Optimal Franchising

Author(s): Roger D. Blair and David L. Kaserman

Source: Southern Economic Journal , Oct., 1982, Vol. 49, No. 2 (Oct., 1982), pp. 494-505
Published by: Southern Economic Association

Stable URL: https://www.jstor.org/stable/1058499

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Southern Economic Association is collaborating with JSTOR to digitize, preserve and extend
access to Southern Economic Journal

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 Optimal Franchising*
ROGER D. BLAIR

University of Florida
Gainesville, Florida

DAVID L. KASERMAN

University of Tennessee
Knoxville, Tennessee

I. Introduction

An important series of papers has shown that a firm possessing monopoly power over a
intermediate product which is employed by a competitive industry that combines its inputs
in variable proportions will have an incentive to exercise some form of vertical contro
- that is, to influence the input mix and output decisions of firms at the downstrea
stage.' Later papers have shown that such vertical control may assume at least five gener
forms that, under the stylized conditions postulated, provide economically equivalen
results. These forms are: ownership integration, input tying, output royalties, sales roy
ties, and lump-sum entry fees.2 Thus, within the context of the extant literature, t
intermediate product monopolist is seen to face a smorgasbord of vertical control al-
ternatives from which to choose, with all dishes equally capable of yielding the essent
control necessary to maximize profits.
If the alternatives available were, in fact, economically equivalent in real world
situations, an input monopolist would view them as perfect substitutes and would be
indifferent as to which particular mechanism was employed. Moreover, under these
conditions, we would not expect to observe the monopolist making use of more than o
vertical control mechanism in any given market situation. Such indifference and specia
ization, however, are seldom observed in the complex contracts that are negotiated in

*The authors express appreciation for the financial support provided by the Public Policy Research Center a
the University of Florida. Also, Kaserman received financial support from the College of Business Faculty Resear
Awards Program at the University of Tennessee. Helpful comments on a previous draft of this paper were provided b
Sidney Carroll and an anonymous referee. The usual disclaimer applies.
1. The main contributors have been [5; 9; 15; 17; 18].
2. Formal proof of the economic equivalence of tying and vertical integration is provided in [2]. Equivalenc
of output royalties, lump-sum entry fees, and vertical integration is contained in [3]. And equivalence of sales royalti
and vertical integration is proved in [11].

494

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 OPTIMAL FRANCHISING 495

franchise, labor union, and patent licensing agreements.3 Typically, these con
corporate two or more of the above control options. Most commonly, perhaps
combine the royalty on output or sales with the lump-sum entry fee.4
This paper examines the use of this particular mixed strategy in the conte
franchise agreement. While the combined use of lump-sum entry fees and outpu
royalties is representative of several other types of vertical contracts, this narr
focus to the franchise agreement allows us to incorporate some interesting inst
detail in our discussion and seems justified by the absolute importance of this
business organization. As Caves and Murphy [6, 572] recently noted:
Franchised businesses account for over 38 percent of all retail sales in the United
States and originate 12 percent of the gross national product, yet the franchise
relationship has largely escaped economic analysis.

The reader should bear in mind, however, that the results we present are also ap
other vertical contractual agreements that combine these strategies.
By viewing the franchise contract as a combination of lump-sum entry fe
royalty on output, we are able to shed new light on both the purposes served b
these two vertical control mechanisms together and the underlying factors tha
expected to influence the structure of the optimal contract. In addition, the a
provides some insight regarding the sources of important post-contractual co
between the parties involved in the agreement.5

II. Franchisor Optimization of Contract Terms

To examine the combined use of entry fees and output royalties, we make use of
two-period model in which a downstream firm (franchisee) pays an upstre
(franchisor) a lump-sum (franchise) fee equal to F in period one in exchange for
to produce and sell a final product in period two. In addition to this franchise
franchisee agrees to pay the franchisor a per-unit output royalty equal to R in t
period. The simplifying assumption of only two periods does not appear restrict
views the second-period profits of the two firms as the present worth of all future
over the lifetime of the contract.

Regarding market structure at the two stages of production, we assume that the
franchisor holds a monopoly over a differentiated intangible asset (a trademark) that is
employed as an input by the franchisee. And we assume competition among franchisees in
the purchase of the franchise.
Next, to allow for the possibility of uncertainty and/or divergent expectations
concerning the future level of final product demand, we assume that the discount rates

3. These various kinds of vertical contracts can easily be seen to be isomorphic in nature, with the only
economically meaningful distinction being the particular source of upstream monopoly power (i.e., a trademark versus
a labor union or a patent).
4. Caves and Murphy [6, 579] point out that, where inputs are employed in fixed proportions to output, a
tying arrangement is analytically equivalent to a royalty on output or sales.
5. As Goldberg [8] and Klein [13] recently pointed out, such conflicts are currently providing the subject for a
great deal of statutory, regulatory, and antitrust activity.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
496 Roger D. Blair and David L. Kaserman

implicit in the behavior of the two firms at the time the contract is negotiated may differ
from one another. Due to what Williamson [19] has called "strategic misrepresentation
risk," the franchisor will have an obvious incentive to overestimate the future demand for
the franchised product, and the franchisee will have an equally obvious incentive to
underestimate it. Letting ru represent the franchisor's implicit discount rate and rD the
franchisee's implicit discount rate, this sort of opportunistic behavior implies ru 5 rD,
where the equality will hold only in the rare case of complete certainty on the part of both
negotiating parties regarding future demand.6
Under the above assumptions, the present value of the franchisor's profits at the time
the contract is negotiated will be

7ru = F + [1/(1 + ru)]RQ(R), (1)

where the rate of output at the downstream stage in the

upon the level of the output royalty but not the fixed fe
franchisee's profits are

rD = [1/(1 + rD)]{P[Q(R)]Q(R) - C[Q(R)] - RQ(R)} - F, (2)

where C[Q(R)] is the total cost of production at the downstream stage net of the
royalty and franchise fee, and P[Q(R)] is inverse demand.
Assume that the downstream firm faces a linear inverse demand for the final pro
that is given by P(Q) = a - bQ and a linear total cost curve that is given by C(Q
where a, b, and c are positive constants.8 Then, the downstream firm's profits
rewritten as

7rD = [1/(1 + rD)] {aQ(R) - b[Q(R)]2 - cQ(R) - RQ(R)} - F. (3)

In the absence of any monopsony power in the purchase of the franchise,9 the
franchisee will attempt to maximize lrD over Q with R and F taken as given. Such
maximization implies an optimal level of final-product output in the second period of

Q* = (a - c - R)/2b. (4)

This optimal output, then, will respond inve

royalty imposed by the upstream firm. We can s

8Q*/IR = - 1/2b < 0, (5)

which is the inverse of the slope of the final-produ

At the time the contract is negotiated, the f
combination of the entry fee and output royalty t

6. Other reasons might exist for a divergence in implicit d

example, Rubin [14] presents an argument that the franchisor's c
Allowing for different implicit discount rates is an analytically c
variety of specific sources of uncertainty, divergent perceptions,
7. Note that this simple specification ignores any costs associa
future period. Since our concern here is with the design of the o
convenient to interpret the effects of policing through the implic
8. These assumptions of linearity are made for convenience
9. Competition among other potential franchisees prior to
monopsony power. As Goldberg [8, 108] points out, franchisors ty
potential franchisees on a take-it-or-leave-it basis without any ne

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 OPTIMAL FRANCHISING 497

that the potential franchisee be able to earn at least a normal profit. At the same
competition among potential franchisees for the franchise will ensure that no more than
normal profit need be offered in the terms of the contract. We may, then, characterize
franchisor's optimization problem as

Max rr,
F, R
s.t. rrD = 0,

where ir* is maximized profit at the downstream stage with F and R taken as given.
The Lagrangian for this problem is given by

L = F + [1/(1 + r,)] RQ(R) + X(F - [1/(1 + rD)]{aQ(R)

- b[Q(R)] - cQ(R) - RQ(R)}),

with first-order conditions:

aL/aF = 1 + X = 0, (6)

8L/8R = [1/(1 + r,)][Q + R

[a(oQ/8R) - 2bQ(8Q/
8L/8X = F- [1/(1 + rD)](aQ - bQ2 - cQ - RQ)= 0. (8)
From equation (6), we have A = --1, i.e., profits earned by the franchisee are
the franchisor. Substituting this value for X, equation (4) for Q, and e
8 Q/R all into equation (7), we can solve for the optimal level of the out

R* = g (a - c), (9)
where

g = [1/(1 + ru,)- 1/(1 + rD)]/[2/(1 + ru,)- 1/(1 + rD)]. (10)

Now, substituting (9) and (4) into (8), we may solve for the optimal level of the lump-
sum entry fee as

F* = [1/(1 + rD)][(l - g)(a - c)]2/4b. (11)

Equations (9) through (11) define the terms of the optimal (profit-maximizing)
contract from the franchisor's point of view. Given our assumption of competition among
franchisees in the purchase of the franchise, these are the terms that will result. The
remainder of the paper is devoted to an analysis of the implications of this optimal
contract.

III. Uncertainty and the Use of the Mixed Strategy

Previous authors have presented arguments that what we refer to as a mixed strat
which we simply mean the utilization of more than one vertical control mechan

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
498 Roger D. Blair and David L. Kaserman

simultaneously) will arise only in response to some kind of uncertainty at the negotiating
stage. Caves and Murphy [6, 577] indicate that, with perfect information, the fixed fee
would be used alone. In this situation, the franchisor would simply place the franchise up
for bid, and the successful bidder would be the franchisee that is most efficient in using the
intangible asset.'0 The addition of other contractual provisions for ongoing franchisee
payments is attributed to risk aversion on the part of the franchisee and/ or the need for
some assurance that the franchisor will, in fact, provide the (costly) policing service
necessary to maintain the value of the franchise. Klein [13, 360] also emphasizes the
possibility of franchisor cheating by failure to satisfactorily police the performance of
other franchisees."

The above model may be used to confirm these arguments. Since failure of the
franchisor to perform the policing function will have the effect of reducing the future level
of demand for the franchised product through degradation of the trademark, the potential
for this sort of hold-up problem will have the effect of increasing the value of rD above ru in
our model. In other words, it is the potential franchisee's uncertainty that the franchisor
will fulfill his obligation to maintain the value of the trademark that leads him to discount
the future level of demand more highly than does the franchisor. Moreover, the greater the
incentive and opportunity for such franchisor hold-ups, the greater the expected diver-
gence in implicit discount rates. Thus, in the absence of this (and other forms of)
uncertainty, implicit discount rates will be equal.
The divergence or equality of implicit discount rates has a strong influence on the
terms of the optimal franchise contract. This influence is summarized in the following
three propositions:
Proposition 1: With future levels of final product demand certain (i.e., with ru = rD),
the franchisor will not make use of the output royalty but will, instead, extract all
monopoly rent through the fixed fee.12
The reason for this result is fairly simple. Since the future demand is certain, the
stream of future profits is also certain. Competition among potential franchisees will bid
up the entry fee to the present value of that profit stream.
Proposition 2: Under broadly defined conditions of uncertainty concerning future
levels of final product demand (i.e., with r, < rD), the franchisor will empoy both a fixed
fee and an output royalty.13

10. With perfect information, the lump-sum entry fee is equivalent to vertical integration [3].
I 1. Klein [ 13] points out that the incentive for such franchisor cheating is at least partially offset by the effect of
such cheating on the ability of the franchisor to market new franchises and any cost differences that exist in operating
company-owned versus franchised outlets. On the latter subject, see [16].
12. This proposition is fairly easy to establish. From (10), r, = rD implies g = 0. Equation (9) then yields
R* = 0. Substituting g = 0 into equation (11) yields

F* = [1/(1 + rD)](a - c)2/4b.

Substituting equation (4) into (3) and setting R = 0 and F = 0, we find the maximum profit at the downstream
(franchisee) stage to be

rDlI Ro = [l/(l + rD)](a - c)2/4b.

FO0

Thus, F* = 7rI F-0

R=O', and the proof is complete.
13. This proposition is established directly. From (10), ru < rD implies 0 < g < 1. Then, (9) and (11)
R*, F* > 0.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 OPTIMAL FRANCHISING 499

Proposition 3: The existence of uncertainty and the mixed strategy that i

reduces the franchisor's potential profit below that attainable under perfect informa
These propositions demonstrate that the use of an output royalty in conjunction
an entry fee is not the preferred way of doing business. Rather, the output royalty
to supplement the entry fee in an effort to deal with the annoying complication of
discount rates.15 Since the value of the franchise contract at negotiation depends
post-contractual behavior of both parties, the use of a mixed strategy may be s
adaptation to the firms' inability to write a completely contingent costlessly en
contract. Thus, where uncertainty of future profitability makes full capitaliz
monopoly rents impossible, we shall observe a combination of an entry fee and
royalty.
Such uncertainty can be expected to be pervasive at the time the contract is nego-
tiated. To calculate the optimal entry fee, the upstream firm must know the demand curve
for final output in each geographic market not only in the present period but in all future
periods to which the contract applies. In addition, it must know the downstream firm's
production costs [3]. These informational difficulties are not confined to the upstream
firm. Divergent estimates of the future profitability of producing the final good over some
specified period may render a mutually agreeable entry fee nonexistent. Even under
uniform expectations, divergent attitudes toward risk may yield the same result. More-
over, the downstream producer is not apt to pay the present value of full monopoly profit
without some guarantee of being able to collect such profit in the future [1]. Once the
franchisor collects an exhaustive franchise fee, the franchisee must be concerned about the
franchisor's incentive to sell more franchises.16 Thus, the franchisor must find a way to

14. Perfect information implies that ru = rD and, hence, g = 0. Thus, we want to show that

uIg=0 >rUlg>0"
Substituting equations (4), (9), and (11) into equation (1), and setting g = 0 and rD = r on the LHS, we have

[1/(1 + ru)](a - c)2/4b > [1/(1 + rD)][(l - g)(a - c)]2/4b

+ [1/(1 + ru)][g(l - g)(a - c)2/2b].

Some algebra yields

{[1/(1 + ru)] - [1/(1 + rD)]}(a - c)2 / 4b > {[1/(1 + ru)] - [1/(1 + rD)]}

g(a - c)2/2b - {[2/(1 + ru)] - [1/(1 + rD)]}g2(a - c)2/4b.

Now, divide both sides by [2/(1 + ru)] - [1/(1 + rD)] and recall equation (10),

g(a - c)2/4b > g2(a - c)2/4b

or

I >g,

and the proof is complete.

15. It is worth noting that when rD < ru, i.e., when the franchisee is more optimistic about future demand fo
franchised product than the franchisor (perhaps because of information about a local market or superior man
skills that, due to impacted information, cannot be communicated), the model implies F* > 0 and R* < 0. In th
then, the franchise contract contains provisions for franchisor payments to the franchisee in the future period.
helps to explain the observed phenomenon of franchisors providing advertising, management training, etc. at
cost. The potential for post-contractual franchisor cheating in this case is, of course, more severe.
16. The problem here is analogous to that analyzed by Coase [7]. Coase considered the plight of the mono
owner of a fixed quantity of a completely durable good. In order to charge a supra-competitive price some of the
supply would have to go unsold. The monopolist could not charge such a price because he could not guarante
buyers that he would resist the temptation to sell additional units in the future.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
500 Roger D. Blair and David L. Kaserman

guarantee the absence of future competition. Such guarantees may be extremely difficult
to provide and, in the case of multiple geographic markets, may have to be buttressed with
exclusive dealing contracts or territorial restrictions [6, 576-7]. Since these ancillary
restrictions can cause antitrust problems, this is a serious drawback to relying solely upon
entry fees. Moreover, markets change over time. As customer density increases, a given
geographic market may support more than one franchise. If there are diseconomies of
scale, optimality may require additional franchisees. Finally, there is the danger of new
entry at the upstream stage. For example, a franchisor may not be able to prevent the
development of a close substitute for the franchised product.'7 Consequently, franchisees
would have reduced future profits. For all of these reasons, the franchisor is apt to be
unable to capitalize fully the monopoly profits at the downstream stage through use of the
entry fee alone and will be forced to make use of a mixed strategy.

IV. Factors Affecting Contract Terms

The next two propositions describe how the optimal franchise contract will be altered by
variations in the implicit discount rates of the two firms. As might be expected, the
optimal level of the fixed fee increases and the output royalty decreases as the franchisor's
discount rate rises or as the franchisee's discount rate falls. In other words, when rD - ru is
large, more reliance is placed on the output royalty R. In contrast, when rD - ru is small,
more reliance is placed on the entry fee F.
Proposition 4: Under a mixed strategy (i.e., R*, F* > 0), the optimal level of the
entry fee will increase (decrease) with increases in the franchisor's (franchisee's) implicit
discount rate.8
Proposition 5: Under a mixed strategy (i.e., R*, F* > 0), the optimal level of the
output royalty will increase (decrease) with decreases in the franchisor's (franchisee's)
implicit discount rate.19
The significance of these results lies in focusing our attention on those aspects of the
franchised business that can be expected to influence the implicit discount rates of the two

17. In the fast food field, one can see extensive and continuing entry by imitators of the original and highly
successful fast food franchisors.
18. We can prove this proposition quite easily. From (11), we have

OF*/dg = - [1/(1 + rD)](l - g)(a - c)/2b

which must be negative since (1 - g) > 0 and (a - c) must be positive for negatively sloped demand curves or
output would be negative. From (10), we have

'g/laru = - (1 + rD)/[(l + rD) + (rD - ru)]2 < 0.

Further,

Og/ roD = (1 + ru)/[(l + rD) + (ro - ru)]2 > 0.

Consequently,

F* / ru = ( F* / g)(ag/ ru) > 0, and a F* / arD = (a F* / g)(ag/ rD) < 0.

19. From (9), we have OR*/ag = a - c > 0. Then, since ag/lru < 0 and ag/lar > 0, OR*/dru
(R* / g)(dg/dru) < 0 and OR*/r rD= (R* / g)(ag/ rD) > 0.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 OPTIMAL FRANCHISING 501

negotiating parties. Given these aspects and corresponding hypotheses regarding

direction of their impacts on r, and rD, these propositions provide testable hypo
concerning determinants of the expected structure of observed contracts. A comp
with the prior literature is instructive.
Rubin [14, 229-30] discusses two basic factors that may be expected to influenc
relative use of the fixed fee versus the output or sales royalty. First, he argues
franchised businesses that require a relatively large degree of managerial discretion
sale (and service) of the final product will tend to make relatively heavy use of the
fee. In terms of our model, this simply means that the future period demand is rel
dependent upon the behavior of the franchisee as opposed to the franchisor. Such
would imply relatively low franchisee discount rates and relatively high franchisor disc
rates since the burden of uncertainty regarding future demand will be borne primarily
the latter. As we have just seen, both of these effects will be reflected in higher fixed
and lower output royalties.20
Second, Rubin argues that increases in the relative value to the business o
trademark itself should lead to a relatively heavy reliance on the output royalty. I
model, a greater value of the trademark means that future demand will be more depend
upon the post-contractual behavior of the franchisor in policing the performance of
franchisees to avoid devaluation of the trademark. This, in turn, implies a relative
discount rate for the franchisor and a relatively high discount rate for the franchisee,
of which lead to decreases in the optimal level of the fixed fee and increases in the opti
level of the output royalty.
Other hypotheses may be derived by examining additional aspects of the bus
that can be expected to influence the implicit discount rates of the negotiating pa
Within the context of the hold-up problem discussed by Klein [13], r, will be increa
anything that increases the opportunities and incentives for franchisee cheating w
will be increased by anything that increases the chances that the franchisor will cheat.

V. Post-Contractual Conflicts

Post-contractual conflicts between franchisors and franchisees have provided employm

opportunities for a great many lawyers (and more than a few economists). The abo
model may be used to shed light on some of the sources of such conflicts. In the sect
below, we discuss two of these.

Franchisor Perception of Franchisee Profits

We showed earlier that the franchisor will resort to the use of a mixed strategy on
response to a divergence in the implicit discount rates of the two firms participating in t
franchise contract, i.e., when ro - r, > 0. We now show that this same divergence
the franchisor to perceive the existence of positive profits at the downstream stage after
contract is negotiated.

20. This interpretation is also consistent with Klein's [ 13, 358] description of the fixed fee as a forfeitab
will collateral bond that is posted by the franchisee to ensure desirable post-contractual performance.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
502 Roger D. Blair and David L. Kaserman

Proposition 6: Under a mixed strategy (i.e., R*, F* > 0), the franchisor will perceive
that positive profits are being earned by the franchisee.21
This proposition indicates that the present value of the future economic profits to the
franchisor, [1/(1 + ru)] [(1 - g)(a - c)]2/4b, exceeds the present value to the franchisee,
[1/(1 + rD)] [(1 - g)(a - c)]2/4b. Clearly, this result is due to the difference in discount
rates between the franchisor and the franchisee. But we have specified such a difference to
reflect different expectations or attitudes toward risk. This difference in valuation prevents
the franchisor from capturing all of the downstream profits that it perceives and may
provide a strong incentive for forward integration. With positive rents being perceived at
the downstream stage, the franchisor may open downstream subsidiaries in an attempt to
capture these rents.22
Moreover, as the franchised business matures, the risk faced by the franchisor may
decline as it is spread over an increasing number of franchisees [14]. Consequently, there is
some reason to believe that ru may fall over time leading the mature franchisor to pursue
an integrated structure more vigorously than a new franchisor.23 This implication is
consistent with the empirical evidence discussed by Caves and Murphy [6].
In practice, however, it may not be possible for the franchisor to vertically integrate
forward. For one thing, the costs of doing so may exceed the differential profit [6]. An
example of how this may occur is provided by Shelton [16]. He analyzed the business
results of a franchise with company-owned outlets as well as independently owned
franchise operations. Shelton reported that the independent franchisees performed better
than the company-owned outlets. This performance differential was attributed to X-
inefficiency, which may not be the correct term. There may be simple managerial dis-
economies of scale at work. But whatever one wants to call this phenomenon, one thing is
apparent: there are costs to vertical integration that may offset the gains of vertical
integration.
For another thing, the franchisor may be concerned about the possibility of antitrust
action.24 As forward integration proceeds, the franchisees may complain about being
squeezed by the franchisor and file a civil suit against him. Alternatively, the Antitrust

21. To reflect the upstream firm's perception of profits at the downstream stage, substitute ru for rD in
equation (3). Then, substituting equation (9) for R and (11) for F, the upstream firm's perception of downstream
profits is

Pu(ir) = [1/(1 + ru)](a - bQ2 - cQ - g(a - c)O]

- [1/(1 + rD)][(l - g)(a - c)]2/4b.
Substituting (4) for Q and simplifying yields

P,(ir*) = {[1/( + ru)] - [1/(1 + rD)]}[(l - g)(a - c)]2/4b >0

when ru < ro.
22. In addition, the operation of downstream subsidiaries may provide the franchisor valuable information
concerning costs at the final-good stage. This cost information is necessary in determining the optimal level of the
franchise fee and output royalty.
23. This may explain the experience of the Holiday Inn franchise system. More systematic evidence has been
provided by Hunt [10]. He analyzed some 95 fast food franchise systems to determine any relationship between the age
of the system and the percentage of outlets that were company operated. He found a statistically significant and
positive relationship.
24. Such a concern does not arise out of idle speculation. Rather, it springs from the Antitrust Division's
expressed interest in dual distribution situations [12]. Since William Baxter became head of the Antitrust Division,
however, this interest has waned considerably.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 OPTIMAL FRANCHISING 503

Division could charge the franchisor with attempting to monopolize the downstre
industry. Given the Antitrust Division's present attitude toward vertical business relation
this may not be too likely. Nonetheless, if the franchisor feels unable to vertically integr
forward, the perceived presence of uncaptured profit will be a continuing source
frustration and will cause tension between the franchisor and its franchisees.

Post-Contractual Hold- Up Incentives

To this point, the model that we have been analyzing requires that the potential franchise
be able to earn a normal profit from the operation of the franchised business. Th
requirement is certainly valid at the time the contract is negotiated; but, given th
necessarily incomplete terms of such relational contracts [8], it may be possible for
franchisor to increase profits after the contract is in effect by encouraging or forcing t
franchisee to produce beyond the profit-maximizing point (Q* in equation (4)). Suc
behavior corresponds to Klein's [13] description of the hold-up problem.
Obviously, if an output royalty is in effect, the franchisor's post-contractual pr
will be maximized if the franchisee can be forced to maximize output. And if a royal
sales is employed, the franchisor would like to see sales revenue maximization at t
downstream stage. In either case, an output beyond Q* is desired by the franchisor, w
places the franchisee in a loss situation. At least two avenues may be open to the franchis
for the pursuit of franchisee output expansion.
First, the franchisor may impose maximum resale prices.25 With the output roy
the franchisor's post-contractual profit-maximizing final output price would be zero. Wit
a sales royalty and the linear demand curve employed above, the franchisor's unco
strained post-contractual optimum price would be P = a/ 2. Since these optimum m
mum prices are both below the franchisee's profit-maximizing price of P = (a + c + R
imposition of such maximum prices will be effective in eliciting the desired outp
expansion.
Second, the franchisor could attempt to force the franchisee to the optimum lev
output by establishing equivalent performance standards or quotas. With the outp
royalty, the franchisor could impose a minimum output of Q = a/b; and, with a s
royalty, this minimum could be set at Q = a/2b. These quotas are both above
franchisee's profit-maximizing output of Q = (a - c - R)/ 2b and will, therefore, forc
franchisee into an ex post loss situation.
Although a franchisor might conceivably make use of these hold-up strategies, severa
important constraints exist to discourage such behavior.26 First, under the cu
Supreme Court interpretation, fixing maximum resale prices is a per se violation of
antitrust laws.27 Second, since maximum prices or performance standards that resu

25. Caves and Murphy [6, 579] cite evidence that indicates that the preponderant pressures exerte
franchisors on franchisees are designed to lower output prices.
26. Hold-up opportunities also exist for the franchisee, and the franchise contract may be viewed
instrument that is designed to balance the incentives of the two parties to engage in these various fo
opportunistic behavior [13].
27. See Albrecht v. Herald Co., 390 U.S. 145 (1968), for the controlling decision. In that case, the publis
the Globe-Democrat tried to impose maximum resale prices on its newspaper distributors. The Court was offen
the publisher's substitution of his business judgment for that of the distributor and ruled against the publisher.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
504 Roger D. Blair and David L. Kaserman

franchisee losses must be imposed after the contract is in effect, franchisees may call upon
recent court decisions in which "unfair" contractual provisions have been declared un-
enforceable [13]. And third, any such hold-up behavior on the part of a franchisor will
result in a loss of both current and future franchisees.28 As Klein [13] points out, the higher
costs of operating the business through owned subsidiaries can provide an effective
constraint against such franchisor cheating. Despite these institutional and market con-
straints, however, it seems likely that much of the observed conflict between franchisors
and franchisees could be traced to these kinds of hold-up incentives.

VI. Concluding Remarks

Franchising is a form of vertical integration by contract. In the United States economy, it

is a very important form since it accounts for such a large portion of national income. In
this paper, we have extended our understanding of the franchise relationship. In particular,
we have examined the widespread use of a lump-sum franchise fee in conjunction with a
royalty on sales revenue or output as a means of extracting rent by the franchisor. We
have seen that the use of this combination can be attributed to uncertainty emanating
from various sorts of strategic behavior as well as different expectations of future demand.
The basic consequence of this uncertainty is an inability to capitalize fully the monopoly
profit inherent in the franchise. We have seen that this causes a tension between the
franchisor and its franchisees. This tension creates incentives for forward integration by
ownership or for other forms of vertical control such as sales quotas or maximum resale
prices.
Public policy, as embodied in the antitrust laws and some state statutes, should
benefit from an enhanced understanding of the tensions between franchisors and their
franchisees. For example, we have seen that a franchisor may resort to fixing maximum
resale prices in an effort to expand its own profit. The courts have been quite hostile
toward such a business practice. In fact, fixing maximum resale prices is a per se violation
of the antitrust laws.29 Although the franchisor is motivated by his own self interest, fixing
maximum resale prices will improve consumer welfare as output expands. Judicial recog-
nition of this fact could alter the current Supreme Court doctrine. Given the pervasiveness
of franchising in our economy, public policy governing franchisor - franchisee relations
should be grounded in a clear understanding of the basis for these relations.

28. This provides another potential explanation for observed trends toward vertical integration among mature
franchises. With fewer new franchises to sell, the disincentive to hold-up existing franchisees may fall. Such hold-ups,
in turn, lead to the eventual loss of existing franchisees, and the franchisor will be forced to take over the operation of
downstream outlets.
29. For an evaluation of the current judicial attitude toward fixing maximum resale prices, see [4].

References

1. Bierman, H. and R. Tollison, "Monopoly Rent Capitalization and Antitrust Policy." Western Economic
Journal, December 1970, 385-89.
2. Blair, R. and D. Kaserman, "Vertical Integration, Tying, and Antitrust Policy." American Economic
Review, June 1978, 397-402.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms
 OPTIMAL FRANCHISING 505

3. , "Vertical Control with Variable Proportions: Ownership Integration and Contractual Equiva
Southern Economic Journal, April 1980, 1118-28.
4. , "The Albrecht Rule and Consumer Welfare: An Economic Analysis." University of Florida
Review, Summer 1981, 461-84.
5. Burstein, M., "A Theory of Full-Line Forcing." Northwestern University Law Review, March/ Apr
62-95.
6. Caves, R. and W. Murphy, II, "Franchising: Firms, Markets, and Intangible Assets." Southern Economic
Journal, April 1976, 572-86.
7. Coase, R., "Durability and Monopoly." Journal of Law and Economics, April 1972, 143-49.
8. Goldberg, V., "The Law and Economics of Vertical Restrictions: A Relational Perspective." Texas Law
Review, December 1979, 91-129.
9. Hay, G., "An Economic Analysis of Vertical Integration." Industrial Organization Review, 1973, 188-98.
10. Hunt, S., "The Trend Toward Company-Operated Units in Franchise Chains." Journal of Retailing,
Summer 1973, 3-12.
11. Inaba, F., "Franchising: Monopoly by Contract." Southern Economic Journal, July 1980, 65-72.
12. "Justice Takes Aim at Dual Distribution." Business Week, July 7, 1980, 24-5.
13. Klein, B., "Transaction Cost Determinants of 'Unfair' Contractual Arrangements." American Economic
Review, May 1980, 356-62.
14. Rubin, P., "The Theory of the Firm and the Structure of the Franchise Contract." Journal of Law and
Economics, April 1978, 223-33.
15. Schmalensee, R., "A Note on the Theory of Vertical Integration." Journal of Political Economy,
March/ April 1973, 442-49.
16. Shelton, J. "Allocative Efficiency v. 'X-Efficiency': Comment." American Economic Review, December
1967, 1252-58.
17. Vernon, J. and D. Graham, "Profitability of Monopolization by Vertical Integration." Journal of Political
Economy, July/August 1971, 924-25.
18. Warren-Boulton, F., "Vertical Control with Variable Proportions." Journal of Political Economy, July/
August 1974, 783-802.
19. Williamson, O., "The Vertical Integration of Production: Market Failure Considerations." American
Economic Review, May 1971, 112-23.

This content downloaded from

103.126.229.39 on Mon, 18 Mar 2024 19:54:05 +00:00
All use subject to https://about.jstor.org/terms