Macroeconomic Policy Since the Financial Crisis

By Matteo Iannizzotto

Economic policymakers use various macroeconomic models, but how reliable are they in real-world conditions? Starting from the premise that all models are wrong, but some are useful, Matteo Iannizzotto introduces and explains the workings of the key economic models available for policymaking. He shows that the inconsistencies and contradictions evident in the real world require the economist to make choices about which models to adopt in certain circumstances and when not to rigidly adhere to a single approach.

The book uses a clear and critical step by step analysis to consider the strengths and weaknesses of each model, in a way that enables students to develop their own critical engagement with macroeconomic policymaking. In so doing, the book provides an understanding of the world economy’s fluctuations since the global financial crisis that embraces the uncomfortable fact that inconsistency and the need for a multiplicity of models is central to macroeconomic policy choices.

For the many students bewildered by the disconnect between the models in their textbooks and the policy choices so hotly debated in the press, the book will be essential reading.

• Language: English
• Publisher: Agenda Publishing
• Release date: Sep 28, 2023
• ISBN: 9781788216579

Matteo Iannizzotto

Matteo Iannizzotto is Associate Professor in Macroeconomics at Durham University.

Book preview

MACROECONOMIC POLICY SINCE THE FINANCIAL CRISIS

Also by Matteo Iannizzotto and published by Agenda

Post-Keynesian Theory Revisited: Money, Uncertainty and Employment

MACROECONOMIC POLICY SINCE THE FINANCIAL CRISIS

Matteo Iannizzotto

Image:logo is missing

© Matteo Iannizzotto 2023

This book is copyright under the Berne Convention.

No reproduction without permission.

All rights reserved.

First published in 2023 by Agenda Publishing

Agenda Publishing Limited

PO Box 185

Newcastle upon Tyne

NE20 2DH

www.agendapub.com

ISBN 978-1-78821-654-8 (hardcover)

ISBN 978-1-78821-655-5 (paperback)

British Library Cataloguing-in-Publication Data

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

Typeset by Newgen Publishing UK

Printed and bound in the UK by TJ Books

Contents

Preface

Acknowledgements

List of figures and tables

1 The three-equations model

1.1 Why is the IS-LM model no longer suitable?

1.2 The three-equations model

1.3 The three-equations model in operation

1.4 The double lag and the Taylor rule

1.5 All models are wrong but some are useful

1.6 Suggested further reading

2 Behind the three equations I: the monetary rule and the IS curve

2.1 Behind the monetary rule

2.2 Behind the IS curve

2.3 Suggested further reading

3 Behind the three equations II: inflation and the Phillips curve

3.1 Inflation and the Phillips curve

3.2 The imperfectly competitive labour market narrative

3.3 The conflicting claims narrative

3.4 Integrating the two narratives

3.5 Differences with the DSGE

3.6 Suggested further reading

Appendix: Perfectly and imperfectly competitive labour market

4 Expectations

4.1 Endogenous or model consistent (rational) expectations

4.2 Risk and uncertainty

4.3 All models are wrong but some are useful

4.4 The policy ineffectiveness proposition (Sargent & Wallace)

4.5 Barro Gordon (1983)

4.6 All models are wrong but some are useful

4.7 Suggested further reading

Appendix: Calvo pricing and the New Keynesian Phillips curve

5 The financial crisis of 2007/08

5.1 Deflation

5.2 Not ignoring the banking sector

5.3 A smoothly functioning system

5.4 The financial crisis

5.5 All models are wrong but some are useful

5.6 Suggested further reading

Appendix: Unconventional measures of monetary policy

6 Financial instability

6.1 The financial instability hypothesis

6.2 Hedge finance

6.3 Speculative finance

6.4 All models are wrong but some are useful

6.5 Suggested further reading

7 The three-equations model for the open economy

7.1 The legacy of the Mundell-Fleming model

7.2 Extending the three-equation diagrams to the open economy

7.3 The Dornbusch overshooting model

7.4 All models are wrong but some are useful

7.5 Suggested further reading

Appendix: An alternative Mundell-Fleming

8 Fiscal policy

8.1 The Ricardo-Domar approach

8.2 Fiscal rules (the eurozone)

8.3 Other fiscal rules

8.4 All models are wrong but some are useful

8.5 Suggested further reading

9 Broken shards of fiscal policy

9.1 Immediate aftermath 2007–08

9.2 Expansionary austerity 2010

9.3 The Reinhart-Rogoff controversy

9.4 The sectoral balance approach

9.5 Sectoral balances approach (after Krugman 2009)

9.6 Modern monetary theory

9.7 Conclusions

9.8 Suggested further reading

10 Ambiguities and problems

10.1 The energy crisis

10.2 Profit inflation and the interest rate

10.3 Policy responses

10.4 Alternative policies

References

Index

Preface

Anybody who has taught macroeconomics at an undergraduate intermediate level in the UK has faced a major problem in the choice of texts as the market is largely dominated by American textbooks that fall into two broad categories: introductory texts and advanced graduate books. The former can be very good for first-year undergraduates in the UK who have typically never studied macroeconomics before, and the latter may be suitable for advanced electives in the third year, once the compulsory course of macro in the second year has been done. But it is precisely this second-year course that falls in between and never finds a book that is pitched at the right level. The disconnection or the gap has also been reflected in the macroeconomic models being presented in the two classes of books. Introductory texts (often called intermediate by US publishers) have largely relied on some version of the venerable IS-LM model, however revised in presentation. Advanced texts focus instead often on largely very technical models of some long-run equilibrium, and take for granted most policy questions and implications faced by central banks and treasury departments. The undergraduates who have elected to do economics often find themselves bewildered and unable to bridge the gap between the models imposed on them to study and the current policy choices hotly debated in the press. In this large gap, an alternative model was introduced in 2006 by Wendy Carlin and David Soskice proposing a version of the consensus model that would be pedagogically viable to an undergraduate audience and would be consistent, albeit with some crucial changes, with the models used by central banks for forecasting and policy decisions.

At the outset I should acknowledge my debt to Carlin and Soskice because it is unquestionably on the basic structure of the three-equations model that I have built and expanded my teaching for many years, and this entire book is largely based on that model.

Having said this, teaching and textbooks are inevitably the product of compromises where time constraints, the desire to say something about current policy choices, and individual model preferences play a very large role. It should not be a matter of surprise then that the compromises that are acceptable to some are not acceptable to others. This book is effectively the presentation of the compromises that I have considered acceptable and have therefore included in my teaching. It may help to go through them as they also effectively provide a map of the book.

In the basic rendition of the model, I face much the same problem that Carlin and Soskice had to deal with, in that there is a lot of background theory that underlies each of the three equations and entire books have been written on them. But doing so would detract from the immediate desire of making the case for using the new model for understanding policy choices. A compromise is therefore necessary in choosing to present the mechanics of the model first (Chapter 1), and relegating the background material to subsequent chapters that may be skipped or examined at different stages. In presenting the basic model, I have chosen, however, to break up some diagrams over different periods in order better to emphasize the passage of time, that is a desirable feature of the model. This is very much a pedagogical choice as experience has taught me that students often struggle with the stepped adjustments of these diagrams and need to see, at least at the beginning, each time period in isolation, rather than the combined overall result. I have also preferred to focus on negative shocks in order to be closer to current policy choices where the need to react to negative demand and supply shocks has been predominant, in a way that certainly was not the case during the great moderation.

The subsequent two chapters deal with the background of each of the three equations and as stated, they may be read out of sequence. While the specification of the policymaker reaction function is pretty uncontroversial, I have struggled a lot in presenting what I felt were the many ambiguities behind the IS curve (Chapter 2). I deliberately chose not to delve too deeply into intertemporal consumption models and present instead the IS curve as arising from two possible renditions: a macroeconomic one that has a distinct Keynesian flavour, and a microeconomic one that hails from the theory of loanable funds. I am well aware that a lot more could have been said, but again I chose what I saw as most intuitive and directly relatable.

Examining the background to the Phillips curve (Chapter 3) should be done, in my mind, within the terms of a broader look at inflation, which I have attempted to do, introducing elements of other theories that I felt were significant. This is because all too often inflation is seen as a purely monetary phenomenon which is largely demand driven or a wage induced one when supply considerations are examined. I felt that a more general view had to be presented first which is exploited much later in the last chapter. I have also added an appendix on the neoclassical labour market model, because it is the kind of benchmark or reference point against which all the rest is measured and I felt that it was useful to have this material available to the reader in some form, even though it would not fit at all in the main narrative of the text. Obviously the appendix may be disregarded without detriment to the rest.

Chapter 4 arises out of my dissatisfaction with the fact that an inflation-targeting central bank is assumed at the outset without saying how such an institutional arrangement has come to be chosen. But to revisit the terms of this choice triggers an extended excursus into the theory of rational or endogenous expectations which I felt needed an explicit and detailed treatment, and a reworked rendition of the Barro Gordon (1983) problem in order to show that most monetary policy arrangements in the subsequent 30-odd years could be understood as an attempt at answering that issue of dynamic inconsistency. In this the greatest influence on the material covered is unquestionably given by De Grauwe (2012), which is densely woven with macroeconomic theory and intuitive diagrams that I have attempted to recast in the framework of the three-equations model in order to be consistent with the rest.

The financial crisis constitutes a watershed that dictates that financial matters cannot be ignored in the teaching of macroeconomics. This I have attempted to deal with in two chapters and I feel that it is these two where I have most radically departed from what is to be found in most other texts.

Chapter 5 is my own rendition of Howells (2009) largely following the progress of the diagrams as stated, because ever since I first read it, I immediately considered it the best and clearest way of introducing a banking sector and connecting it to the rest of the economy as modelled. It is here that the most complex diagrams will be found, but a case can be made that they do serve a purpose, despite their complexity. I have included an appendix on unconventional measures of monetary policy but have deliberately chosen to do so without extensive recourse to charts and data, or very intricate details on financial derivatives.

Chapter 6 is an attempt on my part at making Minsky’s financial instability hypothesis accessible through some diagrams in a way that is not too incompatible with the rest of the book. In so doing I am aware of going against the grain of much of what Minsky argued, but again a compromise was needed in order to avoid too radical a departure with what had been seen particularly in the preceding chapter on banking.

Abandoning the IS-LM model as the basic framework of presentation inevitably presents the challenge of what to do with the open economy. This is what I do albeit awkwardly in Chapter 7 where I have chosen to prioritize a rendition of the Dornbusch overshooting result that can be clearly stated in nominal terms. This is done primarily as a justification for the great volatility of the nominal exchange rate which is a macroeconomic feature that has become standard ever since the world as a whole has settled on a flexible exchange standard in the 1970s. However, this model replicates the notion that fiscal policy has no stabilization role to play, which runs counter to the history of the immediate response to the financial crisis which was unquestionably also fiscal in nature. I have chosen to suggest an alternative open economy model in the appendix where fiscal policy has an effect on output even under flexible exchange rates, as a prelude to the subsequent chapters.

The three-equations model was also the product of a consensus assignment during the period of the great moderation, whereby it fell to monetary policy to restrain inflation while fiscal policy was tasked purely with maintaining the stability of government debt over the longer run. This consensus on policy has been shattered since the financial crisis. I have therefore dedicated two chapters to fiscal policy matters.

Chapter 8 deals with what could be called the Ricardo-Domar approach that largely justifies the consensus assignment and the concerns on the stability of government debt. The following Chapter 9 is unquestionably and unavoidably uneven in that it tries to account for the disparate and somewhat heretical arguments that have surfaced once the consensus was broken. In its nature it is open-ended and should only be attempted after a thorough understanding of the material covered in the preceding chapter.

Lastly in Chapter 10 I have elected to present some ambiguities and reconnect with the rendition of inflation in Chapter 3 in order to present some of the policy choices that are hotly debated at the time of writing (spring 2023). Too many details would need to be included for a fuller rendition of them, and I have chosen not to do so because the idea was to direct the reader towards using the model and understanding its limitations when dealing with actual very uncertain policy choices.

An omission that many will lament is economic growth. This is another of the unavoidable compromises where my choice may well not be shared. But the scope of the material chosen is limited to what has historically been covered in 20 lectures over the first of two terms at Durham. Given the time limit, the material itself has to be focused on policy questions and models whose time horizon is limited between the current period and at most two to five years further ahead. Models of growth whether exogenous or endogenous are therefore outside its scope and inconsistent with an overall limited dimension of the book chosen for editorial reasons. I would hold that omitting growth is not as limiting as it may seem at first, as it is arguably the case that it is actually very difficult to come up with anything better than C. I. Jones’ Introduction to Economic Growth (various editions since 1998).

Lastly I feel the need to say a few words on the pedagogical approach that I have taken here.

At a most fundamental level, the pedagogical philosophy of the present book is not to present any single model as either the true or preferred depiction of the macroeconomy. This is very much at odds with many other texts that understandably seek to persuade the reader of the merits of the models presented. But modelling is inevitably a matter of compromise and choices that are justifiably open to question and debate. Following a famous dictum by the British statistician George Box, I hold instead that all models are wrong, but some are useful. This I have used as a convenient way of doing a summary in (most) of the chapters of this book. Sometimes this made no sense (Chapter 2, 3, 9 and 10), but in the other chapters I have tried to be consistent, showing how the models presented could be seen to be wrong in many different (sometimes trivial) ways, but arguably potentially useful in some contexts. I have routinely suggested to students that this scheme can also be used as a plan for an essay on a given economic model, in that the model itself needs to be stated first and then shown to be wrong but arguably potentially useful. This approach is arguably at odds with most modern textbooks which have prized consistency above all. This has to be interpreted mainly as internal consistency whereby a simpler model is progressively elaborated into more complex ramifications while essentially retaining the original structure and logic and further confirming the validity of it all. The desire for such consistent purity has come, however, at the price of relevance to the world macroeconomics purports to explain. Consistency, in its absolute purity, does not belong to that world or central banks would have had no need to resort to unconventional policies. I very much hold the view that is time for the teaching of macroeconomic policy choices to accept the uncomfortable fact of inconsistency and embrace the need for a multiplicity of models not all of which can be, at all times, internally consistent with each other.

As even a cursory glance will reveal, the book relies a lot on diagrams, and much more on diagrams than on equations, although a fair few of those are present too. This is the product of my own pedagogical bias because I was taught economics primarily with diagrams and I have come to need a diagram to visualize an argument that I want to present to my students. This inevitably comes at a cost, as the presentation is biased towards what can be presented in diagrams. Again this is a compromise that I saw as desirable, being fully aware that many may hold a different view. In relying on diagrams, I have produced several that may seem forbidding at first, but it is hoped they will not be so after some effort on the part of the reader. I am reminded in saying so, of what my late friend Victoria Chick once related to me about the time she found herself sitting next to Sir John Hicks at a conference. What was been presented to them at the time was no longer to be recollected. But it must have had fairly involved diagrams because Sir John turned to her and commented: I consider that having more than two schedules in a diagram is bad taste, don’t you think? The reader must be warned that what follows in the present book is in exceedingly bad taste in the Hicks sense of the term, but it should still make sense at least in some circumstances.

Acknowledgements

At the very outset I wish to acknowledge my gratitude first towards Alison Howson of Agenda Publishing for being supportive and understanding on a project that has taken a lot longer that she and I had envisaged. Second, and most importantly, I owe an enormous debt of gratitude towards Peter Howells who not only provided much of the inspiration for the book with his own chapter on banking dating back to 2009, but has also carefully read each of these chapters and provided very detailed and thoughtful comments on each. To have had a careful check on my own diagrams and thoughts has been an enormous help in a process that has not proven linear or straightforward in many ways. Obviously, I remain entirely responsible for any errors of omission and commission as not always have I followed what he suggested. I should also acknowledge, with the customary disclaimer, the helpful feedback on the open economy diagrams of Chapter 7 from my longstanding colleague and friend David Chivers with whom I have shared the teaching of macroeconomics for some years.

I equally acknowledge with thanks the crucial help with some data in Chapter 8 from Filip Stefanovic of the OECD, again with any possible errors remaining unquestionably mine.

A book like this, that in many ways represents the distillation of lessons learned and revisited over more than 20 years, inevitably implies exchanging ideas and comments with many, too many to mention here. Unquestionably my biggest debt remains with the many students who at various stages have changed the way I teach. I have no qualms in saying that I have learned a lot from them and it is obviously to them all that I wish to dedicate this work.

List of figures and tables

Figures

1.1The IS curve

1.2The Phillips curve and the monetary rule

1.3The three-equations model overall

1.4The three-equations model schematic decision process

1.5A (temporary) positive inflation shock

1.6A (permanent) negative demand shock

1.7A (permanent) negative supply shock

1.8The double lag structure

2.1The indifference curves of different policymakers

2.2The monetary rule of different policymakers

2.3The macroeconomic IS curve

2.4The natural rate of interest

2.5Consumption and the rate of interest

2.6Consumption and investment possibilities

3.1The imperfectly competitive labour market

3.2Conflicting claims equilibrium (confrontational)

3.3Conflicting claims equilibrium (consensual)

3.4Conflicting claims dynamic changes

3.5Combined narratives diagram 1

3.6Combined narratives diagram 2

3.7Combined narratives diagram 3

3.8Expansionary disinflation

3.9Neoclassical labour market

3.10Leisure and consumption optimization

3.11Perfectly and imperfectly competitive labour market

4.1Rational expectations as a weighted average

4.2The random shock

4.3Sargent and Wallace (1975)

4.4Barro and Gordon 1

4.5Barro and Gordon 2

4.6Barro and Gordon 3

4.7Solution 2 ERM

4.8Solution 3 New Zealand

5.1The deflationary trap (after Carlin & Soskice 2014)

5.2The banking circuit

5.3The banking sector diagram

5.4The extended model diagram (Howells 2009)

5.5A smoothly functioning system during the great moderation (Howells 2009)

5.6Financial crisis 1 (Howells 2009)

5.7Financial crisis 2 (after Howells 2009)

5.8Financial crisis 3 (after Howells 2009)

5.9Policy reactions to the financial crisis

6.1The dot-com bubble (after De Grauwe 2012)

6.2Minsky (after Iannizzotto 2020)

6.3Hedge finance

6.4Euphoric stage (after Iannizzotto 2020)

6.5The Minsky moment (after Iannizzotto 2020)

7.1The open economy extended diagram

7.2Dornbusch overshooting 1

7.3Dornbusch overshooting 2

7.4Dornbusch overshooting 3

7.5Open versus closed economy interest rate changes

7.6A different Mundell-Fleming

7.7A fiscal expansion 1

7.8A fiscal expansion 2

8.1Primary deficit with real interest rate smaller than growth rate of GDP: stable dynamics

8.2Primary surplus with real interest rate bigger than growth rate of GDP: unstable dynamics

8.3Stable and unstable dynamics over the years

8.4Revision of the risk premium

9.1Sectoral balances (Godley) 1

9.2Sectoral balances (Godley) 2

9.3Sectoral balances (Krugman) 3

9.4Sectoral balances (Krugman) 4

9.5Sectoral balances (Krugman) 5

10.1Energy price shock

10.2Profit inflation after an energy price shock

10.3Interest rate changes after an energy price shock

Tables

8.1Primary surplus relative to GDP

8.2Real interest rate minus real income growth

1

The three-equations model

In the last 20–30 years of the twentieth century a peculiar disconnect developed between the kind of economic policy model that was taught in universities and the practice of economic policy and monetary policy in particular. Over time the disconnect widened so that the theoretical and empirical models used to inform and analyse policy actions were equally disconnected from what was still being taught in universities. The issue obviously is that the latter remained the venerable IS-LM first introduced by John Hicks (1937) in the UK and Alvin Hansen in the United States (1938). That model remains the simplest and most intuitive one that many introductory textbooks still use.¹ But understandably the disconnect referred to above created many problems because so many contemporary policy decisions could only be represented in that context with convoluted and implausible steps. Understandably the search for a model that could bridge the gap and eliminate the disconnect started in earnest. The bulk of this book is dedicated to – arguably – the most versatile of such replacements: the three-equations model pioneered by Carlin and Soskice (2006, 2009, 2014).

1.1 Why is the IS-LM model no longer suitable?

To understand why the IS-LM model has to be abandoned it is useful to distinguish between two separate sets of problems about it. One set was always there from the very beginning and its existence had been known since the 1930s, but the other advantages of the model, in terms of its simplicity and intuitiveness, were enough to compensate. These problems are the following.

First, the IS is in flow terms whereas the LM is in stock terms. That is, one relationship measures the amount of consumption, investment, and net exports per unit of time, while the other is a statement of how much liquid money there is at a specific point in time.² This mismatch was essentially unavoidable because the IS-LM model was intended to be a mathematical/diagrammatic translation of what Keynes had written in the General Theory (1936), and such a mismatch was already there.

Second, the IS depends on the real interest rate (r) whereas the LM depends on the nominal interest rate (i). That is, because decisions to invest inevitably take into consideration the passage of time, it is logical therefore to conceptualize them as trying to net out any intervening change in purchasing power, which means that the interest rate negatively affecting investment has to be interpreted as the real one. That is not the case for the decision to remain liquid because in that case the main consideration is the opportunity cost of remaining liquid that is represented by a nominal interest rate that can be earned on deposits or very short-term assets. Again the difference between the two was already present in Keynes (1936) and was therefore unavoidable. The problem is therefore that the interest rate measured on the vertical axis of all IS-LM diagrams would need to be the real one for the IS curve and the nominal one for the LM. In practice putting the two curves together in a single diagram blurs this distinction.

The two definitions of the interest rate are related by the Fisher relationship but they cannot be considered identical and depend on what the expected rate of inflation is (πe). This can be done as follows. The gross³ real interest rate can be defined as the ratio of the (gross) nominal rate divided by the (gross) expected rate of inflation:

The reason why the expected rate of inflation has to be taken into account is because the earning of interest involves the passage of time, during which obviously prices may vary in ways that are unknown at the moment of making the investment decision.⁴ Some simple manipulation can translate the above into the following:

This last statement is the Fisher relationship and it holds as an approximation in that the term (rπe) in the second last line is considered of second order of magnitude as the product of two percentages and disregarded. The diagrams are not therefore totally implausible in so far as one also parallelly assumes that the expected rate of inflation is of a given magnitude and unlikely to change in the timeframe being considered. This extra assumption is always there when making use of an IS-LM diagram but it is not always explicitly stated.

These problems were all known from the very beginning of the model but have not detracted from its flexibility and eclectic versatility, which meant that, in the end, it just remained too useful and too convenient, so that it still has its advocates (e.g., Krugman 2009), following Keynes (1936) in preferring to be approximately right rather than precisely wrong.

However, with the 1990s (if the not the 1980s) a separate set of problems arose that made continuing to use the