The Rational Outsourcing Blog

Wednesday, December 20, 2006

Why globalization does not mean your job will get outsourced to India

We live in a “flat world” where physical location does not matter and where jobs will migrate to the least expensive locations. An Indian worker may cost 60% less than an equivalent American worker. Does this mean your white collar job will necessarily move to India and that you might as well give up the fight and reconcile yourself to Ross Perot’s “giant sucking sound”? Not necessarily.

Labor costs are not a perfect indicator of overall costs. To make an apples-to-apples comparison between an American worker and an Indian worker we have to consider the total costs of their performing equivalent tasks. The three major additional components of total cost are: productivity, quality, and management overhead. Let’s tackle these in order:
  • Productivity: First, some basic math: If one person costs $10 / hr and takes two hours to do a task, while a second person costs $15 / hr and takes just one hour to do the same task, which would you rather employ? If you didn’t sleep through math, you realized that the first person costs $20 per task, which is more than the $15 per task that the second person costs. Thus, you would select the second person even though her hourly rate is higher. So, the question is: who is more productive, an American worker or an Indian worker? It is tough to say. What we can say for certain is that our productivity is under our control and can be improved by orders of magnitude through process innovation. My favorite example of this was described by Dr. Michael Hammer in his “Reengineering Work: Don't Automate, Obliterate,” article in the Harvard Business Review. He described how Mutual Benefit Life, an insurance company, transformed a customer application process from a typical turnaround of 5-25 days and a best effort time of 24 hours to a typical turnaround of 2-5 days and a best effort time of just 4 hours. Mutual achieved this by creating a new position called a case manager and empowering these workers to process entire applications instead of having applications “go through as many as 30 discrete steps, spanning 5 departments and involving 19 people.” “Case managers have total responsibility for an application from the time it is received to the time a policy is issued. Unlike clerks, who performed a fixed task repeatedly under the watchful gaze of a supervisor, case managers work autonomously. No more handoffs of files and responsibility, no more shuffling of customer inquiries.” The productivity gain achieved by this company would be more than sufficient to offset the 60% difference in labor costs between Indian and American workers. Moreover, the job of the case manager is far more complex and “high-touch” than the tasks performed by the original clerks and is far less likely to be outsourced in the future. Dr. Hammer concludes his article with “We must have the boldness to imagine taking 78 days out of an 80-day turnaround time, cutting 75% of overhead, and eliminating 80% of errors.” Such boldness and innovation are far greater assets than 60% labor cost differentials.
  • Quality: Well, you guessed it, time for some more math. Let’s say processing an insurance claim correctly costs $1. How much do you think correcting an error in an insurance claim costs? Well, once you add up the cost of quality control, the call center costs for fielding customer complaints, and the cost of reissuing a corrected claim, the costs climb quite high. Let’s say the cost is $100. If the error rate is 3%, then this company would be spending three times as much on the downstream cost of errors as on the original claim processing costs. Thus a 1% change in error rates would have the same total cost impact as a 3% difference in labor costs. The specific ratio of the cost of errors to original processing costs varies from case to case; however, the cost of errors is almost always greater than the processing costs. Thus, output quality is almost always more important than labor cost. In other words, if outsourcing increases your error rate even slightly, it can wipe out the benefits of lower labor costs. In one case, an US insurance company outsourced its claims processing to a BPO vendor that delivered 30% lower claims processing costs. Unfortunately, the vendor’s error rate was also slightly higher than the customer’s: just 1.1% instead of the original 1.0% error rate. Such a small 10% difference in error rates seems trivial, but it is sufficient to wipe out the benefits of the 30% cost difference from outsourcing. [For details, see the Case Study at]
  • Management and training overhead: It is not easy to manage a task from across the world. You need proper oversight mechanisms which often require expensive international travel, managers who are trained in cross-cultural interactions, information security safeguards, etc. Other factors such as the high employee churn rate in India and the resultant training costs contribute to management overhead as well. In general, due to relatively high management overhead the overall cost benefits of outsourcing to India are often reduced to just 10 to 20%.

Once we consider the total cost differences between an American worker and an Indian worker, we find that the 60% labor cost differential is much less important than all the other factors that are included in the total cost. Thus, if American workers focus on their productivity and quality while leveraging their inherent management overhead advantages, they can effectively compete against Indian workers notwithstanding the labor cost difference. The unfortunate reality is that instead of focusing on these goals Americans are focusing on protectionism. In the meantime, Indians are focusing on improving these exact same factors. Let’s revisit them again:

  • Productivity: If you visit a major Indian Business Process Outsourcing vendor you will be amazed by the way it manages its productivity. In many ways the major Indian firms have replicated the assembly line in a business process environment. While Indian firms aggressively adopt methodologies often invented in America, US firms are beginning to lag behind the Indians in process improvement in the service industry.
  • Quality: The quality of Indian providers varies widely. I have evaluated vendors who had critical errors in more than 17% of processed documents, and I have evaluated others that demonstrated less than 0.5% errors. What is uniformly true is that most major Indian firms are investing heavily on quality improvement methodologies and software.
  • Management and training overhead: While employee churn remains a significant problem, Indian firms have come a long way in tackling this problem through improved training systems. They are also beginning to invest in business service operations management software and some of the leading firms have even created home-grown management software. The most obvious change is in the corporate cultures of the larger firms. In 1999, I remember being surprised by the lack of sophistication of many Indian outsourcers. Today when I deal with the larger Indian vendors, it is easy to imagine that the meeting is taking place in New York or London and the vendor’s managers are invariably steeped in western corporate culture.

When the Japanese motorcycle manufacturers first entered the US market, the dominant British manufacturers laughed at them. The larger Japanese motorcycles leaked oil, “looked ridiculous,” and broke down regularly. The smaller Honda Cubs were considered “toys” by serious motorcycle enthusiasts. Their only advantage was that they were cheaper than British motorcycles. While the British laughed, the Japanese improved their motorcycles until they essentially drove the British out of the motorcycle business. It is true that Indian BPO vendors still have many problems with employee churn, security, infrastructure, quality; and today I truly believe Americans could give Indians a serious run for their money based on the overall cost of business processes. However, if Americans remain distracted by protectionism they will lose the chance to improve themselves and compete fairly for their slice of the global business process market.

I have a fierce belief in the inherent abilities of Indians and Americans. Americans today believe that the game is unfairly stacked against them due to the low Indian labor costs and they are essentially refusing to play the game. You can’t win a game that you don’t show up for! Of course Americans can’t beat the Indians on labor costs, but these costs are only a small portion of overall costs. Americans need to rejoin the game and figure out how their inherent strengths in innovation, management / training overhead and possibly quality can counter the core strengths of the Indians in productivity and costs. That would be a fair match worth competing in, and may the one with the lowest overall cost win.

Labels: , , , ,

Tuesday, December 19, 2006

The outsourcing bogeyman is the real threat to the US economy

It seems I managed to confuse some of my readers with my "Why a US ban on offshore outsourcing is the best possible thing for India!" post. I was using sarcasm to make a point, and am not really suggesting that the US ban outsourcing. Any ban on outsourcing would harm both countries overall, but would be especially harmful to the US. Let me make the same point using economic data rather than humor:

Right now the focus is on Business Process Outsourcing (BPO), however, during 1998 to 2002 everyone was convinced that IT outsourcing (ITO) to India would gut the American IT services industry. Let us consider the IT services most directly affected by offshore outsourcing, namely: “computer and data processing services” and “data base and other information services.” According to the “Digital Economy 2003” report published by the U.S. Department of Commerce, US imports of these services rose from $0.3 billion in 1995 to $1.2 billion in 2002 (with a peak of $1.6 billion in 2000). This seems to be clear proof of Ross Perot’s “giant sucking sound” and any number of Lou Dobbs’ tirades. However, in the same time period, US exports of these services rose from $2.4 billion in 1995 to $5.4 billion in 2002 (with a peak of $5.7 billion in 2000). Thus, the U.S. trade surplus in these services expanded from $2.1 billion to $4.2 billion over the same years when the US faced the greatest threat from Indian outsourcing firms due to the Y2K contracts and the Internet boom.

If the US had magically managed to “ban IT offshoring” and other countries had done the same, the US economy would have lost $26.2 billion over these 8 years. How many US jobs do you think that $26.2 billion translates to? I realize data provides cold comfort to people who have lost their jobs due to outsourcing. However, the above analysis highlights how overall global trade creates far more jobs in America than it destroys. If you don’t believe the analysis above, you should at least learn from US history. The following excerpt from the US Department of State website highlights how US protectionism contributed to and exacerbated the Great Depression.
The Smoot-Hawley Tariff Act of June 1930 raised U.S. tariffs to historically high levels. The original intention behind the legislation was to increase the protection afforded domestic farmers against foreign agricultural imports. … But once the tariff schedule revision process got started, it proved impossible to stop. Calls for increased protection flooded in from industrial sector special interest groups and soon a bill meant to provide relief for farmers became a means to raise tariffs in all sectors of the economy. When the dust had settled, Congress had agreed to tariff levels that exceeded the already high rates established by the 1922 Fordney-McCumber Act and represented among the most protectionist tariffs in U.S. history.

The Smoot-Hawley Tariff was more a consequence of the onset of the Great Depression than an initial cause. But while the tariff might not have caused the Depression, it certainly did not make it any better. It provoked a storm of foreign retaliatory measures and came to stand as a symbol of the ‘beggar-thy-neighbor’ policies (policies designed to improve one’s own lot at the expense of that of others) of the 1930s. Such policies contributed to a drastic decline in international trade. For example, U.S. imports from Europe declined from a 1929 high of $1,334 million to just $390 million in 1932, while U.S. exports to Europe fell from $2,341 million in 1929 to $784 million in 1932. Overall, world trade declined by some 66% between 1929 and 1934.

Thus, between 1929 and 1934, US imports were reduced by just $944 million while exports were reduced by $1,557 million and so the US economy lost hundreds of millions of dollars and more importantly tens of thousands of jobs due to US protectionism. Moreover, today due to the high growth rate of countries like India and China, the rest of the world counts for a significantly greater proportion of global economic growth than they did in the 1930s. As such, US protectionism would probably harm the US even more in the current environment because American companies would be locked out of the rapidly growing Asian economies while Asian and European companies would probably benefit from the vacuum created by the absence of American companies.

Daniel W. Drezner in his thought-provoking article titled “The Outsourcing Bogeyman” (Foreign Affairs, May/June 2004) provides two excellent examples on how US protectionism has already caused US job losses.
Consider the example of candy-cane manufacturers: despite the fact that 90 percent of the world's candy canes are consumed in the United States, manufacturers have sent much of their production south of the border in the past five years. The attraction of moving abroad, however, has little to do with low wages and much to do with protectionism. U.S. quotas on sugar imports have, in recent years, caused the domestic price of sugar to become 350 percent higher than world market prices. As candy makers have relocated production to countries where sugar is cheaper, between 7,500 and 10,000 workers in the Midwest have lost their jobs -- victims not of outsourcing but of the kind of protectionism called for by outsourcing's critics.

A similar story can be told of the steel tariffs that the Bush administration foolishly imposed from March 2002 until December 2003 (when a ruling by the World Trade Organization prompted their cancellation). The tariffs were allegedly meant to protect steelworkers. But in the United States, steel users employ roughly 40 times more people than do steel producers. Thus, according to estimates by the Institute for International Economics, between 45,000 and 75,000 jobs were lost because higher steel prices made U.S. steel-using industries less competitive.

Proponents of American protectionism should keep such recent misadventures in mind while pushing for higher tariffs.

US protectionism impacts more than just the economy. The previously quoted US Department of State article goes on to say: “More generally, Smoot-Hawley did nothing to foster trust and cooperation among nations in either the political or economic realm during a perilous era in international relations.” Does this sound eerily applicable in the current international relations context? George Santayana wrote: 'Those who cannot remember the past are condemned to repeat it.' Unfortunately, the cost of repeating this past mistake could be greater than the anti-globalization brigade can even imagine.

Labels: , , , , , ,

Friday, December 08, 2006

Nasscom’s new Data Security Watchdog: it has bark, will it have bite?

It seems Nasscom is trying to address the information security perception problem in India by creating a new Data Security Watchdog (as reported in CIO India).
The National Association of Software and Services Companies (Nasscom) is setting up a watchdog organization that will focus on the introduction and monitoring of best data security and privacy practices in the country's IT services, call center and business process outsourcing industries. The move is one of several measures by Nasscom and the IT industry to strengthen data security and privacy in the Indian call center and BPO industries.

"We are planning a self-regulatory organization (SRO) that will be initially set up by Nasscom, but will operate independently with an independent chief executive officer and board," said Sunil Mehta, vice president of Nasscom in Delhi.

"Being a member of the SRO will in effect be a certification, as member companies will have to follow the best practices specified by the SRO," he said.

Besides setting benchmarks and training companies on the best data protection and data privacy practices, the new organization will also have the authority to punish and expel erring member companies

The SRO will be funded for one year by Nasscom, which has budgeted Rs 1.35 crore for the purpose. After the first year, the SRO is expected to finance itself from membership, training, and audit fees.

This sounds like a great idea, especially as this organization can become a forum for sharing information security best practices. I have been impressed by the information security and fraud detection methodologies used by some Indian vendors and if they start helping each other they can improve even more rapidly. I am a firm believer in incentives, and here I think the outsourcing vendors’ incentives are properly aligned: When one Indian outsourcer has an information security or fraud problem, every Indian outsourcing vendor suffers from the negative press. Japanese manufacturers helped each other build the “made in Japan = quality” perception, Indian outsourcing firms have to do likewise.

I am however not sanguine about the incentives for the enforcement component of this watchdog. After the first year, the watchdog will be funded by the dues paid by its membership and the only way for it to punish a member would be to “expel” the “erring member” and thus lose their “membership, training, and audit fees.” This sounds like classic incentive misalignment. I hope that the Nasscom leadership will address this problem before the organization goes live. Perhaps the organization could be funded by the outsourcing customers instead? Rs 1.35 crore (approximately $300,000) split among even thirty large outsourcing customers sounds like a very good investment. If that $10,000 a year helps them avoid a single information security breach, or more likely a PR headache, it would be money well-spent.

Labels: , , , ,

Thursday, December 07, 2006

Who would you trust with your credit card number: an Indian college grad, or an American felon?

Recently there have been a series of articles on Indian companies’ problems with information security and fraud. However, the BPO vendors I personally evaluated in India tended to have as good if not better information security procedures than most US companies. I was recently semi-joking with the CEO of an Indian BPO that workers in California would never accept the kind of restrictions Indian companies regularly place on their employees to prevent information theft. Some of these restrictions (such as keystroke monitoring) may even be illegal in countries like Germany. However, right now when it comes to outsourcing, perception is reality and the pervasive perception is that India has a data security problem.

I must admit, I am a bit confused by this. If Americans are OK with felons in US prisons accessing their information, surely they would be OK with trusting a college graduate in India? Strange as this sounds, I am not making this up. In July 2004, USA Today reported:
About a dozen states — Oregon, Arizona, California and Iowa, among others — have call centers in state and federal prisons, underscoring a push to employ inmates in telemarketing jobs that might otherwise go to low-wage countries such as India and the Philippines. Arizona prisoners make business calls, as do inmates in Oklahoma. A call center for the DMV is run out of an all-female prison in Oregon.

At least 2,000 inmates nationwide work in call centers, and that number is rising as companies seek cheap labor without incurring the wrath of politicians and unions. At the same time, prison populations are ballooning, offering U.S. companies another way to slash costs.

As expected, there are some “information security problems” with using prison labor in call centers:
executives shudder at the prospects of inmates sharing the personal information of customers with fellow prisoners, as some did in Utah in 2000.

An article from NPR comments:
Labor unions and some states say they believe it's too much of a security risk to have prisoners talking to the outside world, even if they're being monitored. Private businesses say it's also a security risk to have prisoners taking down customers' credit card information.

Maybe it is just me, but it seems that if information security outcries restrict offshore outsourcing, companies are more likely to shift the work to similarly priced prison-based call centers than to hire American workers who even at the minimum legal wage are several times more expensive. To quote the UNICOR Federal Prison Industries website: “Imagine... All the benefits of domestic outsourcing at off shore prices. It's the best kept secret in outsourcing!” By the way, you also have to see the slick marketing video on their website. Maybe it is just me, but I would prefer to share my credit card information with an Indian college graduate than with a felon. [To be fair, I am sure UNICOR works hard to restrict their operators access to private information, as do Indian BPOs. The problem arises when the system does not work as planned.]

I am not trying to belittle the information security problem in India. It exists, just like similar problems exist in the US, and needs to be addressed proactively. But, let’s please take the politics out of business decisions and stick to Rational Outsourcing.

Labels: , , , ,