Outsourcing and the Risks of Dependent Autonomy

Outsourcing affords those who engage in it a type of autonomy, albeit a highly dependent, and thus risky, one. The autonomy associated with outsourcing comes from how it allows entities to pursue their activities in a more self-directed manner (e.g., it gives entities greater control over what activities they perform as well as how they are performed). However, this autonomy is only made possible by an entity becoming dependent on those who take over its outsourced activities. In other words, it is only by becoming dependent on others that outsourcing frees entities to peruse their activities in a more self-direct manner. Recognizing the dependent nature of the autonomy associated with outsourcing is important because of the risks it poses both to the entity engaging in outsourcing and to some broader social concerns. This article focuses on three such risks: how this autonomy is dependent on the availability of suppliers willing and able to take on one’s outsourced affairs; how this kind of dependency can lead to a loss of an entity’s internal capabilities; and how an entity can inadvertently outsource, and thus lose control over, its core activities. Given that the dependence inherent in outsourcing relationships is the source of so many of outsourcing’s benefits as well as its risks, this article highlights the highly precarious nature of this dependent form of autonomy.


Original Research
In its Final Report to Congress, the Commission on Wartime Contracting in Iraq and Afghanistan (2011) highlighted how "U.S. agencies engaged contractors at unprecedented levels to help achieve mission objectives in Iraq and Afghanistan" (p. 16). On one hand, this Report notes how this extensive reliance on "contractors frees the military to use service members primarily for warfighting," presumably one of its core activities (Commission on Wartime Contracting in Iraq and Afghanistan, 2011, p. 29). On the other hand, this reliance on contractors " [c]reates unreasonable risks to mission objectives or other key U.S. interests" not found when the military keeps its affairs in-house (Commission on Wartime Contracting in Iraq and Afghanistan, 2011, p. 19). One such risk comes from the possibility of a contractor becoming unavailable as a result of them "walking off the job … [especially] when there is no timely back-up available" (Commission on Wartime Contracting in Iraq and Afghanistan, 2011, p. 48). Another is how relying on contractors "for so much professional and technical expertise eventually leads to the government's losing much of its mission-essential organic [i.e. internal] capability" (Commission on Wartime Contracting in Iraq and Afghanistan, 2011, p. 19). A third relates to how even though some activities are so intimately tied to the core operations of the State they are designated as being inherently governmental thereby mandating they be performed by government personnel, there is evidence showing that "[a]gencies violate [these] inherently governmental standards" by outsourcing some of these core governmental activities to private actors (Commission on Wartime Contracting in Iraq and Afghanistan, 2011, p. 42). In other words, while outsourcing affords the military the freedom to concentrate on certain activities (e.g., warfighting), this freedom results from the military becoming dependent on the use of outside contractors. In turn, this dependence introduces risks that would not be present in the absence of outsourcing (i.e., risks related to contractor availability, the possibility that outsourcing can lead to a loss of an entity's internal capabilities, and how entities can outsource, and thus lose control over, their core activities). This Report ultimately concludes these risks are the result of the military's "over-reliance" on private contractors.
What I want to show in this article is how these risks are not confined to outsourcing in the military context, nor do 845177S GOXXX10.1177/2158244019845177SAGE OpenLair research-article20192019 1 Gettysburg College, PA, USA they arise from an over-reliance on suppliers. Rather, I argue that such risks are inherent in all outsourcing relationships and the type of dependent autonomy they produce. Indeed, previous research (Lair, 2012) has shown that outsourcing allows for possibilities of action and non-action, or "freedoms," not found when an entity keeps its affairs in-house that, alone and in combination, allow entities to pursue their activities in a more self-directed manner (e.g., they give entities greater control over what activities they perform as well as how they are performed). This ability to perform one's activities in a more self-directed manner (e.g., the ability to concentrate on a certain activity) can be seen as a form of autonomy. However, this type of autonomy is distinctive in that it results from an entity becoming dependent on those who take over its outsourced activities (e.g., the military is only free to concentrate on its war-fighting activities because of its dependence on contractors that perform activities on its behalf). Thus, what outsourcing offers those who engage in it is a highly dependent form of autonomy.
Recognizing the dependent nature of the autonomy associated with outsourcing is important because of the risks this kind of dependency produces and the threat they pose both to the entity engaging in outsourcing and to some broader social concerns. As indicated above, this kind of dependency produces risks related to contractor availability, the possibility that outsourcing can lead to a loss of an entity's internal capabilities, and how entities can outsource (and thus lose control over) their core activities. And as these risks result from the dependency inherent in outsourcing, this means they can manifest themselves in any outsourcing relationship. This is significant given that there has been an overall increase in businesses (Anner, 2012;Gereffi & Frederick, 2010;Sturgeon & Kawakami, 2010), the government (Singer,2011;Stanger, 2014), and individuals and families (Epp & Velagaleti, 2014;Hochschild, 2011;Lair, 2012) outsourcing their affairs. Thus, as outsourcing becomes more prevalent, so too are more areas of social life subjected to the risks associated with this kind of dependent autonomy.
This article contributes to the existing literature in two ways. The first is that while some works in sociology (e.g., power-dependence and social exchange theory) (Cook & Yamagishi, 1992;Emerson, 1962) and economics (i.e., transaction cost economics [TCE]) (Williamson, 1991(Williamson, , 1996 identified a link between dependence and risk, these analyses have tended to focus on the risks resulting from power imbalances produced under high levels of asymmetrical dependence. In contrast, I focus on risks associated with the kind of dependence inherent in outsourcing relationships. As such, these are risks that can manifest themselves regardless of the level of power or dependence that exists between the parties in an outsourcing relationship. Second, while some works (McIvor, 2010;Quinn & Hilmer, 1995) have identified some of the risks associated with the dependent nature of outsourcing, they have tended to confine their analysis to one area of social activity (e.g., to business or the military). However, if certain risks stem from the kind of dependence inherent in outsourcing relationships, then it is possible for these risks to manifest themselves in any instance of outsourcing. As such, I hope to show that the risks associated with outsourcing's dependent form of autonomy are both more pervasive and systematic than the literature suggests.
Before addressing these issues, however, it is first important to define what I mean by outsourcing and to illustrate the dependent form of autonomy that results from it.

Outsourcing
Outsourcing occurs when activities conventionally performed by a certain type of entity (e.g., a business, the government, individuals or families) are disembedded from their organizational environment and transferred to source(s) located in the market (Lair, 2012). To clarify what exactly is (and is not) meant by this definition, a number of points are worth stressing.
The first is that outsourcing operates on the level of social organization, that is, in terms of which entities are performing what particular activities. As such, outsourcing implies that an activity once performed with an entity's organizational confines ("in-house") is moved to an outside, organizationally separate, supplier. Noting the organizational nature of outsourcing is important because while some accounts treat outsourcing primarily (or exclusively) as a geographical process (i.e., one where activities are transferred across large spatial distances), the reality is that outsourcing can occur on any geographical scale, be it local (e.g., the outsourcing of janitorial functions within the same physical location) or global (e.g., when American call center jobs are outsourced to overseas suppliers).
Second, even though many works frame outsourcing as something exclusively engaged in by businesses, the reality is that many different types of entities (e.g., government agencies, individuals, and families) also disembed activities once performed in-house and transfer them to market-based sources. Indeed, one of the most striking aspects of outsourcing is its increased presence across many areas of social life.
Third, while the ability to remove an activity from one's organizational confines and reinsert it into those of another may not seem that significant, the reality is otherwise. This is because the ability to disembed activities from one entity's organizational environment and remebed them in those of another offers entities certain possibilities of action and nonaction that are not evident when an entity keeps its affairs in-house. These possibilities of action and non-action can be thought of as freedoms. Moreover, it is these freedoms that afford entities the type of autonomy that is associated with outsourcing.

Autonomy and the Freedoms of Outsourcing
One of the most striking aspects of outsourcing is how it enhances an entity's ability to control what activities it performs as well as how these activities are performed. As mentioned above, this increased control can be thought of as a type of autonomy. In turn, this autonomy stems from how outsourcing frees entities to do (and not do) certain things that would not be possible if an entity were to keep its activities in-house. Five of these freedoms are worth noting. The first is how outsourcing allows an entity to free itself of activities it does not wish to, or cannot, perform in-house. In this regard, outsourcing allows entities to divest themselves of certain activities but still have access to them (as they are being performed by others on one's behalf). In this regard, outsourcing gives entities greater control over the activities that they do not have to perform.
The second freedom relates to how by outsourcing certain activities, entities are then freed to concentrate on others (Prahalad & Hamel, 1990). By transferring one's activities to outside suppliers, resources (e.g., time, energy, manpower, etc.) that would normally be involved in performing these activities are freed up which can then be devoted to the performance of other functions. Thus, this freedom enhances both an entity's ability to control which activities it performs and how they are performed (e.g., in terms of the internal resources it has available to devote the activities that remain in-house).
The third freedom is how outsourcing frees entities to access others that have more, different, and/or better resources than those found in-house (McIvor, 2010). These resources can include such things as skills, technical capabilities, manpower, and time. Thus, even if an entity lacks certain resources internally, it can still have access to them via outsourcing. Moreover, it can access these resources without the time and expense associated with developing them in-house. This freedom gives entities greater control over how their activities are performed.
Another freedom associated with outsourcing is how it frees entities to be more operationally flexible. That is, outsourcing frees entities to change the conditions under which their activities are performed in ways that are not possible if an entity were to keep its affairs in-house. In particular, an entity can change these conditions by switching suppliers as needed and/or desired. This freedom is enhanced by the fact that outsourcing allows these changes in conditions to be made without an entity fearing that it will lose any/all of the previous investments that it has made in developing its internal capacities (e.g., the costs of building a factory) as these are something borne by the supplier. This freedom gives entities greater control over how their activities are performed.
Finally, outsourcing also frees entities to move their activities to different organizational and/or institutional environments. When an entity disembeds an activity from its organizational confines, it is freed to move it to locations where the organizational and/or institutional conditions are different from those found in-house. The most common example of this is the movement of outsourced activities to lower cost organizational environments. However, this same freedom is evident when entities move their activities to environments with different institutional regulations (e.g., one of the purported benefits of outsourcing governmental activities is that they are removed from the State's overly bureaucratized institutional environment and into the more flexible context of the market) (Rasor & Bauman, 2007). Here, again, this freedom offers entities greater control over how their activities are performed.
As can be seen from the above, the freedoms associated with outsourcing give entities greater control of what activities they perform as well as how they are performed. In this regard, they offer entities a type of autonomy. However, the type of autonomy afforded by outsourcing is unique in that it is only made possible by an entity becoming dependent upon others to perform the activities that it has freed itself of.

Outsourcing and the Dependence of Autonomy
The dependent nature of the autonomy afforded by outsourcing can be seen by looking at a number of specific cases. In the business context, the electronics industry is especially illustrative of this phenomenon. In this industry, most electronics "producers" do not actually make the products that are sold under their name. Rather, the production of these goods is outsourced to contract manufactures (Lim & Culpan, 2010;van Liemt, 2007). This development has been driven largely by the autonomy afforded by outsourcing. Thus, outsourcing gives these firms greater control over which activities they perform by allowing them to free themselves of certain tasks (e.g., manufacturing) which, in turn, allow them to concentrate on others (i.e., research and development, marketing, and brand building) (International Labor Organization, 2014). For example, the primary reason that Apple is able to focus on what it sees as its core competencies (i.e., design, marketing, and retailing) is because it has become dependent upon an extensive array of contractors that actually manufacture the goods sold under its name (Weil, 2014).
Outsourcing also gives electronics companies greater control over how their activities are performed. In part this is because outsourcing allows these companies access to suppliers with specialized resources they may lack internally (Cassia, 2010). In addition, as electronics companies are "not tied to production sites through property ownership and permanent staff contracts" as a result of outsourcing their manufacturing activities, they have a high degree of operational flexibility (Astill & Griffith, 2004, p. 7). This flexibility in turn enhances electronics companies' ability to control the organizational and/or institutional environment under which their activities are performed (e.g., how if "wages in one country rise, outsourcing makes it easier to move somewhere cheaper"; Astill & Griffith, 2004, p. 7).
However, the autonomy afforded by outsourcing and its associated freedoms is only possible because electronic companies have become dependent upon others to perform the activities they outsourced. One specific measure of this can be seen in the operations of Apple and how in 2012 for every one worker directly employed by this company, approximately 12 others were employed by one of its contractors (Weil, 2014). On a more general level, this dependency can be seen in how beginning in the 1990s many brand name electronics companies began divesting themselves of their production facilities. By freeing themselves of the means to produce their goods in-house, most electronics companies now have no choice but to depend upon others to produce their goods for them (Ernst, 2004).
The same pattern of dependent autonomy can be seen in the automotive industry. The auto industry was once the paragon of vertical integration with car producers attempting to keep virtually all elements of production in-house (Cairola & Revelli, 2004). Thus, in 1960s and 1970s when levels of vertical integration in this industry were at their zenith, automakers routinely used 80% or more self-produced components (Kwon, 2004). However, over time, automakers began to rely more upon outsourcing to capture the benefits associated with the autonomy that this practice affords. Thus, many automakers began to focus their efforts on overall automobile design and assembly by freeing themselves of the design and production of a number of a car's subsystems (Anner, 2012;Frigant & Lung, 2002;Zirpoli & Becker, 2011). Automakers also turned to outsourcing because this gave them access to specialized resources not possessed inhouse (e.g., the skills associated with producing a particular module such as sunroofs; Frigant & Lung, 2002;Zirpoli & Becker, 2011). In addition, outsourcing enhanced automakers' operational flexibility by allowing them switch between suppliers as needed or desired (Anner, 2012;Ciravegna, Romano, & Pilkington, 2013;Helper & Kleiner, 2003) while also freeing them to move their activities to different organizational and/or institutional environments as needed or desired (Graham, 2010).
However, automakers have only been afforded these freedoms because they have become dependent upon others to perform the activities they have freed themselves of. One measure of this dependency is how it is now estimated that approximately 80% of a car's value now comes from purchased (i.e., outsourced) components (Klier & Rubenstein, 2008). More specific measures of this dependency can be seen in how Ford and General Motors, once characterized by very high levels of vertical integration, now depend, respectively, on 1,260 (Trudell, 2013) and 2,700 (Colias, 2014).
This dependent form of autonomy can also be seen in the context of governmental outsourcing. This development is particularly evident in the Logistics Civil Augmentation Program (LOGCAP) which outsources the bulk of U.S. Army's general support and logistics functions. Begun in the 1980s, LOGCAP contracts are awarded to a private contractor(s) for a certain period of time (e.g., 5 or 10 years) that becomes responsible for providing support and logistical activities to the military if a contingency situation arises (Rasor & Bauman, 2007). As such, this program frees the military from having to provide many of the support activities (e.g., base construction) it historically performed itself. Moreover, by being freed of these non-combat activities, the military is then freed to concentrate its efforts on its warfighting (Singer, 2004). This program also promised the military more operational flexibility given the institutional differences that exist between the government and the free market (e.g., how it is easier to both hire and fire workers in the private sector as opposed to the public one) (Goldberg & Gilmore, 2005;Rasor & Bauman, 2007;Schwartz, 2011).
But, in attempting to capitalize upon these freedoms, the military had to become highly dependent upon private contractors. The extensive scope of this dependency is particularly evident when the use of private contractors is placed in historical context. Until the United States's most recent war efforts in Iraq and Afghanistan, contractors generally made up about 15% of the military's total force (soldiers plus contractors) (Commission on Wartime Contracting in Iraq and Afghanistan, 2009). In 2011, the last year that U.S. forces were officially engaged in Iraq, contractors made up roughly 58% of the total American force (Schwartz & Swain, 2011). Ever since the official withdrawal of American troops, this pattern of dependence remained so that, as of 2015, there were about twice as many private contractors in Iraq as military personnel (Zenko, 2015). In Afghanistan, the story is much the same: in 2011, contractors made up just under half (48%) of America's total force in this country (Schwartz & Swain, 2011). However, even after officially leaving Afghanistan in 2014, contractors still outnumber soldiers by a ratio of more than 3 to 1 (Zenko, 2015).
This dependent form of autonomy is also evident in the outsourcing of individual and family affairs. For example, Sandholtz, Derr, Buckner, and Carlson (2002) profile a very busy family that outsources a number of activities to "just spend time together" (p. 71). However, this freedom is only possible because of this family's reliance upon a number of people outside of the family. A measure of this dependence can be seen in a comment made by one member of this family: "We have a family of four and a staff of eight" (Sandholtz et al., 2002) (p. 10). On a more general level, Cortés and Tessada (2011) have found that one of the main reasons why high-earning women have been freed to spend more time working, and thus to take more time-demanding jobs, is because of their dependence upon low skilled migrant workers who have taken on many of the domestic tasks that these women (and/or their families) are too busy to perform themselves.
The above cases highlight how outsourcing produces a highly dependent form of autonomy. Recognizing this is significant because dependency introduces certain risks that are not evident when an entity keeps its affairs inhouse. Of course, that a dependence upon others can be risky is something that has been noted in the sociological and economic literature. However, the literature has tended to concentrate on the risks associated with power imbalances that result from asymmetrical levels of dependence. Perhaps the most famous example of this is power-dependence theory developed by Richard Emerson (1962) and colleagues (e.g., Cook, 1977;Cook & Yamagishi, 1992). From this perspective, the power that one actor has over another is directly related to how dependent the latter is upon the former in terms of acquiring a particular resource (Emerson, 1962). The more dependent one party is on another for a particular resource, the more power the latter has over the former.
TCE also sees dependency upon one other entity as being risky because of the power imbalance it can induce. For TCE, asymmetrical dependence manifests itself through asset specificity, or the "degree to which an asset can be redeployed to alternative uses and by alternative users without sacrifice of productive value" (Williamson, 1991) (p. 281). The more one party sinks high asset-specific resources into an exchange relationship, the more dependent this party will become upon the continuation of this particular relation as most, if not all, of the value of these resources would be lost if this particular exchange relationship were to end. As such, situations of high asset specificity are problematic in that if one party in a transaction realizes that another is highly dependent upon them because of asset specificity, they can opportunistically exploit this situation to their advantage (e.g., by asking to renegotiate a contract on terms that are considerably more favorable to it).
As can be seen, both power-dependency theory and TCE highlight the risks associated with power imbalances that can emanate from asymmetrical levels of dependence in an exchange relationship. However, the dependence found in outsourcing relationships can be risky in ways that are independent of how much one party is dependent upon another. And while a number of these risks have been identified in the outsourcing literature (McIvor, 2005;for example, McIvor, 2010;Quinn & Hilmer, 1995), these works have largely confined their analysis to one context of social life. This is a significant limitation given how prevalent outsourcing is in so many areas of social life. Below, I attempt to overcome this limitation by showing that these risks are more prevalent and pervasive than is suggested by the outsourcing literature.

The Precariousness of Dependent Autonomy
In this section, I identify three risks associated with the dependent form of autonomy afforded by outsourcing. These are how this autonomy is dependent on the availability of suppliers, how outsourcing can lead to a loss of internal capabilities, and how entities can inadvertently outsource their core activities.

The Availability of Suppliers
One risk associated with the type of autonomy afforded by outsourcing is how it is dependent upon the availability of suppliers willing and/or able to perform one's outsourced activities. If for whatever reason a suitable supplier cannot be found, or those that are found are not willing and/or able perform what one wishes to outsource to them, then one has no access to the autonomy afforded by outsourcing. This unavailability of suppliers can occur either before or after an outsourcing relationship has been established.
In terms of the former, an example can be seen in how electronics companies are only able to capture the benefits of outsourcing because suppliers are willing to, and capable of, taking on their manufacturing activities. As Sturgeon and Kawakami (2010) put it, "this sort of outsourcing would be impossible without suppliers with the capabilities to accept the work and efficiently meet the requirements of lead firms" (p. 17). And while "[s]uch firms exist today" so that outsourcing in this industry is routine, this "was not always the case." Indeed, many contract manufacturers "were asked and in some cases forced by de-verticalized 'manufacturers' in the West" to upgrade their production capabilities so that outsourcing could occur (Sturgeon and Kawakami, 2010). If this upgrading had not taken place, then electronics firms would have been forced to keep their manufacturing activities inhouse thereby missing out on the autonomy afforded by outsourcing and its associated freedoms.
Similarly, Lacity and Willcocks (2006) note that a number of participants in their case studies of information technology outsourcing "were willing to outsource their large data centers but could not find suppliers who could do it cheaper" (p. 15) than they could which forced them to keep these activities in-house. Along the same lines, even though the arts and crafts retailer Michaels wanted to outsource the building of its custom frames, it was unable to do so because "no supplier was large enough to assemble and quickly deliver daily orders" (Gordon, 2002). As such, Michaels had to abandon its desire to outsource this activity and the autonomy this would have afforded.
Situations like the above are not confined to the business sector. For example, one of the factors that has historically limited the use of domestic workers has been a lack of individuals willing to engage in this type of work, not a demand for them. Katzman refers to this as the "servant problem" and notes how this "was the bread and butter of women's magazines between the Civil War and World War I, and it filled volumes of general-circulation weeklies and monthlies as well as the earliest issues of social science journals." As a result of this "problem," many who wanted to reap the benefits associated with outsourcing their domestic activities could not. But even if one is able to secure a supplier and form an outsourcing relationship, there are times when these suppliers can become either temporarily or permanently unavailable. This unavailability is problematic because it forces an entity into a situation where it must either do without what it has outsourced, which may not be possible if an activity is necessary and/or critical to an entity's operations; switch suppliers, which can be costly in terms of the time, money, and/or other resources that are needed to find a new provider of a good service; and/or resume performing its activities in-house thereby losing out on autonomy afforded by outsourcing (as will be discussed below, resuming the inhouse performance of activities may not be practical given how outsourcing tends to drain entities of their internal capabilities). Two examples illustrate the risks associated with existing suppliers becoming unavailable particularly well: the outsourcing of military activities and child care.
In terms of the former, if the military outsources an activity that is critical to its operations, and the supplier of this activity becomes unavailable, the result can be a quite dangerous. An example of this can be seen in how, in 2004, a number of military contractors decided not to deliver much needed supplies in Iraq as a result of a spike in convoy attacks. The result of this refusal was a potentially dangerous situation for the military given that its "fuel and ammunition stocks [began] to dwindle" jeopardizing both the mission and those assigned to carry it out (Singer, 2008). Similarly, when other contractors refused assignments to "dangerous parts" of Iraq, U.S. troops were left "sitting in the mud, and without hot food" again jeopardizing both the well-being of these troops and their overall mission (Surowiecki, 2004). In these cases, we can see the risks that are evident when a supplier becomes unavailable in an existing outsourcing relationship.
In terms of the latter, child care is a classic case of dependent autonomy in that for many families, its only by relying on this service that members of a family (particularly mothers) are freed to work (Epp & Velagaleti, 2014). So long as providers are available to perform this activity, this dependency is not especially problematic. However, as child care providers can have disruptions in their operations that render them temporarily unavailable (e.g., a center will be closed on a holiday, because of bad weather, as a result of illness, etc.), it is not uncommon for families to temporarily lose access to their providers (Gordon, Kaestner, & Korenman, 2008;Usdansky & Wolf, 2008). When this occurs, families are forced to either find alternative care arrangements (which can consume at least some of the time they had freed up through outsourcing) and/or miss work (thereby losing out on the ability to concentrate their efforts on this activity) (Usdansky & Wolf, 2008). These examples show how the autonomy offered by outsourcing is dependent both on the existence of suppliers and their continued availability.

Outsourcing and the Loss of Internal Capabilities
Another risk associated with the dependent autonomy afforded by outsourcing relates to how, when entities outsource their affairs, they tend to free themselves not only of the specific tasks being disembedded but also many of the resources (e.g., human, physical, and/or informational) associated with producing these affairs in-house. However, by freeing themselves of these resources, entities can lose (or never develop) the capabilities associated with the in-house production of these activities. As such, outsourcing can drain entities of their internal capabilities and leave them dependent upon outsourcing.
An early identification of this risk was made by Quinn and Hilmer (1995) in their discussion of how as a result of some U.S. companies deciding to outsource the production of what "seemed to be only minor components" of the products they sold (e.g., bicycle frames), they "lost the skills" (p. 53) needed to actually manufacture these components. This situation proved to be problematic when these companies learned that their suppliers were "unable or unwilling" to provide these components to these companies as these companies could no longer produce these components themselves (Quinn & Hilmer, 1995, p. 53). Outsourcing had made these companies dependent upon outsourcing.
A similar example can be seen the case of "Alpha," the pseudonym of a European car maker studied by Zirpoli and Becker (2011). In the 1990s, Alpha began to outsource entire subsystems of their cars in the hope this would increase the company's flexibility, reduce design time and cost, and improve quality by allowing Alpha to focus on the integration of these various modules. However, over time, Alpha's managers began to realize that outsourcing was eroding the company's ability to actually integrate the subsystems it had outsourced. This erosion was the result of Alpha losing the engineers who had formerly designed the now outsourced modules and how the engineers who remained in-house became more removed from, and thus less familiar with, the technical aspects of the outsourced systems. The result was that "Alpha had no choice but to lean more heavily on outside suppliers" when designing their cars. Here again, we see how outsourcing can drain an entity of its internal capabilities and, thereby, make it dependent upon outsourcing.
A study by Rand highlights this same risk in the context of military outsourcing. In particular, this report notes how when military activities are outsourced, "Army personnel lose training opportunities they need to develop missioncritical skills," which causes the military to become even "more dependent on contractors" (Camm & Greenfield, 2005, p. 171). As noted above, this risk manifested itself in the United States's war efforts in Iraq and Afghanistan when the extensive use of contractors resulted in the military losing "much of its mission-essential organic capability" (2011, p. 29). The end result of this was that the military had no choice but to increase its reliance upon contractors (Camm & Greenfield, 2005, p. 171). Along the same lines, Singer (2011) has noted how one of the structural "benefits" of providing private military services to developing nations relates to how as these states never develop the "expertise and capabilities" associated with these activities when they are outsourced, "client dependency grows each time" they are outsourced (p. 78). Here again, we see how outsourcing can drain an entity of its internal capabilities causing it to become increasingly dependent upon outsourcing. Rosen (2005) has argued that the same dynamic is evident in the outsourcing of domestic affairs. For Rosen, the extensive use of domestic outsourcing can result in "a form of domestic learned helplessness" given that the skills and capabilities needed to perform household functions are either atrophied or never developed when they are outsourced (cf. Bradley, 2010). Ehrenreich (2003) makes a similar point when she notes that [s]ince most of us learn to clean from our parents (usually our mothers), any diminution of cleaning skills is transmitted from one generation to another… Upper-middle-class children raised in the servant economy…are bound to grow up as domestically incompetent as their parents and no less dependent on people to clean up after them. (p. 100) This same dynamic is evident in other, perhaps less obvious, cases of outsourcing on the individual level. For example, aside from the dishonesty involved, one of the concerns that surrounds the outsourcing of writing academic papers (e.g., hiring a service to write one's paper) is how students "are depriving themselves of the opportunity to strengthen their communications and writing skills" (Laws, 2015). Moreover, if students continually deprive themselves of the ability to develop these skills, then they will have little choice but to depend upon outsourcing when they are called upon to demonstrate these skills in the future. Similarly, a common criticism of "bought" (i.e., outsourced) greeting cards (as opposed to ones that are "made" in-house) is that they are part of "a general trend in consumer culture whereby we look to the market and to experts to fill our needs, becoming deskilled, passive and dependent on the market in the process") (West, 2010, p. 452).
These examples illustrate how outsourcing can lead to a loss of internal capabilities which, in turn, can leave those who outsource their affairs increasingly dependent upon the act of outsourcing. This shows how the dependence associated with outsourcing can actually undermine the very thing it promised to enhance: that is, an entity's autonomy.

The Outsourcing of Core Activities
As noted above, outsourcing allows an entity to free itself of certain activities which can, in turn, free it to focus on others. This often takes the form of entities outsourcing their noncore (i.e., peripheral) activities to concentrate on their core ones, or those activities that are seen as being an entity's raison d'être (Reve cited in McIvor, 2003. In principle, pursuing this freedom is a relatively uncomplicated process: an entity simply inventories its activities to determine whether a particular activity is core or not. If it is the former, then it should be kept in-house; if it is the latter, this is something that can be outsourced. However, in practice, determining whether an activity is core or not is not as straightforward as it seems. This is partly because entities can have a hard time identifying what exactly are their core activities given that these are often "vague and intangible" (McIvor, 2005, p. 67). However, even if one is clearly able to determine whether an activity is core or not, it is also the case that "when a process is considered to be non-core, there can be implicit and tacit interdependencies with processes that are considered to be core" (McIvor, 2010, p. 245). As such, even outsourcing seemingly peripheral activities can "have a detrimental effect upon the performance of core activities that remain" in-house (McIvor, 2010, p. 245). As a result of problems associated with identifying an entity's core activities as well as the interdependencies that can exist between an entity's core and non-core activities, many entities have outsourced core activities or those that are crucially tied to them.
One example of this can be seen in the history of Mattel. Initially, Mattel outsourced a great deal of its production activities to overseas vendors. However, "in the 1980s executives became concerned that outsourcing toy-making put trademarks at risk: the market could be flooded with imitation Barbies" (Barboza & Story, 2007). Thus, while Mattel decided that "[n]oncore products, like trinkets made under movie-licensing deals, could be outsourced," the production of its core products (e.g., Barbie dolls and Hot Wheels cars) should only occur in-house (Barboza & Story, 2007). As such, Mattel reclaimed some of its manufacturing activities because it realized these were core parts of their operations.
Similarly, even though the core competencies of pharmaceutical companies are increasingly "focused around the final stages of drug development, which include large-scale, expensive, clinical studies to establish efficacy, regulatory interactions for marketing approval, manufacturing in large quantities, and marketing, sales and distribution of the final product," final stages of drug development are now being outsourced by a number of pharmaceutical companies (Mehta & Peters, 2007, p. 30). For example, even though personnel from pharmaceutical firms interviewed by Mehta and Peters indicated that drug trial design is something they should "never outsource" given how crucial this activity is in determining a drug's potential efficacy, about one-third of contracts they reviewed called for the outsourcing of this very activity (see also Christensen, Grossman, & Hwang, 2009).
But even if firms keep their core activities in-house, many are outsourcing tasks that have strong (if unrecognized) ties to their core activities. Because of this, many companies have harmed their core operations by outsourcing what seemed to be peripheral to them. For example, Pisano and Shih (2012) have argued that many U.S. firms have undermined what they see as their core competency-the ability to innovate new products-as a result of outsourcing their manufacturing activities, something they see as being peripheral to their operations. The reason for this is that in many cases, "R&D and manufacturing are tightly intertwined" so that " [u]nless you know how to manufacture a product, you often cannot design it. And, to understand how to manufacture it, you have to have manufacturing competencies and experience" (Pisano, 2009). Thus, by outsourcing this supposedly peripheral activity, these companies are undermining the very thing they see as being core to their operations.
As noted above, the outsourcing of core activities is also evident in governmental outsourcing. The most significant cases of this occur when activities defined as being "inherently governmental"-that is, as something "that is so intimately related to the public interest as to require performance by Federal Government employees" (Federal Activities Inventory Reform Act of 1998)-are disembedded from the confines of government and transferred to sources in the private sector. Despite inherently governmental activities being explicitly defined as those that should remain within the confines of the government, the recent past has witnessed the outsourcing of many such activities. Perhaps the most significant example of this was the government's use of private contractors in the "enhanced interrogation" of detainees. As the Senate Intelligence Committee's Report on Torture notes, the CIA hired two contractors to develop, operate, and assess its interrogation program. In this capacity, these contractors "carried out inherently governmental functions, such as acting as liaison between the CIA and foreign intelligence services, assessing the effectiveness of the interrogation program, and participating in the interrogation of detain-ees…in foreign government custody" (Senate Select Committee on Intelligence, 2014, Kindle location 389). The morality, efficacy, and legality of enhanced interrogation techniques notwithstanding, as Verkuil (2007) notes, "[w] hatever else it is, torture is surely a governmental 'function' that cannot be privatized" (p. 130). The problem is that it was.
However, this is not the only case in which inherently government functions have been outsourced. For example, a 2013 report by the Governmental Accountability Office (2013) found that in (FY) 2011, the Army had the full-time equivalent of 936 contractors performing inherently governmental functions, while this same number was 473 for the Air Force. In addition to this, the Air Force outsourced inherently governmental activities related to its CSAR-X search and rescue helicopter. In particular, the Air Force outsourced the drafting of a response to the Inspector General's Office despite the fact that this is clearly defined as being an inherently governmental activity (Office of Inspector General, 2010).
In addition to this, there are many cases where the government has outsourced activities that, even if not officially defined as being inherently governmental, are so close to, and so bound up with, them that their outsourcing poses a significant challenge to the State's ability to retain control over its core functions. For example, a Government Accountability Office (2011) audit of the professional and management support service contracts of five governmental agencies (the Departments of Homeland Security, Transportation, and Housing and Urban Development; the United States Agency for International Development; and the National Science Foundation) found that more than half of these contracts "requested services that closely support the performance of inherently governmental functions" (p. i). The Government Accountability Office (2013) also found that in (FY) 2011, the Army had the full-time equivalent of 44,541 contractors performing work officially defined as being "closely associated with inherently governmental functions," while this same number was 1,398 for the Air Force (p. 23).
Many of these same concerns are evident when personal and, more importantly, intimate tasks are outsourced. This, for example, was a theme expressed by a number of people in Hochschild's (2011) exploration of the emotional tolls of intimate outsourcing. For example, one father had "taken small stands against … forms of what he considered the over-outsourcing of domestic tasks" because he felt that, at some point, this type of outsourcing results in people being "estranged from their own intimate life" (Hochschild, 2011, p. 27). Similarly, a mother who generally believed that you should "outsource what you can to experts" was, nevertheless, wary of outsourcing in the context of parenthood because "[i]f you outsource all these tasks to a different specialist, your kid is going to feel like the car you take in for the tune-up, oil change, wheel rotation, lube job. How would [your child] remember [his/her] childhood?" (Hochschild, 2011, p. 30). In other words, these persons resisted outsourcing certain aspects of family life because they felt they were core parts of domestic and family life (cf. also Epp & Velagaleti, 2014).
But even if core parts of individual and family life are left in-house, there is still room for concern when individual and family affairs are outsourced. This is because of the interdependencies that can exist between an entity's core and noncore activities noted above. Jackson (2002) puts this concern in the context of family life: The domestic work we often dread-raking the leaves, scouring the kitchen sink, washing the curtains-doesn't necessarily make a house a home…Yet outsourcing domestic life does have a symbolic cost, because such chores in sum are more than just drudgery. In accomplishing seemingly small domestic tasks again and again, we're creating much more than a tidy shelter from the elements. Each time we cook, clean, or mend, we're creating opportunities for talking, for being together with those who share our home. We're creating the glue that binds us to the humans we love. (p. 71) Jackson is pointing out that the outsourcing of even seemingly mundane and peripheral domestic activities might inadvertently squander chances to come together as a family. Thus, even if outsourcing is used to free up resources so that a family can bond, depending on others to perform certain activities may actually take some opportunities for this away. Pollan (2013) makes a similar argument in terms of the outsourcing of cooking. While noting how the outsourcing of food preparation has many liberating aspects to it, it is also the case that it has "undermined the institution of the shared meal, by encouraging us to eat different things and to eat them on the run and often alone" (Pollan, 2013, p. 8). For Pollan (2013), this "is no small thing" given that family meals are "a foundation of family life, [a] place where our children learn the art of conversation and acquire the habits of civilization: sharing, listening, taking turns, navigating differences, arguing without offending" (p. 8). Here again, we see that the outsourcing of a seemingly peripheral activity (food preparation) can negatively affect one that is more core to the family (meals together).

Conclusion
This article has shown that while outsourcing and its associated freedoms can offer entities a form of autonomy, certain risks are attached to this autonomy given its dependent nature. Three principal risks associated with this dependent form autonomy were highlighted: how this type of autonomy is dependent upon the availability of suppliers to take on one's outsourced affairs, how outsourcing tends to drain an entity its internal capabilities, and how entities may inadvertently outsource a core activity (or ones that are intimately tied to them).
Noting the risks that stem from this dependent form of autonomy is important for two reasons. The first is that it highlights how most, if not all, of the benefits associated with outsourcing are derived from the exact same source as the risks noted above: the dependence inherent in outsourcing relationships. In other words, it is the ability to rely on others to perform one's disembedded affairs that is the source of this autonomy's benefits and risks. This, in turn, highlights outsourcing's highly precarious nature: that is, how the risks and benefits of this activity are so intimately tied to one another that it is impossible to have the former without the threat of the other. As such, the more entities attempt to secure the autonomy afforded by outsourcing, the more they expose themselves to the risks inherent in this practice (even if these risks are not intended or foreseen).
The second is that the dependent form of autonomy offered by outsourcing is also potentially destructive. In part, this is because outsourcing can be quite detrimental to the entities that engage in this activity. The clearest example of this is how outsourcing can drain an entity of its internal capabilities thereby undermining its ability to perform certain activities. This risk is especially concerning given that entities can outsource activities that are, or are closely related to, core to their operations. In these cases, it is possible for outsourcing to drain entities of the very activities they see as being most central to their operations. However, the destructive potential of outsourcing relationships can also can extend beyond the confines of the entity that outsources its affairs. For example, the democratic foundations of the American political system are seriously undermined when core governmental activities are outsourced. In particular, these kinds of activities are supposed to remain within the confines of government because this allows for public oversight of them and in ways that are not possible when such activities are performed in the private sector. Moreover, the outsourcing of inherently governmental activities allows the interests of private entities to creep into, and potentially override, what are supposed to be the public's most central concerns. Thus, the dependent nature of outsourcing has the potential to not be detrimental not just to the specific government agencies involved in this activity but also the entire institutional foundation of government.
Given the autonomy it affords, it is likely that outsourcing is going to continue well into the future. However, as this article has shown, this autonomy comes at the price of certain risks associated with autonomy's dependent nature. While it does not seem possible to eliminate such risks from outsourcing relationships, given that they stem from the very structure of such relationships and are also the source of what can make outsourcing so beneficial, it is hoped that by shedding light on this issue, this article can help to mitigate some of the more precarious aspects of outsourcing.

Author's Note
The research discussed in this article was originally presented at the 113th Annual Meeting of the American Sociological Association, an abstract of which was published in https://convention2.allacademic.com/one/asa/asa18/index.php?cmd=Online+Program+View +Paper&selected_paper_id=1378363&PHPSESSID=fa8urqkfjuo1 blekur04e1dpp4

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.