Winkler et al (eds ) a flow of funds perspective on the financial crisis; vol i, money, credit and sectoral balance sheets (2014)

Against this background, the Directorate General Economics andDirectorate General Financial Stability of the European Central BankECB jointly hosted a workshop on ‘A flow-of-funds perspec

Palgrave Studies in Economics and Banking

Series Editor: Professor Richard Werner

This series focuses on the economic implications of banking, bridging the usual divide between economics and banking/finance Titles in the series facilitate a deeper understanding of the interaction between banking and the economy, enabling readers to better understand the role and importance of banking in economic activity, and promote a better integration of banking and finance into policy models at theoretical and empirical levels.

Titles include:

A Flow-of-Funds Perspective on the Financial Crisis, Volume I: Money, Credit and Sectoral Balance Sheets

Bernhard Winkler, Ad van Riet and Peter Bull (editors)

A Flow-of-Funds Perspective on the Financial Crisis, Volume II: Macroeconomic Imbalances and Risks to Financial Stability

Bernhard Winkler, Ad van Riet and Peter Bull (editors)

Palgrave Studies in Economics and Banking

Series Standing Order ISBN: 978–1137–33135–9

(outside North America only)

You can receive future titles in this series as they are published by placing a ing order Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above.

stand-Customer Services Department, Macmillan Distribution Ltd, Houndmills, ingstoke, Hampshire RG21 6XS, England

Director General Statistics (retired), European Central Bank,

Frankfurt am Main, Germany

Editorial matter, selection and introduction

© Bernhard Winkler, Ad van Riet and Peter Bull on

behalf of the European Central Bank 2014

Foreword and remaining chapters

© Respective authors or their affiliations 2014

All rights reserved No reproduction, copy or transmission of this

publication may be made without written permission.

No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency,

Saffron House, 6–10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages.

The authors have asserted thier rights to be identified as the authors of this work

in accordance with the Copyright, Designs and Patents Act 1988.

First published 2014 by

PALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin.

A catalogue record for this book is available from the British Library.

A catalog record for this book is available from the Library of Congress.

Softcover reprint of the hardcover 1st edition 2014 978-1-137-35297-2

ISBN 978-1-349-46944-4 ISBN 978-1-137-35298-9 (eBook)

DOI 10.1057/9781137352989

Bernhard Winkler, Ad van Riet and Peter Bull

Allocation: A Contribution to Assessing Risks to Price

Roberto A De Santis, Carlo A Favero and Barbara Roffia

Claudio Borio, Robert N McCauley and Patrick McGuire

Ulrich Bindseil and Adalbert Winkler

Riccardo De Bonis, Luigi Infante and Francesco Patern`o

v

vi Contents

Part II Sectoral Analysis of the Flow of Funds

Tobias Adrian and Hyun Song Shin

Celestino Gir´on and Silvia Mongelluzzo

Sanvi Avouyi-Dovi, Vladimir Borgy, Christian Pfister, Michael

Scharnagl and Franck S´edillot

Jacob Isaksen, Paul Lassenius Kramp, Louise

Funch Sørensen and Søren Vester Sørensen

Laurent Maurin

Riccardo Bonci

portfolio investment between non-monetary financial

structural instability and four-step ahead out-of-sample

vii

viii List of Figures

12.5 Financial liabilities of euro area NFCs 277

12A.1 Response of endogenous variables to a shock to bank

12A.2 Response of endogenous variables to a shock to bank loans 29212A.3 Response of endogenous variables to a shock to

liquidity measures: MSFE computed over the period

liquidity measures: bias computed over the period

liquidity measures: variance of the forecast error

liquidity measures: MSFE computed over the period

x

10.2 System estimation on French data 249

The biggest financial crisis in a lifetime has shown how important it

is to have a deep understanding of the financial balance sheets of themain sectors of the economy and the financial flows that take placebetween them This type of information is essential for a proper under-standing of the transmission of monetary and financial shocks throughthe economy

Against this background, the Directorate General Economics andDirectorate General Financial Stability of the European Central Bank(ECB) jointly hosted a workshop on ‘A flow-of-funds perspective onthe financial crisis: lessons for macrofinancial analysis’ in Frankfurt amMain, Germany, on 28–29 November 2011 This publication of theworkshop proceedings, complemented by a few invited contributions,provides a comprehensive overview of a broad range of uses of the flow

of funds within the central bank community as well as in the academicfield The flow-of-funds perspective on the financial crisis is presented

in two volumes The first volume on ‘Money, credit and sectoral balancesheets’ focuses on the role of flow-of-funds analysis in complementingtraditional monetary analysis centred on bank balance sheets and exam-ines the portfolio and financing behaviour of non-financial sectors Thesecond, companion volume on ‘Macroeconomic imbalances and risks tofinancial stability’ explores the use of flow of funds for macrofinancialanalysis

Financial flows and sectoral balance sheets are the ‘bread and ter’ of flow-of-funds analysis They lie at the heart of the financialcrisis, while debt, default and financial intermediation have been largelyabsent from mainstream macroeconomic models As monetary policy-makers we had to navigate through uncharted territory in confrontingthe fall-out from the crisis, steering a delicate course between pre-empting disruptive disorderly deleveraging and adverse real–financialfeedback loops while buying time, without soliciting moral hazard, withrespect to necessary structural adjustments and balance sheet repair inthe financial and non-financial sectors (Praet, 2012) A particular chal-lenge for the ECB throughout the crisis has been the need to safeguardthe functioning of the monetary transmission mechanism in the euroarea This had become impaired, first, by the abrupt shocks to finan-cial intermediation in the wake of the money market freeze in August

but-xii

2007 and the Lehman Brothers insolvency in September 2008, and,second, by emerging signs of financial market fragmentation across dif-ferent parts of the euro area in the course of successive stages in the twinsovereign debt and banking crisis.

Unlike most of macroeconomic mainstream thinking, the ECB hasfrom its inception placed considerable emphasis on the analysis of bankbalance sheets, money and credit, under the monetary pillar of ourmonetary policy strategy Hence, the financial sector and financial quan-tities were squarely on our radar screen, as was the need for a centralbank to accommodate liquidity preference shocks during the financialcrisis In the wake of the financial and sovereign debt crisis, it alsobecame obvious that monetary policy transmission could not be cap-tured only by a single policy rate, as in textbook macromodels, but asoperating though a whole array of financial prices and quantities, inthe presence of so-called ‘financial frictions’ which have taken centrestage in much of the policy discussions In such uncharted territory theflow of funds provides a very useful map of financial flows and balancesheets, as well as on the structure of financial intermediation Hence,flow-of-funds analysis will continue to be an important element in ourcontinuous efforts to enhance monetary analysis in the wake of thecrisis, as also documented in Papademos and Stark (eds) (2010)

Following an introductory overview by the editors, the present ume I of the workshop proceedings is subdivided into two parts, each

vol-of which covers contributions from experts in the field Part I, tled ‘Money, Credit and Liquidity in the Flow of Funds’, illustrates howflow-of-funds analysis can be seen as a natural extension of and com-plement to monetary analysis by looking at money in the context ofthe full range of assets held by different sectors and by bringing intothe picture a number of financial intermediation channels beyond thetraditional nexus of bank loans and bank deposits This suggests taking

enti-a new look enti-at the trenti-aditionenti-al quenti-antity theory of money, including therole of banks in the creation of credit, and the international allocation

of portfolios A few chapters also address the linkages between the ability of liquidity in the economy and its implications for credit boomsand busts

avail-Part II, on ‘Sectoral Analysis of the Flow of Funds’, brings togethercontributions dealing with the balance sheet of financial intermedi-aries and bank leverage, with implications for the supply of credit tothe other sectors of the economy An understanding of the demandside of credit, in turn, requires, in particular, taking a closer look atboth sides of the household and corporate sector balance sheets The

xiv Foreword

chapters cover a number of key elements in this respect, related to folio choices, savings behaviour and debt financing of households andnon-financial corporations, partly also in conjunction with other sectors

port-of the economy within the integrated flow-port-of-funds framework.What can monetary policy-makers take away from this flow-of-fundsperspective on monetary transmission, the availability of liquidity andthe consequences for credit supply and demand? This volume highlightsthe ongoing efforts in the central bank and academic community togain a deeper understanding of the implications of the financial crisisfor monetary and financial analysis and to develop empirical tools toextract regularities from the rich flow-of-funds dataset that is alreadyavailable and is being further expanded Over time, this should enablemonetary policy-makers to draw on new insights and instruments inthe analysis of credit cycles and the transmission of monetary policy viaflow-of-funds variables This renewed recognition of the valuable role

of flow-of-funds analysis for addressing current monetary and conomic challenges in an environment of sectoral deleveraging andrebalancing, left aside by much of mainstream economics over the pastdecades, is an appropriate tribute 50 years after the early work by Tobinand Brainard (1963) on the role of financial intermediaries in monetarytransmission

macroe-Peter Praet European Central Bank

References

Praet, P (2012) Deleveraging and monetary policy, Speech at the Hyman

P Minsky Conference, Berlin, 26 November.

Papademos, L and J Stark (eds) (2010) Enhancing Monetary Analysis (Frankfurt am

Main: European Central Bank).

Tobin, J and W.C Brainard (1963) ‘Financial intermediaries and the effectiveness

of monetary controls’, American Economic Review (Papers and Proceedings),

53 (2), 383–400.

Notes on Editors

Bernhard Winkler worked at the Deutsche Bundesbank before

join-ing the European Central Bank (ECB) in July 1998 as Economist in theDirectorate-General Research Subsequently he held positions as SeniorEconomist in the Monetary Policy Strategy Division and in the Coun-sel to the Executive Board as Advisor to Prof Otmar Issing Since May

2005 he is Senior Advisor in the Directorate Monetary Policy

responsi-ble, inter alia, for flow-of-funds analysis at the ECB and the co-ordination

of financial projections as part of the quarterly macroeconomic tions exercises He has published on issues related to monetary and fiscalpolicy in a monetary union, on monetary policy communication and onthe Stability and Growth Pact as well as on cross-checking and the flow

projec-of funds

Ad van Riet joined De Nederlandsche Bank as Economist in 1987

and took up a (Senior) Economist position at the European MonetaryInstitute in 1994 He joined the European Central Bank as PrincipalEconomist when it was established in 1998 and was then in charge

of the Monetary Policy Stance Unit He became Head of the EU tries Division in 2000 and Head of the Fiscal Policies Division in 2007.Since September 2011 he is Senior Advisor in the Directorate GeneralEconomics and Secretary of the ECB Occasional Paper Series He haspublished on European money demand, fiscal policy and structuralreforms

Coun-Peter Bull joined the Bank of England in 1964 After some years in

the economics, international and foreign exchange areas, he moved tostatistics, as Head of the Statistics Department in 1987–94 In 1994 hejoined the European Monetary Institute in Frankfurt as Head of Statis-tics, and remained as Director General Statistics when the EuropeanCentral Bank was established in 1998 After retirement in autumn 2002

he has continued to work on related matters in the ECB and elsewhere.His more recent publications are in the field of national accounts andstatistics

xv

Notes on Contributors

Tobias Adrian is Vice President of the Federal Reserve Bank of New York

and Head of the Capital Markets Function of the Research and StatisticsGroup His research covers asset pricing, financial intermediation andmacroeconomics, with a focus on the aggregate implications of capitalmarket developments He has contributed to the New York Fed’s finan-cial stability policy and to its monetary policy briefings He holds a PhDfrom MIT and an MSc from LSE He has taught at MIT and PrincetonUniversity

Sanvi Avouyi-Dovi is Senior Advisor at the Microeconomic and

Struc-tural Analysis Department and was previously Head of the ResearchDivision of the Banque de France Before joining the Banque de France,

he was Head of the Department of Economic and Financial Analysis at

at Paris-Dauphine University His main research interests are in conomic modelling and applied econometrics He is a graduate of ´EcoleNationale de la Statistique et de l’Administration ´Economique (ENSAE)and University of Paris 1, Panth´eon-Sorbonne (PhD) He has pub-lished papers on monetary policy, the labour market, macroeconomicmodelling and financial econometrics

macroe-Ulrich Bindseil is Head of the ECB’s Directorate General Market

Opera-tions since May 2012, after having been Deputy Director General of thesame Directorate General since September 2009 Previously he was Head

of the ECB’s Liquidity Management Section and Deputy Head and Head

of the ECB’s Risk Management Division He had joined central banking

in 1994, in the Economics Department of the Deutsche Bundesbank,

after obtaining a PhD in Economics His publications include Monetary

Policy Implementation (2004) and Risk Management for Central Banks and Other Public Investors (co-editor, with F Gonz´alez and E Tabakis, 2009).

Riccardo Bonci graduated in Statistics at the University of Siena, Italy.

After brief periods working at the National Statistical Institute (Istat) and

at the Italian Ministry of Economics, he joined Banca d’Italia in 2001,where he was employed in the Financial Accounts Unit at the ResearchDepartment Between 2007 and 2009 he worked at the European Cen-tral Bank on the flow of funds and the estimation and analysis of debt

xvi

and wealth and of the portfolio allocation of households Back in Italy,

he has been responsible for the Regional Research Office in the Perugiabranch of the Banca d’Italia Since his return to the ECB at the end of

2012, he has dealt with flow-of-funds projections and with topical issuesregarding private sector debt and deleveraging, both at the euro area and

at the country level

Vladimir Borgy is Head of the Public Finance Division in the Business

Conditions and Macroeconomic Forecasting Directorate of the Banque

de France He was previously Senior Economist in the Financial nomics Research Division Before joining the Banque de France, heworked as an economist at the CEPII (French research centre in interna-tional economics) and in the International Studies Unit of the FrenchMinistry of Economy, Finance and Industry He holds a PhD fromthe University of Paris 1, Panth´eon-Sorbonne His research interestsinclude fiscal policy, financial macroeconomics, portfolio choice andalso demographics and ageing

Eco-Claudio Borio is Head of the Monetary and Economic Department of

the Bank for International Settlements and was previously Deputy Head

of that Department and Director of Research and Statistics At the BISsince 1987, he covered various responsibilities in the Monetary and Eco-nomic Department, including Head of the Secretariat for the Committee

on the Global Financial System and the Gold and Foreign ExchangeCommittee (now known as the Markets Committee) From 1985 to 1987

he worked as an economist at the OECD in the country studies branch ofthe Economics and Statistics Department Prior to that he was Lecturerand Research Fellow at Brasenose College, Oxford University He holds aPhD in Economics from the same university He is the author of numer-ous publications in the fields of monetary policy, banking, finance andissues related to financial stability

Riccardo De Bonis is at Banca d’Italia, in the Economics, Research and

International Relations Area, where he is Co-Deputy Director in the nomic and Financial Statistics Department Previously he worked in theMonetary Sector of the Research Department and in the Banking Pru-dential Supervision Area of Banca d’Italia His major fields of interestinclude banks, financial systems, household wealth, and economic andfinancial statistics

Eco-Roberto De Santis is Senior Economist in the Capital Markets and

Financial Structure Division of the Directorate General Economics of theEuropean Central Bank since August 2009, having been an economist

xviii Notes on Contributors

in the External Developments Division of the same Directorate Generalsince November 2000 He joined the ECB in 2000, after a three-yearworking experience as an economist in the Kiel Institute of World Eco-nomics He obtained a PhD in Economics from Warwick University in

1998 He has published widely on international economic issues, moneyand finance, central banking and economic modelling

Carlo Favero holds a PhD from Oxford University, where he was a

mem-ber of the Oxford Econometrics Research Centre In 2009 he joinedthe Department of Finance at Bocconi University, where, as Professor

of Economics, he teaches financial econometrics He has published inscholarly journals on the econometric modelling of bond and stockprices, applied econometrics, monetary policy and time-series modelsfor macroeconomics and finance He is a research fellow of CEPR in theInternational Macroeconomics programme, president of the InnocenzoGasparini Institute for Economic Research at Bocconi University and

a member of the scientific committee of the Centro InteruniversitarioItaliano di Econometria (CIDE)

Louise Funch Sørensen has an MSc in Economics from the

Univer-sity of Copenhagen She has been employed at Danmarks Nationalbanksince September 2006 (in Market Operations, Department of Eco-nomics) Her research interests include exchange rate and monetarypolicy, monetary policy instruments and market operations, finan-cial market surveillance, IMF-related issues and panel co-integrationanalysis

Celestino Gir ´ on is employed at the European Central Bank, in the

Division of Macroeconomic Statistics He has developed his career, asboth compiler and analyst, in the field of flow of funds and national

at the ECB since 1999 His current areas of interest include leveragedynamics and interaction across agents He holds a Licenciatura (BA andMA) in Economics and Business Administration from the University ofAlicante

Luigi Infante graduated in Economics from the University of Foggia

and received an MSc in Economics from the University Pompeu Fabra.Since 1994, he has been working for Banca d’Italia, first in the Eco-nomic Research Department (Balance of Payments Office), then in theBank’s Regional Economic Unit in Milan, and currently in the Eco-nomic Research and International Relations Area (Financial Accounts

Sector) He has published on different topics, including internationaltrade, migration and global imbalances.

Jacob Isaksen has an MSc in Economics from the University of Aarhus.

He has been employed at the Department of Economics, DanmarksNationalbank since August 2011 His research interests include flow

of funds, international economics, cross-country research and appliedpanel co-integration analysis

Paul Lassenius Kramp holds a PhD in Economics from the University

of Copenhagen He has been employed at Danmarks Nationalbank sinceDecember 2002 (Statistics Department, Financial Markets, Department

of Economics) His research interests include flow of funds, aggregateconsumption, debt and interaction between the financial sector and thereal economy

Laurent Maurin is Senior Economist in the Capital Markets and

Finan-cial Structure Division of the Directorate General Economics of theEuropean Central Bank He is a former student of ´Ecole NormaleSup´erieure de Cachan and holds a PhD in Economics from the Univer-sit´e de la M´editerran´ee in Marseille After having worked for two years

at the Banque de France, he joined the ECB in 2003 on issues related

to monetary analysis He has worked since on forecasting the euro areamacroeconomy and the analysis of the banking sector

Robert N McCauley is Senior Adviser in the Monetary and Economic

Department of the BIS Before that, he served as Chief Representativefor Asia and the Pacific of the Bank for International Settlements in2005–08, after joining the BIS Asian Office in 1998 Prior to that, heworked for 13 years at the Federal Reserve Bank of New York, serving

at times as Chief Economist for the inter-agency committee of banksupervisors that rates country risk There he wrote on international com-parisons of the cost of capital, foreign bank lending to US corporationsand the unprofitability of foreign direct investment in the US In 1988,

he worked for the Joint Economic Committee of the US Congress In

1992 he taught international finance and the multinational firm at theUniversity of Chicago’s Graduate School of Business He serves on theCouncil of Management of SUERF, the European Money and FinanceForum

Patrick McGuire is Head of the International Data Hub and formerly

Senior Economist in the Financial Institutions Section of the Bank forInternational Settlements Prior to joining the BIS in 2002, he completed

xx Notes on Contributors

his PhD in Economics at the University of Michigan with a dissertation

on the Japanese financial system He studied for several years in Japanand was a visiting scholar at METI and the Bank of Japan His currentresearch is focused on international financial markets and financial sta-bility issues, with a specific interest in international banking, emergingmarket financing and hedge funds

Silvia Mongelluzzo is Consultant at KPMG Advisory, Italy She obtained

her PhD in Statistics from Bocconi University in Milan Her researchfocuses on the application of Bayesian statistics to a variety of fields,ranging from economics to biostatistics Since September 2012, she hasbeen part of the team of financial risk management at KPMG Advisory,focusing on the development of statistical models within the Opera-tional Risk area for some of the largest Italian banking groups Previousworking experience includes the analysis of Euro Area Accounts, sen-sitivity analysis for hierarchical non-linear models on different priorspecifications at Novartis in the Department of Modelling & Simulation(Basel), and frailty models (survival analysis) at the Max Planck Institute(Rostock)

Francesco Patern ` o studied at the London School of Economics, where

he obtained an MSc in Economics in 1995, and at the Kiel Institute forWorld Economics in 1996 From 1996 to 2000 he worked as a researcher

at Confindustria Research Office He joined Banca d’Italia in 2000, ing initially in the Economic Research Department, International Sector,and then from July 2007 in the International Economic Analysis andRelations Department He is a member of the G20 International Finan-cial Architecture Working Group and of the ECB International RelationsCommittee’s Task Force on IMF issues He has published articles on theNew Economy, global imbalances, and the macroeconomic impact ofworkers’ remittances

work-Christian Pfister is Deputy Director General for Statistics at the Banque

de France He was previously Director for Economic Analysis andResearch and then Deputy Director General for Economics and Interna-tional Relations of the Banque de France, which he joined in 1980 Heteaches at the ´Ecole Nationale de la Statistique et de l’Administration

´Economique (ENSAE) He graduated from ´Ecole des Hautes ´Etudes merciales (HEC) in 1976 and from Institut d’ ´Etudes Politiques de Paris(Sciences Po) in 1978 He has published on monetary policy, monetaryunification, financial stability, labour market, structural reforms andinternational economics

Com-Barbara Roffia is Secretary of the Monetary Policy Committee of the

European System of Central Banks and also works as a Senior Economist

in the Directorate Monetary Policy of the European Central Bank Sheholds a PhD in Economics from the University of Warwick She has pub-lished a number of articles related to monetary policy and internationalmacroeconomics

Carmelo Salleo is Advisor in the Secretariat of the European Systemic

Risk Board (ESRB) He is the Secretary of its Advisory Scientific tee and coordinates analytical work in the fields of financial stabilityand macroprudential policy He was formerly Economist and Head ofUnit in the Research Department of Banca d’Italia His research interestscover financial stability, bank funding, financial constraints to invest-ment and bank mergers He holds a PhD in Economics from HarvardUniversity

Commit-Michael Scharnagl is an economist at Deutsche Bundesbank His main

research interests are monetary policy (money demand, monetary icy rules, identification of loan supply shocks) and time series analysis(Bayesian VARs, Bayesian Model Averaging) as well as wavelet analysis

pol-Franck S´edillot is Head of the Financial Accounts Division in the

Monetary and Financial Directorate of the Banque de France Hewas previously seconded to the European Central Bank (DirectorateGeneral Economics) and the OECD (Economic Department) In bothorganisations, he developed short-run econometric models to assess eco-nomic activity He has published articles on world models and scenarioanalysis

Hyun Song Shin is Hughes-Rogers Professor of Economics at Princeton

University and has been appointed as of May 2014 as Economic sor and Head of Research at the Bank for International Settlements Hisresearch interests cover financial institutions, risk and financial stabilityissues Before moving to Princeton in 2006, he was based in the UnitedKingdom, holding academic positions in Oxford and the London School

Advi-of Economics In 2010, he was on leave from Princeton, serving in a icy role in Korea as Senior Advisor to President Lee Myung-bak He is afellow of the Econometric Society and of the British Academy

pol-Søren Vester pol-Sørensen holds a PhD in Economics from the University

of Aarhus He has been employed at Danmarks Nationalbank since April

2007 (Department of Economics) His research interests include flow offunds, international economics and macrofinancial linkages

xxii Notes on Contributors

Richard Werner is Professor in International Banking at the University

of Southampton Management School in England, and is Director of theUniversity of Southampton Centre for Banking, Finance and SustainableDevelopment He is also Chair of LocalFirst Community Interest Com-pany and a member of the ECB Shadow Council In the past he wasalso Visiting Professor of Macroeconomics and Monetary Economics atHouse of Finance, Goethe-University Frankfurt; Assistant Professor ofEconomics at Sophia University, Tokyo; Senior Managing Director andSenior Portfolio Manager at Bear Stearns Asset Management Ltd.; ChiefEconomist at Jardine Fleming Securities (Asia) Ltd.; Visiting Researcher

at the Bank of Japan; Visiting Scholar at the Japanese Ministry of Financeand Senior Consultant to the Asian Development Bank He obtained aDPhil in Economics from the University of Oxford Major works include

New Paradigm in Macroeconomics (2005) and Princes of the Yen (2003) He

is Series Editor of Palgrave Studies in Economics and Banking.

Adalbert Winkler is Professor for International and Development

Finance at the Frankfurt School of Finance & Management Before ing the Frankfurt School, he pursued a career in development financeand central banking, serving in the European Central Bank’s DirectorateGeneral International and European Relations and in the InternationalDepartment of Deutsche Bundesbank He holds a PhD from TrierUniversity and a post-doc (‘Habilitation’) from the Bayerische Julius-

publi-cations focus on monetary policy, the global monetary and financialsystem, as well as development finance and microfinance

Bernhard Winkler, Ad van Riet and Peter Bull

1.1 Introduction

Flow-of-funds accounts are a component of the national accounts tem reporting the financial transactions and balance sheets of theeconomy, classified by sectors and financial instruments As described

sys-by Winkler (2010), the financial accounts track funds as they move fromsectors, such as households, that serve as sources of funds (net lenders),through intermediaries (financial corporations) or financial markets, tosectors that use the funds to acquire physical and financial assets (non-financial corporations, government, rest of the world) These flows,together with valuation changes, result in changes to sectoral (net) assetpositions and the composition of the corresponding balance sheets.The financial crisis has driven home the importance of financial flowsand balance sheets for an understanding of real–financial linkages and

it has spurred a renewed academic and policy interest in flow-of-fundsanalysis During the crisis policy-makers could rely neither on receivedwisdom and assumptions on liquid and efficient markets underlying thefunctioning of the financial system, nor on standard macroeconomicworkhorse models, to give ready answers on the origins, transmissionchannels and policy implications of the financial crisis In such circum-stances flow-of-funds data could be seen, at least, to provide a promisingframework to articulate relevant questions to be asked when confrontingnew challenges for monetary policy and financial stability, such asrelated to debt and asset market dynamics, leverage cycles, financial

should not be reported as representing the views of the ECB The views expressed are those of the authors and do not necessarily reflect those of the ECB.

1

flow-of-In this respect, the flow of funds provides a nexus between the ‘flow’dynamics of money, credit and other financial intermediation flowsand the implications for ‘stock’ dynamics in terms of sectoral balancesheets and the evolution of assets and liabilities On this basis one can,for example, construct early warning indicators for financial boom-bustcycles In particular, private and public sector debt indicators based

on financial accounts data have become an important element in theenhanced surveillance of macroeconomic imbalances (in both the EUand the G20 context) Moreover, flow-of-funds approaches can be usedfor macroprudential risk analysis

Central banks have traditionally taken a close interest in the working

of the financial system and have for a long time invested in compilingfinancial accounts, most notably at the US Federal Reserve, but also atthe Bank of Japan and at many European national central banks For acomprehensive compilation of key academic papers and applications seeDawson (ed.) (1996) The set of studies included in De Bonis and Pozzolo(eds) (2012) is also highly recommended Flow-of-funds analysis for theeuro area is a relatively recent endeavour For the European Central Bank

it offers a natural platform for cross-checking and ‘bridging’ analysisunder the economic and monetary ‘pillars’ that are a key feature of itsmonetary policy strategy (see Winkler, 2010)

The remainder of this introduction and overview summarises the tributions to the workshop proceedings collected in the present volume,sub-divided into two thematic parts, each covering a specific field ofinterest

con-1.2 Part I: Money, credit and liquidity in the

flow of funds

While mainstream macroeconomic models had got used to largelyignoring financial developments, the growing economic importanceand complexity of financial sectors and markets has spurred a renewedacademic and policy interest in flow-of-funds analysis The contribu-tions in Part I of this volume, entitled ‘Money, credit and liquidity in the

flow of funds’, discuss in more detail the role of flow-of-funds analysis

as a natural extension of and complement to monetary analysis

By way of introduction to this theme, Carmelo Salleo (European

Sys-temic Risk Board) recalls that the original version of the quantity theory

terms of total transactions in the economy (T) This comprised both realand financial transactions (as well as intermediary transactions alongthe production chain) Multiplied by the price (P), the nominal value

of all transactions must be equal to the quantity of money (M) adjustedfor the number of times it is used, i.e the velocity of circulation (V).Only subsequently was the transactions variable replaced (or approxi-mated) by a measure of income (Y) for practical reasons As the stability

of this relationship between money and economic activity broke down,

a search started for new interpretations for its components tively, the equation of exchange may be rearranged in terms of the threemotives for holding money in varying economic and financial condi-tions: to finance transactions, as a store of value, for portfolio purposes.This could prompt searching for a ‘financial transactions-augmented’model of money demand drawing on flow-of-funds statistics of thestocks and flows in the financial system While a revisited quantity the-ory of money still offers a solid basis for understanding the dynamics

Alterna-of money, Salleo remains somewhat sceptical on the ability Alterna-of centralbanks to use the quantity relation for monetary policy purposes Instead,

an extended monetary analysis framework is argued to play a potentiallymore useful role with respect to financial stability considerations

Richard A Werner (University of Southampton) presents a simple

macroeconomic model that incorporates the special role of banks ascreators of credit, the ‘Quantity Theory of Credit’, which can also beseen as a parsimonious flows-of-funds model He stresses the importance

of distinguishing between ‘good’ credit creation for income-generatingtransactions that contribute to GDP and ‘bad’ credit creation for financ-ing transactions that may generate unsustainable capital gains that donot contribute to GDP, and how this is linked to the separation of eco-nomic activity into income accounts and financial accounts He thenshows how the Quantity Theory of Credit solves ten seeming ‘puzzles’(such as the recurrence of banking crises and the ineffectiveness of fiscalpolicy in promoting GDP) that traditional macroeconomic or monetarymodels have struggled with, and discusses a number of policy implica-tions that are relevant today Given the importance of detailed data onthe use of bank credit, he calls on central banks to collect and makeavailable such data in a far more detailed and timely fashion than is

4 Winkler et al.

currently the case Furthermore, he urges central banks to publish thewealth of data they have concerning total transactions in the economy– thanks to their role as settlement system of bank flows – on a real-time,daily basis

Roberto A De Santis (European Central Bank), Carlo A Favero

(Boc-coni University) and Barbara Roffia (European Central Bank) characterise

money demand as part of a broader portfolio allocation problem whereboth domestic and foreign asset prices influence money holdings Bymodelling international portfolio shifts they are able to obtain a sta-ble broad money demand for the euro area over the period 1980–2011.This implies that fluctuations in international financial markets areamong the key determinants of the observed path of euro area moneygrowth The authors conclude that model-based excess liquidity mea-sures, namely the difference between actual M3 growth (net of theinflation objective) and the expected money demand trend dynamics,can be useful to predict inflation

Claudio Borio, Robert McCauley and Patrick McGuire (all Bank for

Inter-national Settlements) explore insights from combining the BIS tional financial statistics with national flow-of-funds data to produceindicators of global liquidity, with a focus on global credit aggregates.Their aim is to better understand the international dimension of creditalong two dimensions: foreign currency credit to residents, regardless

interna-of the lender’s location; and cross-border (external) credit, regardless interna-ofthe currency of denomination This is badly needed, given that finan-cial globalisation and the use of international currencies outside theircountry of origin mean that the monetary authorities of these currencieshave a direct influence on financial conditions in other jurisdictions Bycontrast, the countries whose residents denominate a significant frac-tion of their debt (and assets) in these foreign currencies are constrained

in their room for policy manoeuvre Moreover, cross-border credit has

a history of outpacing the growth of overall credit in economies encing credit booms, and therefore could raise concerns from nationalsupervisors The authors show that an increasing share of outstand-ing credit and a significant part of the recent boom-bust cycle wasdriven by external sources of credit, especially for US dollar lendingand, to a lesser degree, for lending in euro They note that moni-toring direct cross-border credit, which is not channelled through thedomestic banking system, presents challenges The BIS data could beused for cross-checking the authorities’ estimates of residents’ inter-national debt positions, especially the part owed by firms to banksabroad

experi-Ulrich Bindseil (European Central Bank) and Adalbert Winkler

(Frank-furt School of Finance & Management) address the role of the centralbank as lender of last resort in financial crises within a closed system offinancial accounts They compare the ability of central banks to respond

to a dual liquidity crisis (a confidence crisis involving both the ment and the banking sector) under the gold standard and within theframework of a monetary union The system of financial accounts offers

govern-a frgovern-amework to trgovern-ace liquidity flows, identify qugovern-antitgovern-ative constrgovern-aintsand relevant policy options and, hence, allows conclusions to be drawn

on the ability of central banks to absorb shocks under alternative etary regimes The authors find that a central bank in a monetary union

mon-is much more able to respond to liquidity shocks than a central bankunder the gold standard Their analysis also suggests that a sufficientlystrong underpinning of the monetary union in terms of banking and fis-cal union is needed in order to be able to deal with solvency issues thatmight arise when fighting a liquidity crisis When this ability is ensuredand the integrity of monetary union is beyond doubt, the common cen-tral bank is as unconstrained in providing liquidity as a central bank of

a nation state issuing a currency under a flexible exchange rate regime

Riccardo De Bonis, Luigi Infante and Francesco Patern`o (all Banca d’Italia)

analyse linkages between financial and real variables for the UnitedStates and the euro area They find that a deterioration in bank cap-ital leads to tighter lending standards applied by banks to firms andhouseholds (as measured by bank lending surveys) In turn, tighterlending standards reduce credit, which finally affects different categories

of spending (machinery and residential investment and consumption).Based on this three-step estimation, the authors quantify the first roundimpact of bank capital losses on GDP A 1 per cent bank capital losscauses three years after the shock a GDP loss of about 0.5 per cent in theUnited States, and of about 0.2 per cent in the euro area They explainthe larger effect recorded for the United States by the greater influence

of bank capital losses on consumer loans and consumption in the US

1.3 Part II: Sectoral analysis of the flow of funds

The second part of this volume, entitled ‘Sectoral analysis of the flow

of funds’, collects a number of contributions which study the folio and/or financing behaviour of individual institutional sectors,covering both financial intermediaries and non-financial private sectors(households and non-financial corporations)

port-Tobias Adrian (Federal Reserve Bank of New York) and Hyun Song Shin (Princeton University) examine the balance sheet management

6 Winkler et al.

by financial intermediaries, which they contrast with conventionaldiscussions of the balance sheet management by non-financial firms.Concerning the latter, the size of the assets of the firm is determined

by taking the set of positive net present value projects as given Thecomposition of equity and debt is then the key factor in funding suchassets By contrast, the balance sheet management of financial inter-mediaries reveals that equity behaves like the predetermined variable,while the asset size of the financial intermediary is determined by thedegree of leverage that is permitted by market conditions The fact thatequity is relatively sticky suggests that there are possible non-pecuniarybenefits to bank owners As a result, bank owners are reluctant to raisenew equity, even during boom periods The authors explore the empir-ical evidence for market-based financial intermediaries such as the WallStreet investment banks, as well as the commercial bank subsidiaries

of the large US bank holding companies They also explore the gate consequences of such behaviour by the banking sector for thepropagation of the financial cycle and securitisation

aggre-Celestino Gir´on (European Central Bank) and Silvia Mongelluzzo

(Boc-coni University) focus on the behaviour of bank leverage and bankbalance sheet growth in the euro area financial sector as derived fromflow-of-funds data, taking advantage of the fact that the system requiresmarked-to-market valuation of assets, liabilities and equity Studying theperiod 1999–2011, they find that, before the financial crisis of autumn

2008, the bank leverage ratio behaved in a strongly procyclical ner for most of the euro area countries in their panel Afterwards, theleverage ratio was less procyclical, signalling a precautionary reaction onthe part of banks This evidence suggests that strong credit and balancesheet growth are accompanied by an insufficient build-up of precau-tionary capital buffers, while severe downturns in the credit cycle arelinked to a fast accumulation of capital The authors note that this pro-cyclical behaviour of the bank leverage ratio might contribute to theamplification of the credit cycle

man-Sanvi Avouyi-Dovi, Vladimir Borgy, Christian Pfister, Franck S´edillot

(all Banque de France) and Michael Scharnagl (Deutsche Bundesbank)

present a detailed empirical analysis of households’ financial portfoliostructure during the period 1978–2009 in France and Germany In theGerman case, the main data source is the newly compiled quarterly flow-of-funds dataset for households according to ESA 1995 that was built inthe Bundesbank Concerning the French portfolio data, the main dataare taken from the quarterly financial accounts collected and published

by the Banque de France The aggregate financial portfolio structures of

households have evolved dramatically in France and Germany over theperiod 1978–2009; however, at the end of the sample, both aggregatehousehold portfolios display somewhat similar structures in the twocountries The authors also study the portfolio choices of German andFrench households, by estimating a Financial Almost Ideal Demand Sys-tem (FAIDS) model The analysis of cross-interest rate elasticities allowsthem to assess substitution effects between asset shares in the aggregatefinancial portfolios of households.

Jacob Isaksen, Paul Lassenius Kramp, Louise Funch Sørensen and Søren Vester Sørensen (all Danmarks Nationalbank) undertake a cross-country

comparison of household debt and balance sheets for a large sample ofOECD economies Their study decomposes changes in net wealth due tosavings and capital gains, linking the latter to determinants like finan-cial liberalisation and private pension wealth Regression of householddebt ratios also highlights the interaction between debt and assets (hous-ing or pension wealth) as well as with other sectors (public net assets)together with macro variables like the short-term real interest rate, infla-tion or NAIRU They find that higher gross household debt ratios lead

to a greater sensitivity of households to changes in interest rates andunemployment, as well as a greater volatility in private consumption.This may be accompanied by increased losses on bank lending to firmsand risks to financial stability

Turning to analysis of the non-financial corporate sector, Laurent

Maurin (European Central Bank) presents preliminary results on the

links between investment and different sources of external financing,the role of internally generated funds and substitution between bankand market-based funding in the face of credit supply restrictions Anincreased role for market funding could reflect a response to constrainedaccess to bank finance Using a VAR framework estimated on euro areaquarterly data for 1992–2011, his results point to a positive response ofinvestment to loan shocks, debt shocks and shocks to internal financing.The additional finding of a negative correlation between loan growthand debt issuance in 2009–10 and again in 2011 may be due to spe-cific events in debt markets Maurin cautions that the relationshipfound may differ for large and small non-financial firms and may there-fore be dependent on the production structure of individual euro areacountries

Riccardo Bonci (Banca d’Italia) extends a VAR model to flow-of-funds

variables to present new evidence on the impact of a change in tary policy on the borrowing and lending decisions of euro area institu-tional sectors His results show that a monetary policy tightening leads

mone-8 Winkler et al.

households to increase precautionary savings and reduce net borrowing

in the short term Bank loans to firms decline, partly compensated byrecourse to inter-company loans and by drawing down liquidity buffers

to offset a decline in revenues Hence, the interest rate hike is associatedwith a significant fall in bank loans to the private sector, especially ofshort-term maturity Non-bank loans provide some offsetting increase.Net borrowing by the public sector increases, as following the economicslowdown the government faces the costs of automatic stabilisers andfalling tax receipts

As stressed in many contributions to this publication, the financial sis has underlined the usefulness of flow of funds for macrofinancialanalysis and financial stability issues The flow of funds supports ourunderstanding of the origins and the successive evolution of the finan-cial crisis globally as well as in the euro area in a number of dimensions.This relates, in particular, to the need to:

interconnectedness of private, government and financial sectors;

interest rates, when assessing financial conditions;

in the context of assessing deleveraging needs and balance sheetrepair from the perspective of stock-flow adjustment;

debt is another sector’s asset;

understand the dynamics of real–financial linkages

We hope and expect that the contributions in this book, and those inthe companion volume, will stimulate additional analysis and research

to further deepen our understanding of stock-flow adjustments from across-sectoral perspective as well as on the role of the financial systemand its interaction with the real economy

References

Dawson, J (ed.) (1996) Flow of Funds Analysis – A Handbook for Practitioners

(Armonk, NY and London: M.E Sharpe).

De Bonis, R and A.F Pozzolo (eds) (2012) The Financial System of Industrial Countries – Evidence from Financial Accounts (Heidelberg: Springer).

European Central Bank (2011) ‘The financial crisis in the light of the euro area

accounts: a flow-of-funds perspective’, Monthly Bulletin, October, 99–120.

European Central Bank (2012) ‘Comparing the recent financial crisis in the United States and the euro area with the experience of Japan in the 1990s’,

Monthly Bulletin, May, 95–112.

Winkler, B (2010) ‘Cross-checking and the flow of funds’ in L Papademos and J.

Stark (eds), Enhancing Monetary Analysis (Frankfurt am Main: European Central

Bank), 355–80.

Part I

Money, Credit and Liquidity in the Flow of Funds

go directly to the papers quoted for the original, detailed arguments.

2.2 A stable relation between money and transactions

One of the cornerstones of economics is what is known as the QuantityTheory of Money Its original formulation is due to Irving Fisher (Fisherand Brown, 1911):

Where: M: the quantity of money; V: the velocity of circulation ofmoney; P: prices; T: quantities transacted

In other words, the nominal value of the sum of all transactions must

be equal to the quantity of money multiplied by the number of times

it is used With a few reasonable assumptions, this accounting identitybecomes a very powerful tool to think about monetary policy (togetherwith its statistical counterpart, the flow-of-funds matrix)

necessarily reflect those of the European Systemic Risk Board or of its Secretariat.

13

14 Carmelo Salleo

The first assumption is that quantities transacted are proportional toGDP (originally this assumption was actually motivated by a scarcity ofdata on transactions in the economy) Then PT becomes proportional

to nominal GDP

The second assumption is that the velocity of circulation is anexogenous, slow-moving technological parameter determined by pay-ment systems technology, regulation and deep habits of consumers(Friedman, 1968)

With these two assumptions, the equation becomes:

The quantity of money is proportional to GDP (Y) and the price level, ornominal GDP Therefore, if the goal of the institution that influences Mdirectly or indirectly (the central bank) is to keep P stable, the equationoffers a simple policy rule: track changes in nominal GDP, and, based

transactions taking place in the economy and no more

2.3 Revisiting the quantity theory

It is clear that the key measure is a stable relationship between moneyand economic activity And this relationship has been relatively stableacross a number of countries for a long time However, this relationshipstarted breaking down in the 1970s, first in the US (Sargent and Surico,2011), then in Europe

In particular, what has been happening is that money has grownmuch faster than what would be warranted by nominal GDP, which hasbeen growing slowly, or by inflation, which has been consistently low.This could be explained simply by a slowdown of velocity, but, intu-itively, technological and financial innovation and deregulation shouldhave fostered a more efficient use of a scarce resource, rather than a lessefficient (slower) use

Given that the quantity theory starts with what in essence is very close

to an accounting identity, it couldn’t be simply discarded Research hasfocused on finding new interpretations for its components:

substitutes, could be changing over time If ‘money’ is a larger gate and we observed only the growth of one component, we weresimply victims of a fallacy of observation This larger aggregate could

aggre-be due to the increase of financial instruments with cash-equivalent

functions (Poszar, 2011), or to financial integration, due to which oneshould look at monetary and real aggregates of groups of integratedcountries (De Santis, 2012).

assume that velocity could be different for different types of etary assets’, or for different sectors of the economy, such as firmsversus households (Bricongne, 2011), we might find that the rela-tionship is still relatively stable at the disaggregated level and thatfluctuations are due to the changing weights of the building blocks

‘mon-of the economy Alternatively, velocity could vary according to action types, for example real versus financial (Werner, 2011); again,changes in the mix of transactions could explain the instability ofthe aggregate relationship between money and nominal GDP

the nominal value of transactions is too strong Transactions includereal economy transactions and financial transactions As long as thesize of the financial sector was relatively stable in proportion to thereal economy, or grew slowly and predictably, the assumption wasworkable, but the explosion of the financial sector of the past decadesand its increasing opacity (the ‘shadow banking system’), could haveweakened the assumption from both a fundamental and a statisticalreporting perspective Including measures of (real and financial) assettransactions, subject to data availability, could restore the originaltheory and possibly lead to a more stable relationship (Werner, 2011).But should money really be proportional to anything? We know fromthe fundamentals of monetary theory that money is held mainly for

a combination of three purposes: to finance transactions, as a store ofvalue, for portfolio motives The (partial) explanations of the breakdown

of the quantity theory of money illustrated above, which looked at theelements of the quantity theory, can be rearranged in terms of changes

in the motives for the demand for money stated above:

be relatively stable at a disaggregated level, in terms of instruments

or sectors, and changes are due to changes in relative weights (seeCollins and Edwards, 1994 for a discussion of how M2 augmentedwith bond and equity mutual funds fits a traditional money demandmodel) Alternatively, the demand for money could be proportional

to economic activity, as in the traditional version, but we would need

to recognise that, thanks to financial liberalisation and integration,

16 Carmelo Salleo

we should not stop at the national level but look for broader gates that allow us to consider cross-border portfolio shifts (De Santis,2012) Or, transactions should also include assets (Werner, 2011) Inall three cases, the demand for money is driven by the need to financetransactions: what is needed is a ‘better’ definition of the relevanttransactions

increasingly large gross financial flows and become more cated, they face growingly complex choices for the intertemporalallocation of resources The understanding of real options meansthat these agents have an increasing demand for an instrument thatallows them to time their choices optimally This instrument must

sophisti-be a store of value, as the optimal choice depends on the knowledgethat the value of these resources is exactly predictable for the uncer-tain moment in which it will be mobilised This instrument is money,

or ‘quasi-money’, and there is a clear trend for financial institutions,firms, governments to hold ‘pools of cash’, in order to mobilise theseresources when needed at short notice and without fluctuations intheir value (see Gorton and Metrick, 2010 for a discussion of the role

of shadow banking as a source of money-equivalent instruments)

Keyne-sian liquidity preference motive): this motive might actually havecaused a decline in the demand for money in industrialised countries,since the smoothing of the business cycle (the ‘Great Moderation’)and the greater availability of instruments for risk sharing wouldimply less need for money (lower returns from holding money, ifone takes a broader view of the return on money – zero volatility andzero nominal return but still providing utility as a store of value).However, if one takes a broad view of the financial system, consider-ing that globalisation has in fact integrated most economies (in thespirit of De Santis, 2012), then the picture is quite different First,

we must acknowledge that global savings are quite high (to the pointthat the ‘savings glut’ hypothesis was put forward to explain the lowlevel of long-term interest rates) Second, the relatively low yields

of fixed-income securities prevailing before the crisis, a side effect ofthe Great Moderation, have made money a relatively more attractiveasset, as the opportunity cost of holding it decreased Third, there hasbeen a shift among large savers towards more risk aversion: emerg-ing markets’ central banks and sovereign wealth funds are more riskaverse than financial institutions and institutional investors of indus-trialised countries So the combination of the increase in savings to

be invested on a global scale, the decrease in the returns gap betweenmoney and alternative comparable instruments and the increase ofrisk aversion could have all contributed to an increase in the demandfor money The question would be to what extent this demand hasbeen accommodated by an increase in the supply.

So, if we take into account the changing definitions of instrumentsthat meet the criteria of ‘moneyness’, the dynamics of the differentdemand motives for different segments of the economy, and the effects

of financial integration, we can conclude the following: if we had a morecomprehensive model that takes all these factors into account then wemight well have again a stable relationship between money and nom-inal GDP, all else being equal That is, if we could also overcome dataissues related to all these factors

The asset transactions explanation in particular seems to fit thestylised facts of the past two decades In its simple form (Werner, 2011),

it posits that credit creation by banks, which substitutes more tive definitions of money, finances more and more asset transactions,driving up asset prices Therefore, the spirit of the quantity theory ofmoney is still well and alive, although transposed to a wider concept:excess money creation spurs credit growth and creates asset inflation.And, since credit is procyclical, we have booms and busts Also, with thisquantity theory the policy implications are clear: if banks were made toallocate credit mainly to financing productive investment, there would

restric-be less intermediation, fewer asset bubbles and more growth

2.4 The role of the financial sector

This view of a revisited quantity theory of money is intuitively ing, but can be extended further Asset bubbles can’t be the onlyexplanation for the ‘abnormal’ growth of money and credit, especiallysince there would be an asymmetry: with the bust the quantity of liq-uidity hasn’t decreased, but has, rather, increased This can be explained

appeal-by a simultaneous increase in the store of value/precautionary motive,but maybe a further exploration of the asset transactions motive couldalso help

A fundamental fact that we learn, for example, from flow-of-fundsstatistics, which are the key instrument to understand the national andinternational dynamics of financial flows and stocks, is that over thepast two decades the size of the financial sector increased dramatically

18 Carmelo Salleo

in industrialised countries (Winkler, 2010) This is, coincidentally, whenthe traditional quantity theory of money was breaking down

Part of the growth is due to the increased use of financial instruments

by the non-financial sector, for example, credit card debt and gages for households But a significant part of this growth comes fromtransactions within the financial sector

mort-The increase of transactions within the financial sector is due to manycauses First of all, changes in business models and an increased empha-sis on focus and efficiency, due also to the increase in competitivepressure that followed widespread deregulation and liberalisation, havemeant that the chain of transactions has increased What was a simplecircuit from central bank to commercial bank to customer has become

an increasingly complex web of transactions, each with its own andcounterparty risk (for a description of the interplay of financial interme-diaries in the euro area, see ECB, 2012) Individual financial institutionshave also become more adept at dynamically adjusting their risk profilesthanks to increasingly sophisticated risk management models (althoughthe sub-prime crisis exposed the many weaknesses of such models).And innovation and deregulation have contributed to shortening thetime horizon of profit opportunities: we have gone from the extreme oforiginating and holding mortgages with maturities of over 30 years tohigh-frequency trading within (fractions of) seconds

All this means that the number and nominal value of financial actions have literally exploded over the past two decades, within andoutside regulated or monitored sectors So, even though we don’t havethe full picture, it is commonly accepted that the financial sector hasgrown massively, especially with intra-sector trades

trans-Does this affect the demand for money? Just as asset transactions aresettled with money that is redeposited into the system and so on, with

a multiplier that contributes to determining the demand for money,transactions within the financial sector also ‘consume’ money (and col-lateral – for an exploration of the link between the two see Gorton andMetrick, 2012) and therefore they should also impact on the demandfor money

The effect of all these intra-sector transactions on the demand formoney is difficult to measure, also because of interlinkages betweenmoney-creating institutions such as banks and partly money-holdingones such as other financial intermediaries Measurement problems arepartly due to data issues, since we don’t have (yet) a comprehensivepicture of the whole web of transactions But they are also due to theinherent instability of a set of transactions that depends on financial

innovation, competition, regulation and risk aversion These are allfactors that influence the more traditional channels, but to a height-ened degree: these intra-sector transactions are only loosely related to

an end-use and can therefore be compacted again into one, as in the

old intermediation model, or, vice versa, further extended into a longer

chain of intermediation, depending on circumstances

2.5 Policy implications

One policy implication is, of course, that we need more and better data.The Office for Financial Research in the US is working in this direction,with a project to map all transactions with at least one US counterparty.Such a database, especially if extended globally, would allow centralbanks to trace the path of liquidity in all the crevices of the financialsystem and let them explore its determinants Aggregating these datawithin a framework such as the flow of funds would give researcherssome very useful information about changing patterns of the finan-cial system For the moment we can further develop the flow-of-fundsapproach, which allows us to look into intra-financial sector transac-tions, albeit aggregated by sub-sectors This way we can keep track ofliquidity even as the distinction between banks and market-based vehi-cles becomes more blurred, and we can see how credit underwritten

by some sectors becomes debt in others and monitor the build-up ofmacrofinancial imbalances

However, one could draw also a more pessimistic policy implication.While a financial transactions-augmented model of money demandwould probably have a higher explanatory power than a traditional one,

it is doubtful whether it would be stable enough to become a useful icy tool, except perhaps for short-term purposes The size and shape

pol-of the financial sector depends on so many variables that interact incomplex ways and at different speeds that reduced-form models wouldbecome unreliable, and more structural models too complex

So maybe the bottom line is that the quantity theory of money is still

a solid intellectual framework to think about monetary economics, and

a good starting point to understand the general dynamics of money But

as a simple direct policy tool its best days might be already behind us(which, by the way, is no news to central bankers ) Money remainscentral to monetary policy, but its definition and the interpretation ofits role should be recast in the broader flow-of-fund analysis of devel-opments in financial intermediation In fact, new theories are already