Over the past few weeks I have received a tremendous amount of hate mail regarding my column The coming death of Indian outsourcing. It seems to have catalysed debate on the problems facing the Indian outsourcing sector, such as annual 15 per cent salary hikes and 20 per cent attrition.

What India needs to do to survive as the world's back office boils down to this: It must first adjust its value system, and then it must broaden and strengthen its education programs to increase the supply of trained professionals.

Video: India's outsourcing fate
Slideshow: Fastest-growing tech companies

Let's start with value systems: India has created a generation of terribly mercenary young information technology workers. Young professionals in Bangalore get numerous job offers a month. They switch jobs all the time, and they don't learn to build anything. Employers are now in the business of buying and selling people in the open market, and Indian youths are squandering the best years of their lives hopping from one company to another.

For such young Indians, I have one piece of career advice: Your resume is becoming unmarketable. If you sit across the interview table from me, I would not hire you. Nor would anyone else worthy of your respect.

And to employers, I would say this: If you're fueling this behavior, you are shooting yourself in the foot.

Of course, the shortage of skilled professionals is acute. The world's hunger for Indian back-office workers is insatiable--for the moment. Demand far exceeds supply, so whatever supply exists is getting bid up in price. This is Economics 101.

Slideshows:
Where India outsources
World's 10 hottest job markets

Basic economics also offers the solution: Increase the supply.

How?

Spread things out. Move away from the big cities and take advantage of the heartland of India. Bring work to people who want to stay close to their families, speak their own language and eat their own food. Smaller cities. Small towns. Even villages.

It is a strategy that has been obvious to most industry leaders for a long time but has not yet been adopted widely.

UST Global is one of the rare outsourcing companies that is following this approach.

"We have stayed away from the model of building mega centers and built our operations in Kerala," says president Sajan Pillai. "We have 3,000 people in Trivandrum and 1,000 in Kochi. Our attrition is less than 10 per cent. We believe in not making social displacement. We will continue to grow in a hub-and-spoke model, with lots of centers in the tier-three cities."

Slideshows:
America's best states for tech jobs
America's worst states for tech jobs

An interesting experiment is being conducted by Satyam Computer Services' [Get Quote] founder, Ramalinga Raju. His philanthropic endeavor GramIT aims to ship business-process outsourcing projects to some of the 600,000 odd villages in India, dropping the cost-structure further while creating jobs in India's poverty-stricken heartland. GramIT employees earn $1,000 a year vs. the $10,000 to $15,000 that urban employees at outsourcing companies make.

Venkatesh Roddam, chief executive of Satyam's business process outsourcing business, which outsources work to GramIT, describes the GramIT workers as 'exceptionally motivated'. They will, he says, 'wade through heavy rain to come to work'.

That's a world apart from Bangalore.

GramIT's attrition rate is close to zero. Employees successfully take on jobs such as data entry and transaction processing. Seven such centers are operating in rural Andhra Pradesh and 10 more are slated to start up in the next few months. Each center averages 70 to 100 employees and is often the pride of the village.

GramIT's model can become the survival formula for the low end of the Indian outsourcing industry. If the cost structure is pegged at the right level, India can also do work for other emerging markets. And, more important, India can do work for itself.

The government should put in place incentives to ramp up this distributed development model. And it needs to be adopted commercially by major BPO vendors, because rather than just being a philanthropic experiment, it makes business sense.

Where this formula fails, however, is the call center industry, where conversational English requirements are paramount. The average salary hike in the call-center industry was "only" 14.8 per cent in 2007, down from 17.2 per cent the previous year. (The Indian call center industry employs about 350,000 people.)

It could also become a ripe target for US companies seeking to move jobs elsewhere. Indian employees sleep through the day and work through the night. Burnouts abound. This is socially unsustainable for India.

The next important issue is education. India's education system below the top tier is deplorable. There are 1,346 degree-granting engineering colleges in India with an annual student intake of 440,000, plus 1,244 polytechnics with an annual intake of 265,000.

These institutions are facing tremendous faculty shortages, and graduates have questionable mastery of their respective disciplines. Taking advantage of remote learning could be one part of the solution. Wide adoption of MIT's Open Courseware content, which makes lectures by institute faculty widely available, would be a step in the right direction.

Skyrocketing salaries and attrition would automatically be reduced if the number of IT and IT-enabled services workers grew from 4 million to 40 million, especially if many of those new workers lived and worked throughout the heartland of India. Distributed supply--matched with distributed demand--could create a truly sustainable model for India's development.

With that, the world's back office will survive.