GENUINE SAVING AND POSITIONAL EXTERNALITIES

• Date: 01 November 2017
• Published date: 01 November 2017
• DOI: http://doi.org/10.1111/iere.12248

INTERNATIONAL ECONOMIC REVIEW

Vol. 58, No. 4, November 2017

GENUINE SAVING AND POSITIONAL EXTERNALITIES∗

BYTHOMAS ARONSSON AND OLOF JOHANSSON-STENMAN1

Ume˚

a University, Sweden; University of Gothenburg, Sweden

Much evidence suggests that people are concerned with their relative consumption. Yet, positional external-

ities have so far been ignored in savings-based indicators of sustainable development. This article examines the

implications of relative consumption concerns for measures of sustainable development by deriving analogues to

genuine saving when people are concerned with their relative consumption. Unless the positional externalities

have been internalized, an indicator of such externalities must be added to genuine saving to arrive at the proper

measure of welfare change. We also show how relative consumption concerns affect the way public investment

ought to be reflected in genuine saving.

1. INTRODUCTION

How to measure social welfare, and correspondingly welfare change over time, is a classical

and much discussed question in economics that has received increased attention recently, e.g.,

through the report of the Stiglitz-Sen-Fitoussi Commission (Stiglitz et al., 2009), initiated by

French president Sarkozy, and its aftermath.2According to this report (p. 8), “it has long

been clear that GDP is an inadequate metric to gauge well-being over time particularly in its

economic, environmental, and social dimensions, some aspects of which are often referred to

as sustainability.” The idea of sustainability, or sustainable development, has also grown in

importance over time and is highlighted by the 17 sustainable development goals (Sachs, 2015)

adopted by the United Nations in September 2015. Although there are many definitions and

interpretations of sustainable development, it is often, as expressed by the World Commission

on Environment and Development, defined to reflect a development that meets “the needs of

the present without compromising the ability of future generations to meet their own needs”

(Our Common Future, 1987, p. 54).3

The concept of genuine saving, which is at the core of this article, plays a key role in the

measurement of both social welfare change and sustainable development. The genuine saving

of a country is a measure of comprehensive net investment, i.e., the value of all capital formation

undertaken by society over a time period. In other words, it summarizes the value of the

net investment in all relevant capital stocks, potentially including net investments in man-

made capital and human capital, changes in natural resource stocks and environmental capital,

as well as biodiversity. Earlier research shows that the genuine saving constitutes an exact

measure of economy-wide welfare change over a short time interval if the resource allocation

is first best.4Furthermore, in the aftermath of the World Commission on Environment and

∗Manuscript received June 2015; revised March 2016.

1The authors would like to thank three very constructive anonymous referees, an associate editor, Sofia Lundberg,

and Karl-Gustaf L¨

ofgren for helpful comments and suggestions as well as Catia Cialani for helpful research assistance.

Research grants from the Swedish Research Council (ref 421-2010-1420) are gratefully acknowledged. Please address

correspondence to: Thomas Aronsson, Department of Economics, Ume˚

a School of Business and Economics, Ume˚

a

University, SE – 901 87 Ume˚

a, Sweden. E-mail: Thomas.Aronsson@econ.umu.se.

2See in particular Fleurbaey and Blanchet (2013) for a thorough and rigorous analysis of social welfare measures

beyond GDP.

3ThisreportisoftenreferredtoasThe Brundtland Report.

4The seminal contributions are Pearce and Atkinson (1993) and Hamilton (1994, 1996). See also Hamilton (2010)

for an overview of the literature and van der Ploeg (2010) for a political economy analysis of genuine saving. A

1155

C

(2017) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

1156 ARONSSON AND JOHANSSON-STENMAN

Development some 30 years ago, genuine saving has also become an indicator of sustainable

development. Indeed, a natural and frequently used economic interpretation of sustainable

development is that welfare must be nondeclining over time, meaning that genuine saving

becomes an exact indicator of sustainable development over a short time interval.5Another

possible definition of sustainable development is that the instantaneous utility must not exceed

its maximum sustainable level, on the basis of which Pezzey (2004) shows that nonpositive

genuine saving constitutes an indicator of unsustainable development, although positive genuine

saving does not necessarily imply that development is sustainable.6For each of these definitions,

genuine saving gives information of clear practical relevance for economic welfare, which is

further emphasized by the attention paid to genuine saving by the World Bank, which regularly

publishes estimates on genuine saving (referred to as adjusted net saving) for a large number of

countries (see also Section 6).

Yet, the literature dealing with genuine saving has so far focused on traditional neoclassi-

cal textbook models, where people derive utility solely from their own absolute consumption

of goods and services (broadly defined). Thus, it neglects the possibility discussed in behav-

ioral economics literature that people also enjoy consuming more, and dislike consuming less,

than others—an idea that appeared rather obvious to many leading economists of the past,

including Adam Smith, John Stuart Mill, Karl Marx, Alfred Marshall, Thorstein Veblen, and

Arthur Pigou, before it became unfashionable in the beginning of the 20th century (see Mason,

1998).

The purpose of this article is to examine how relative consumption concerns, through a pref-

erence for “keeping up with the Joneses,” affect the principles for measuring welfare change.

It will then present a correspondingly modified measure of genuine saving. Arguably, such a

study is relevant for several reasons. First, there is now a large body of empirical evidence that

people are concerned with their relative consumption (and not just their absolute consump-

tion as in standard economic models).7Questionnaire–experimental research often concludes

that up to 50% of an individual’s utility gain from increased consumption may actually be

due to increased relative consumption (e.g., Alpizar et al., 2005; Solnick and Hemenway, 2005;

Carlsson et al., 2007). Similarly, many happiness-based studies find that a large (or sometimes

even dominating) share of consumption-induced well-being in industrialized countries is due

to relative effects (e.g., Easterlin, 2001; Luttmer, 2005; Easterlin et al., 2010). In turn, this may

distort the incentives underlying consumption and capital formation. Second, based on the es-

timates referred to above, wasteful conspicuous consumption generates positional externalities

(often referred to as consumption externalities in the macroeconomic literature). These are

likely to result in significant welfare costs, which—if not properly internalized—may change

the principles for calculating welfare change-equivalent measures of saving. Indeed, we show

that the more positional people are on average, the more the conventional measure of genuine

saving (where people are assumed not to have positional preferences) will overestimate the true

welfare change. Third, recent literature shows that optimal policy rules for public expenditure

public economics approach is taken by Aronsson et al. (2012), who derive a second-best analogue to genuine saving

in a representative-agent economy with distortionary taxes and public debt accumulation, whereas Li and L¨

ofgren

(2012) address genuine saving in an economy where growth is stochastic. See also the recent empirical application by

Geasley et al. (2014) and Fleurbaey (2015) for an examination of relationships between social welfare and measures of

sustainable development (including genuine saving).

5See, e.g., Arrow et al. (2003a).

6See also Asheim (1994) and Pezzey (1994). This false-message problem is further analyzed by Valente (2008), who

derives conditions under which a long-run measure of genuine saving is negative (which it may be even if the current

genuine saving is positive). Relationships between genuine saving and consumption (or instantaneous utility) are also

examined by, e.g., Aronsson et al. (1997) and Hamilton and Hartwick (2005).

7See, e.g., Easterlin (2001), Johansson-Stenman et al. (2002), Blanchflower and Oswald (2004), Ferrer-i-Carbonell

(2005), Luttmer (2005), Solnick and Hemenway (2005), Carlsson et al. (2007), Clark and Senik (2010), and Corazzini

et al. (2012). See also Fliessbach et al. (2007) and Dohmen et al. (2011) for evidence based on brain science and Rayo

and Becker (2007) for an evolutionary approach.

GENUINE SAVING AND POSITIONALITY 1157

are modified in response to relative consumption concerns,8suggesting that such concerns may

also affect the value of public investment in the context of genuine saving. This will be further

discussed below.

The analysis is based on a dynamic general equilibrium model where each consumer derives

utility from his/her own consumption and use of leisure, respectively, and from his/her relative

consumption compared with a reference consumption level (reflecting other people’s current

consumption). Our main contribution is that we show how positional concerns, and correspond-

ing positional externalities, influence the way welfare change-equivalent savings ought to be

measured. We distinguish between a social optimum where all externalities are internalized and

market economies without externality correction. We also distinguish between first-best and

second-best social optima by extending the benchmark model to allow for asymmetric infor-

mation between the consumers and the social planner (or government). Furthermore, by using

insights developed in the literature on tax and other policy responses to relative consumption

concerns, we are also able to relate genuine saving to empirical measures of “degrees of posi-

tionality,” i.e., the extent to which relative consumption is important for individual well-being.

The paper closest in spirit to ours is Aronsson and L ¨

ofgren (2008). They consider the prob-

lem of calculating an analogue to Weitzman’s (1976) welfare-equivalent net national product

in an economy where the (identical) consumers are characterized by habit formation. Their

results show that if the habits are fully internalized through consumer choices, habit formation

does not change the basic principles for measuring welfare (except that the individual’s own

past consumption affects his/her current instantaneous utility). However, with external habit

formation, i.e., if the habits partly reflect other people’s past consumption, the present value of

this marginal externality also affects the welfare measure. Our study differs from Aronsson and

L¨

ofgren (2008) in at least four distinct ways: We (i) consider measures of genuine saving (or

analogues thereof) instead of comprehensive net national product measures, (ii) focus attention

on the empirically well-established keeping-up-with-Joneses type of comparison,9(iii) allow for

redistributive aspects by considering a case with heterogeneous consumers, and (iv) introduce

public investments into the study of welfare change-equivalent savings.10

The article is outlined as follows. In Section 2, we present the benchmark model where each

individual derives utility from consuming more than other people. We also present useful indi-

cators of the extent to which relative consumption matters for individual well-being. Following

earlier literature on genuine saving, we assume in the benchmark model that individuals are

identical and that the population size is fixed. In Section 3, we use the benchmark model to an-

alyze economy-wide measures of welfare change. In addition, we present two extensions: First,

in Subsection 3.3 we analyze a case of endogenous population growth. Second, in Subsection 3.4

we examine the consequences of replacing the conventional welfarist social objective (which

reflects all aspects of consumer preferences) with a nonwelfarist objective that does not reflect

the consumer preference for relative consumption. Section 4 extends the benchmark model by

analyzing the role of public investments. As a further extension, Subsection 4.2 analyzes the

case with more than one externality. This means that the model encompasses a case where

private consumption gives rise to environmental externalities beyond the positional ones. Sec-

tion 5 examines a more general model with two ability types that differ in productivity, where

8See Ng (1987), Brekke and Howarth (2002), Aronsson and Johansson-Stenman (2008, 2014b), and Wendner and

Goulder (2008), who analyze different aspects of public good provision in economies where people are concerned with

their relative consumption.

9In a background working paper, Aronsson and Johansson-Stenman (2016), we also analyze the implications of

catching-up-with-the-Joneses comparisons, where the measure of reference consumption refers to other people’s past

consumption. The results show that the associated externalities affect the welfare change measure in the same general

way as the externalities following from keeping-up-with-the-Joneses comparisons.

10 Yamaguchi (2014) uses the model developed in Aronsson and L¨

ofgren (2008) to derive an indicator of genuine

saving. He examines conditions under which the value of investment in physical capital and the value of investment in

the stock of habits jointly contribute to increased welfare in a social optimum. Thus, his analysis has a very different

focus from ours.