Econometria

Journal of International Economics 52 (2000) 235–257
www.elsevier.nl / locate / econbase

A unification of second best results in international trade
Pravin Krishna a , *, Arvind Panagariya b
a

b

Department of Economics, Box B, Brown University, Providence, RI 02912, USA
Department of Economics, University of Maryland, College Park, MD 20742 -7211, USA
Received 3 August 1998; accepted19 July 1999

Abstract
The received wisdom from the theory of second best is that the presence of distortions in
some sectors of the economy would in general require intervention in other sectors as well.
This paper offers a simple set of propositions which help unify the results of numerous
papers in the international trade literature involving the theory of second best, many of
whichappear to contradict this theory. The propositions identify how, in optimization
problems in economics, pre-imposed quantitative restrictions enter differently from price
restrictions. The implications of this difference for the conduct of second-best optimum
policies are also analyzed. In particular, the paper identifies and discusses the conditions
under which the presence of distortions in somesectors does not undermine the case for
non-intervention in other markets. © 2000 Elsevier Science B.V. All rights reserved.
Keywords: Theory of second best; Welfare; Commercial policy; Trade policy; Taxation; Piecemeal
reform
JEL classification: F11; F13; H20; D50

1. Introduction
Stimulated by the pioneering contribution by Bhagwati and Ramaswami (1963),
an extraordinarily large body ofliterature has developed on optimal policies in the
*Corresponding author. Fax: 1 1-401-863-1970.
E-mail addresses: pk@econ.pstc.brown.edu (P. Krishna), panagari@wam.umd.edu (A. Panagariya).
0022-1996 / 00 / $ – see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S0022-1996( 99 )00057-4

236

P. Krishna, A. Panagariya / Journal of International Economics 52 (2000) 235– 257

presence of distortions in one or more sectors of the economy.1 The received
wisdom from this theory, as stated in the original contribution by Lipsey and
Lancaster (1956), is that in the presence of distortions in one or more sectors,
welfare can be improved by intervening appropriately in other sectors of the
economy. Papers dealing with specific second-best problems generally attemptto
identify such interventions.
In this paper, we unify the results of numerous papers on the theory of the
second best by proving a set of propositions with reference to a generic
optimization problem, as in Lipsey and Lancaster (1956), and then applying these
propositions to several specific second-best problems in the pure theory of
international trade.2 We show that many results relatingto the global optimality of
free trade, piecemeal trade reform, customs unions theory, the principle of
targeting, immiserizing growth, the theory of non-economic objectives, and the
welfare effects of aid and directly unproductive profit seeking (DUP) activities in
the presence of distortions are specific applications of these propositions.3 Our
basic propositions are, however, general enoughthat they can be readily applied to
second-best problems in other branches of microeconomics, especially public
economics.
The key steps in our unification of the literature are to show how, in
optimization problems, quantitative restraints enter differently from price restraints, to derive the implications of this difference for the conduct of policy in
1

As Panagariya (1999) carefullydocuments, though Meade (1955b) had discussed many second-best
problems in his now celebrated Trade and Welfare, it was not until the forceful statement and clear
demonstration, by Bhagwati and Ramaswami (1963), of the targeting principle and the superiority of
free trade over intervention, once the distortion had been neutralized, that the literature on domestic
distortions and commercial...