You are on page 1of 16

Journal of Business Venturing 18 (2003) 403 – 418

Franchise turnover and failure


New research and perspectives
Stevan R. Holmberga,*, Kathryn Boe Morganb,1
a
Kogod School of Business, American University, 4400 Massachusetts Avenue, Northwest,
20016 Washington, DC, USA
b
International Franchise Association Educational Foundation, Washington, DC, USA

Abstract

Entrepreneurship literature has long noted likely significant new venture failure rates. This paper’s
new franchise failure concept reconciles many prior, seemingly inconsistent study results based largely
on franchisor’s surveys. Data include more than 800 franchise systems and 250,000 franchise outlets
over four consecutive years and are based on state-required Uniform Franchise Offering Circular
(UFOC) registration data. Overall franchisee turnover rates are significant and appear to have
increased. The result is a 1997 median franchisee turnover with a transfer rate of 10.49% (8.86%
in 1994). Franchisee failure and turnover are important areas that merit further study and analysis.
D 2002 Elsevier Science Inc. All rights reserved.

1. Executive summary

Entrepreneurship literature has long noted significant new venture failure rates. Further-
more, few franchise topics have generated more interest over the years than franchise failure
rates. Past franchisee failure literature, based largely on franchisor surveys, used very
divergent concepts and indicated widely ranging failure estimates. On the low end, many
studies used the old Department of Commerce numbers in the 4–5% range. On the higher
end, some estimates were in the 25–35% range (Hartnett, 1992).

* Corresponding author. Tel.: +1-202-885-1921 (office), +1-301-340-2567 (home); fax: +1-301-340-2714.


E-mail addresses: s.r.holmberg@worldnet.att.net (S.R. Holmberg), TnKMorgan@aol.com (K.B. Morgan).
1
Tel.: + 1-703-312-0001.

0883-9026/02/$ – see front matter D 2002 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0883-9026(02)00085-X
404 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

This paper presents both a broader concept of franchisee failure and new longitudinal data.
The new concept is significant because it reconciles many of the prior, seemingly
inconsistent study results. These new data are significant because of their scope (more than
800 franchise systems and 250,000 franchise outlets), the duration (four consecutive years),
and the validity of being provided under oath (required by state registration statutes). The
new concept and data combination facilitates new explorations of research questions sur-
rounding franchisee failure.
The facet of failure studied was ‘‘turnover,’’ the term specified in the Uniform Franchise
Offering Circular (UFOC) disclosure documents. Whether viewed as means or medians, and
with or without transfers, overall franchisee turnover rates appear to have increased
noticeably over the 4-year study period. The result is a 1997 median turnover with transfer
rate of 10.49% (8.86% in 1994).
A broader stakeholder view of franchisee failure that goes beyond the franchisor and
franchisee reveals many additional actual and/or opportunity costs/losses. Suppliers, financial
institutions, customers, investors, and others all suffer as the circle of real and/or opportunity
cost/loss widens because of franchisee turnover. Franchisee turnover itself, however, is only
one step on the failure continuum. If recognized and managed at earlier steps on the continuum,
systematic risk mitigation strategies might be developed, losses minimized, and turnover
possibly avoided. Hence, these strategies may avoid real and/or opportunity costs if franchi-
sor–franchisee relationships, business issues, and failure risk are managed appropriately.
This paper’s increased UFOC data validity and franchisee turnover insights represent new
and important information for existing and prospective franchisee ventures. Franchisee failure
and turnover are important areas that merit further study and analysis.

2. Divergent franchisee ‘‘failure’’ concepts/definitions and research methodologies in


perspective

Entrepreneurial new venture creation and entrepreneurial activity, in general, has long been
seen as a significant contributor to economic growth and development and individual
enterprise advancement (Falbe et al., 1998). Franchise systems constitute one form of
entrepreneurial ventures that have contributed significantly to economic development results
in the United States and other countries. Franchising has also been seen as an alternative to
individual self-employment decisions to start an independent small business (Kaufmann,
1999; Williams, 1998). Central to these new venture alternative decisions is the likelihood of
success or failure. This study focuses on the critical importance of failure prospects and
failure management strategies on entrepreneurial franchise decisions.
Prior research studies (Cross, 1994, 1998; Castrogiovanni et al., 1993) and the
International Franchise Association have pointed to the need for more reliable franchise
failure studies. Franchise failure research is strongly influenced by the failure concept
underlying the study’s research perspectives, data collection, analysis, and resulting research
conclusions (Bates, 1995b; Stanworth and Dandridge, 1994). Diverse franchisee ‘‘success’’
or ‘‘failure’’ realities and perceptions exist among researchers and franchise system stake-
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 405

holders, such as: franchisees, franchisors, financial institutions, general public and/or
consumers, government agencies, legal system components, and others (Falbe and Welsh,
1998; Hoy, 1994).
Comparing past franchise failure research is difficult because failure concepts and
definitions are not consistent. Several recent studies have suggested that the need is not to
define failure definitively, but rather to view the process comprehensively (Holmberg and
Morgan, 1996, 2000; Cross, 1994, 1998). An eight-step franchise failure model has been
proposed, suggesting that the first indicators of failure can often be seen very early. The
following progression is described: (1) franchisee core competency misfit; (2) franchisee–
franchisor dissatisfaction; (3) franchisee discontent; (4) royalty delinquency, etc.; (5) com-
plaints to FTC, et al.; (6) turnover/termination; (7) defaults/other losses to creditors; and
(8) closure (Holmberg and Morgan, 1996, 2000). This eight-stage continuum provides a
context within which a number of earlier studies make sense and even complement one another.
This paper provides critical new data on Step 6 of the failure continuum: turnover/
termination. Turnover is a broader concept than closure and ‘‘a more encompassing term than
failure’’ (Walker and Cross, 1988). Turnover is the term used to denote a change in ownership
of a franchise outlet. The change could occur because the unit was cancelled or terminated by
the franchisor, reacquired by the franchisor, not renewed by the franchisor, transferred to
another entity, or not doing business for other reasons. The turnover rate more appropriately
‘‘indicates the overall likelihood of remaining in business’’ (Cross, 1998).
Turnover includes, however, the category of transfers—a category particularly susceptible
to double counting. For example, if a unit is terminated and transferred to a new franchisee,
rather than being counted as one unit ‘‘turning over,’’ it could be counted separately in each
category of ‘‘terminations’’ and ‘‘transfers.’’ Instead of setting the ‘‘transfer’’ category aside,
one study suggested that transfers should be included ‘‘in order to yield a more complete
picture’’ (Stanworth et al., 1998).
A multistep failure perspective is important to evaluate this data. Turnover is only Step 6 in
the failure continuum. It is legally significant and it is externally verifiable, but it is not the
whole picture. ‘‘Not every franchisee failure will hit every point on the continuum, and
realities are easier to measure at some points on the continuum than others. Researchers will
certainly show that some of these points may have much more predictive validity than others
for a given individual and more representational validity for franchising in general.
Nevertheless, franchise failure should be considered by looking at all of its various
manifestations and the studies which have focused on them individually’’ (Holmberg and
Morgan, 1996).

3. New franchisee failure turnover rate data

Turnover information is now available summarizing four consecutive years of franchisor


registration documents. State laws in the 12 registration states require that these documents be
prepared according to a specific format (UFOC) and that they be filed with the state
governments. These documents contain franchisor and franchise business, financial, contract,
406 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

and turnover information. The information is provided under penalty of law. The research
commissioned by the International Franchise Association Educational Foundation (IFA
Foundation, 2000, 1999, 1998 and 1996) was extensive, including the collection and analysis
of UFOCs for every registered company. For 1997, that meant 834 franchisors with over
250,000 franchised outlets. Table 1 presents the number of franchisor companies studied in
each of the 4 years.
Franchise turnover data in UFOC Item 20 presents 3 years of data for five turnover event
categories that might represent business failures:

. Transfers—A franchisee unit is transferred from the existing owner to a new franchisee
owner.
. Cancellations—The franchisee unit is cancelled for failure to comply with quality control
reasons or for other reasons.
. Nonrenewals—The franchisee unit contract is not renewed and the unit may be closed, sold
to a new franchisee, or converted to a company-owned unit.
. Reacquisitions—The franchisee unit is purchased by the franchisor and becomes a
company-owned unit.
. Other—Units which are no longer doing business for other reasons.

Despite the strong validity of the new turnover data, five caveats must be given. First, it is
important to note that turnover does not necessarily mean failure. Unfortunately, in reviewing
Item 20 data, there is no way to distinguish units that have ‘‘turned over’’ for positive reasons
(retirement or opportunity to sell at a profit) from those that have ‘‘turned over’’ for negative
reasons (business failure). The only way to determine which are failures would be to contact
affected owners for their individual stories. There is no doubt, however, that many transfers
do represent failure from the original franchisee’s perspective. In addition, Bates (1995b)
found that young retail franchisee transfers, where a new franchisee purchases the unit from
an ongoing franchisee, is riskier than a franchisee starting an entirely new unit since ‘‘ongoing
franchise units that are available for purchase appear to be disproportionately prone to going
out of business’’.
The second warning is that a potential double counting problem exists since the five event
categories in Item 20 are not necessarily mutually exclusive. For example, a franchised unit
may have been canceled and reacquired by the franchisor in the same time period and
included as two separate events in Item 20. Although UFOC Guidelines have provisions to

Table 1
Number of franchise systems analyzed
Total franchisors
UFOCs collected UFOCs used
1994 472 447
1995 1174 753
1996 1178 841
1997 1226 834
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 407

account for several events with the same unit, they are not followed often. This oversight
leads to double counting and a potential upward aggregate turnover rate bias. The transfer
category (individual units transferred from one franchisee to another existing franchisee, to a
new franchisee, or converted to a company-owned unit) is most susceptible to double
counting. Without investigating circumstances behind each outlet’s transfer, there is no way
of knowing the extent of the double counting problem. Therefore, to deal with this problem,
this study presents turnover rates both with and without transfers to provide, perhaps, ‘‘worst-
case’’ and ‘‘best-case’’ perspectives.
Third, only median or mean data are available. In 1994, median data were collected as a
better central tendency measure where data sets are skewed. In years 1995 to 1997, both
median and mean data were collected. Insights based on these data are preliminary due to the
short longitudinal period, availability of only summary median and mean data, and individual
firm data confidentiality.
Fourth, although the new longitudinal data provide the most thorough analysis of
franchisee turnover, the study does not include all US franchisors. Franchisors that have
no units in registration states and are not soliciting sales there would not be included. Since
registration states include major population centers (e.g., New York, California, Illinois), most
major franchisors would have at least some presence in those states, however. Existing data
from other studies suggest that a few franchisors represent a significant proportion of
business-format franchisee establishments (Lafontaine and Shaw, 1998). Nevertheless, some
smaller or newer franchisors may not have expanded their franchised units into the
registration states and, excluding these franchisors, which tend to have disproportionately
higher failure or turnover rates, may understate franchisee failure or turnover rates.
Finally, the study would not have included franchise systems that totally disappeared and,
therefore, did not register in the following year. Some researchers believe that 75% of
franchise systems fail within the first 10 years of existence (Shane, 1996; Shane and Spell,
1997). This is a complex issue, however, because the closing down of the franchisor does
not necessarily mean the closure of the individual franchisee units. For purposes of this
paper, however, suffice it to say that there is no current governmental method of tracking
such units. It is likely, however, that franchisee failure or turnover rates may be understated
due to these factors.

4. Findings and new venture strategic perspectives

Past franchisee-focused failure studies have largely used franchisor survey data and
single-year cross-sectional data. Notable exceptions are a multiyear restaurant franchisee
study (English and Willems, 1991–1999) and two longitudinal franchisor studies (Shane
and Spell, 1997; Lafontaine, 1994). New venture strategic perspectives may be enhanced
by this study’s use of actual UFOC franchisee failure longitudinal data and analysis by
industry category, number of franchised units, number of franchisor company-owned units,
rate of franchisee unit expansion, length of time in franchising, and average non-real estate
initial investment.
408 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

4.1. Total franchisee turnover rates

Median and mean data for total franchisee turnover rates are presented with transfers
included (turnover with transfers) and without transfers (turnover-only). Franchisee turnover
with transfers represents a broader stakeholder view. Including transfers recognizes that
ownership changes did occur for the original franchisee.
Overall franchisee turnover rates, presented in Table 2, indicate a high level of franchisee
turnover on an annualized basis (all data are median data unless otherwise noted). The 1997
franchisee turnover with transfer rate was 10.49%, indicating that 10.49% of all franchisees
that were in existence in 1996 turned over. Turnover-only rates (5.54%) are lower due to the
exclusion of transfers of the original franchisee’s unit to a new franchisee, to another existing
franchisee, or to a company-owned unit. Transfers must be included for turnover rates to be
fully indicative of a franchisee’s turnover risk. Interestingly, the 10.49% US franchisee
turnover with transfer rate is comparable to UK franchisor self-reported survey data, ‘‘which
suggests franchisee failure and turnover combined amount to around 10% per annum’’
(Stanworth et al., 1997).
If one looks at medians, turnover increased consistently from 1994 to 1995 to 1996, and
declined slightly in 1997. Median turnover with transfers increased 18.4% from 8.86% in
1994 to 10.49% in 1997. Rates excluding transfers rose 19.1% from 4.65% in 1994 to 5.54%
in 1997. The peak turnover rates both with and without transfers occurred in 1996. The
increasing turnover rates during this 4-year period are somewhat surprising since the US
economy was expanding significantly during this entire time.
In the following subcategory analysis, median data are used as a more accurate central
tendency representation due to the significant impact that one or two large franchisors can
have on mean data and the relatively small number of franchisors in some categories.

4.2. Franchisee turnover rates by industry category

Industry category franchisee turnover rates vary both in absolute magnitude and in
respective trends over time, as shown in Table 3. Using 1997 data, the absolute magnitude
of turnover with transfer rates ranged from a high of 14.7% for travel franchisees to a low of
5.9% for restaurants. The turnover rate distribution over 18 industry categories divides with
the top half having a 10.9% turnover rate or above. Using turnover-only data, franchisee
turnover rates in 1997 range from a high of 10.6% in the travel industry to a low of 3.6% in

Table 2
Overall franchisee turnover rates for all industry categories
Turnover rate with transfers Turnover rate without transfers
Median (%) Mean (%) Median (%) Mean (%)
1994 8.86 – 4.65 –
1995 9.83 14.32 5.19 10.0
1996 10.8 14.7 6.3 10.1
1997 10.49 15.99 5.54 11.33
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 409

Table 3
Franchisee turnover rates by industry category
Total all categories Turnover rates (medians) by industry category
Median with transfers Median without transfers
1994 1995 1996 1997 1994 1995 1996 1997
8.9% 9.8% 10.8% 10.49% 4.7% 5.2% 6.3% 5.54%
Franchisee turnover by industry category
Automotive 9.43 8.70 9.5 11.2 4.77 5.82 5.2 6.8
Baked goods 8.22 10.26 10.3 12.2 2.55 2.53 6.6 4.6
Building and construction 5.41 8.81 9.7 10.2 3.03 4.67 8.7 6.3
Business services 10.16 9.85 13.1 10.9 5.51 3.85 7.7 7.8
Child related 22.31 17.09 6.1 11 14.79 7.23 3.1 4.9
Education related 10.18 9.09 10.1 13.8 6.22 5.34 8.2 8.8
Fast food 7.53 9.54 11.7 10.6 3.38 5.14 5.6 3.9
Lodging 7.52 12.35 10.8 10 2.54 6.74 5.4 3.3
Maintenance services 8.00 12.50 8.3 11.1 3.44 6.51 6.9 7.1
Personnel services 8.31 5.26 9.1 11 3.25 2.63 4.9 6.8
Printing NA 8.10 9.6 9 NA 5.71 5.9 6.2
Real estate 10.81 14.06 12.0 10 7.16 7.75 7.2 7.9
Restaurants 10.44 8.06 9.5 5.9 4.96 5.65 6.0 3.6
Retail food 11.54 9.14 9.1 10.3 8.00 4.57 5.2 6.0
Retail 7.18 10.63 9.4 9.6 4.22 4.70 5.6 4.9
Service businesses 10.53 11.52 14.4 12.8 6.25 4.35 7.3 6.7
Sports and recreation NA 6.76 11.8 8.9 NA 3.57 7.9 6.6
Travel NA 16.67 10.8 14.7 NA 13.18 6.3 10.6

the restaurant industry. Hence, whether one examines turnover with transfer or turnover-only
rates, there are significant industry variations. Retail sales, fast food restaurants, and customer
service franchisees were found to have low average closures while beauty and health,
automotive, and hotels and motels had the highest failure rates (Justis et al., 1992).
An annual 10.5% franchisee turnover rate suggests that over time, franchisees face
substantial failure risks. These data are similar to findings by Bates (1995a, 1995b) using
the Census Bureau’s Characteristics of Business Owners (CBO) database of young retailing
franchise firms starting in 1986 and 1987. Bates found that by 1991, ‘‘only 52.4% of these
firms were still operating with the original owner of record in 1987.’’ Thus, over the 4-year
period, an average annual rate of 13.1% of retail franchisees changed hands, were reacquired
by the franchisor, or went out of business.
Franchisee turnover rates over time suggest an increasing pattern from 1994 to 1997 for
11 of 16 industry categories that reported 1994 data, while five decreased over the same
period. The turnover with transfer rates for a total of 11 of the 18 industry categories also
increased from 1994 to 1996, while 10 of the 18 industries showed an increasing turnover rate
from 1996 to 1997.
Industry category turnover rates are impacted by franchisor, franchisee, and competitive
influences. Various other factors influence industry category turnover rates: ‘‘fluctuating
business cycles, industry life cycles, and other environmental factors can affect a unit’s
410 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

profitability and its chance of survival or failure. . .’’ (Castrogiovanni, 1991; Castrogiovanni
et al., 1993). Multiunit franchisees and master franchisees, who own, operate, or control more
than one franchisee unit, are found in many industries and are prevalent in some industry
segments, such as fast food. (Stanworth and Curran, 1999). Kaufmann and Dant (1996) found
that single-unit, owner-operated franchisees may be rare: 88% of the franchisors surveyed in
1994 had multiunit franchisees. Other studies have confirmed the importance of multiunit
franchisees by documenting that a majority of new fast food franchise units were opened by
existing franchisees (Kaufmann, 1995; Kaufmann and Lafontaine, 1994).
Multiunit franchisees may result from incremental or sequential expansion or from a
master franchise agreement that grants area development rights. A multiunit franchisee
strategy may lead to lower franchisee unit failure or turnover rates compared to a single-unit
franchisee-only strategy. Multiunit franchising strategies yield faster growth rates, enhance
capital acquisition, utilize more experienced management, enhance local market understand-
ing, increase individual unit monitoring frequency and effectiveness, leverage franchisor and
multiunit franchisee distinctive competencies, and reduce overall franchisee unit risk
(Kaufmann and Dant, 1996, 1998). Multiunit franchisee realities may lower franchisee unit
turnover in numerous areas (by industry, by growth rate, by number of franchisee units, time
in franchising, etc.). However, these positive multiunit impacts may be influenced by other
variables such as franchisor newness, experience, size, industry, and market niche. Whether
positive or negative, the impact of multiunit ownership is not tracked or measured in UFOCs.
Therefore, the turnover rates in this paper do not distinguish between units owned by single-
unit owners or multiunit owners.

4.3. Franchisee turnover rates by number of franchisee units

Franchisors move along the learning curve as they successfully develop more franchise
units. One would expect more standardized operations, quality maintained at desired levels,
purchasing power increases driving down franchisor and franchisee costs, and advantages of
scale from geographic concentration resulting in improved brand name recognition and
advertising effectiveness (Shane and Spell, 1997; Castrogiovanni et al., 1993). As volume
grows, franchisee and franchisor revenues grow, and profitability is increased. If some or all
of these factors materialize, one would expect that total franchisee turnover rates would
decline the more franchised units are present in a given system and the longer a franchisor
had been franchising. The new study results are surprising in this regard.
Table 4 presents franchisee turnover rate data by number of franchised units. Examining
the most recent data, franchisee turnover rates do not start to drop until a franchisor has
reached 501-or-more franchise units. In the three prior years, franchisee turnover rates show a
stable pattern or slight decrease as we move from 51–100 units to 101–500 units, but there is
no sharp drop. Turnover-only data show very similar overall patterns to turnover with transfer
data. However, a sharper proportional decline occurs as a franchisor moves from the 101-to-
500 category to the 501-and-over category. The results seem contrary to expectations that
turnover rates would decline as franchised unit numbers increased. One explanation found in
a franchise failure study using 140 UFOCs suggests the possibility that ‘‘. . . most learning
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 411

Table 4
Total franchisee turnover rates by number of franchised units
Total all categories Turnover rates (medians) by number of franchised units (%)
Zero 1 – 10 11 – 50 51 – 100 101 – 500 501 and over
Median with transfers
1994 0.0 9.52 11.43 11.08 10.08
1995 0.0 8.70 11.11 11.11 9.85
1996 0.0 0.0 10.5 11.5 11.6 10.4
1997 0.0 0.0 9.1 10.4 12.1 11.0

Median without transfers


1994 0.0 4.55 6.25 6.49 5.04
1995 0.0 3.85 6.56 6.54 4.79
1996 0.0 0.0 6.7 7.0 6.9 5.7
1997 0.0 0.0 5.0 6.8 7.3 5.2

benefits accrue early, possibly during the first 3–5 years of the franchisor’s existence’’ and
after that period, failure rates may rise due to unit obsolescence (Castrogiovanni et al., 1993).
They found no significant effect of system size on franchisee failure rates. From a franchisor’s
perspective, UK franchisor failure research shows that franchisee unit numbers need to move
to the 21–50 outlet range for the franchisor to break even with franchising and management
franchisee support network costs (Stanworth et al., 2000).

4.4. Franchisee turnover rates by number of company-owned units

Franchisors can also gain experience through opening and running company-owned units.
Once the company starts to franchise, company-owned stores provide valuable experience
that should increase franchisee success. Moreover, company-owned stores provide new
product and service testing sites, offer new franchisees training locations, provide market
place monitoring vehicles, and increase the franchisor’s general experience base that can
transfer to franchisees (Fulop and Forward, 1997). Mitenko’s (1991) franchise failure study
found, ‘‘If the franchisor has no company-owned units, this may raise a number of questions
as to the business acumen of the franchisor or their management. Also, how and where are
new products test-marketed?’’
While UFOC turnover with transfer data suggest some slight, but variable, reduction in
franchisee turnover rates once a franchisor has 11 or more company-owned units, there are no
consistent trends as company-owned unit numbers increase. Turnover-only data produce an
even more inconclusive picture with franchisee turnover rates varying across six company-
owned store size categories. Three of the four years of data suggest that those with no
company-owned units had the highest franchisee turnover rates.
From 1994 to 1997, franchisee turnover with transfer and turnover-only rates have
increased, with only one exception, as shown in Table 5. Relatively small changes in
turnover with transfer data occur between 1994 and 1997 in all company-owned unit size
categories below-50 units. The largest 4-year period changes occur in 51-and-more company-
412 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

Table 5
Total franchisee turnover rates by number of company-owned units
Total all categories Turnover rates (medians) by number of company-owned units (%)
Zero 1 – 10 11 – 50 51 – 100 101 – 500 501 and over
Median with transfers
1994 10.46 10.38 9.52 6.79 8.05 6.85
1995 11.03 10.62 9.09 9.14 7.60 7.79
1996 11.1 11.0 10.9 8.6 11.0 7.8
1997 11.0 11.0 9.0 10.0 11.0 9.0

Median without transfers


1994 5.91 4.55 5.43 4.84 3.88 3.65
1995 6.27 5.38 4.69 4.88 4.37 4.30
1996 7.2 5.9 5.9 5.4 7.1 3.9
1997 6.0 6.0 5.0 5.0 7.0 5.0

owned unit size categories with the 51-to-100 size category rising 47.3% from 6.79% to
10.0%; the 101-to-500 category rising 36.6% from 8.05% to 11.0%; and the 501-and-over
category rising 31.4% from 6.85% to 9.0%.

4.5. Franchisee turnover rates by percentage growth in franchised units

In addition to examining franchisee turnover rates by the number of franchise units and
company-owned units, UFOC data also facilitate a franchisee turnover rate analysis by
percentage growth in number of franchisee units. The number of franchise units and the rate
of franchise unit expansion are central to building brand name consumer awareness (Shane
and Spell, 1997), building per franchisee unit revenues, and achieving per unit and overall
franchise system advertising and operations economies of scale. Whatever the franchisee unit
net growth rate, systematic and effective franchisee screening is critical (Jambulingam and
Nevin, 1999). However, at some rapid growth pace, the franchisor theoretically has less time
to properly screen new franchisees and evaluate their core competencies relative to franchise
concept and demands, not to mention the declining ability to support existing franchisees
(Cross, 1994). This adverse selection potential may increase franchisor and franchisee risk
(Prescott and Visscher, 1980). Conversely, where new franchisee units are launched
sequentially by multiunit franchisees or launched by master or area franchisees, adverse
selection difficulties are minimized because existing multiunit franchisees have prior
franchisee experience and success (Dant and Nasr, 1998) and these systems grow at faster
rates (Kaufmann and Dant, 1996).
Once individual franchise units exist, whether operated by individual franchisees or by
multiunit franchisee owners, two additional franchisor strategies may impact franchisee
turnover. First, typical franchisee contracts provide for franchisees paying franchisors a
royalty based on gross sales. Franchisors, due to the nature of the franchise contract, have
incentives to push franchisees to implement sales-gain strategies that may result in lower
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 413

levels of franchisee profits (Phan et al., 1996). Lack of goal congruence occurs because
franchisor goals are linked to sales growth while franchisee goals are concerned with both
their sales and profits (Dant and Nasr, 1998). If lower franchisee profit levels are sustained
over time, franchisees may seek to abandon that particular franchise unit since it no longer
provides the anticipated return on investment, hence, increasing franchisee turnover.
Second, strong franchisor sales growth motivations may also materialize in the form of
placing additional franchise units into existing markets resulting in franchisee encroachment.
As more and more new units siphon sales and lure away customers from existing franchisees
(Vincent, 1998), they potentially increase franchise unit turnover in an increasingly saturated
market. Also, franchisors have multiple encroachment paths ‘‘. . .in addition to worrying
about other locations edging closer and closer geographically, franchisees have to deal with
them creeping through the Internet, mail order, kiosks, airports, gas station mini-marts and
grocery stores’’ (Chun, 1996).
The data suggest that franchisee turnover with transfer and turnover-only rates are
materially higher for systems with negative growth than those with some positive growth,
as shown in Table 6. When turnover with transfer data are examined in relation to system
growth, turnover does not seem to decline until a growth rate of over 50% is reached. In 1997,
the turnover with transfer rate for the three growth categories from 0-to-50% franchise unit
growth is 9.0% compared to a 6.0% turnover rate for the over-50% franchise unit growth
category. High growth rates in franchise systems may, in some cases, represent relatively
small numbers of franchise units where a high percent growth rate is easy to achieve.
Viewed over time, turnover-only rates have increased for systems with a negative percentage
growth in franchise units and decreased in varying amounts for systems with some positive
increase in the percentage growth rate in franchise units. Franchisee turnover-only rates
increased from 1994 to 1997 for negative growth and 0–10% positive growth in franchise
unit categories, while decreasing in the three highest positive growth percent rate categories.

Table 6
Total franchisee turnover rates by percent growth in franchised units
Total all categories Turnover rates (medians) by percent growth in franchised units (%)
Less than 0% 0 – 10% 11 – 25% 26 – 50% Over 50%
Median with transfers
1994 11.72 9.46 9.52 9.09 10.40
1995 11.16 9.68 9.09 10.30 0.98
1996 12.9 10.3 9.1 10.2 7.2
1997 14.0 9.0 9.0 9.0 6.0

Median without transfers


1994 7.39 4.57 5.60 3.65 6.63
1995 7.24 4.90 4.31 6.25 0.00
1996 9.1 5.5 4.7 6.3 0.0
1997 11.0 5.0 4.0 2.0 0.0
414 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

4.6. Franchisee turnover rates by length of time in franchising

Franchise system age has been cited as one of several factors impacting franchisee failure
rates (Stanworth and Curran, 1999). Length of time in business prior to starting franchising
has been shown to be a factor in franchisor failure research (Lafontaine and Shaw, 1998). One
would expect that the longer franchisors have been franchising, the more benefits would
accrue to both franchisor and franchisee, including a reduction in franchisee failures or
turnover rates. However, data shown in Table 7 are mixed and do not seem to support this
proposition. Even comparing the 1-year-or-less experience in franchising category to all other
longer experiences in franchising categories, the picture is not conclusive. Turnover with
transfer rates for 1-year-or-less in 1996 are higher, 14.7%, but in 1997, the 7.0% turnover rate
is actually lower than four of the five more-time-in-franchising-experience categories.
Over time, using turnover with transfer data, three of the six length of time in franchising
categories show increases from 1994 to 1997, while the other half show decreases. Here
again, the data do not present any useful trends that can be differentiated longitudinally over
time. It would appear that both number-of-years-in-franchising data within a given year as
well as the number-of-years-in-franchising category data over time do not produce any
patterns/trends and/or are overwhelmed by industry and/or other effects.

4.7. Franchisee turnover rates by non-real estate initial investment

The magnitude of non-real estate initial investment potentially impacts franchisee turnover
rates. The larger the franchisee upfront investment (franchisee fee and initial investment), the
more ‘‘franchisees would be inclined to exert maximal effort since they would have larger
investments at stake’’ (Castrogiovanni et al., 1995). Furthermore, the larger the initial
investment, the less willing franchisees would be to walk away from their investment,
lowering turnover. In short, as investment goes up, turnover should go down. Industry factors

Table 7
Total franchisee turnover rates by length of time in franchising
Total all categories Turnover rates (medians, %) by length of time in franchising (years)
1 or less 2–3 4–5 6–8 9 – 11 12 or more
Median with transfers
1994 0.00 2.47 9.20 10.91 11.43 8.86
1995 0.00 12.50 11.41 10.26 13.03 9.33
1996 14.7 8.5 10.5 12.2 9.5 10.5
1997 7.0 0.0 9.0 12.0 11.0 11.0

Median without transfers


1994 0.00 0.00 4.08 6.61 6.12 4.72
1995 0.00 5.34 7.14 6.52 5.17 5.14
1996 11.2 4.0 5.5 7.4 5.7 6.0
1997 7.0 0.0 2.0 6.0 6.0 6.0
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 415

may affect this supposition, since some industry categories may require substantially larger
non-real estate initial investments than other industry categories.
Table 8 presents turnover data in relation to non-real estate initial investment. Viewed
across investment categories, a consistent turnover pattern does not appear. Depending on the
year analyzed, the lowest non-real estate initial investment category, US$25,000-or-less, has
higher franchisee turnover rates than middle initial investment categories as franchisees move
from the US$75,000-to-US$99,000 category (12.0%) to the US$100,000-to-US$249,000
category (11.1%) to the US$250,000-to-US$499,000 category (8.0%). However, the three
largest investment categories have a mixed pattern with franchisee turnover rates actually
rising in the US$500,000-to-US$749,000 category, falling in the US$750,000-to-
US$999,999 category and then rising in the largest category, over US$1 million.
Consistent initial non-real estate investment category data exist for 1995, 1996, and 1997.
The two largest non-real estate initial investment categories (over US$750,000) have lower
turnover rates than the three smallest investment categories under US$74,999, especially the
two smallest categories with under-US$49,999 in non-real estate initial investment. This
pattern is consistent with the El Paso multiyear restaurant study that compared franchisee
failures to independents and corporate chain units. The El Paso study authors contend ‘‘that
other factors, particularly the greater initial investment of the franchisees, have more to do
with their greater success [compared to independents] than their association with the
franchising format’’ (English and Willems, 1991–1999).
Longitudinal analysis within each investment category may provide insights on franchisee
turnover rate changes. Because of investment category structure shifts after the first year, only
categories used in the last 3 years are consistent across time. The three smallest investment
categories, under US$74,999, show a declining pattern over 3 years using both turnover with
transfer and turnover-only data. The middle investment categories, from US$75,000 to
US$499,999 show turnover rates increasing from 1995 to 1997. The two largest non-real

Table 8
Franchisee turnover rates by non-real estate initial investment
Total all Turnover rates (medians, %) by non-real estate initial investment (in thousand dollars)
categories
Under US$25 – US$50 – US$75 – US$100 – US$250 – US$500 – US$750 – Over US$1
US$25 US$49 US$74 US$99 US$249 US$499 US$749 US$999 million
Median with transfers
1994 9.72 NA NA NA NA 3.85 8.62
1995 18.53 12.04 12.36 8.55 9.52 7.65 6.86 11.48 11.50
1996 12.2 11.1 12.3 10.5 10.8 8.7 12.0 7.0 12.2
1997 12.0 12.0 10.0 12.0 11.1 8.0 10.0 3.0 11.0

Median without transfers


1994 5.60 NA NA NA NA 0.30 4.65
1995 13.33 7.33 7.91 3.80 4.15 4.49 4.86 6.50 6.12
1996 7.8 7.1 6.7 5.3 5.9 5.8 9.1 4.5 6.6
1997 9.0 7.0 5.0 6.0 6.0 4.0 8.0 2.0 4.0
416 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

estate investment categories, above US$750,000, show franchisee turnover rates decreasing
from 1995 to 1997.
Comparing across category and longitudinal data patterns, it would appear that, although
somewhat mixed results are present, there is a general convergence trend. The smallest
investment categories that had the highest franchisee turnover rates in the early years have
had falling franchisee turnover rates over the 3-year period. The middle investment categories
that had somewhat lower franchisee turnover rates have seen rising franchisee turnover rates
over the 3-year period. Finally, the largest two investment categories (over US$750,000) that
were high in 1995 have decreased by 1997. These results are preliminary and merit additional
research and analysis.

5. Conclusions

Past franchisee failure literature has included numerous divergent accounts and statistical
franchisee failure estimates. Prior studies, based largely on franchisor surveys, contain potential
definitional and data problems. In addition, the failure concept utilized has often placed
franchisees into a relatively simplistic position of being either ‘‘successes’’ or ‘‘failures’’.
This paper uses a broader franchisee failure concept and provides new actual longitudinal
franchisee failure data. Combined, they present a better understanding of evolving franchisor
and franchisee business success/failure dimensions and realities. The emergence of more
reliable franchisee turnover UFOC data provides valuable new franchisee failure turnover
information and perspectives. With 4 years of data for over 800 franchisors and over 250,000
franchisee units, many research questions can be examined more extensively.
Franchisee overall turnover rates appear to have increased over the 4-year period when
examining total franchisee turnover with transfer or turnover-only data. By 1997, the median
turnover rate was 10.49% with transfers and 5.54% without transfers. Franchisee turnover was
also analyzed relative to variables of industry, system size, growth rates, and initial investment.
While these data may provide preliminary insights, data for these variables are not conclusive,
show a variety of patterns, or provide mixed indicators. More sophisticated subcategory data
analysis was constrained by the lack of individual franchisor and franchisee company data.
Further research is needed to resolve some ambiguities in present subcategory data.
Broadening the franchisee failure perspective in two regards expands the potential real or
opportunity cost/loss. First, a stakeholder perspective mandates inclusion of parties beyond
the franchisor and franchisee. Suppliers, financial institutions, customers, investors, and
others may have real costs/losses and/or opportunity costs due to franchisee turnover. Second,
placing franchise failure on a continuum (Holmberg and Morgan, 1996, 2000) implies that
franchisors, franchisees, and other stakeholders could see real costs/losses and/or opportunity
costs at stages prior to and after the turnover phase. There are costs if franchisor–franchisee
relationships and business issues are not managed appropriately. These real or opportunity
costs could be minimized by systematic franchisee failure risk mitigation strategies deployed
early and throughout the phases described in the Holmberg–Morgan franchise failure
continuum model phases.
S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418 417

This paper’s increased UFOC data validity and franchisee turnover insights represent new,
important information for existing and potential franchisee ventures. Franchisee turnover and
franchisee failure are critical areas that merit further study and analysis.

References

Bates, T., 1995a. Analysis of survival rates among franchise and independent small business startups. J. Small
Bus. Manage. 33, 26 – 36.
Bates, T., 1995b. Survival rates among newcomers to franchising. J. Bus. Ventur. 13, 113 – 130.
Castrogiovanni, G., 1991. Environmental munificence a theoretical assessment. Acad. Manage. Rev. 16, 542 –
565 (July).
Castrogiovanni, G.J., Justis, R., Julian, S.D., 1993. Franchise failure rates: an assessment of magnitude and
influencing factors. J. Small Bus. 31, 105 – 114.
Castrogiovanni, G.J., Vennett, N., Combs, J.G., 1995. Franchisor types: reexamination and clarification. J. Small
Bus. Manage., 45 – 55 (January).
Chun, J., 1996. Crossing the line. Entrepreneur, 150 (May).
Cross, J., 1994. Franchising failures: definitional and management issues. Society of Franchising Conference
Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Cross, J., 1998. Improving the relevance of franchise failure studies. Society of Franchising Conference Proceed-
ings. International Society of Franchising, St. Paul, Minnesota, USA.
Dant, R.P., Nasr, N.I., 1998. Control techniques and upward flow of information in franchising markets: con-
ceptualization and preliminary evidence. J. Bus. Ventur. 13, 3 – 28.
English, W., Willems, J., 1991 – 1999. Franchise vs. Non-franchise Restaurant Attrition (various years of a nine
year study of a yellow pages longitudinal analysis). Papers published in Society of Franchising Conference
Proceedings.
Fable, C.M., Dandridge, T.C., Kumar, A., 1998. The effect of organizational context on entrepreneurial strategies
in franchising. J. Bus. Ventur. 14, 125 – 140.
Falbe, C., Welsh, D.H.B., 1998. NAFTA and franchising: a comparison of franchisor perceptions of characteristics
associated with franchisee success and failure in Canada, Mexico and the United States. J. Bus. Ventur. 13,
151 – 171.
Fulop, C., Forward, J., 1997. Insights into franchising: a review of empirical and theoretical perspectives. Serv.
Ind. J. 17, 603 – 625.
Hartnett, M., 1992. Franchising: how safe are these things? Restaur. Bus. 5, 66 (March 20).
Holmberg, S., Morgan, K.B., 1996. The franchising failure continuum. Society of Franchising Conference Pro-
ceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Holmberg, S., Morgan, K.B., 2000. Franchise failure: new research and perspectives. Society of Franchising
Conference Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Hoy, F., 1994. The dark side of franchising, or appreciating flaws in an imperfect world. Int. Small Bus. J. 12,
26 – 38.
IFA Educational Foundation, 1996. The Study of Franchised Unit Turnover. IFA Educational Foundation, Wash-
ington, DC.
IFA Educational Foundation, 1998. The Profile of Franchising. Volume II: A Statistical Abstract of 1996 UFOC
Data. IFA Educational Foundation, Washington, DC.
IFA Educational Foundation, 1999. The Profile of Franchising. Volume II: A Statistical Abstract of 1997 UFOC
Data. IFA Educational Foundation, Washington, DC.
IFA Educational Foundation, 2000. The Profile of Franchising. Volume III: A Statistical Abstract of 1996 UFOC
Data. IFA Educational Foundation, Washington, DC.
Jambulingam, T., Nevin, J.R., 1999. Influence of franchisee selection criteria on outcomes desired by the fran-
chisor. J. Bus. Ventur. 14, 363 – 395.
418 S.R. Holmberg, K.B. Morgan / Journal of Business Venturing 18 (2003) 403–418

Justis, R.T., Castrogiovanni, G.J., Chan, P., 1992. Examination of franchise failure rates. Society of Franchising
Conference Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Kaufmann, D., 1995. The case against mandatory earnings claim disclosure. Franch. Law J. 15 (1) 3, 6 – 9.
Kaufmann, P.J., 1999. Franchising and the choice of self-employment. J. Bus. Ventur. 14, 345 – 362.
Kaufmann, P.J., Dant, R.P., 1996. Multi-unit franchising: growth and management issues. J. Bus. Ventur. 11,
343 – 358.
Kaufmann, P.J., Dant, R.P., 1998. Franchising and the domain of entrepreneurship research. J. Bus. Ventur. 14,
5 – 16.
Kaufmann, P.J., Lafontaine, F., 1994. Costs of control: the source of economic rents for McDonald’s franchisees.
J. Law Econ. 37 (2), 417 – 453.
Lafontaine, F., 1994. Risk in franchising: a look at some measurement issues. Society of Franchising Confer-
ence Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Lafontaine, F., Shaw, K.L., 1998. Franchising growth and franchisor entry and exit in the U.S. market: myth and
reality. J. Bus. Ventur. 13, 95 – 112.
Mitenko, G., 1991. A methodology to predict the success of a franchise. Society of Franchising Confer-
ence Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Phan, P.H., Butler, J.E., Lee, S.H., 1996. Crossing mother: entrepreneur – franchisees’ attempts to reduce franchi-
sor influence. J. Bus. Ventur. 11, 379 – 402.
Prescott, E., Visscher, M., 1980. Organizational capital. J. Polit. Econ. 88, 446 – 461.
Shane, S., 1996. Hybrid organizational arrangements and their implications for firm growth and survival: a study
of new franchisors. Acad. Manage. J. 39, 216 – 234.
Shane, S., Spell, C., 1997. Enhancing new franchisor survival: a model and empirical test. Society of Franchising
Conference Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Stanworth, J., Curran, J., 1999. Colas, burgers, shakes, and shirkers: towards a sociological model of franchising
in the market economy. J. Bus. Ventur. 14, 323 – 344.
Stanworth, J., Dandridge, T., 1994. Business franchising and economic change: an overview. Int. Small Bus. J. 12,
12 – 14.
Stanworth, J., Purdy, D., Price, S., 1997. Franchise growth and failure in the U.S. and the U.K.: a troubled
dreamworld revisited. Society of Franchising Conference Proceedings. International Society of Franchising, St.
Paul, Minnesota, USA.
Stanworth, J., Purdy, D., Price, S., Zafiris, N., 1998. Franchise versus conventional small business failure rates in
the US and UK: more similarities than differences. Int. Small Bus. J. 16, 56 – 69.
Stanworth, J., Purdy, D., English, W., Willems, J., 2000. Unravelling the evidence on franchise system surviv-
ability. Society of Franchising Conference Proceedings. International Society of Franchising, St. Paul,
Minnesota, USA.
Vincent, W.S., 1998. Encroachment: legal restrictions on retail franchise expansion. J. Bus. Ventur. 13, 29 – 41.
Walker, B.J., Cross, J., 1988. Franchisee failures: more questions than answers. Society of Franchising Confer-
ence Proceedings. International Society of Franchising, St. Paul, Minnesota, USA.
Williams, D.L., 1998. Why do entrepreneurs become franchisees? An empirical analysis of organizational choice.
J. Bus. Ventur. 14, 103 – 124.