If there's been one constant during the years that Dr. Arie Lewin has studied the corporate offshoring experience, it's this: Very few companies understand \n\nthe strategic value of global sourcing. Even today, just five percent of companies that offshore IT services do it really well, says Levin, professor of strategy \n\nand international business at Duke University's Fuqua School of Business and director of its Center for International Business Education and Research \n\n(CIBER).According to results from the sixth annual offshoring survey conducted by CIBER in conjunction with The Conference Board, average cost savings \n\nachieved by offshoring (both third-party outsourcing and captive center operations) has declined consistently during the last five years, in part because just a \n\nhandful of companies have an optimized enterprise-wide sourcing strategy. And despite all the knowledge readily available about offshoring management \n\nrequirements, companies continue to get tripped up by the so-called "hidden" costs of global sourcing.Nonetheless, more than half of the companies surveyed expect to expand their offshoring initiatives during the next 18 to 36 months, the study found. \n\nNearly a third have already moved up the value chain to offshore innovation-related services.CIO.com talked to Dr. Lewin about eroding offshore labor arbitrage advantage, the effect of offshoring on domestic IT job opportunities, how \n\ncompanies who succeed at offshoring look beyond cost cutting, and why executives continue to be surprised by the management overhead required to \n\nmake offshoring work.CIO.com: You found that the average savings yielded by offshoring IT services has decreased from around 38 percent a decade ago to 27 \n\npercent today. To what do you attribute that decline?Dr. Arie Y. Lewin: First there is the fact that the labor arbitrage effect is eroding, and there's nothing you can do about it. There are \n\ninflationary cost pressures in offshoring across most countries and especially at hot spots like Bangalore. For most locations labor arbitrage dissipates over \n\nthree years. If you are only counting on labor arbitrage you will be disappointed.The second factor is whether you are doing captive or third party outsourcing. Most large IT services offshoring is done via a third party. On the \n\nexternal provider side of the equation, [many] are getting better at what they do, which allows the most efficient to charge at market rates but make higher \n\nprofits. An important source of savings is excelling in process optimization. One company that is at the state of the art with process optimization and \n\nprocess reengineering is the Penske truck rental company. Penske lives by and breathes process optimization. They're perhaps the only example that \n\n[offshore outsourcing provider] Genpact can brag about. They understand that process reengineering means steep jump in savings.CIO.com: Well, 27 percent savings on traditional IT services is still nothing to sneeze at-particularly today when every dollar counts. But you \n\npropose that cost reduction alone is no longer enough to justify moving operations offshore \u2014 that companies need a "multidimensional value \n\nproposition."Lewin: Companies that stand out in our database are those that have implemented a corporate-wide strategy to guide offshoring decisions. \n\nFor example, they already had in place strategy for tier two and tier three cities in India. They are among early companies locating in China. They were first \n\nin demanding that their providers provide nearshore options.Unfortunately very few companies understand the strategic value of global sourcing. They think of [offshoring management] as a core capability, not \n\nsomething to patch on. This is a discipline. You can't do it just from one day to the next. You need to be thinking through the next thing that you'll be ready \n\nfor. Think "cloud." Now that cloud is here, will you be a company that wait until it is perfect or be an early adopter on a fast learning trajectory? CIO.com: You found that as companies expand offshoring activities, they actually see a decline in overall efficiency, likely attributed to a loss \n\nof managerial control or the "hidden costs" involved such as training, staff recruitment and retention, and government and vendor relations. Why are \n\ncompanies still blindsided by the overhead required to make offshoring work?Lewin: There are many companies just starting offshoring. But it is amazing to me that somehow they started down this path, and they don't \n\nrecognize what they're getting into.Two companies asked me to come in after they commissioned a McKinsey report [on their offshore option]. The report outlines all the important \n\ndetails about hidden costs. But most of the executives did not read or comprehend beyond the first two or three sections of the report. Much of the \n\nimplementation process was one of trial-and-error discovery.The bigger source of inefficiency is in companies in which offshoring gets C-level attention. They lay out guidelines and risks and may also prioritize \n\nstrategic drivers. But they don't direct attention to creating and internal organization that can manage and coordinate the global sourcing strategy. So the \n\ndifferent and functions find themselves reinventing the wheel. There's not a central organizational unit\u2014a center of excellence for global sourcing of business services\u2014that, for example, does the \n\nvendor and country selection so that the business unit doing the outsourcing only has to deal with processes that they want to outsource. When you have an \n\noverall strategy, but much is still left to the business or function manager, there's going to be inefficiency.When it comes to IT services offshoring, only five percent of companies are really phenomenal.CIO.com: The perceived connection between offshoring and domestic job loss has driven a great deal of anti-offshoring sentiment in the U.S. \n\nparticularly given the current levels of unemployment in America. Yet just 34 percent of your respondents said offshoring had resulted \nin local layoffs.Lewin: The more telling finding is the ratio of offshore employment to domestic employees. For most industries, it's a small ratio. But for \n\nsoftware development, the ratio is very high. To me, this is a worrying signal. At time when unemployment is high, companies report that access to talent is \n\nthe reason they are offshoring.There are at least two factors at play. Small, entrepreneurial firms are far more likely to offshore innovation services than large companies. They need \n\nengineers. In some cases, the skills they need\u2014like VSLI [very-large-scale integration] design\u2014no longer exist in the U.S. In other cases, the \n\nventure capitalists ask the new firm, 'What is your innovation outsourcing strategy?' So even though the talent might exist in the U.S., the VCs believe that \n\nfor speed to market, cost, or some other reason, the start-up firm ought to find a way to get the work done abroad. And they find that the passion and \n\ncommitment to get work done offshore is higher than hiring someone in-house or on a part-time basis.Secondly, there is a structural shift going on in the employment of technical talent. Companies are learning to substitute full-time positions with hiring \n\nengineers, programmers, and technical personnel on-demand for a particular project. That decreases the desirability of the profession to young people.Another problem is that American companies no longer invest in keeping up the skills of their technical personnel. Unlike other countries who \n\nunderstand the importance of keeping up the technical skills of their people, they'd rather let them go. So unless the affected employees invest in \n\nthemselves, at some point in their careers they become not employable. We used to have a proud tradition of companies investing in maintaining their human capital. Some of the things I see are crazy. We have an example of \n\na major technology company in this country, that rather than hire new people to replace those who are retiring, are engaging an Indian engineering company \n\nto in source their engineers do the work. When those people leave, where does that intellectual property go? Back to India in the brains of those people. \n\nSuch practices only solve a short-term problem.From a long-term point of view, the statistics are clear that fewer Americans are entering science and engineering careers as indicated by the number \n\nof advanced degrees awarded. The trend began in 1995. Part of our research approach is to look for "small, initiating events" whose impact may not be \n\nseen for a number of years. In 2003, Congress allowed the H-1B quota to lapse which greatly reduced number highly skilled workers that companies could \n\nbring in. The H-1B shortage was about 130,000 visas. In addition the cumulative shortage of American nationals earning advanced degrees in science and \n\nengineering reached about 49,000. Indeed, in 2004, the unemployment rates in almost all engineering subfields were at historical lows. In 2008, something else has happened. Companies began to realize the strategic importance of entering new markets like Brazil and China and they \n\nneeded learn how to do product development and other work in those countries. If you're Boeing, where is your source of major top line growth? It's not \n\nin the United States. Companies are facing the challenge of creating an organization and processes that are aligned with their new markets.CIO.com: When you asked service providers about their biggest concerns, pressure on margins was number one. Should that concern \n\ncustomers at all?Lewin: Customers will benefit greatly. The new variables [putting pressure on provider profits] are that some countries have developed \n\nnational policies to attract the outsourcing industry. Go to Dalian, China and there are at least 200,000 Japanese-speaking Chinese doing work for \n\nJapanese companies. According to KPMG, the current value of executed contracts is $5 billion\u2014just in Dalian. Many countries are following the \n\nlead of China to establish national aspirations to attract the outsourcing industry\u2014Sri Lanka, Morocco, and Egypt. The competition is not just among providers, but also between countries. At a new tech park being built outside of Shanghai, a deputy mayor told us \n\nthat a million people will be conducting back-office work for financial services companies within five years. In India, the total number of employees directly \n\nemployed in the provider industry is only about one million people.CIO.com: Companies are starting moving up the value chain with their offshoring\u201432 percent of your participants indicated that they \n\nnow offshore innovation services. But it's not clear what exactly is driving that. You found that 88 percent of those who offshore innovation services say \n\naccess to qualified offshore talent is the driver, but less than a third say a domestic shortage of qualified personnel is a driver. And, as you note, \n\nunemployment rates in the U.S. have reached historical highs due to the recession. So what's going on here?Lewin: It's important to point out that the vast majority of engineering work\u201486 percent, according to a study by [India's IT services \n\ntrade group] NASSCOM and [consulting firm] Booz Allen Hamilton\u2014is still done in the Western world.But you have to watch for those small, initiating events. Those companies that let their engineers retire and didn't replace them are an early trend. What \n\nare all the factors that are going to drive innovation offshoring? I don't have a good answer for that. Different companies and different industries have \n\ndifferent reasons.It's all part of an emerging trend that we call the disassembly and reassembly of the organization. If you're the CIO, you're always in the hot seat \n\nbecause of the high cost of IT and because it is so highly visible at the level of C-suite. The CEO must be always questioning these costs. The pressure is \n\nalways on. But there's still a lot to be done offshore. This is not the end of the story by any means.