FormalPara Definition

Vertical integration involves a single company having ownership and control of two or more stages of the supply chain, such as manufacturing and distribution, or components and assembly.

Vertical integration exists when any two stages of the input-to-end-user supply chain are brought under common ownership and control within a single organization. Often, the rationale for vertical integration is compensating for the absence of, or deficiencies in, market-based contracting. Managerial actions and commands inside the vertically integrated company are used to achieve the necessary coordination in place of arm’s-length contracts. In a sense, vertical integration results from the ‘make’ outcome of the make-or-buy decision; it often requires as much management focus as corporate diversification (Harrigan 1986). Vertical expansion can take the form of backward integration (e.g., Samsung designing the processors that go into its phones) or forward integration (e.g., Sony or Apple opening their own chain of retail stores).

In practice, vertical integration is not a binary choice. In many instances, firms can both make and buy the same input, or, in the downstream direction, own some retail outlets while also using independent distributors. In the early 1970s, the industrial organization literature referred to this as quasi-vertical integration (Blois 1972). Jacobides and Billinger (2006) call such hybrid outcomes a ‘permeable architecture’ and identify one of the benefits as the ability to benchmark internal activities against their external counterparts. The ability to optimally balance concurrent insourcing and outsourcing of a single good or service is an important aspect of the firm’s asset orchestration capability (Rothaermel et al. 2006).

Another type of quasi-integration occurs when a customer maintains ownership of specialized tooling used by a supplier (Monteverde and Teece 1982a). Apple, for example, owns no factories but spends billions of dollars each year on machines, often under exclusive licence, that are used by its suppliers to produce Apple products (Satariano 2013).

Explanatory Factors

The explanations advanced over time for why a firm would find it advantageous to integrate backwards or forwards have included market power, the need for technology integration, transactions cost and control rights. These have been developed and assessed in the economics literature and also in the strategic management literature. Other theories of vertical integration, particularly capabilities, have been explored primarily in the strategic management field.

In the market power case, a firm with some level of freedom from price competition might theoretically be able to increase (total) profits by investing in another stage of production for a number of possible reasons related to its pricing power (Perry 1989). The market power hypothesis has been of particular interest in the field of antitrust because the resulting integration would raise prices to customers without creating more economic value. Empirically, however, in most cases that have been studied, ‘efficiency considerations overwhelm anticompetitive motives’ for integration (Lafontaine and Slade 2007: 677).

A non-technological type of efficiency motive, the minimization of transaction costs (Williamson 1985), has become one of the dominant explanations for vertical integration. The focus in transaction cost explanations is on the risks associated with the possibility of opportunistic recontracting after one party has made transaction-specific investments. Investments at one stage of production that have limited alternative uses (‘asset specificity’) can be subject to a renegotiated contract (‘hold up’) by parties in the next stage if the two stages are under separate ownership.

The first empirical test of the transaction cost economics paradigm was by Monteverde and Teece (1982b), followed by Masten (1984). Both studies showed a statistically significant relationship between asset specificity and the choice of firm boundaries. Since then, the predictions of transaction cost economics for firm boundaries have been confirmed many times. The evidence is overwhelmingly supportive of an effect, particularly for backward integration, that is, buyer–supplier linkages (Lafontaine and Slade 2007: 658).

However, there is much more at work in outsourcing/insourcing decisions in addition to transaction costs. Property rights theorists (e.g., Grossman and Hart 1986) combine elements of the transaction cost and technological explanations in an explicitly contractual approach. In property rights models, the essential trade-off is between control and efficiency, under the assumption that management of a stage of production is most efficient when it operates independently. Vertical ownership may prove optimal when there is a high cost to specifying a complete contract. This approach has found the most empirical support in cases of forward integration, that is, manufacturer–retailer or franchisor–franchisee (Lafontaine and Slade 2007: 660). The framework has also been applied to R&D/innovation (Chesbrough and Teece 1996).

Another factor driving vertical integration that has not garnered much attention in the economics field is the absence of firms in the market with the requisite capabilities. For example, in the early stages of an industry’s evolution when certain inputs are not yet available in competitive supply, vertical integration may be necessary to assure the quality or quantity of supply (Langlois 1991). Business historian Alfred Chandler documented how Pabst Brewing, Singer Sewing Machine, McCormick Harvester and Ford invested in in-house supply or retail at a time when ‘the supply network was unable to provide the steady flow of a wide variety of new highly specialized goods essential to assure the cost advantages of scale’ (Chandler 1992: 89). The problem is not that markets had failed because of ‘transaction costs that can be attenuated by substituting internal organization for market exchange’ (Williamson 1971: 114), but rather that no ‘market’ able to provide the requisite supplies in the necessary quantity had yet come into existence. Such industries are likely to dis-integrate as the supply base matures and/or the links between stages of the supply chain become better defined, enabling product (and organizational) modularity (Stigler 1951; Langlois 2002).

It is not generally claimed that capabilities, by themselves, explain integration decisions. More typical are claims that capabilities are a necessary complement to transaction costs for fully explaining the decision. This is hardly surprising, because transaction cost economics generally abstracts from production-related considerations, but vertical integration is more likely in cases where relevant internal capabilities are already present (Masten et al. 1991). In the computer industry, IBM in the 1990s demonstrated the benefits of bucking the then-current trend towards dis-integration by building on the firm’s well-established capabilities to provide complete hardware and service solutions, even if this meant supporting competitors’ hardware in some cases (Davies et al. 2007).

Vertical structures can also bring informational advantages that the leading theories have not yet fully embraced (Arrow 1975; Teece 1976, 2007). These informational benefits have been noted in the context of some applied studies. For example, in the case of natural gas pipelines and the ‘merchant’ function (buying and selling gas), integration permits ‘informational efficiencies’ from the sharing of data about supply interruptions, demand shifts and transportation bottlenecks. This degree of transparency might be too transitory and/or too business-sensitive to be worth sharing between a stand-alone pipeline and multiple merchant partners (Teece 1990). Similarly, a high or low need for ‘unstructured technical dialog’ between design and manufacturing engineers for various product types in the semiconductor industry is correlated with the vertical and specialized structures, respectively, of firms in the industry (Monteverde 1995).

Strategic Considerations

A vertical integration strategy has inherent benefits and risks. A vertically integrated structure may bring better access to resources, the ability to tightly coordinate the integrated stages of production and the ability to obtain parts or services without paying for someone else’s profit margin. However, the integrated firm also has less flexibility to reduce costs in a downturn or following a radical change in technology. Furthermore, because the integrated activity often loses the discipline of the market, the integrated firm may sacrifice access to best-in-class goods or services in favour of using its internal source.

As mentioned above, there are times when investment in self-supply is necessary to launch a new market. Chandler (1977: ch. 12) described the evolution of forward integration in retail sales, service and education by the Singer Sewing Machine Company in the late 19th century. Singer’s high-volume production needed a marketing organization capable of not just demonstrating its products, but actually teaching customers how to use sewing machines and providing repairs and maintenance. Although Singer started with a network of independent agents, it replaced them over time with salaried employees. A similar logic explains the purchase of networks of service stations by petroleum refiners (Teece 1976, 2010).

A modern example of this logic is Apple’s decision to open its own retail stores. These stores allow Apple to control the level of knowledge and service provided by sales staff. The careful training of store employees is an essential element of the strategy.

A similar situation can arise when suppliers have the desired capabilities but are unwilling to make required investments. In 1985, Qualcomm, now a leading supplier of mobile phone technology, was offering an untested digital cellular technology that some doubted would work. Faced with the prospect of limited support by the leading telecom equipment suppliers, Qualcomm decided that it needed to offer an end-to-end solution of its own. Beginning in 1995, when its technology began to be deployed commercially, Qualcomm entered into the design and manufacture of the infrastructure equipment, handsets and key chips that they require. In 1999, as its technology began to gain traction in the market, Qualcomm exited the infrastructure and handset businesses, focusing on microchips and licensing.

In general, vertical integration should be considered whenever a supply chain is being constructed in the service of a systemic innovation (e.g., an aircraft). It is also potentially beneficial in any business model that relies on the tight coordination of its elements for success.

The profiting from innovation framework (Teece 1986, 2006), which recognizes transaction cost considerations as relevant to the make-or-buy decision, provides some guidance for firms as they enter – or create – new markets. The framework shows that vertical integration makes sense in the case of inputs that are not available in competitive supply and are difficult to replicate, conditional on in-house capabilities, time-to-market requirements and other concerns. If such an input resides outside the firm, it may not only become a bottleneck for growth, but the owners of the bottleneck asset may also drain away some of the economic rents accruing from innovation elsewhere in the supply chain.

Viewed from the other side, it’s often strategically advantageous to own a bottleneck asset. The ‘Wintel’ nexus in the personal computer (PC) industry is one example (Morris and Ferguson 1993); other participants in the PC value chain receive at best moderate returns compared with the owners of the bottleneck assets, that is, the microprocessor and operating system.

The integration calculation is dynamic; in some cases, the value of an input may be anticipated to rise when the focal firm introduces its new product, making it worthwhile to integrate in advance. However, there are methods of achieving the necessary degree of control short of complete ownership, such as joint ventures and alliances. For example, Apple doesn’t own any factories, but it maintains exclusive to capacity and know-how by making direct, non-equity investments with its suppliers (Gobry 2011).

Vertical integration is becoming less necessary in many instances because globalization has made a growing number of goods and services readily available on a competitive basis. Vertically integrated structures have been replaced by ‘modular production networks’ in numerous industries, such as electronics (Sturgeon 2002). Competitive supply raises the attraction of ‘buy’ over ‘make’ because it reduces (but does not eliminate) the potential for building (or acquiring) an internal substitute that can provide competitive differentiation.

This tendency of global specialization to reduce vertical integration has an analogue at a local level. In Italian industrial districts, for example, where many small, specialized firms are clustered together, vertical integration is less common (Cainelli and Iacobucci 2012). There is seldom a reason to build internally what can be bought on the market at competitive prices.

See Also