SORTING, FRANCHISING AND REAL ESTATE BROKERAGE FIRMS
John D. Benjamin
Kogod School of Business, American University
4400 Massachusetts Ave., NW
Washington, DC 20016
(202) 885-1892
jbenj@american.edu
Peter Chinloy
Kogod School of Business, American University
4400 Massachusetts Ave., NW
Washington, DC 20016
(202) 885-1951
chinloy@american.edu
Daniel T. Winkler
Bryan School of Business and Economics
University of North Carolina at Greensboro
Greensboro, NC 27412-5001
(336) 334-3094
dt_winkler@uncg.edu
We are grateful to Paul Bishop, Anton Haidorfer, Ellen Roche, and Marion Steele for their
helpful comments and discussions. The data set and survey information was provided by the
National Association of Realtors.
SORTING, FRANCHISING AND REAL ESTATE BROKERAGE FIRMS
Abstract
A franchise serves as one mechanism for sorting real estate residential brokerage firms.
Each firm evaluates the inputs of marketing and systems provided by franchisors. If the firm is
more profitable by accepting a franchise, then that structure is selected. In a sample of 1,143
United States residential brokerage firms, large size is associated with franchising. However, the
benefits vary regionally. Local, non-franchised firms are able to hold their own in the Northeast
generally a slower-growing areas. Franchises have a larger contribution to net margins in the
faster-growing South and West.
Keywords: Franchise, Brokerage, Residential Real Estate, Profit Margin
2
SORTING, FRANCHISING AND REAL ESTATE BROKERAGE FIRMS
1.
Introduction
This paper develops a model where a franchise serves as one mechanism for sorting by
size and scale within real estate residential brokerage firms.1 If a franchisor’s inputs such as
national marketing are complementary with size, then larger firms tend to become franchisees.
The empirical analysis is divided by region. This allocation allows for testing whether
franchising is more prevalent in faster-growing markets.
There has been related research on franchising by real estate brokerage firms, although
there is less emphasis on sorting theories of the firm. Having a franchise pays off on a gross
revenue basis and supports positive royalty fees. Jud, Rogers, and Crellin (1994) show a 9%
increase in revenue for real estate brokerage firms with franchise affiliation.2 Richins, Black,
and Sirmans (1987) and Frew and Jud (1986) also support the idea that franchise affiliation
increases revenues. Regarding cost and technology, Anderson, Lewis, and Zumpano (2000)
demonstrate that franchised firms are more efficient than their non-franchised counterparts, but
franchises may not be more profitable.3 Lewis and Anderson (1999) show that franchised
brokerage firms have lower costs than non-franchised firms, but the average brokerage firm
operates close to its efficient frontier. Franchised firms are found to be more efficient in
allocating resources by Anderson and Fok (1998), but non-franchised firms have more scale and
technical efficiency.
This paper is complementary, examining franchising as a selective process. Variables
such as size are tested for the decision on whether to acquire a franchise. Firm net margin, or net
operating income divided by revenue, reflects the self-selection decision of holding a franchise.
If the franchise decision depends on size and scale, that decision would be accounted for in the
net margin. There is no requirement that this selection structure be identical across regions,
allowing for comparisons across markets.
1
In franchising, the franchisor or parent firm offers inputs such as common marketing, technology, and training in
exchange for an upfront fee and royalty payments (typically a percentage of revenue). There are over one million
Realtors® according to the National Association of Realtors® (see www.realtors.org) and approximately one-third
are employed with franchised brokerage firms. The International Franchise Association (www.franchise.org)
estimates that in 2000, franchisors and their franchisees had volume of $1 trillion, more than 40 percent of all U.S.
retail sales. These sales originated from over 300,000 franchised businesses in 75 industries. Franchising is
estimated to employ more than 8 million people.
2
The estimates of Jud, Rogers, and Crellin (1994) are after subtracting royalties, fees, and other charges associated
with franchise affiliation. The role of brokerage firms in helping to set listing prices in a competitive marketplace
has been examined by Sirmans and Turnbull (1997) and Knight, Sirmans, and Turnbull (1994).
3
Zumpano, Elder, and Anderson (2000) note that greater firm costs may have led to increased consolidation among
brokerage firms to take advantage of economies of scale. Consolidation has encouraged even more franchising as
franchising allows brokerage firms to substitute variable costs for fixed costs so as to reduce break-even output
levels.
3
The franchisor offers a set of inputs not directly available to non-members or nonaffiliates. These inputs include common marketing, affiliation programs, branding, training and
technology such as databases and software. The franchisees specialize in inputs in their specific
markets, such as local labor and facilities and clientele networks. If the franchisor-provided
inputs are complementary with the local inputs, then there is a benefit from retaining a franchise.
Franchisors have incentives to seek out those inputs that are complementary with their
franchisees.
On the demand side, prospective franchisees evaluate net profits and returns from being
in a franchise or remaining independent. Comparing the two involves a selection decision.
Those accepting a franchise have a higher expected return even when subtracting the royalties
and fees from the value of franchisor-provided inputs.
On the supply side, the franchisor will accept the firm offering the highest expected profit
or royalty revenue less costs of franchisor-provided inputs. The equilibrium sorts the type of
firm by ability and equates demand with supply.
There are resulting profit or performance conditions for franchise and non-franchise
firms. The return to having a franchise is the difference between the expected return with having
a franchise and with not having one, controlling for self-selection. This test is possible because
real estate residential brokerage firms enter and exit franchising, and because franchise and nonfranchise firms coexist. That flexibility is not possible in industries such as new car sales where
there is no coexistence or hotels where much of the franchise requirement is unobservable capital
investment.
To test for firm sorting, residential brokerage firm data are used from a National
Association of Realtors survey of 1,143 member firms in 2001. In the sample, 313 firms are
franchisees and the remaining 830 are non-franchisees. The results show that there is sorting by
franchise selection. The probability of holding and being accepted for a franchise is increasing at
each level of firm size and with measures of technological investment. At each size grade, larger
firms tend to be franchisees. Once self-selected, both franchised and non-franchised firms’ net
margins increase in technology investment and decline with age. Even though National
Association of Realtors data do not contain complete firm cost information, the net margin or
revenues less expenses data take costs into account.
Larger firms earn a return from holding a franchise. However, the higher return does not
obtain in all regions. The self-selection return to holding a franchise is positive in high-growth
areas for the South and West. In the slower-growing Northeast and the Midwest, holding a
franchise has a low or zero added return.
2. Franchising and Real Estate Firms
Franchises for residential real estate brokerage firms impose limited and uniform initial
capital requirements and have similar royalty fees. Real estate franchisors do not operate
4
company-owned stores, as in the fast-food business. Moreover, unlike in the new car business
protected by state regulatory barriers, franchisees and non-franchisees coexist.
For illustrative purposes, Table 1 compares the costs of acquisition for three of the most
widely-held real estate brokerage franchises, Century 21, ERA, and Coldwell Banker, with
popular restaurant franchises.4 The Subway restaurant franchise is generally considered to be
one with relatively low acquisition and entry costs.
Table 1. Comparative Costs for Franchises, Inside and Outside Real Estate
Century 21 ERA
Coldwell
Subway
McDonalds
Banker
Total
$11-522
$43-206
$23-477
$86-213
$506-1,600
Investment
($’000)
Franchise
To $25
To $20
$13-20
To $12.5
$45
Fee ($’000)
Royalty on
6%
6%
6%
8%
12.5%
Sales
Term of
10
10
10
20
NA
Franchise in
Years
Personal Net $25
$25
$25
$30-90
$750
Worth
($’000)
Source: Entrepreneur Magazine and various franchisor websites
Dunkin
Donuts
$2551,100
$40-80
5.9%
NA
$600
Barriers to entry for real estate brokerage franchises are lower as compared with
restaurant franchises. For the three real estate firms, the royalty fee is 6% of sales but generally
lower than the royalty fees at the restaurant chains. The terms of the franchise agreement are
also shorter at real estate brokerage firms and personal net worth requirements are lower. For
example, with a $25,000 personal net worth requirement, a 10-year term for the agreement, and a
franchise fee not exceeding $25,000 (and capital expenditures which can be financed by the
franchisor) entry into the real estate brokerage industry is relatively affordable as compared to
the restaurant industry.
From Table 1, franchisor firms such as Century 21, ERA, and Coldwell Banker are not
interested in size or scale. While single-person “mom-and-pop” entities may not qualify, the
relatively low net worth and franchise fee indicate that in advance, franchisors are not selfselecting by size. Any subsequent observed distribution or sorting of firms by size is likely, to
come from structural aspects of the industry.
Franchisor-provided inputs involve economies of scale, such as common advertising to a
mass or national market. Other franchisor inputs have similar economies of scale properties,
including purchasing, training, capital access, and affinity programs as well as maintaining
4
An additional franchise is ReMax, which has costs and fees that are less homogeneous. ReMax operates with sales
associates paying fixed costs for a desk fee and a chargeback on overhead and office expenses. In exchange, sales
associates receive a larger percentage of the upside, essentially viewing the cost charged by ReMax as a call option
on their earnings.
5
databases and making technology improvements. In exchange for the set of franchisor inputs,
the franchisee pays a royalty on gross revenue. Most franchising arrangements also require an
up-front payment as an initial fee. There has been extensive research on franchising and
incentives. The incentives for boosting output motivate the franchise relationship (Martin
(1988)), while competition for space and cannibalization within the geographic trade area drives
down franchisee returns (Mazzeo (2002)).5
For a franchise model of the real estate brokerage firm, production occurs with directly
purchased inputs x such as the payments to sales associates, office space, utilities. The
technology includes shift variables a such as size, location, and age. The real estate firm has a
choice between selecting a franchise or not. A franchisor offers inputs z including referral
networks, national marketing, affiliate services, training and technology such as database
services. For holding a franchise, the franchisee pays a fixed franchise fee and a royalty on total
revenue at rate s .6 Franchises are available to all firms in the industry, and there is no
prohibition against operating without one. If the brokerage firm decides to acquire a franchise,
an indicator variable is I  1 , otherwise the decision is I  0 .
The brokerage firm’s revenue without a franchise is r ( x, a) depending on the sales
associate inputs and other purchased services and the fixed shift variables. The cost of directly
purchased sales associate commission splits, support labor, and office overhead is c( x, a) ,
increasing and convex. If the firm obtains a franchise, the revenue is (1  s )r ( x, z , a) , including
the franchisor marketing and other support z . The profit is revenue less cost
or (1  s)r ( x, z , a)  c( x, a) . The net margin y for a franchisee is the profit per dollar of net
(1  s)r ( x, z, a)  c( x, a)
revenue, or
. The firm’s net profit margin from becoming a member of a
r ( x, z , a )
franchise network is
(1)
y ( x, z, a, I  1)  (1  s) 
c ( x, a )
.
r ( x, z , a )
The franchisee hires its sales associates and other direct purchases x . The franchisorprovided inputs z and its size as well as its scale a help the franchisee to generate revenue and
profits. When the decision is not to have a franchise, the firm’s net profit margin is
5
Franchising contracts are methods of sharing and reallocating risk, as discussed in Gallini and Lutz (1992) and
Lafontaine (1992). There have been other justifications for franchising. Prendergast (2002) views a franchise as a
method for delegation when there is uncertainty in output. When there is more uncertainty, compensation is based
on output. The franchise has been viewed as a mechanism similar to sharecropping, although there are enforcement
issues should one party fail to comply (Lafontaine and Raynaud (2002)).
6
The franchisor may charge fixed costs or up-front fees. These franchise fees are a tradeoff for the royalty. In a
model where these are both fees and the franchisee’s benefit is added revenue, at maximum profit there is a negative
tradeoff between the royalty and the franchise fee. Using the notation of (1), if the franchise fee is k , differentiating
totally at maximum profit r ( x, z , a )ds  dk  0 , so
dk
  r ( x, z , a )  0 .
ds
6
(2)
y( x, z, a, I  0)  1 
c( x, a)
.
r ( x, 0, a)
Here the residential brokerage firm sets the level of franchise-provided inputs z at zero, in
exchange for retaining all the gross commission revenue. No royalty fee s is paid.
The firm chooses to hold a franchise when the profit exceeds that from being
independent, or
(3)
 y ( x, z , a, I  1)  y ( x, z, a, I  0)

c ( x, a )
c ( x, a )

.
 s  r ( x, z , a )  r ( x, 0, a)

The ratio of expenses to revenue for a franchise including all fees must be less than the expense
ratio for a non-franchised firm. Regardless of size and given the franchise terms, the firm selects
to be a franchise when the net profit is positive. The condition (3) establishes the demand for a
franchise. It may be optimal for the firm not to hold a franchise, in which case the weak
inequality in (3) is reversed. A firm of a given size, location, or scale a will choose a franchise
when it provides an enhancement, leading to the condition
(4)
y ( x, z, a, I  1) y( x, z, a, I  0)

.
a
a
The above conditions apply to the demand side for franchises. There is a corresponding
supply side from the franchisors. Franchisors, including firms across the quality spectrum of the
market from Century 21 to Coldwell Banker to Sotheby’s, provide national marketing, training,
affinity relationships, technology, and branding. It is offers inputs z in exchange for royalty s .
A given franchisor will operate within the set of firms ranked by structure a  1,..., A that
provides the highest profit return. The franchisor supplies the same inputs to all firms, and its
cost function is c ( z ) . The profit condition is
(5)
 F  s max r ( xa , z, a, I  1)  c( z)
xa
a  1,..., A .
If franchisor-provided inputs z and its size and scale a help the franchisee to generate increased
revenue, then the franchisor selects that brokerage firm which benefits the most from
franchising.
The equation (5) implies that the decision of an individual residential brokerage firm to
acquire a franchise depends on its net profit margin after expenses. This selection on which firm
acquires the franchise is
(6)
I *  I * y( x(s, z, a), z, a, I  1)  y( x,0, a, I  0) 
7
where I * is an index of preference intensity. If the argument is positive, the firm selects a
franchise. Conditional on the decision, the profit is
y ( I  1)  y ( x( s, z, a), z, a, I  1)
(7)
for franchised firms.
In the contract ( z , s ) , the franchisor delivers marketing, training, affinity relationships,
technology, and branding inputs z in exchange for a royalty s . The result is a two-stage selfselective process. In the first stage, conditional on firm fixed attributes such as size and location
a , firms select or reject the franchise contract. In the second stage, conditional on the first stage
self-selection, firms achieve their success through the profit margin.
A testable hypothesis is whether size is one of the underlying characteristics that leads to
franchise selection. The model allows for determining whether size itself leads to higher or lower
profitability, or whether indirectly through selecting a franchise. With the self-selection on size,
the model tests whether franchisor-provided inputs z are complementary with or substitutable for
purchased labor, equipment, and real estate services.
Empirical implementation requires the specification of the selection test (6) and profit (7)
with parameters and a disturbance term. The decision can be structured on whether or not the
firm decides to hold a franchise. If the firm decides not to obtain a franchise, then for a linear
specification in variables W
(8)
 I *  y ( x( s, z, a), z, a, I  1)  y ( x(0, 0, a), 0, a, I  0)  W   

if I *  0
I  0
The parameters  and disturbance  . That decision is determined by whether their profit is
higher with a franchise than without. Conditional on the decision to decline the franchise the net
profit margin is
(9)
y  y ( x( s, z, a), z, a, I  0)  X    .
The disturbance is  with variance   and covariance with the decision of   . The
covariance matrix of the disturbances for the selection and profit margin is
(10)
  
1
cov( ,  )  
.
    
The expected net return for firms not holding a franchise is
(11)
E  y | I  0  X    
1  f (W  )
F (W  )
8
1  f (W ˆ )
. If the self-selection condition
F (W ˆ )
is negative, remaining outside a franchise provides a net advantage to a firm. If the overall term
is positive, then a having a franchise provides an increase in the net profit margin. Smaller firms
with lower a have fewer economies of scale that would enable taking advantage of the
franchisor’s inputs.
The self-selection term for a franchise is ˆ
The empirical objective is to test for the benefits of a franchise. Those benefits may vary
not only by the size and structure of the firm, but also regionally. With some parts of a country
growing faster than others, less-informed new residents may be attracted to a franchise by the
assurance of uniform quality signaled by the franchised brand.
3. Data and Empirical Results
The sample is from a survey conducted by the National Association of Realtors in March
2001, of residential brokerage firms in the United States on financial performance. If more than
50% of a respondent’s business was from commercial brokerage, it was removed in order to
obtain a sample of real estate firms that focus primarily on residential real estate. The result is
1,143 eligible firms.
Table 2 presents the descriptive statistics. For the sample, 26.51% or 313 of the firms are
members of a franchise. The remaining 830 are not franchised. The firm’s profit is expressed as
its net margin, which is the difference between revenues and expenses. For franchised firms, the
revenues are net of the royalty and include the added potential revenue from the franchisorprovided inputs. The firms, both franchised and non-franchised are distributed across four
regions of the country, the Northeast, Midwest, South and West. There are at least 60 firms in
any one category by region and franchise status, providing sufficient degrees of freedom for
estimating the self-selection.
The first set of tests is (8), the decision to acquire a franchise. That equation is specified
as probit, with I *  y ( x( s, z, a), z, a, I  1)  y( x, 0, a, I  0)  W    . The variable list in W
includes firm size and technology, as demonstrated by the number of Internet websites. The
results appear in Table 3. The dependent variable is whether the firm decides not to select a
franchise.
9
Table 2. Residential Real Estate Brokerage Firm Sample Statistics
(National Association of Realtors, U.S., 2001)
Variable
Net Margin
Lnetmargin
Reloc
Age
Oneoff
Mfirm
Lfirm
West
South
Midwest
Dedicated
Weblist
Email
Numwebs
Description
Franchise = 0
Mean
Sample size
Net margin,
percentage points
Natural log, net
margin
=1 if member of a
relocation network
Length of time in
business, in years
=1 if firm has only
one office
Medium-sized firm,
11-200 salespersons
Large firm, more than
200 salespersons
=1 if West
=1 if South
=1 if Midwest
=1 if firm has
dedicated website
=1 if firm posts
listings on web
=1 if sales staff
accessible by e-mail
Number of websites
firm is found on
830
20.02
19.80
313
14.70
16.37
2.51
1.06
2.21
0.99
0.98
0.14
0.98
0.13
21.30
18.54
23.86
19.06
0.62
0.49
0.50
0.50
0.38
0.49
0.45
0.50
0.04
0.20
0.11
0.31
0.28
0.33
0.22
0.85
0.45
0.47
0.41
0.36
0.25
0.31
0.26
0.92
0.44
0.46
0.44
0.27
0.79
0.41
0.88
0.32
0.60
0.49
0.64
0.48
2.83
1.53
3.46
1.39
10
Franchise =0
Standard
deviation
Franchise =1
Mean
Franchise = 1
Standard
deviation
Table 3. Franchise Selection Probit, Regions and National
Variable
Dependent Variable
FRAN = 1
if franchise
Independent Variables
CONSTANT
MFIRM
Medium-Sized
11-200 salespersons
LFIRM > 200
Salespersons
NUMWEBS
Number of websites
listed on
Ln L
N
Chi-squared
Midwest
Coefficient
(z-statistic)
Northeast
Coefficient
(z-statistic)
South
Coefficient
(z-statistic)
West
Coefficient
(z-statistic)
National
Coefficient
(z-statistic)
-0.95
(-4.77)**
-0.17
(-0.94)
-1.48
(-5.91)**
0.035
(0.16)
-1.14
(-6.91)**
0.28
(1.82)
-1.24
(-6.54)**
0.40
(2.39)**
-1.15
(-12.00)**
0.15
(1.76)
0.30
(0.95)
0.16
(2.67)**
0.78
(1.65)
0.27
(3.80)**
0.59
(1.89)
0.11
(2.33)**
0.72
(2.39)**
0.11
(1.99)**
0.55
(3.30)**
0.15
(5.24)**
-159.10
265
9.67**
-104.70
194
20.06**
-206.80
373
16.03**
-167.90
311
16.67**
-645.08
1143
51.81**
** significant at the 5% level
The first test is for aggregation across regions. A Chow test statistic was performed on
the sample selection regression, comparing the error sum of squares of the aggregated sample
with the four regions. The F- test statistic of 35.55 exceeded the critical value at the 5% level.
The finding indicates that the residential brokerage market probit and regression models should
be conducted by region. For completeness, the national results for the United States are reported
in the last column of Table 3.
Larger firms are more likely to be organized as franchises, but the results differ around
the country. In all regions, a large firm with more than 200 salespersons is more likely to be a
member of a franchise. However, only in the West is the estimate significant at the 5% level,
although in the South the estimate is significant at an 8% level. Similar results occur for
medium-sized firms with 11-200 employees. Franchises are more likely to be tied to size in the
West, and South. Conversely, while parameter estimates are positive, size does not significantly
influence the decision to hold a franchise in the Midwest and the Northeast. Turning to the
number of websites, a measure of scale, in all cases this variable is positively associated with
being a member of a franchise.
The national effects mask the differences in regional performance. At the national level,
size at medium and large firms consistently leads to an increased probability of joining a
11
franchise, but that effect occurs because of dominance in the high-growth regions of the West
and South.
The conditional probability of not having a franchise for the non-franchised firms is
1  f (W ˆ )
. The overall effect is the product of this conditional probability and its coefficient,
F (W ˆ )
an estimate ˆ of the covariance between the selection and the return to the non-franchisees.
ˆ (1  f (W ˆ ))
So 
is the overall effect. If this effect is negative, then not having a franchise
F (W ˆ )
lowers the net profit margin for the non-franchised firms. The test for the productivity of a
ˆ (1  f (W ˆ ))
franchise is therefore 
 0 and this varies by region.
F (W ˆ )
The net margin is gross revenues less all expenses including franchise and royalty fees,
divided by gross revenues. This dependent variable is expressed in percentage points. The
estimation uses weighted least squares to correct for sample heteroskedasticity. The weights
used in this procedure are the sample weights from the survey, and they are designed to reflect
the differential probability of firm and item non-response.7 Results for the non-franchised firms
appear in Table 4.
ˆ (1  f (W ˆ ))
The total effect of the franchise on the net margin of a firm is 
at the
F (W ˆ )
sample means reported in Table 4. In the Northeast, the franchise effect is not significant.
Franchise and non-franchise firms are similar in performance. In the Midwest, South and West,
not having a franchise reduces the profitability of the non-franchised firms. Correspondingly,
franchised firms have higher profitability in these regions.8
The results suggest that the benefits of holding a franchise vary regionally. In fastergrowing markets a larger proportion of homebuyers are moving to the area. Those buyers are
likely to be less familiar with the local market players, and must rely on national names such as
those provided by a franchise. Conversely, in the slow-growing markets of the Northeast, the
market is dominated by existing residents. Those residents are more familiar with local firms,
reducing or eliminating the benefits of holding a franchise.
7
Historically, the National Association of Realtors (NAR) surveys of real estate brokerages had suffered from a
biased response where smaller brokerages responded at a rate significantly higher than that of larger brokerages. To
correct for this in their 2001 survey, NAR stratified the brokerage industry’s firms into four different groups. NAR
then “over sampled” firms with 11 to 200 agents and those with more than 200 agents relative to firms with just one
agent and those with 2 to 10 agents. These “larger” firms received the survey twice to induce a greater response. A
weight was developed to control for both the over sampling of firms with 11 or more agents and for the different
response rate for each of the four stratified groups.
8
To determine the true benefits of holding a franchise, the various costs of holding a franchise must be added to
firm expenses. Net profit margin contains firm expenses including franchise costs.
12
Table 4. Net Margin, Non-Franchised Firms
(parameter estimates, t-statistics in parentheses, sample means below)
Variable
Midwest
Northeast
South
West
Dependent
MARGIN: Net margin
Independent
CONSTANT
RELOCO = 1
if member of
relocation network
AGE, Length of time
in business
ONEOFF = 1 if one
office
1  f (W ˆ )
F (W ˆ )
self-selection
R2
N
51.69
(1.38)
54.95
(4.06)**
58.34
(1.44)
41.62
(1.50)
17.99
(0.67)
0.98
-0.45
(-1.92)
17.42
23.57
(2.49)**
0.81
92.84
( 2.05)**
-0.47
-45.99
(-3.88)**
0.98
0.32
(3.06)**
22.09
18.51
(3.31)**
0.87
-3.35
(-0.28)
-0.36
-2.45
(-0.07)
1.00
-0.30
(-1.67)
14.95
4.19
(0.83)
0.68
67.67
(2.25)**
-0.32
53.27
(2.92)**
0.94
-0.61
(-1.49)
10.54
-4.56
(-0.40)
0.81
121.33
(2.11)**
-0.28
0.10
270
0.28
232
0.35
183
0.20
140
** significant at the 5% level
The impact from other variables supports the regional differences. Firms in the
Northeast and Midwest with one office have higher profit margins, without being franchises.
From the probit estimates, small firms have no disadvantages; and from the inverse Mills ratio,
franchises themselves have limited positive effect. The results by age are supportive.
4. Summary and Concluding Remarks
This data set allows the direct testing of hypotheses about franchising and firm
profitability and size along with regional effects. First, the data allows for self-selection testing
as to whether or not a residential brokerage firm adopts a franchise. Second, the sample allows
for testing about whether the existing royalty structure provides for effective sorting between
franchise and non-franchise brokerage firms.
Locality affects the performance of real estate firms. Firms in slower-growing but dense
areas such as the Northeast are able to substitute for a franchise and remain independent because
the clientele is primarily local as opposed to being out of town. Firms in the Northeast and
13
Midwest substitute a personal touch when the buyers are local, leading to a return for one-office
entities. In the West, South and Midwest a franchise carries a positive return. That return is
particularly beneficial for larger firms that are able to take advantage of national and regional
advertising.
14
References
Anderson, Randy and Robert Fok (1998) “The Efficiency Of Franchising In The Residential
Real Estate Brokerage Market,” Journal of Consumer Marketing, 15, 386-398.
Anderson, Randy, Danielle Lewis, and Leonard Zumpano (2000) “Residential Real Estate
Brokerage Efficiency from a Cost and Profit Perspective,” Journal of Real Estate Finance and
Economics, 20, 295-310.
Frew, J. R. and G. Donald Jud (1986) “The Value of a Real Estate Franchise,” Journal of the
American Real Estate & Urban Economics Association, 14:2, 374-383.
Jud, G. Donald, Ronald C. Rogers, and Glenn E. Crellin (1994) “Franchising and Real Estate
Brokerage,” Journal of Real Estate Finance and Economics, 8, 87-93.
Gallini, Nancy and Nancy Lutz (1992) “Dual Distribution and Royalty Fees in Franchising,”
Journal of Law, Economics and Organization, 8, 471-501.
Lafontaine, Francine (1992) “Agency Theory and Franchising: Some Empirical Results,” Rand
Journal of Economics, 23, 263-283.
Lafontaine, Francine and Emmanuel Raynaud (2002) “The Role of Residual Claims and SelfEnforcement in Franchise Contracting,” National Bureau of Economic Research, NBER Working
Papers: 8868.
Lewis, Danielle and Randy Anderson (1999) “Residential Real Estate Brokerage Efficiency and
the Implications of Franchising: A Bayesian Approach,” Real Estate Economics, 27, 543-60.
J. Knight, C.F. Sirmans, and G. Turnbull (1994) “List Price Signaling and Buyer Behavior in the
Housing Market,” Journal of the American Real Estate and Urban Economics Association, 9,
177-192.
Martin, Robert (1988) “Franchising and Risk Management,” American Economic Review, 78,
954-968.
Mazzeo, M.J. (2002) “Competitive Outcomes in Product-Differentiated Oligopoly,” Review of
Economics and Statistics, 84, 716-728.
Prendergast, C. (2002) “The Tenuous Trade-Off Between Risk and Incentives,” Journal of
Political Economy, 110, 1071-1102.
Richins, M. L., W. C. Black, and C.F. Sirmans (1987) “Strategic Orientation and Marketing
Strategy: An Analysis of Residential Real Estate Brokerage Firms,” Journal of Real Estate
Research, 2, 41-54.
15
Sirmans, C. F. and G. Turnbull (1997) “Brokerage Pricing under Competition,” Journal of
Urban Economics, 41, 102-117.
Zumpano, L.V., H. W. Elder, and Randy I. Anderson (2000) “The Residential Real Estate
Brokerage Industry: An Overview of Past Performance and Future Prospects,” Journal of Real
Estate Research, 6, 237-250.
16