank of England shouldonly lend to solvent institutions

But the Bank of England had no supervisory powers then, or the right toinspect other financial institution’s books So how could the Bank of England knowwho was solvent, and who not? Instead, what he meant, and said clearly, in hissecond rule for LOLR is that the Bank of England should lend freely on all ‘good

securities’.4The criterion for Bagehot was the quality of the collateral, which could

be assessed,5rather than the solvency of the borrower, which could not be

There is a common view, more prevalent in the USA than in Europe, that theauthorities, including the CB, should intervene as little as possible in markets, and/orthat markets are better informed (efficient market hypothesis) than any authority can

be, (despite CB’s role as supervisor) If so, so it is asserted, in a panic the CB shouldprovide liquidity to the market as a whole via open market operations, and leave thedistribution of such liquidity to the market, which will sort out those deserving of

support from those who should be let go, (better than CB).

This is, I believe, wrong, because it fails to grasp the dynamics of contagion In

a panic, the weakest is forced to close Its failure will worsen the crisis The marketwill then withdraw funds from the next weakest, further amplifying the downwardsspiral To prevent total collapse at some point the authorities will have to step in to

support every institution which can meet certain criteria, as the G20 did in October

2008 Bagehot’s criterion was the availability of ‘good collateral’

Such was the political revulsion from public sector support, ‘bail out’, of thebanking sector in the USA, that the conditions under which the Fed could provideliquidity support to individual financial institutions were made somewhat more

good security to offer’, p 198, (1999 version: John Wiley: NY HG3000 L8283).

are two answers to this, the first being more applicable to the nineteenth century, the second more

to subsequent centuries, twentieth and twenty-first First, during panics financial markets tend to become dysfunctional, with no one being prepared to part with cash at any reasonable price In such circumstances, the Central Bank is not only the Lender of Last Resort, but also the market maker of last resort In such a situation what interest rate should it set? As Bagehot states, a ‘high’

one, but obviously not a ‘penalty’ rate Bagehot never uses the word ‘penalty’ in this context.

Second, such has become the stigma of being seen to borrow on LOLR terms from the Central Bank that banks tend to use up all their good quality collateral to borrow from the market, before turning, if all else fails, to the Central Bank for succour With banks also of the view, prior to 2007–2009, that they could always borrow cash in wholesale markets (funding liquidity), they had run down their holdings of high quality liquid assets to almost nothing at the start of the Great Financial Crisis So amongst the various unconventional monetary measures then taken were those that swapped less liquid assets (held by banks) for more liquid assets, e.g Treasury Bills The Bank

of England’s Special Liquidity Scheme is a prime example In the aftermath of the Great Financial Crisis various requirements have been put in place, such as the Liquidity Coverage Ratio, to try to ensure that banks will always have enough high quality liquid assets to enable banks to be rescued from a panic, and associated liquidity troubles, without forcing the Central Bank to choose between accepting poor collateral, i.e taking a credit risk, and letting that bank fail.

restrictive under Title XI, Sections 1101–1109, of the Dodd-Frank Act, passed bythe House of Representatives in H.R 4173, pp 738–752, passed in 2011 Underthis,

A Section 13.3 lending, under which previously the Fed could lend to anybody,

not just to banks, under ‘unusual and exigent’ circumstances has been curtailed.

In future the Fed can only lend to eligible banks, and/or to “any participant inany program or facility with broad-based eligibility” What does this mean inpractice?

B The Fed cannot now lend to ‘insolvent’ borrowers; though [the CEO of] theborrowing bank may certify the solvency of her bank, with a duty to update anymaterial information on such solvency

C More information on such emergency liquidity assistance has to be provided,and sooner

Also provision of additional guarantees to depositors and other creditors of

financial institutions can only be provided after a ‘liquidity event’ is agreed by theFederal Reserve Board, Federal Deposit Insurance Corporation and the Executive(President and Secretary of the Treasury) This must then be accepted by bothHouses of Congress

All this could make emergency liquidity assistance in a crisis less flexible,and make the Federal Reserve Board’s freedom of action constrained by legalinterpretation of the Dodd-Frank Act Would, for example, the Fed be expected

to audit the books of a potential borrower, prior to granting emergency liquidityassistance, or could it rely on the borrower’s self-certification? If Bank A borrowedfrom the Fed in, say, May 2017 and then subsequently went into bankruptcy in July

2017, would there be (political) penalties, and, if so what, on the Fed and/or theself-certifier?

It is my view that the Dodd-Frank Act has already imposed undesirably rigidconstraints on the Fed’s flexible freedom of manoeuvre to respond to financialcrises, though this is contentious The Warren/Vitter Bill would have made such

constraints much tighter, and the Fed has accepted, in November 2015, that

‘broad-based’ means at least five participants

The contrasting view is that the Fed used a legal loop-hole, in Section 13.3, toexpand its powers to act in a way that was close to, if not beyond, its proper capacity,i.e ultra vires, as even Volcker complained Rules of behaviour and accountabilityshould be made by the legislature The problem with that is that no one can foresee

the future So, binding the hands of the authorities tightly in advance, ex ante, may

force them to stand idly by as the financial system unravels, as turned out to be thecase with the failure of Lehman Bros Accountability to the legislature for actionsalready taken, ex post, is necessary, but tight prescription in advance overlooks theinherent uncertainty of an ever changing financial system The future will not just

be a re-run of the past, not just a different draw from an unchanging probabilitydistribution

A retort is that either the CB would have to lend in crisis conditions nately to everybody, which would generate moral hazard, or would have to rely onambiguity, which in a crisis would be the reverse of ‘constructive’ So if one wants a

indiscrimi-CB to be flexible and accommodating not only in a crisis but even in the instance of

a potentially dangerous disturbance, how might one proceed to limit ‘moral hazard’?

One of the traditional roles of Central Banks has been to maintain financial stability,

as emphasized in my1988book on The Evolution of Central Banks But policies,

such as LOLR and market maker of last resort to achieve this end, represent a form

of insurance to commercial banks, and hence entail a risk of less prudent behaviour,since the banks believe that the Central Bank may save them from the adverseconsequences of risk-taking, i.e moral hazard

If we want, as I do, the Central Bank to continue to have responsibility forfinancial stability, and hence use its various policy instruments flexibly to this end,then what precepts can be applied to limit the accompanying moral hazard?There are some potential precepts:

5.1 Treat First Worst

The first casualty of a financial downturn is likely to have been the most egregiouslyexposed and the most aggressive risk-taker So a general precept should be totreat this most toughly, and become increasingly accommodating as contagion thenthreatens Thus when the CB allows the first institution in a class to be liquidated,

‘pour encourager les autres’, it needs to stand ready to support the rest of that class.

That principle was adopted by Governor Eddie George in 1995, when Barings wasnot supported, but steps were taken to prepare support for the remaining class ofBritish merchant banks

The liquidity problems of foreign-headquartered banks, and the foreign-currencyproblems of domestic banks, within a global financial system involve somewhatseparate issues which will not be tackled here

5.2 Involve the Other Banks, if Possible

While the CB should take the lead, it would be helpful to involve other banks in anyprocess of cross-guarantees and mutual self-help This not only reduces the public-sector burden, but also encourages additional information on status and reputation.This was done in the UK between Barings 1 (1891) and the Fringe Bank Crisis

(1974/1975), but collapsed under the strains of global banking (Johnson MattheyBankers 1984) A US example is Long Term Capital Management (1998).

With the potentially failing bank being a competitor, other commercial banks arenot likely to agree to coordinated mutual assistance unless they are persuaded that

the damage to the working and reputation of the banking system as a whole merits

it So a need to persuade other banks to participate in a rescue can act as a brake on

a Central Bank which otherwise might seek to rescue a ‘bad’ bank whose demisewould actually have benefitted the remaining system

How far can this be resurrected?

5.3 Change the Incentive Structure

In the first half of the nineteenth century, most bank shareholders had unlimitedliability, and this was a check on consciously risky behaviour But when businessborrowers became large and the efficient size of banks increased in line, in thelatter half of the nineteenth century, it became necessary to attract bank equity fromoutside shareholders, who could not individually control or monitor the bank Suchoutsiders could not be tempted to purchase bank equity if it had unlimited liability.But for many decades in the USA, until the 1930s, bank shareholders had doubleliability in the sense that, if their bank became distressed, not only would their sharevalue go to zero, but also they could be required legally to inject further fundinginto their bank equal to the original par value of their shares Clearly one could

go beyond such double to multiple liability for certain shareholders, and require alldesignated insiders to hold, or to be allotted, shares with such multiple liability.Among such insiders could be:

• Those with large share-holdings,

• Board members,

• Senior executives,

• Staff earning more than X over some period

Other steps to enhance more prudent behaviour might be:

1 Bail-inable debt for large creditors

2 Require all bonuses to be paid in bail-inable debt

3 Establish a Supervisory Board with a wider cast of stakeholders, includingrepresentatives of staff and creditors Workers have more of their human capitaltied up in their firm than more diversified shareholders and hence should be morerisk averse

4 Current regulatory proposals, e.g on the use of bail-inable debt and claw-backs

of prior bonuses in the event of failure, have gone some slight way in thisdirection, but not, in my view, far enough

6 Conclusions

My own preference would be to change the incentive structure dramatically, ‘makethe punishment fit the crime’, but leave the CB with flexibility and leeway to balancethe requirements of preventing financial instability against concerns about ‘moralhazard’

The Dodd-Frank Act has given too much ground to the moral hazard talists We may live to regret that

fundamen-References

Bagehot, W (1873) Lombard Street: A description of the money market In E Johnstone &

H Withers (Eds.), The library of economics and liberty London: Henry S King and Co.

http://www.econlib.org/library/Bagehot/bagLom.html (11 Sept 2017)

Goodhart, C (1988) The evolution of central banks Cambridge, MA: MIT Press.

US House of Representatives (2010) Dodd-Frank Wall Street Reform and Consumer Protection Act Public Law 111–203 [HR 4173].

Charles Goodhart is a professor of banking and finance (emeritus) at LSE, and a former member

of the Bank of England’s Monetary Policy Committee His main research interests are central banking, monetary policy and financial regulation.

in Different Financial Systems

Falko Fecht and Marcel Tyrell

Abstract In a framework closely related to Diamond and Rajan (J Polit Econ

109:287–327, 2011) we characterize different financial systems and analyze thewelfare implications of different central bank policies in these financial systems

We show that in case of a large negative liquidity shock, liquidity demand has lowerinterest rate elasticity in a bank-based financial system than in a market orientedfinancial system Market interventions, i.e non-standard monetary policy measures

to inject liquidity need to be much larger in a bank-based financial system in order tobring down interest rates to sustainable levels Therefore, in financial systems withrather illiquid assets an individual liquidity assistance might be welfare improving,while in market oriented financial systems, with rather liquid assets in the banks’balance sheets, liquidity assistance provided freely to the market at a penalty rate islikely to be efficient While the costs of individual support might not be worthwhile

in a market oriented financial system in which deadweight losses of market basedsupport are small, in a bank based system the deadweight losses of unconventionalmonetary policy are large and thus individual support more efficient

JEL Classification: D52, E44, E52, E58, G21

© Springer International Publishing AG 2017

F Heinemann et al (eds.), Monetary Policy, Financial Crises,

and the Macroeconomy, DOI 10.1007/978-3-319-56261-2_3

27

crisis of 2007–2009 For instance, the Federal Reserve (FED) acted at these times

by executing its open market operations (OMOs) with a small group of bank) primary dealers as direct trading counterparts Other banks only had indirectaccess to the central bank via Primary and Secondary Credit Facilities (PCF andSCF), the so called ‘Discount Window’, which however were left largely unusedbecause the visibility of drawing on the discount window created a stigma effect.That stood in contrast to the European Central Bank (ECB) system where allfinancial institutions subject to reserve requirements had access to the regular ESCBauctions and standing facilities, and the recourse to a marginal lending facility(MLF) led to no stigma since it was not inferable Thus, the FED interveneddirectly in financial markets by outright purchases and sales with primary dealerand relied on a proper functioning of financial market for the reallocation ofliquidity from primary dealers to the rest of the financial system reflecting themarket-based nature of its financial system In contrast, the ECB allotted liquiditydirectly through repurchase agreements to large number of banks against collateral,reflecting the bank-dominated structure of the euro area Hence the different centralbank policies frameworks in place before the Great Financial Crisis (GFC) reflectedthe differences in the financial systems

(non-During the GFC the need for broad liquidity support became obvious Anextraordinary liquidity shock affected the global financial system and impaired thefunctioning of national and international financial markets In particular the FEDreacted rapidly Since the FED’s monetary policy framework relied on a functioninginterbank market and on central bank’s counterparties being willing and able tolend to the institutions which are most liquidity-starved, the FED had to adapt itsoperational framework immediately when markets dried-up The FED introducedthe Term Auction Facility (TAF), available to all depository institutions, whichpreviously had only access to the discount window In addition, a Primary DealerCredit Facility (PDCF) was created for securities dealers Thus during the GFCthe FED expanded its operational framework to provide liquidity directly to furthercounterparties, which made the FED’s framework effectively more similar to that ofthe ECB

However, in response to the enduring crisis central banks all over the world alsoadapted their operational framework and introduced non-standard monetary policymeasures to inject liquidity on a large scale In this respect the FED was much moreaggressive and proactive Already in 2008 the FED enacted it quantitative easingprogram centred on outright asset purchases of Treasury securities in particular.Only in July 2012 Mario Draghi announced to do “whatever it takes” and launchedthe Outright Monetary Transactions (OMT) program of the ECB, followed by an

‘expanded asset purchase program’ of euro-area bonds which was started in March

2015 Thus notwithstanding some differences in detail and motivation, the ECBcopied the large scale liquidity allotment through outright asset purchases from theFED but with a significant delay However, given the prevailing differences of thefinancial systems on the two sides of the Atlantic one might wonder whether thisresponse by the FED is indeed also most efficient for the bank-dominated financialsystem of Continental Europe

More general, the main question we want to address in the paper is the following:Should a lender of last resort (LOLR) in a traditional bank-dominated financialsystem respond differently to liquidity shortages than in a more market-orientedfinancial system? Using a framework that strongly relates to Diamond and Rajan(2001) we argue that the relevance of relationship lending and securitization can becaptured in differences in the pledgeability of bank returns and the liquidation value

of bank assets Thus we find that in bank-dominated financial systems liquidityshortages spike asset prices more severely given no governmental intervention.Therefore, asset purchase programs that provide liquidity through the asset market,i.e quantitative easing, generate in our framework a windfall gain for liquidityrich banks These windfall gains are the larger the less liquid bank assets are.Hence in a financial system like the one of the U.S in which the bank assets aremostly securitized and tradeable, in which shadow banks play an important roleand in which relationship lending is of minor importance, these windfall gains atthe expense of the government are rather modest In contrast, in the Euro areafinancial system in which firm bank relationships are essential to ensure financing

of households and firms and in which securitization and loan sales to other financialintermediaries play a subdued role liquidity rich banks receive larger windfallprofits from asset purchases Consequently, governments confronted with the lattertype of financial system have stronger incentives to avoid these windfall profits.Governments in those countries benefit from providing individually tailored creditlines that ensures that liquidity support is allocated efficiently but entails largerinformation costs on behalf of the lender of last resort

What is the intuition for our results? In our model banks serve as relationshiplender They must refinance themselves at least partially through demand deposits.Since households have no loan collection skills, they have to rely for efficientinvestments on the collection skills of a relationship lender, i.e the bank But bankscan commit to repaying households only by issuing deposits As Diamond andRajan (2001) show, the demandable nature of deposits create a collective actionproblem for depositors They will individually run to demand repayment in casethey anticipate that the bank cannot, or will not, pay the promised amount Sincebankers will lose all rents when there is a run on the bank, they will repay wheneverthey can For that reason deposits serve as a commitment device

Now consider a situation where a regional business cycle shock leads to a certainquantity of loans being overdue Banks can either call loans due or rollover its loans

On the one hand, rollover requires further refinancing, which however can only beprovided by entrepreneurs whose projects already were successful On the otherhand, collecting loans which are called due leads to firm defaults Firm profits arelost, but the bank can seize the assets of the firm and redeploy them to the next-bestuse, resulting in a loss of overall surplus Thus rolling over the loan would be the bestsolution Yet the bank’s borrowing capacity against the pledgable returns on rolled-over loans is determined by the interest rate prevailing in the financial market For

a sufficiently large liquidity shock a bank might be unable to borrow sufficiently torepay depositors and roll over its loans Depositors anticipate that the bank obtainsnot enough liquidity The collective action problem occurs and all depositors will

start running on the bank Depositors will seize all bank asset and call all delayedloans due Firms with delayed cash-flow default and the entrepreneurs’ rents fromtheir human capital are lost Thus liquidity shocks with elevated interest rates canspark off banks runs and generate negative externalities for entrepreneurs.

Inflicting negative externalities on entrepreneurs delivers a foundation for vention by a LOLR In principle, two options are possible for a central bank First,the central bank can provide direct and discretionary liquidity support just to anailing bank and second, the central bank provides liquidity to the market in order

inter-to stabilize the interest rate at a sustainable level, which might be interpreted asquantitative easing Direct LOLR support only to ailing banks generates no windfallprofits for other banks and therefore requires less central bank liquidity injection Onthe other hand, for being effective it is essential that direct LOLR support is provided

on the basis of very precise information Otherwise, liquidity will be wasted even

by using this policy option

How does the configuration of the financial system influences the choice betweenthese two options? In our modelling framework we grasp the differences in financialsystems by assuming that bank assets in a bank-dominated financial system areless liquid than in a market-oriented financial system In our view bank-dominatedfinancial systems are characterized by a strong relationship-lending orientationwhich typically leads to a higher illiquidity of bank loans Furthermore the higherilliquidity of loans also affects the liquidity demand of banks which is moreheterogeneous in bank-dominated systems Therefore banks are more inclined toroll-over loans in a bank-dominated financial system Liquidity demand increasesand has a lower interest rate elasticity As a result, market interventions need to

be much larger in a bank-based financial system in order to bring down interestrates to sustainable level as compared to a market-oriented financial system, therebygenerating more windfall profits for sound banks Thus, the benefits of directdiscretionary liquidity support relative to market liquidity interventions seem to belarger in bank-based financial systems

Of course, individual liquidity assistance, bailout policy, recapitalization orclosure is more demanding for government and regulators Furthermore, it mightimply that liquidity assistance is unevenly provided to banks from different countries

in a monetary union, which implies huge political costs But unconventionalmonetary policy and the massive liquidity allotment by central banks also carrysome well-established costs To mention just a few arguments: It distorts bankinvestment incentives.1 It undermines market discipline.2 And, last but not least,

it might cause financial repression by taxing savers In fact, while the costs ofindividual support might not be worthwhile in a market-oriented financial system

in which deadweight losses of market-based support are small, in a bank-basedsystem the deadweight losses of unconventional monetary policy are large and thusindividual support more efficient However, the political costs from country specific

1 See for instance Drechsler et al ( 2016 ) and Abbassi et al ( 2016 ).

2 See for instance Fecht et al ( 2015 ).

liquidity support seems to be extraordinarily high for the ECB And that might also

be one important reason why in recent times the ECB in its crisis response more orless follows the blueprint of the FED

1.1 Related Literature

The basic consideration about an optimal design of lender of last resort policies

goes back to the principles formulated by Bagehot (1873), based on the work byThornton (1802) He suggested that in a crisis, the lender of last resort should lend

freely, at a penalty rate, on the basis of collateral that is marketable in the ordinary cause of business when there is no panic.

This doctrine apparently follows the view that interbank markets are not alwaysefficient in reallocating funds to the most illiquid banks This assumption isparticularly criticized by Goodfriend and King (1988) who argue that given today’srepo markets banks that have sufficient collateral to turn to the LOLR should also

be able to receive funding in the interbank market Thus no LOLR is needed toovercome individual liquidity shortages that result from idiosyncratic shocks Incase of an aggregate liquidity shortage only liquidity provision to the market isrequired since the most illiquid banks will be willing to pay the highest rates toreceive funding in the interbank market so Goodfriend and King’s reasoning Toensure that the most illiquid banks can sustain the aggregate liquidity shortage thecentral bank has to provide sufficient liquidity to keep money market rates at asustainable level Our model is very much in line with this argument However,

we show that to contain the spike in the money market rates to level sustainable tothe most illiquid bank the central bank has to inject liquidity that will be partiallyabsorbed by banks that could sustain the liquidity shortage Thus we show theliquidity provision to money markets as proposed by Goodfriend and King leads

to a waste of liquidity This might carry some cost as it inflates the central bank’sbalance sheet which in turn might result in an inflation tax and excessive credit risksencountered by the central bank

The vast majority of the recent literature on the design of LOLR measures,however, challenges the applicability of the Bagehot doctrine because of theimplicit assumption that solvent banks dispose of sufficient collateral As argued,for instance, by Calomiris and Kahn (1991) and Diamond and Rajan (2000) themain reason why banks run a liquidity risk is because it serves as a commitmentdevice which allows them to raise funding for opaque and thus illiquid assets Thishowever means that the collateral value of those assets is limited Thus neither therepo market nor fully collateralized liquidity provisions might provide sufficientinsurance for banks against adverse liquidity shocks Our paper builds on thisliterature and studies what implications the differences in the collateral value ofassets have for the functioning of the interbank market during an aggregate liquidityshortage Some of our results are in line with those of Diamond and Rajan (2001,

2005,2011,2012) In particular, Diamond and Rajan (2011) also show that private

and public benefits from rolling over overdue loans might deviate They argue thatimpaired banks have private incentives to hold rather than sell illiquid assets even

if they may be forced in the future to sell those assets But in their theoreticalanalysis they look at the incentives of liquid buyers to lend to these ailing banks.Diamond and Rajan (2012) analyze the optimal interest rate policy of central banksand show that central banks should raise rates in normal times above the market-determined levels to offset reduced interest rates at times of financial stress This isnecessary to preserve bank incentives to maintain low leverage and high liquiditywhich otherwise, expecting that interest rates would be reduced in adverse times,would take on more short-term leverage and make more illiquid loans But againthey do not study to what extent the structural difference in the pledgeability ofbanks’ returns in different financial systems affects the severity of crises and theoptimal policy design

Our analysis focuses on aggregate liquidity shocks and optimal measures to dealwith them If interbank markets are inefficient there might also be a role for anLOLR that contains individual liquidity shortages as pointed out by Freixas et al.(2004) For instance Rochet and Vives (2004) show in such a context that a lender

of last resort can avoid inefficient liquidation of banks Repullo (2005) investigatesthe question whether the existence of a lender of last resort really increases theincentives of banks to take risk In our model, though, the interbank market is alsonot efficient in the sense, that it channels always the liquidity to those banks that areneeding it the most Freixas et al (2004) discuss how the optimal LOLR policy isaffected by moral hazard problems on side of the banks, an aspect not considered inour analysis.3

Our view of the differences of financial systems boils down the major insightsfrom the extensive literature on comparative financial systems This literature showsthat there are many dimensions in which financial systems differ.4 It includestheoretical analysis, e.g Allen and Gale (2000a) and, with respect to corporategovernance systems, Magill et al (2015), as well as more empirically oriented worksuch as Franks and Mayer (1995), Schmidt et al (1999) and Levine (2002) Mostinterestingly, in a just published paper Langfield and Pagano (2016) provided strongempirical findings that differences in financial systems between the US and Europeare still persistent and even increasing in recent times Furthermore, they docu-mented evidence that the strongly bank-based financial structure in Europe increasessystemic risk and lowers economic growth in comparison to economies which arecharacterized by a market-based financial structure, thereby resuscitating the debate

on the relative merits of bank-based and market-based financing However, we focus

3 For a discussion of the different lender of last resort function(s), see Freixas et al ( 1999 ) We

do not want to touch the issue if there should (and could) be an institutional separation between

a central bank which is responsible for the conduct of monetary policy and a lender of last resort;

and Santos ( 2005 ).

4 See Allen and Gale ( 2004 ) for a survey.

our very simple analysis on just one aspect, namely the differences in the importance

of relationship banking, securitization and tradability of banks’ assets in oriented and bank-dominated financial systems and its impact on central bank policymeasures in adverse times Referring to the banking dimension, by conducting ameta-analysis Kysucky and Norden (2016) have shown that relationship lending

market-is more prevalent in the bank-dominated financial systems in Europe and Japan.Furthermore, with respect to central bank policy, Cour-Thimann and Winkler (2013)recently emphasized in its analysis of the ECB’s non-standard monetary policymeasures that the institutional set-up of the EMU and the mostly bank-basedfinancial structure of the euro area economy are framing the ECB’s monetary policy

2.1 The Setup

Following Diamond and Rajan (2001) we consider an economy with threedates .t D 0; 1; 2/ and a large number of entrepreneurs, bankers and investors.

Entrepreneurs are wealthless, however each of them has a project at his disposal

which requires an investment I D 1 at t D 0 Each investor is endowed with a

small amount of the consumption good in comparison to the required investmentsize, hence many investors are needed to fund a project In addition, we assumethat the aggregate endowment of all investors in the economy is lower than thetotal investment possibilities Because of this shortage of investment capital at date

0 entrepreneurs and bankers must offer an expected return as high as possible toattract funding Entrepreneurs, investors and bankers, whose role will be clarifiedbelow, are risk-neutral but differ in their preferences: Investors and bankers have astrong preference for consumption at date 1, i.e they have a very high discount rate

 for consumption at date 2, whereas entrepreneurs value consumption at each dateequally Investors can store their initial endowment earning a return of1 for everyunit invested, or they can invest it in the project

Financing the projects includes some difficulties which have to be overcome.Entrepreneurs have specific abilities vis-a-vis their projects, i.e the cash flow eachentrepreneur can generate from his project exceeds what anyone else can get out of

it But entrepreneurs cannot commit their human capital to the project, except on

a spot basis From this it follows that a lender can extract future repayment only

by threatening to take away the project from the initial entrepreneur The project

returns C generated by the initial entrepreneur are uncertain in terms of their time structure The project pays out C either at t1if the project produces early or at t2ifthe project is delayed All uncertainty about projects is resolved at date 1

We consider two alternatives when taking away the project from an entrepreneur

The project can be restructured at any time until date 1 which will yield a payoff c1immediately and nothing at date 2, or the entrepreneur can be replaced with assets