i've recently read Stephen Cohen and Brad DeLong's Thinking About Outsourcing, an excellent perspective on the issue. with my first-person perspective on hi-tech outsourcing, i have generally seen it fail badly. the lack of a legal framework for ip and information safeguards in many nations, the time zone and cultural differences, the communication barriers, and the lack of control from HQ all conspire to make the outsourcing strategy fail. this is especially acture where the work involved requires a high level of expertise and in-depth, hands-on experience.
i have worked at two organisations which had outsourced software development efforts for enterprise products, and both efforts were dismal failures.
one was with a team in bangalore, india; this team was erudite and possessed the necessary acumen for the work, but the other factors listed above doomed the effort. there was never enough presence from the architects and designers of the products at the remote site to keep the project on track, and despite daily conference calls and emails aimed at ensuring maximum communication between both sides, it always took at least a day to turn over even the minimum tasks or research. the work they produced didn't appropriately meet requirements, even though the coding was sound, and they always had to be pushed to ask the appropriate questions. their impl had to be sent back to be redone more than once, and in the end still required rework from the engineers in the home sites.
another was with a team in singapore. where bangalore is a tech centre, singapore is most definitely not (you can find a pool of talented investment bankers there, but your choice of hi-tech professionals is quite limited). the company could not find the appropriate personnel for the projects to be completed in singapore, and they managed to botch every project sent their way; every one required considerable engineering intervention from the engineers in the home sites. the company eventually decided to shut down the outsourcing effort and bring back all development to the U.S. (but quietly, unfortunately, so the illusory allure of outsourcing was never impacted).
in both cases, there were no actual cost savings for the companies involved. more money was spent in the outsourcing effort than would have been spent keeping the same work at the home sites. however, both claimed cost savings from a line item basis (the usual U.S. corporate technique used to hide failure).
although this isn't a statstically relevant sample, it does highlight how assuming outsourcing success using present metrics is poorly based. salary rate arbitrage and employment tax evasion - the main benefits of outsourcing - do not guarantee operative success, and ip based endeavours always involve barriers in expertise, practises, process, and communication. outsourcing is not the panacea as claimed by U.S. executives; it is simply a way to continue hiding the cost of doing business in an ambiguous expense category, and persist tried and true pennywise/poundfoolish management practises.
however, i have worked closely with two organisations which have successfully made use of offshore resources. these did not pursue outsourcing, but a globalisation strategy, where they invested and developed technology centres across the globe - over a number of years - to tap local research, development and marketing expertise. in both cases, the companies had these centres focus on specific technology areas, and did not unduly over-divide responsibilities. for example, one company made its bangalore centre entirely responsible for the implementation of a particular EAI stack, following specifications which were developed in north america, japan and europe, taking advantage of the communication and cultural infrastructure it had spent years developing.
what was the difference between these two globalising companies and the other two outsourcers? the former focussed on a long term strategy which would best leverage talent pools and resources at each centre, while the latter focussed on a short term strategy to cut salaries and other business costs.
this latter trend is most troubling, because organisations following it are in essence looking to exploit a foreign workforce through salary arbitrage. the focus is on slashing costs for the quarter and fiscal year and not on global presence. as soon as the workforce currently used becomes too 'expensive' for the organisation's taste, it picks up and moves elsewhere. there is no overall strategy other than to preserve executive compensation (one of the two outsource failures i mentioned above cut 15% of its workforce - in essence taking it out of the market in several product offerings - in the space of 6 months, while giving its CEO a 10% pay raise, where that money could have paid for those employees for more than a year and kept the company in lucrative markets).
another disturbing effect of this trend is the negative impact the development of U.S. hi-tech talent. the less jobs there are in the U.S. for hi-tech professionals, the less incentives university students will have to enter hi-tech fields. and the so-called 'higher-level' tech jobs (architecture & design) are not available to university graduates without considerable work experience. where are these grads going to get such experience? China? India? not bloody likely, where U.S. citizens have fewer chances of getting a work permit than Chinese or Indian citizens have in the U.S. flooding these markets with foreign workers would kill their hi-tech industries.
the last recession ('90-'93) illustrated that the U.S. educational system had a steep drop off in hi-tech grads and did not recover until 1998, when the boom had already been running for at least two years (the boom didn't really start until 1996). of the '90-'93 hi-tech undergrads, it has been estimated that 50% were lost to other fields/industries because no jobs were available to them until late 1994; until late '94 5-8 years experience were required for even the most basic jobs. approximately 25% of hi-tech masters/doctorate grads were lost to other fields/industries during this period for the same reasons.
what is worse in the '01-'04 recession is that the jobs from which required experience is developed won't be around in large part in the U.S.. U.S. citizens are largely barred from jobs in the hot foreign hi-tech markets unless they are hired by U.S. companies. yet U.S. companies have shown a disturbing trend of not hiring citizens for such jobs because they can still demand higher salaries than foreign workers. it is a vicious cycle, and the portents are not good for U.S. hi-tech and r&d primacy.
however, it may take the loss of this primacy to motivate the rank and file in the U.S. to act. presentlly, it seems there is not enough motivation among the electorate to enact any substantive change. the votes - whether actual or not - keep going to politicians who only serve to increase the incentives for outsourcing. the election of Schwarzenegger as governor of California was particularly distressing: California in particular has undergone an especially grim experience in the '01-'04 recession, yet the electorate blissfully continues to place political candidates who only serve to exacerbate it's economic woes. this political naïveté goes beyond 'panem et circenses', since the electorate continues to directly place (or at least believe that they are) into power those who are openly acting against their interests.
at one point or another, it will become apparent that U.S. corporate executives neither have a G-d given right to a job, nor a G-d given right to disproportional compensation, especially at the risk of shareholders, employees and the industry. hopefully this will happen before the U.S. experiences further economic damage, but i doubt it.
