scispace - formally typeset
Search or ask a question
Journal ArticleDOI

A retrospective on friedman's theory of permanent income*

01 Jun 2004-The Economic Journal (Institute for Fiscal Studies)-Vol. 114, Iss: 496, pp 293-306
TL;DR: Friedman's Permanent Income Hypothesis (PIF) is one of the great works of economics as mentioned in this paper, and it has been used extensively in modern economics and discusses some relevant empirical results and the way they relate to the original approach.
Abstract: Friedman’s book on the “Consumption Function” is one of the great works of Economics demonstrating how the interplay between theoretical ideas and data analysis could lead to major policy implications. We present a short review of Friedman’s Permanent Income Hypothesis, the origins of the idea and its theoretical foundations. We give a brief overview of its influence in modern economics and discuss some relevant empirical results and the way they relate to the original approach taken by Friedman.

Summary (2 min read)

Introduction

  • The theory of the Consumption function played an important role in explaining why traditional Keynesian demand management, through transitory tax policy or other transitory income boosting measures can have little or no effect on real consumption and on the desired policy outcomes.
  • The apparently simple ideas in this excellent book have been so insightful and powerful that they have given rise to a huge amount of research, both theoretical and empirical, which continues to this date.
  • Particularly that which has been based on microeconomic data, and I demonstrate the relevance of these ideas for their current way of thinking about consumption, savings and income processes.

A statement of the Hypothesis

  • Milton Friedman’s PI hypothesis originates from the basic intuition that individuals would wish to smooth consumption and not let it fluctuate with short run fluctuations in income.
  • The basic hypothesis posited is that individuals consume a fraction of this permanent income in each period and thus the average propensity to consume would equal the marginal propensity to consume.
  • He brings together empirical results and statistical theory, and develops new results by combining these with his economic ideas.
  • By comparing averages over time or across different groups of individuals, Friedman is implicitly using Instrumental variables, now a well recognised technique for dealing with classical measurement error in linear models.
  • Wald (1940) suggested comparing group averages to overcome measurement error, although the way he approached grouping was not quite right.

The notions of Permanent Income

  • The PIH provides a flexible framework for the study of consumption and savings.
  • Much of the subsequent research has filled in those elements necessary for explaining aspects of the data, thus defining the agenda for research on consumption and savings.
  • The basic hypothesis requires a rolling horizon, which allows some degree of smoothing of consumption.
  • The flexibility of the hypothesis comes at a price, since it is hard to define the theoretical underpinnings of the model, and consequently it is hard to make more detailed statements about policy.
  • Friedman points this out in the Consumption function book when he states that that permanent income is best defined “… to be whatever seems to correspond to consumer behaviour.”.

The theoretical foundations of the PIH

  • There have been other attempts to explain consumption behaviour.
  • The key implication of this model ttt E R cU λ β =)(' is that the marginal utility of consumption in each period is equal to the expected marginal utility of wealth multiplied by a factor depending on the interest rate and the rate of time preference.
  • Whether this will be detectable in the data will depend on the proportion of individuals who are thus constrained.
  • This is not too say that this is not important, but to lay the ground for arguing that a more general version of the PIH, with similar foundations is in fact consistent with the data, although the version of the PIH developed here is probably too restrictive.

Risk aversion, and the PIH

  • Given that markets are never found to be complete (e.g. Cochrane, 1991) and Attanasio and Davis, 1996) and given aggregate uninsurable shocks this must be a central issue.
  • While these ideas were not formalised at the time they turn out to be key elements in demonstrating that some generalised version of the PIH does in fact fit the data very well and can explain observed consumption patterns.
  • It may for example be reasonable to think that more consumption is shifted towards the household when it consists of more members.
  • Heckman (1974) pointed out that if consumption is not additively separable from hours of work and this is ignored, the resulting consumption choices over time will look as if they track income, in a way that would reject the basic premise of the permanent income hypothesis.
  • The study of income processes has been motivated in part by the study of consumption and savings because it has become apparent that knowing the time series properties of income may be informative about consumption behaviour.

Conclusions

  • Most important discoveries and insights are simple, economical, have important implications for a broad range of issues and withstand the test of time.
  • Moreover, they generate large amounts of research, verifying it and refining it.
  • This is exactly the case with Friedman’s PIH.
  • At the end of all this, the original idea has not only survived, but has formed the basis for developing a coherent analysis of consumption and savings.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

A RETROSPECTIVE ON FRIEDMANS
THEORY OF PERMANENT INCOME
Costas Meghir
THE INSTITUTE FOR FISCAL STUDIES
WP04/01

A Retrospective on Friedman’s Theory of Permanent Income
Costas Meghir
1
University College London and Institute for Fiscal Studies
November 2002
This Version January 2004
Abstract
Friedman’s book on the “Consumption Function” is one of the great works of Economics
demonstrating how the interplay between theoretical ideas and data analysis could lead to
major policy implications. We present a short review of Friedman’s Permanent Income
Hypothesis, the origins of the idea and its theoretical foundations. We give a brief
overview of its influence in modern economics and discuss some relevant empirical
results and the way they relate to the original approach taken by Friedman.
1
Acknowledgements: This article was prepared for a conference in honour of Milton Friedman at the
University of Chicago in November 2002, on the occasion of his 90
th
Birthday. I thank the organisers for
the invitation. I also thank Orazio Attanasio, Richard Blundell, Martin Browning, Jim Heckman, Hide
Ichimura, an anonymous referee and colleagues at IFS and UCL for useful discussions that helped me
organise this presentation. I am of course responsible for all errors and interpretations.

Introduction
Friedman’s book on the “Consumption Function” is one of the great works of Economics
demonstrating how the interplay between theoretical ideas and data analysis could lead to
major policy implications. The theory of the Consumption function played an important
role in explaining why traditional Keynesian demand management, through transitory tax
policy or other transitory income boosting measures can have little or no effect on real
consumption and on the desired policy outcomes. The apparently simple ideas in this
excellent book have been so insightful and powerful that they have given rise to a huge
amount of research, both theoretical and empirical, which continues to this date. In this
short article I trace out some of the research relating to the Permanent Income Hypothesis
(PIH), particularly that which has been based on microeconomic data, and I demonstrate
the relevance of these ideas for our current way of thinking about consumption, savings
and income processes.
Friedman and Kuznets (1954) “Incomes from Independent Professional Practice” which
was actually written in the early 1940s but delayed in publication, first formulated many
of the ideas of the PIH, including the permanent/transitory decomposition of income in
the volume. The core of these ideas, together with further tests, was brought together in
Friedman’s “Consumption function”. In my review of the PIH I draw mainly from this
latter work.

The Permanent Income Hypothesis
A statement of the Hypothesis
Milton Friedman’s PI hypothesis originates from the basic intuition that individuals
would wish to smooth consumption and not let it fluctuate with short run fluctuations in
income. In fact the model was developed to explain important empirical facts in a unified
framework. For example, why is income more volatile than consumption and why is the
long run marginal propensity to consume out of income higher than the short run one. To
answer these questions Friedman hypothesized that individuals base their consumption on
a longer term view of an income measure, perhaps a notion of lifetime wealth or a notion
of wealth over a reasonably long horizon. The basic hypothesis posited is that individuals
consume a fraction of this permanent income in each period and thus the average
propensity to consume would equal the marginal propensity to consume. The propensity
itself could vary with a number of factors, including the interest rate and taste shifter
variables, or could reflect uncertainty – we will return to these important insights below.
Friedman set himself the task of testing his hypothesis against an increasing set of
empirical facts from time series data and budget studies. The standard least squares
regression of consumption on income would always point to a marginal propensity to
consume below the average propensity. Conditioning on extra regressors seems to make
things worse. It is at this point that Friedman’s ingenuity, brought together the literature

on budget studies by Margaret Reid, Morgan and others
2
, as well as time series analyses
with Econometric ideas on measurement error to devise estimation techniques, that not
only allowed the testing of the basic hypothesis, but led to the estimation of underlying
parameters that directly characterised the Permanent Income Hypothesis (PIH). As far as
the role of measurement error is concerned the works of great economists and
statisticians of the time, namely Harold Hotelling
3
and James Durbin, influenced
Friedman. He brings together empirical results and statistical theory, and develops new
results by combining these with his economic ideas.
The ingredients of Friedman’s model are permanent consumption (
p
c
), permanent
income (
p
y ), transitory consumption (
t
c ), transitory income (
t
y ). Measured income is
the sum of
permanent and transitory income (
t
y
) and measured consumption is the sum
of
permanent and transitory consumption (
t
c
), i.e.
tp
ccc +=
and
tp
yyy +=
Permanent consumption is determined by the equation
pp
yzrkc ),(=
where
),( zrk is the average (or marginal) propensity to consume out of permanent
income which depends on the rate of interest and on taste shifter variables z. The
2
See Reid (1952) and Morgan (1951).
3
See Hotelling (1933) and Friedman (1992) on the regression fallacy.

Citations
More filters
01 Jan 1987

268 citations

Report SeriesDOI
TL;DR: In this article, the authors discuss recent developments in the literature that studies how the dynamics of earnings and wages affect consumption choices over the life cycle, highlighting the role of persistence, information, size and insurability of changes in economic resources.
Abstract: We discuss recent developments in the literature that studies how the dynamics of earnings and wages affect consumption choices over the life cycle. We start by analyzing the theoretical impact of income changes on consumption—highlighting the role of persistence, information, size and insurability of changes in economic resources. We next examine the empirical contributions, distinguishing between papers that use only income data and those that use both income and consumption data. The latter do this for two purposes. First, one can make explicit assumptions about the structure of credit and insurance markets and identify the income process or the information set of the individuals. Second, one can assume that the income process or the amount of information that consumers have are known and test the implications of the theory. In general there is an identification issue that has only recently being addressed with better data or better “experiments”. We conclude with a discussion of the literature that endogenizes people’s earnings and therefore change the nature of risk faced by households.

216 citations

Posted Content
TL;DR: For example, if Alice out-earns Bob by a dollar, then on average she will outspend him by at least 90 cents, and that's only the beginning as discussed by the authors.
Abstract: Suppose you believe your economy is in the doldrums because people are somehow not spending enough. How do you get them to open up their pocketbooks? Start by perusing some data. You’ll quickly discover that spending is highly correlated with income. It’s well documented that if, in any given year, Alice out-earns Bob by a dollar, then on average she’ll outspend him by at least 90 cents.1 Aha! Problem solved! If you want people to spend more, you should start by raising their incomes. Encourage your government to hire Alice and raise her salary by a dollar. She’ll spend an extra 90 cents or so—and that’s only the beginning. If she spends that 90 cents at the butcher shop or the hair salon or the craft brewery, then the butcher or the beautician or the brewer earns an extra 90 cents and probably spends about 90 percent of that, which raises yet someone else’s income, and off we go. When all is said and done, one dollar of additional government spending can raise total spending (and total income) by $10 or more. That’s the story of the so-called “Keynesian multiplier.” Once upon a time, pretty much all economists considered it a cornerstone of policymaking. Here’s the problem: Income is indeed highly correlated with spending. But correlation is not causation. When Alice out-earns Bob by a dollar, she typically outspends him

157 citations

Journal ArticleDOI
TL;DR: In this article, the authors present new evidence that households are unable to protect themselves from rainfall failure that occurs on average every five years in rural Ethiopia and that other less extreme rainfall variat...
Abstract: We present new evidence that households are unable to protect themselves from rainfall failure that occurs on average every five years in rural Ethiopia. However, other less extreme rainfall variat...

113 citations


Cites background from "A retrospective on friedman's theor..."

  • ...Townsend (1994) tested the full insurance hypothesis in the development context and a body of research has emerged on this theme (see Ravallion and Chaudhuri, 1997; Bardhan and Udry, 1999 for an overview)....

    [...]

  • ...In a parallel literature, various empirical specifications of the Permanent Income Hypothesis have been tested (Friedman, 1957; for a review, see Meghir, 2004), and its insights have become widely used in models of intertemporal choice investigating whether and how consumption is smoothed when…...

    [...]

Posted Content
TL;DR: In this article, the authors introduce and apply a general framework for evaluating long-term income distributions according to the equality of opportunity principle, which allows for both an exante and an ex-post approach to EOP.
Abstract: In this paper, we introduce and apply a general framework for evaluating long-term income distributions according to the Equality of Opportunity principle. Our framework allows for both an ex-ante and an ex-post approach to EOp. Our ex-post approach relies on a permanent income measure defined as the minimum annual expenditure an individual would need in order to be as well off as he could be by undertaking inter-period income transfers. There is long-term ex-post inequality of opportunity if individuals who exert the same effort have different permanent incomes. In comparison, the ex-ante approach focuses on the expected permanent income for individuals with identical circumstances. Hence, the ex-ante approach pays attention to inequalities in expected permanent income between different types of individuals. To demonstrate the empirical relevance of a long-run perspective on EOp, we exploit a unique panel data from Norway on individuals' incomes over their working lifespan.

100 citations


Cites background from "A retrospective on friedman's theor..."

  • ...A benchmark case in intertemporal choice theory uses the annuity of the income stream as measure of permanent income (see e.g. Meghir, 2004)....

    [...]

  • ...intertemporal choice theory uses the annuity of the income stream as measure of permanent income (see e.g. Meghir, 2004 )....

    [...]

References
More filters
ReportDOI
TL;DR: In this article, the authors argue that a life cycle model can replicate observed patterns in household wealth accumulation after accounting explicitly for precautionary saving and asset-based, means-tested social insurance and demonstrate theoretically that social insurance programs with means tests based on assets discourage saving by households with low expected lifetime income.
Abstract: Micro data studies of household saving often find a significant group in the population with virtually no wealth, raising concerns about heterogeneity in motives for saving. In particular, this heterogeneity has been interpreted as evidence against the life cycle model of saving. This paper argues that a life cycle model can replicate observed patterns in household wealth accumulation after accounting explicitly for precautionary saving and asset-based, means-tested social insurance. We demonstrate theoretically that social insurance programs with means tests based on assets discourage saving by households with low expected lifetime income. In addition, we evaluate the model using a dynamic programming model with four state variables. Assuming common preference parameters across lifetime income groups, we are able to replicate the empirical pattern that low-income households are more likely than high-income households to hold virtually no wealth. Low wealth accumulation can be explained as a utility-maxim...

946 citations

Journal ArticleDOI
TL;DR: An error structure that offers a rich statistical framework for panel data analysis is presented that includes as special cases most of the error specifications found in longitudinal studies of wages and earnings.

841 citations


"A retrospective on friedman's theor..." refers methods in this paper

  • ...A seminal paper on this issue has been written by MaCurdy (1982), where he shows that earnings from the PSID are best described by the sum of a random walk and an MA(1) component....

    [...]

Journal ArticleDOI

837 citations


"A retrospective on friedman's theor..." refers background in this paper

  • ...Wald (1940) suggested comparing group averages to overcome measurement error, although the way he 2 See Goldberger (1972) for an account of instrumental variables as a method for dealing with endogeneity and errors in variables with non-experimental data....

    [...]

ReportDOI
TL;DR: This paper showed that saving should be at least as good a predictor of declines in labor income as any other for ecast that can be constructed from publicly available information, even when income is stationary in first differ ences rather than levels.
Abstract: The permanent income hypothesis implies that people save because they rationally expect their permanent income to decline; they save "for a rainy day." It follows that saving should be at l east as good a predictor of declines in labor income as any other for ecast that can be constructed from publicly available information. Th e paper tests this hitherto ignored implication of the permanent inco me hypothesis, using quarterly aggregate data for the period 1953-84 in the United States. By contrast with much of the recent literature, the results here are valid when income is stationary in first differ ences rather than levels. Copyright 1987 by The Econometric Society.

780 citations

Frequently Asked Questions (1)
Q1. What have the authors contributed in "A retrospective on friedman’s theory of permanent income" ?

The authors present a short review of Friedman ’ s Permanent Income Hypothesis, the origins of the idea and its theoretical foundations. The authors give a brief overview of its influence in modern economics and discuss some relevant empirical results and the way they relate to the original approach taken by Friedman. This article was prepared for a conference in honour of Milton Friedman at the University of Chicago in November 2002, on the occasion of his 90 Birthday.