THE STRATEGIC USE OF REGULATORY INVESTMENT

BY INDUSTRY SUB-GROUPS

SHARON OSTER*

In the last ten years, consumer regulation at the federal, state and local
levels has burgeoned. Of the eight federal agencies directly concerned
with consumer protection, three - the Office of Consumer Affairs, the
Consumer Product Safety Commission, and the Consumer Product Infor-
mation Center - were created in the 1970’s. At the same time, the
emphasis placed by existing agencies on consumer regulations has
increased. These regulations have had, of course, important effects on
industry as well as on the consumers they may have intended to serve.
Early work investigating the effect of regulation on industry (see, for
example, Stigler, 1971), treated firms in the industry as a single coalition
- equally affected by the regulation, and therefore equally interested in
those regulations. In this paper, I focus on the groups of firms within
industries, and in particular examine the way particular regulations may
favor one group or another within an industry, and thus alter the competi-
tive balance within that industry.
The argument I will make is two-pronged. Firstly, I will argue that
new regulations often impose very different benefits and costs on groups of
firms within an industry, Secondly, and this is the more controversial link
in the argument, I will argue that groups within an industry often recog-
nize and indeed may promote new regulations which will be to their com-
parative advantage. In short, regulatory barriers created by firm conduct
may be used by groups in the industry as a competitive weapon against
other groups.
I. INDUSTRY GROUPS

The argument in this paper depends on the existence of groups within

an industry with different interests, as well as on the power of these groups
to participate actively in the process by which entry barriers are created.
Traditional statistical work in industrial organization generally used the
industry as its primary unit of observation. The focus in that work was on
the structure of the industry, and on how structural differences across
industries create differences in the performance of those industries; details
of thesub-structures of industries were for the most part suppressed.
The major exception to this treatment of the industry as a homogeneous
unit in the older literature was the case study. McKie, in his classic study of
cans, for example, emphasizes the differences among can manufacturers

‘Yale University. The research on which this work is based was funded by the Federal Trade
Commission and the Small Business Administration; however, the ideas expressed herein are the
author’sand in no way are intended to express the views of the Federal Trade Commission. I would like
to thank Ray Fair, Richard Levin, Richard Nelson and Michael Spence for helpful comments.

604
Economic In uiry
Vol.xx.Oct%er 1982
 OSTER: T H E STRATEGIC USE OF REGULATORY INVESTMENT 605

not only in their size, but in their scope, type of product produced and
policies. (McKie, 1959). Similarly Peck in his study of the aluminum
industry also highlighted differences among firms. (Peck, 1961). In the
post-war period, the capacity expansion strategies of the three major pro-
ducers - Kaiser, Reynolds, and Alcoa - were quite different, and quite
important for the performance of the three companies.
Recent economics literature has focussed on what have been termed
strategic groups within an industry. The group here is defined by the com-
mon strategies followed by the members in setting key decision variables
- investment levels, R&D, etc. (Hunt, 1975; Caves and Porter, 1977;
Porter, 1979). Not only do groups have different strategies, they are, in
part because of these different strategies, differentially protected from the
market. From this central proposition about the particularity of entry bar-
riers to the group, the empirical literature has proceeded to examine the
effect that these differences in entry barriers have on the performance of
the firms in each group. What are the results of the different strategic
choices made by groups within an industry, both in terms of differences in
the height of the entry barriers facing potential entrants into that group,
and in terms of the performance of that group? Work along this line has
been done by Newman (1978), Hatten and Schindall (1977), Porter
(1979), Porter and Spence (1978), and Oster (1982).
Coincident with this literature on the existence of groups within an
industry, there has been new work on the strategic use by firms of entry
barriers. The early industrial organization literature characterized entry
barriers as being largely outside of the control of firms within an industry;
in Salop’s terms entry barriers were characterized as being “innocent”
(Salop, 1979). In part, this had to do with the focus on economies to scale
as the main source of entry barriers, coupled with the view that the tech-
nology creating these barriers was largely exogenous to the firm.
The earliest change in the treatment of entry barriers came in the dis-
cussion of advertising, where it was recognized that firms could through
advertising purposively change product differentiation and thus increase
barriers to entry in the industry. More recently, strategic use of research
and development as a way to increase the minimum efficient scale of an
industry and thus raise entry barriers has been treated (see, for example,
Levin, Southern Economic Journal, 1978). Salop (1979) discusses the
recent literature on the use of both innovation and advertising as strategic
entry deterrence. The connection between the literature on group-specific
entry barriers and the strategic use of entry barriers is clear. As long as
entry barriers were thought of as largely technological and independent of
firm conduct, it was hard to see firms in the industry actively manipulat-
ing these barriers. Once barriers to entry are seen to come in part from
firm conduct, the potential for using these barriers as competitive
weapons becomes clearer.
606 ECONOMIC INQUIRY

II. THE USE OF REGULATION TO ERECT MOBILITY BARRIERS

Systematic treatment of the central role of industry in fashioning its

own regulation by government was first provided by Stigler (1971), and
later developed and refined by Peltzman (1976). In contrast to the old
theory which emphasized the market failure/public interest source for
regulation, Stigler argued that economic regulations were designed not to
correct a resource misallocation but rather to transfer wealth; in most
cases, the intent was to transfer wealth to politically powerful producers’
groups. (Stigler, 1971). The industry in this work, both in the theoretical
discussion and in the empirical examples, was treated as a single coalition
- with essentially the same interests. The game, then, was to use regula-
tion as a barrier to entry against possible new competitors, and to exploit
as fully as possible the existing market power of the industry. Empirical
support for thisview has been mixed.
If we take seriously the Caves, Porter idea of groups within industries,
it becomes apparent that this characterization of the link between indus-
try and the regulatory process is too simple. Regulations typically do not
affect all firms within an industry in the same way; thus all firms in an
industry should not be expected to respond to these regulations in the same
way.
As long as there is some initial difference among firms in an industry,
different firms in that industry may push for regulations which increase
the relative rate of return to their peculiar characteristics. Indeed, a firm
may even encourage passage of a regulation which reduces industry
demand or increases industry costs. The firm may encourage such regula-
tions because they differentially damage its rivals, and thus rearrange
market shares at the same time they reduce the total market. To go one step
further, the firm may even encourage a regulation which lowers its short-
term profits if that regulation simultaneously reduces the ability of its
rivals to compete effectively. In this case, there are some links between the
discussion of rivalrous regulatory investment and the limit pricing litera-
ture. In the limit pricing discussion, the firm may find it profitable in the
long run to set a low price in order to prevent entry and protect its long-run
position. Regulatory investments may be used similarly. Of course the reg-
ulation case is a bit more complicated since regulations may themselves
affect barriers to entry.
There are three general conditions which must be met before one
would expect any rivalry in regulatory investments by firms to occur.
Firstly, the industry must contain differentiated sub-groups of firms. This
differentiation can be created either by firms producing different product
mixes, or through differences in size, technology, levels of integration or
whatever. These differences create the possibility that a regulation can
create a comparative advantage for some firms in the industry, and lead to
a shifting of market shares within the industry.
The second requirement for this rivalry in regulatory investments to
 OSTER: THE STRATEGIC USE OF REGULATORY INVESTMENT 607

occur is that the firms in the industry must be fairly interdependent. That
is, we must be dealing with groups in an industry and not with truly sepa-
rate industries. The essence of rivalry in investments is that a firm will try
to exploit the differential effect of a regulation on itself versus its rivals; this
differential effect is only important if firms are interdependent, though
this interdependence can be either on the demand side or on the supply
side.
Finally, if firms are to engage in strategic regulatory behavior there
must be some actual or potential barriers to entry to the industry, and some
barriers to mobility within the industry. Without barriers, any quasi-rents
created by the firm’s regulatory intervention will be dissipated. Barriers
protect the return associated with the change induced by the regulation.
This case is similar to the case of limit pricing. Without some entry bar-
riers, neither limit pricing nor rivalrous investments are profitable
strategies.
If an industry meets these three requirements, then there is at least
some potential for rivalrous regulatory investments to occur. Firms with
different bundles of specific capital and hence different operating strate-
gies can try to influence the regulatory process to increase the pay-off to
the particular strategies they have adopted. In the discussion which fol-
lows, I sketch out the specific factors which will lead to a situation in
which some firms in an industry gain from a regulation, while others lose,
creating the potential for rivalrous investments. I consider regulations
affecting both supply and demand conditions facing the firm.
Demand differentiation: sub-groups produce different but substitutable
products
In order to make the problem simpler, I will work with an industry
with two firms and linear demand functions. The demand system can be
derived from a quadratic utility function (see Dixit, 1979).We have:

Pz = a , - B,Q2 - V Q 1

(3)

Firm 1produces only good 1, and so faces demand function 1, and firm
2 produces only good 2 . “V” is a measure of cross-product substitution.
Equality of V’sin equations (1)and (2) assures us of symmetry in the cross-
product elasticities. This is a one-period model, so that we have not
allowed for any entry across product classes.
This is a simple case of a differentiated duopoly. We are interested in
the context of this model in considering the effect on firm profits of
608 ECONOMIC INQUIRY

changes in the demand and cost parameters which might be produced by

regulatory action. There are a number of solution concepts which might
be applied in solving for an equilibrium in this problem. Perhaps the sim-
plest would be to assume joint profit maximization. But in the context of
this problem cooperative behavior of this sort is unrealistic. Given that the
firms are differentiated, one firm may gain from a regulation while
another loses. Cooperation or joint profit maximization thus seems
implausible here. If we allow for side payments, joint maximization is
perhaps more plausible. But in many situations such payments are impos-
sible. Thus either a Cournot or Stackelberg model seems to be more sensi-
ble. In the discussion which fdlows, both Cournot and Stackelberg solu-
tions are used.
The specific results which emerge in this paper obviously depend some-
what on the solution concept used, and on the specifics of the demand and
cost functions, but I think the intuition behind the results follows in a more
general case. In what follows, I have tried to point out the general results.
The Cournot equilibrium to the simple problem above is

If we assume instead firm 1 is a Stackelberg leader, the equilibrium

becomes

In the context of this model, we can consider the effects on firm’s profits
of two different broad types of demand-side regulations: a regulation
which affects the intercepts of the firm’s demand curves, or, alternatively
one which affects V, the cross-goodsubstitution parameter.
The most straightforward case in which rivalrous investments can be
expected is the case of a regulation which directly affects only the demand
for one of the two goods. For example, we might have a regulation which
reveals some adverse information about the characteristics of good 2 . One
way to think of this regulation is that it reduces u 2 , or shifts the demand
curve to the left. Firm 2 obviously will oppose such a regulation. I n this
case, however, the industry will not be in agreement. In particular, given
 OSTER: T H E STRATEGIC USE OF REGULATORY INVESTMENT 609

the substitution between goods, the reduction in production of Q, will

shift the demand for good 1to the right; the intercept for good 1is simply
a, - VQ, . So firm 1 has some incentive to support an a, shiftingregulation.
This result is symmetrical. That is, firm 2, the smaller, follower firm
also has an interest in promoting regulations which decrease a , . Since
small firms generally have a smaller influence on the regulatory process
than large firms, in what follows I concentrate on the regulatory incen-
tives facing the dominant firm.
As long as goods are substitutes, firms will gain from a shift in a,. How-
ever, the size of the gain depends on’the particular characteristics of the
problem at hand. If we consider the profit of firm 1 at the Stackelberg
equilibrium and change a , , we find:

(7)

This result follows Dixit (1979), who is primarily concerned with analyz-
ing optimal firm behavior with and without entry. First, of course, the
larger the shift in a , the greater the loss to firm 2 and gain to firm 1. The
lower is B,, the more elastic the own demand for good 2, the larger the
potential gain to firm 1. A more elastic demand implies that any shift in a,
will have a large effect on Q,, and this, of course, benefits firm 1 through
the substitution factor. Finally, a large “V,” implying strong substitution
between goods, increases the potential gain available to firm 1 from
firm 2’s misfortune. In the extreme, if V is zero (no substitution), rivalrous
investments will not occur. In sum, rivalrous investments are likely in this
case if the regulation has a large effect, rival’s demand is price elastic, and
cross-good substitution is large.
Consider now a second, more complicated case. In particular, consider
a regulation which reduces demand for both good 1 and good 2, but does
so unevenly, firm 2 being more heavily affected. In this case, we may or
may not have regulatory rivalry, depending on the particular structure of
the problem. The effect of reducing a , is to decrease firm 1’sprofits. In the
case at hand, the effect on profits can be represented as:

The firm then in considering whether to support a particular regulation

will compare (7) and (8).It is interesting to note that if a , and a, are equally
reduced by the regulation, firm 1 will never have an incentive to encour-
age the regulation, since V / B ,must be less than 1.
In the previous two cases, I considered regulations which changed the
intercept term in the demand curve. A second type of regulation affecting
demand is one which changes the degree of substitution between the two
610 ECONOMIC INQUIRY

goods. A regulation might well, for example, provide information which

tells consumers that two products are better substitutes than they previ-
ously believed. One way to model this is to consider a change which simply
increases V. This is the case considered by Dixit (1979) in his analysis of the
effect of product differentiation on entry. Dixit gets a rather curious result,
which is of interest here. In the Dixit model, there are some fixed costs of
entry so that there is a lack of continuity which influences the incentives
facing an already established firm. In particular, changes in the parame-
ters of the model will affect firms differently depending on whether they
are successfully impeding entry or not. Once entry has occurred, which is
the primary case I am considering, increases in V hurt the firm. If firms
are already rivals, the more differentiated their individual products, the
less they will be affected by changes in V. Indeed both firms in the industry
will lose from a simple increase in V; The size of the loss will again depend
on the size of the substitution parameter, and the demand elasticity. For
the leader firm, the loss in profits from an increase in V is

(9)

If the established firm is preventing entry, an increase in V (i.e,, better

substitution) increases profits. As Dixit points out, it is easier to prevent
entry as V increases. So there is a conflict between the interests of the firm
in preventing entry through regulatory intervention and its interests in
improving its market position vis-u-vis its rivals.
Sub-groups diflerentiated by cost functions
Government regulations can affect the cost curves of firms, as well as
their demand curves. Cost-changing regulations will not necessarily affect
all firms within an industry in the same way. This point is well known (and
indeed a subject of concern) in the area of pollution regulation. Leone
(1977), for example, argues that water pollution requirements in the metal
finishing industry raise the minimum efficient scale in that industry by a
factor of five. This clearly affects both entry conditions into the industry,
and the profitability of existing large versus small firms. A recent study
suggests that the costs per dollar of sales of complying with federal pollu-
tion control regulations in the automobile industry also decline radically
with firm size (cited in Berney, 1979). In what follows, I consider rivalrous
behavior by firms in manipulating cost changing regulations.
The simplest case is one in which firms produce substitute products,
and differ in technology, and the anticipated regulation adversely affects
only one of the firms. The effect of a change in the average costs of the firm
is identical to the effect of changes in the firm's demand intercept. I n par-
ticular, an increase in the costs of firm 2 will in general increase the profits
of firm 1. In the case being considered here, the size of this effect is:
 OSTER: THE STRATEGIC USE OF REGULATORY INVESTMENT 611

The rivalry effect varies with the substitution parameter, the initial cost
change, and demand elasticity. Marvel (1977) provides an excellent early
example of this situation, in his discussion of the effect of child labor laws
on 19th Century British textiles. At the time, textiles were produced in two
types of mills: steam-powered and water-powered mills. The former
employed few children, the latter many. Marvel interprets the evidence to
suggest that child labor laws were passed at the behest of the steam mills to
encourage higher costs in their rivals, the water mills.
In the case discussed above, only one firm was influenced by the regu-
lation. A more interesting case occurs when the anticipated regulation
raises costs of both firms, but unevenly. An increase in own costs decreases
own profits as:

So, the firm will calculate the effect of a regulation which increases every-
one’s costs by comparing 10 and 11. As in the demand-side cases, three
factors are critical in setting the groundwork for rivalrous investments: the
size of the cost differential, the substitutability between products, and the
own elasticity of goods. Indeed, an increase in per unit costs is equivalent
in the model to a reduction in the intercept of the demand curve.

In each of the previous examples, I have concentrated on the short-run

gains or losses accruing to firms from regulation. There is a long-run story
as well. Suppose there are barriers to entry into an industry, but exit is
relatively easy. In this case, we know that firms may sometimes pursue
unprofitable short-run pricing strategies in order to push rivals out of the
industry and enable themselves to raise prices. The same strategy may be
used in the area of regulation. In particular, firms may encourage a regu-
lation which in the short-term results in a reduction in profits, if that
regulation induces exit by the firm’srivals. If goods are substitutes then the
exit of a rival may create sufficient new profits to compensate for the ear-
lier loss. Several factors encourage this situation: high barriers to entry
relative to barriers to exit, a large differential firm effect, and strong cross-
good substitution.
In short, I would argue that the regulatory process can be used as a
powerful strategic weapon by sub-groups within an industry against other
sub-groups. Given this, one should analyze new regulations not as the
result of a dialectic between consumers and producers, but at least in part
as a result of (and a contributor to) the competitive balance within the
industry. In the next section of this paper, I consider a case study of one
612 ECONOMIC INQUIRY

particular FTC rule and trace its evolution under the influence of various
industry groups. This material is intended as a case study in the creation of
a new barrier to mobility through the use of a government regulatory
agency.

111. BRAND DRUGS VERSUS GENERICS: A CASE STUDY

In January, 1979, the Federal Trade Commission proposed a Model

Drug Product Selection Rule (hereafter, the Act). This regulation will be
used as a case study to investigate sub-group investment within the indus-
try in the regulatory area. The Act is clearly only one of many regulations
affecting this important industry, but it does reveal quite well the move-
ment of sub-groups within the industry.
In the 1950’sand 605, many states passed anti-substitution laws requir-
ing pharmacists to fill prescriptions with the exact brand-name drug pre-
scribed by the physician. These laws appear to have been passed largely in
response to pressure by large drug manufacturers, who argued that these
laws were necessary to ensure quality control in the industry. Critics have
recently noted that these laws also raised barriers to entry in the industry
(FTC, 1979). In the 19705, a number of states began to repeal these anti-
substitution laws by passing generic substitution laws. In January 1979,
the Federal Trade Commission proposed a model Drug Product Selection
Rule (the Act) designed to act as a model for the assortment of generic
substitution laws then existing in the states. In this paper, I will focus on
industry’s reaction to this FTC Rule.
As of 1979, all but 13 states permitted pharmacists to engage in some
substitution of generic drugs for prescribed brand name drugs.’ There are,
however, large differences in the form of the various substitution laws,
New York, for example, has the most stringent law: pharmacists are
required to fill all prescriptions with the cheapest generic equivalent avail-
able. A formulary is provided to pharmacists to inform this process. In
other states, for example Alaska, the physician must explicitly consent to
generic substitution on the prescription, no formulary is available, and
substitution is at the discretion of the pharmacist.
The FTC is intermediate in its requirements. Substitution by the phar-
macists is permitted, rather than required. In order to prevent substitu-
tion, the prescribing doctor must write “medically necessary” on the pre-
scription; this option encourages more substitution than the procedure
used by Alaska or the alternative used by several other states of using
reprinted signature lines in the prescription to prevent substitution.
Finally, a drug formulary is to be developed indicating drug equivalences.
The reaction of firms in the drug industry to the FTC Act has not been
uniform. There are 1300 firms in the pharmaceutical industry. In general,

1. States not allowing substitution were: Alabama, Hawaii, Idaho, Indiana, Louisiana,
Mississippi, Nevada, New Hampshire, North Carolina, North Dakota, Oklahoma, Texas, Wyoming.
 OSTER: THE STRATEGIC USE OF REGULATORY INVESTMENT 613

the industry is divided into two sub-groups: large firms which specialize in
the production of patented brand-name drugs, and generally smaller
firms which specialize in generic drugs. The former group, consisting of
about 130 firms, account for about 90 % of domestic pharmaceutical sales
(Pharmaceutical Manufacturers Association Fact Book, 1976). Many
firms, however, produce both brand-name and generic drugs, and there is
clearly substitution both in consumption and in production in the two
drug types. Thus, one would consider the two firm types to be in the same
industry using standard industrial organization industry criteria; they are
clearly sub-groups in the industry. The Pharmaceutical Manufacturers’
Association - an industry trade association - stresses the existence of
sub-groups in the industry:

“The U.S. prescription industry is made up of two distinct segments -

‘innovators’ that research and develop new and improved drugs, and
‘imitators’ (commonly referred to as generic houses) that produce only
follow-on copies of the original version of a drug.” (PMA Press Release)

Indeed the PMA represents only the large innovative firms; the smaller
generic houses have their own trade association - the National Associa-
tion of Pharmaceutical Manufacturers.
Since the FTC Act is still only a proposal, it is not yet possible to docu-
ment its effect on the various sub-groups within the industry. At the
present time, almost 90 % of all prescriptions are written by brand-name
(FTC, 1979, p. 4). Approximately 45% of the ethical drug market is no
longer protected by patents. (Business Week, October 29, 1979). So, the
potential effect of eliminating anti-substitution laws is large. While state
substitution laws of one sort or another have been in effect for several
years, evidence on these laws is also quite limited. Evidence from Califor-
nia and Florida suggests that the law is, as one would expect, tilting con-
sumption in favor of the output of the smaller generic firms: early evi-
dence indicates that in 60 % of the cases the drug substitution laws were
used, products of firms other than the 125 large research-oriented firms
were used. (FTC, 1979, 47). In general, firms other than the 125 large
firms control only 5-10 % of the market. The generics’share of the market
rose from 8 % in 1970 to 12% today (BusinessWeek, Nov. 12,1979).
The record suggests that the large firms recognize the potential effect of
the FTC Act on the competitive structure of the industry. In a January 9,
1979 statement, C. J. Stetler, past president of the trade association repre-
senting the large drug firms, argues “Drug firms that specialize in cheaper
drugs and spend nothing on research may temporarily gain advantages
here and there from a market artificially restructured in their favor by
government.’’ (emphasis added) Similar comments were made by Eli Lilly,
oneof the largest firms in the industry:
614 ECONOMIC INQUIRY

“Repeal of the anti-substitution laws has regulatory implications

when thought of as changing the environment upon which the current
institutional structure of the pharmaceutical industry is based.”
(Comments on the FTC Drug Substitution Inquiry, April 1978)

Large research firms have followed two strategies in response to this

environmental change. A number of large firms in the industry have
begun to increase their production of generic drugs. Indeed, the president
of the trade association representing the large drug firms - the Pharma-
ceutical Manufacturers Association - argued that the repeal of anti-
substitution laws was bad precisely because they would “force” big com-
panies to emphasize generics (Business Week, Nov. 6, 1978). The large
firms have thus far been idiosyncratic in their use of this strategy; Eli Lilly,
Upjohn and Smith-Kline have been fairly aggressive in expanding into the
generic market; a number of the other major ones have still not penetrated
this market. In table 1, I report growth rates and product mix of the sev-
eral major drug companies I was able to contact. It is interesting to note
that the diversified firms have been growing faster than the non-
diversified.

TABLE 1
Growth Rates of Major Drug Firms

% Generic Sales ’78 /Sales ’75

Lilly 7.5 1.50

Merck 0 1.33
Searle 0 1.38
Smith Kline 10 1.89
Upjohn 10 1.49
CIBA Geigy 0 1.26
Bristol Meyers 0 1.34

The second strategy followed by the large research firms has been to
invest in anti-regulatory lobbying. This is precisely what we would expect
from the analysis in section I1 given industry characteristics. In this indus-
try, the potential effect of the regulatory change appears to be large, the
cross elasticity of demand is high, and demand for generic firms’ drugs is
price elastic. The Pharmaceutical Manufacturers Association, represent-
ing the large drug firms, has been aggressively against the FTC Act. The
PMA argues that substitution laws reduce the pay-off to branded drugs
 OSTER: T H E STRATEGIC USE OF REGULATORY INVESTMENT 615

and thus reduce both the incentive for and ability to innovate. The PMA
position is put well by Armistead Lee, current president of the PMA, in a
letter:

“Those in Government proposing the MAC and substitutions pro-

grams are using an Adam Smith model of perfect competition as a
measure for maximizing consumer welfare. It is a static model, and it
implies that departures from pure competition are contrived by pro-
ducers through product differentiation based on advertising and pro-
motion, which forces consumers to spend more than they should.
“We, on our side, are using a dynamic model - associated with the
name of Schumpeter, even though most who share this philosophy
may never have heard of that great economist. We recognize that com-
petition is imperfect. We say that this is inevitable in an environment
where one is dealing with a high-technology product, where quality
differences are critical, where (admittedly) buyers or their agents are
not perfectly informed on price, and where brand loyalties, as in so
many other commodity areas, are very important. We would recall
Schumpeter’s assertion that under perfect competition one could not
expect to find innovation.” (Letter from Lee to James Mitchell, May
1979).

The generic firms have been less vocal, though they too have behaved
as we predicted. The smaller drug companies are represented by their own
trade association - the National Association of Pharmaceutical Manufac-
turers (NAPM). The NAPM supported the various states’ repeal of anti-
substitution laws, and also supports the FTC position. In response to a
questionnaire which I sent, one of the major generic houses stated the
position of this segment of the market:

“There should be a substantial monetary saving if effective substitu-

tion laws are passed . . . major companies will devote more time to
research in order to maintain a lead in original pharmaceutical prod-
ucts and development .”

The positions taken by the brand drug companies versus the generics are
familiar ones: the brand drug companies have emphasized the carrot as a
way to encourage innovation; the generic firms instead place their faith in
the stick of competition.
The large drug companies have tried to change two provisions in the
FTC Act. In the original, physicians were required to write “medically
necessary” to prevent substitution. The large drug firms have suggested
that this provision be replaced by one similar to that used by Arizona and
Delaware, in which the physician must positively request generic
substitution.
The attempt to use the regulatory environment strategically is most
clearlyseen in the PMAS response to the formularies. A formulary is simply
616 ECONOMIC INQUIRY

a listing of equivalent drugs; evidence from the states indicates that substi-
tution is increased substantially when the pharmacist is provided with a
formulary. (Federal Trade Commission, 1979, p. 282). In the FTC Act,
the Food and Drug Administration is primarily responsible for the devel-
opment of the formulary. The PMA has argued that “all drug products
with the same active chemical ingredients do not necessarily have the same
therapeutic effect on each patient.” (PMA Press Packet, 7); and further,
that the FDAk interpretation of therapeutic equivalence, as embodied
in the formularies “represents a radical departure from the past interpre-
tation.” (PMA Press Packet, 7). To counteract this change in policy the
PMA has proposed changing FDA regulations on drug testing.
Under existing FDA rules, all new drugs must hold a New Drug Appli-
cation (NDA) which requires a set of elaborate (and expensive) tests of the
drug in man (in vivo). Chemical equivalents made by other manufac-
turers after the patent lapses are only required to have an Abbreviated
New Drug Application (ANDA). In general, unless some problem exists
(i.e., documented therapeutic failure), an ANDA does not require tests in
man; cheaper in vitro (outside body) tests may be used instead. (Federal
Trade Commission, 1979,124-25).
In order to assure the therapeutic equivalence needed for a formulary,
the PMA has proposed that “each manufacturer of any multi-source prod-
uct should meet monograph standards and hold either a full NDA or an
ANDA that includes comparative bioavailability in man (in uiuo)” (PMA
Press Packet, 8). Should the FDA/FTC accept this proposal, two effects
would follow: our confidence inthe bioequivalence of drugs on the formu-
lary might be increased; and the testing costs to the generic houses would
rise substantially. In short, the PMA has proposed a new regulation which
would improve its comparative position vis-u-vis the encroaching
generics. It is a clear case in which one sub-group in the industry is
attempting to use the regulatory process to change the structure of the
industry - in this case to reestablish the old structure.2
The PMA has also been active in trying to use regulation to shore up its
position on the demand side. In particular, in early discussions of the var-
ious anti-substitution laws, the PMA argued that new regulations should
be passed requiring that the manufacturer be identified on the package
label for each drug product dispensed. The PMA firms, in addition to
being heavy innovators, are also heavy advertisers relative to the generic
houses. (Burack, 1976). Clearly the large advertised firms will gain more
from product identification than will the smaller drug houses. This PMA
proposal too can be seen as a way to reestablish barriers between the two
markets .
2. This is not to say the PMA proposal is not a “good one: this is in part a scientific question. The
industry evidence does suggest that products of brand-namefirms have a lower incidence of recall due
to poor quality than those of the generic houses (Lilly, 1979). FDA Commissioner, Donald Kennedy,
testifying before Con ess in 1977, reported, however, that the FDA “has found no evidence of
consistent differencesretween the products of large and small firms, or between brand name and
generic producers.”(Federal Trade Commission report, 1979).
 OSTER: THE STRATEGIC USE O F REGULATORY INVESTMENT 617

IV. CONCLUSION

In this paper, I have considered the role of the regulatory process in

altering the competitive balance within an industry. In an industry with
differentiated groups of firms, some groups in that industry may have an
incentive to manipulate the regulatory process to tilt the regulations
against rival groups. Gains to firms from manipulations of this sort have
been shown to depend on both their own and cross elasticities of demand
for the goods produced by the rivalrous firms. A proposed FTC regulation
in the drug area was used to illustrate a case in which strategic use of
regulation occurred.

REFERENCES

Ar+a, Phillip, Antitrust Analysis, Boston: Little Brown & Co., 1974.
Berney, Robert, “The Cost of Government Regulation in Small Business,” SBA paper.
Brock, Gerald, “Competition, Standards and Self-Regulation in the Computer Industry,” in R. Caves
and M. Roberts, 75-96.
Burack, Richard, The 1976 Handbook of Prescription Drugs, New Jersey: Pantheon Books, 1976.
Caves, Richard, American Industry: Structure, Conduct and Performance, New Jersey: Prentice
Hall, 1964.
Caves, Richard and Porter. Michael. “From Entrv Barriers to Mobilitv Barriers: Coniectural
Decisions and Contrived Deterrence to New Competition,” Quarterly h - n a l of Econohics 91,
(May 1977), 421-441.
,“Market Structure, Oligopoly and the Stability of Market Shares,”Journal of Industrial
Economics, June 1978.
Caves, Richard and Roberts, Marc, ed. Regulating the Product, Mass.: Ballinger Press, 1975.
Dixit, Avinash, “A Model of Duopoly Suggesting a Theory of Entry Barriers,” Bell Journal of
Economics 10 (Spring 1980) 20-32.
United States, Federal Trade Commission, Drug Product Selection: Staff Report to the FTC, Bureau
of Consumer Protection, January 1979.
Hatten and Schendel, “Heterogeneity Within an Industry: Firm Conduct in U.S. Brewing, Journal of
Industrial Economics, December 1977.
Hunt, Michael, “Trade Associations and Self-Regulation: Major Home Appliances,” in Caves and
Roberts, 39-56.
Leone, Robert, “The Real Costs of Regulation,” Haroard Business Reoiew , November-December
1977,57-66.
Levin, Richard, “Technical Change and Optimal Scale,” Southern Economic Journal, 44 (October
1977), 208-221.
Eli Lilly & Co., “Implications of Drug Substitution Laws: Analysis and Assessment,” Staff Report,
June 1978.
Marvel, Howard, “Factory Regulation: a Reinterpretation of early English Experience,” Journal of
Law and Economics, 20 (October 1977) 379-402.
McKie, James, Tin Cans and Tin Plate, Cambridge: Harvard University Press, 1959.
Nelson, Richard and Winter, Sidney, “Dynamic Competition and Technical Progress,” in Belassa and
Nelson, eds., Economic Progress, Private Values and Public Policy: Essays in honor of William
Fellner, North Holland Press, 1977.
Newman, Howard, “Strategic Groups and the Structure-Performance Relationship,” Reoiew of
Economics and Statistics, 60 (August 1978) 417-427.
618 ECONOMIC INQUIRY

Oster, Sharon, “An Analysis of Some Causes of Interstate Differences in Consumer Regulations,”
Economic Inquiry, 18 (January 1980) 39-54.
, “Intraindustry structure and the ease of strategic change,” Reoiew of Economics and
Statistics, 1982.
Peck, M. J., Competition in the Aluminum Industry, Cambridge: Harvard University Press, 1961.
Peltzman, Sam, “Toward a More General Theory of Regulation,” Journal of Law and Economics, 19
(August 1976) 211-240.
Pharmaceutical Manufacturer’s Association, Various Press Releases, 1978 and 1979.
Porter, Michael, “The Structure Within Industries and Companies’ Performance,” Reoiew of
Economics and Statistics, 61 (May 1979) 214-227.
Porter, Michael and Spence, Michael, “The Capacity Expansion Process in a Growing Oligopoly: the
Case of Corn Web Milling,” HIER Discussion Paper 670, November 1978.
Salop, Steven, “Strategic Entry Deterrence,” American Economic Reoiew, May 1979.
Stigler, George, “The Theory of Economic Regulation,” Bell Journal (Spring, 1971).
Wishard, William, “Corporate Response to a New Environment,” Law and Contemporary Problems,
41 (Summer 1977) 222-244.