Foreign operation mode flexibility: tradeoffs and managerial responses

Abstract

Firms’ ability to change foreign operation modes appears highly desirable in an increasingly volatile and unpredictable global environment. We propose and discuss mode flexibility as a management capability, with the aim at curbing the potential downsides of flexibility; in particular, the extra costs of coordination and contracting as well as revenue losses due to diminished partner commitment. We model the balancing and shifting of essential tradeoffs in relation to the two dimensions of mode flexibility – multiplicity and switchability – and highlight modularization and reciprocal use of real options as examples of tradeoff-shifting mechanisms that may improve the cost-benefit balance of mode flexibility.

Keywords:

• Foreign operation mode
• flexibility
• tradeoffs
• real options

JEL CLASSIFICATION:

• F23
• L24
• M16

Acknowledgement

We thank the editor David Paton for his guidance and an anonymous reviewer for constructive comments. This article has evolved from earlier versions presented at the Strategic Management Society Special Conferences in Banff (2017) and Oslo (2018), the 2018 European International Business Academy Annual Conference (Poznan), and the 2019 Academy of International Business Annual Conference (Copenhagen) where it got the FIU/AIB Best Theory Paper Award. We are grateful for valuable feedback received at these conferences as well as at a research seminar at BI Norwegian Business School.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We adopt Thompson’s classic distinction between three basic types of interdependencies (Thompson 1967): pooled, sequential, and reciprocal. Pooled (or modular) interdependency is associated with the lowest coordination costs. The various organizational units (in casu, operation modes) provide inputs to a central unit that coordinates and reallocates the pool of inputs. The coordination of inputs and related activities takes place on a bilateral basis between the central and affiliated units. Hence, the central unit administrating the resource pool guides the other units as to what to deliver to the central pool. The key difference between pooled and sequential interdependence is that in the latter case the coordinating unit not only has to coordinate what the other units have to deliver, the unit also has to coordinate when each unit has to deliver inputs\resources and to whom. This implies extra coordination costs. The importance of timing of inter-firm delivery resonates with the transaction cost theory concepts of “temporal specificity” (Masten, Meehan, and Snyder 1991) or “time specificity” (Malone, Yates, and Benjamin 1987), where an asset is time-specific if its value is highly dependent on its reaching the user within a specified time-period. Reciprocal interdependence implies that each unit coordinates with all other units in the value chain. Moreover, the coordination among the units is done in a simultaneous way given the time specificity. In other words, the units are integrated but with no central, coordinating unit in the foreign market. The units coordinate bilaterally. This type of interdependence is cost-sensitive to the number of units (in casu, operation modes). Whereas pooled and sequential interdependencies only experience linearly and monotonically increasing coordination costs when new units are added, coordination costs increase exponentially.

2 We exclude the highly unlikely situation where a licensor or franchisor transfers its international trademarks and other copyrights to the local licensee or franchisee, see also Lafontaine and Bhattacharyya (1995).

3 We are indebted to an unknown referee for pointing this out; but we also notice that the findings of studies of risk preferences of entrepreneurs (e.g., Brockhaus 1980; Forlani and Mullins 2000; Ketchen, Short, and Combs 2011; Vereshchagina and Hopenhayn 2009) are very mixed which makes it difficult to establish whether or not entrepreneurs at large are more risk-willing than business people in general.

4 As an example, Martin (1988, 954) challenges the view that franchisees in general are the more risk averse vis-à-vis the franchisor because the latter has a more diversified business portfolio: “It has been observed by Rubin (1978, 225–26) and others that the representative franchisee's investment is undiversified relative to the representative franchisor. From this it is concluded that the franchisee will be more risk averse than the franchisor. I suggest that the choice of a less diversified investment reflects less risk aversion on the part of franchisees. Recall that franchisees are drawn from the more entrepreneurial agents in our society, since they chose to start their own business rather than seek employment.” [Emphasis (italics) added by the authors]. So even though the general contention found in the literature on franchising is that franchisees are the more risk averse party in these arrangements there are indeed proponents of opposite views such as the one aired by Martin (1988). For an overview of this academic controversy, see for example Lafontaine and Bhattacharyya (1995).

5 In the same vein, one may think of entrant MNCs that are geographically diversified but at the same time are characterized by delegating decision rights in a very decentralized manner (e.g., Robinson and Stocken 2013). Hence, the MNC-managers making entry mode decisions are disinclined to take into account the risk appetite prescribed by the business diversification of the wider corporation.

6 Specific details about the call and put options, including the transfer fees (calculated as multiples), are held strictly confidential by the company and are only available to approved joint-venture partners. In October 2018, we approached the company’s board of directors (the parent company is Zebra A/S) and asked for access to this information but received a polite refusal.

7 While we have only come across one example of reciprocal use of real options, we know of several practical examples of non-reciprocal use of call options described in the literature (e.g., Benito et al. 2013; Jensen and Petersen 2013a, 2013b) in addition to the Joe & The Juice example mentioned in the introduction.

8 Even though a call option in itself seems unsuitable as an instrument for achieving mode flexibility, there might be other good reasons to use a call option without necessarily combining it with a put option. One argument for using a call option as a substitute for an irrevocable contract with a fixed length relates to the free riding risk associated with contract expiry. A call option may eliminate this risk. It is well described in the literature (e.g. Brickley and Dark 1987; Brickley, Dark, and Weisbach 1991; Ghemawat 1991; Lafontaine and Bhattacharyya 1995) how economic agents are inclined to free ride as a contract comes closer to expiration and has to be either terminated or renegotiated. If market prospects turn out to be better than expected the principal/contractor will be tempted to either end the collaboration and take over the territory, or renegotiate the contract and take the lion’s share of any future quasi rent or windfall gain. In anticipation of this adverse outcome the agent (in casu the local contractual partner) is enticed to appropriate as much market value as possible during the time up to contract expiry, e.g. by compromising on service quality and/or postponing investments in maintenance and marketing – even though this effectively means free riding on the reputation of the contractor. Alternatively, the agent may deliberately underperform in order to make the local market look less lucrative and therefore less attractive for the principal to take over (Ellis 2005; Petersen et al. 2006). Such conduct is probably at least as harmful to the contractor by taking on the characteristics of a Nash equilibrium (e.g. Fudenberg and Tirole 1991). The conclusion is that a call option eliminates this growing free riding risk towards contract expiry inasmuch as the transfer fee at any time compensates the agent adequately for exercising the call option.

9 Because of the franchisee’s 2-year grace period assumed in this numerical example, a call option only allows the franchisor to exercise its call option from the third year and get half of the earnings in the lucrative scenario the next eight years. The franchisor therefore gets less than half of the gross earnings during the ten years that the contract runs. This is why the dotted vertical line that in Figure 6 and 7 indicates the franchisor’s and franchisee’s sharing of the earnings intersects $ 3m and not $ 5m/2 = $ 2.5m. In the lucrative scenario the franchisee’s gross earnings are 2 @ $ 500,000 = $ 1m during the 2-year grace period and 8 @ $ 250,000 = $ 2m in the remaining 8 years of the contract. All in all $ 3m. The franchisor’s gross earnings when exercising the call option immediately after the grace period are 8 @ $ 250,000 = $ 2m.