The politics of offshoring: all talk, no action

Key Insights (9): There are few options for developed countries to respond to this trend.

Continuing with the Key Insights list, but jumping ahead a bit. I did a guest post a week or so ago on Phil Ferst's Horses for Sources blog. Since the post is already written, I will jump ahead to Key Insight #9. Back to numbers 2, 3, and 4 in the coming days.

--- ---

Picking up on Phil’s April Fool’s day post, I wanted to share a few thoughts on why we see lots of anti-offshoring rhetoric from politicians, but (thankfully) very little actual policy.

There is certainly a heavy demand from the man on the street to “do something.” It’s easy to understand why. We have become accustomed to trade in manufactured goods and natural resources. But manufacturing accounts for only about 15% of US employment, and developing countries generally enjoy a fully-delivered cost advantage of 10-30%. This is disconcerting for developed country workers, and we frequently observe moves against trade in manufactured goods, such as spurious anti-dumping actions and “buy America” provisions in various pieces of legislation.

Services, by contrast, account for about 78% of U.S. employment, and developing countries enjoy delivered cost savings of 40-70%.

So developing countries’ advantage is much larger in services, and many more people are potentially affected. That explains the political heat and the politicians’ rhetoric, but not the lack of policy.

So, why haven’t we seen offshoring regulations? Because trade in services is incredibly difficult to regulate. Regulating trade effectively requires that two conditions exist:

1. The government can observe what actions firms are taking, and

2. Any proposed policy must be credible and enforceable.

To see why regulating offshoring is so difficult, compare the “offshoring” of auto components with that of IT services.

Auto parts -- Trade displacement in manufacturing is easy to observe. If Delphi closes a component plant in Michigan and opens one in Mexico, it is easy to see what happened. 500 Mexican workers are now doing the same tasks in the same way that the Michigan workers did. Production from the Mexican plant now goes to the customers formerly serviced by the Michigan plant. It is easy and accurate to conclude that the Mexican workers replaced the Michigan workers.

Second, if the Congress chooses to regulate this trade, it is fairly easy to do. Don’t let trucks from the plant cross the U.S. – Mexico border, or slap a tariff on auto components from Mexico. (Note that 99% of economists would recommend that Congress not do this, but most politicians are immune to the logic of comparative advantage).

Offshoring IT services -- IBM has been fairly aggressive about moving software support services offshore. In boom times, IBM is hiring many people in India and a few in the US. So, in 2006 and 2007, IBM hires SAP specialists, software testing, and wireless telecomm engineers in its India operation. These IT specialists work with other IBM teams in Australia, China, Japan, Germany, the UK, and the United States to service global customers.

Then, following the financial crisis in late 2008, IBM decides to lay off systems engineers, maintenance engineers, and COBOL programmers in the United States.

Is it in any way accurate to claim that the SAP specialists hired in India in 2006 displaced the systems engineers laid off in Philadelphia in 2009? Of course not. These are people in different functions, hired at different times.

IBM is a global firm servicing global companies. Except at the most aggregate level, the US government has almost no ability to independently observe whether IBM is “exporting jobs.” They could, of course, require that IBM report on what it’s doing. But there is no chance IBM would report in a way that indicts itself. There is simply no way a regulator could accurately observe what the firm’s 300,000 employees are doing, and who they are servicing.

Second, even if Congress wanted IBM to stop hiring people in India, what could it do? Would the Congress threaten to cut IBM off from the Internet? Or from communications satellites? Would Congress be willing to impose fines that are massive enough to cause IBM to withdraw from international markets, where it realizes 68% of its revenues? Unlike trucks crossing the US-Mexico border, regulators have no ability to monitor the bits and bytes zipping around the Internet, knitting various work teams together and allowing the firm to service global clients.

--- ---

The bottom line is that politicians are nearly helpless in the face of offshoring. Firms have a tremendous incentive to locate service activities in low-cost, good quality locations. Doing so cuts costs, increases capabilities, and creates competitive advantage. Politicians would like to regulate this activity, but they can neither accurately observe what firms are doing, nor come up with regulations that can be credibly enforced.

So, expect lots more rhetoric and, perhaps, some efforts to require new reporting by firms. But the offshore trend will continue unabated, despite politician’s hopes. In the short run, offshoring will cause more dislocation and pushback. In the medium and long-run, it will continue to raise productivity and living standards in both the developing and developed countries.

Good discussion unfolding on Policy issues

The guest blog on Horses for Sources is generating some interesting discussion (read from the bottom up) -- almost all positive.

I hope you'll join in.

Policy discussion on Horses for Sources blog

I did a guest post over on Phil Fersht's Horses for Sources blog.

Phil's blog is one of the hubs for outsourcing.offshoring discussion on the Internet. It will be interesting to see what reaction we get.

The post, titled The Politics of Offshoring: All Talk, No Action, should generate a good discussion. I encourage you to check out the blog and participate in the discussion.

Obama Town Hall: Partial Sanity on Offshoring

I'm still traveling in India (more on that in another post), so I'm running a bit behind.

An interesting development in President Obama's on-line townhall meeting. A woman named Harriet, from Georgia, asked the following question (all quotes from the NYT transcript ):

When can we expect that jobs that have been outsourced to other countries to come back and be made available to the unemployed workers here in the United States?

The president gave a surprisingly coherent response -- at least for a liberal democrat. Here it is:

Now, a lot of the outsourcing that was referred to in the question really has to do with the fact that our economy -- if it's dependent on low-wage, low-skill labor, it's very hard to hang on to those jobs because there's always a country out there that pays lower wages than the U.S. And so we've got to go after the high-skill, high-wage jobs of the future. That's why it's so important to train our folks more effectively and that's why it's so important for us to find new industries -- building solar panels or wind turbines or the new biofuel -- that involve these higher-value, higher-skill, higher-paying jobs.

So I guess the answer to the question is, not all of these jobs are going to come back. And it probably wouldn't be good for our economy for a bunch of these jobs to come back because, frankly, there's no way that people could be getting paid a living wage on some of these jobs -- at least in order to be competitive in an international setting.

So what we've got to do is create new jobs that can't be outsourced.

[He then went on to talk about "Green jobs" and having union workers building a smart grid, making $80,000-90,ooo. In the interest of space, I am leaving that out because it's very long and kind of wandering]

--- --- ---

I offer a few thoughts:

1. First, he didn't take the bait and make references to "our jobs" or "American jobs." That's progress.



2. He didn't roll out the "no tax breaks for companies that export jobs" line that nativists seem to like so much. There are no such tax breaks for "exporting jobs," and if the people who use such lines think they can take away other breaks for companies they don't like, they will run into the problem of identification and designing a program that actually works (discussed below Offshoring and Public Policy -- a Conundrum )



3. He seems to be somewhat mistaken about the nature of the work going offshore. From 10 to 5 years ago (as BPO gained steam) much of work was medium skilled -- things like call center jobs, accounting and finance, and claims processing.

What the president misses is that software jobs (in India, a bit more than half the total) are fairly high skill, certainly much more so than the union construction and electrical positions he talks about. Furthermore, the fastest growth today is in high skilled KPO (knowledge sector offshoring) work. Things like engineering services, business analytics, product design, teleradiology, and others.

My overall position is that, while disruptive in the short run, this is ultimately beneficial in the long run.

Bottom line: The president gets a B+. He didn't fall into the typical political trap of mischaracterizing the trade as win-lose (or zero sum). That puts him ahead of 75% of politicians, and 95% of democrats in Congress.

On the other hand, he doesn't understand (or at least communicate) the real nature of what's going on. It will be interesting to see what actions he takes going forward. One of the first tests will be how his administration treats the auto companies. Will they force them to cut back on software offshore outsourcing (GM does huge amounts of work offshore), or on back office stuff (Ford is ramping up its business analytics unit in India). Time will tell.

Offshoring and Public Policy -- a Conundrum

Big news from IBM. Multiple sources are reporting that IBM will lay off about 5,000 positions in the United States, while expanding hiring overseas (Reuters story). Also did an interview with the WSJ, but they story is behind a subscription wall.

This story illustrates several of the challenges facing firms and policymakers as they grapple with the Services Shift.

First, the advantage that developing countries have in services for medium skilled positions is much larger than the advantage they enjoy in manufacturing. This is primarily because services are so labor intensive. If you save 60% on labor, but labor is only 20% of your cost base (as in auto parts) that gives you a 10-12% advantage. If labor is 75% of your cost base, a 60% savings gets you a 45% advantage.

Second, unlike with manufacturing, it is very difficult to match jobs lost in the US to jobs gained elsewhere. If Delphi closes a plant in Michigan and opens on in Mexico, you have about the same number of people doing the same tasks. It's easy to say "these jobs replaced those jobs." With services, however, it's rarely so simple. Say IBM hires lots of SAP and networking specialists in India in 2007-08. These people service many markets, including the US. Now in the midst of the financial crisis, they lay off systems engineers and COBOL programmers in the US. Is it really accurate to say "the SAP people replaced the COBOL programmers."?

The positions are not matched up in either time or function. So is one really replacing the other?

I'm sure the anti-trade, nativist zealots would respond that "there is a US programmer 'willing' to do the job," but that's a far cry from saying that person is qualified. Is it reasonable to ask IBM to spend tens of thousands of dollars retraining an expensive and under qualified US worker? Does this make sense even if the person will be serving global markets (including India)? Does it make sense if doing so will cause IBM to become uncompetitive with rivals who do not follow a policy of giving preference to US workers?

Third, services offshoring is much more disruptive than trade in manufactured goods. Why? Because the advantage that developing countries have is much larger in services (see above). Also, many more people are affected by globalization of services. Manufacturing accounts for approximately 15% of US employment. Services for about 70%. Tens of millions of people who thought they were safe from global competition are finding out that they are not. They want the government to do "something," but no one is exactly clear on what.

Fourth, what can the government do? Actually, very little. As mentioned above, it is very difficult to identify which jobs are displacing which other jobs. Foreign workers account for 71% of IBM's global workforce. Foreign sales account for just over 2/3 of IBM's sales. So, the workforce and sales distribution are roughly in alignment.

But say your congressman wanted to do "something." What would this be? It is unlikely that Congress is willing to cut IBM off from the Internet or satellite communication? Does anyone think congress can micromanage IBM's (or any other firm's) global resource allocation. Beyond making a fuss, there is little that policymakers can do without crippling the very firms that prop up their home economies.

Bottom line, Congress is impotent.

There are really two issues at play here.

1. In the long run, does the globalization of services raise of lower living standards in the rich countries? Almost everyone agrees that it is disruptive and inconvenient in the short run. I believe that the long run benefits outweigh the short run costs (as do most economists, the McKinsey Global Institute, and many others). But intelligent people can disagree on this.

2. What policies can the government put in place to slow of stop services offshoring? Even if you believe offshoring harms the United States, the onus is on the opponents to suggest a policy that would be effective. I can't think of one. Sputtering and whining won't cut it. Most policies (such as the Grassley-Sanders fiasco on H-1B visas) cause more harm than good.
--- --- ---

So, it's a tough problem. Everyone wants to focus on question #1. But without addressing question #2, it's just whining. Still waiting for an answer on that.