You can be forgiven for thinking that the days of the offshore captive center were numbered. \n\nCitigroup, Unilever, Deutsche Bank, and Dell are just a few of the Fortune 500 companies who have sold or shut down their one of their wholly owned \n\noffshore service centers in the last year. Some 60 percent of all such captive centers fail to meet expectations. \n\nAnd the third-party offshore outsourcing industry, particularly in India, has reached a level of provider maturity that one wonders why a company would \n\ntake on the hassle of setting up their own shop abroad. (Indeed, many industry watchers have.)But the captive center lives on. Captive activity reached a two-year high in the fourth quarter of 2009, with the opening of 40 new wholly-owned \n\nservice centers offshore, according to outsourcing consultancy Everest, and the last five quarters have seen growth in the captive market. We talked to Ilan \n\nOshri, author of the forthcoming book, Offshoring Strategies: Evolving Captive Center Models (February 2011, MIT Press) about the resurgence of \n\noffshore captives, the centuries-old history of the model, and the continued difficulty of making them \n\nwork.CIO.com: Why write about captive centers now?Ilan Oshri: I have published six books on offshoring and outsourcing. In 2006, I got interested in captive centers while I was visiting Bangalore \n\nlooking into a problem area between a captive center and a local vendor. That made me think if captive centers outsource work which has already been \n\noffshored, they were probably pursuing other strategies. The academic literature offers very little about captive centers and the professional press also did \n\nnot take interest in captive centers unless it was a failure story. This industry deserves a thorough study starting with the basics: taxonomy, strategies and \n\ncapabilities.CIO.com: For the past few years, we've been hearing that interest in captive offshore centers\u2014particularly IT centers combined IT and \n\nbusiness process centers\u2014were waning, and companies from AOL to Unilever have shuttered or sold one of their offshore captives. Are they really \n\ncoming back?Oshri: According to Everest Research, they are. 2009 saw an increase in the number of captive center setups. My view is that captive center \n\nis one sourcing model from the many sourcing models available nowadays. It has its advantages and disadvantages and there will always be firms that will \n\nfind this sourcing model suitable for their needs.CIO.com: What was behind the divestitures of so many captive centers around 2008, and into early 2009?Oshri: In the middle of the financial crisis, some multinationals divested their captives to improve their cash position. There was also interest \n\nfrom local vendors to buy captives that have built scale or have specialized in an area complementary to the line of services the local vendor is providing \n\nfrom offshore.CIO.com: Has that sell-off slowed down?Oshri: Indications are that it did not slow in 2009. I do not have full data on 2010. But it has become far more challenging to divest a captive \n\nfor two reasons. First, local vendors have built massive scale offshore and the acquisition of a captive today will probably not have a dramatic impact on \n\ntheir economies of scale. Second, private equity firms\u2014the other candidate buyers\u2014are not cash rich as they were between 2002 and 2006. \n\nHaving said that, vendors will consider buying a captive if the purchase includes a long-term service contract.CIO.com:You say there will always be a need for captive centers, even with the maturation of the offshore IT services industry over the last \n\ndecade. Why?Oshri: Some executives believe that certain business processes or an IT function are too important for the firm to leave it in the hands of a \n\nthird party. Others set up a captive center in a growth market hoping to turn their captive into a success story. There will always be those who hope that the \n\ncaptive center will save costs, but to achieve that the parent firm needs to invest in the captive and ensure that it builds scale.CIO.com: In your book you note that 60 percent of offshore captives will struggle and perhaps not survive. Why is this model so difficult to \n\nset up and maintain successfully?Oshri: There are two stories here. The first is about captive centers which were set up for the wrong reason\u2014usually a "me too" \n\nstrategy. These captives often fail to develop scale, cannot follow the evolutionary path described in my book, and therefore will struggle. Other captive centers had the potential to become successful with the right management attention and allocation of resources. In many of these cases, \n\ncaptive centers become unsuccessful because of the gap between the perceived potential and execution. Most parent firms are attracted to the idea that the \n\ncaptive center can produce massive cost savings, so they go for this sourcing model. They set up captives and expect to see cost savings almost \n\nimmediately. In reality captive centers face challenges that can erode potential savings if not handled properly: high attrition levels (affecting knowledge \n\nretention and higher training costs), tense relationships with the parent firm, inefficient governing structure, and other issues. The captive can cope with \n\nthese challenges, but it needs resources and management attention from the parent firm.CIO.com: Your research found that "talent" was cited as the main driver for setting up a captive center offshore. But with unemployment \n\nlevels in Western countries so high, isn't that just code for "cheaper talent"?\n\nOshri: True, but I won't call it "cheaper talent." It is talent at the right price and value. CIO.com: Did it surprise or concern you that 37 percent of company executives you surveyed said they did not know why they set up a \n\ncaptive center\u2014or would not answer the question.Oshri: There are many reasons for setting up a captive center, beyond talent, costs, infrastructure and government policy, which firms tend not \n\nshare with the public.CIO.com: What are the best examples of a successful IT or IT-BPO captive centers?Oshri: Some experts suggest that the nature and purpose of captive centers must evolve for them to be successful. WNS, previously owned \n\nby British Airways, has evolved from a basic captive providing services to a parent firm to a larger center that now provides services to international \n\ncustomers as well. Genpact, formerly a captive center owned by GE, is another successful example. They have transformed from transform service and \n\ncost centers to profit centers.CIO.com: So in these examples, divestiture is not a sign of failure but part of the evolution of a successful captive?Oshri: Yes.CIO.com: What industry owns the most offshore IT or IT-plus captives and why?Oshri: The electronics and computing industry is the biggest operator of captives. The simple reason for that is that this industry has utilized the \n\ncaptive model to do research and development offshore as well as set up shared service centers around the globe.CIO.com: You found that India is still the prime location for offshore IT captives. Are there any near- or long-term competitors?Oshri: Countries in Central and Eastern Europe are emerging as serious contenders to India.CIO.com: It's clear that captives have to be pretty big to succeed. How can small or mid-size companies take advantage of them?Oshri: SMEs will explore ways to use existing captive facilities. I cannot see the logic for such firms to set up their own captives. There isn't \n\nenough scale.CIO.com: Many of us think of the offshore captive center as a 20th century innovation. But you say it dates back to the seventeenth century \n\nwhen the East India Company first established its factories in India, recognizing the cost-effectiveness, flexibility, and viability of having a company foothold \n\nin the targeted trade country. Are captives here to stay?Oshri: I think it is here to stay but the concept is going to evolve over time.Ilan Oshri is associate professor of strategy and technology management at Erasmus University's Rotterdam School of Management in the \n\nNetherlands.