Get the full ebook with Bonus Features for a Better Reading Experience on ebookgate.

com

Structural Macroeconometrics Second Edition David

N. Dejong

https://ebookgate.com/product/structural-macroeconometrics-
second-edition-david-n-dejong/

OR CLICK HERE

DOWLOAD NOW

Download more ebook instantly today at https://ebookgate.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Sea Ice Second Edition David N. Thomas

https://ebookgate.com/product/sea-ice-second-edition-david-n-thomas-2/

ebookgate.com

Sea Ice Second Edition David N. Thomas

https://ebookgate.com/product/sea-ice-second-edition-david-n-thomas/

ebookgate.com

Structural Equation Modeling Foundations and Extensions

Advanced Quantitative Techniques in the Social Sciences
Second Edition David W. Kaplan
https://ebookgate.com/product/structural-equation-modeling-
foundations-and-extensions-advanced-quantitative-techniques-in-the-
social-sciences-second-edition-david-w-kaplan/
ebookgate.com

The Structural Basis of Architecture 2nd ed. Edition Bjørn

N. Sandaker

https://ebookgate.com/product/the-structural-basis-of-
architecture-2nd-ed-edition-bjorn-n-sandaker/

ebookgate.com
Polysaccharides Structural Diversity and Functional
Versatility Second Edition Severian Dumitriu

https://ebookgate.com/product/polysaccharides-structural-diversity-
and-functional-versatility-second-edition-severian-dumitriu/

ebookgate.com

Applied structural and mechanical vibrations theory and

methods Second Edition Gatti

https://ebookgate.com/product/applied-structural-and-mechanical-
vibrations-theory-and-methods-second-edition-gatti/

ebookgate.com

Inorganic Structural Chemistry Second Edition Dr. Rer.

Nat. Ulrich Muller(Auth.)

https://ebookgate.com/product/inorganic-structural-chemistry-second-
edition-dr-rer-nat-ulrich-mullerauth/

ebookgate.com

Economic Growth 3rd Edition David N. Weil

https://ebookgate.com/product/economic-growth-3rd-edition-david-n-
weil/

ebookgate.com

Economic Growth 3 edition David N. Weil

https://ebookgate.com/product/economic-growth-3-edition-david-n-weil/

ebookgate.com
Structural Macroeconometrics
This page intentionally left blank
Structural Macroeconometrics

SECOND EDITION

David N. DeJong and Chetan Dave

PRINCETON UNIVERSITY PRESS PRINCETON AND OXFORD

Copyright © 2011 by Princeton University Press
Published by Princeton University Press, 41 William Street, Princeton,
New Jersey 08540

In the United Kingdom: Princeton University Press, 6 Oxford Street,

Woodstock, Oxfordshire OX20 1TW

All Rights Reserved

ISBN: 978-0-691-15287-5

Library of Congress Control Number: 2011933704

British Library Cataloging-in-Publication Data is available

This book has been composed in Galliard

Printed on acid-free paper. ∞

press.princeton.edu

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1
For Denise, Andrew, and Alex: Enjoy the lemonade! DND

For my grandparents: S. B. Pande, P. Pande, M. S. Yagnik,

and K. Yagnik. CD
This page intentionally left blank
Contents

Preface xiii
Preface to the First Edition xv

Part I Introduction
1 Background and Overview 3
1.1 Background 3
1.2 Overview 4
2 Casting Models in Canonical Form 9
2.1 Notation 9
2.1.1 Log-Linear Model Representations 11
2.1.2 Nonlinear Model Representations 11
2.2 Linearization 12
2.2.1 Taylor Series Approximation 12
2.2.2 Log-Linear Approximations 14
2.2.3 Example Equations 15
3 DSGE Models: Three Examples 18
3.1 Model I: A Real Business Cycle Model 20
3.1.1 Environment 20
3.1.2 The Nonlinear System 23
3.1.3 Log-Linearization 26
3.2 Model II: Monopolistic Competition and Monetary Policy 28
3.2.1 Environment 28
3.2.2 The Nonlinear System 33
3.2.3 Log-Linearization 34
3.3 Model III: Asset Pricing 38
3.3.1 Single-Asset Environment 38
3.3.2 Multi-Asset Environment 39
3.3.3 Alternative Preference Specifications 40

Part II Model Solution Techniques

4 Linear Solution Techniques 51
4.1 Homogeneous Systems 52
4.2 Example Models 54
4.2.1 The Optimal Consumption Model 54
4.2.2 Asset Pricing with Linear Utility 55
4.2.3 Ramsey’s Optimal Growth Model 56
viii Contents

4.3 Blanchard and Kahn’s Method 57

4.4 Sims’ Method 61
4.5 Klein’s Method 64
4.6 An Undetermined Coefficients Approach 66
5 Nonlinear Solution Techniques 69
5.1 Projection Methods 71
5.1.1 Overview 71
5.1.2 Finite Element Methods 72
5.1.3 Orthogonal Polynomials 73
5.1.4 Implementation 74
5.1.5 Extension to the l -dimensional Case 78
5.1.6 Application to the Optimal Growth Model 79
5.2 Iteration Techniques: Value-Function and Policy-Function
Iterations 87
5.2.1 Dynamic Programming 87
5.2.2 Value-Function Iterations 89
5.2.3 Policy-Function Iterations 94
5.3 Perturbation Techniques 95
5.3.1 Notation 95
5.3.2 Overview 97
5.3.3 Application to DSGE Models 99
5.3.4 Application to an Asset-Pricing Model 105

Part III Data Preparation and Representation

6 Removing Trends and Isolating Cycles 113
6.1 Removing Trends 115
6.2 Isolating Cycles 120
6.2.1 Mathematical Background 120
6.2.2 Cramér Representations 124
6.2.3 Spectra 125
6.2.4 Using Filters to Isolate Cycles 126
6.2.5 The Hodrick-Prescott Filter 128
6.2.6 Seasonal Adjustment 130
6.2.7 Band Pass Filters 131
6.3 Spuriousness 134
7 Summarizing Time Series Behavior When All Variables
Are Observable 138
7.1 Two Useful Reduced-Form Models 139
7.1.1 The ARMA Model 139
7.1.2 Allowing for Heteroskedastic Innovations 145
7.1.3 The VAR Model 147
Contents ix

7.2 Summary Statistics 149

7.2.1 Determining Lag Lengths 157
7.2.2 Characterizing the Precision of Measurements 159
7.3 Obtaining Theoretical Predictions of Summary Statistics 162
8 State-Space Representations 166
8.1 Introduction 166
8.1.1 ARMA Models 167
8.2 DSGE Models as State-Space Representations 169
8.3 Overview of Likelihood Evaluation and Filtering 171
8.4 The Kalman Filter 173
8.4.1 Background 173
8.4.2 The Sequential Algorithm 175
8.4.3 Smoothing 178
8.4.4 Serially Correlated Measurement Errors 181
8.5 Examples of Reduced-Form State-Space Representations 182
8.5.1 Time-Varying Parameters 182
8.5.2 Stochastic Volatility 185
8.5.3 Regime Switching 186
8.5.4 Dynamic Factor Models 187

Part IV Monte Carlo Methods

9 Monte Carlo Integration: The Basics 193
9.1 Motivation and Overview 193
9.2 Direct Monte Carlo Integration 196
9.2.1 Model Simulation 198
9.2.2 Posterior Inference via Direct Monte Carlo Integration 201
9.3 Importance Sampling 202
9.3.1 Achieving Efficiency: A First Pass 206
9.4 Efficient Importance Sampling 211
9.5 Markov Chain Monte Carlo Integration 215
9.5.1 The Gibbs Sampler 216
9.5.2 Metropolis-Hastings Algorithms 218
10 Likelihood Evaluation and Filtering in State-Space
Representations Using Sequential Monte Carlo Methods 221
10.1 Background 221
10.2 Unadapted Filters 224
10.3 Conditionally Optimal Filters 228
10.4 Unconditional Optimality: The EIS Filter 233
10.4.1 Degenerate Transitions 235
10.4.2 Initializing the Importance Sampler 236
10.4.3 Example 239
x Contents

10.5 Application to DSGE Models 241

10.5.1 Initializing the Importance Sampler 243
10.5.2 Initializing the Filtering Density 245
10.5.3 Application to the RBC Model 246

Part V Empirical Methods

11 Calibration 253
11.1 Historical Origins and Philosophy 253
11.2 Implementation 258
11.3 The Welfare Cost of Business Cycles 261
11.4 Productivity Shocks and Business Cycle Fluctuations 268
11.5 The Equity Premium Puzzle 273
11.6 Critiques and Extensions 276
11.6.1 Critiques 276
11.6.2 Extensions 279
12 Matching Moments 285
12.1 Overview 285
12.2 Implementation 286
12.2.1 The Generalized Method of Moments 286
12.2.2 The Simulated Method of Moments 294
12.2.3 Indirect Inference 297
12.3 Implementation in DSGE Models 300
12.3.1 Analyzing Euler Equations 300
12.3.2 Analytical Calculations Based on Linearized Models 301
12.3.3 Simulations Involving Linearized Models 306
12.3.4 Simulations Involving Nonlinear Approximations 307
12.4 Empirical Application: Matching RBC Moments 308
13 Maximum Likelihood 314
13.1 Overview 314
13.2 Introduction and Historical Background 316
13.3 A Primer on Optimization Algorithms 318
13.3.1 Simplex Methods 319
13.3.2 Derivative-Based Methods 328
13.4 Ill-Behaved Likelihood Surfaces: Problems and Solutions 330
13.4.1 Problems 330
13.4.2 Solutions 331
13.5 Model Diagnostics and Parameter Stability 334
13.6 Empirical Application: Identifying Sources of Business
Cycle Fluctuations 337
14 Bayesian Methods 351
14.1 Overview of Objectives 351
14.2 Preliminaries 352
Contents xi

14.3 Using Structural Models as Sources of Prior Information

for Reduced-Form Analysis 355
14.4 Implementing Structural Models Directly 360
14.5 Model Comparison 361
14.6 Using an RBC Model as a Source of Prior Information
for Forecasting 364
14.7 Estimating and Comparing Asset-Pricing Models 373
14.7.1 Estimates 380
14.7.2 Model Comparison 384

References 387
Index 401
This page intentionally left blank
Preface

Scratch where it itches.

—Alice Roosevelt Longworth

If it ain’t broke, don’t fix it.

—Bert Lance

The usual excuse for crafting the second edition of a text is rapidly changing
subject matter. We could certainly invoke this excuse here: the empiri-
cal analysis of dynamic stochastic general equilibrium (DSGE) models has
evolved considerably over the past five years. Instead, we confess that this
excuse is only second-order important in this instance. First-order im-
portant is our graduation from the steepest portion of a critical learning
curve.
In particular, when we set out to characterize likelihood evaluation and
filtering for nonlinear model representations, our comprehension of ap-
plicable numerical methods did not extend beyond the standard particle
filter. Moreover, our understanding of the particle filter was limited to the
mechanics of its execution (which is exceedingly simple), and did not ex-
tend to a sufficient appreciation of its shortcomings (which, alas, can be
significant in empirically relevant situations). Indeed, as we were putting
the finishing touches on the first edition, one of us (DeJong) was develop-
ing plans to implement the particle filter in pursuit of a promising empirical
application. Shortly thereafter, the particle filter was found to be not up to
the task, and despite the passage of time, the application remains on the
back burner (thank goodness for tenure).
The good news is that, with necessity serving as the mother of inven-
tion, we have come to better appreciate the strengths and weaknesses of
the particle filter, as well as alternative methods. Even better, our emer-
gence from the learning curve has led to the development (by DeJong, in
collaboration with Hariharan Dharmarajan, Roman Liesenfeld, Guilherme
Moura, and Jean-François Richard) of an alternative filtering method: the
efficient importance sampling (EIS) filter. As we shall demonstrate, imple-
mentation of the EIS filter is well suited to DSGE models, and the payoff
to its adoption can be high.
The emphasis here on filtering reflects our appreciation of the im-
portance that departures from linearity hold for the future of empirical
work involving DSGE models. Five years ago, the use of nonlinear model
xiv Preface

representations in empirical applications was exotic; five years from now,

we expect this use to be routine. Accordingly, while in the first edition the
treatment of nonlinear model representations was isolated to a pair of clo-
sing chapters, linear and nonlinear treatments are now characterized in
tandem throughout the text.
Users of the first edition will notice additional changes as well: details
are sketched in chapter 1. However, we remain satisfied with the bulk of
the subjects covered in the first edition, and have heeded the wisdom of
Bert Lance in their treatment. We hope we have done so judiciously.
Since the publication of the first edition, we have benefited from input
on the text received from students we have taught at Carnegie-Mellon and
Di Tella universities, the Kiel Institute for the World Economy, the Uni-
versity of Pittsburgh, and the University of Texas at Dallas. We have also
been alerted to many errors of commission: we are particularly indebted
to Roberto Chang, Andres Fernandez, Michael Hauser, Jan Christopher
Ocampo, and Surach Tanboon for their careful reading. Certainly we
committed many errors of omission in the first edition when it came to
describing literatures related to the topics we covered. We have sought
to update and expand our treatment of these literatures here, though in
so doing we have surely left a great many excellent papers unmentioned:
apologies in advance for this.

Dave DeJong and Chetan Dave

Preface to the First Edition

Let me only say that what econometrics—aided by

electronic computers—can do, is only to push for-
ward by leaps and bounds the line of demarcation
from where we have to rely on our intuition and
sense of smell.
—Ragnar Frisch, Nobel Prize Lecture,
June 1970.

It is a capital mistake to theorize before one has

data. Insensibly one begins to twist facts to suit
theories, instead of theories to suit facts.
—Sir Arthur Conan Doyle

A steady stream of conceptual and computational advances realized over

the past three decades has helped to bridge the gulf that has historically sep-
arated theoretical from empirical research in macroeconomics. As a result,
measurement is becoming increasingly aligned with theory. The purpose
of this text is to provide guidance in bringing theoretical models to the
forefront of macroeconometric analyses.
The text is suitable for use as a supplementary resource in introductory
graduate courses in macroeconomics and econometrics, and as a primary
textbook in advanced graduate courses devoted to the pursuit of applied
research in macroeconomics. The lecture notes that ultimately gave rise
to the text were designed for this latter purpose. The text’s historical per-
spective, along with its unified presentation of alternative methodologies,
should also make it a valuable resource for academic and professional
researchers.
Readers of the text are assumed to have familiarity with multivariate cal-
culus, matrix algebra, and difference equations, and cursory knowledge of
basic econometric techniques. Familiarity with dynamic programming is
also useful but not necessary. This is the tool used to map the class of models
of interest here into the system of nonlinear expectational difference equa-
tions that serve as the point of departure for the empirical methodologies
presented in the text. However, familiarity with dynamic programming is
not needed to follow the text’s presentation of empirical methodologies.
We decided to write this text because through our own teaching and
research, we have sought to contribute towards the goal of aligning theory
xvi Preface to the First Edition

with empirical analysis in macroeconomics; this text is a natural extension

of these efforts. We set out to accomplish two objectives in writing it. First,
we wished to provide a unified overview of this diverse yet interrelated area
of active research. Second, we wanted to equip students with a set of tools
that would expedite their own entry into the field.
The content of this text reflects much that we have learned over many
years spent pursuing collaborative research with a fabulous group of coau-
thors: Sheryl Ball, Patty Beeson, Dan Berkowitz, Stefan Dodds, Scott
Dressler, Catherine Eckel, Emilio Espino, Steve Husted, Beth Ingram,
Roman Leisenfeld, John Nankervis, Jean-François Richard, Marla Ripoll,
Gene Savin, Werner Troesken, and Chuck Whiteman. We are deeply
indebted to them for their implicit and explicit contributions to this pro-
ject. We have also benefited from input provided by Charlie Evans, Jim
Feigenbaum, Jesús Fernández-Villaverde, Peter Ireland, Naryana Kocher-
lakota, Chris Otrok, Barbara Rossi, Juan Rubio-Ramírez, Tom Sargent,
Thomas Steinberger, Yi Wen, and Tao Zha on various aspects of the text.
Finally, Hariharan Dharmarajan provided valuable research assistance on
the empirical application presented in chapter 11.
Part I

Introduction
This page intentionally left blank
Chapter 1

Background and Overview

Science is facts; just as houses are made of stones,

so is science made of facts; but a pile of stones is not
a house and a collection of facts is not necessarily
science.
—Henri Poincaré

1.1 Background
The seminal contribution of Kydland and Prescott (1982) marked a sea
change in the way macroeconomists conduct empirical research. Under
the empirical paradigm that remained predominant at the time, the fo-
cus was either on purely statistical (or reduced-form) characterizations of
macroeconomic behavior, or on systems-of-equations models that ignored
both general-equilibrium considerations and forward-looking behavior on
the part of purposeful decision makers. But the powerful criticism of this
approach set forth by Lucas (1976), and the methodological contributions
of, for example, Sims (1972) and Hansen and Sargent (1980), sparked a
transition to a new empirical paradigm. In this transitional stage, the for-
mal imposition of theoretical discipline on reduced-form characterizations
became established. The source of this discipline was a class of models
that have come to be known as dynamic stochastic general equilibrium
(DSGE) models. The imposition of discipline most typically took the form
of “cross-equation restrictions,” under which the stochastic behavior of a
set of exogenous variables, coupled with forward-looking behavior on the
part of economic decision makers, yield implications for the endogenous
stochastic behavior of variables determined by the decision makers. Never-
theless, the imposition of such restrictions was indirect, and reduced-form
specifications continued to serve as the focal point of empirical research.
Kydland and Prescott turned this emphasis on its head. As a legacy of
their work, DSGE models no longer serve as indirect sources of theoretical
discipline to be imposed upon statistical specifications. Instead, they serve
directly as the foundation upon which empirical work may be conducted.
The methodologies used to implement DSGE models as foundational em-
pirical models have evolved over time and vary considerably. The same
4 1 Background and Overview

is true of the statistical formality with which this work is conducted. But
despite the characteristic heterogeneity of methods used in pursuing con-
temporary empirical macroeconomic research, the influence of Kydland
and Prescott remains evident today.
This book details the use of DSGE models as foundations upon which
empirical work may be conducted. It is intended primarily as an instruc-
tional guide for graduate students and practitioners, and so contains a
distinct how-to perspective throughout. The methodologies it presents
are organized roughly following the chronological evolution of the empir-
ical literature in macroeconomics that has emerged following the work of
Kydland and Prescott; thus it also serves as a reference guide. Throughout,
the methodologies are demonstrated using applications to three bench-
mark models: a real-business-cycle model (fashioned after King, Plosser,
and Rebelo, 1988); a monetary model featuring monopolistically com-
petitive firms (fashioned after Ireland, 2004a); and an asset-pricing model
(fashioned after Lucas, 1978).

1.2 Overview
The empirical tools outlined in the text share a common foundation: a sys-
tem of nonlinear expectational difference equations representing a DSGE
model. The solution to a given system takes the form of a collection of
policy functions that map a subset of the variables featured in the model—
state variables—into choices for the remaining subset of variables—control
variables; the choices represent optimal behavior on the part of the deci-
sion makers featured in the model, subject to constraints implied by the
state variables. Policy functions, coupled with laws of motion for the state
variables, collectively represent the solution to a given DSGE model; the
solution is in the form of a state-space representation.
Policy functions can be calculated analytically for only a narrow set of
models, and thus must typically be approximated numerically. The text pre-
sents several alternative methodologies for achieving both model approxi-
mation and the subsequent empirical implementation of the corresponding
state-space representation. Empirical implementation is demonstrated in
applications involving parameter estimation, assessments of fit and model
comparison, forecasting, policy analysis, and measurement of unobservable
facets of aggregate economic activity (e.g., measurement of productivity
shocks).
This book is divided into five parts. Part I contains three chapters. Fol-
lowing this brief introduction, chapter 2 establishes the consistent set of
notation used throughout the text, and presents an important preliminary
1.2 Overview 5

step—linearization—of the model solution stage. Chapter 3 then intro-

duces the three benchmark models that serve as examples throughout
the text.
Part II contains two chapters devoted to model-solution techniques.
Chapter 4 describes solution techniques applicable to linear approxima-
tions of systems of expectational difference equations. Chapter 5 then
describes three alternative classes of methods for obtaining nonlinear
model approximations: projection, iteration, and perturbation. Although
nonlinear approximations are necessarily more accurate than their linear
counterparts, linear approximations are nevertheless valuable, as they of-
ten serve as the foundation upon which nonlinear representations are
constructed.
Part III is devoted to data preparation and representation, and contains
three chapters. Chapter 6 presents two important preliminary steps often
needed for priming data for empirical analysis: removing trends and iso-
lating cycles. The purpose of these steps is to align what is being measured
in the data with what is being modeled by the theory. For example, the
separation of trend from cycle is necessary in confronting trending data
with models of business cycle activity.
Chapter 7 presents tools used to summarize properties of the data. First,
two important reduced-form models are introduced: autoregressive mov-
ing average models for individual time series, and vector autoregressive
models for sets of time series. These models provide flexible characteriza-
tions of the data that can be used as a means of calculating a wide range
of important summary statistics. Next, a collection of popular summary
statistics (along with algorithms available for calculating them) are intro-
duced. These statistics often serve as targets for estimating the parameters
of structural models, and as benchmarks for judging their empirical perfor-
mance. Empirical analyses involving collections of summary statistics are
broadly categorized as limited-information analyses.
Chapter 8 concludes Part III with an overview of state-space represen-
tations. Such representations characterize the evolution over time of two
types of variables: those that are observed (possibly subject to measurement
error) by the empirical analyst, and those that are not. Conditional on the
parameterization of a given model, and the evolution of the observable vari-
ables through a given point in time, evolution of the unobservable variables
up to the same point in time may be inferred through a process known as
filtering. As a by-product of the filtering process, the likelihood function of
the parameterized model can be derived. Chapter 8 first demonstrates the
mapping of a given DSGE model into a generic state-space representation.
It then provides a general overview of likelihood evaluation and filter-
ing, and then characterizes the Kalman filter, which achieves likelihood
evaluation and filtering analytically for the special case in which stochastic
6 1 Background and Overview

innovations and measurement errors are Normally distributed, and model

representations are linear. Chapter 8 concludes by characterizing a special
class of reduced-form state-space representations: dynamic factor models.
With the exception of the linear/Normal class of state-space repre-
sentations, likelihood evaluation and filtering requires the calculation of
integrals that cannot be computed analytically, but instead must be ap-
proximated numerically. Part IV contains two chapters that characterize
numerical integration techniques, or Monte Carlo methods. Chapter 9
presents four methods, all in the context of generic integration prob-
lems: direct Monte Carlo integration; model simulation; importance sam-
pling; and Markov Chain Monte Carlo. Chapter 10 then characterizes the
application of numerical integration techniques to the specific objective
of likelihood evaluation and filtering in applications involving state-space
representations. The particle filter is first introduced as a foundational
method for achieving these objectives. Next, concerns regarding numerical
efficiency are introduced, with an eye towards achieving efficiency enhance-
ments through the process of adaption. The EIS filter is then introduced
generically as a specific means of achieving such enhancements. Finally, the
specific application of the EIS filter to state-space representations associated
with DSGE models is outlined in detail.
Taking as input the representation of a given DSGE model in a state-
space framework, Part V concludes the text with four chapters devoted to
alternative methods for achieving empirical implementation. Specifically,
chapters 11 through 14 present the following empirical methodologies:
calibration, methods of moments, maximum likelihood, and Bayesian in-
ference. Each chapter contains a general presentation of the methodology,
and then presents applications to the benchmark models in pursuit of
alternative empirical objectives. (For an alternative textbook presentation
of these methods, see Canova, 2007.)
Chapter 11 presents the most basic empirical methodology covered in
the text: the calibration exercise, as pioneered by Kydland and Prescott
(1982). Original applications of this exercise sought to determine whether
models designed and parameterized to provide an empirically relevant ac-
count of long-term growth were also capable of accounting for the nature
of short-term fluctuations that characterize business-cycle fluctuations,
summarized using collections of sample statistics measured in the data.
More generally, implementation begins with the identification of a set of
empirical measurements that serve as constraints on the parameterization
of the model under investigation: parameters are chosen to ensure that
the model can successfully account for these measurements. (It is often
the case that certain parameters must also satisfy additional a priori con-
siderations.) Next, implications of the duly parameterized model for an
additional set of statistical measurements are compared with their empirical
1.2 Overview 7

counterparts to judge whether the model is capable of providing a success-

ful account of these additional features of the data. A challenge associated
with this methodology arises in judging success, because this second-stage
comparison is made in the absence of a formal statistical foundation.
The method-of-moments techniques presented in chapter 12 serve as
one way to address problems arising from the statistical informality associ-
ated with calibration exercises. Motivation for their implementation stems
from the fact that there is statistical uncertainty associated with the set of
empirical measurements that serve as constraints in the parameterization
stage of a calibration exercise. For example, a sample mean has an associated
sample standard error. Thus there is also statistical uncertainty associ-
ated with model parameterizations derived from mappings onto empirical
measurements (referred to generally as statistical moments). Method-of-
moment estimation techniques account for this uncertainty formally: the
parameterizations they yield are interpretable as estimates, featuring classi-
cal statistical characteristics. Moreover, if the number of empirical targets
used in obtaining parameter estimates exceeds the number of parameters
being estimated (i.e., if the model in question is overidentified), the es-
timation stage also yields objective goodness-of-fit measures that can be
used to judge the model’s empirical performance. Prominent examples
of moment-matching techniques include the generalized and simulated
methods of moments (GMM and SMM), and indirect-inference methods.
Moment-matching techniques share a common trait: they are based on
a subset of information available in the data (the targeted measurements
selected in the estimation stage). An attractive feature of these methodolo-
gies is that they may be implemented in the absence of explicit assumptions
regarding the underlying distributions that govern the stochastic behav-
ior of the variables featured in the model. A drawback is that decisions
regarding the moments chosen in the estimation stage are often arbit-
rary, and results (e.g., regarding fit) can be sensitive to particular choices.
Chapters 13 and 14 present full-information counterparts to these metho-
dologies: likelihood-based analyses. Given a distributional assumption re-
garding sources of stochastic behavior in a given model, chapter 13 details
how the full range of empirical implications of the model may be assessed
via maximum-likelihood analysis, facilitated either by use of the Kalman
filter or a numerical counterpart. Parameter estimates and model evalu-
ation are facilitated in a straightforward way using maximum-likelihood
techniques. Moreover, given model estimates, the implied behavior of un-
observable variables present in the model (e.g., productivity shocks) may
be inferred as a by-product of the estimation stage.
A distinct advantage in working directly with structural models is that,
unlike their reduced-form counterparts, one often has clear a priori guid-
ance concerning their parameterization. For example, specifications of
8 1 Background and Overview

subjective annual discount rates that exceed 10% may be dismissed out
of hand as implausible. This motivates the subject of chapter 14, which
details the adoption of a Bayesian perspective in bringing full-information
procedures to bear in working with structural models. From the Bayesian
perspective, a priori views on model parameterization may be incorpo-
rated formally in the empirical analysis, in the form of a prior distribution.
Coupled with the associated likelihood function via Bayes’ Rule, the corre-
sponding posterior distribution may be derived; this conveys information
regarding the relative likelihood of alternative parameterizations of the
model, conditional on the specified prior and observed data. In turn, condi-
tional statements regarding the empirical performance of the model relative
to competing alternatives, the implied behavior of unobservable variables
present in the model, and likely future trajectories of model variables may
also be derived.
In the spirit of reducing barriers to entry into the field, we have devel-
oped a textbook website that contains the data sets that serve as examples
throughout the text, as well as computer code used to execute the method-
ologies we present. The code is in the form of procedures written in the
GAUSS programming language. Instructions for executing the procedures
are provided within the individual files. The website address is http://www
.pitt.edu/∼dejong/text.htm. References to procedures available via this
site are provided throughout this book. In addition, a host of freeware is
available throughout the Internet. In searching for code, good starting
points include the collection housed by Christian Zimmerman in his Quan-
titative Macroeconomics web page, and the collection of programs that
comprise DYNARE: http://dge.repec.org/, http://www.dynare.org/.
Much of the code provided at our website reflects the modification of
code developed by others, and we have attempted to indicate this explicitly
whenever possible. Beyond this attempt, we express our gratitude to the
many generous programmers who have made their code available for public
use. Finally, we thank Ryan Davis and Lauren Donahoe for assistance with
updating the data sets reported in the first edition of the text.
Chapter 2

Casting Models in Canonical Form

We could, of course, use any notation we want; do

not laugh at notations; invent them, they are pow-
erful. In fact, mathematics is, to a large extent,
invention of better notations.
—Richard P. Feynman

One of the cleverest features of the rational ex-

pectations revolution was the application of the
term “rational.” Thereby, the opponents of this ap-
proach were forced into the defensive position of
either being irrational or of modelling others as ir-
rational, neither of which are comfortable positions
for an economist.
—R. J. Barro

2.1 Notation
A common set of notations is used throughout the text in presenting mod-
els and empirical methodologies. A summary of it follows. The complete
set of variables associated with a particular model is represented by the
n × 1 vector of stationary variables zt , the i th component of which is given
by zit . Among other things, stationarity implies that the unconditional
expectation of the level of zt does not depend on time, and thus the in-
dividual elements zit show no tendency towards long-term growth. For
a given model, this may necessitate the normalization of trending vari-
ables prior to analysis. Specific examples of normalization are presented in
chapter 6 (specifically, in sections 1.2 and 2.1.4).
The environment and corresponding first-order conditions associated
with any given DSGE model can be converted into a nonlinear first-order
system of expectational difference equations. We represent such systems as

(Et zt +1 , zt , υt +1 ) = 0, (2.1)

where 0 is an n × 1 vector of zeros, υt is an m × 1 vector of structural

shocks, and Et zt +1 is the expectation of zt +1 formed by the model’s
10 2 Casting Models in Canonical Form

decision makers conditional on information available up to and including

period t :
Et zt +1 = E(zt +1 |zt , zt −1 , . . . , z0 )
≡ E(zt +1 |z t ).
In this sense, expectations are said to be formed rationally. In addition
to being conditional on z t , it is typically assumed that expectations are
formed given full information regarding the decision makers’ environment
(including the k × 1 vector µ, denoting the collection of parameters asso-
ciated with the model). For a textbook treatment of extensions wherein
decision makers face uncertainty regarding aspects of their environment,
see Hansen and Sargent (2008).
Rewriting forecasted variables as the composition of ex post realizations
and forecast errors, we introduce expectations errors into the system, which
becomes

(zt +1 , zt , υt +1 , ηt +1 ) = 0, (2.2)
where ηt is an r × 1 vector of expectational errors associated with intertem-
poral optimality conditions. Note that ηt = f (υt ); that is, expectational
errors arise from the realization of structural shocks. The deterministic
steady state of the model is expressed as z, defined in the context of (2.1)
as satisfying

(z, z, 0) = 0. (2.3)
The components of zt belong to one of three classifications: exogenous
state variables, endogenous state variables, and control variables. The lat-
ter are denoted by the nc × 1 vector ct , and represent optimal choices by
decision makers taking as given values of the state variables inherited in
period t . Exogenous state variables evolve over time independently of de-
cision makers’ choices, while the evolution of endogenous state variables
is influenced by these choices. Collectively, state variables are denoted by
the ns × 1 vector st , where nc + ns = n.
Often it will be useful to express variables as logged deviations from
steady state values. This is done using tildes:
 
zit


zit = ln .
zi
In turn, the vector xt denotes the collection of model variables written (un-
less indicated otherwise) in terms of logged deviations from steady state
values. So for a model consisting of output yt , investment it , and labor
hours nt , xt is given by
yt

xt = [
it nt ] .


2.1 Notation 11

2.1.1 Log-Linear Model Representations

Log-linear representations of structural models are expressed as
Ax t +1 = Bx t + C υt +1 + Dηt +1 , (2.4)
where the elements of A, B, C , and D are functions of the structural
parameters µ. Solutions of (2.4) are expressed as
xt +1 = F (µ)xt + G(µ)υt +1 . (2.5)
In (2.5), certain variables in the vector xt are unobservable, whereas
others (or linear combinations of variables) are observable. Thus filtering
methods such as the Kalman filter must be used to evaluate the system
empirically. The Kalman filter requires an observer equation linking ob-
servables to unobservables. Observable variables are denoted by Xt , where
Xt = H (µ) xt + ut , (2.6)
with
E(ut ut ) = u .
The presence of ut in (2.6) reflects the possibility that the observations of
Xt are associated with measurement error. Finally, defining
et +1 = G(µ)υt +1 ,
the covariance matrix of et +1 is given by
Q (µ) = E(et +1 et +1 ). (2.7)
Given assumptions regarding the stochastic nature of measurement er-
rors and the structural shocks, (2.5)–(2.7) yield a log-likelihood function
log L(X |), where  collects the parameters in F (µ), H (µ), u , and
Q (µ). Often it will be convenient to take as granted mappings from µ to
F , H , u , and Q . In such cases the likelihood function will be written as
L(X |µ).

2.1.2 Nonlinear Model Representations

Nonlinear approximations of structural models are represented using three
sets of equations, written with variables expressed in terms of levels (again,
possibly normalized to eliminate trend behavior). The first characterizes
the evolution of the state variables included in the model:
st = f (st −1 , υt ), (2.8)
where once again υt denotes the collection of structural shocks. The second
set of equations are known as policy functions, which represent the optimal
12 2 Casting Models in Canonical Form

specification of the control variables included in the model as a function of

the state variables:

ct = c(st ). (2.9)

The third set of equations map the full collection of model variables into
the observables:

Xt =

g (st , ct , υt , ut ) (2.10)
≡ g(st , ut ), (2.11)

where once again ut denotes measurement error. Parameters associated

with f (st −1 , υt ), c(st ), and g(st , ut ) are again obtained as mappings from µ,
thus their associated likelihood function is also written as L(X |µ).
Note that the solution to the log-linear system (2.5) can be converted
to a form analogous to (2.9) by solving for the control variables contained
in xt as functions of the state variables contained in xt . A straightforward
means of executing this conversion is presented in chapter 5.
In the remainder of this chapter, we outline procedures for mapping
nonlinear systems into (2.4). Then in chapter 3 we introduce the three
benchmark models that serve as examples through the remainder of the
text, and demonstrate their mapping into (2.4).

2.2 Linearization
2.2.1 Taylor Series Approximation
Consider the following simplification of (2.1) given by

(zt +1 , zt ) = 0, (2.12)

where we have abstracted momentarily from the stochastic component of

the model under investigation. We have done so because models are typi-
cally designed to incorporate stochastic behavior directly into the linearized
system (including those introduced in chapter 3). Also, expectational terms
need not be singled out at this point, because they receive no special
treatment in the linearization stage.
Before proceeding, note that although (2.12) is written as a first-order
system, higher-order specifications may be written as first-order systems
by augmenting zt to include variables observed at different points in time.
For example, the p th -order equation

ωt +1 = ρ1 ωt + ρ2 ωt −1 + · · · + ρp ωt −p+1
2.2 Linearization 13

can be written in first-order form as

⎡ ⎤ ⎡ ⎤⎡ ⎤
ωt +1 ρ1 ρ2 · · · · · · ρp ωt
⎢ ω ⎥ ⎢ 1 0 ··· ··· 0 ⎥⎢ ω ⎥
⎢ t ⎥ ⎢ ⎥⎢ t −1 ⎥
⎢ ⎥−⎢ ⎥⎢ ⎥ = 0,
⎢ .. ⎥ ⎢ .. .. .. ⎥ ⎢ .. ⎥
⎣ . ⎦ ⎣ . . ··· ··· . ⎦⎣ . ⎦
ωt −p+2 0 0 ··· 1 0 ωt −p+1

or more compactly, as

zt +1 − zt = 0, zt +1 = [ωt +1 , ωt , . . . , ωt −p+2 ] .

Thus (2.12) is sufficiently general to characterize a system of arbitrary

order.
The goal of the linearization step is to convert (2.12) into a linear system,
which can then be solved using any of the procedures outlined in chap-
ter 4. The reason for taking this step is that explicit solutions to (2.12)
are typically unavailable, rendering quantitative assessments of the system
problematic. For textbook discussions of the analysis of nonlinear systems,
see Azariadis (1993) and Sedaghat (2003).
In light of (2.4), the representation of the system we seek is given by

Ax t +1 = Bx t , (2.13)

where xt represents a transformation of zt . One way of establishing this

representation is via linearization, which is accomplished via a first-order
Taylor Series approximation of (2.12) around its steady state:

∂ ∂
0 ≈ (z) + (z) × (zt − z) + (z) × (zt +1 − z), (2.14)
∂zt ∂zt +1
∂
where (zt − z) is n × 1, and the n × n matrix ∂z t
(z) denotes the Jacobian
of (zt +1 , zt ) with respect to zt evaluated at z. That is, the (i, j )th element
∂
of ∂z t
(z) is the derivative of the i th equation in (2.12) with respect to the
j th element of zt . Defining

∂ ∂
A= (z), B=− (z), xt = (zt − z),
∂zt +1 ∂zt

we obtain (2.13), wherein variables are expressed as deviations from steady

state values. (As we shall see in chapter 4, it is also possible to work with
higher-order approximations of (2.12), e.g., following Schmitt-Grohé and
Uribe, 2004.)
14 2 Casting Models in Canonical Form

2.2.2 Log-Linear Approximations

It is often useful to work with log-linear approximations of (2.12), due in
part to their ease of interpretation. For illustration, we begin with a simple
example in which the system is 1×1, and can be written as
zt +1 = f (zt ).
Taking natural logs and noting that zt = e lnzt , the system becomes
ln zt +1 = ln[f (e ln zt )].
Then approximating,
f  (z)z
ln zt +1 ≈ ln[f (z)] + ( ln (zt ) − ln (z)),
f (z)
or because ln[f (z)] = ln z,
zt +1 f  (z)z zt
ln ≈ ln .
z f (z) z
f  ()z
Note that f () is the elasticity of zt +1 with respect to zt . Moreover, wri-
ting zt as z + εt , where εt denotes a small departure from steady state,
zt εt εt
ln = ln 1 + ≈ ,
z z z
and thus ln zzt is seen as expressing zt in terms of its percentage deviation
from steady state.
Returning to the n × 1 case, rewrite (2.12) as
1 (zt +1 , zt ) = 2 (zt +1 , zt ), (2.15)
because it is not possible to take logs of both sides of (2.12). Again using
zt = e ln zt , taking logs of (2.15) and rearranging yields
ln 1 (e ln zt +1 , e ln zt ) − ln 2 (e ln zt +1 , e ln zt ) = 0. (2.16)
The first-order Taylor Series approximation of this converted system yields
the log-linear approximation we seek. The approximation for the first
term is
∂ ln[1 ]  zt 
ln 1 (zt +1 , zt ) ≈ ln[1 (z)] + (z) × ln
∂ ln (zt ) z
∂ ln[1 ]  zt +1 
+ (z) × ln , (2.17)
∂ ln (zt +1 ) z
where ∂∂ ln[ 1] ∂ ln[1 ] zt
ln (zt ) (z) and ∂ ln (zt +1 ) (z) are n × n Jacobian matrices, and [ln ( z )]
zt +1
and [ln ( z )] are n × 1 vectors. The approximation of the second term
in (2.16) is analogous. Then defining
2.2 Linearization 15
 
∂ ln[1 ] ∂ ln[2 ]
A= (z) − (z) ,
∂ ln (zt +1 ) ∂ ln (zt +1 )
 
∂ ln[1 ] ∂ ln[2 ]
B=− (z) − (z) ,
∂ ln (zt ) ∂ ln (zt )
 zt 
xt = ln ,
z
we once again obtain (2.13). The elements of A and B are now elastici-
ties, and the variables of the system are expressed in terms of percentage
deviations from steady state.
In Part V we will discuss several empirical applications that involve the
need to approximate (2.12) or (2.16) repeatedly for alternative values of µ.
In such cases it is useful to automate the linearization stage via the use of
a numerical gradient calculation procedure. We introduce this briefly here
in the context of approximating (2.12); the approximation of (2.16) is
analogous.
Gradient procedures are designed to construct the Jacobian matrices
in (2.14) or (2.17) without analytical expressions for the required deriva-
tives. Derivatives are instead calculated numerically, given the provision of
three components by the user. The first two components are a specifica-
tion of µ and a corresponding specification of z. The third component
is a procedure designed to return the n × 1 vector of values ς generated
by (2.12) for two cases. In the first case zt +1 is treated as variable and zt is
fixed at z; in the second case zt is treated as variable and zt +1 is fixed at z.
The gradient procedure delivers the Jacobian

∂
(z) = A
∂zt +1

in the first case and

∂
(z) = −B
∂zt

in the second case.

Before turning to the fully specified example models provided in chap-
ter 3, we conclude here by demonstrating the linearization and log-
linearization of two example equations.

2.2.3 Example Equations

Consider the simple resource constraint

yt = ct + it ,
16 2 Casting Models in Canonical Form

indicating that output (yt ) can be either consumed (ct ) or invested (it ).
This equation is already linear. In the notation of (2.12) the equation
appears as

yt − ct − it = 0;

and in terms of (2.14), with

zt = [yt ct it ]
∂
and the equation representing the i th of the system, the i th row of ∂zt (z)
is given by
∂
(z) = [1 −1 −1].
∂zt
In the notation of (2.16), the equation appears as

ln yt − ln[exp ( ln ct ) − exp ( ln it )] = 0,

and in terms of (2.13), the i th row of the right-hand-side matrix B is

   
∂ ln[1 ] ∂ ln[2 ] −c −i
(z) − (z) = 1 . (2.18)
∂ ln (xzt ) ∂ ln (zt ) c+i c+i

Finally, to use a gradient procedure to accomplish log-linear approxima-

tion, the i th return of the system-evaluation procedure is

ςi = ln yt − ln[exp ( ln ct ) − exp ( ln it )].

As an additional example consider the Cobb-Douglas production

function

yt = at ktα nt1−α , α ∈ (0, 1),

where output is produced by the use of capital (kt ) and labor (nt ) and is
subject to a technology or productivity shock (at ). Linear approximation of
this equation is left as an exercise. To accomplish log-linear approximation,
take logs of the equation and rearrange to map into the notation of (2.16) as

ln yt − ln at − α ln kt − (1 − α) ln nt = 0.

With
 
yt at kt nt 
zt = ln ln ln ln ,
y a k n
2.2 Linearization 17

the i th row of the right-hand-side matrix in (2.13) is

 
∂ ln[1 ] ∂ ln[2 ]
(z) − (z) = [1 −1 −α −(1 − α)] . (2.19)
∂ ln (zt ) ∂ ln (zt )

And to use a gradient procedure to accomplish log-linear approximation,

the i th return of the system-evaluation procedure is

ςi = ln yt − ln at − α ln kt − (1 − α) ln nt .
Chapter 3

DSGE Models: Three Examples

Example is the School of Mankind, and they will

learn at no other.
—Edmund Burke, Thoughts on the Cause of the
Present Discontents

Chapter 2 introduced the generic representation of a DSGE model as

(Et zt +1 , zt , υt +1 ) = 0,

and provided background for mapping the model into a log-linear repre-
sentation of the form

Ax t +1 = Bx t + C υt +1 + Dηt +1 .

This chapter demonstrates the representation and completion of this map-

ping for three prototypical model environments that serve as examples
throughout the remainder of the text. This sets the stage for chapter 4,
which details the derivation of linear model solutions of the form

xt +1 = Fx t + Gυt +1 ,

and chapter 5, which details the derivation of nonlinear solutions of

the form

ct = c(st ),
st = f (st −1 , υt ).

For guidance regarding the completion of log-linear mappings for a far

broader range of models than those considered here, see Hansen and
Sargent (2005).
The first environment is an example of a simple real business cycle
(RBC) framework, patterned after that of Kydland and Prescott (1982).
The foundation of models in the RBC tradition is a neoclassical growth
environment, augmented with two key features: a labor-leisure trade-off
that confronts decision makers, and uncertainty regarding the evolution
of technological progress. The empirical question Kydland and Prescott
3 DSGE Models: Three Examples 19

(1982) sought to address was the extent to which such a model, bereft
of market imperfections and featuring fully flexible prices, could account
for observed patterns of business-cycle activity while capturing salient fea-
tures of economic growth. This question continues to serve as a central
focus of this active literature; overviews are available in the collection of
papers presented in Barro (1989) and Cooley (1995).
Viewed through the lens of an RBC model, business cycle activity is in-
terpretable as reflecting optimal responses to stochastic movements in the
evolution of technological progress. Such interpretations are not without
controversy. Alternative interpretations cite the existence of market imper-
fections, costs associated with the adjustment of prices, and other nomi-
nal and real frictions as potentially playing important roles in influencing
business cycle behavior, and giving rise to additional sources of business
cycle fluctuations. Initial skepticism of this nature was voiced by Summers
(1986), and the papers contained in Mankiw and Romer (1991) provide an
overview of DSGE models that highlight the role of, for example, market
imperfections in influencing aggregate economic behavior.
As a complement to the RBC environment, the second environment
presented here (that of Ireland, 2004a) provides an example of a model
within this New Keynesian tradition. The empirical purpose of the model
is to evaluate the role of cost, demand, and productivity shocks in driv-
ing business cycle fluctuations. Textbook references for models within this
tradition are Benassy (2002) and Woodford (2003). In addition, the sem-
inal papers of Christiano, Eichenbaum, and Evans (2005) and Smets and
Wouters (2003, 2007) are must-reads for those interested in models of
this class: based on the empirical performance of the models these papers
describe, New Keynesian models have emerged as important tools for pol-
icymakers, particularly central bankers. Yet, in turn, this emergence is not
without controversy: for a critique of the empirical performance of these
models, see Chari, Kehoe, and McGrattan (2009).
The realm of empirical applications pursued through the use of DSGE
models extends well beyond the study of business cycles. The third envi-
ronment serves as an example of this point: it is a model of asset-pricing be-
havior adopted from Lucas (1978). The model represents financial assets as
tools used by households to optimize intertemporal patterns of consump-
tion in the face of exogenous stochastic movements in income and divi-
dends earned from asset holdings. Viewed through the lens of this model,
two particular features of asset-pricing behavior have proven exceptionally
difficult to explain. First, LeRoy and Porter (1981) and Shiller (1981) used
versions of the model to underscore the puzzling volatility of prices associ-
ated with broad indexes of assets (such as the Standard & Poor’s 500), high-
lighting what has come to be known as the “volatility puzzle.” Second,
Mehra and Prescott (1985) and Weil (1989) used versions of the model to
20 3 DSGE Models: Three Examples

highlight the puzzling dual phenomenon of a large gap observed between

aggregate returns on risky and riskless assets, coupled with exceptionally
low returns yielded by riskless assets. These features came to be known
as the “equity premium” and “risk-free rate” puzzles. The texts of Shiller
(1989), Campbell, Lo, and MacKinlay (1997), and Cochrane (2001) pro-
vide overviews of literatures devoted to analyses of these puzzles.
In addition to the references cited above, a host of introductory
graduate-level textbooks serve as useful references for the specific example
models considered here, and for a wide range of extensions. A partial list-
ing includes Sargent (1987a, 1987b), Stokey and Lucas (1989), Blanchard
and Fischer (1998), Romer (2006), Ljungqvist and Sargent (2004), and
Wickens (2008).

3.1 Model I: A Real Business Cycle Model

3.1.1 Environment
The economy consists of a large number of identical households; aggregate
economic activity is analyzed by focusing on a representative household.
The household’s objective is to maximize U , the expected discounted flow
of utility arising from chosen streams of consumption and leisure:
∞

max U = E0 β t u(ct , lt ). (3.1)
ct ,lt
t =0

In (3.1), E0 is the expectations operator conditional on information avail-

able at time 0, β ∈ (0, 1) is the household’s subjective discount factor,
u(·) is an instantaneous utility function, and ct and lt denote levels of
consumption and leisure chosen at time t .
The household is equipped with a production technology that can be
used to produce a single good yt . The production technology is repre-
sented by
yt = zt f (kt , nt ), (3.2)
where kt and nt denote quantities of physical capital and labor assigned
by the household to the production process, and zt denotes a random
disturbance to the productivity of these inputs to production (that is,
a productivity or technology shock).
Within a period, the household has one unit of time available for division
between labor and leisure activities:
1 = nt + lt . (3.3)
3.1 Model I: A Real Business Cycle Model 21

In addition, output generated at time t can be either consumed or used

to augment the stock of physical capital available for use in the production
process in period t + 1. That is, output can be either consumed or invested:

yt = ct + it , (3.4)

where it denotes the quantity of investment. Finally, the stock of physical

capital evolves according to

kt +1 = it + (1 − δ)kt , (3.5)

where δ ∈ (0, 1) denotes the depreciation rate. The household’s problem

is to maximize (3.1) subject to (3.2)–(3.5), taking k0 and z0 as given.
Implicit in the specification of the household’s problem are two sets of
trade-offs. One is a consumption/savings trade-off: from (3.4), higher
consumption today implies lower investment (savings), and thus from
(3.5), less capital available for production tomorrow. The other is a la-
bor/leisure trade-off: from (3.3), higher leisure today implies lower labor
today and thus lower output today.
In order to explore quantitative implications of the model, it is necessary
to specify explicit functional forms for u(·) and f (·), and to characterize
the stochastic behavior of the productivity shock zt . We pause before do-
ing so to make some general comments. As noted, an explicit goal of
the RBC literature is to begin with a model specified to capture impor-
tant characteristics of economic growth, and then to judge the ability of
the model to capture key components of business cycle activity. From the
model builder’s perspective, the former requirement serves as a constraint
on choices regarding the specifications for u(·), f (·), and the stochastic
process of zt . Three key aspects of economic growth serve as constraints
in this context: over long time horizons the growth rates of {ct , it , yt , kt }
are roughly equal (balanced growth), the marginal productivity of capital
and labor (reflected by relative factor payments) are roughly constant over
time, and {lt , nt } show no tendencies for long-term growth.
Beyond satisfying this constraint, functional forms chosen for u(·)
are typically strictly increasing in both arguments, twice continuously
differentiable, strictly concave, and satisfy

∂u(ct , lt ) ∂u(ct , lt )
lim = lim = ∞. (3.6)
c→0 ∂ct l →0 ∂lt
Functional forms chosen for f (·) typically feature constant returns to scale
and satisfy similar limit conditions.
Finally, we note that the inclusion of a single source of uncertainty in
this framework, via the productivity shock zt , implies that the model carries
22 3 DSGE Models: Three Examples

nontrivial implications for the stochastic behavior of a single corresponding

observable variable. For the purposes of this chapter, this limitation is not
important; however, it will motivate the introduction of extensions of this
basic model in Part V.

FUNCTIONAL FORMS
The functional forms presented here enjoy prominent roles in the macro-
economics literature. Instantaneous utility is of the constant relative risk
aversion (CRRA) form:

ϕ 1−ϕ 1−φ
ct lt
u(ct , lt ) = , (3.7)
1−φ
or when φ = 1, u(·) = ln (·). The parameter φ > 0 determines two at-
tributes: it is the coefficient of relative risk aversion, and also determines
the intertemporal elasticity of substitution, given by φ1 (for textbook discus-
sions, see e.g., Blanchard and Fischer, 1998; or Romer, 2006). Note that
the larger is φ, the more intense is the household’s interest in maintaining
a smooth consumption/leisure profile. Also, ϕ ∈ (0, 1) indicates the im-
portance of consumption relative to leisure in determining instantaneous
utility.
Next, the production function is Cobb-Douglas:

yt = zt ktα nt1−α , (3.8)

where α ∈ (0, 1) represents capital’s share of output. Finally, the log of

the technology shock is assumed to follow a first-order autoregressive, or
AR(1), process:

ln zt = (1 − ρ) ln (z) + ρ ln zt −1 + εt (3.9)
2
εt ∼ NID(0, σ ), ρ ∈ (−1, 1). (3.10)

The solution to the household’s problem may be obtained via standard

application of the theory of dynamic programming (e.g., as described in
detail in Stokey and Lucas, 1989; and briefly in chapter 4 of this text).
Necessary conditions associated with the household’s problem expressed
in general terms are given by
   
∂u(ct , lt ) ∂u(ct , lt ) ∂f (kt , nt )
= × (3.11)
∂lt ∂ct ∂nt
  
∂u(ct , lt ) ∂u(ct +1 , lt +1 ) ∂f (kt +1 , nt +1 )
= βEt + (1 − δ) . (3.12)
∂ct ∂ct +1 ∂kt +1
3.1 Model I: A Real Business Cycle Model 23

The intratemporal optimality condition (3.11) equates the marginal ben-

efit of an additional unit of leisure time with its opportunity cost: the
marginal value of the foregone output resulting from the corresponding
reduction in labor time. The intertemporal optimality condition (3.12)
equates the marginal benefit of an additional unit of consumption today
with its opportunity cost: the discounted expected value of the additional
utility tomorrow that the corresponding reduction in savings would have
generated (higher output plus undepreciated capital).
Consider the qualitative implications of (3.11) and (3.12) for the im-
pact of a positive productivity shock on the household’s labor/leisure and
consumption/savings decisions. From (3.11), higher labor productivity
implies a higher opportunity cost of leisure, prompting a reduction in
leisure time in favor of labor time. From (3.12), the curvature in the house-
hold’s utility function carries with it a consumption-smoothing objective.
A positive productivity shock serves to increase output, thus affording an
increase in consumption; however, because the marginal utility of con-
sumption is decreasing in consumption, this drives down the opportunity
cost of savings. The greater is the curvature of u(·), the more intense is
the consumption-smoothing objective, and thus the greater will be the
intertemporal reallocation of resources in the face of a productivity shock.
Dividing (3.11) by the expression for the marginal utility of consump-
tion, and using the functional forms introduced above, these conditions
can be written as
   α
1 − ϕ ct kt
= (1 − α)zt (3.13)
ϕ lt nt
ϕ(1−φ)−1 (1−ϕ)(1−φ)
ct lt
   1−α 
ϕ(1−φ)−1 (1−ϕ)(1−φ) nt +1
= βEt ct +1 lt +1 αzt +1 + (1 − δ) .
kt +1
(3.14)

3.1.2 The Nonlinear System

Collecting components, the system of nonlinear stochastic difference equa-
tions that comprise the model is given by
   α
1−ϕ ct kt
= (1 − α)zt (3.15)
ϕ lt nt
   1−α 
nt +1
ctκ ltλ = βEt ctκ+1 ltλ+1 αzt +1 + (1 − δ) (3.16)
kt +1
24 3 DSGE Models: Three Examples

yt = zt ktα nt1−α (3.17)

yt = ct + it (3.18)
kt +1 = it + (1 − δ)kt (3.19)
1 = nt + lt (3.20)
ln zt = (1 − ρ) ln (z) + ρ ln zt −1 + εt , (3.21)
where
κ = ϕ(1 − φ) − 1
and
λ = (1 − ϕ)(1 − φ).
This is the specific form of the generic nonlinear model representation

(Et zt +1 , zt , υt +1 ) = 0
introduced in chapter 2, with the slight abuse of notation involving the
interpretation of zt . This representation serves as the point of departure
for deriving solutions of the form
ct = c(st ),
st = f (st −1 , υt ),
using the nonlinear solution methods described in chapter 5.
Steady states of the variables {yt , ct , it , nt , lt , kt , zt } may be computed
analytically from this system. These are derived by holding zt to its steady
state value z, which we set to 1:
y
= η,
n
c
= η − δθ,
n
i
= δθ,
n
1
n=  , (3.22)
1 1−ϕ
1+ 1−α ϕ 1 − δθ 1−α

l = 1 − n,

k
= θ,
n
3.1 Model I: A Real Business Cycle Model 25

where
  1
α 1−α
θ=
1/β − 1 + δ
η = θα.

Note that in steady state the variables {yt , ct , it , kt } do not grow over
time. Implicitly, these variables are represented in the model in terms of
deviations from trend, and steady state values indicate the relative heights
of trend lines. To incorporate growth explicitly, consider an alternative
specification of zt :

zt = z0 (1 + g)t e ωt , (3.23)
ωt = ρωt −1 + εt . (3.24)

Note that, absent shocks, the growth rate of zt is given by g, and that
removal of the trend component (1 + g)t from zt yields the specification for
ln zt given by (3.21). Further, the reader is invited to verify that under this
specification for zt , {ct , it , yt , kt } will have a common growth rate given by
g
1−α . Thus the model is consistent with the balanced-growth requirement,
and as specified, all variables are interpreted as being measured in terms of
deviations from their common trend.
One subtlety is associated with the issue of trend removal that arises in
dealing with the dynamic equations of the system. Consider the law of mo-
tion for capital (3.19). Trend removal here involves division of both sides
g t
of (3.19) by 1 + 1−α ; however, the trend component associated with
g t +1
kt +1 is 1 + 1−α ,so the specification in terms of detrended variables is
 
g
1+ kt +1 = it + (1 − δ)kt . (3.25)
1−α
Likewise, a residual trend factor will be associated with ct +1 in the
intertemporal optimality condition (3.16). Because ct +1 is raised to the
power

κ = ϕ(1 − φ) − 1,
g κ
the residual factor is given by 1 + 1−α :

ctκ ltλ = βEt

 κ   1−α 
g nt +1
× 1+ ctκ+1 ltλ+1 αzt +1 + (1 − δ) .
1−α kt +1
(3.26)
26 3 DSGE Models: Three Examples

With κ negative (ensured by φ1 < 1, i.e., an inelastic intertemporal elasticity

of substitution specification), the presence of g provides an incentive to
shift resources away from (t + 1) towards t .

Exercise 3.1
Rederive the steady state expressions (3.22) by replacing (3.19) with (3.25),
and (3.16) with (3.26). Interpret the intuition behind the impact of g on the
expressions you derive.

3.1.3 Log-Linearization
This step involves taking a log-linear approximation of the model at steady
state values. In this case, the objective is to map (3.15)–(3.21) into the
linearized system
Ax t +1 = Bx t + C υt +1 + Dηt +1
for eventual empirical evaluation. Regarding D, dropping Et from the Eu-
ler equation (3.16) introduces an expectations error in the model’s second
equation, therefore
D = [0 1 0 0 0 0 0] .
Likewise, the presence of the productivity shock in the model’s seventh
equation (3.21) implies
C = [0 0 0 0 0 0 1] .
Regarding A and B, these can be constructed by introducing the follow-
ing system of equations into a gradient procedure (where time subscripts
are dropped so that, for example, y = yt and y  = yt +1 ):
 
1−ϕ
0 = ln + ln c  − ln l  − ln (1 − α) − ln z 
ϕ
− α ln k  + α ln n  (3.27)
0 = κ ln c + λ ln l − ln β − κ ln c  − λ ln l  (3.28)
  
 exp[(1 − α) ln n ]
− ln α exp ( ln z ) + (1 − δ)
exp[(1 − α) ln k  ]
0 = ln y  − ln z  − α ln k  − (1 − α) ln n  (3.29)
  
0 = ln y − ln{exp[ln (c )] + exp[ln (i )]} (3.30)

0 = ln k − ln{exp[ln (i)] + (1 − δ) exp[ln (k)]} (3.31)
0 = − ln{exp[ln (n  )] + exp[ln (l  )]} (3.32)

0 = ln z − ρ ln z. (3.33)
3.1 Model I: A Real Business Cycle Model 27

The mapping from (3.15)–(3.21) to (3.27)–(3.33) involves four steps.

First, logs of both sides of each equation are taken; second, all variables
not converted into logs in the first step are converted using the fact, for ex-
ample, that y = exp ( ln (y)); third, all terms are collected on the right-hand
side of each equation; fourth, all equations are multiplied by −1. Deriva-
tives taken with respect to ln y  , and so on evaluated at steady state values
yield A, and derivatives taken with respect to ln y, and so on yield −B.
Having obtained A, B, C , and D, the system can be solved using any
of the solution methods to be presented in chapter 4, yielding a system of
the form
xt +1 = F (µ)xt + et +1 .

Exercise 3.2
With xt given by
 
yt ct it nt lt kt zt 
xt = ln , ln , ln , ln , ln , ln , ln
y c i n l k z
and
µ = [α β φ ϕ δ ρ σ ] = [0.33 0.975 2 0.5 0.06 0.9 0.01] ,

show that the steady state values of the model are y = 0.90, c = 0.70, i =
0.21, n = 0.47, l = 0.53, and k = 3.49 (and take as granted z = 1). Next,
use a numerical gradient procedure to derive
⎡ ⎤
0 1 0 0.33 −1 −0.33 −1
⎢ ⎥
⎢0 1.5 0 −0.06 0.5 0.06 −0.08⎥
⎢ ⎥
⎢1 0 0 −0.67 0 −0.33 −1 ⎥
⎢ ⎥
⎢ ⎥
A = ⎢1 −0.77 −0.23 0 0 0 0 ⎥,
⎢ ⎥
⎢0 0 0 0 0 1 0 ⎥
⎢ ⎥
⎢ ⎥
⎣0 0 0 −0.47 −0.53 0 0 ⎦
0 0 0 0 0 0 1
⎡ ⎤
0 0 0 0 0 0.33 0
⎢ ⎥
⎢0 1.5 0 0 0.5 0 0 ⎥
⎢ ⎥
⎢0 0 0 0 0 0 0 ⎥
⎢ ⎥
⎢ ⎥
B = ⎢0 0 0 0 0 0 0 ⎥.
⎢ ⎥
⎢0 0 0.06 0 0 0.94 0 ⎥
⎢ ⎥
⎢ ⎥
⎣ 0 0 0 0 0 0 0 ⎦
0 0 0 0 0 0 0.9
28 3 DSGE Models: Three Examples

Exercise 3.3
Rederive the matrices A and B given the explicit incorporation of growth
in the model. That is, derive A and B using the steady state expressions
obtained in Exercise 3.1, and using (3.25) and (3.26) in place of (3.19)
and (3.16).

3.2 Model II: Monopolistic Competition and

Monetary Policy
This section outlines a model of imperfect competition featuring “sticky”
prices. The model includes three sources of aggregate uncertainty: shocks
to demand, technology, and the competitive structure of the economy.
The model is due to Ireland (2004a), who designed it to determine how
the apparent role of technology shocks in driving business cycle fluc-
tuations is influenced by the inclusion of these additional sources of
uncertainty.
From a pedagogical perspective, the model differs in two interesting
ways relative to the RBC model outlined above. Whereas the linearized
RBC model is a first-order system of difference equations, the lineari-
zed version of this model is a second-order system. However, recall that it
is possible to represent a system of arbitrary order using the first-order form

Ax t +1 = Bx t + C υt +1 + Dηt +1 ,

given appropriate specification of the elements of xt . Second, the problem

of mapping implications carried by a stationary model into the behavior
of nonstationary data is revisited from a perspective different from that
adopted in the discussion of the RBC model. Specifically, rather than as-
suming the actual data follow stationary deviations around deterministic
trends, here the data are modeled as following drifting random walks;
stationarity is induced via differencing rather than detrending.

3.2.1 Environment
The economy once again consists of a continuum of identical households.
Here, there are two distinct production sectors: an intermediate-goods
sector and a final-goods sector. The former is imperfectly competitive: it
consists of a continuum of firms that produce differentiated products that
serve as factors of production in the final-goods sector. Although firms in
this sector have the ability to set prices, they face a friction in doing so.
Finally, there is a central bank.
3.2 Model II: Monopolistic Competition and Monetary Policy 29

HOUSEHOLDS
The representative household maximizes lifetime utility defined over con-
sumption, money holdings, and labor:
∞
 ξ

 mt n
max U = E0 β t
at ln ct + ln − t (3.34)
ct ,mt ,nt pt ξ
t =0
bt
s .t . pt ct + + mt = mt −1 + bt −1 + τt + wt nt + dt , (3.35)
rt

where β ∈ (0, 1) and ξ ≥ 1. According to the budget constraint (3.35), the

household divides its wealth between holdings of bonds bt and money mt ;
bonds mature at the gross nominal rate rt between time periods. The
household also receives transfers τt from the monetary authority and
works nt hours in order to earn wages wt to finance its expenditures. Finally,
the household owns an intermediate-goods firm, from which it receives a
dividend payment dt . Note from (3.34) that the household is subject to
an exogenous demand shock at that affects its consumption decision.
Recognizing that the instantaneous marginal utility derived from con-
sumption is given by actt , the first-order conditions associated with the
household’s choices of labor, bond holdings, and money holdings are
given by
 
at
wt ξ −1
= nt (3.36)
pt
ct
     
1 at +1 1 at
βEt = (3.37)
pt +1 ct +1 rt pt ct
 −1       
mt 1 at +1 1 at
+ βEt = . (3.38)
pt pt +1 ct +1 pt ct

Exercise 3.4
Interpret how (3.36)–(3.38) represent the optimal balancing of trade-offs
associated with the household’s choices of n, b, and m.

FIRMS
There are two types of firms: one produces a final consumption good yt ,
which sells at price pt ; the other is a continuum of intermediate-goods firms
that supply inputs to the final-good firm. The output of the i th interme-
diate good is given by yit , which sells at price pit . The intermediate goods
combine to produce the final good via a constant elasticity of substitution
30 3 DSGE Models: Three Examples

(CES) production function. The final-good firm operates in a competitive

environment and pursues the following objective:
 1
F
max t = pt yt − pit yit di (3.39)
yit 0
  θt
1 θt −1 θt −1
θt
s .t . yt = yit di . (3.40)
0

The solution to this problem yields a standard demand for intermediate

inputs and a price aggregator:
 −θt
pit
yit = yt (3.41)
pt
  1
1 1−θt
pt = pit1−θt di . (3.42)
0

Notice that θt is the markup of price above marginal cost; randomness in θt

provides the notion of a cost-push shock in this environment.
Intermediate-goods firms are monopolistically competitive. Because
the output of each firm enters the final-good production function sym-
metrically, the focus is on a representative firm. The firm is owned by
the representative household; thus its objectives are aligned with the
household’s. It manipulates the sales price of its good in pursuit of these
objectives, subject to a quadratic adjustment cost:
∞
   
I at dt
max it = E0 βt , (3.43)
pit ct pt
t =0

s .t . yit = zt nit (3.44)

 −θt
pit
yit = yt (3.45)
pt
 2
φ pit
χ(pit , pit −1 ) = − 1 yt , φ > 0, (3.46)
2 π pit −1

where π is the gross inflation rate targeted by the monetary authority

(described below), and the real value of dividends in (3.43) is given by
 
dt pit yit − wt nit
= − χ(pit , pit −1 ) . (3.47)
pt pt
3.2 Model II: Monopolistic Competition and Monetary Policy 31

The associated first-order condition can be written as

 −θt
pit yt
(θt − 1)
pt pt
    
pit −θt −1 wt yt 1 pit yt
= θt − φ −1 − βφ
pt pt zt pt π pit −1 π pit −1
   
at +1 ct pit +1 yt +1 pit +1
×Et −1 . (3.48)
at ct +1 π pit π pit2

The left-hand side of (3.48) reflects the marginal revenue to the firm gen-
erated by an increase in price; the right-hand side reflects associated
marginal costs. Under perfect price flexibility (φ = 0) there is no dynamic
component to the firm’s problem; the price-setting rule reduces to

θt wt
pit = ,
θt − 1 zt

which is a standard markup of price over marginal cost wztt . Under “sticky
prices” (φ > 0) the marginal cost of an increase in price has two addi-
tional components: the direct cost of a price adjustment, and an expected
discounted cost of a price change adjusted by the marginal utility to the
households of conducting such a change. Empirically, the estimation of φ
is of particular interest: this parameter plays a central role in distinguishing
this model from its counterparts in the RBC literature.

THE MONETARY AUTHORITY

The monetary authority chooses the nominal interest rate according to a
Taylor Rule. With all variables expressed in terms of logged deviations from
steady state values, the rule is given by

rt = ρr
πt + ρg

rt −1 + ρπ
g t + ρo

o t + εrt , εrt ∼ iidN (0, σr2 ), (3.49)

where
πt is the gross inflation rate,
g t is the gross growth rate of output, and


o t is the output gap (defined below). The ρi parameters denote elasticities.
The inclusion of
rt −1 as an input into the Taylor Rule allows for the gradual
adjustment of policy to demand and technology shocks, for example, as in
Clarida, Gali, and Gertler (2000).
The output gap is the logarithm of the ratio of actual output yt to
capacity output  y t . Capacity output is defined to be the “efficient” level of
32 3 DSGE Models: Three Examples

output, which is equivalent to the level of output chosen by a benevolent

social planner who solves:
⎧  ξ ⎫
 ∞ ⎨ 1 1 ⎬
max U S = E0 β t at ln  yt − nit di (3.50)

y t ,nit ⎩ ξ 0 ⎭
0
  θt
1 θt −1 θt −1
θt
s .t . 
y t = zt nit di . (3.51)
0

The symmetric solution to this problem is simply

1
y t = atξ zt .
 (3.52)

STOCHASTIC SPECIFICATION
In addition to the monetary policy shock εrt introduced in (3.49), the
model features a demand shock at , a technology shock zt , and a cost-push
shock θt . The former is iid; the latter three evolve according to

ln (at ) = (1 − ρa ) ln (a) + ρa ln (at −1 ) + εat , a>1 (3.53)

ln (zt ) = ln (z) + ln (zt −1 ) + εzt , z>1 (3.54)
ln (θt ) = (1 − ρθ )ln (θ) + ρθ ln (θt −1 ) + εθ t , θ > 1, (3.55)

with |ρi | < 1, i = a, θ. Note that the technology shock is nonstationary:

it evolves as a drifting random walk. This induces similar behavior in the
model’s endogenous variables, and necessitates the use of an alternative to
the detrending method discussed above in the context of the RBC model.
Here, stationarity is induced by normalizing model variables by zt . For the
corresponding observable variables, stationarity is induced by differenc-
ing rather than detrending: the observables are measured as deviations of
growth rates (logged differences of levels) from sample averages. Details
are provided in the linearization step discussed below.
The model is closed through two additional steps. The first is the im-
position of symmetry among the intermediate-goods firms. Given that the
number of firms is normalized to one, symmetry implies:

yit = yt , nit = nt , pit = pt , dit = dt . (3.56)

The second is the requirement that the money and bond markets clear:

mt = mt −1 + τt (3.57)
bt = bt −1 = 0. (3.58)
3.2 Model II: Monopolistic Competition and Monetary Policy 33

3.2.2 The Nonlinear System

In its current form, the model consists of 12 equations: the household’s
first-order conditions and budget constraint, the aggregate production
function, the aggregate real dividends paid to the household by its
intermediate-goods firm, the intermediate-goods firm’s first-order con-
dition, the stochastic specifications for the structural shocks, and the
expression for capacity output. Following Ireland’s (2004a) empirical im-
plementation the focus is on a linearized reduction to an eight-equation
system consisting of an IS curve, a Phillips curve, the Taylor Rule (specified
in linearized form in (3.49)), the three exogenous shock specifications, and
definitions for the growth rate of output and the output gap.
The reduced system is recast in terms of the following normalized
variables:
yt ct y pt
ÿt = , c̈t = ,  y¨t = t , πt = ,
zt zt zt pt −1
(dt /pt ) (wt /pt ) (mt /pt ) zt
d̈t = , ẅt = , m̈t = , z̈t = .
zt zt zt zt −1
Using the expression for real dividends given by (3.47), the household’s
budget constraint in equilibrium is rewritten as
φ πt 2
ÿt = c̈t + −1 ÿt . (3.59)
2 π
Next, the household’s first-order condition (3.37) is written in normalized
form as
 
at at +1 1 1
= βrt Et × × . (3.60)
c̈t c̈t +1 z̈t +1 πt +1
Next, the household’s remaining first-order conditions, the expression for
the real dividend payment (3.47) it receives, and the aggregate production
function can be combined to eliminate wages, money, labor, dividends,
and capacity output from the system. Having done this, we then introduce
the expression for the output gap into the system:
yt ÿt
ot ≡ = 1. (3.61)

yt
atξ
Finally, normalizing the first-order condition of the intermediate-goods
firm and the stochastic specifications leads to the following nonlinear
system:
φ πt 2
ÿt = c̈t + −1 ÿt (3.62)
2 π
34 3 DSGE Models: Three Examples
 
at at +1 1 1
= βrt Et × × (3.63)
c̈t c̈t +1 z̈t +1 πt +1
c̈t ξ −1 πt πt
0 = 1 − θt + θt ÿt −φ −1
at π π
 
c̈t at +1 πt +1 πt +1 ÿt +1
+ βφEt −1 (3.64)
c̈t +1 at π π ÿt
z̈t ÿt
gt = (3.65)
ÿt −1
yt ÿt
ot = = 1 (3.66)

yt
atξ
ln (at ) = (1 − ρa ) ln (a) + ρa ln (at −1 ) + εat (3.67)
ln (θt ) = (1 − ρθ ) ln (θ) + ρθ ln (θt −1 ) + εθ t (3.68)
ln (z̈t ) = ln (z) + εzt . (3.69)

Along with the Taylor Rule, this is the system to be log-linearized.

3.2.3 Log-Linearization
Log-linearization proceeds with the calculation of steady state values of the
endogenous variables:

z
r= π,
β
 1
ξ
θ −1
c=y = a ,
θ
 1
ξ
θ −1
o= ; (3.70)
θ

(3.62)–(3.69) are then log-linearized around these values. As with the RBC
model, this can be accomplished through the use of a numerical gradient
procedure. However, as an alternative to this approach, here we follow
Ireland (2004a) and demonstrate the use of a more analytically oriented
procedure. In the process, it helps to be mindful of the reconfiguration
Ireland worked with: an IS curve, a Phillips curve, the Taylor Rule, the
shock processes, and definitions of the growth rate of output and the out-
put gap.
Random documents with unrelated
content Scribd suggests to you:
reverence whatever is offered in simple obedience, the most
desirous themselves to learn simply to obey, will also be the first to
feel that no one should venture to break the silence in which inward
prayer may be arising from other hearts except under the influence
(to use the time-honoured Quaker expression) of “a fresh anointing
from above.” The nearest approach to a description of what we hold
to be a right ministry would seem to be—words spoken during, and
arising from, actual communion with God.

118
 CHAPTER V.
SPECIAL TESTIMONIES.

There are certain points of Christian practice upon which we have

been accustomed to lay a degree of stress amounting to peculiarity,
although our “testimony” in regard to them does not involve any
opposition to the beliefs of other Christian bodies, as does that
which we have just been considering respecting the freedom of the
ministry and the disuse of ordinances. The idea of “testimony,” or
practical witness-bearing to a stricter obedience to the teaching of
Jesus Christ than is thought necessary by the mass of those who are
called by His name, has been strongly impressed upon Friends from
the very outset, and the persecutions which it brought upon them
did but burn it irrevocably into the Quaker mind.

The preaching of the early Friends was, above all things, a preaching
of righteousness. I think I cannot be wrong in saying that a greater
value has from the first been attached by Friends to practice, 119
as compared with doctrine, than is the case with most other
Christian bodies. Obedience to the light which convinces of sin was
the sum and substance of George Fox’s preaching, and through his
epistles and other writings there runs a vigorously practical tone
which seems to have been responded to with equal vigour by those
whom he addressed.

The early Friends certainly did, as a rule, wonderfully practise what

they preached; and their character for integrity was very quickly, and
has been permanently, recognized. It seems to myself inevitable that
the appeal to the witness in each heart should reach deeper, and
bring forth correspondingly better fruit of obedience, when
disentangled from all reliance on external passports to Divine favour.
Not only the idea of any possible “efficacy of sacraments” as apart
from righteousness of life, but also the idea of “substitution” as
distinguished from actual experience of the transforming power of
the righteousness of Christ, were vigorously rejected by the early
Friends; and in this insistence upon the identity of righteousness
with salvation lay, as I believe, the main secret of their strength. At
any rate, there has been a remarkably steady endeavour to maintain
a high and definite standard of Christian morality, partly by 120
means of the discipline of the Society, partly by family
tradition, discipline, and example. Certain “testimonies,” i.e. practices
conscientiously adopted, inculcated, and watched over, have been
handed down from generation to generation with a jealous care
which, though sometimes overshooting its mark and tending to
produce reaction, has nevertheless moulded the very inmost springs
of action, and produced and maintained a distinct and somewhat
singular type of Christian character.

The essence of Quaker “testimony” is a practical witness-bearing—a

lifting up in practice of the highest possible standard of
uncompromising obedience to the teaching of Jesus Christ, both as
recorded in the Gospels, and as inwardly experienced as the Light—
the Spirit of Truth. Friends have, as a matter of fact, felt certain
things to be inconsistent with this teaching which, by the great body
of those who profess and call themselves Christians, are not
regarded as being so. They have attacked these things not so much
in words as by enjoining and observing a strict abstinence from them
at any cost, in a spirit not unlike that of the Rechabites of old. The
original Quaker idea was before all things to have “clean hands;” to
stand clear of evil in one’s own person, but to abstain in 121
[16]
silence unless specially called to speak.

It is, of course, impossible to abstain on conscientious grounds from

what is freely practised by others without giving some offence. Any
singularity of this kind will inevitably be understood as casting some
shade of disapprobation, if not of actual blame, on the common
practice. I do not see how we can avoid this offence unless we are
content to sink to the level of least enlightenment. We must, I
believe, nerve ourselves to endure the giving of it, remembering that
the disciple is not above his Master, and that there was a time when
our Master Himself had “no honour in His own country.” If they have
heard His word, they will hear ours also. Meanwhile we may take
heart from the knowledge that conduct destined to have permanent
influence must often displease for a time.

The early Friends, or “children of light,” as they sometimes called

themselves, seem to have been drawn together in a kind of
spontaneous unanimity on the main points in which their view of
Christian duty transcended that of those about them. The 122
Yearly Meeting, which was not constituted till 1672 (or
twenty-four years from the date of George Fox’s beginning to
preach), finding the “testimonies” against war, oaths, and
superfluities already in full practice, expressly recognized them as
belonging to “our Christian profession,” and directed inquiry as to
the faithfulness of Friends in maintaining them to be made in certain
queries addressed from time to time to all the subordinate meetings.

The practice of addressing such queries to the subordinate meetings

is maintained to this day, although the queries themselves have from
time to time been altered, and of late years the greater number of
them are directed to be seriously considered, but not answered. This
change in our practice has probably not been without a balance of
advantage and disadvantage. The system of requiring answers to
the queries was, in truth, a very powerful engine of discipline, for
they were considered and answered with scrupulous care and
precision, and, in case of an unfavourable report, individuals who
were regarded as failing to maintain the testimonies of the Society
were liable to be “put under dealing,” and, in case of obstinacy, to
eventual disownment. This ultimate penalty of disobedience was in
former times inflicted for much slighter causes than would at 123
[17]
the present day render any one liable to it.
It seems to me that there is a serious danger inherent in the very
nature of collective testimonies, especially those which imply the
lifting up of a standard of exceptional severity and purity, lest that
which is in some, perhaps even in the majority, sincere and
spontaneous should be adopted at second hand, and without
personal warrant, by others, and should thus become a mere hollow
profession. For this reason I am thankful that a much greater degree
of freedom is now allowed to our members in all matters as to which
there is room for a conscientious difference of opinion. Our strength
seems to me to depend largely upon our consistency in appealing to
the gospel rather than to the law—in trusting to the purifying power
of an indwelling, informing Spirit, rather than to any external
framework of regulations. To do anything which can stimulate the
profession of more or higher enlightenment than is actually
possessed, is indeed a signal and a grievous departure from 124
our own avowed principles; and I believe that no surer
method could be devised for bringing our Society into disrepute and
decay than the attempt to require a pre-arranged uniformity with
regard to those special testimonies which imply special
enlightenment.

The loftier and more delicate the ideal, the greater the risk of
formulating and attempting to impose it. It seems to me that our
wisdom is to insist more and more boldly upon obedience to the
broad plain laws of duty which all Christians recognize as laid down
for us by those recorded words of our Master Himself, which are our
one supreme standard; and at the same time to leave more and
more scope for the working out in detail of all the lovely and
harmonious yet varying results of individual faithfulness to the
promptings of His Spirit in each heart. Any distinct breach of the
moral law, any falling below that standard of “peaceable innocent
life” which is acknowledged by all as the test of the reality of light
within, may and should surely be made a matter of Church discipline
—a matter, that is, in which brethren should watch over one another
for good, and obedience to which must be a condition of sound
fellowship. But when discipline descends to the regulation of details
whose whole significance and value depend upon their being 125
prompted by conscience, under the living and ever-present
guidance of the light, then surely the human is intruding upon the
province of the Divine, and we are checking and hampering and
weakening that very moulding from within which it is our chief
object as a Society to watch for and to yield to in all things.

But while all attempts at collective self-discipline must involve a

danger of hollowness, which means weakness, if not actual
insincerity, it is to be remembered, on the other hand, that there is
in association not only a well-known source of strength, but a very
valuable shelter; a protection to right instincts of modesty. In
maintaining any exceptionally high standard of action, especially in
matters of detail, there is a real safety as well as comfort in united
action. While we are treading in the steps of our honoured
predecessors, however freely we may have chosen our path, we are
not tempted to claim that we discovered it; nor need we anxiously
vindicate it as though it were but the prompting of some individual
scruple or preference. In the practical results of the collective
exercises of a considerable body of fellow-disciples, we do, I believe,
in fact find, as we might reasonably expect to find, a peculiar 126
purity and propriety. It is not difficult to justify the wisdom by
which our special testimonies have been worked out, though it is
easy also to see the mischief of too rigid an enforcement of them.

Our Society, like the United Kingdom, enjoys the elasticity resulting
from the absence of any written constitution, and the precise
working of its discipline is by no means easily traced. Its
“testimonies,” though clearly recognized and notorious, are not
formulated or defended in any authoritative document. The “Book of
Discipline” consists, as I have said, of extracts from “Epistles” and
“Advices” circulated from time to time by the Yearly Meeting. These
are in the nature of brotherly exhortations, which assume our
principles as undisputed; and though carefully worded, they do not
deal in definitions or arguments. Our testimonies are, in fact, to a
degree which is, I think, hardly understood outside the Society, the
result of individual and spontaneous obedience to the bidding of
individual conscience, and to the guiding of the Divine light shining
in each heart, rather than of conformity to rules enforced or even
precisely laid down by any human authority. They are collective, but
unformulated; subjects for discipline, yet not prescribed or
regulated; familiar and even notorious peculiarities, yet 127
varying indefinitely in the degree in which they are
maintained by individuals.

The traditional reverence for individual conscience is still so strong

that the precise nature of the obstacle felt by Friends to any
particular course of conduct is apt to be shrouded in some degree of
mystery. The phraseology of the Society, which is almost a separate
language, vividly conveys this sense of an insuperable but (to
outsiders) mysterious restraint. “Truth requires” that certain things
should be done or left undone. A Friend “feels a stop in his mind,” or
“is not easy to proceed with” some undertaking. Such and such a
thing “appears” (to John Woolman, for instance) “to be
distinguishable from pure wisdom.”

There are, as is well known, individual Friends who have abundantly

argued, on general grounds, the moral questions involved in our
“testimonies.” Friends have never been wanting in pugnacity,
whatever their scruples as to the use of “carnal weapons” or of
violent language. But yet their practice has in the main been felt out
rather than thought out; their testimonies are instances of problems
solved by going forward rather than of theories built up through any
speculative process; and in regard to each one of them every 128
true Friend feels that to his own Master he stands or falls,
and that there is but one Example to which he ought to look, and
one Guide whom he desires to obey. As in regard of our public
ministry, so in the lesser matters of everyday life and practice, we
jealously guard our individual liberty from human interference in
order that it may be the more unreservedly subjected to all Divine
influences.
And not only do we guard our own liberty—we refrain from
attempting to limit that of others. If our conscientious abstinence
from certain practices is inevitably understood as in some sense
casting a shadow of reproach or blame upon those practices, we yet
are careful to abstain from condemning those who are acting in
obedience to their own measure of light. I believe we must with
boldness and humility acknowledge that such practical witness-
bearing as we believe ourselves called to implies that we are as “a
city set on a hill.” We do not attempt to lay down rules applicable at
once and equally to all. The homeward road cannot be altogether
the same for dwellers on the hill and dwellers on the plain; the goal
[18]
alone is one.

The most important and the best known of the special 129
testimonies of which I have now to speak is that which has
been steadily borne by our members against all war. Friends have
ever maintained and acted upon the belief that war and strife of all
kinds are opposed to the spirit and the teaching of Christ, and have
felt themselves, as His disciples, precluded from engaging in them.
They have steadfastly refused to take up arms at the bidding of any
human authority, or in the presence of any danger. This course of
conduct has, of course, brought them into frequent collision with the
civil power, and needs for its justification, as Friends are the first to
acknowledge, the warrant of a higher than any national authority.

It is, indeed, an awful position which we have thus been bold to take
up—the position of those who feel themselves called upon to act as
the salt of the earth, as leaders who refuse to be led. I do not
hesitate to confess that this attitude of possible resistance to the
demands of our country in the presence of a common danger was
the one part of the Quaker ideal which I for a time seriously
hesitated to accept. So long as I understood it to be 130
accompanied by or based upon any condemnation of those
who conscientiously believe that their duty to God requires them to
yield unqualified obedience to the demands of their country for
military service, I was unable to accept it. But when I came to
understand that the Quaker testimony against all war did not take
the form of any ethical theory of universal application, but was
simply the acting out in one’s own person and at one’s own risk of
obedience to that which one’s own heart had been taught to
recognize as Divine authority, even where its commands transcended
and came into collision with those of the nation, I felt at once that
the position was not only perfectly tenable, but was the only one
worthy of faithful disciples.

So long as our country is but very imperfectly Christianized, it is

impossible not to recognize that an insuperable contradiction may at
any time arise between the demands upon our loyalty and obedience
which may be made in its name, and those of the spirit of Christ. It
would assuredly not be acting in His spirit to make light of
disobedience to law; but neither can any Christian hesitate for a
moment when called upon to choose which Master he shall obey. It
seems to me that if any man be prepared in the true spirit of a
martyr to rise above his country’s sense of right, and to serve 131
his country in the highest sense by disobeying and
withstanding such of its requirements as in his heart he believes to
be wrong and ungodly, it is impossible to withhold from such a man
the respect and the admiration which we all feel for the martyrs of
old. I do not see how the national standard of duty can be raised—
how, in other words, the nation can ever be thoroughly Christianized
—except through individual faithfulness, at all costs and at all risks,
to a higher view of duty than that held by the nation at large.

Here, of course, we are confronted with the question, Is our view of

duty truly a higher one than that of the nation at large? Does the
teaching of Jesus Christ really call us to abstain from all warfare?

It seems to me that not only Friends’ testimony, but the teaching of

our Master Himself on this subject have been much, and in a sense
inevitably, misunderstood. The subject is profoundly complex, and
much of what is said and written about it sounds altogether
unsubstantial and unpractical, because neither the depth and
intricacy of the evil, nor the far-reaching and full significance of the
principles opposed to it, are sufficiently felt. “I say unto you, That ye
resist not evil: but whosoever shall smite thee on thy right cheek,
turn to him the other also.” Surely this does not point to an 132
abject submission or a tame indifference, but to an
undaunted persistence in blessing—a fearless overcoming of evil
with good. It is an appeal to an unchanging and fundamental
principle, rather than a mere rule of conduct. The whole passage
breathes a spirit of ardent confidence in the supremacy of goodness;
in the power of the Perfect One who makes His sun to shine upon
the just and the unjust, and sends rain upon the unthankful and the
evil; it is a call to us to be perfect, even as our Father is perfect; not
a suggestion that we should abandon or relax our conflict with evil,
but an assurance that we are not at its mercy—that He who is with
us is stronger than all they who can be against us, and that in His
strength we can and must meet evil with good and overcome it.

Those who would follow Him who thus spoke, must rise above all
personal considerations, and above every temptation to retaliate—
not fall below them. This, surely, was the spirit in which William
Penn won his victories in the early days of Pennsylvania—the
bloodless victories which make his name to this day a word of love
and honour amongst the Indians, with whom his treaty of peace was
never broken. It was not by lying down like sheep to be 133
slaughtered by them, but by going forward to meet them
with open hands and a trusting heart, and by honourably and
generously recognizing their rights and paying them a fair price for
their lands, that he and his followers turned suspicion and hatred
into firm friendship.

We are called to rise above the level of fighting pagans, not to fall
below it. There is, indeed, a lower depth than that of the military
spirit—the depth of complacent mammon-worship. To our shame be
it confessed that this spirit may clothe itself under the profession of
“non-resistance.” When the salt so loses its savour, it is truly fit for
nothing but to be cast out and trodden underfoot. But we are
concerned here not with the deplorable caricature of that “testimony
against all war” which has for two hundred years been at once the
boast and the reproach of Quakerism, but with its essence and true
significance. These lie in the fact that Friends have, one by one,
individually and unitedly, been led by obedience to the spirit of Christ
to abstain from fighting and from all concern, so far as it has been
possible to clear themselves from it, in strife of any kind. This is
surely clear and solid ground to take, and quite distinct from any
attempt or necessity for laying down general and comprehensive
formulæ of conduct applicable to all cases, to all persons and 134
all bodies. To formulate such general rules is, in truth, foreign
to the spirit of Quakerism. To yield one’s self unreservedly to Divine
guidance; resolutely, and at whatever cost, to refuse to participate in
that which one’s own conscience has been taught to condemn;—this
is the ancient and inestimable Quaker ideal. It is surely the best, the
most effectual, the most Christian way of witnessing against evil,
and of arousing the consciences of others.

There is not, I believe, any possibility of dispute, I will not say

amongst Christians, but amongst rational beings, as to the enormity
of the evil of strife and discord, whether between nations or
between individuals. The question upon which we Friends differ from
other Christians is not the question whether peace be desirable—
whether it be not, in fact, the goal of all political effort—but what are
the means by which it is to be attained or maintained. Other
Christians do not deny that quarrelling is contrary to the spirit of
Christ, and we do not deny that a holy warfare is to be continually
maintained against evil in every form. But we regard the opposing of
violence by violence as a suicidal and hopeless method of
proceeding; we feel, as Christians, that the weapons of our warfare
are not carnal. We cannot, by taking military service, place 135
ourselves at the absolute disposal of a power which may at
any time employ its soldiers for purposes so questionable and often
so unhallowed.
To abstain, on these grounds, from all participation in warfare is
surely a quite different thing from laying down any general theory as
to the “unlawfulness” of war. I own that it does not appear to me to
be right or wise to blame those who are acting in obedience to their
own views of duty, however much they may differ from our own. I
do not think it can serve any good purpose to ignore the force of the
considerations by which war appears to many people to be justified.
I would even go further, and admit that, under all the complicated
circumstances of the world (including historical facts and treaty
obligations), there are cases in which men may be actually bound to
fight in what they believe to be a just cause; although it does not, I
believe, follow that every individual would be justified in taking part
in such warfare. Would any one say that at the time of the Indian
Mutiny the Governor-General of India ought not to have permitted
the use of arms for the protection of the women and children? I
doubt whether any Friend would be found to maintain this. But it is
equally to be remembered that no true Friend could well 136
have occupied the position of the Governor-General. No
nation which had from the beginning of its history been thoroughly
Christian could, I suppose, have found itself in the position which we
occupied in India in 1857. Were all the world, in the true and full
sense of the word, Christian, such events obviously would not occur.
Had we been from the first a thoroughly Christian nation, our whole
history must have been different, and would (as we Friends believe)
have been infinitely nobler.

We do not profess to lay down any general rule, by obedience to

which war can be instantly dispensed with by nations in their
unregenerate state, and without a sacrifice. A fully Christian nation
has never yet been seen on earth. It may well be that such a nation,
could it now come suddenly into existence, would meet with national
martyrdom. Meanwhile it is the imperfection of our Christianity and
the mixed and complex nature of national affairs which make it so
difficult to apply to national action any pure principles of conduct.
This is not to deny the existence of such principles. To recognize the
difficulty, nay, the impossibility, of suddenly or sweepingly applying
them to practice, is not to deny their leavening power. When 137
we Friends speak of what is “right,” we refer not to any
external and rigid rule of conduct, but to that which in each
individual case is truly the best and the highest course open to the
individual. To say this is not to say that right is in itself variable. It is
only to say what will, I think, be denied by few, that the human ideal
of right is progressive, not stationary.

We do, however, further say, what undoubtedly is denied by many,

that the ideal revealed to us in the life and the words of Jesus Christ
our Lord, and gradually being worked out in His own people through
His ever-present inward influence, is the highest and the purest
conceivable; and that, therefore, all real progress must be an
approach towards Him. It is this Christian ideal which, as it gains
possession of the human mind, must extinguish the spirit which
leads to strife and warfare.

It is commonly supposed that Friends have some special scruple

about the use of physical force in any case. This is, I believe, by no
means true of the Society at large, although the popular notion may
very likely be founded upon fact as regards individuals. As a body,
Friends have always recognized “the just authority of the civil
magistracy,” and have, I believe, never disputed the lawfulness of
the use of “the sword” (whatever may be meant by that 138
[19]
expression) in maintaining that authority. George Fox
himself repeatedly reminded magistrates that they should not “bear
the sword in vain,” but that they should use it for the punishment of
evil-doers, not of those who did well.

It is not, as I understand it, the use of physical force, or even the

suffering caused by the use of it, which really makes war hateful in
Christian eyes; but the evil passions, the “lusts” from which it
springs, and to which, alas! it so hideously ministers. The
dispassionate infliction of punishment by an impartial and a lawful
authority surely stands upon a quite different footing from that
“biting and devouring one another” which, whether between nations
or between individuals, it is the very aim and object of law to
suppress. Suffering inflicted for the purpose of maintaining peace
cannot, I think, be condemned by the advocates of peace unless it
be on the ground of failure.

I own that I personally cannot but recognize that upon this 139
view certain wars appear to be not only inevitable but
justifiable, as partaking of the nature of national police operations. I
cannot, therefore, regard all war as wholly and unmitigatedly
blamable, although I can hardly imagine any war which does not
both come from evil and lead to evil.

Again, there are treaty obligations requiring us in certain cases to

take up arms for the protection of weaker nations, from which we
could not suddenly recede without a breach of national good faith. It
surely does not become us, in our zeal for peace, to make light of, or
overlook, such considerations as this. They should, I think, in the
first place lead us to abstain from sweeping generalizations, and
from blaming those who are ready to lay down their lives in
obedience to their country’s call and in our defence, or the defence
of the oppressed in other countries; while yet we resolutely maintain
our own obedience to that higher authority by which we have, as we
believe, been taught a better way—a way incompatible with outward
strife—of giving our lives for the common weal. We should be very
careful how we call that a wicked action which a good man may
honestly do in obedience to his own sense of duty; we should be still
more careful lest, while professing to take higher ground, we 140
do in fact fall far short of such men in our lives; but we must,
for all that, be faithful to the light we have.

And, in the second place, it seems to me that the true inference

from the consideration of the complicated conditions of international
affairs is that the time is not yet ripe for the assumption of all offices
of public authority by thorough-going Christians. Our place surely
still is mainly to leaven, not to govern, the world.
The world must become the kingdom of the Lord and of His Christ
before wars and fightings will cease from amongst men. And the
world is very far as yet from acknowledging that dominion in
anything but words. Yet we do “profess and call ourselves
Christians”; we do live in the full light of the everlasting gospel; and
however it may be rejected or discarded, however far even those
who profess it may be from entering into its spirit, it has yet, in spite
of ourselves, raised us out of the possibility of consistent paganism.
We cannot return to the old condition of things, in which nations
thought it no shame to strive each for its own petty objects, and to
be reckless of each other’s interests. There is no satisfactory resting-
place for us now on any lower level than that which Christ has
brought to light. Under the name of “altruism,” this is 141
recognized even by those who think of Christianity as a worn-
out superstition. We who believe it to be the power of God unto
salvation are surely bound to yield ourselves heart and soul to its
emancipating influence—to its indestructible, irresistible appeal to us
to “live as brethren.” As in the beginning it was felt by some at least
to be as clear as daylight that “Christians cannot fight,” so now, not
only amongst Friends, but in many another Christian body, the same
spirit is working, and consciences are awakening to the utter
incompatibility of strife and retaliation and reckless self-
aggrandizement with the spirit of brotherhood which lies at the very
foundation of Christianity. They had need to awake; now, at the
eleventh hour, with all Europe making itself ready for war, it may yet
be that the few in whom the fire of Christian zeal is burning in its
purity may see their cause and the cause of their Master begin to
prevail against overwhelming odds. But whether the nations will hear
or whether they will forbear, wherever two or three Christians meet
together, there still will be a protest against strife and selfishness.

A protest against strife and selfishness; not only against strife, but
against “the greedy spirit which leads to strife.” If we are willing to
go down to the root in this matter, if we truly desire to do 142
what in us lies towards ridding the earth not only of wars and
fightings, but of all forms of oppression and cruelty, must we not
recognize that the very first step is to be ourselves freed from
covetousness?

For who can doubt that it is mainly about outward and material
things that nations or individuals are led into quarrels? Who will
venture to say that, if none of us desired either to get or to keep
more than our share of this world’s goods, there would be anything
like the amount of fighting, or of preparation for it, which now
devastates the earth? We may be skeptical about the possibility of
any general acceptance of arbitration or disarmament. To be
skeptical about the possibility of personal disinterestedness would
imply a very different sort of blindness. It seems to me that in
struggling to rise and to raise others more and more clearly above
the greedy spirit which leads to war, is the best hope for many of us
of contributing in any real sense to the cause of peace on earth.

It was long ago recognized by Friends that (to use the words of John
Woolman) “in every degree of luxury are the seeds of war and
oppression.” The connection between luxury and cruelty is, indeed,
almost a truism, but it is one of those truisms of which it is
unfortunately easy to lose sight; and I fear that even 143
amongst Friends the familiar testimonies against all war and
against superfluities are apt to be held without any vivid sense of
their vital connection.

No one, surely, will deny that the selfish desire of mere pleasure,
when allowed to rule, will feed itself at the expense of suffering and
privation to others; that it does cause that scramble for gain in
which the weak are trampled upon, and every furious passion is
[20]
stimulated. The difficulty in regard to bearing a practical
testimony against superfluities is not that which some of us feel in
the case of war—that we do not know where to take hold, that our
personal and daily conduct seems to have no immediate bearing
upon questions of international policy, and that the whole problem
eludes our grasp by its very vastness. It is, rather, that we do not
like to put our shoulder to the wheel of simplifying life for ourselves
and others; that we do not see the beauty of severity; that we love
softness, or yield to it for want of any purifying fire of hope.

But yet, in one form or another, often extravagantly, foolishly, 144

even injuriously, an ineradicable instinct has prompted
Christians in all times to free themselves from luxurious and self-
indulgent ways of living; to walk as disciples of Him who “had not
where to lay His head;” to lay aside, not only every sin, but every
weight, that so they may run the race set before them, not as
beating the air, but as those who strive for the victory.

It is, indeed, not easy to define the precise kind or amount of luxury
which is incompatible with Christian simplicity; or rather it must of
necessity vary. But the principle is, I think, clear. In life, as in art,
whatever does not help, hinders. All that is superfluous to the main
object of life must be cleared away, if that object is to be fully
attained. In all kinds of effort, whether moral, intellectual, or
physical, the essential condition of vigour is a severe pruning away
of redundance. Is it likely that the highest life, the life of the
Christian body, can be carried on upon easier terms?

The higher our ideal of life, the greater, indeed, must be the
sacrifices which it will require from us. As we rise from the lower to
the higher objects of life, many things of necessity become
superfluous to us—in other words, we become independent of them,
or outgrow them. This is a widely different idea from that of ascetic
self-discipline or self-mortification; and it is surely a sounder 145
and a worthier idea.

The Quaker ideal, as I understand it, requires a continual weighing

of one thing against another—a continual preference of the lasting
and deep over the transient and superficial. “Weightiness” is one of
the Friends’ characteristic and emphatic forms of commendation. To
sacrifice any deep and substantial advantage to outward show is
abhorrent to the Quaker instinct. To “stretch beyond one’s compass”
grasping at shadows, and encumbering oneself with more than is
needed for simple, wholesome living, is at variance with all our best
traditions.

If we bear in mind the essentially relative meaning of the word

“superfluous,” it is obvious that such a testimony against
“superfluities” does not require any rigid or niggardly rule as to
outward things. To my own mind, indeed, this view of the matter
seems to require at least as clearly the liberal use of whatever is
truly helpful to “our best life” as the abandonment of obstructing
superfluities. No doubt a testimony against superfluities is very liable
to degenerate into formality, and to be so misapplied as to cut off
much that is in reality wholesome, innocent, and beautiful. Art has
to a great extent been banished from many Quaker homes; 146
and a considerable amount of injury has no doubt been done
by such rigid severity, and perhaps still more by the very natural
consequent reaction. But it would, I believe, be quite a mistake to
suppose that the extreme plainness in dress and other surroundings
adopted by the stricter Friends, and formerly made a matter of
discipline by the Society, was originally adopted with any intention of
[21]
self-mortification or asceticism.

I believe that asceticism is in a very deep sense contrary to the real

Quaker spirit, which desires in all things to abstain from any
interference “in the will of man” with Divine discipline and guidance,
and which would, I believe, regard the idea of self-chosen exercises
in mortification of the flesh with the same aversion as it entertains
for pre-arranged forms of worship. Friends, no doubt, have often
believed themselves required to submit to the adoption of the plain
dress “in the cross” to natural inclination, and have felt it a valuable
exercise to do so; but the plainness was not devised for that
purpose, but chosen (or rather, as Friends would say, they were led
into it by Truth) because of its inherent suitableness and 147
rightness. It is an outcome of the instinctively felt necessity
of subordinating everything to principle. Its chief significance is that
of a protest against bondage to passing fashions, and for this reason
it is a settled costume. It is also felt that our very dress should show
forth that inward quietness of spirit which does not naturally tend
towards outward adornment, and the Friends’ recognized dress is
therefore one of extreme sobriety in colour and simplicity in form.

It is a significant fact that there is really no such thing as a precisely

defined Quaker costume. The dress is certainly precise enough in
itself, and to the naked eye of the outside observer it may appear to
present an undeviating uniformity; but it is really not a uniform in
the sense in which a nun’s or a soldier’s dress is a uniform. It is in all
respects a growth, a tradition, a language; and it is subject to
constant though slow modification. Any perfectly unadorned dress of
quiet colour, without ornament or trimming, if habitually worn, is in
fact, to all intents and purposes, the Quaker costume, though one or
two details have by a sort of accident acquired a traditional meaning
as a badge, which one may adopt or not according to one’s feeling
about badges. Some Friends nowadays object on principle to 148
anything of the kind. Others still see a “hedge” or shelter in
them. Others, again, feel that they serve a useful and innocent
purpose in enabling Friends readily to recognize one another, and
that it is not amiss for them to be easily recognized even by
outsiders. But the one important matter of principle which the
Society as a body have recognized, is that it is a waste of time and
money for which Christian women can hardly fail to find better
employment, to condescend to be perpetually changing the fashion
of one’s garments in obedience to the caprice or the restlessness of
the multitude. “Plain Friends” are those who are resolved to dress
according to the settled principles which commend themselves to
their own mind, not enslaving themselves to passing fashion.

It is easy to say that they do but exchange one bondage for another.
That may, indeed, have been the case at times, and may even still
be so in some families or meetings. But the crystallizing into rigid
formality, though a possible tendency, is no real part of the true
Quaker ideal. My own strong feeling is that the adoption of a settled
costume, at any rate in mature life and from conviction, is not only
the right and most dignified course on moral grounds, but also that
it has in actual experience afforded one more proof of the 149
truth that the lower aims of life can thrive only in proportion
as they are kept in subordination to the higher. The freedom from
the necessity of perpetual changes, which commends itself to
Friends as suitable to the dignity of “women professing godliness,”
has also the lower advantage of admitting a gradual bringing to
perfection of the settled costume itself. We all know how exquisite,
within its severely limited range, can be the result. The spotless
delicacy, the precision and perfection of plain fine needlework, the
repose of the soft tints, combine, in the dress of some still lingering
representatives of the old school of Quakerism, to produce a result
whose quiet beauty appeals to both the mind and the eye with a
peculiar charm. I cannot think that such mute eloquence is to be
despised; or that it is unworthy of Christian women to be careful
that their very dress shall speak a language of quietness, gentleness,
and purity—that it shall be impressive even with a touch of eternity.

This principle of Christian simplicity should, in our view, run through

everything—dress, furniture, habits of life, and forms of speech; all
should be severely purged from redundance, and from mere
imitation and conventionality. The “plain language,” best known as
leading to the use of thee and thou for you in speaking to 150
one person, and of first, second, etc., for the days of the
week and the months, instead of the ordinary names “derived from
heathen deities,” was an instance of this endeavour to winnow away
every superfluity and every taint of flattery and superstition from our
speech. These special peculiarities of speech are, as is well known,
completely dropped by many of the present generation of Friends.
The changes which have taken place in two hundred years in our
language and habits have deprived these expressions of much of
their original significance, and the tendency of the present time is no
doubt towards the effacing of all peculiarities. But some special
attention is still paid amongst us to simplicity and guardedness of
language in a wider sense, and surely this is an object well worthy of
attention on the highest as well as the lowest grounds.

The idea of a scrupulous guard over the lips, which is so strongly

characteristic of all Friends at all worthy of the name, culminates in
their united testimony against oaths. This has, indeed, been always
regarded by Friends as a matter of simple obedience to a plain
command of Jesus Christ; and I think that nothing but long habit
could reconcile any sincere disciple to the ordinary interpretation of
[22]
His words as intended to forbid “profane swearing” only. 151
Many others besides Friends have felt this scruple; but to our
Society belongs the indisputable credit of having, through a long and
severe course of suffering for their “testimony,” obtained a distinct
recognition of the sufficiency in their case of a plain affirmation,
thereby vindicating a principle which is beginning to be generally
recognized—the principle of having but one rule for all cases, that of
plain truth; of being as much bound by one’s word as one’s bond. I
think it can hardly be questioned that, through this simple and
unflinching course of obedience to the plain injunction of Jesus
Christ, Friends have done much to raise the standard of veracity in
[23]
our country.

The refusal to pay tithes is a part of the testimony against a paid

ministry, of which I have sufficiently spoken in the last chapter; and
I need here only say that in all these cases of resistance to the
demands of authority, for military service, for oaths, or for tithes, the
idea has been that of witnessing at one’s own cost against 152
unjust or unrighteous demands. It is, I think, fair to claim
that it is at one’s own cost that one refuses a demand even for
money when it is made by those who have the power to take the
money or its equivalent by force, and when no resistance is ever
offered to their doing so. Friends have again and again submitted
patiently to the levying of much larger sums than those originally
claimed, as well as to severe and sometimes lifelong imprisonments,
and other penalties, rather than by any act of their own give consent
to exactions which they believed to be unrighteous in their origin or
purpose. While such unmistakable proofs of disinterestedness were
given, the motive for withholding money could hardly be
misunderstood. With regard to tithes, however, the circumstances
have, since the Tithes Commutation Act, become so complicated,
that few Friends now feel a refusal to pay them a suitable method of
testifying against a paid ministry, and the Yearly Meeting has placed
on record this sense of the alteration of the state of the case in a
[24]
minute dated 1875: “This meeting believes that the time has
arrived when the mode of bearing this testimony must be left to the
individual consciences of Friends.”

Amongst lesser matters as to which Friends have made a 153

stand upon principle against prevailing customs, may be
mentioned “the superstitious observance of days,” especially that of
fasts or thanksgivings prescribed by the civil government (a power
which we do not regard as competent to prescribe religious
exercises), and the practice of wearing mourning, and placing
“inscriptions of a eulogistic character” on tombstones. In Friends’
burial-grounds nothing beyond the name, and the dates of birth and
death, is permitted. The objections to wearing mourning are
obvious, both on the ground of unnecessary expense and trouble at
a time when the mind should surely be left as much as possible
undisturbed, and also on that of its being an expression (and an
expression so formal as to be of doubtful sincerity) of grief and
gloom in regard to providential dispensations which, however
painful, we should desire to accept with cheerful submission. There
is obviously much to be said for this application of the principle of
simplifying our customs, and adjusting our dress and other
surroundings to the permanent rather than the transient
circumstances of our lives.

To simplify life to the very uttermost—is not this truly in itself a

worthy aim; nay, the one inexorable condition of excellence?
We have just now been engaged with comparatively trivial 154
matters—straws which show as no more solid thing can do
which way the wind blows. These things are important, not in
themselves, but in relation to the principles in honest obedience to
which they have been worked out. Simplicity—“the simplicity which
is in Christ”—the simplicity, not of exclusion, but of Divine all-
subduing supremacy—this is the keynote of our ideal; and it is a
keynote to which the human heart must always in some degree
respond. At the bottom of all art, of all beauty, and surely, we may
say, of all goodness, lies the principle of subordination—the
necessity of a perpetual choice between the permanent and the
transient, the essential and the superficial. Quakerism is an honest
endeavour to carry out this principle in the Christian life; to weigh “in
the balance of the sanctuary” the meat that endureth against the
meat that perisheth; to cleave to the eternal at the sacrifice, if
necessary, of all that is temporal.

I am, of course, not absurd enough to claim that this endeavour is

peculiar to Quakerism. My object throughout is to show what are the
eternal and unassailable principles of truth to which Quakerism
appeals, to which it clings as to its strongholds. And I believe that
the severe sifting away of non-essentials which lies at the 155
foundation of our revolt from accepted ecclesiastical
practices, and which has ramified in detail into these minor
testimonies, often rigidly and at times even laughably worked out by
individuals, is a process more and more urgently needed in these
days of rapid growth in all material and intellectual resources.

The permanent danger of giving our labour and our lives “for that
which satisfieth not” was surely never more desperate than in these
days of hurry and fulness, when merely to stand still needs a
resolute effort of will. Are not half the lives we know carried along in
a current they know not how to resist towards objects they but
vaguely recognize, and in their heart of hearts do not value? Was the
bondage of outward things ever more oppressive than it is to many
of those who are ostensibly, and ought to be really, in a position of
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com