NBER WORKING PAPER SERIES
QUANTITY CONTROLS, LICENSE TRANSFERABILITY,
AND THE LEVEL OF INVESTMENT
Kala Krishna
Ling Hui Tan
Ram Ranjan
Working Paper 8796
http://www.nber.org/papers/w8796
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
February 2002
This paper was written while Kala Krishna was a visiting scholar at the IMF Institute. We are grateful to
Andrew Feltenstein for helpful comments. The views expressed herein are those of the authors and not
necessarily those of the National Bureau of Economic Research.
© 2002 by Kala Krishna, Ling Hui Tan and Ram Ranjan. All rights reserved. Short sections of text, not to
exceed two paragraphs, may be quoted without explicit permission provided that full credit, including ©
notice, is given to the source.
Quantity Controls, License Transferability, and the Level of Investment
Kala Krishna, Ling Hui Tan and Ram Ranjan
NBER Working Paper No. 8796
February 2002
JEL No. F3
ABSTRACT
This paper models investment/entry decisions in a competitive industry that is subject to a
quantity control on an input for production. The quantity control is implemented by auctioning licenses
for the restricted input. The paper shows that liberalizing the quantity control could reduce investment
in the industry under certain circumstances. Furthermore, the level of investment is quite different when
licenses are tradable than when they are not. Key factors in the comparison include the elasticity of
demand for the final good and the degree of input substitutability. Two examples are computed to
illustrate the results.
Kala Krishna Ling Hui Tan
The Pennsylvania State University IMF Institute
Department of Economics European Division, IS3-1612
608 Kern Graduate Building International Monetary Fund
University Park, PA 16802 700 19th Street, NW
and NBER Washington, DC 20431
kmk4@psu.edu
Ram Ranjan
The Pennsylvania State University
Agricultural Economics and Rural Sociology
106 Armsby Building
University Park, PA 16802
- 3 -
I. INTRODUCTION
Quantity controls on production inputs are commonly used throughout the world. In
transition countries, input rationing is still prevalent in many industries. In some developing
countries, import licensing and foreign exchange controls may restrict the quantity of inputs
available to domestic firms. In industrial countries, input quantity controls are often used to
regulate natural resource industries such as timber and fishing; they have also been imposed
in response to supply shocks such as those experienced with petroleum or electricity, and for
environmental reasons such as pollution reduction.
Input rationing is usually implemented by issuing licenses or permits for the restricted
factor. In this regard, for analytical purposes, there is a close similarity between input quotas
and output quotas: in the latter case, a license is required in order to produce a certain amount
of output, so the license can be thought of as a restricted input that is necessary for
production.
The initial allocation of licenses may be based upon certain criteria, such as firms’
historical performance, or by auction. In some cases, these licenses are tradable; in other
cases, they are not. This is a particularly important issue in agricultural and natural resource
economics, where production controls are most likely to be implemented. For example, in
fisheries management, some countries—notably Iceland and New Zealand—have introduced
a system of individual transferable quotas (ITQ) whereby an annual quota for a particular
species (the total allowable catch) is distributed among individual firms by means of licenses
that are tradable; other countries—such as the United States and the United Kingdom—are
still weighing the merits of this system vis-à-vis alternative measures such as nontransferable
- 4 -
individual fishing quotas. Similarly, in the dairy and poultry industry, there is an ongoing
discussion within the United States and the European Union on the advantages of a supply
management system like Canada’s, where farmers have to purchase a tradable permit to
produce a specified amount of output.
Although they are widely used, quantity controls and input rationing have not been
widely studied. Squires (1994) estimates the effect of an input quantity control on factor
demand, output supply, and capacity utilization by a competitive multiproduct firm in a
certain environment, using data on wet rice production in Indonesia. However, there has been
little work on the effect of input rationing on entry and investment in the affected industries
under production uncertainty. This paper considers the investment/entry decision in a
competitive industry with no fixed costs which is subject to input rationing such that licenses
or permits are required for production. The objective is to compare the level of investment in
the industry under two scenarios: when the licenses are transferable and when the licenses are
nontransferable. A similar problem is analyzed in Spencer (1997), which considers the
effects of a licensing requirement on imported capital equipment, comparing the outcome
under an exogenous bureaucratic allocation (nontransferable quota licenses) with that under a
market allocation (transferable licenses). Unlike Spencer (1997), however, the analysis used
in this paper draws on the model(s) developed in Krishna and Tan (1999), which compares
the endogenous outcomes from transferable and nontransferable regimes of quota licenses.
They show that—contrary to the common belief that transferable licenses are always worth
more than nontransferable licenses and always yield more efficient outcomes—the price of
transferable licenses can fall below the price of nontransferable licenses if the quota is
sufficiently large and that nontransferability could result in higher welfare if license revenue
- 5 -
is heavily weighted in the overall welfare function. To the extent that license possession
enables a firm to produce the final product, the license market will be closely linked to the
market for the final product. This paper thus extends the analysis in Krishna and Tan (1999)
by explicitly modeling the product market as well. In this paper, the demand for licenses
(i.e., the restricted input) is modeled as a derived demand arising from production in a final
output market. It is assumed that each firm has to utilize a certain number of licenses in order
to produce a unit of output.
It is necessary to introduce an element of uncertainty into the model in order to make
a meaningful comparison between transferable and nontransferable licenses. After all, as
explained in Krishna and Tan (1999), if every economic agent knows with full certainty his
or her valuation of a license at all times, and if the initial allocation of licenses is determined
endogenously (reflecting the voluntary choice of each agent), then there will be no scope for
resale and transferability will confer no benefit (or equivalently, nontransferability will imply
no cost).
1
Furthermore, this uncertainty has to be at the individual rather than the aggregate
level: if all agents faced a common shock then, again, there will be no scope for resale and no
benefit from transferability. In this paper, uncertainty is modeled as a firm-specific
productivity shock which affects the efficiency of final good production. Firms have to
purchase licenses for production before this productivity shock is realized; in the transferable
1
In many practical applications, the initial allocation of licenses is not endogenous. For
example, the initial allocation of tradable fishing quotas in New Zealand and Iceland were
not effected by auction but by free distribution to existing firms based primarily on historical
catch (see OECD (1999)). However, such schemes are generally considered inferior to
market-based mechanisms such as allocation by auction, on efficiency and equity grounds
(see Morgan (1997), for example.)
