Anda di halaman 1dari 13

International Journal of Research in Marketing 34 (2017) 252–264

Contents lists available at ScienceDirect

IJRM
International Journal of Research in Marketing
journal homepage: www.elsevier.com/locate/ijresmar

Full Length Article

Outsourcing to convert fixed costs into variable costs: A


competitive analysis
Yunchuan Liu a, Rajeev K. Tyagi b,⁎
a
College of Business, University of Illinois, Urbana Champaign, Champaign, IL, 61820, USA
b
Paul Merage School of Business, University of California, Irvine, Irvine, CA 92697, USA

a r t i c l e i n f o a b s t r a c t

Article history: Outsourcing of production, services, and various economic activities is a pervasive phenomenon
First received on May 10, 2011 and was across industries. One of the key economic benefits of this popular practice, as mentioned by its
under review for 6 months sellers and buyers, is that it allows the outsourcing firm to reduce its fixed costs such as expen-
Available online 24 August 2016
ditures on equipment, information technology, fixed salaries of employees, etc., and convert
those into a variable cost in the form of the purchase price that the outsourcing firm then
Area Editor: David Soberman
pays the outside industry. This paper examines the strategic implications of this role of
outsourcing in an oligopolistic setting. We show how these strategic considerations imply
Keywords:
that, even absent any cost savings from outsourcing, competing firms can find it profitable to
Outsourcing
outsource. Furthermore, in such a setting, one firm may outsource while the ex-ante similar
Fixed and variable costs
Competition competing firm may not outsource. We also examine how consumer and social welfare are
Game theory adversely affected when outsourcing plays this fixed-cost-to-variable-cost conversion role in
an oligopolistic setting.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction

Outsourcing of production, services, and various economic activities is a pervasive phenomenon across industries. A key eco-
nomic benefit of outsourcing mentioned prominently by both sellers and buyers of this practice is that it allows the outsourcing
firm to reduce its fixed cost and convert it into variable cost (Kakabadse & Kakabadse, 2000; Kelleher, 1990; Kremic, Tukel, & Rom,
2006; Razzaque & Sheng, 1998). Specifically, the outsourcing firm can avoid a significant part of the fixed costs of facilities, equip-
ment, information technology, rents, personnel salary, insurance, and logistic and overhead expenses. These reductions in fixed
costs are replaced with increases in variable costs in the form of purchasing prices that the outsourcing firm must pay the outside industry.
This benefit of outsourcing is advertised prominently by outsourcing consultancy houses, and highlighted in industry reports
and formal surveys of outsourcing firms. The Accenture website mentions “migrate fixed cost to variable cost” as an advantage
of offshoring1; HCL BPO highlights “convert fixed cost to variable cost business,” as the advantage of outsourcing2; the website
of Emineo Partners, a consultancy house for outsourcing, says “conversion of fixed cost to variable cost” is a key benefit to its
clients3; and Cap Gemini gives “convert fixed costs to variable costs” as the first benefit of their internal audit service
(Capgemini Internal Audit Services Survey, 2009). In the Strategic Outsourcing Survey 2004, conducted jointly by CAPS Research

⁎ Corresponding author.
E-mail addresses: liuf@illinois.edu (Y. Liu), rktyagi@uci.edu (R.K. Tyagi).
1
https://microsite.accenture.com/concadia/Pages/default.aspx.
2
http://www.hclbpo.com/pdf/brochures/bfs/Mortgage%20Services.pdf.
3
http://www.emineopartners.sk/en/outsourcing.php.

http://dx.doi.org/10.1016/j.ijresmar.2016.08.002
0167-8116/© 2016 Elsevier B.V. All rights reserved.
Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264 253

and A.T. Kearney, out of 165 companies surveyed from across 24 global industries, 58% of respondents cited “turn fixed costs into
variable costs” as one of the key reasons for outsourcing (Monczka, Markham, Carter, Blascovich, & Slaight, 2005).
Such surveys and industry reports also mention which specific fixed costs get converted to variable costs by outsourcing. A
report on outsourcing by Deloitte and Touche (2014) calls fixed costs on Information Technology, Human Resources, Finance
and Accounting, and Procurement as the big four types of fixed costs being changed to variable costs by outsourcing. A report
by Ernst and Young (2013) mentions software and computer systems outsourced to cloud services as an example. In a survey
by Purchasing Magazine, more than 50% of outsourcing firms surveyed cited cutting the fixed costs of transportation/distribution
costs, freeing up or reducing staff, focusing on the core business and cutting internal administrative costs as major reasons for
outsourcing (Razzaque & Sheng, 1998). Small Business Authority (2012) and Regus (2010) describe fixed costs on marketing
staff, designers, and office space as being converted to variable costs; similarly, ADP Total Source (2012)mentions fixed costs on
HR, payroll, and compliance; and Berry-Wehmiller International Resources mentions fixed costs involved in engineering services
as examples of fixed cost being converted to variable cost by outsourcing (Barry-Wehmiller International Resources, 2012).
Yourvoice Outsourcing Company, which offers both front office and back office services (e.g., Document Administration, Analytics,
Finance, Accounting, Customer Care, Customer Surveys, Consumer Research, and Quality Audits), mentions “These services are of-
fered on a fee-for-service basis, which helps your business become more flexible by transforming fixed costs into variable costs
(Outsourcing—Your Voice CMR, 2013).” Tabuchi (2011) and Sutton (2012) describe how Sony saved fixed costs by closing produc-
tion facilities for LCD screens and outsourcing them, while its competitor Samsung maintains more such production facilities to
produce in-house.
In spite of the immense popularity of both the outsourcing phenomenon and of the idea that it converts a firm's fixed cost into
variable cost, it is surprising that little rigorous academic work has been done to analyze the economic implications of this key
feature of outsourcing. This paper aims to fill this void, and formally examines the effects of this role of production/service
outsourcing on outsourcing firms and customers in a competitive setting.
Our analysis highlights the following key economic effects and results. First, we show that outsourcing, when it plays the fixed-
cost-to-variable-cost conversion role, can benefit competing firms by allowing them to sustain higher prices. Intuitively, when
firms produce the products or service in-house, their fixed costs rationally become sunk costs when they decide their prices
and hence price competition is intense. On the other hand, when they outsource and then compete on prices, then the conversion
of fixed costs into variable costs implies that they compete with reduced fixed costs and higher variable costs, allowing them to
sustain higher prices. We show that this implies that firms may outsource even when production is more costly for the outside
industry than for the outsourcing firms. In other words, competing firms can engage in outsourcing even when it does not lead
to any cost savings. Of course, any cost savings provided by outsourcing only adds to this beneficial effect.
Second, we show that besides the above-mentioned beneficial price increase effect, outsourcing by a firm, if done unilaterally,
creates a detrimental competitive cost disadvantage for it and a beneficial competitive cost advantage for its competitor. Specifi-
cally, when only one firm outsources to an outside industry, its variable cost will be higher than the variable cost of the competing
firm which chooses in-house production. This higher variable cost leads to a disadvantage for the outsourcing firm in its price
competition with the competing firm.
We show how, depending on the relative magnitudes of the above-described competitive cost disadvantage effect and the
beneficial price increase effect, we can have different outsourcing equilibria in a market: an equilibrium where both competing
firms do not outsource; an equilibrium where both competing firms outsource; or even an asymmetric equilibrium where one
firm outsources and the competing firm does not. This last possible outcome implies that the optimal outsourcing strategy can
be asymmetric even for symmetric firms. Intuitively, when a firm chooses to outsource, it benefits the competing firm since
both the competitive cost disadvantage effect and the beneficial price increase effect favor the competing firm. This increases
the profit that the competing firm gets from not outsourcing, and hence effectively reduces the additional beneficial price increase
benefit that this competing firm will get if it also chooses to engage in outsourcing. Therefore, we can have an equilibrium where
ex-ante symmetric firms can become ex-post asymmetric in that only one firm chooses to outsource.4
Finally, we also show how consumer welfare can be adversely affected when outsourcing plays the fixed-cost-to-variable-cost
conversion role examined in this paper. In this framework, the fixed costs converted to variable costs are shifted to the shoulder of
consumers in the form of increased prices. Although firms may benefit from the beneficial price increase effect of outsourcing, this
benefit does not always offset the loss of consumer welfare in our framework, and hence social welfare can also reduce. Of course,
if outsourcing also leads to cost savings, i.e., the outside industry can produce at a lesser cost than the outsourcing firms, then that
can generate a beneficial effect on consumer and social welfare, mitigating or even overwhelming the reduction in consumer and
social welfare caused by the strategic effects of outsourcing discussed above.

1.1. Relationship to literature

In addition to the papers cited earlier, our paper is also related to the following streams of work in the marketing, strategy, and
operations management literatures.

4
One can see examples of industries where among approximately similar competing firms, some choose outsourcing and others do not. For example, Apple does not
outsource its call centers much, while other main PC manufacturers, such as Dell and HP, outsource heavily (Mourdoukoutas, 2013). In the LCD television market,
Samsung produces most of its LCD in-house in expensive factories, while other competitors such as Sharp and Sony outsource this production to a much larger extent
(Tabuchi, 2011). Similarly, Intel produces most of its processors in its own factories while AMD outsources much more of its production (Franco, 2015).
254 Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264

A number of academic papers and a vast number of industry publications make the point that outsourcing allows a firm to save
costs and to focus on its core competencies (e.g., Arnold, 2000; Laarhoven, Berglund, & Peters, 2000; Quinn & Hillmer, 1995; Vining
& Globerman, 1999). Chalos and Sung (1998) and Qu and Brocklehurst (2004) highlight the trade-off of outsourcing benefits against
the increased coordination and transaction costs. Cachon and Harker (2002) show that when competing firms have economies of
scale, then each firm has an incentive to overproduce, leading to intense price competition. In such a situation, outsourcing by both
firms can soften price competition by taking away this overproduction incentive. Gilbert, Xia, and Yu (2006) show how outsourcing
can act as a mechanism to dampen firms' incentives to overinvest in cost reduction and avoid the resultant intense price competition.
Arya, Mittendorf, and Sappington (2008) show that if competing firms outsource to the same supplier, then it can be in the interest of
the supplier to not offer a better price deal to an outsourcing firm selectively, again avoiding intense price competition among the
outsourcing firms. Our paper focuses on the strategic effects of a different aspect of outsourcing—conversion of fixed costs into vari-
able costs, and shows how it leads to none, one, or both competing firms to engage in outsourcing.
In the marketing literature, a number of papers examine firms' decisions on whether to sell directly or through downstream
intermediaries (Choi, 1991; McGuire & Staelin, 1983; Trivedi, 1998). In contrast to these papers, we examine firms decision on
whether to have production in-house or use upstream suppliers for production. More closely related to our paper are the papers
by Anderson (1985), Horsky (2006), and Stremersch, Weiss, Dellaert, and Frambach (2003), who also examine outsourcing deci-
sions. Anderson (1985) uses transaction cost analysis to examine when firms keep selling function in-house versus delegating to
outside firms, and finds that firms use in-house salespeople when it is more difficult to measure their performance. Horsky (2006)
examines outsourcing of advertising function, and finds that this happens when firms do not have internal abilities and have high-
level requirements in creative areas. Stremersch et al. (2003) examine firms' decisions on whether to outsource system integration
of modular systems, and find that a firm's internal know-how and the technological volatility it faces influences its preference for
outsourcing system integration. Our paper adds to this literature by examining a widely mentioned reason by practitioners for
outsourcing—conversion of fixed cost into variable cost.5 Specifically, we focus on this role in a competitive setting and examine
its implications for competing firms, and consumer and social welfare.
Finally, before moving on to describing a formal model in the next section, we emphasize that our paper focuses on one
important reason firms outsource, but there are usually a number of reasons for outsourcing. Our search of industry, consultancy,
and academic studies found that converting fixed cost into variable cost, reducing total cost, and access to specialized knowledge
are usually the top three reasons. For example, Abraham and Taylor (1996) use data from bureau of labor statistics on manufactur-
ing employer's use of outsourcing for various business services, such as janitorial services, machine maintenance services,
engineering and drafting services, accounting services, and computer services, and find empirical evidence in favor of each of
the above-mentioned three main reasons. In a survey of 165 outsourcing firms across various industries (Monczka, Blascovich,
& Carter, 2004), 58% of the respondents mention using outsourcing to convert fixed cost into variable cost, 89% mention reducing
total operating costs, and 55% mention gaining access to needed skills. Lau and Zhang (2006) conduct in-depth case studies and
interviews of Chinese firms that outsource, and found evidence in favor of these three reasons.
Overall, our paper focuses on one important reason for outsourcing, and examines its competitive effects. Models focusing on
some other reason for outsourcing, say to access specialized knowledge, would lead to different competitive implications. The
effects shown in this paper coexist with other economic effects in reality, and hence managerial actions must consider a compre-
hensive and integrative framework that includes factors shown in this and prior work in the area of outsourcing.

2. A model of outsourcing

Our goal is to build the simplest possible model that allows us to focus attention on competition between two firms that can
conduct an economic activity in-house and incur the associated fixed costs, or outsource to incur lower fixed costs but higher
variable costs in the form of price paid to an outside industry.
There are two ex-ante symmetrically differentiated firms, 1 and 2, which compete on price to sell to consumers. We use the
popular Hotelling framework to model the situation. In particular, consumers' ideal points are distributed uniformly on a unit
line bounded by 0 and 1, firm 1 is located at 0, and firm 2 is located at 1. The reservation price for consumers when they buy
their ideal products is V, and V is sufficiently large such that the market is covered and firms 1 and 2 are in de facto competition.6
When buying a product that is not ideal, consumers will incur some mismatching costs that increase in the distances between
consumers' locations and the firms' locations. For a consumer located at x on the Hotelling line, it incurs a mismatching cost of
tx if it buys from firm 1 or t(1 − x) if it buys from firm 2. A higher value of parameter t implies higher consumer disutility
from buying a product away from its ideal point, and is generally taken as a proxy for softer price competition or more
differentiation between firms. The size of the market is normalized to unity without loss of generality.
Each firm can decide whether to conduct the focal economic activities in-house or outsource them to an outside industry. If a
firm conducts it in-house, it incurs (i) a fixed cost of f, and (ii) a variable cost, which we normalize to zero without loss of

5
There is also a vast literature on a firm's decision to vertically integrate and produce its input materials in-house, rather than buy from another firm. That vertical
integration literature focuses on effect of factors such as asset specificity, hold up costs, uncertainty, cost of searching and imperfection in matching for suitable outsider
partners, etc. (e.g., Grossman & Helpman, 2002). Outsourcing decision in our framework has a different focus and is a particular form of make-or-buy decision in which
outsourcing converts fixed costs into variable costs, and the outside industry is perfectly competitive.
6
We replicate our analyses with a reduced-form linear demand function in Section 4, and show that our main qualitative results are robust. That linear demand func-
tion allows for an outside option for consumers, and also allows a consumer to have non-unitary demand.
Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264 255

generality. As mentioned earlier, this fixed cost captures costs such as expenditures on facilities, equipment, information technol-
ogy, rents, personnel salary, insurance, and logistic and overhead expenses, etc. If a firm chooses to outsource, (i) its fixed cost
reduces, which we normalize without loss of generality to zero; and (ii) variable cost increases to w, which is the per-unit
price it must pay the outside industry. The assumption that the outsourcing contract consists of a variable price w is of course
a simplification of potentially complex real-life contracts to capture and focus on the idea that outsourcing allows firms to
avoid fixed costs that are incurred irrespective of the potentially fluctuating demand and instead incur variable costs which
track demand more directly.7
The outside industry is assumed to be perfectly competitive, and have no advantage on the fixed production cost over firms 1
and 2. The assumption that the outside industry incurs the same fixed cost, f, as the outsourcing firm producing in-house would,
controls for the cost-savings based rationales for outsourcing. The assumption of perfectly competitive outside industry allows us
to focus away from strategic actions by outside industry; for example, of the type shown by the monopolist supplier in Arya et al.
(2008). Furthermore, this assumption can be seen substantively as holding in markets where the outside industry is competitive.
Recalling that the variable cost of firms 1 and 2 was normalized to zero, we let the variable cost of the outside industry to be c.
Of course, the interesting case is where c ≥ 0, i.e., where the outside industry does not have any cost advantage over the firms
contemplating outsourcing; in this case, if outsourcing still occurs, it does for a reason other than cost savings.8 That would
then control for the already known cost-savings rationale for outsourcing. We also assume that neither outsourcing firms nor
the outside industry enjoy more economy of scales. This removes the alternative explanation that production outsourcing is due
to the effect of economy of scales in the manufacturing industry.
We next analyze the three possible actions by firms—neither firm outsources, only one firm (either firm 1 or firm 2) outsources,
and both firms outsource.

2.1. No outsourcing

We first analyze the benchmark case where both firms 1 and 2 choose to not outsource, and each incurs a fixed cost f. In this
case, firms 1 and 2 decide their respective prices p1 and p2 to maximize their respective profits. Given p1 and p2, a consumer
located at x has a surplus of V −tx − p1 if buying from firm 1 and V − t(1 − x) −p2 if buying from firm 2. Therefore, the indifferent
consumer between firm 1 and firm 2 is located at ^x ¼ 12 þ p2 −p
2t . The profit functions of firms 1 and 2 are given by
1

   
1 p2 −p1 1 p2 −p1
Π1 ¼ p1 þ −f ; Π2 ¼ p2 − −f : ð1Þ
2 2t 2 2t

Maximizing firms' profits, we have

    t
p1 ¼ p2 ¼ t; Π1 ¼ Π2 ¼ −f : ð2Þ
2

We note that the equilibrium prices do not depend on the fixed cost f. This is because the fixed costs are sunk costs for firms
when they decide their prices. Also, note that for firms to have positive profits, we must have f bf ¼ 2t, a condition we use through-
out our analysis.

2.2. Outsourcing by one firm

Next, we analyze the case where one firm chooses to outsource to an outside industry and the other firm chooses to not do it.
Due to symmetry, we only analyze the case where firm 1 outsources and firm 2 does not.
In particular, the game unfolds as follows: first, the outsourcing firm 1 quotes a price w1 that it will pay the outsourcing
industry, keeping in mind the zero-profit condition of the perfectly competitive outside industry; second, the competing firms 1
and 2 choose their product prices; and finally, consumers choose which product to buy. We analyze the model following backward
induction, and first derive the optimal product prices p1 and p2 given the contract price w1 offered by firm 1 to the outside indus-
try. Firms' profits are given by
   
1 p2 −p1 1 p2 −p1
Π1 ¼ ðp1 −w1 Þ þ ; Π2 ¼ p2 − −f : ð3Þ
2 2t 2 2t

7
As we discussed earlier, almost all industry surveys, industry reports, and popular press articles on outsourcing list “ conversion of fixed cost into variable cost” as a
prominent reason for outsourcing. A 2012 report on cloud computing by a European Commission, states “ When purchased as a service, cloud computing is highly cost
effective as it is based on pay-per-use” and that “ Customers are spared the expense of purchasing, installing and maintaining hardware and software locally (Commu-
nication from the Commission to the European Parliamen).” Defining what we call per-unit or demand-based price as “ fixed price contracts” which charge a fixed price
per transaction per billing cycle, Mani, Barua, and Whinston (2013) report that about 33% of the 100 largest outsourcing initiatives implemented between 1996 and
2005 were of this type.
8
Since the marginal cost of the outsourcing firms is normalized to zero, a negative value of c represents the situation where the outside industry has a lower marginal
cost of production, compared to the outsourcing firms. At the cost of introducing more variables in the analysis, one can get the same results by representing marginal
cost of outsourcing firms by CN 0 and the marginal cost of the outside industry to be C±c and impose restrictions such that C−c≥0.
256 Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264

Note that since firm 1 outsources, it does not incur the fixed cost f. Firm 2, however, still incurs the fixed cost f. Maximizing
firms 1 and 2's profits, we have

2w1 w
^1 ¼ t þ
p ^2 ¼ t þ 1 :
;p ð4Þ
3 3

^1 and p
Therefore, both the product prices p ^2 increase with the contract price w1 offered by firm 1 to the outside industry.9 The
location of the marginal consumer indifferent between buying from firm 1 and firm 2 is located at ^x ¼ 12 − w6t1 . Firm 1's profit is
given by
 
^ ¼ ðp 1 w
Π1
^1 −w1 Þ − 1 ; ð5Þ
2 6t

and the sales by ð12 − w6t1 Þ. For the perfectly competitive outside industry to supply to firm 1 at price w1, we must have ðw1 −cÞð12 − w6t1 Þ−
f ≥0. Maximizing firms' profits given the constraint of an acceptable contract to the outside industry, we have10
 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 1
w1 ¼ c þ 3t− 9t 2 þ c2 −6ct−24tf ; ð6Þ
2

 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 1  1
p1 ¼ c þ 6t− 9t 2 þ c2 −6ct−24tf ; p2 ¼ c þ 9t− 9t 2 þ c2 −6ct−24tf ; ð7Þ
3 6

 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2  pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2

3t−c þ 9t 2 þ c2 −6ct−24tf 
9t þ c− 9t 2 þ c2 −6ct−24tf
Π1 ¼ ; Π2 ¼ −f : ð8Þ
72t 72t

We note from the above expressions that when the fixed cost f increases, the contract price w1⁎ offered by firm 1 to the outside
industry increases, and the product prices p1⁎ and p2⁎ also increase. In addition, the contract price and the product prices also
increase with the marginal production cost of the outside industry. This is because when production in the outside industry
becomes more costly either on the fixed production cost or the variable product cost, firm 1 has to offer a better contract price
w1⁎ for that outside industry to supply to firm 1.
Using (2) and (8) to compare firms' profit here with those in the case where neither firm outsources, we get the following
result.

Proposition 1. When firm 2 does not outsource, then firm 1 can benefit from outsourcing to the outside industry even if that industry
does not have a cost advantage (c ≥ 0). Firm 2 is better off with firm 1's outsourcing unless the outside industry has a large cost
advantage over firms 1 and 2.

Proof. See Appendix A.

Intuitively, there are two important effects when firm 1 outsources to the outside industry. First, when firm 1 does not out-
source, then its fixed production cost does not affect firms' prices since the fixed costs become sunk costs for firms when they
choose to not outsource. However, when firm 1 outsources to outside industry, the fixed cost, now borne by the outside industry
rather than firm 1, becomes relevant to pricing since that industry will not agree to the outsourcing contract if the outsourcing
contract term w1 is too low to meet their zero-profit condition. Therefore, firms 1's effective marginal cost—the contract price
quoted to the outside industry—is higher when it outsources (mc = w1⁎ N 0) than its marginal cost when it does not (mc = 0).
This consequently increases the product prices by both firm 1 and firm 2 and benefits both firm 1 and firm 2.
Second, outsourcing by firm 1 also leads to a higher effective marginal cost for firm 1 (mc =w1⁎ N 0) than for firm 2 (mc = 0).
This disadvantages firm 1 in its price competition with firm 2—the competitive cost disadvantage effect. This effect hurts firm 1 but
benefits firm 2, and increases in magnitude as the outside industry becomes more cost inefficient relative to firms 1 and 2.
Overall, as long as the outside industry is not too cost disadvantaged (c ≥ 0 is small), the magnitude of the competitive cost
disadvantage effect stays lower than that the beneficial price increase effect, and hence firm 1 benefits from outsourcing.

9
Given that the outside supplier market has been modeled in our paper as a perfectly competitive one (a commodity market), it is reasonable to assume that contract
price w1 is public knowledge and is observable by the rival firm 2. In a different context, say in a model where the suppliers are monopolists, one can also model contract
price as private information. Traditionally, much of the marketing analytical literature and economics literature in distribution channels have assumed public observ-
ability of wholesale prices, but there does exist a specialized literature that explicitly focuses on private nature of wholesale prices (Coughlan & Wernerfelt, 1989; Kuhn,
1997; O'Brien & Shaffer, 1992). There is room for additional papers on outsourcing which allow different market structures for the supplier market and correspondingly
different information structures.
10
There are two roots of w1, but the one given in (6) is the optimal price. We need fb (3t−c)2/(24t) for the outsourcing firm to have a positive market share and profit.
Note (3t−c)2/(24t)b t/2 always holds for c≥0.
Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264 257

For firm 2, both the price increase effect and the competitive cost effect are beneficial as long as the outside industry used by
firm 1 does not have a large marginal-cost advantage. Intuitively, when that industry has a large marginal-cost advantage, then the
outsourcing firm 1 may gain a price competition advantage due to its lowered marginal cost, making firm 2 worse off.

2.3. Outsourcing by both firms

Now, we analyze the case where both firm 1 and firm 2 outsource to outside industries. Here the game unfolds as follows: first,
the outsourcing firms 1 and 2 quote prices w1 and w2 that they will pay their outsourcing industries, keeping in mind the zero-
profit condition of a perfectly competitive industry; second, firms 1 and 2 choose their product prices p1 and p2; and finally,
consumers choose which product to buy. We analyze the model following backward induction, and first derive the optimal
product prices p1 and p2 given the outsourcing terms w1 and w2 offered by firms 1 and 2 to the outside industries. Firms 1 and
2's profits are given by
   
1 p2 −p1 1 p −p1
Π1 ¼ ðp1 −w1 Þ þ ; Π2 ¼ ðp2 −w2 Þ − 2 : ð9Þ
2 2t 2 2t

Note that when both firms outsource, neither firm incurs the fixed cost f. Maximizing firms' profits, we have

1 1
^1 ¼
p ^2 ¼ ð3t þ w1 þ 2w2 Þ:
ð3t þ 2w1 þ w2 Þ; p ð10Þ
3 3

Again, both the product prices p ^1 and p^2 increase with the contract prices w1 and w2 offered by firms 1 and 2 to the outside
industries.11 The new location of the marginal consumer indifferent between buying from firm 1 and buying from firm 2 is located
x ¼ 12 − w1 −w
at ^ 6t . Firms' profits are given by
2

   
^ ¼ ðp 1 w1 −w2 ^ 1 w1 −w2
Π1
^ 1 −w1 Þ − ; Π2 ¼ ðp^ 2 −w2 Þ þ ; ð11Þ
2 6t 2 6t

with the respective sales for the two firms being ð12 − w1 −w
6t Þ and ð2 þ
2 1 w1 −w2
6t Þ. For the perfectly competitive outside industries to
supply to these firms at prices w1 and w2, we must have ðw1 −cÞð12 − w1 −w 6t Þ−f ≥0 and ðw2 −cÞð2 þ
2 1 w1 −w2
6t Þ− f ≥0. Maximizing
firms' profits given the constraints of acceptable contracts to the outside industries, we have

      t
w1 ¼ w2 ¼ c þ 2 f ; p1 ¼ p2 ¼ c þ t þ 2f ; Π1 ¼ Π2 ¼ : ð12Þ
2

We note again that the contract prices w1⁎ and w2⁎ and the product prices p1⁎ and p2⁎ increase with f, the fixed cost.12

2.4. Outsourcing equilibrium

Using firms' profit expressions from the above-examined cases where (i) neither firm outsources (Eq. (2)), (ii) only one firm
outsources (Eq. (8)), and (iii) both firms outsource (Eq. (12)), we calculate the conditions under which in equilibrium neither firm
outsources, only one firm outsources, and both firms outsource. When the cost of the outside industry is sufficiently high (cNc1 ¼ 3t−
qffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffi
9t 2 −2tf þð9t−8 f Þ tðtþ2 f Þ
ð3t−4f Þ t−2 t
f
), neither firm outsources. When the cost of the outside industry is sufficiently low (c b c 2 ¼ 3t−2 f
), both
pffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi
9t 2 −2tf þð9t−8 f Þ tðtþ2 f Þ
firms outsource. When the cost is moderate (c2 ¼ 3t−2 f
b c b c1 ¼ 3t−ð3t−4 f Þ t−2 t
f
), only one firm will outsource.
Proposition 2 summarizes the main insights. We also illustrate the conditions for different equilibria in Fig. 1 for t= 1.

Proposition 2. Even when the outside industry does not have a cost advantage (c≥ 0), (i) both firms may choose to outsource, (ii) one
firm may choose outsourcing and the other one does not.

Proof. See Appendix A.

When both firms 1 and 2 outsource, then compared to the case where only firm 1 outsources, the price increase effect is
further enhanced since the fixed cost f for firm 2 becomes relevant when firm 2 also outsources. The price offered by firm 2 to
the outside industry again has to be sufficiently high for it to cover the fixed costs, now borne by the outside industry, to supply
the product/service. This price increase effect again benefits both firms 1 and 2. In addition, outsourcing again leads to the

11
As mentioned earlier, given that the outside supplier market has been modeled in our paper as a perfectly competitive one (a commodity market), it is reasonable to
assume that contract prices w1 and w2 are public knowledge and observable by rival firms.
12
The magnitude of price increase resulting from increase in marginal cost, caused by outsourcing, depends on the importance of outside good. The lesser important
the outside good is, the more firms can increase their prices in response to increase in their marginal costs. For example, under the market covered assumption, firms can
increase price more than they can under a linear demand function which has an outside good present (e.g., the model in Section 4). However, in both cases, the qual-
itative nature of the result is the same – increase in marginal cost from outsourcing increasing the price.
258 Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264

Fig. 1. Outsourcing equilibria and social welfare (for t = 1).

competitive cost disadvantage effect for firm 2 when it outsources. When c is small, the beneficial price increase effect of
outsourcing dominates the competitive cost disadvantage effect, and firm 2 is better off with outsourcing to the outside industry.
When c is not small, the competitive cost disadvantage effect dominates the beneficial price increase effect, and firm 2 is worse off
with outsourcing and thus does not outsource.
Interestingly, for ex-ante symmetric firms 1 and 2, their outsourcing strategy can be ex-post asymmetric when c takes an
intermediate value. The intuition is as follows. Recall that when only firm 1 outsources, firm 2 benefits due to both the price
increase effect and the competitive cost disadvantage effect. Therefore, the incentive of outsourcing by firm 2 is reduced since
price increase has already occurred. That is, the incremental benefit of price increase from outsourcing by firm 2 is reduced. In
this case, firm 2 is less likely to outsource when firm 1 outsources, compared to the case that firm 1 does not outsource.
Overall, Fig. 1 illustrates our results about both firms finding it optimal to outsource when the outside industry has a cost
advantage (c b 0), and even when the outside industry has no cost advantage (c ≥ 0); an asymmetric equilibria with only one
firm outsourcing for intermediate values of c; and obviously neither firm outsourcing when the outside industry has a large
cost disadvantage, i.e., c is very high.
Is there a first-mover advantage in asymmetric equilibria? We showed above the existence of asymmetric equilibria where
one firm outsources while the competing one does not. This naturally raises the question whether, in this asymmetric equilibria,
the firm making the outsourcing decision first (first mover) would be better off than the firm making this decision second (second
mover). Intuitively, in the parametric space where asymmetric equilibria exist, if a firm makes its outsourcing decision first, then it
knows that if it chooses to outsource, then the competing firm will choose to not outsource; conversely, if the firm making its
outsourcing decision first chooses to not outsource, then its competing firm will outsource.13 Given that firms in the asymmetric
sourcing equilibrium make different profits,14 there is a first-mover advantage since, depending on the cost parameters f and c, the
first-mover can choose whether to outsource or not—knowing that the competing firm will take the opposite decision. Of course,
in the parametric space described earlier where both firms choose to outsource or both choose to not outsource, there is no first-
mover advantage.

3. Welfare implications of outsourcing

We provide a discussion of the welfare effects next, relegating all the proofs in this section to Appendix A.
Consumer utility specification, V − p − tx, implies that consumer welfare is affected by both (i) the price p paid by the
consumers and (ii) the mismatch between consumers' ideal points and the specification of the product they choose (−tx).

3.1. Price

Outsourcing converts fixed costs into variable costs for firms in our framework. Since fixed costs do not affect optimal prices
but variable costs do, a direct effect of outsourcing in our model is a change in consumer price.
We showed earlier that firms in our model may outsource even when the outside industry has the same or higher marginal
cost of production than the outsourcing firms. In these cases, outsourcing causes consumer price to rise. This, in combination
with the assumption of fixed market size, implies that consumer welfare reduces when firms outsource in these cases.15
Of course, if the outside industry has a large advantage in marginal cost of production, then outsourcing can lower consumer
prices, improving consumer welfare.

13
The perfect competition assumption for the outside industry in our setup implies that the contract price wi when only firm i outsources (and similarly contract
prices w1 and w2 when both firms outsource) remains the same whether firms make outsourcing decisions simultaneously or sequentially.
14
Comparing the profits of firm 1 and firm 2 when firm 1 outsources to the outside industry in Section 2.2, we can check that firm 1 gets higher (lower) profit than
firm 2 if the fixed cost f is higher (lower) than (6t+2c)/72t.
15
If the market size is allowed to vary with prices, then increased price from outsourcing can reduce market size, thus moderating the loss in consumer welfare.
Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264 259

3.2. Product mismatch

There is an interesting adverse impact on consumer welfare when only one firm outsources. Specifically, when only one firm 1
outsources in equilibrium, outsourcing leads to more product mismatching disutility for consumers. To see this, note when neither
^x
firm outsources, the indifferent consumer is located at ^x ¼ 21 and the average product mismatching disutility is ∫ 0 ð−txÞdx þ ∫ ^x ð−tð1
1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
−xÞÞdx ¼ 4t . When only firm 1 outsources, the indifferent consumers are now at ^x ¼ 3t−cþα
12t , where α ¼ c2 þ 9t 2 −6ct−24tf ; hence,
^x 1 f −αÞ t
the average product mismatching disutility now becomes ∫ 0 ð−txÞdx þ ∫ ^x ð−tð1−xÞÞdx ¼ cðc−αÞþ3tð9t−4
72t N 4. Therefore, consumers
on average incur higher product mismatching cost when only one firm outsources.
Overall, unless the outside industry has a large advantage in marginal cost of production over the outsourcing firms, consumer
welfare decreases in outsourcing owing to the price and the product mismatch effects.
To consider effect on social welfare, we also need to look at the change in firms' profits. Since firms' profits increase when they
find it optimal to use outsourcing, they provide a counterforce to the decrease in consumer welfare arising from outsourcing. In
our model, the loss in consumer welfare exceeds the gain in firms' profits when firms outsource, unless the outside industry
has a large advantage in marginal cost over the outsourcing firms. Interestingly, consumer and social welfare can still decrease
even when the outside industry has an intermediate level of advantage in production costs over the outsourcing firms and
hence when society has lowered its cost of production; both price and product mismatching effects play a role in these results.
Fig. 1 illustrates these results on social welfare for t= 1.

4. An alternative formulation

In the last section, we illustrated the effects of outsourcing by employing the commonly used Hotelling model. One feature of
that model is that each consumer has a unit demand and hence either buys zero or one unit of product. To focus on the effect of
competition, we had further assumed that the market is fully covered, i.e., the total market size is fixed. So, for example, when
firms 1 and 2 both lower their product prices, the combined market does not expand in that model.
The goal of this section is to use a linear demand function and show the robustness of our two key results—namely, that even if
the outside industry does not have a production-cost advantage over the two competing focal firms, we can have an equilibrium
where both focal firms outsource to convert fixed cost into variable costs, and also an asymmetric equilibrium where one focal
firm outsources to convert fixed cost into variable cost while the ex-ante symmetric focal firm chooses to incur the fixed cost
by not outsourcing.
As is well known, this linear demand function does not have the unitary demand property of the commonly-used Hotelling
model, and also allows total market size to change with product prices.
Specifically, let the demand for the two products charging prices p1 and p2 be

q1 ¼ 1−p1 þ δp2 ; q2 ¼ 1−p2 þ δp1 ; ð13Þ

where parameter δ ∈(0,1) represents the degree of product substitutability; the larger the value of δ, the more substitutable the
products are and the more the competitive price affects the demand of a product.
We again first analyze the case where neither firm outsources. In this case, firms 1 and 2 decide their respective prices to max-
imize their profits given by

Π1 ¼ p1 ð1−p1 þ δp2 Þ−f ; Π2 ¼ p2 ð1−p2 þ δp1 Þ−f : ð14Þ

Solving, we get

  1   1
p1 ¼ p2 ¼ ; Π ¼ Π2 ¼ −f : ð15Þ
2−δ 1 ð2−δÞ2

Next, we analyze the case where firm 1 outsources and firm 2 does not. We again analyze the model following backward
induction, and first derive the optimal product prices p1 and p2 given the contract price w1 offered by firm 1 to the outside
industry. Firms' profits are given by

Π1 ¼ ðp1 −w1 Þð1−p1 þ δp2 Þ; Π2 ¼ p2 ð1−p2 þ δp1 Þ−f : ð16Þ

Maximizing firms 1 and 2's profits, we have

2 þ δ þ 2w1 2 þ δ þ δw1
^1 ¼
p ^2 ¼
;p : ð17Þ
4−δ2 4−δ2

^1 and p
Therefore, as in the Hotelling model in the last section, both the product prices p ^2 increase with the contract price w1
offered by firm 1 to the outside industry. For the perfectly competitive outside industry to supply to firm 1 at price w1, it must be
260 Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264

^1 ; p
that ðw1 −cÞq1 ðp ^2 Þ− f ≥0. Maximizing firms' profits given this constraint on an acceptable contract to the outside industry, we
get

 c 2 þ δ−Ω
w1 ¼ þ  ; ð18Þ
2 2 2−δ2

 6 þ 2c þ 3δ−ð2 þ cÞδ2 −δ3 −Ω  8 þ ð6 þ 2c−ΩÞδ−3δ2 −ð2 þ cÞδ3


p1 ¼    ; p2 ¼    ; ð19Þ
2−δ2 4−δ2 2 2−δ2 4−δ2

h   i2 h n   oi2

2 þ δ−c 2−δ2 þ Ω 
8−δ −6 þ δð3 þ 2δÞ−c 2−δ2 þ Ω
Π1 ¼  2 ; Π2 ¼  2  2 −f ; ð20Þ
4 4−δ2 4 2−δ2 4−δ2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
where Ω ¼ ð2 þ δ þ cð2−δ2 ÞÞ −4ð2−δ2 Þðcð2 þ δÞ þ ð4−δ2 Þf Þ:
We next compare the profits of firms 1 and 2 before and after firm 1 outsources to outside industry, and summarize the main
results in the following proposition.

Proposition 3. When firm 2 does not outsource, then firm 1 can benefit from outsourcing even if the outside industry does not have a
cost advantage (c≥0). Firm 2 is better off with firm 1's outsourcing unless the outside industry has a large cost advantage over firms 1
and 2.

Proof. See Appendix A.

Proposition 3 confirms the findings in the base Hotelling model. The intuition for all the results remain the same as given in
Proposition 1.
Finally, we analyze the case where both firms 1 and 2 outsource. Following the same three-stage game structure as in the last
section, we get
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
1 þ cð1−δÞ− ð1−cð1−δÞÞ2 −4ð2−δÞð1−δÞ f
w1 ¼ w2 ¼ ; ð21Þ
2ð1−δÞ

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
3−2δ þ cð1−δÞ− ð1−cð1−δÞÞ2 −4ð2−δÞð1−δÞf
p1 ¼ p2 ¼ ; ð22Þ
2ð2−δÞð1−δÞ

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
1−cð1−δÞ þ ð1−cð1−δÞÞ2 −4ð2−δÞð1−δÞf
 
Π1 ¼ Π2 ¼ : ð23Þ
4ð2−δÞ2

Comparing these profits to the profits firms get when neither outsources (Eq. (15)) and when only firm 1 outsources
(Eq. (20)), we get the following main results.

Proposition 4. Even when the outside industry does not have a cost advantage (c≥0), (i) both firms may choose outsourcing, (ii) one
firm may choose outsourcing and the other one not.

Proof. See Appendix A.

Proposition 4 replicates the main results obtained from the Hotelling model, and the rationale remains the same as given in the
base model.

5. Concluding remarks

Whether to produce products or services in-house or outsource these tasks to an outside manufacturing/service industry is an
important strategic decision for firms. This decision has significant implications for competing firms, consumers, and the society.
The literature has examined a number of cost-savings and strategic reasons that firms ought to consider in making this decision.
This paper focused on one particular reason for outsourcing mentioned in the popular press and by practitioners: reducing fixed
cost and converting it into variable cost.
We showed a number of ways in which this economic role of outsourcing affects strategic price competition among competing firms
choosing whether to outsource. For example, we showed how it can tilt the playing field against a firm if it is the only one doing it, and at
Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264 261

the same time can benefit it indirectly through a price increase in the marketplace. Hence, even if the outside industry is more cost-
efficient, competing firms may choose to not outsource to it; or vice versa, one or both competing firms may choose outsourcing even
if the outside industry is less cost-efficient than the outsourcing firms. We also showed how symmetric firms may adopt asymmetric
outsourcing strategies, with one firm choosing to outsource and the competing one to not outsource. We also discussed how the
fixed-cost-to-variable-cost-conversion role of outsourcing adversely affects consumer and social welfare.
Of course, in reality, firms choose outsourcing due to multiple reasons and outsourcing decisions lead to multiple tactical and
strategic effects. Thus, managerial and public policy actions must consider a comprehensive and integrative framework that
includes factors shown in this and prior work in the area of outsourcing.
The area of outsourcing offers many other avenues for research. For example, one can consider a longer term game in which
fixed costs also affect firm entry decisions. Our model starts with a setting where two competing firms already exist, and are
considering whether to outsource to convert fixed cost into variable cost. If we were to add to this setup an earlier stage where
firms choose whether to enter the market or not, then the fixed costs will also matter in whether the expected profit is large
enough to make it profitable for both firms to enter the market. To that extent, fixed costs also matter in entry decisions. However,
once the two firms are in the market, then our analysis will hold and the fixed costs will have the same impact on outsourcing
decisions as shown in our analysis. Outsourcing decisions also affect and interact with other important variables such as product
quality, flexibility, time-to-market, and responsiveness to changing market conditions, which can be examined in future studies.
Future research can also differentiate between inshore versus offshore outsourcing, with offshoring involved with issues of
taxes, tariffs, and government regulations; in addition, in the case of offshore outsourcing, each government may only care
about the consumer welfare and social welfare of its country. Overall, outsourcing is an economically significant practice, and
deserves more attention from economics, strategy, and marketing scholars.

Appendix A

Proof of Proposition 1. Let ΠON NN


1 be firm 1's profit when firm 1 outsources and firm 2 does not, and Π1 be firm 1's profit when
both firms do not outsource. Using (2) and (8), we have
 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2
3t−c þ 9t 2 þ c2 −6ct−24tf  
ON NN t
ΔΠ ¼ Π1 −Π1 ¼ − −f :
72t 2

Solving ΔΠN 0 in terms of c, we get


sffiffiffiffiffiffiffiffiffiffiffiffi
t
c b c1 ¼ 3t−ð3t−4f Þ :
t−2 f

Note that c1 N 0 if f b 3t
8 ; c1 =0 if f ¼ 8 ; and c1 b 0 if f N 8 .
3t 3t

Next, define ΠON2 be firm 2's profit when firm 1 outsources and firm 2 does not, and ΠNN
2 be firm 2's profit when both firms do
not outsource. Using (2) and (8), we have
0 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 1
 
B 9t þ c− 9t þ c −6ct−24tf
2 2
ON NN C t
ΔΠ ¼ Π2 −Π2 ¼@ −f A− −f :
72t 2

Solving ΔΠN 0 in terms of c, we get c b −2f. For c ≥0, we always have ΔΠb 0.

Proof of Proposition 2. Define ΠON OO


2 as firm 2's profit when firm 1 outsources and firm 2 does not, and Π2 as firm 2's profit when
both firms outsource. Using (8) and (12), we get

0 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 1
t B 9t þ c− 9t þ c −6ct−24tf
2 2
OO ON C
ΔΠ ¼ Π2 −Π2 ¼ −@ −f A:
2 72t

Solving ΔΠN 0, we must have

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
9t 2 −2tf þ ð9t−8 f Þ t ðt þ 2 f Þ
c b c2 ¼ :
3t−2f
262 Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264

Noting that c2 b c1 (defined above in Proposition 1), and combining along with the results from the Proposition 1, we can con-
clude that (i) both firms choose outsourcing when c b c2, and (ii) one firm chooses outsourcing and the other one chooses in-house
production when c2 b c b c1. Also, a necessary condition for c2 N 0 is f b 32
9t
.

Proof of Results in Section 3. (Welfare implications of outsourcing)

First, when neither firm outsources its production, consumers who are indifferent between buying from firms 1 and 2 are lo-
cated at ^x ¼ 12. Therefore, consumers located in the interval x ∈ ½0; 12 buy from firm 1 and enjoy a surplus of V − tx− p1. Consumers
located in the interval x ∈ ½12 ; 1 buy from firm 2 and enjoy a surplus of V − t(1− x)− p2. The total consumer surplus is calculated as
follows:
Z ^x Z 1
5t
CW 1 ¼ ðV−tx−p1 Þdx þ ðV−t ð1−xÞ−p2 Þdx ¼ V− : ð24Þ
0 ^x 4

Social welfare (SW1) is calculated by adding up firms' profits and consumer welfare. We have

t
SW 1 ¼ V− −2f : ð25Þ
4
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
3t−cþ c2 þ9t 2 −6ct−24tf
Second, when firm 1 outsources, the indifferent consumers is located at ^x ¼ 12t . Define α ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
c2 þ 9t 2 −6ct−24tf . Then, consumers located in the interval x ∈ ½0; 3t−cþα
12t  buy from firm 1 and enjoy a surplus of V − tx − p1.
Consumers located in the interval x ∈ ½3t−cþα
12t ; 1 buy from firm 2 and enjoy a surplus of V −t(1 −x)− p2. The total consumer sur-
plus is given by
Z ^x Z 1 cð18t−c þ α Þ þ 3t ð4f þ 45t−5α Þ
CW 2 ¼ ðV−tx−p1 Þdx þ ðV−t ð1−xÞ−p2 Þdx ¼ V− ð26Þ
0 ^x 72t

Social welfare is calculated as

cð18t−5c þ 5α Þ þ 3t ð44f þ 9t−α Þ


SW 2 ¼ V− : ð27Þ
72t

Third, when both firms 1 and 2 outsource, the indifferent consumer is again located at 12. Following the same procedure, we
have

5t t
CW 3 ¼ V− −c−2 f ; SW 3 ¼ V− −c−2f : ð28Þ
4 4

Note that CW3 − CW1 = −c − 2f b 0. Thus, consumer welfare reduces when both firms outsource compared to the case where
neither firm outsources.
Next, note that CW2 − CW1 b 0 if c N − 2f. Thus, consumer welfare reduces when one firm outsources compared to the case
where neither firm outsources unless the outside industry has a large marginal cost advantage.
We note that SW3 − SW1 = −c b 0. Thus, social welfare reduces when both firms outsource compared to the case where neither
firm outsources. pffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffi
9t 2 −2tf þð9t−8 f Þ tðtþ2 f Þ
Finally, when c2 b c b c1, i.e., 3t−2 f
b c b 3t−ð3t−4f Þ t−2 t
f
, we compare SW2 and SW1. We get
   
cð18t−5c þ 5α Þ þ 3t ð44f þ 9t−α Þ t
SW 2 −SW 1 ¼ V− − V− −2 f :
72t 4

Solving for SW2 − SW1 N 0, we must have


pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
9t 2 −16tf −ð3t−2f Þ t ð9t−20f Þ
cb :
5t−20 f
Note that we always have

2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
9t −16tf −ð3t−2f Þ t ð9t−20f Þ 9t −2tf þ ð9t−8f Þ t ðt þ 2f Þ
b :
5t−20 f 3t−2f
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
9t 2 −16tf −ð3t−2 f Þ tð9t−20 f Þ
Therefore, SW2 − SW1 N 0 only when c b minf0; 5t−20 f
g:
Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264 263

Proof of Proposition 3. Let ΠON NN


1 be firm 1's profit when firm 1 outsources and firm 2 does not, and Π1 be firm 1's profit when
ON NN
both firms do not outsource. The expressions of Π1 and Π1 were derived in (15) and (20). Then,
h   i2
2 þ δ−c 2−δ2 þ Ω  
ON NN 1
ΔΠ ¼ Π1 −Π1 ¼   − −f ;
4 4−δ2 2 ð2−δÞ2

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
where Ω ¼ ð2 þ δ þ cð2−δ2 ÞÞ −4ð2−δ2 Þðcð2 þ δÞ þ ð4−δ2 Þf Þ. Solving ΔΠN 0 in terms of c, we get

2þδ 2 þ δ−2ð2−δÞf
c b c3 ¼ − qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi :
2−δ2 2−δ2  1−2−δ2  f

Note that
pffiffiffiffiffiffiffiffiffiffi
pffiffiffiffiffiffiffiffiffiffiffiffiffi ð3−2f Þ 1−f
Lim c3 ¼ 1− 1−2f N0 and Lim c3 ¼ 3− N0
δ→0 δ→1 1−f

for the feasible range of f. Thus, c3 can take positive value for δ∈ (0, 1).
Next, define ΠON NN
2 be firm 2's profit when firm 1 outsources and firm 2 does not, and Π2 be firm 2's profit when both firms do
not outsource. Using (15) and (20), we have
0hn   oi2 1
8−δ −6 þ δ ð3 þ 2δ Þ−c 2−δ
2
þ Ω  
ON NN B C 1
ΔΠ ¼ Π2 −Π2 ¼@     −f A− − f :
4 2−δ2 4−δ2
2 2
ð2−δÞ2

Solving ΔΠN 0 in terms of c, we get c b −(2− δ)f b 0.

Proof of Proposition 4. Define ΠON OO


2 as firm 2's profit when firm 1 outsources and firm 2 does not, and Π2 as firm 2's profit when
both firms outsource. The expressions of ΠON
2 and Π OO
2 were derived in (20) and (23). Then, we need to solve for ΔΠ(c, f, δ) =
ΠOO ON
2 − Π2 N 0.

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 0h 1
n   oi2
1−cð1−δÞ þ ð1−cð1−δÞÞ2 −4ð2−δÞð1−δÞ f 2
B 8−δ −6 þ δð3 þ 2δÞ−c 2−δ þ Ω C
ΔΠðc; f ; δÞ ¼   −@  2  2 −f A:
4 2−δ2 4 2−δ2 4−δ2

We note that there is no closed-form explicit solution for ΔΠ(c, f, δ) = 0 in terms of parameter c. So, we use Taylor series
expansion of ΔΠ(c, f, δ) to get a linear approximation in the neighborhood of c =0, which is as follows
 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 h n qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   ffioi2
  1 þ 1−4ð1−δÞð2−δÞ f −8 þ δ −6 þ δð3 þ 2δÞ þ ð2 þ δÞ 2 þ δ−4ð2−δÞ 2−δ2 f

ΔΠðc ¼ 0; f ; δÞ þ ðc−0Þ ΔΠðc; f ; δÞ j c¼0 ¼ f þ −  2  2
∂c 4ð2−δÞ2 4 2−δ2 4−δ2
3
"
2
c 2δð1 þ δÞ 2ð1−δÞf1−2ð2−δÞð1−δÞ f g 2δð2 þ δÞ þ 4ð2−δÞδ f 7
− 2− þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi þ: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   ffi5:
2ð2−δÞ2 ð2 þ δÞ 1−4ð2−δÞð1−δÞ f ð2 þ δÞ ð2 þ δÞ 2 þ δ−4ð2−δÞ 2−δ2 f

Solving this (i.e., ΔΠ(c,f,δ) =0) gives us

2  pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 h n    ffioi2 3
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 þ 1−4ð1−δÞð2−δÞf −8 þ δ −6 þ δð3 þ 2δÞ þ ð2 þ δÞ 2 þ δ−4ð2−δÞ 2−δ2 f
26 7
2ð2−δÞ 4 f þ −  2  2 5
4ð2−δÞ2 4 2−δ2 4−δ2
c ¼ c4 ¼ 2 3
2
6 2δð1 þ δÞ 2ð1−δÞf1−2ð2−δÞð1−δÞf g 2δð2 þ δÞ þ 4ð2−δÞδ f 7
42− þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi þ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
   5
ð2 þ δÞ 1−4ð2−δÞð1−δÞf ð2 þ δÞ ð2 þ δÞ 2 þ δ−4ð2−δÞ 2−δ2 f
We note that
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi! rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 pffiffiffiffiffiffiffiffiffiffiffiffiffipffiffiffiffiffiffiffiffiffiffiffiffiffi 4 4
−3 1− f ð9−9 1− f −8 f
− 1−4f þ 1−8 f 1−8f 3 3
Lim c4 ¼  pffiffiffiffiffiffiffiffiffiffiffiffiffi2 b 0 and Lim c4 ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi! N0
δ→0 δ→1 4
1 þ 1−8f 6þ6 1− f þ 4f
3
264 Y. Liu, R.K. Tyagi / International Journal of Research in Marketing 34 (2017) 252–264

for the feasible range of f. Thus, c4 can take positive value for δ∈ (0, 1).
Therefore, along with the results from the Proposition 3, we can conclude that (i) both firms choose outsourcing when c b c4,
and (ii) one firm chooses outsourcing and the other one in-house production when c4 b c b c3.

References

Abraham, K. G., & Taylor, S. K. (1996). Firms' use of outside contractors: Theory and evidence. Journal of Labor Economics, 14(3), 394–424.
ADP Total Source (2012). Outsourcing and the future of HR. https://www.adp.com/126/media/White%20Papers/ReportOutsourcing%20%20the%20Future%20of%
20HRfinal.ashx
Anderson, E. (1985). The salesperson as outside agent or employee: A transaction cost analysis. Marketing Science, 4(3), 234–254.
Arnold, U. (2000). New dimensions of outsourcing: A combination of transaction cost economics and the core competencies concept. European Journal of Purchasing &
Supply Management, 6(1), 23–29.
Arya, A., Mittendorf, B., & Sappington, D. E. M. (2008). The make-or-buy decision in the presence of a rival. Management Science, 54(10), 1747–1758.
Barry-Wehmiller International Resources (2012). Engineering services outsourcing—A research on drivers and trends. http://www.barry-wehmillerinternational.com/
docs/bwir/bwir-engineering-outsourcing-trends-whitepaper7E53713E1606.pdf
Cachon, G. P., & Harker, P. T. (2002). Competition and outsourcing with scale economies. Management Science, 48(10), 1314–1333.
Capgemini Internal Audit Services Survey (2009). Balancing budget pressures and increased enterprise risk. https://www.nl.capgemini.com/resource-file-access/
resource/pdf/b_Capgemini_Internal_Audit_Services.pdf
Chalos, P., & Sung, J. (1998). Outsourcing decisions and managerial incentives. Decision Sciences, 29(4), 901–919.
Choi, S. C. (1991). Price competition in a channel structure with a common retailer. Marketing Science, 10(4), 271–296.
Communication from the Commission to the European Parliament, The Council, & The European Economic and Social Committee and the Committee of the Regions
(2012n). Unleashing the potential of cloud computing in Europe. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2012:0529:FIN:EN:PDF
Coughlan, A., & Wernerfelt, B. (1989). On credible delegation by oligopolists: A discussion of distribution channel management. Management Science, 35(2), 226–239.
Deloitte's 2014 global outsourcing and insourcing survey (2014). http://www2.deloitte.com/content/dam/Deloitte/us/Documents/strategy/us-2014-global-
outsourcing-insourcing-survey-report-123114.pdf
Ernst, & Young (2013). Outsourcing in Europe—An in-depth review of drivers, risks and trends in the European outsourcing market. http://www.ey.com/Publication/
vwLUAssets/Outsourcing_in_Europe_2013/$FILE/EY-outsourcing-survey.pdf
Franco, C. (2015). Fab or fabless: An overview of the semiconductor debate. https://www.linkedin.com/pulse/fab-fabless-overview-semiconductor-debate-
christopher-franco
Gilbert, S. M., Xia, Y., & Yu, G. (2006). Strategic outsourcing for competing OEMs that face cost reduction opportunities. IEEE Transactions, 38(11), 903–915.
Grossman, G. M., & Helpman, E. (2002). Integration versus outsourcing in industry equilibrium. Quarterly Journal of Economics, 117, 85–120.
Horsky, S. (2006). The changing architecture of advertising agencies. Marketing Science, 25(4), 367–383.
Kakabadse, A., & Kakabadse, N. (2000). Sourcing: New face to economies of scale and the emergence of new organizational forms. Knowledge and Process Management,
7(2), 107–108.
Kelleher, J. (1990). The dollars and sense of outsourcing: Sometimes a great notion. Computerworld, 24(2), 75–77.
Kremic, T., Tukel, O. I., & Rom, W. O. (2006). Outsourcing decision support: A survey of benefits, risks, and decision factors. Supply Chain Management, 11(6), 467–482.
Kuhn, K. (1997). Nonlinear pricing in vertically related duopolies. RAND Journal of Economics, 28(1), 37–62.
Laarhoven, P. V., Berglund, M., & Peters, M. (2000). Third-party logistics in Europe—Five years later. International Journal of Physical Distribution and Logistics
Management, 30(5), 425–442.
Lau, C., & Zhang, J. (2006). Drivers and obstacles of outsourcing practices in China. International Journal of Physical Distribution and Logistics Management, 36(10),
776–792.
Mani, D., Barua, A., & Whinston, A. B. (2013). Outsourcing contracts and equity prices. Information Systems Research, 24(4), 1028–1049.
McGuire, T. W., & Staelin, R. (1983). An industry equilibrium analysis of downstream vertical integration. Marketing Science, 2, 161–190.
Monczka, R. M., Blascovich, J. D., & Carter, J. R. (2004). Strategic outsourcing: Emerging trends and success factors. AT Kearney.
Monczka, R. M., Markham, W. J., Carter, J. R., Blascovich, J. D., & Slaight, T. H. (2005). Outsourcing strategically for sustainable competitive advantage. CAPS Research and A.
T. Kearney Inc.
Mourdoukoutas, P. (2013, April 14). How Hewlett-Packard and Dell destroyed their PC advantage… Piece-by-piece. Forbes.
O'Brien, D. P., & Shaffer, G. (1992). Vertical contracts with bilateral contracts. RAND Journal of Economics, 3(3), 299–308.
Outsourcing—Your Voice CMR (2013). http://yourvoice.co/outsourcing.html
Qu, Z., & Brocklehurst, M. (2004). What will it take for China to become a competitive force in offshore outsourcing? An analysis of the role of transaction costs in
supplier selection. Journal of Information Technology, 18(1), 53–67.
Quinn, J. B., & Hillmer, F. G. (1995). Strategic outsourcing. The McKinsey Quarterly, 1.
Razzaque, M. A., & Sheng, C. (1998). Outsourcing of logistics functions: A literature review. International Journal of Physical Distribution and Logistics Management, 28(2),
89–107.
Regus (2010, March 17). Smart small business: Turning fixed costs into variable costs. http://www.prnewswire.com/news-releases/smart-small-business-turning-
fixed-costs-into-variable-costs-88158887.html
The Small Business Authority, "3 Ways to Reduce Fixed Costs," Blog. Posted on September 25, 2012 http://www.thesba.com/2012/09/25/3-ways-to-reduce-fixed-
costs/.
Stremersch, S., Weiss, A., Dellaert, B., & Frambach, R. T. (2003). Buying modular systems in technology-intensive markets. Journal of Marketing Research, 40(August),
335–350.
Sutton, J. (2012, February 22). 1 in 3 LCD TV production will be outsourced this year. http://www.hdtvtest.co.uk/news/lcd-tv-production-outsourced-201202221679.
htm
Tabuchi, H. (2011, December 26). Sony to cease its flat screen partnership with Samsung. New York Times.
Trivedi, M. (1998). Distribution channels: An extension of exclusive retailership. Management Science, 44(July), 896–909.
Vining, A., & Globerman, S. (1999). A conceptual framework for understanding the outsourcing decision. European Management Journal, 17(6), 645–754.