On Wednesday, CIO magazine's Stephanie Overby posted a fascinating interview  with Arjun Sethi, partner and head of the outsourcing practice at A.T. Kearney  consultancy. Sethi believes that the rise of cloud computing and the move from on-premise to hosted IT services in the mainstream enterprise will render the majority of the current outsourced IT workload--programming, maintenance, and system integration--extinct, perhaps in as soon as five years.
For those who equate "outsourcing" with "unemployment," the reaction to this prediction is probably jubilation. Cloud computing has the potential to restore American IT and IT services giants HP, Accenture, and IBM to peerless prominence, elevate Amazon, Google, and perhaps some heretofore-unknowns into the forefront of a new industry, and lift the fortunes of software companies like Microsoft, SAP, and Oracle if they are able to pivot their business models from licenses to access fees.
There are even some compelling efficiencies to hosting the cloud data centers in the U.S. rather than overseas, creating viable knowledge economies in domestic rural communities and post-industrial city centers rather than in Bangalore and Pune. Hooray for the USA!
But are the celebrations premature? Cloud computing today is just past its infancy. Security and network service problems still occur. Enterprises are reluctant to turn the keys to their mission-critical systems and their proprietary data over to remote providers. Large amounts of data still reside on legacy systems that can't be migrated to the cloud, and which still require the kind of care and feeding that only dedicated IT services staff--either employed or outsourced--can provide.
Maybe we'll have solved these problems in 4-5 years. But then again, 15 years of dedicated investment and innovation have not put an end to email spam, despite the rosy predictions of industry leaders. Cloud computing is not an area where 90 or 95 or 98 percent service quality is "good enough." The costs of a security breach or a service outage are too high, particularly in the high-revenue enterprise space where collateral damage may include brand reputation and hard-won customer confidence. The progress we make in the next five years will have to be comprehensive, not just significant, to sway risk-averse CIOs.
But let's say Sethi's forecasts are largely correct. That means that parts of the world that have just become accustomed to the income potential of the outsourcing industry will have to become unaccustomed in a hurry.
NASSCOM , the association of the Indian business process and IT services industry, estimated  that outsourcing was worth $71.7 billion in 2008, or 5.8% of GDP. Software and services revenue was about $60 billion. Worldwide, outsourcing was a $1.6 trillion business in 2007. New players in Southeast Asia, Latin America, Eastern Europe and Africa are coming on strong as they build greater workforce capacity through training and equipment.
These businesses and industries which have demonstrated remarkable entrepreneurial zeal and transformative potential for their countries, will need to pivot quickly to provide services appropriate to a world where the cloud, not the on-premise data center, sits at the center of their clients' IT universe.
This could prove to be a challenge in places like India. Top-tier outsourcers such as Wipro and Infosys have succeeded by harnessing the massive scale and comparatively low costs of Indian engineering talent, throwing lots and lots of bodies (and brains) at IT and programming challenges. They have thrived by taking an existing business model and running with it, but they are largely unproven as strategic innovators.
If their familiar world is suddenly vaporized by the gathering storm of cloud computing, it is highly uncertain whether they can discover the agility to reinvent their value proposition. And if that happens, what becomes of the promise of the overseas knowledge economy, and its prospects of creating prosperity and stability for a rising global middle class?