INSTRUCTOR’S GUIDE
Instructor’s Instructor’s Guide Gu ide
to accompany
Macroeconomics: A European Text
Michael Burda Charles Wyplosz
OXFORD 1997
Instructor’s Instructor’s Guide Gu ide
to accompany
Macroeconomics: A European Text
Michael Burda Charles Wyplosz
OXFORD 1997
Oxford University Press, Walton Street, Oxford OX2 6DP Oxford New York Toronto Delhi Bombay Calcutta Madras Karachi Kuala Lumpur Singapore Hong Kong Tokyo Nairobi Dar es Salaam Cape Town Melbourne Auckland Madrid and associated companies in Berlin Ibadan
Oxford is a trade mark of Oxford University Press
Published in the United States by Oxford University Press Inc., New York © Michael Burda and Charles Wyplosz 1997
All rights reserved. reserved. No part of this publication publication may be reproduced, reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press. Within the UK, exceptions are allowed in respect of any fair dealing for the purpose of research research or private study, or or criticism or review, review, as permitted under the Copyright, Designs and Patents Act, 1988, or in the case of reprographic reprographic reproduction in accordance with the terms of the licences issued by the Copyright Copyright Licensing Agency. Agency. Enquiries concerning concerning reproduction outside these terms and in other countries should be sent to the Rights Department, Oxford University Press, at the address above
This book is sold subject to the condition that it shall not, by way of trade or otherwise, be lent, re-sold, hired out or otherwise circulated without the publisher’s prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser purchaser
ISBN : 0-19-877378-1 Printed in Great Britain by Docuprint, Bath, Avon
INTRODUCTION
In this Instructor’s Guide to the second edition of Macroeconomics: Macroeconomics: A European Text , we would like to share with teachers our approach to teaching the subject. More concretely, we review some of the most important ways in which this book differs from other textbooks on the market, and provide instructors with some hints -- based on our experience -- in making the material digestible and even appealing to students. Each chapter provides a short summary of the important concepts, as well as a short list of further reading which may help users of this textbook to obtain more perspective on the strengths and weaknesses of our approach to teaching macroeconomics.
Lectures and readings
There is a big difference between what is said in class and what the students read in the textbook. This is why the time spent in class and on the textbook should be seen as complements, not substitutes. We have found that lectures are best devoted to a limited number of the issues covered in the book, usually at a basic level. While it is important to make the material look as easy, clear, and interesting as possible in class, students can be made responsible for a much larger range of material not covered in class. Those students who study the book carefully before lectures -- hopefully the majority -- are rewarded in two ways: they can check that they have understood what matters and why, and they have a chance of asking detailed questions. Following this recipe, each chapter can be covered in a class lasting about 90 minutes, although Chapters 4, 10, 11, and 13 are best spread over several lectures. Of course, teachers who have more time might spend commensurately more hours on all the material.
Examples, not proofs or literature reviews
What makes this textbook different is the large number of examples drawn from all over Europe and elsewhere. Examples serve two purposes. First, they break the
usual monotony and offer a breather. Second, they offer a reality check on the theory with which students are trying to understand for the first time. The immediate relevance of the theory has been, in our experience, an important incentive for the students to retain what they have learned. This is why we have tried systematically to present an example whenever a new concept is introduced or a new result is established. This is a recipe we have experimented with in class with great success. Students enjoy relating their own experiences to the principles they learn. We strongly encourage teachers to present some of these examples in class (overhead projectors are great teaching devices) and take the time to comment extensively on them. The improved quality of the transparencies supplied as an add-on should make this option more attractive. Of course, examples are not proofs. Yet it is our belief that it is inappropriate to submit theories to formal analysis in an introductory textbook. Informal inference and introspection can generate many of the propositions and theories that we present, which to our minds represent a consensus view of macroeconomics in the 1990s. We intentionally shield the student at this level from the nuances of the empirical (and econometric) literature; we feel that our presentation of macro is relatively well-established in the literature and are confident that the ’facts’ will continue to be supported by further research. In the forefront was our concern not to reinforce the familiar stereotype of ’twohanded economists’ unwilling to take a stand .
Topics
A shorter course may limit itself to the first fifteen chapters, possibly skipping Chapter 7 which, like Chapters 18 and 19, deals with exchange rates while Chapters 20 and 21 cover special topics. For an even shorter course the teacher may want to focus on the basics and accordingly drop Chapters 15 to 17. Older texts typically cover growth (Chapter 5) and labour markets (Chapter 6) at a much later stage. Nowadays this is hardly acceptable. Since the real economy occupies a central role in our presentation it is essential
Instructor’s Guide
Introduction
to establish long-run growth and labour markets at the heart of macroeconomics. Similarly, we understand better now how dangerous it is to separate too cleanly the short run (business cycles) from the long run (economic growth). Yet, if hard pressed by a tight programme, a teacher could drop Chapter 5, and, as a last resort, Chapter 6.
Order
The preface to the textbook suggested two possible tracks: an aggregate demand/business cycles track moving quickly to the IS-LM and demand/supply approach familiar from previous popular textbooks; and a neoclassical/microfoundations track which moves slowly to equilibrium emphasising behavioural relationships first. In the end, it is a matter of taste. Yet we encourage teachers to cover at least Chapter 3 early on. The IS-LM analysis is very useful but students should understand that it is a short run approach, not only because of price and wage rigidity, but also as a result of intertemporal budget constraints. It is our experience that students quickly grasp and retain the particularly useful message that most macroeconomic choices (consumption and spending, investment, budget imbalances) are fundamentally intertemporal, that intertemporal budget constraints bite, and that with forward-looking financial markets they bite relatively quickly.
Changes from the first edition
The second edition is different from the first in a number of significant ways. Obviously, we have updated and streamlined tables and figures and have tried to add more current examples. Second, we added some new exercises and deleted other which were too difficult or repetitive. Third, we edited and revised most of the chapters significantly -- examples being Chapters 4, 7, and 11 (old Chapter 10). Fourth, we have switched the order of Chapters 5 and 6, reflecting a desire for more continuity. Most significantly, we have added two new and important chapters, responding to many demands from instructors: Chapter 10 is the much-sought after integration of the real parts of the macroeconomy with the monetary sector in the usual classical (flexible price) but also Keynesian (fixed price) analyses. Rather than calling it a "closed economy detour" we prefer to think of it as
ii
the global economy context, highlighting determinants the world rate of interest, inflation, output, etc. We are particularly excited about our survey of business cycles, Chapter 14. Not only is this an opportunity to apply the AS-AD model and explain a durable feature of capitalist economies, but also an opportunity to highlight two very different ways of looking at the world. This treatment parallels that of Chapter 10: the AS-AD, sticky price variant corresponds to the Keynesian short-run analysis, while the real business cycle view parallels the classical flexprice analysis. To motivate discussion, we provide some "stylised facts" on the cycle -- some of which we think are not well-known -- using the reference cycle methodology of Burns and Mitchell, and considering averages over a number of OECD countries. Our conclusion is agnostic, meaning that the both views of the business cycle have merits and diff iculties.
Mathematics
Wherever the students’ level permits, we encourage teachers to use maths in class. Most chapters have a mathematical appendix which offers the backbone of the material covered in the main text. It is primarily designed for classroom use when possible. On the other side, if the students are not at ease with mathematics, there is a real danger of their focusing most of their learning efforts on the formalization instead of grasping the underlying economic meaning. This is especially true of introductory macroeconomics courses. In such cases, mathematics is better assigned as optional reading.
Exercises
Each chapter (with the exception of the first and last two) contains roughly twenty exercises. A first group, labelled theory, directly relates to the material presented in the text. These exercises are designed to check the understanding of key results; sometimes they offer extensions. A second group, labelled applications, is meant to train the students to translate concepts and results into useful tools. These applications are sometimes ambiguous, with more than one acceptable answer, just like real life. Teachers may use them to expose students to the limitations of a social science. Within each group, exercises are normally presented in order of ascending difficulty.
Instructor’s Guide
Introduction
The second edition of Macroeconomics: A European Text contains a number of important changes which are aimed at improving pedagogy as well as streamlining and unifying content. In addition to describing these changes, the accompanying Instructor’s Guide offers a complete set of solutions to both the theoretical and applied exercises at the end of each chapter. We hope that this improvement will make teaching with the book a more convenient and pleasurable experience.
Acknowledgements, present and future
In preparing the guide, we have kept in mind the reactions of early users of the textbook. Their questions have reminded us that what is clear and simple to one person may be problematic to another, which makes teaching challenging and fun. In addition to the scores of contributors mentioned in the acknowledgements, we are particularly grateful to Simon Burgess for his helpful review in the Economic Journal of May 1994. Successive waves of students at INSEAD, Berlin and Geneva continue to tell us how they like and dislike the exercises and have led us to rethink a great many of them. We continue to receive helpful comments and suggestions from both colleagues and students and will continue to exploit them in future editions of the textbook and this guide.
Michael Burda and Charles Wyplosz
iii
CHAPTER 1 WHAT IS MACROECONOMICS?
This chapter corresponds to the very first lecture, maybe just fifteen minutes if the schedule is tight. Advance reading by the students is desirable, but not indispensable. We see this chapter as both a motivating introduction to the subject matter and a declaration of intent. In this introduction we make it clear that, in our view, it is important to take time early on to look carefully at some macroeconomic data. This activity conveys the message that macroeconomics is about explaining facts and that these facts are given by aggregated data. The danger exists, however, that eager students will want to explain everything immediately. For that reason the lecturer should avoid an overload of data which may be dizzying or even discouraging. We begin with ’this is what I shall talk about’: it presents the key concepts of macroeconomics: (real) GNP, growth, and cycles; factors of production and income distribution; inflation; the link between the real economy and financial markets; and, of course, openness. No precise definition is offered, as the sole purpose is to appeal to intuition and stimulate interest. Section 1.2 moves from issues to the social role of macroeconomics. It is designed to alert the student to the broad implications of what he is about to study. We have chosen to claim that policies inspired by macroeconomic theory have altered the shape of business cycles, because we believe that this is the case, but we know that it is controversial. Even the staunchest anti-Keynesians and real business cycle theorists would agree that the behaviour of prices has changed since World War II (Fig. 1.5) and lay the blame on (bad?) macroeconomic policy! The tone changes in Section 1.3, which stresses that macroeconomics is not a description but an analysis of the facts. The distinction between exogenous and endogenous variables is introduced early on to emphasise that we work with models. We find it useful to warn that while macroeconomics is a scientific field -- in its rigour and the methods that it uses -- it is nevertheless plagued or blessed with the particular difficulty of dealing with social phenomena. This is the time, we feel, to refute the newspaper allegations that economists chase irrelevant theories.
CHAPTER 2
MACROECONOMIC ACCOUNTS
Objectives
There is no short cut: students must know the bare minimum about national income accounting before they can study macroeconomics. We have clearly chosen the ’light touch’ with respect to accounting, but this dry material can be effectively used to paint the ’big picture’ both efficiently and rigorously. The presentation can be structured around the flow diagram in Fig. 2.2, which maps out the flow of goods and services in the macroeconomy. The large circle represents how income flows from producers to customers and back to producers. The smaller circles correspond to the three sectors of the economy: the private sector, the government, and the rest of the world. This sectoral breakdown of the economy is the backbone of the book. Students should understand that what comes in is not the same as what goes out because any of the three sectors can be out of balance, with the imbalance matched by a build-up or draw-down of (net) assets. Imbalances arise when one sector attempts to spend more, or less, than it earns. Despite the fact that total balance requires that the three sector imbalances cancel out as shown in (2.7), intertemporal budget constraints imply restrictions for each sector in the future. Playing up this result exposes students to the concept of market equilibrium and to the distinction between ex ante and ex post behaviour as well as preparing for the next chapter (intertemporal balances).
GDP and related concepts
Like the rest of the industrialized world, we now emphasise GDP over GNP. This required us to rework many of the definitions, but with little loss along the way. The first two definitions of GDP -- total spending and total factors income -- serve later on to clinch the concept of general aggregate equilibrium. This is why it is essential to stress this point over and over again. Students typically like to discuss ad infinitum about the underground economy and other limitations of GDP data. They should be told early on that aggregate data
are inaccurate but that most of the time we use them to measure growth rates, not levels. The margin of error is reduced with growth rate as long as distortions do not vary systematically. The next important distinction is nominal versus real. Having defined deflators it is natural to contrast them with price indices. Another more important distinction, between GDP and GNP, is often too subtle to grasp at first blush. Just mentioning it early on should suffice as we shall return to it more formally in Chapter 3. On the other hand, the concepts ’factor income’ and ’factors of production’ recur frequently and it is useful to stress them early on.
The circular diagram
It pays to work carefully through the diagram, because it leaves students with a good insight flow to understand the accounting terms. It is best to start from the left where GDP is shown and ask students what happens to sellers’ incomes. As we pass the bifurcation between consumption and saving on the right of the large circle, we move from the incomes breakdown of GDP ( Y =C +S +T ) to the spending breakdown (Y =C + I +G+ X - Z ). It is worth showing the two relationships at this stage and later on to derive the identity as ( I -S )+(G-T )+( X - Z )=0, noting what it implies for the three circles. The diagram misses out a few connections or details which may be brought to the students’ attention (as proposed in some exercises): - corporate and personal savings (and therefore private income and private disposable income) are not distinguished. Corporate savings may be represented explicitly by a ’pipe’ going to the private sector circle and originating where national income is written: after the bifurcation we would have national disposable income. - trade in assets is not shown. Each sector’s imbalance is financed somehow, but the diagram does not say by whom. Our view was that with a fully integrated world capital market, it really doesn’t matter.
Instructor’s Guide
Chapter 2
Additional ’pipelines’ would be necessary -- perhaps in another colour -- to illustrate this counterbalancing flow of assets. These would all hook up with the ’world capital market’ which allocates world savings and investment. The logistics of the current diagram are already quite daunting, and we have decided to leave it be at this stage.
Balance of payments
In the same way as for national income accounts, the challenge is to make accounting interesting. It is relatively easy to do so, emphasising that the current account is the pivotal concept: it separates out ’real’ (trade in goods and services) from financial transactions ones (trade in assets). As is well known, the distinction between trade in goods and services and trade in assets is not completely airtight, but it is very important to stress the distinction early on. It leads directly to stressing that, because the current account represents the net external lending or borrowing of the nation, the lower part of the balance of payments, private and official financial transactions, simply matches the current account, hence (2.9). This is the time to recall the identity Y = C + I +G+CA and show that CA = Income - Spending. It is also useful to signal early on the difference between fixed and flexible exchange rate regimes by explaining the role of official interventions and of the monetary authorities. As the residual ex ante overall imbalance, official interventions show what the monetary authorities have done to prevent the price of domestic currency -- the exchange rate -- from moving all the way until the overall account is balanced ex ante.
3
CHAPTER 3
INTERTEMPORAL BUDGET CONSTRAINTS
Objectives
This chapter provides students with an understanding of intertemporal trade and its graphical representation. The chapter can be extended, according to the instructor’s preferences, to include more detailed discussions of bond prices and interest rates as well as other aspects of intertemporal budget constraints. Two-period diagrams are used throughout as a simplifying but intuitive device. The main drawback of this approach is that it rules out second period investment because the economy ends afterwards. Intentionally, we do not delve into the associated difficulties. These are discussed in more detail below. This is one chapters where mathematics is really essential; most instructors will agree that the simple algebra of discounting should be part of everyone’s tool-kit. An important distinction is introduced for the first time here, which the instructor will should be familiar with for the inevitable questions which arise. This is the distinction between overall public or external deficits or surpluses versus primary deficits or surpluses, which exclude interest payments or more generally investment income receipts. This distinction will prove helpful when, later on, debt service will be shown to be a major 1 source of instability.
makes it clear that each term refers to net saving, i.e. a shift of resources over time.
Constraints and optimisation
This chapter sets out budget constraints but refrains from dealing with preferences or optimal behaviour. An alternative is to teach consumption in one shot -- that is combining constraints and optimal choice, followed by investment and the current account. This ’functional approach’ is possible by pairing the corresponding sections of Chapters 3 and 4: - Consumption: Sections 3.3 and 4.2 - Production and Investment: Sections 3.4 and 4.3 In our view, there are good reasons for separating constraints and behaviour. First, the very notion of intertemporal trade is hard to grasp when first introduced. Limiting this first contact to constraints is a way of reducing complexity. Second, showing the similarities and differences between the three sectors’ constraints has great pedagogical appeal. Third, the aggregation of sectors offers a natural link with national accounts, and extends neatly to the foreign sector (current account). Finally, it allows us to give a nearly complete and yet relatively simple treatment of Ricardian equivalence without studying the relevant 2 behavioural assumptions.
Motivation Varying the level of difficulty
A good way of starting is to recall the circular flow diagram of the previous chapter (Fig. 2.2) and point out that one task of macroeconomics is to study the relationship between output, inflation, interest and exchange rates, to imbalances in the three circles. Then the accounting identity which shows the link between the imbalances: CA = (S - I ) + (T - G)
Since the chapter starts with accounting and ends at the frontier (Ricardian equivalence), the teacher must take a decision on how far to go. This in turn may have ramifications for material which can be treated later on. For a short course, focusing on essentials means using the two-period diagram to explain that the interest rate is a relative price and to show graphically what is a 2
1
It is also one reason why the IS - LM model is a short-run analysis.
What is lost is mainly the whole question of bequests and altruism across generations. One exercise (Theoretical Exercise 6) provides an opportunity to introduce the idea.
Instructor’s Guide
Chapter 3
present value; this is applied to the consumer, the firm, and the government but consolidation is not shown (skip Sections 3.4.4, 3.5.2, 3.5.3, 3.5.4 and 3.6.). For longer courses and/or advanced audiences, one key issue the instructor must decide is how extensively to treat Ricardian equivalence. One possibility is to just present the consolidation of accounts in successive logical steps (Sections 3.4.4, 3.5.2 and 3.6) and leave it at that. Another is to state the equivalence proposition (Section 3.5.3) and explain in a few words why it may fail to hold in practice. Full treatment implies using the material presented in the more demanding Section 3.5.4. We take the middle-road view that Ricardian equivalence is an interesting theoretical idea with mixed empirical 3 support.
productive technology, an alternative means of converting resources today into resources tomorrow, is the production function. The desirability of this technology is determined by how well it stacks up against the opportunity cost of resources today versus tomorrow (the market interest rate). Making this clear and simple is the real challenge of this chapter. Net wealth of the consumer can be read in terms of today’s consumption on the horizontal axis of the twoperiod diagrams. (It can also be read in terms of tomorrow’s good as well on the vertical axis, but this is suppressed to avoid confusing students.) The value of the firm can also be read -- in terms of tomorrow’s good -- as the vertical distance between the production function F (K ) and the cost-of-borrowing line OR in Fig. 3.5. Note that we assume the absence of existing productive capital (fruit-bearing trees) so that investment today and capital stock tomorrow are identical. This makes the presentation simpler but unrealistic. Box 3.3 alerts students to this fact and Chapter 4 will extend the model appropriately. (In addition, the ’planting season’ restricts Crusoe from planting coconuts he could borrow in the financial markets, a point that the smartest students will quickly pick up!). The discussion of the production function at this juncture will give the instructor an opportunity to remind students of the distinction between physical investment (expanding the productive capital stock) and financial investment (swapping existing assets).
Two-period Crusoe
Irving Fisher introduced Robinson Crusoe to economics in his pioneering work on intertemporal aspects of economic decisions. Since then, there have been two categories of textbooks which present the topic: those with Robinson, and those without. We think the Crusoe model represents an important and robust microfoundation of macroeconomics, and is the source of much useful intuition about the subject. We have toned down the parable in deference to those who may find the device too simple or even offensive. No doubt, there are two categories of teachers, those who use Robinson and those who don’t. With Fisher’s two-period framework almost all that must be understood in an introductory course can be presented compactly with two periods (present and future). In addition it prepares the students for thinking in terms of short and long run. This is why, throughout the text, we interpret the first period (today) as the short run and the second period (tomorrow) as the long run. 4 It is a trick which works almost all the time . Some indications of its shortcomings are given below. Appendices to this and other chapters show the transition from the two-period case used in the text to an infinite horizon.
Consolidation
The consolidation of the private sector -- consumers and firms -- requires that we flip the production function around the vertical axis. Indeed in Fig. 3.7 investment is read off the horizontal axis from right to left in contrast with Fig. 3.4. In Fig. 3.7 it is worth emphasising the fact that the production rise above BB indicates that productive technology raises wealth, the last term in the second line of (3.9). Of course, the optimal level of investment can be deduced from Fig. 3.7, but this task is left f or Chapter 4.
The intertemporal budget constraint Consolidation in the two-period framework with investment in both periods
For both the consumer and the government, the budget constraint is a line whose slope is given by the gross ’market’ interest rate (1+r). It is a ’technology’ which allows resources today to be converted into resources tomorrow. For the firm which has access to a 3 4
It was assumed that there is no productive capital to start with, so that investment today and the capital stock tomorrow are identical. An alternative presentation is as follows. Endowments are the outcome of
Two references are Barro (1974) and Bernheim (1987). A good reference is Frankel and Razin (1987).
5
Instructor’s Guide
Chapter 3
previously accumulated capital -- trees in existence at time 0:
preferences and behaviour. Our intention was twofold. First we wanted to use the equivalence principle as a convenient application of the intertemporal budget constraint without taking a stand on its ultimate veracity. Second, we thought it important to stress that Ricardian equivalence indeed arises first and foremost from budget constraint considerations: once the aggregate private sector realises that it will pay future taxes, the solvency of the government implies that purposeful and rational private agents will take note of this fact. The discussion which follows then allows to focus on a number of points that students may remember. These are: the deeper meaning of consolidation ( ex post it is just a matter of accounting while ex ante it carries the strong implication of equivalence); the difference between interest rates faced by the public and private sectors and the income effects associated with public borrowing; the notion of credit constraints; distortionary taxation; the disconnectedness of taxpaying generations; and uncertainty stemming from the mortality of taxpayers.
Y 1 = F (K 0) and Y 2 = F (K 1).
Resources available for consumption in both periods are now: C 1 = F (K 0) - I 1 and C 2 = F (K 1).
In present value terms: C 1 + C 2 /(1+r ) = Y 1 - I 1 + Y 2 /(1+r ).
Although there is no second period investment (end of the world) so that I 2=0, it is correct and more general, therefore, to write: C 1+ I 1 + (C 2+ I 2)/(1+r ) = Y 1 + Y 2 /(1+r ).
It can then be shown that (3.20) is just the consolidated budget constraint of the nation by recalling the two budget constraints: - private sector: C 1+ I 1 + (C 2+ I 2)/(1+r ) = Y 1-T 1 +(Y 2-T 2)/(1+r )+rF 0.
which is (3.9) modified in two ways: 1) I 2 has been added; 2) if Y is GDP and not GNP, we need to add the income earned on net foreign wealth F 0. - public sector References
G1 + G2 /(1+r ) = T 1 + T 2 /(1+r ).
Barro, Robert (1974) ’Are Government Bonds Net Wealth?’, Journal of Political Economy, 82: 1095-117.
which is (3.11) with r G = r . Adding up these two equations gives:
Bernheim, Douglas (1987) ’Ricardian Equivalence: An Evaluation of Theory and Evidence’, NBER Macroeconomics Annual, 2: 263-303.
C 1+ I 1+G1 +(C 2+ I 2+G2)/(1+r ) = Y 1+Y 2 /(1+r )+r F 0.
which is (3.23). Note that Y =C + I +G+PCA since Y is GDP. So the last equation can be r ewritten as:
Frenkel, Jacob, and Razin, Assaf (1987), Fiscal Policies and the World Economy , The MIT Press, Cambridge, Mass.
PCA1 + PCA2 /(1+r ) = - F 0.
or assuming that interest is paid at the end of the period as in (3.21) PCA1 + PCA2 /(1+r ) = - (1+r )F 0.
If we want Y to represent GNP, then it includes the return on net foreign assets and F 0 disappears in the private sector budget constraint as well as in the consolidated account. Now, however, Y =C + I +G+CA (see (3.22) and we have: CA1 + CA2 /(1+r ) = 0. Ricardian equivalence
Some teachers may be surprised that the issue of Ricardian equivalence is taken up before consumer
6
CHAPTER 4
DEMAND OF THE PRIVATE SECTOR
Objectives
The student should finish Chapter 4 equipped with a consumption function and an investment function. Given a level of output and government purchases, these two functions are the primary determinants of the primary current account. The strategy is to begin with ’first principles’ and then to inject realism. Teachers impatient to go to the IS - LM analysis faster may skip this chapter -- and return later -- provided that they offer justification for the behavioural relationships (4.4) and (4.7) or (4.23), as well as the primary current 1 account function of Chapter 7. This chapter is probably the most difficult to teach. The big stumbling block is the q-investment function. In response to many suggestions, the second edition has been modified in a number of ways to make this complex material more palatable, even to students with a limited knowledge of microeconomics.
Consumption
There is no major difficulty in shifting from the intratemporal apparatus of standard consumption theory to the intertemporal interpretation. Students only need to be warned that consumption today or tomorrow really represents a basket of goods. Impressing students with the central result that the consumption function, in theory, depends on wealth alone, is justified by the principle of consumption smoothing -- a principle which does not generally apply to other components of national expenditure, such as investment, government spending, or exports. Yet it is healthy to follow up by pointing out the well-known limitations of this elegant theory: borrowing constraints, uncertainty about future income and rates of interest. It is also helpful for the short run IS - LM
1
Chapter 11 now begins with a quick motivation for the primary current account (net export) function, so this is less important than was the case with the first edition.
analysis to derive results which will justify a Keynesian consumption function. The distinction between permanent and temporary changes in income is not only a good way for students to check their understanding, it is also an important tool of analysis. The examples provided are designed to illustrate the importance of this distinction. The role of the real interest rate in the consumption function is usually more difficult for students to grasp. It often helps to start by asking whether saving (the mirror image of consumption) should increase or decline when the real interest rate rises. Details may be 2 skipped if time is short.
Net wealth
The emphasis on endowment may leave the impression that wealth is just the present value of earned incomes. It is important to remind students that in general, financial assets and liabilities inherited from the past also enter in Ω .
Physical investment
In the first edition we pushed the q-theory of investment for a number of reasons, which we still consider valid. First, is the only one consistent with the intertemporal forward-looking approach adopted in this text and by modern macroeconomics. Second, it establishes a clean link between aggregate demand and the stock markets. Finally, even though empirical support for the q-theory is not as strong as one might like, empirical support for the alternative (that the real 2
Users of the first edition will no doubt note that Box 4.4 has disappeared. Many found it too advanced for an undergraduate text; others found the distinction made by Summers (1981b) and others to be uninteresting. On the other hand, some found it useful for sorting out the channels by which interest rates influence consumption. To accommodate these demands, we have introduced Figure 4.9 and the accompanying text.
Instructor’s Guide
Chapter 4
interest rate is the prime determinant of investment) is even weaker. The sad truth is that the only ’theory’ which works well empirically is the accelerator, but we know this has as much to do with the limitations of the 3 data as with those of theory. The treatment of investment was arguably the hardest part of the first edition. We have now changed it to allow for a modular treatment of interest rates (Sections 4.3.1 and 4.3.2) and the accelerator (Section 4.3.3) for those instructors who would like to omit Tobin’s q. For those who would like to offer a "baby version" of the q-theory, we offer Section 4.3.4. The harder underlying economics -- which still derive from the two-period model and are intact from the first edition -- is laid out in Section 4.3.5. With this new structure, it is possible to teach investment in four steps, with increasing degree of difficulty. First, use microeconomic principles to find the (long-run) optimal capital stock as a function of the 4 real interest rate. Second, provide a simple justification 5 for the investment accelerator. Third, define Tobin’s q and link investment to the market value of capital already installed (in place). Finally (optionally) introduce costs of adjustment to obtain the q-investment function.
the second panel of Fig. 4.18 the horizontal axis represents investment, not the capital stock; the vertical axis is discounted back to ’today’. Ways to make this difference clear: - recall magnitudes: that the capital stock is considerably larger than annual investment ( K / Y is about 3, I / Y is about 0.2). What we explain in step 2 is the (small) addition to the existing capacity of production. - stress that investment represents new bets on the future, while the optimal capital stock is the sum of many more decisions which turned out, on average, to be correct (otherwise the capital stock would have been depreciated away). Installation costs are less intuitive for students and often appear too insignificant to justify the centre stage that they are often given. One way to clarify the idea is that the representative firm is an approximate stand-in for the economy; although individual firms do not recognise these installation costs, the economy as a whole behaves as if this were the case (because of pecuniary or nonpecuniary negative externalities, shortrun decreasing returns in the investment goods sector, etc.). One way to stimulate students’ interest is the following sequence of points: - define Tobin’s q as the ratio of the value of installed capital to that of un-installed (or replacement) capital. Thus q is a relative price in the same way that 1/(1+r ) is the price of coconuts tomorrow in terms of coconuts today. - observe that the value of installed capital (and all other forms of capital, for that matter) is determined by stock markets. Why does a firm’s value often exceed the replacement cost of its capital? (Why can’t Daimler-Benz be reproduced merely by buying de novo all the physical equipment which comprises it?) - note that the present value of expected dividend payments (plus realisable capital gains at selling time) is the market valuation of a firm, hence the numerator in Tobin’s q. - finally note that when Tobin’s q is larger than unity it pays to install capital -- all at once! This all leaves the students with some intuition for investment. It also poses a puzzle. The intuition is that the larger q is the stronger are the incentives to invest. The puzzle is: how can q remain above unity? The 7 answer is: installation costs.
Deriving Tobin’s q
It is the final step which is hardest to digest. There are many reasons: installation costs are hard to make intuitive and students often find it hard to believe that investment moves slowly to the optimal capital stock (especially in the absence of uncertainty); implicitly at 6 least this is not a two-period analysis; and what is meant by the marginal return on investment -- the full stream of expected returns on a marginal increment to the capital stock -- often appears obscure. In presenting this material, it is essential that students understand the important differences between Fig. 4.14 and 4.18: in 3
For an extensive survey of the empirical evidence, see Chirinko (1993). 4 Recall that, because the second period is the last, all capital is lost at the end, hence MPK=1+r and not MPK=r when capital remains in place (possibly depreciated in which case the rate of depreciation must be subtracted). This is stressed in Boxes 4.6 and 4.7 in the second edition. 5 In the long run the optimal capital stock is in place and MPK=1+r . With a homogeneous production function, MPK is a function of the ratio K / Y so K / Y= g(r ). In the special case of a Cobb-Douglas production function we obtain the simple linear relationship (4.12) in the text. 6 Alternatively, we cut ’today’ into smaller sub-periods (as is explicitly shown in the Appendix). This is needed to represent the fact that we do not jump straight ahead to the optimal capital stock because it is costly to do so in one leap.
7
Some instructors might prefer to stress time-to-build considerations, which would require a multiperiod setting to treat adequately. For a nice review of the q-theory of investment in a multiperiod setting, see Summers 1981a).
8
Instructor’s Guide
Chapter 4
The next step is to show why investment depends upon q, represented on the vertical axis when the cost of new capital is unity. Maybe the hardest part is to convince the students that 1 on the vertical axis is the price of capital in terms of consumption goods. It is possible, in fact, to start the graphical exposition with the nominal cost of new capital on the vertical axis. What corresponds to point A is not q, but the nominal expected return on investment. A short-cut -- recommended for shorter courses -consists of Section 4.3.4. and restates the q-theory exactly as Tobin (1969) originally did. Firms can raise money on the stock market to buy new equipment. Once installed, though, equipment is worth more to the firm because it is combined with previously installed capital and the firm’s labour force. The ratio of the value of installed capital to new equipment, Tobin’s q, is thus a measure of how desirable it is to borrow and invest, hence I=I(q). Tobin’s q, on the other hand, depends on expected future returns from the newly installed capital, i.e. future MPKs.
References
Chirinko, Robert (1993) "Business Fixed Investment Spending: A Critical Survey of Modeling Strategies, Empirical Results, and Policy Implications," Journal of Economic Literature 31, 1875-1911. Frenkel, Jacob, and Razin, Assaf (1987), Fiscal Policies and the World Economy , The MIT Press, Cambridge, Mass. Sachs, Jeffrey D. (1981) ’The Current Account and Macroeconomic Adjustment in the 1970s’, Brookings Papers on Economic Activity, 81/1: 201-68. Summers, Lawrence H. (1981a) ’Taxation and Corporate Investment: A q-Theory Approach’, Brookings Papers on Economic Activity, 81/1: 67-140. Summers, Lawrence H. (1981b) ’Capital Taxation and Accumulation in a Life-Cycle Model’, American Economic Review, 71: 533-44.
The primary current account
The PCA function is now postponed until Chapters 7 and 11. At this point we motivate that to a large extent it can be understood from the national income identity:
Tobin, James (1969) ’A General Equilibrium Approach to Monetary Theory’, Journal of Money Credit and Banking, 1: 15-29.
PCA = Y - C - I - G
given the consumption and investment functions. Thus it is simply derived from previous results, which will be inadequate later on when two goods and relative prices are introduced. More theorising on this function is provided in Chapter 7. The interpretation of the current account as national savings can be repeated at this juncture. The reaction of the current account of an economy of ’consumption smoothers’ in response to investment booms (Spain in the 1980s), sudden increases in government spending (wars), or changes in terms of trade (Fig. 4.6) will be to respond in the same direction. The irrationality of running persistent primary current account surpluses (at least from the consumer’s point of view) can be justified using the theory presented in this chapter. A quick look at Fig. 3.16 will remind students that high surplus countries have also seen periods of current account deficits and will see them again in the future (for example, Germany as a consequence of unification, and probably in Japan as consumers begin to enjoy 8 more consumption and leisure).
8
For more on the current account in an intertemporal context, see Sachs (1981) or Frenkel and Razin (1987).
9
CHAPTER 5
EQUILIBRIUM OUTPUT AND GROWTH
General equilibrium Objectives
As we stress in the introductory chapter, economic growth may well be the most important topic in macroeconomics. Over periods of a decade or more, the average person’s well-being is more closely linked to issues of growth in per capita output and income than to business-cycle fluctuations. Thanks to recent work at the frontier, these issues are now firmly rooted in the realm of macroeconomics, where they belong; yet despite the revival of the Solow (1956) model inspired by the newer empirical work summarised in Barro and Sala-i-Martin (1995), considerable ignorance remains. A mixture of enthusiasm and caution sets the tone of this chapter. The Maddison data serve to catch the student’s eye while the Solow decomposition, and its mysterious residual, reminds us that the residual ’technical progress’ still explains a large part of growth. That growth is presented early on in the book follows from the real-nominal dichotomy which is later stressed in Chapter 10. It is more natural, in our view, to elucidate a long run toward which the economy gravitates. The chapters’ objectives are simple. First, motivate economic growth as an equilibrium process (Section 5.2) resulting from growth or accumulation of factors of production working through the production function. Second, demonstrate using the Solow decomposition just how much (or little) of growth can be accounted for in this way. Third, introduce the Kaldor stylised facts as a guidepost for viable growth theories (and introduce the notion of a stylised fact in general, which will help in Chapter 14, among other places). Next, introduce the Solow model of balanced growth and point out the importance of technological change in this model. Finally, take the student to the frontier of the field in the discussion of what technical progress really is.
Section 5.2 sits a bit uncomfortably at the beginning of this chapter and can be skipped if time is short. It does serve two important functions. First, after a chapter linking output and capital (Chapter 4) it meets the need for bringing output, capital, and labour together in a 1 form of general macroeconomic equilibrium. Second, it introduces the aggregate production function, the work-horse of both growth theory and analysis of the economy’s supply side. Fig. 5.1 is meant to symbolise that we now operate in three dimensions rather than sub-spaces. In addition, this section is used to introduce some concepts which will be needed later on: returns to scale and technological change, in particular. While some teachers may find it a bit too dry to sustain their audience’s interest, it is important to define the aggregate production function and explain what returns to scale mean for the macroeconomy. This section also fills an important gap: Section 5.2.2 provides a quick account of the determination of the world interest rate in the long run. The ’kissing tangency’ in Fig. 5.4 is a classic. It will be taken up in more detail again in Chapter 10 (Figure 10.7).
The Solow decomposition and balanced growth
The Solow decomposition is a central organising framework for the material of the chapter. One way to look at it is as just an exercise in accounting and the early part of the chapter takes this approach. Once we realise that the residual accounts for only about half of growth performance, the attention shifts from 1
In swapping the order of the labour and growth chapters in the second edition, we are forced to downplay the household’s decision to work, implicitly assuming completely inelastic labour supply. Later in Chapter 6 this omission is amended. As a result, we treat labour as a fixed input to the production process and derive the labour demand curve informally from the MPL=w rule.
Instructor’s Guide
Chapter 5
accounting to analysis. An alternative approach is to derive the decomposition rigorously from a general production function with the usual attributes. Equation (5.6) is the cornerstone of this chapter and deserves special emphasis in the classroom. It is also useful to fix students ideas about the magnitudes involved: the ’normal’ rate of growth of countries, the contribution of inputs, and the size of the residual. Following tradition, we focus on balanced growth, which occurs when particular ratios of economic interest are constant. Balanced growth paths can be thought of as a subset of steady-state growth paths, in which all variables are growing at constant but not necessarily equal rates. We chose, as Kaldor did, to focus on the relative stability of K / Y (the US is the most outstanding example). It is important to explain to students that especially for countries like Germany and Japan which lost considerable capital stock in the war, / Y increasing can be consistent with a transition to a K steady state value. The balanced growth condition, combined with constant returns and the Solow decomposition, generates a tight link between economic growth, population growth, and technical progress. We do not pretend that this standard choice is obvious. In fact, the data shown in Tables 5.6 and 5.7 should remind the student that stylised facts are regularities, not iron laws. Countries vary in many ways that are not captured by the model. The stylised facts are useful in that they impose restrictions on the aggregate production function, which in turn lead to an emphasis on the role of total factor productivity in determining per capita growth.
Later in the chapter we note that all is not well with the simple two-factor growth model: high savers seem to grow faster than low savers (despite the fact that savings rates do not affect steady-state growth in the Solow model); and that poor countries fail to catch up richer countries. A good pedagogical approach for highlighting these issues is to propose the convergence hypothesis: that GDP per capita converge ’asymptotically’ so that per capita income levels in poor countries should catch up to those in richer ones (Figure 5.13). The fact that convergence seems to occur only among the wealthier countries invariably generates much interest. Three resolutions of these difficulties are then proposed, leaving the reader free to choose among, or 2 accept all three extensions. First, it is shown that once human capital is added as a third factor to the aggregate production function both facts can be explained. The rehabilitation in Mankiw et al. (1992) has given new life to the neoclassical, constant returns approach to growth. Second, the same is true if one adds public 3 infrastructure. Finally, endogenous growth, the theory developed in the late 1980s, gives a role to increasing returns to scale and externalities and also allows us to account for the role of saving and the absence of convergence. It may therefore be useful to close the presentation by suggesting extensions of the two-factor model. Human capital is the most frequently and successfully modelled third factor. Further additions may include natural resources or public infrastructure, that can be incorporated into the Solow decomposition. Some of
Bringing in theory
2
This agnosticism is motivated by the fact that, at the time of writing, the verdict on endogenous growth was still out. Writing this guide in late 1996, we are unsure about where the empirical literature on growth is taking us (for a critical review see Solow (1994)). On the empirical side (see Levine and Renelt (1992)), three results seem important: 1) investment in human capital is positively associated with growth (Mankiw et al. 1992); 2) public infrastructure also seems to matter; 3) convergence of per capita income seems to occur within regions of a country (Barro and Sala-i-Martin 1991) at roughly 2% per year, when the steady state to which the local economies are converging is appropriately controlled for. Externalities and returns to scale may explain the distribution of activities within a country -- the new discipline of economic geography -- but may be less useful for national growth performance. 3 These two explanations must be combined with the idea of conditional convergence. Technically, if one is willing to admit ’out-of equilibrium’ behaviour, conditional convergence may also account for the positive association of growth rates and savings rates (countries with higher savings rates have a higher steady state to which they converge, will accumulate capital at a faster rate, and will therefore grow more rapidly as in Figure 5.13).
Balanced growth is the accepted way of putting more structure on the analysis. The distinction between balanced growth and steady state deserves perhaps more emphasis than it receives in the text. Balanced growth paths are a restricted subset of steady-state growth paths which requires that some selected variables grow at the same rate. The next step is the Solow model, which has a pedagogical elegance which is seldom found in our field. We derive the model in the usual way, although leaning more heavily on the diagrams than on the maths. The key result, of course, is the invariance of growth with respect to the savings rate -- a difficult result to explain but nevertheless one of central importance. As promised, an appendix with this and other elementary formal aspects of growth theory has been included in the second edition. Limitations and Extensions
11
Instructor’s Guide
Chapter 5
the exercises at the end of the chapter drive home this point.
Feldstein-Horioka
Why introduce the Feldstein-Horioka puzzle in a chapter devoted to growth? The answer is that our textbook is dedicated to the open economy, and as such 4 needs to confront this fascinating fact. In texts devoted to the closed economy, the link between saving and growth is assumed since saving (public and private) equals investment (public and private) by definition. In the open economy, this link can be broken by international borrowing or saving. Yet it survives, as Feldstein and Horioka showed. In a sense, the solution to the puzzle might have to do with sovereign risk (no country can sustain growth solely on foreign capital without being tempted to confiscate it in the end and avoid repayment) or the high correlation of permanent investment opportunities across countries in the long run (as opposed to transitory ones, to which optimally smoothed consumption would respond with currentaccount fluctuations).
References
Barro, Robert J. and Sala-i-Martin, Xavier (1995) Economic Growth, New York: McGraw Hill. Feldstein, Martin and Horioka, Charles (1980) "Domestic Saving and International Capital Flows," Economic Journal, 90: 314-29. Levine, Ross and Renelt, David (1992), ’A Sensitivity Analysis of Cross-Country Growth Regressions,’ American Economic Review, 82: 942-63. Mankiw, N. Gregory, Romer, David, and Weil, David (1992), ’A Contribution to the Empirics of Economic Growth,’ Quarterly Journal of Economics , 107: 407-38. Solow, Robert M. (1956), ’A Contribution to the Theory of Economic Growth,’ Quarterly Journal of Economics, 70: 65-94. ______ (1994), ’Perspectives on Growth Theory,’ Journal of Economic Perspectives, 8: 44-54.
4
The reference is Feldstein and Horioka (1980).
12
CHAPTER 6
LABOUR MARKETS AND EQUILIBRIUM UNEMPLOYMENT
Objectives
This chapter explains unemployment in the long run: why does the rate of unemployment fluctuate around a level which is far from small in most countries, and has risen considerably in Europe over the last two decades? One effective way we have found to teach this chapter is first to propose a standard demand and supply analysis, in which all ’unemployment’ is the outcome of voluntary choice. The paradox of how to interpret the unemployment which we observe arises immediately, and the teacher then proceeds to unearth 1 the sources of involuntary unemployment. The central message is that labour is a very particular ’commodity’: once we take into account what makes it particular, the paradox disappears. Given the many different reasons why labour is special, the chapter does not offer an all-encompassing theory of unemployment. Instead it looks at each explanation one by one, leaving the reader to add them up, and allowing the instructor leeway to stress his or her own preferred (or locally relevant) cause. Many of these aspects have been removed from Chapter 6 and can now be found in Chapter 17 (supply side).
distinction of equilibrium unemployment between frictional and structural. We find these distinctions very helpful in the classroom and, we hope, roughly correct.
Approach
It is thus fruitful for the teacher to remember that the results are ultimately summarised in (6.7) Equilibrium = Frictional + unemployment unemployment
Structural unemployment
This distinction refers to the two classes of reasons why the demand-supply framework is inadequate: - static causes of unemployment, i.e. reasons why wages are prevented from clearing the market. This is interpreted as the cause of structural unemployment. - dynamic causes of unemployment, i.e. reasons which increase the inflow into unemployment or slow down the process of job take-ups. This is interpreted as the determinants of frictional unemployment.
Static causes of unemployment Controversial distinctions
We have chosen to separate out actual (i.e. observed and quoted in newspapers) unemployment into two parts: equilibrium and cyclical. While this accords well with intuition and current econometric practice, it may be at variance with the recent flow approach to unemployment or with recent developments of the 2 disequilibrium view. The same applies to the 1
Some might argue that the distinction is largely meaningless, i.e. that all unemployment has an involuntary element; for a convincing case, see Lucas (1978) or Pissarides (1989). We take a neutral stand. 2 Two references are C. Pissarides (1989, 1990) and C. Bean and J. Drèze (eds.) (1991). The text presents the flow view in Section 6.4, and part of the disequilibrium view in Section 6.3.
Trade union theory (see Booth, 1995, for a recent review) relies on the distinction between individual labour supply decisions and the outcome of collective bargaining. The trade-union mediated "collective labour supply curve" (wage-offer curve) cannot be to the right of the individual supply curve because trade unions cannot force workers to work more than they wish. It is further to the left the more the trade union values real wages relatively to jobs. A good illustration is to show the effect of increased labour supply (demography, immigration, entry of women into the labour force) as a shift of the individual supply curve -- possibly matched by a shifting labour demand curve as the result of capital accumulation or technological change. If the trade
Instructor’s Guide
-
Chapter 6
union does not change its wage offer schedule, 3 involuntary unemployment can increase. efficiency wages (see Katz, 1986, for a survey) can be introduced to justify rigid real wages. minimum wages is a straightforward example of wage rigidity. regulations and taxes may be represented as drawing a wedge between supply (both individual and collective) and demand: they both increase the cost of labour to the firm without raising net aftertax workers’ income. If net after-tax real wages are shown on the vertical axis, the demand curve shifts down: employment declines as real wages fall.
Facts and institutions
There is hardly any other branch of macroeconomics which is so intertwined with local institutions and traditions. The text emphasises this point in various ways: the choice of topics (the flow approach is highly tied to institutional aspects including benefits), discussions of the effects of national institutions (e.g. collective bargaining structures), and by examples (contrasts between European and US labour markets is a natural way of illustrating the issue). Teachers profit from drawing on their own national experiences for alternative examples.
Static causes of unemployment Numbers
The dynamic considerations of Section 6.4 are not easily cast in the demand-supply framework. This is what may make this part hard to convey in class. Table 6.5 which presents unemployment flows, as well as the standard diagram in Fig. 6.18, signal the change of approach. These flows include those who are fired or whose firms go bankrupt (more important in Europe), as well as those who enter unemployment from the labour force or quit into unemployment (less important). The magnitudes shown in the table are convincing evidence that gross movements are not trivial, and are part of the mechanism by which the stock relationships, which form the core of the analysis, are maintained. For more evidence in the European context see our paper (Burda and Wyplosz 1994).
Students want to know how high equilibrium unemployment actually is (so do a lot of policymakers!). There are few good estimates around, unfortunately, and those that exist do not always coincide. References are the studies in Bean and Drèze (1991) or various studies by the OECD and IMF (which are really estimates of the NAIRU studied in Chapter 12). Table 6.8 produces some unpublished OECD estimates which, as always, should be sold as estimates surrounded by the usual bands of statistical imprecision.
Economics and politics
Markets are not perfect and economists must deal with that as best they can. This is especially true for labour markets. There is however a serious risk that students, frustrated by the persistence of high unemployment in Europe, will react to the material with sweeping conclusions: ban or restrict trade unions, or abolish minimum wages, slash unemployment benefits, etc. It is the teacher's responsibility to remind them that economics is just one component of a larger social game. Indeed, such conclusions can make sense from a narrow economic viewpoint (efficiency wages, for example). But political science and sociology also have much to say about the unemployment phenomenon, and they may contradict economic reasoning, which means that civil order and social harmony have a price. In the end, we economists can even explain why economic principles should not be followed too closely!
Off-the-curve equilibria
Many interesting results occur when either workers are off their individual (or even collective) supply curves or firms are off their demand curves. This is one way of capturing the popular wisdom that unemployment is painful and that firms suffer because of redundancies or unfilled vacancies. Bargaining models in which neither firms nor workers are on their demand and supply curves are not discussed in the text but may be worth exploring (for a review see Booth, 1995).
3
Andrew Oswald has rightfully pointed out that monopolists do not have "supply curves"; by calling the outcome a "collective labor supply curve" we try to avoid pinning the mechanism on a monopoly union. The term "wage offer curve" used in the first edition may be preferred by the purists. It should also be that the slope of the curve will generally depend on the nature of the shock; only if everything is linear will all shifts to labor demand result in the same collective labor supply schedule.
14
Instructor’s Guide
Chapter 6
References
Bean, Charles and Drèze, Jacques (1991) Unemployment in Europe, Cambridge, Mass., MIT Press. Booth, Alison (1995) The Economics of the Trade Union, Cambridge, UK: Cambridge University Press. Burda, Michael and Wyplosz, Charles (1994) "Gross Worker and Job Flows in Europe," European Economic Review, 38:1287-1315. Katz, Lawrence (1986), 'Efficiency Wage Theories: A Partial Evaluation', NBER Macroeconomics Annual, 1: 235-76. Lucas, Robert E. Jr. (1978) 'Unemployment Policy,' American Economic Review Papers and Proceedings, 68: 353-357. Pissarides, Christopher (1989), 'Unemployment and Macroeconomics', Economica, 56: 1-14. ____________ (1990) Equilibrium Unemployment , London, Basil Blackwell.
15
CHAPTER 7
THE REAL EXCHANGE RATE
Objectives
So far it has been implicitly assumed that there is just one good in the world; this chapter introduces a second. This step is necessary to give content to the question: what is the role of the real exchange rate -- an intratemporal price -- for an open macroeconomy? It also allows us to deal with a number of ideas and results normally overlooked in textbooks which mostly focus on the closed economy: why do price levels differ across countries? What are the terms of trade? Could there be a link between growth and inflation (the Balassa-Samuelson effect)? Because of its central importance in the open economy, Chapter 7 has been modified and updated in a number of ways. The real exchange rate is used so frequently that, we now begin Section 7.2 with a thorough discussion of both the concept and its practical implementation and measurement. We proceed then to motivate heuristically the primary current account function: how the real exchange rate -still loosely defined as the price of foreign goods in terms of domestic goods -- positively influences the current account (surplus). The use of the notation PCA(λ,...) is meant to signal both that everything else is held constant and that more is to come. As before, we firmly establish that the exchange rate is measured in European terms (how many units of domestic currency or goods are required to purchase one unit of foreign currency or goods). Next we take one way of looking at the real exchange rate and explore it more deeply, namely the real exchange rate as the relative price of traded goods in terms of nontraded goods. Users of the first edition will notice the shift in emphasis away from a second definition used more extensively in the first edition, namely the relative price of imports in terms of 1 exports.
Following the general strategy of first anchoring the long run, Section 7.4 derives the equilibrium real exchange rate as that necessary to balance the intertemporal budget constraint. It begins with the observation that a country’s intertemporal budget constraint imposes a steady-state restriction on the primary current account and suggests that one mechanism by which the intertemporal budget constraint is obeyed is via reallocation of productive resources towards goods that can be exported (rather than a reduction of absorption). Following this line, the long-run equilibrium or fundamental real exchange rate is defined as the one which balances intertemporal trade. Two examples of this approach in the literature for infinite horizon models are Sachs (1982) and Dornbusch (1983) (although the latter addresses somewhat different issues); a more recent application to an interesting historical episode which stresses the exports/imports distinction is Wyplosz (1991).
Position of the chapter
This chapter is placed at the end of the sections dealing with the real economy. As such it can be seen as an extension of what precedes it, and is consistent with our treatment of the real economy, microeconomic foundations, and intertemporal budget constraints. It may, however, be taken up at different stages in a course. For example, it could as well fit between Chapters 18 and 19 in a course on international monetary economics. The order of topics could be: the exchange rate: institutions and markets (Chapter 18), the exchange rate in the long run (this chapter), and the exchange rate in the short run (Chapter 19). In that case, before starting Chapter 11 and the open economy treatment of the IS - LM model, the teacher should remember that Chapter 7 defines the nominal and real exchange rate in section 7.2.1.
1
Feedback from users signalled difficulties with the notion of "exports" versus "exportables" so we ended up putting the exports/imports distinction into a Box 7.3. This frees up
teaching time to focus on difficult issues raised in Section 7.4.
Instructor’s Guide
Chapter 7
Dornbusch, Rudiger (1983), ’Real Interest Rates, Home Goods, and Optimal External Borrowing,’ Journal of Political Economy, 91: 141-53.
The long-run budget constraint and the equilibrium real exchange rate
Sachs, Jeffrey D. (1982), ’The Current Account in the Macroeconomic Adjustment Process, ’ Scandinavian Journal of Economics, 84: 147-59.
Sections 7.4 conveys a simple message already emphasised in Section 3.6: in the long run the primary current account ceases to be a choice variable. In practice, because most developed countries’ net external positions are relatively small the steady-state primary current account is close to balance (the highly indebted countries reached a debt at the peak of about 40% of their GDP implying a debt service to GDP ratio of 2 about 5%). Students are often surprised, even sceptical, when presented with this conclusion. Fig. 7.1 shows that Germany can run current account deficits -- data f or the 1990s show that this can repeat itself -- while Italy can also have surpluses! Additional long term data can be found in Maddison (see references in the textbook). We find it easy to convince the students that ’in the long run, on average, the primary current account must be small’. The novelty is that a new relative price - the real exchange rate - supplies an economy with an additional degree of freedom for meeting the national intertemporal budget constraint. A real appreciation reduces the production (and consumption) of 3 nontradable goods in favour of tradables. The real exchange rate thereby becomes endogenous, and the teacher can follow up with Section 7.4 and the conclusion that ’in the long run, on average, the real exchange rate or "competitiveness" must be such that the primary current-account imbalance is small enough’, hence motivating the equilibrium real exchange rate. The rewritten version of Chapter 7 stresses this even more by postponing the discussion of the equilibrium real exchange rate until the end.
Wyplosz, Charles (1991), ’A Note on Exchange Rate Effects of German Monetary Union,’ Weltwirtschaftliches Archiv, 127: 1-17.
References
2
In addition, growing countries can afford even lower current-account imbalances if the objective is to stabilise the net external position as a proportion of GDP. 3 Arguing from the perspective of Box 7.3, it could also involve a shift in resources away from the production of importables towards exportables, as well as a decrease in the consumption of importables.
17
CHAPTER 8
MONEY AND THE DEMAND FOR MONEY
Objectives
This chapter is standard. It reviews the definitions and functions of money and prepares for the next chapter with a presentation of consolidated balance sheets (Fig. 8.1). The money-demand function is not derived from first principles: it is simply presented and motivated by the transactions approach, recognising the dependence of demand on opportunity cost of holding money, the 1 nominal interest rate. The Appendix derives the inventory theory of money demand in the tradition of Baumol and Tobin. The chapter is somewhat innovative in two directions, buttressing an otherwise descriptive and institutional chapter with central analytical results. First, the chapter discusses money-market equilibrium, assuming an exogenously set real money supply (Section 8.6). The student is thus exposed early on to 2 the equilibrating role of the interest rate. Second, a long run interpretation locks in the principle of homogeneity of degree 1 in nominal magnitudes, here between money, prices and the nominal exchange rates. It also provides the first opportunity to present the Fisher principle. In contrast to the first edition, the second edition postpones discussion of the concepts of dichotomy and monetary neutrality to Chapter 10, in which all major markets of the economy can be considered simultaneously.
What is highlighted and what is de-emphasised
1
More formal models (cash-in-advance or money-in-theutility function) would increased the level of complexity well beyond that of an introductory text. See Blanchard and Fischer’s (1989) textbook for a nice derivation of the most useful approaches. 2 The assumption is, of course, that the central bank can fix the real money supply. This cannot be true in a world with flexible prices, as Section 8.7 makes clear, so an instructor must either assume a given (but perhaps not fixed) price level, or derive the demand for money in nominal terms.
We emphasise a number of well-known ideas and results that do not always get attention in macro textbooks. First, we stress that money has aspects of a public good -- namely its acceptability in transactions vis-à-vis unknown or unpredictable trading partners. Furthermore, an individual’s use of money makes it more valuable for others by increasing this acceptability. Confidence sustains the value of money, and the lack of it can undermine the stability of banking systems. Second, care must be taken to distinguish carefully between nominal and real variables affecting money demand. Third, the nature of market clearing depends on the time horizon. In the short run (given a price level), real balances are given so the nominal interest rate clears the money market. In the long run, the price level and the inflation rate are endogenous, and are determined by the level of real activity and the rate of growth of the money supply chosen by the monetary authority. On the other side, we pay relatively little attention to some features which often figure prominently in other textbooks. First, we gloss over details concerning monetary aggregates and institutions. Although more information is provided in Chapter 9, we do not spend much time on the definitions of money because they vary from country to country and even over time within a particular country. Teachers may want to elaborate these aspects using a supplementary, perhaps nationalbased text. Showing and explaining a central-bank balance sheet is a good use of classroom time. Second, we do not spend much energy on the concept of velocity. While a number of teachers use this concept, we feel it easily confuses students, who tend to give it a life of its own. We prefer to emphasise the fact that it is nothing more than a transformation of the demand for money and stress the role of nominal interest rates and 3 conversion costs.
3
Velocity is shown in (8.2) to be a function of real GDP, the interest rate, and the technology of money or liquidity services (captured by parameter c).
Instructor’s Guide
Chapter 8
The effects of price on money demand
Students often find it difficult to separate out the two effects of price increases on money demand. Fig. 8.3 attempts to clarify the issue. The first is a level effect which works through the fact that money demand is a demand for real cash balances. Nominal money demand can be written as M =PL: ceteris paribus, there is a one-to-one effect of an increase in the price level on nominal money demand or, equivalently, no effect on real money demand. The second effect is the rate of growth effect . The Fisher effect implies that an increase in the inflation rate increases the nominal interest rate by the same amount; real money demand declines because the cost of holding money increases. Nominal money demand increases over time, but not quite as fast as the price level, so real balances decline. The examples shown in Fig. 8.12 may seem to weaken the latter argument: it is a good occasion to remind students that the Fisher principle involves expected ( ex ante), not actual (ex post ) inflation. Such a discussion serves as a lead-in for future discussions on the credibility of monetary policy.
References
Blanchard, Olivier J. and Fischer , Stanley (1989), Lectures in Macroeconomics (Cambridge, Mass.: MIT Press), 4, esp. 4.1-4.3. Goldfeld, Stephen M. (1990), ’The Demand for Money’ in Friedman and Hahn, eds., Handbook of Monetary Economics (Amsterdam: North Holland).
19
CHAPTER 9
THE SUPPLY OF MONEY AND MONETARY POLICY
Objectives
There are three good reasons for dealing with money supply after money demand and money market equilibrium. First, an understanding of the equilibrating role of the interest rate in the money market greatly helps the exposition of open-market operations. Second, openness and international capital movements profoundly affect monetary policy via the interest rate. Third, if the mechanics of money supply gets relatively less emphasis, it is because monetary policy in most European countries emphasises interest rate or exchange rate policy as well as bank regulatory aspects. Indeed, a key difficulty of presenting monetary policy to students living in an open economy is that they know that foreign interest rates are the central constraint. This is why this chapter de-emphasises the money multiplier and stresses the linkage between monetary policy and exchange-rate policy, establishing the link between a money-market intervention and an exchange-market intervention. While a full resolution of this issue will have to wait until Chapter 11 and the open-economy version of the IS - LM model, students are ready to think about these issues. Another aspect of monetary policy often ’saved for later’ receives here a special early treatment: monetary financing of budget deficits and independence of the central bank. This issue is paramount in several European countries and has been given a central treatment in the discussions surrounding monetary union in Europe. Presenting the idea early on is advisable, despite details provided later in Chapter 15. Overall, the instruments available to the monetary authorities can be presented quickly in a summary form. Similarly, teachers short of time may skip the balance-sheet approach which is very useful but timeconsuming. One important change in this edition is a rewriting of the parts of Section 9.3 to reflect the increasing importance of open market operations in European monetary policy -- as well as the prospect of a European Central Bank by the year 2000. We see this bank -- to the extent it becomes a reality -- as operating
in a manner similar to the Bundesbank and have added Boxes 9.3 and 9.4 to highlight these issues.
Different national institutions, yet the same process
Because monetary institutions differ from country to country, the chapter adopts a neutral presentation of the money-supply process: it does not go into national 1 specifics and instead focuses on common elements. For instance, reserve-holding behaviour is central to legal reserve requirements, yet reserve holding may or may not be regulated in the particular country under study. Similarly, the relationship between central and commercial banks also differs across countries, and this has important policy implications. For example, Fig. 9.7 studies three approaches: i) a strict monetary control as practised in the US in the early 1980s; ii) strict interest rate control as currently practised in many smaller European countries as well as the UK and France; iii) the newer hybrid approach targeting a short term interest rate with open market policy, either via auctions of reserves (via repos as in Germany) or dealings in Treasury securities (as in the US). Yet, it is useful to emphasise the crucial role of bank reserves, be they voluntary or regulated. Again, using the balance sheets introduced in the previous chapter can be an effective way of presenting in one step the two main aspects of money supply: the moneymultiplier process and the effective control of money supply. In doing so, the text introduces two multipliers: the reserves multiplier (the ratio of M1 or any other money aggregate to banks reserves) in Section 9.2.2.1, and the monetary-base multiplier (the ratio of M1 or any other money aggregate to the monetary base M0) in Section 9.2.2.2. In our view the latter is a more effective, and
1
Those interested in delving into national detail might well consider a historical approach which outlines how the institutions arose. For an excellent treatment, see Goodhart (1990).
Instructor’s Guide
Chapter 9
less misleading, way of presenting the two concepts of 2 multiplier and control.
their capital base to the levels required by the ratios; second, with the fall in house prices, a number of bank customers have failed to keep up payments on their loans leaving banks with collateral -- such as houses -which are worth less than their book values, thus 4 eroding the asset side of the bank ’s balance sheets. The other example is the issue of lender of last resort treated in Section 9.6.2. There is no agreement on whether or how this function should be performed by a future European Central Bank. The German Bundesbank seems to oppose making commitments for fear of entering into a commitment to create money. The Bank of England is less reticent, probably having been more sensitised to the risk of instability in its (much more developed) financial markets. This is great material for classroom discussions or examination questions.
Step-by-step multiplication versus aggregate effect
Section 9.2.3 deals in great detail with the money multiplier. In principle, the one-step aggregate effect presented earlier should suffice in terms of macroeconomics. Yet students often correctly suspect that there is more going on among banks. The step-bystep presentation fills that gap. Yet it can be quite timeconsuming in class, and impossible to stop once started. One way of dealing with this is simply to refer curious students to the full treatment in the text and stick to the one-step approach represented in Fig. 9.2.
Policy dilemmas
Students often note that, according to the principles, central banks should be able to control money supply effectively and yet they often miss their pre-announced targets. This provides a good lead-in: teachers should not be afraid of telling students that central banks may well have several objectives: long-run price stability which calls for the control of money growth; a shorterrun preoccupation with the business cycle which can be addressed by varying the interest rate; and very shortrun exchange-rate targets -- even under a flexible exchange rate regime central banks care about the value of the currency. More often than not, these objectives stand in conflict with each other, creating a dilemma for the monetary authorities and leading to compromises. Several examples of this problem are presented in the text (a more recent example is monetary policy in Germany after unification and the UK after the EMS crisis of September 1992).
References
Brunner, Karl and Meltzer, Allan (1990), ’Money Supply’ in Friedman and Hahn, eds., Handbook of Monetary Economics, (Amsterdam: North Holland). Goodhart, Charles A., (1990), The Evolution of Central Banks (Cambridge, Mass., MIT Press).
Regulations
While not traditionally presented in macroeconomic texts, the regulatory aspects of central banks merit some attention, especially in Europe. Two examples might convince the sceptical teacher. The ’Cooke Ratios’ presented in Section 9.6.3 are widely believed to have been the main source of monetary stringency in the early 1990s, in the USA, Europe, and Japan. Two reasons have been presented and may be discussed in class with the help of press reports, especially in Japan 3 and France. First, it takes time for banks to build up
2 3
4
The recent collapse of Barings and enormous reported trading losses of Metallgesellschaft and Sumitomo have important implications for the banking sector. Students may need to be told that banks are virtually always involved when businesses suffer losses or go bankrupt.
A useful reference here is Brunner and Meltzer (1990). See also Theoretical Exercise 9.
21
CHAPTER 10
OUTPUT, EMPLOYMENT, AND PRICES
Objectives
This chapter stitches together the patchwork of the previous nine chapters into the general equilibrium framework most commonly used in macroeconomics. The centerpiece of the chapter is the so-called neoclassical model, which assumes perfectly flexible nominal prices and fixed output, given factor endowments. It is the product of the first three sections, culminating in Figure 10.6. In an effort to be balanced, we conclude the chapter with the fixed price, variable output version of the model, which can be understood using the same IS and LM curves developed in preceding sections and is meant to provide a stepping stone for the Mundell-Fleming analysis of Chapter 11 (previously Chapter 10). Key concepts introduced in this chapter are: macroeconomic general equilibrium (an extension to the discussion of Chapter 5); the IS-LM diagram, which is presented for the first time; the concepts of monetary neutrality and dichotomy; and the Keynesian assumption and its implications.
The return of the closed economy
A number of users of the f irst edition will be relieved at the addition of this chapter. Initially, we were confident that deriving the open economy IS - LM model from the outset would be a straightforward exercise. In the end, many users as well as our own experience convinced us that the closed economy version was necessary, not because the closed economy is particularly relevant (especially in the European context) but because it is the most suitable means of drawing together the loose ends of the previous chapters. (Chapter 11 retains the emphasis of the Mundell-Fleming open economy model and the regime-dependency of fiscal and monetary policy.) We thought that students deserved a last look at the long run, stressing the value of the classical framework for understanding long run trends, while continuing to think about these issues within the confines of the two-period model.
One interesting application of the closed economy model is to motivate more fully the long run determination of the world real interest rate. Figure 10.7 (Box 10.1) adds flesh to the bones of the first attempt in Figure 5.4 and shows how "crowding out" can occur in a purely classical framework, even if there is Ricardian equivalence, when the government increases its claim on resources. This may be supplemented by analyses of related changes, such as an increase in the economy’s endowment today, which would tend to decrease the interest rate in equilibrium, versus a "technology shock" (an increase in marginal productivity at any particular capital stock), which would tend to increase the interest rate.
Structure
Although the analysis is classical as in Patinkin (1948) or Sargent (1987: Chapter 1), it begins with the traditional "Keynesian cross" or 45° line diagram used to interpret demand-determined cyclical fluctuations and in deriving the IS schedule. Next comes the derivation of the IS and LM schedules. They are derived ceteris paribus, i.e. without reference to other constraints on output and interest rates. Derivation of the aggregate supply side occurs in a natural way and provides the first explicit link between the labour market (Chapter 6) and equilibrium output (which figures importantly in Chapter 12 and 13). Both IS and LM schedules are derived in the standard way, but with the exception of Figure 10.12 they are not "moved around"; that chore is left for Chapter 11. In this sense Chapter 10 is a treatment of the theoretical issues, and should be understood by the student as a preparation for the "action" in subsequent chapters. Finally these curves are integrated in a six-panel diagram that will be familiar to some and new to many. We think it will provide a crystallisation point for the many ideas of previous chapters -- in a single picture. This diagram allows the instructor to illustrate the key
Instructor’s Guide
Chapter 10
propositions of monetary neutrality and dichotomy, as well as the failure of these when prices are not fully flexible.
few of the short cuts we have taken in writing Chapter 10 which may turn into stumbling blocks down the road: Two periods or one? This is the last time the two-period framework is mentioned in any significant fashion. The chapter completes the transition to the short-run, suppressing subscripts and treating the entire analysis as if it were "first period." This is a transition also for the next chapter, which is exclusively concerned with the (Keynesian) short run. Observant students will ask: what happens if the potential output of the economy changes? Clearly, teachers can finesse the point by focusing on temporary changes -- those occurring in with no long-term implications for wealth, investment, and output. Longer run changes bring us back to Chapter 5.
Equilibrium
A central concept in this chapter is the idea of macroeconomic equilibrium. Equilibrium is introduced early on as a state in which no forces exist which would move the economy away from that state. Of course, equilibrium is always defined with reference to a choice of exogenous and endogenous variables and we have tried to use the development of the chapter to convey this distinction. One example is the Keynesian cross diagram, which we decided to leave in for those who like to emphasise it. Given the nominal interest rate i, the desired demand curve can be thought of as in equilibrium with output, assuming that output is supplied elastically. Similarly, the IS and LM curves intersect to give a level of output and interest rates given that this is supplied -- leaving open either a classical or Keynesian interpretation.
"The fourth market." The classical model is usually thought of in terms of four markets: goods, labour, money and "bonds" -- where bonds mean all interest-bearing alternatives to money. The bond market is usually ignored (an application of Walras’ Law for a system of asset demands in equilibrium), and this may cause some confusion among better students. It is important to stress that bonds and money are markets for stocks -- as opposed to markets for flows (goods and labour services both today and tomorrow). The short cut is invalid for the flow markets: since investment is occurring, there are goods tomorrow and labour tomorrow as well. Difficult stuff best left for a graduate level course.
Deriving the IS and LM schedules
Even when derived mathematically as in the Appendix, teachers should reinforce students’ intuition for what the IS and LM curves stand for, to prevent their being used mechanically, and possibly incorrectly. We have found that it is best to proceed in a heuristic manner initially. The text performs the ’what if ’ exercise to derive the slopes of the schedule (going from A to C to B in Fig. 10.2 and 10.3). Second, ask what off-schedule 1 points represent and how equilibrium can be restored (both Y and i can move and do the job; explain why and how). Chapter 11 provides detailed information about the slopes of the schedules, so this may be left out at the time of the first presentation: the whole apparatus is often hard to grasp and students should be led to focus on the essentials. Thus, asking what happens as one moves along each schedule requires a more extensive development of the components of aggregate demand (especially the primary current account, which is not addressed in Chapter 10).
IS-LM, aggregate demand and equilibrium. Strictly speaking, our presentation of the IS and LM curves yields the level of aggregate demand, but in Figure 10.6 we draw the demand for money defined at the equilibrium level of output. In that sense the money demand curve in the upper right panel of Figure 10.6 does not shift when the economy is out of equilibrium in the flexible price case.
The neoclassical assumption: price flexibility
While not particularly plausible in the short run, the idea that price-flexibility defines the long run and is of central importance in macroeconomics. In the model presented in this textbook, the assumption of price flexibility and that all nominal prices are indexed to the price level delivers the neutrality of money and the dichotomy proposition -that the real and nominal sides of the economy do not influence each other in a substantial way. It reinforces the intuition that, in the long run, increases in real
A few stumbling blocks and helpful simplifications
Perceptive students always find problems with any instructor’s presentation, and we have written down a 1
This is the object of Exercise 3 in the theory section.
23
Instructor’s Guide
Chapter 10
output are only possible if aggregate supply makes it possible. This point can be made effectively using Figure 10.9.
The Keynesian assumption
towards the more general AS-AD without any particular policy usefulness of its own.
References
It is important to warn the students of the problematic nature of the fixed-price assumption that underlies the IS - LM analysis. One idea is to announce that this chapter deals with the very short-run (a year, say), while the next one introduces inflation covering longer horizons, all the way to the dichotomised long-run. Another is to treat it exclusively as a stepping stone
24
Patinkin, Don (1948) "Price Flexibility and Full Employment," American Economic Review, 38: 54364. Sargent, Thomas J. (1987), Macroeconomic Theory, New York: Academic Press, Ch. 1.
CHAPTER 11
AGGREGATE DEMAND AND OUTPUT
Instructors familiar with the previous edition will notice that the PCA function, previously introduced in Chapter 4, has now been moved to the present chapter. The PCA function is inherently “Keynesian” and is naturally introduced here. It is based on the traditional import function. For the sake of continuity with the microfoundations, we start by modelling imports as a function of total absorption, and then link up absorption to the GDP. The end result is in fact the Keynesian marginal propensity to import --although we don’t call it that in order to avoid “stuffing” students with concepts which are not really essential.
Objectives
Following the closed economy IS-LM model, the present chapter moves to its open economy version, the so-called Mundell-Fleming model. The flexible price version is abandoned, i.e. the focus is decisively on the short-run. The opening-up involves two steps: introducing the trade linkage (via the primary current function) and the international financial link (via a simplified version of the interest parity condition, further elaborated upon in Chapter 19). The main goal is to derive the benchmark “Mundell-Fleming table” which depicts up the policy ineffectiveness results. The second goal is to use this table to help students think about intermediate cases.
The open economy IS and LM schedules
The IS and LM schedules have already been introduced in Chapter 10. They are derived again here. Repetition is not only superficial, however. A second, slower, exposition of this essential tool of macroeconomic analysis has pedagogical merit. In addition, the open economy versions differ from their closed economy counterparts; this justifies a careful treatment. For mathematically-oriented courses, much can be learnt by contrasting the formulae in the Appendices to both chapters. Another advantage of this new exposition is that here we take time to insist on the importance of separating out endogenous and exogenous variables. This is made possible as we now adopt the Keynesian assumption of fixed prices. This distinction is used to address one key difficulty that students face at this stage: grasping when to move along the schedules, and when do the schedules shift.
The Mundell-Fleming table
The key difficulty of this chapter is that the exchange rate regime radically affects the working of the model: with full capital mobility monetary policy is ineffective under fixed exchange rates while under flexible rates it is fiscal policy which fails to affect the economy. Consequently, we cannot just “open up” and show how the results are modified. We have to separate out the two polar cases: full flexibility and perfectly credible fixity of exchange rates. These are extreme cases, especially as they additionally assume full capital mobility. The result is Table 11.4, the chapter’s key result, which can be used to: 1) help the students grasp the dizzying variety of sharp results; 2) guide the student to less clear-cut cases (less than perfect capital mobility with an upward-sloping financial integration line), expected de- or revaluations (shift the financial integration line); 3) discuss intermediary regimes (managed float) and the case of monetary union (in fact the permanently fixed exchange rate case).
The financial integration line
Teachers will know that in the Mundell-Fleming model expectations are omitted, or assumed to be static so that i = i* + a constant term. Later chapters, starting with Chapter 13, will introduce more realism and complexity. In a first pass it is essential, we believe, to stick to this simplification, even if some alert students
The PCA function
25
Instructor’s Guide
Chapter 11
suspect that it is not quite right. Going in the direction of introducing endogenous exchange rate expectations is like opening Pandora’s box at this stage. As with the IS and LM schedules, students should understand what is happening off this schedule, why does the economy promptly returns onto the schedule, and what shifts the schedule - including possibly expectations of appreciation or depreciation. The financial integration schedule assumes full capital mobility and substitutability: then, and only then, are all assets perceived as identical. Of course, that is not a very realistic assumption, just a benchmark. Two interesting deviations may be considered. The first case is that of full mobility but imperfect substitutability. Early models in the Mundell-Fleming tradition used to draw an upward sloping schedule to account for less than perfect substitutability. Subsequent research has tried to understand the corresponding "exchange risk premium". It is hardly plausible that an upward sloping schedule makes sense: the premium is believed nowadays to be small and highly variable. This is why we do not follow the older tradition, and instead tell the students to assume full asset substitutability. Again, as a first order of approximation, it is perfectly acceptable. And there is no graphical second order of approximation. The second case of less than full mobility corresponds to the existence of capital controls. These controls - presented in Chapter 19 and 21 - break the link between domestic and foreign interest rates. There is no simple way of representing them graphically, certainly not with an upward sloping schedule. In the advanced countries with developed financial markets, the best approximation remains the horizontal line because it is known that, given time, controls are easily circumvented.
Policy mix
Sections 11.3.3 and 11.4.3 present how a fiscalmonetary policy mix operates. This may seem a little bit like hair splitting. In fact, it is an excellent way of checking students’ understanding. It can easily be left out of a busy classroom schedule.
26
CHAPTER 12
AGGREGATE SUPPLY AND INFLATION
Everything else being equal, lower rates of unemployment are associated with rising inflation. This is the Phillips curve, and it is indeed visible when everything else remains equal. But, in general, everything else does not remain equal. Inflationary expectations -- which enter the determination of core inflation can and do change -- as well as the equilibrium unemployment rate itself, due to factors that often have nothing to do with the b usiness cycle.
Objectives
This chapter derives the aggregate supply curve. This analytical tool remains controversial; while most economists would probably accept the notion of a short-run aggregate supply curve and admit that the schedule becomes vertical in the long run, disagreements persist regarding the microeconomic foundations. This is why we propose a fairly openminded approach: inflation accounting in the tradition of Friedman and Phelps, with an eclectic view with respect to the foundations. More strongly opinionated teachers can choose to rationalise the aggregate supply curve in terms of a specific model: Lucas’s monetarymisperceptions or error-extraction mechanism, the Fischer-Taylor overlapping contracts, staggered price 1 setting, or menu costs, for example. The proposed strategy starts with the following ’fairy-tale’ presentation of the Phillips curve: it used to exist, it disappeared, but may have returned, and in any case theory says it has to end up vertical. The inflation accounting exercise is reasonable, and makes sense of the ’stylised fact’ of a short-run inflation unemployment trade-off which is nonetheless interrupted by detectable breakdowns in the relationship. Most importantly it need not take a stand on the source or nature of the underlying nominal rigidities. This is followed by a rehabilitation of the Phillips curve in its modified form. To move from the unemployment-inflation space to the output-inflation space we use Okun’s law. Okun’s law is often less robust than a law should be -- and Chapter 6 presents many reasons why. Still, we believe that Okun’s law ought to be in every student ’s toolkit. It may be useful to warn students that employment fluctuations typically lag behind those in output.
Price level or inflation?
In constructing the aggregate supply curve, theory often leads us to operate in the output-price space rather than in the output-inflation space. A frequent strategy is to start with the price level and then go to inflation; some textbooks keep the discussion in the levels throughout. We have found that students are puzzled, if not confused, by the change (they often have a hard time distinguishing price levels from the inflation rate), but want to know about inflation, which seems more relevant politically and comparable internationally. Indeed, we must eventually be able to explain inflation because realistic steady states are ones in which prices rise at a constant rate. This is why the textbook moves directly to inflation, at the cost of some short cuts which at first glance may be less appealing to some instructors, but which ultimately allow students to direct their energies to a smaller number of models.
Core inflation
The concept of core inflation is intentionally left somewhat vague in the main text. It is the component of current inflation which is not attributed to capacity utilization, tension in the labour market, or supply disturbances. The term appears frequently in the financial press and seems to have assumed a meaning not too different from ours (close substitutes are underlying or trend inflation, but we steered clear of the term ’inflationary expectations’, the term used in popular macroeconomic textbooks from the last decade). Our loose treatment is partly due to the lack of conclusive empirical evidence on the subject, partly due to the role that nominal wage indexation, collective bargaining, and other national institutions may play in
The Phillips curve as a ’stylised fact’
The Phillips curve is an important component of this chapter, and we recognize it as an important historical statistical regularity. What the students should remember from the chapter comes in two steps:
1
For a catalogue of sources of nominal rigidities which might lead to a less than vertical supply curve, see Blanchard (1990), or Romer (1995).
27
Instructor’s Guide
Chapter 12
its evolution. In the text we have tried to present a story acceptable to the largest number of colleagues; teachers with strong opinions will always stress whatever they 2 believe is the most convincing approach. In developing the concept of core inflation we find it useful and convincing to tell students to think o f what is being bargained over in wage negotiations. Especially in Europe, negotiators explicitly assume a rate of inflation to factor into wage settlements, and they generally end up agreeing on a rather precise figure. Because national institutions differ, it is not always clear how much this figure corresponds to a catch-up on past inflation and how much it represents an attempt at guessing future inflation. Such aspects as the frequency at which wage negotiations take place, whether wages are indexed, and how high and uncertain the inflation rate is, seem to play important roles in determining core inflation. To a large extent, disagreements largely centre around the relative importance of backward- and forward-looking elements 3 in the formation of core inflation.
inflation is needed, so convergence will include points 4 above point Z. None of these complexities are mentioned in the text because we believe that it is far too complicated. Even the appendix to Chapter 13, which presents a formal model, depicts dynamics in a simple fashion. Clever students, however, will often see the need to overshoot point Z during the convergence process.
Which space?
Section 12.6 constructs the aggregate supply curve simultaneously in the two spaces: unemployment inflation and output inflation. Since the chapter starts with the Phillips curve and ends up with the aggregate supply curve -- most naturally represented in the output inflation space -- it is unavoidable that a step be taken at some point from one space to the other. This is time-consuming in class, though. Teachers who are short of time may have to make a choice. Our advice, then, is to skip the Phillips-curve and Okun ’slaw part and build up the aggregate supply curve directly in output-inflation space. In equations (12.7), (12.9), (12.10) and (12.11) this requires replacing the cyclical indicator (U -U ) by (Y -Y ) as in (12.12). All the reasoning goes through with this limited change.
Dynamics
The presentation in Section 12.5.3 of the transition from the short to the long run is intentionally sketchy. What is certain are both the starting point (point A in Fig. 12.11) and the steady state under long-run neutrality of money (point Z). How we move from A to Z depends on the dynamics of the underlying model which is not fully specified here, among other things because the model is incomplete -- the demand side is missing and is introduced in Chapter 13. Even the postpolicy change point (point B) depends on the relative speed of the effects of monetary policy (the movement along the short-run Phillips curve) and of the shifts of the Phillips curve. Most models will predict that the convergence of inflation to point Z over time is not monotonic, for the simple reason that as long as we are below point Z, inflation has not risen by as much as the rate of money growth so the real money supply has increased. Since the real money stock must, in steady state, return to its initial level, a period of higher
References
Blanchard, Olivier J. (1990), ’Why does Money Affect Output: A Survey’, in B. Friedman and F. Hahn, eds., Handbook of Monetary Economics, (Amsterdam: North Holland). Bruno, Michael, and Sachs, Jeffrey D. (1985), The Economics of World Stagflation, (Cambridge, Mass.: Harvard University Press). Romer, David (1995) Advanced Macroeconomics, New York, McGraw-Hill.
2
See Bruno and Sachs (1985) for a detailed discussion of wage and price behaviour and the Phillips curve in OECD countries. 3 Autoregressive inflationary expectations are another potential source of persistence in core inflation, although this idea is not stressed in the text. Judging from the performance of economic forecasters (whose view are often consulted in collective bargaining in Europe), it reasonable in our view to assume weak rational expectations, or that expectational errors are orthogonal to information available when the forecast was made. This puts the blame for non-neutrality of money squarely on nominal price rigidities.
4
Naturally, this is not exactly correct in a growing economy. With a sufficiently rapid real growth rate, steady state real money demand may rise fast enough to eliminate the need for higher inflation.
28
CHAPTER 13 AGGREGATE DEMAND AND AGGREGATE SUPPLY
Objectives
This is the central chapter which provides the synthesis of the macroeconomic model, which in some form or another, is the common basis for dialogue between 1 macroeconomists of whatever direction or school. As is customary, it presents the aggregate demand and supply analysis. A key difference is, in line with Chapter 11, that this chapter deals separately with fixed and flexible exchange rates, by-passing the closed economy and thus differing from presentations found in most macroeconomics textbooks. Having derived the aggregate supply schedule in the previous chapter, the first order of business here is to derive the aggregate demand curve. It is done by introducing inflation into the IS - LM model in a fairly conventional manner. Under fixed exchange rates, nominal money is endogenous and adjusts to inflation so that the money supply remains in line with demand. Inflation has shortrun real effects on aggregate demand because, with a fixed nominal exchange rate, it leads to a real appreciation. Under flexible exchange rates, nominal money growth is exogenous. Higher inflation means a contraction in the real money supply which, through the usual mechanisms, reduces aggregate demand. In both cases the aggregate demand curve is downward sloping. Yet, students should be reminded that the mechanism at work is different. A careful derivation of the demand schedule along the lines sketched above is one way of driving the point home. Another way is to manipulate the AD and AS curves in tandem with the IS - LM system when studying shifts in exogenous real spending or policy variables. This can, however, be a difficult exercise. Those who are interested should refer to theoretical problem 14.3 and the proposed answer.
1
We think that even real business cycle theories can be couched in terms of this framework, although it does not stress the sources of real fluctuations (i.e. productivity, labour supply, or taste shocks). Empirical evidence that demand and supply disturbances are both important for the USA can be found in Gali (1992).
29
Two-step reasoning: the short-run and the long-run anchor
A large component of this chapter is about dynamics, but dynamics is difficult and, by and large, beyond the 2 level of this textbook. To circumvent the difficulty and still allow students to understand the general direction of dynamic adjustment, the treatment focuses on two points in time, the short run and then the steady state. Filling the gap in between is done heuristically through reasoned guess-work. The short-run effect is found by asking whether the disturbance affects the demand side, the supply side, or both. Moving the corresponding schedule(s) accordingly gives the new temporary position. The long run is found through a different reasoning: domestic inflation is determined by foreign inflation under fixed rates and by domestic money growth with flexible rates; output returns to its trend level, which in the meantime may be higher. The dynamics can only be sketched because the full dynamic behaviour is not completely specified but in most cases it involves gradual adjustments of the supply (and possibly demand) schedule until both curves pass through the long-run 3 equilibrium point.
The output gap
In the output inflation space the horizontal axis represents the output gap, not output itself. The reason is practicality: with steady state growth output, the long 2
Graphically it requires the use of phase diagrams; more importantly, it requires a stand to be taken on the source (i.e. wages versus prices) and nature (overlapping contracts, misperceptions cum signal extraction, menu costs) of nominal rigidities. 3 In the notes to the preceding chapter this point is developed. The position of the aggregate demand curve may also depend on output last period, which will complicate the analysis considerably (for those who remember, Dornbusch and Fischer’s (1987) textbook took this tack in a linear framework).
Instructor’s Guide
Chapter 13
run aggregate supply curve would move to the right continuously, rendering the graphical analysis cumbersome and messy. Using the output gap gets around this difficulty but creates a problem of its own. Some disturbances may have a permanent level effect on output, i.e. they change the trend growth path but not the rate. In the output gap inflation space this might go unnoticed. The way to capture this is somewhat complicated; a simpler alternative is to change the horizontal axis back to the level of output rather than the output gap (this is done in Exercise 1 of Chapter 17). Teachers might find a sketch of the solution to be 4 useful. Fig. 1a illustrates a once and for all drop in trend output which then resumes its old growth rate. Actual output is shown to recover its new trend gradually. In Fig. 1b this is shown as a leftward shift of the long-run aggregate supply line and the associated short-run supply schedule AS ’. At the initial point A, as in Fig. 1a output is above its new trend growth path. If, for example, money growth is unchanged, long-run inflation remains the same and the long-run equilibrium must be at point B. The post-shock position is at point C. Convergence will include shifts in both the demand and supply schedules. While the shifts of the supply schedule are familiar, that of the demand schedule is not. What happens though is that wealth (or permanent income) falls (see Fig. 1a) so demand falls accordingly. If the downward adjustment in demand were immediate, the AD schedule would instantaneously move to its new position, passing through point B. The various paths depicted in Fig. 1a are all possible, depending on the dynamics (as well as assumptions about the consumption function and other components of spending).
interpretation in class. This can be accompanied by relevant press clippings. Since the chapter is meant to be a synthesis of previous chapters, teachers should not hesitate to return to earlier results (e.g. emphasising the role of labour markets in determining the speed of adjustment to shocks and possibly trend output). Later chapters are mainly devoted to doing just that but any time devoted to that effect is well spent. An important issue in this regard is largely sidestepped: the empirical slopes of the AD and AS curves. This is clearly important for predicting the outcomes of various shocks and policy interventions. A few estimates exist (see Gali, 1992) but, in principle, simple reasoning can take us a long way. For example, in Japan, the near perfect harmonization of wage bargaining plus extensive profit sharing reduces nominal rigidities originating in labour markets to near zero; those coming from the price-setting side remain, leading to a relatively steep AS curve. In the USA, a semi-open economy in which long-term nominal contracts are important, explicit cost-of-living adjustment is rare and where unions represent a minor factor in wage determination, one would expect the AS curve to be relatively flat (Sachs, 1980). Europe, with annual but overlapping bargaining and a strong tradeunion movement and often institutionalized inflation compensation, is somewhere in between. In more advanced courses, instructors may also wish to stress the endogeneity of the slope of the AS curve: it may depend on the variability of inflation (Lucas, 1973) or the level (Ball et al., 1988).
Using the AS-AD framework
Merely understanding the AS-AD framework is not really sufficient. Since this is the single most important and useful tool to be brought away from a macroeconomics course, it is essential that students recognize how powerful it is for thinking about the real world. The examples provided in Section 13.4 deal with the most obvious cases. The model is also extensively used in the next chapter to provide an interpretation of business cycles. Several exercises are also designed to provide good practice. One recipe for hammering in the usefulness of the AS-AD model when first exposed, is to identify a head-line economic issue at the time of teaching and provide an AS-AD
Figure 1a and 1b from chap. 12 of the Instructor’s Guide, 1/e here
4
It might be used to provide an extension of the treatment of the oil shocks in Section 13.4.1 where this point is intentionally downplayed.
30
Instructor’s Guide
Chapter 13
Dornbusch, Rudiger, and Fischer, Stanley (1987), Macroeconomics, 4th ed. (New York: McGraw-Hill). Gali, Jordi (1992), ’How Well Does the IS-LM Model Fit Postwar US Data?,’ Quarterly Journal of Economics, 107: 709-38. Lucas, Robert E. Jr. (1973), ’Some International Evidence on Output-Inflation Tradeoffs,’ American Economic Review, 63: 326-34.
Interest parity
The financial integration line in the underlying IS-LM model can be troublesome because the interest parity condition involves exchange rate expectations. In the fixed exchange rate case, it is simply assumed that the existing parity is credible so that i=i*. With flexible exchange rates, this is no longer a tenable assumption. The reasoning implicitly assumes that the financial integration line does not move, which is actually incorrect. A full treatment is presented in Chapter 19. As an example for advanced students, consider the case of a monetary expansion under flexible exchange rates as in Section 13.3.3. The rightward shift of the AD schedule in Fig. 13.10 corresponds to a shift of the LM schedule met along a supposedly unchanged financial integration line by a new IS which has moved rightward to reflect the real depreciation. But as inflation rises along the AS schedule (from A to B) it must be the case that the exchange rate depreciated even more than prices rose. Does it not affect the expected rate of depreciation and therefore the financial integration line? One answer is the overshooting result (see Chapter 19 and references therein) which implies an expected appreciation and a downward shift of the financial integration line. This provides the appealing result that the domestic interest rate temporarily falls after a monetary relaxation. Over time, though, the financial integration line rises to reflect the fact that, in the long run, a higher rate of money growth leads to more inflation and a permanently higher rate of depreciation (or a lower rate of appreciation). The underlying IS and LM schedules both shift to the left during the transition to reflect the fact that rising inflation ( AS moves up along the AD schedule) reduces competitiveness and the real money supply.
References
Ball, Lawrence, Mankiw, N.Gregory, and Romer, David (1988), ’The New Keynesian Economics and the Output-Inflation Tradeoff,’ Brookings Papers on Economic Activity, 1: 1-65.
31
Sachs, Jeffrey D. (1980), ’The Changing Cyclical Behavior of Wages and Prices in the United States, 1890-1976,’ American Economic Review, 70: 78-90. Students can also be referred to existing macroeconomic models. Most of them are built around the AS-AD model. Instructors will be familiar with the models in use in their own countries. At the international level, the IMF’s MULTIMOD has the merit of using rational expectations. It is presented in: Masson, Paul (1990) MULTIMOD Mark II: A Revised and Extended Model, IMF Occasional Paper.
CHAPTER 14 BUSINESS CYCLES
Objectives
This chapter is new to the second edition. It presents the student with an exposition of: 1) the main empirical features of business cycles; 2) the role of price stickiness in explaining cyclical fluctuations, thus introducing the Real Business Cycle (RBC) theory. The other chapters mostly attempt to downplay controversies between the various schools of thought. It is impossible to do so for business cycle theories. Instead, we use this chapter to illustrate the importance of assumptions about the degree of price flexibility. Indeed, one way of presenting this chapter is that it provides another exposition of the principles presented in Chapter 13. The AS-AD framework is put to work to “reproduce” business cycles. Its flexible price version is the backbone of our presentation of the RBC. While it is not completely possible to subsume the debate on business cycles to the issue of price flexibility, doing so offers continuity and coherence of exposition. The strategy adopted here parallels that used in the growth chapter (Chapter 5), and the two are complementary as both deal with GDP growth, the former focusing on the trend, the present one with fluctuations around trend. In both cases, we present stylised facts, setting objectives for the theory to be developed. Thus students know what to look for, and they are given a motivation to work through the theoretical material.
The stylised facts
We have resurrected the old Burns-Mitchell diagrams, which had fallen into some sort of disrepute following the attacks from the Cowles Commission in the early post-war period. This approach has enjoyed a revival, recently (Quah (1994), King and Plosser (1994)). Nowadays, we feel, economists are sufficiently well equipped in rigorous statistical methods to study business cycle regularities (e.g. spectral analysis) that there is little risk of being carried away by the naked eye. In any case, the results that we brand as stylised facts are also well-established in the professional
32
literature (see some references at the end of this Chapter. The procedure that we used to construct these diagrams is described in Figure 14.3. Because we use standardised data we only look at period since 1970, which limits the number of completed cycles and may reduce the “generality” of the stylised facts that we wish to reveal. For this reason, we have decided to use averages over several countries. We have not detrended the data mainly because we wish students to see the relative size of the growth trend and of cyclical fluctuations. In the classroom, instructors may present Burns-Mitchell diagrams which include the uncompleted cycle under-way: this allows a discussion about the particularities of the current cycle which is likely to echo familiar debates in the media. While presenting the stylized facts we find it sound to refer to the microeconomic principles developed much earlier (Chapters 2 to 6). At this stage, students often think that “IS-LM is enough” and that the microeconomic foundations have not be much used in the subsequent chapters. The variability of components of aggregate demand offers a nice opportunity to revive interest for microeconomic principles and also to prepare the presentation of the RBC theory.
Deterministic versus stochastic cycles
Students are often told about either deterministic cycles or about standard cycles (as presented in Box 14.1). We take a skeptical view about these rather mechanical views, partly on the basis of modern statistical analysis. The difference should not be overplayed. For example, Kondriateff cycles can be seen as stochastic cycles triggered by major discoveries which turn out to occur at frequencies of 40-60 years. More interesting is the generality of the impulse propagation mechanism. Yet, the presentation is tricky. On the one hand it is important to show students how noise is transformed into cycle-like fluctuations. This will equip them with a healthy skepticism when it comes to explaining month-to-month wiggles. On the other hand, we do not want to give the impression that “anything can be explained away”. Instructors may
Instructor’s Guide
Chapter 14
want to emphasise that macroeconomics is precisely useful because it explains the nature of propagation. We present a few lags (Robertson, Lundberg) as examples of the more general fact that empirical macreoconomic relationship include various lags or various duration. What must be made clear to the student is that it is the presence of lags which makes propagation more than just transmission. This is what the multiplier-accelerator example is designed to show. This is just an example: real life is more complex, of course, much like different wave-lengths interfere to produce a rich visual or sound outcome. Mathematically, we just want to have difference equations to produce dynamics. Wherever possible, instructors should spend some time presenting difference equations, including the possibility of generating cyclical responses. Box 14.2 and Figure 14.10 provide some material to that effect.
underplayed in the IS-LM and AS-AD analyses. Even if they have doubts that RBC theory in and by itself provides a complete interpretation of business cycles, instructors may point out that these mechanisms can be at work in economies with sticky-prices. Another way of presenting the RBC theory is that it asserts that, given exogenous productivity shocks, economic agents are as well-off as they can be, given the circumstances. For this reason, no policy action can improve their inter-temporal welfare: any benefit provided by a temporary policy action will have to be more than paid for later on.
RBC vs. sticky prices
Our intention here is to build up a solid basis for the debate Keynesian vs. Classical economics which appears in Chapter 16 (instructors should be aware that this topic is taken up explicitly later on). We attempt to point out how data can be made the referee of this controversy. This is why we display some additional Burns-Mitchell diagrams at the end of the chapter. Three stylised facts presented there merit attention. First, real money is usually found to be a procyclical leading indicator, which accords with sticky prices models, but not easily with RBC theory. Second, as predicted by the RBC, productivity is pro cyclical, while it is expected to be counter-cyclical in sticky price models. Third, real wages turn out to be acyclical, not procyclical as expected in RBC models or countercyclical as predicted by most Keynesian models.
Using AS-AD
Implicitly at least, the dynamic use of the AS-AD framework involves many lags. The result is a fairly complicated dynamic behaviour, as the Appendix shows. Accordingly, in the main text, we refrain from pretending to show the detailed evolution of the system. We posit the short-run effect, the long-run equilibrium, and sketch a few principles which can guide the analysis of the transition. Students often want to see more at this stage. It is not a good idea to attempt to study the details since the exercise is rather daunting (e.g., whether convergence is oscillatory or not depends on parameter values). Instructors are advised to stick to the few principles that are unambiguous and show how they help broadly map out the system’s evolution. One of us has had enormous success "simulating" live a simple Hicks-Samuelson model on white noise, using a spread-sheet program. On the other hand, for technically advanced courses, instructors may want to show the details: one of the simplest possible AS-AD models is presented in the Appendix and displays quite some richness (e.g. it shows when and why loops occur). Note that this model assumes perfect foresight and sticky prices, unlike the Sargent-Wallace models which rely on imperfect information to obtain deviations from full employment equilibria.
Trends and cycles
For more technically-oriented courses, this chapter provides an opportunity of discussing standard treatments of macroeconomic time series. Here are few ideas of what can be done: - seasonality: time series like private consumption real money (not to mention output in agriculture) typically exhibit seasonality (Christmas shopping dominates). Procedures to deseasonalize include the use of moving averages or the so-called X-11 Census method which removes estimated seasonal factors. - stationarity: students may be presented with the definition of stationarity (stable moments) and with the difference between zero and first-order of integration, and to be told that most macroeconomic time series are I(1). - procedures can be discussed which separate trends from cycles. They are briefly presented in the Appendix.
Real business cycles
There are two ways of looking at the material presented under this heading. One is that it is possible to explain business cycles even with purely flexible prices. Second, it shows how the macroeconomic implications of the principle of intertemporal substitution, largely
33
Instructor’s Guide
Chapter 14
Simkins, Scott P. (1994): “Do Real Business Cycles Really Exhibit Business Cycle Behavior?”, Journal of Monetary Economics, 33: 381-404.
Additional literature
A simple and lucid presentation of the RBC theory is in McCallum (1989). A critical review is in Summers (1986) (see text). King (1995) presents in a clear way the modern methodology of business cycle research. Simkins (1994) offers a critical review. Some of the references provided refer to debates on the role of price rigidities to explain historical episodes, and may be more informative than stylised facts alone. Students often enjoy (re)visiting the Great Depression: two classics are quoted in the text -- Kindleberger (1986) and Temin (1989). Additional readings to suggest include Mankiw (1989) and Bernanke and Parkinson (1991): they deal with the procyclicality of productivity during the Great Depression. Romer (1986) looks at long time series to challenge the view that real GDP has become more stable after World War II.
References
Bernanke, Ben and Parkinson, Martin (1991) “Procyclical Labor Productivity and Competing Theories of Business Cycles: Some Evidence from Interwar US Manufacturing Industries”, Journal of Political Economy, 99: 439-59. King, Robert G. (1995) “Quantitative Theory and Econometrics”, Federal Reserve Bank of Richmond Economic Quarterly, 81: 53-105. King, Robert G. and Plosser, Charles I. (1994) “Real Business Cycles and the Test of the Adelmans ”, Journal of Monetary Economics, 33: 405-38. Mankiw, N. Gregory (1989) “Real Business Cycles: A New-Keynesian Perspective”, Journal of Economic Perspectives, 3: 79-90. McCallum, Bennett (1989) “Real Business Cycles”, in: R. Barro (ed.) Modern Business Cycle Theory, Chicago University Press, Chicago. Quah, Danny (1994) “Measuring Some UK Business Cycles”, unpublished, London School of Economics. Romer, Christina (1986) “Is the Stabilization of the Postwar Economy a Figment of the Data?” American Economic Review, 76(3): 314-34. Sargent, Thomas (1987) “ Macroeconomic Theory”, Academic Press, New York.
34
CHAPTER 15 FISCAL POLICY, DEBT, AND SEIGNIORAGE
material here may be presented in a more cursory fashion.
Objectives
Thus far fiscal policy has been presented rather mechanically. This chapter plugs a number of holes left over by the previous chapters’ preoccupation with reaching the AS-AD synthesis as fast as possible. First, the chapter presents the welfare arguments for public spending. This is not really macroeconomics, so the presentation is brief. Second, the text returns to the early chapters’ focus on optimal intertemporal choices. This is used here to ask when and how the government should use its taxing and spending powers to ease out cyclical fluctuations. Consumption smoothing plays a central role here and leads to tax and public-spending smoothing results. Attention is drawn to the fact that active fiscal policy must be based on market failures and some cases are discussed, e.g. credit rationing or price and wage rigidities. Third, the conventional automatic stabilizers are presented. This allows us to draw attention to the partial endogeneity of budget figures, i.e. the distinction between discretionary and non-discretionary spending. Fourth, the explosive nature of the public-debt process is discussed in depth, since national indebtedness is a perennial favourite of policy-makers and politicians alike. Among other things, this allows a full treatment of the budget constraint in a growing economy and the link between the deficit, seigniorage, and inflation.
Use of the material
This chapter is a first opportunity to draw together a number of results from previous chapters. It also offers background for discussing current issues. Now that the students understand the broad picture (the AS-AD framework) they can fully appreciate debates about fiscal policies, and press clippings may profitably be distributed and discussed in class or in take-home assignments. Of course, in shorter courses, most of the
35
Welfare
Because debates on the size and role of the government involvement are so highly politicized, the text adopts a fairly narrow economic point of view. Using comparative data, as in Tables 15.2, 15.3 and 15.9, is often a good way of escaping parochial debates. More comparative data on budgetary issues are available in OECD publications and provide the basis for interesting class discussions.
Public debt
In principle, what matters are levels of net public debt. However, in Table 15.6 we present gross debts (as a percentage of GDP). Available net debt figures (e.g. from the OECD, like the gross figures that we use) take into account public holdings of state-owned firms and other commercial properties. Our choice is due to the serious limitations of net figures. First, the valuation of state properties is highly arbitrary (e.g. state owned firms are not priced on stock markets). Second, some state assets include loans which may never be repaid in full. Third, these estimates overlook a number of potential and important assets and liabilities. For example, current retirement legislation implies future pensions which will be quickly rising when the baby boom generation born after World War II reaches retirement age: these are unmeasured liabilities (Table 15.4 elaborates on this point). Probably in recognition of the limits of net debt figures, the Maastricht treaty 1 sets limits on gross, not net national debt.
Debt dynamics
1
See Buiter et al. (1993) for details on the debt ceiling and the debate over its necessity in a future European Monetary Union.
Instructor’s Guide
Chapter 15
The material in Section 15.4 is most easily presented using mathematics, as in the Appendix. The text attempts to avoid maths as much as possible and is written to be comprehensible to students with no proper maths background. Teachers may want to check that the technical difficulties are not a barrier to understanding. An intuitive way of making the point on debt stabilization is to describe the evolution of the debtGDP ratio as a race between the numerator (growing at the rate of the public sector borrowing requirement divided by the current debt stock) and the denominator (growing at the GDP growth rate). When the primary budget is in balance, the numerator growth rate is determined by debt service, i.e. by the interest rate 2 times the debt level. This is why it is essential to compare the interest rate and the GDP growth rate: either the nominal interest rate versus the nominal GDP growth rate, or the real interest rate versus the real GDP growth rate.
International Evidence and the Swedish Experience”, Swedish Economic Policy Review . The May 1996 issue of IMF’s bi-annual publication, World Economic Outlook , focuses on fiscal policy. It includes a wealth of analyses and data which provide excellent classroom material. See also the June 1996 issue (No. 59) of the OECD Economic Outlook .
Stabilising debts
European public debts have risen to high levels over the 1980s, an evolution unheard of in peacetime. This is one reason why nearly everywhere debt reduction has become the main objective of economic policy. The priority given to debt reduction is further reinforced by the Maastricht Treaty (which is fully described in Chapter 21). This has led to a new economics of fiscal stabilization. Instructors who wish to develop this question can use two recent articles. Alesina and Perotti (1995) show which measures work (cutting spending on public employment) and which ones fail (cutting public investment). Giavazzi and Pagano (1996) show that in some cases fiscal retrenchment can be expansionary (a strongly non-Keynesian effect), as may have been the case in Ireland and Denmark in the late 1980s.
References
Alesina, Alberto and Perotti, Roberto (1995) ‘Fiscal Adjustments: Fiscal Expansions and Adjustments in OECD Countries’, Economic Policy, 21:205-248. Buiter, Willem H., Corsetti, Giancarlo, and Roubini, Nouriel (1993), 'Excessive Deficits: Sense and Nonsense in the Treaty of Maastricht,' Economic Policy, 16: 57-100. Giavazzi, Francesco and Pagano, Marco (1996) “NonKeynesian Effects of Fiscal Policy Changes: 2
Of course, in reality the national debt is rarely financed at the current short term interest rate, but rather at a whole spectrum of short and long maturities.
36
CHAPTER 16 THE LIMITS OF DEMAND MANAGEMENT link with policy-making which can be easily shown with the AS-AD apparatus, and which relates directly to the issue of voluntary versus involuntary unemployment. Finally, it also provides a lead into another inevitable and related topic: the costs of inflation.
Objectives
This chapter is meant to be fun for both students and instructors alike. It reviews old and highly popular material -- the inevitable discussion of Keynesians versus Monetarists -- as well as going over more recent and exciting research on the interaction between expectations and policy. Like Chapter 15, it provides many opportunities to apply results established in earlier chapters and to discuss current issues. The main purpose of this chapter is to continue the task of the previous one, namely to convey some of the more subtle aspects of macroeconomic policy and to remove some of the optimism conveyed by the AS-AD framework of Chapter 13. Chapter 15 hinted that fiscal policy is not as effective, even in the short run, as suggested by the AS-AD framework. Chapter 16 is intended to amplify this scepticism to government demand policy in general.
Time and expectations
Section 16.3 presents the results which have so profoundly changed the field of macroeconomic policy in the late 1970s. They all hinge on the importance for policy of expectations, and how policy and expectations affect each other. This leads to the concepts of reputation and credibility. The general implication is that the effectiveness of policy (demand management) is further constrained by the fact that policies are predictable; once they are predicted, they lose some or all of their effectiveness. Students usually find it intuitive, and enjoy the idea that the public tries to out-guess the authorities (and has an interest in doing so!) and that the authorities, having once lost their ability to surprise the public, are better off bound by rules.
The great debate
No macroeconomics course can be complete without some discussion of the on-going debate between Keynesians and Monetarists. In the text, so far, this debate has been suppressed because we believe that students should not be told that "economists disagree about everything" and therefore know nothing. This is both untrue (we hope) and highly demotivating for the students (we know). We also think that classes should first be presented with a framework that they can understand and then shown which of its components are controversial. In presenting the debate we have intentionally suppressed discussion of issues which, we believe, have been resolved: the slopes of the IS and LM curves, the liquidity trap, whether or not the slope of the long-run Phillips curve is vertical. Rather we have cast the debate in terms of the desirability of government interventions. As many have pointed out, the central issue is the rigidity of nominal wages and prices. Assuming the foundation of Chapters 6, 12, and 13 have been well laid, this will be relatively easy for the student to grasp. In addition the chapter establishes a
Political economy
Recent work has reduced the gap between economics and political science. Technically, policy actions used to be considered as the archetypal exogenous variables. The new literature endogenizes the behaviour of policymakers -- by noting that they merely respond to political conditions which are themselves shaped by economics. Put in an extreme form, policy-making is reactive, not active. Because the literature is still in its infancy and shaped by US institutions we give only a 1 brief summary.
References
Alesina, Alberto (1988), ’Macroeconomics and Politics’, NBER Macroeconomics Annual, 3: 13-62. 1
Readers interested in more details can see the surveys in Nordhaus (1989) and Alesina (1988).
37
Nordhaus, William (1989), ’Alternative Approaches to the Political Business Cycle,’ Brookings Papers on Economic Activity, 2: 1-49.
38
CHAPTER 17 SUPPLY-SIDE AND UNEMPLOYMENT POLICY
useful because it represents the evolution of policymaking and also because most of the issues are politically controversial. For this reason, we have presented a large selection of policy issues to motivate the theory. Hopefully, at the time this class is taught, instructors will be able to refer to current policy debates and ask students to prepare relevant economic arguments. Alternatively, the chapter can be taught in reverse: starting with either tax reform or unemployment as a policy issue, then ask what principles are needed to think through the policy options, and thus come back to the principles exposited in Section 17.2.
Objectives
The two previous chapters showed the limits of demand-side policies. Supply-side policies have been believed to be the response to disenchantment with demand-side policies. Usually it means less government but not always. The common idea is that every economy contains pockets of inefficiencies which can be rooted out either by more efficient functioning of markets or by the state. This chapter organizes the ideas which lie at the heart of most o f the supply-side policies which have been tried over the past decade or so. By way of examples, we focus on two areas where the principles of supply-side policies are readily applicable: taxation and labour markets. This choice is obvious: tax reform is on the political agenda of many countries in Europe while unemployment is arguably Europe’s worst supply-side failure.
Public goods and taxation
Section 17.3 returns to a theme already discussed in Chapter 15: public goods are needed but the taxes which finance them are distortionary and are associated with welfare costs. While Chapter 15 used this idea to caution against activist fiscal policies, the approach taken here asks rather what are the public goods that governments must provide and how they can be provided in the most efficient way. Most of the answers come from the literature on public finance (e.g. the principles of efficient taxation). The principles involved are mostly applied, however, to macroeconomic concepts developed in earlier chapters -- savings, labour, capital accumulation -- thus closely echoing the themes developed in the chapter on growth (Chapter 5).
Micro- and macroeconomics
We call supply-side policies all those interventions which aim at improving an economy's overall efficiency or productivity. Because a large patchwork of policies falls under this rubric, the chapter starts by examining the most common sources of inefficiency. This has the virtue of providing a unifying framework, strengthening the analytical content of the chapter, and establishing links with previous results. We start by a reminder of the principle of laissez faire in economies characterised by perfect competition. This is used to motivate a review of the main sources of deviations from the perfect competition paradigm. Here again the students will note that the border between micro- and macroeconomics is fuzzy. Supply-side policies mainly draw the macroeconomic implications of a variety of microeconomic principles.
Structural unemployment
Section 17.4 can be seen as a sequel to Chapter 6. It can be taught separately from the rest, if only to develop the analysis of equilibrium unemployment that was only briefly exposited. Instructors who consider dropping this chapter could -- in our view, should -still use the material of this section, either on its own or when presenting Chapter 6.
A difficult chapter to teach
This is a difficult chapter to teach. It covers a lot of (mostly microeconomic) ground. It deals with a domain where there is still little empirical evidence available on which to judge the effectiveness of policies. Yet, it is
39
Instructor’s Guide
Chapter 17
Barro, Robert (1991), 'Economic Growth in a Cross Section of Countries,' Quarterly Journal of Economics, 106, May: 407-44. The three curves
Blanchard, Olivier J. (1991) “Two Tools for Analysing Unemployment”, in: M. Nerlove (ed.) MacroIEA Conference economics and Econometrics, Volume 99: 102-27, New York University Press, New York.
This chapter introduces three empirical curves. Each of them requires qualification but can be used to motivate theoretical developments. The Laffer curve easily attracts attention. Its logic seems inescapable and, for this reason, should be 1 heavily qualified. Students typically see its relevance and they recognize that the issue is being publicly debated. The Beveridge curve, an empirical regularity with respectable microfoundations (see Blanchard (1991)), provides a possible pretext for the discussion of information and a search in the macroeconomy. The Calmfors-Driffill (1988) “hump-shape” curve has also received some microfoundation, but its empirical status is often debated. It is a very useful instrument to focus the student’s mind on labour bargaining, the role of trade unions and governments, the existence of labour strikes, etc.
Calmfors, Lars and Driffill, John (1988), "Bargaining Structure, Corporatism and Macroeconomic Performance," Economic Policy 6: 16-61. Lindsey, Lawrence (1987), 'Individual Taxpayer Response to Tax Cuts: 1982-1984', Journal of Public Economics, 33: 173-206.
Other avenues
One subject which has received little attention in this chapter but which can be developed alongside public goods, is the role of infrastructure investment and education. These topics were discussed in Chapter 5. While the evidence on the role of public investment in cross-country growth performance is weak (Barro, 1991), this line of thought is clearly behind the European Commission’s plans to increase spending on highways, roads, and bridges as part of a long term supply-side policy to improve total factor productivity. The transformation of Eastern and Central Europe can be seen in a similar light. Reference can also be made to the literature on economic development and to strategies applied in developing countries. The usual growth-literature papers citing the importance of human capital (again Barro, 1991 or the references in Chapter 5) can be used to motivate supply-side policies which target improved education and training.
References
1
There is some evidence from the United States (Lindsey 1987) that the behaviour of taxpayers changed after the tax cuts of the Reagan administration -- due less to increased labour supply than to increased income reporting and fewer tax avoidance activities.
40
CHAPTER 18
FINANCIAL AND EXCHANGE MARKETS
Objectives
This chapter marks a transition between open-economy macroeconomics and a section devoted to the determination of the exchange rate. It has two main purposes: 1) to present key features of financial and exchange markets; 2) to develop the principle of market efficiency.
Further references
Several commercial banks publish booklets for their customers, designed to explain financial and exchange markets. These booklets are usually very simple and contain many details on institutions and/or instruments. The Bank for International Settlements in Basel presents useful data in its Annual Report, especially on Euro-markets.
Market institutions and instruments
The sections on financial and exchange markets are mostly descriptive. They can be left to reading at home, teachers simply summarizing the key points in class as they are listed in the summary section. Depending on student interest, there may be no need to explain these 1 instruments in detail. It may also be useful to check that students know how to compute a one-week or onemonth yield at annual rate.
Market efficiency
Students must clearly understand the concept of market efficiency: it rules out systematic, but not random and unpredictable gains. They should be able to distinguish arbitrage (no risk involved) from speculation (risk taking). The discussion on bubbles in Section 18.3.3.3 may seem esoteric but it is worth going over for three reasons. First, it provides a check on students ’ understanding of arbitrage. Second, it illustrates clearly what is meant by ’fundamentals’. Third, it may be highly relevant to the early 1990s, which saw spectacular declines of stock and house prices in many countries. In the UK, Sweden, France, Finland, Japan, among others, banks have been nearly bankrupted as a result of the heavy losses suffered in real estate investments.
1
An exception is that of forward contracts which play an important role in Chapter 19.
41
CHAPTER 19 EXCHANGE RATES IN THE SHORT RUN
presenting the parity relationships is to prevent students from being confused about when each one holds, and how it all relates to forward premiums. The use of different symbols for different premiums and the summary Table 19.3 are designed to clarify matters. It should be emphasised however that the risk premia introduced in (19.4) and (19.9) are really symbolic rather than derived from a proper theory, and that there is no presumption whatsoever as to their sign. In fact they are known to be very unstable, both in sign and 2 size. It is also well worth going over the decomposition (19) of the forward forecast error between the risk premium and the market forecast error. Note that forecast errors are no proof of market inefficiency as long as they are not systematically exploitable for profit.
Objectives
The theory of exchange rate determination is both difficult and exciting. This chapter proposes an efficient short cut without giving up much substance. The challenge is to cover the material in a way that is clear and simple. We have proposed the following sequence of topics: - Establish the tension between the ’long run’ view of the exchange rate in Chapter 7 with the data, as well as with Mussa’s stylised facts. - Start with the interest parity conditions already presented in Chapter 11. Then, by manipulating this condition, it is possible to show the essentials of exchange rate determination: forward-looking, unpredictable ’jumping’. - With a little bit more complexity and simply using a graphical apparatus, it is then possible to establish the overshooting proposition and, at the same time, to link the short run, developed in this chapter, with 1 the long run as established in Chapter 7.
Forward looking exchange rate
Interest parity has been briefly introduced -- and extensively used -- earlier. Here it is explained in detail and with greater precision. The main difficulty when
Notice that all terms in the interest parity condition (19.15) are endogenous so that the whole exercise conducted in Section 19.3 should not be interpreted as a theory of exchange rate determination. Yet repeated substitutions yield an important intuition for the forward-looking nature of the exchange rate (as for all assets). It is often difficult to make Fig. 19.5 understandable to students on the first try. There are two ways of proceeding. The easy way is to say that when the interest rate rises, capital flows in and the exchange rate appreciates. This is true of course but it misses the whole idea that exchange rates are forward looking. So it is worth confusing students a bit by observing that a higher interest rate means that the exchange rate is expected to depreciate: as long as the expected rate of depreciation equals the interest differential (plus a risk premium factor, if applicable) there are no capital flows at all. To understand what happens to the exchange rate
1
2
Mussa’s stylised facts
Stating ex ante what is to be explained serves three main purposes. First, it gives students a clear view of where the lecture is leading. Second, students’ attention can be maintained by reminding them which stylized fact is being explained each time a new result is achieved. Finally, it represents an anchor in a storm of fairly complex material. Use of a transparency listing the stylised facts and checking them, is advisable.
The interest parity relationships
If Chapter 7 has been skipped earlier, as is entirely possible -- see instructions in this Guide -- it must be covered before Chapter 19.
Some instructors may prefer to assume that the risk premium is zero, and attribute deviations from UIP to the ’peso problem’.
42
Instructor’s Guide
Chapter 19
now, there is no short cut: we must use theory to help us pin down the nominal exchange rate in the long run (see below). Time spent interpreting (19.19) is a good investment. In particular, it is a good idea to point out very early that the link between the short run (and the enormous amount of volatility) and the long run is that the long run exchange rate is always present in the background as captured by the ’mast’ or leading term E t +n+1.
Frankel, Jeffrey, and MacArthur, Alan (1988), ’Political vs. Currency Premia in International real Interest Differentials’, European Economic Review, 32: 1083114. Frankel, Jeffrey, and Meese, Richard (1987), ’Are Exchange Rates Excessively Variable?’, NBER Macroeconomic Annual, 2: 117-62.
Overshooting
The formal presentation of the ’Dornbusch model’ in the Appendix can easily be avoided, where necessary, thanks to the graphical apparatus developed in Section 19.4. There are three key intuitions which must be established. First it is price rigidity which forces an exchange rate to move ’too much’. Appealing to the principle of physics, that when there is excess pressure in a system it must find one way or another to escape, is a useful analogy. Second, note that the mix of highly volatile exchange rates and sticky prices implies that the real exchange rate, by and large, moves like the nominal exchange rate in the short run. This means that nominal fluctuations -- nominal interest rates pushing around nominal exchange rates -- have real effects via the real exchange rate. This is a key channel for transmission of monetary policy in open economies under flexible rates as argued in Chapter 11 and after. Third, the so-called fundamental determinants of the exchange rate mostly consist of expected future variables such as money and real aspects of competitiveness. The past matters relatively little. This immediately establishes a link, somewhat underemphasised in the text, with the authorities’ credibility in pursuing a particular set of policies. Note also that the overshooting result predicts that the exchange rate is likely to deviate frequently by a wide margin from what is warranted by the equilibrium real exchange rate presented in Chapter 7.
References
Dornbusch, Rudiger (1976), ’Expectations and Exchange Rate Dynamics’, Journal of Political Economy, 1161-76.
43
CHAPTER 20
THE INTERNATIONAL MONETARY SYSTEM
Objectives
In this chapter many of the earlier results are illustrated as the history of international monetary relations unfolds. The gold and other metallic standards, which occupied centre stage for so long, bring home to roost many of the important ideas of Chapters 8 and 9. The issues underlying the Hume mechanism are simply Chapters 11 and 13 cast in a historic setting. The issue of credibility of policy (Chapter 16) is important for understanding the working of the gold standard: the often cited increasing nominal rigidity in the world economy may be an result of moving to a fiat money standard in which downward adjustment is no longer assumed to be necessary. A section covers the standard debate on the choice of an exchange-rate regime. Reflecting on why systems emerge and collapse it is useful to think about current issues such as the European Monetary System and discussions on monetary co-operation among the larger countries. There are no exercises for this chapter.
44
CHAPTER 21
POLICY CO-ORDINATION AND EXCHANGE RATE CRISES: THE EMS AND EMU
Objectives
A study of the EMS provides the background for covering a number of analytical issues such as: policy co-ordination, balance of payments crises, the theory of an optimum currency area. Like the preceding chapter, it offers a lot of data and experience to which the results developed earlier can be applied. In the years preceding the launch of a monetary union, these are burning issues of independent interest for European students. This chapter has been entirely rewritten following the dramatic events of 1992-93. No doubt, the years 1997-1999 will be historical for European monetary integration. As we wrote this chapter, we were painfully aware that events will surprise us, one way or another. Instructors will fill the vacuum and correct unfulfilled speculation on our part!
Self-fulfilling speculative attacks
The crises of 1992-93 have revived interest in speculative attacks. Box 21.2 presents the theory of speculative attacks and concludes that a currency which pursues policies incompatible with a fixed exchange rate must eventually give up and float. Six currencies have come under sharp attack during the autumn 1992. Only two of them conform to this view: the lira and sterling indeed have left the ERM and begun to float after a sharp (about 10%) depreciation. This has led to the development of the theory of self-fulfilling attacks which is briefly presented in Section 21.3.4. The paper by Obstfeld (1995) quoted in the text, as well as Eichengreen, Rose and Wyplosz (1995), offer a fuller treatment.
Teaching Optimal currency area
This chapter is relatively easy to teach, and a nice conclusion of the course. European students know the facts and the challenges. The chapter is structured along historical lines, which is a natural, but timeconsuming, way of introducing the issues. An alternative is to select some topics and use them while teaching the relevant principle. A list of topics is: - the Mundell-Fleming model under fixed exchange rate: the N-1 problem and German dominance (Section 21.3.3) - the role of perfect capital mobility in the MundellFleming model: the impossible trilogy (Section 21.3.5) - credibility: one reason for German dominance (Section 21.3.3) - the role of expectations: crises in the EMS (Section 21.3.4) - the long-run real exchange rate (Chapter 7) and the optimum currency area (Section 21.4.1) - balance of payments and mutual assistance (Section 21.2.3)
It is now fashionable to research the conditions under which monetary union is feasible and welfare will improve. Following Mundell (1961), the two criteria seem to be a high correlation of regional supply and demand shocks and/or a high degree of factor mobility. In the USA the latter seems to obtain either because of necessity or national culture (see Blanchard and Katz 1992). This does not seem to be the case in Europe, nor has the integration of product and factor markets removed sources of asynchronization of disturbances in European regions (von Hagen and Neumann (1994)), Decressin and Fatás (1995). The imaginative instructor might introduce some of the convergence discussion that appeared in Chapter 5 'through the back door' as well as arguments for and against a more federal 1 European tax and expenditure system.
1
See Bean (1992) for a survey of the debate surrounding EMU.
45
