What Apple’s Supply Chain Says About US Manufacturing and Middle-Skill Training

In January, The New York Times released a front-page report on the iEconomy, Apple’s vast and rapidly growing empire built on the production of tech devices almost exclusively overseas. The fascinating story created a wave of attention when it was published, and it’s back in the news after NPR’s “This American Life” retracted its story about working conditions at Foxconn, one of Apple’s key suppliers of iPhones and iPads.

The end of the “This American Life” episode includes a discussion (audio | transcript) between host Ira Glass and Charles Duhigg, the NYT reporter who wrote the iEconomy piece, on Apple’s supply chain and the reason the tech giant doesn’t produce its insanely popular devices in the U.S. Perhaps you thought the main reason was labor costs; Apple would have to pay American workers much more than the estimated $17 a day (or less) many Chinese workers at Foxconn make. That’s part of it, but “an enormously small part,” Duhigg told Glass.

Duhigg explained that, in terms of labor costs, producing the iPhone domestically would cost Apple an additional $10 (on the low end) to $65 (on the high end) more per phone. “Since Apple’s profits are often hundreds of dollars per phone, building domestically, in theory, would still give the company a healthy reward,” he wrote in the NYT piece.

Instead, what matters is Apple’s intricate Chinese supply chain. Duhigg went into detail in his conversation with Glass:

Compared to the cost of buying chips or making sure that you have a plant that can turn out thousands of these things a day or being able to get strengthened glass cut exactly right within, you know, two days of this thing being due, that’s what’s important. Labor is almost insignificant. What is really important are supply chains and flexibility of factories. You want to be able to be located right next to the plant that makes the screws so that when you need a small change to that screw factory, you can go next door and say, “Give it to me in six hours,” and they can say, “Here you go.” Because if that factory was in another state or on another continent, it would take two weeks. It’s the flexibility within the Chinese manufacturing system, that’s what you can do in Asia that you can’t do in the United States.

Another major reason why Apple produces the iPhone and iPad abroad is the huge (and available) skilled workforce in China. When Apple needed 8,700 industrial engineers to oversee 200,000 assembly-line workers involved in making the iPhone, the firm looked at how long it would take to find that many workers in the U.S. Its answer: as long as nine months — compared to 15 days in China.

This boils down, according to Duhigg’s report, to the lack of American workers with “mid-level skills that factories need, executives say.”

Companies like Apple “say the challenge in setting up U.S. plants is finding a technical work force,” said Martin Schmidt, associate provost at the Massachusetts Institute of Technology. In particular, companies say they need engineers with more than high school, but not necessarily a bachelor’s degree. Americans at that skill level are hard to find, executives contend. “They’re good jobs, but the country doesn’t have enough to feed the demand,” Mr. Schmidt said.

From Whose Bourn No Manufacturing Returns

So, what does this mean for manufacturing in the U.S.?  For the most part there are no surprises here. These are known, core issues in the U.S. We’re well aware of the skilled worker drought. We’re aware that we’ve shipped much of our production process off-shore. But, we keep saying to ourselves, we’ve increased our productivity. We’re innovating and we’re keeping the profits, right? To answer that we can say “maybe” and “sorta.”

There’s currently a debate raging about how we calculate our productivity. It’s a complex issue that we won’t get deep into here (check out this recent article in the Washington Post for more), but the gist is that our methodology for calculating productivity might be overstated. Like, a lot. Biases within this methodology have tended to convert a factory’s price savings through outsourcing into an increase in output. As Peter Whoriskey of the Post puts it:

For example, suppose a U.S. factory decides to offshore the production of a part for which it used to pay $1. With the switch to an overseas supplier, it might pay 50 cents for the part. If U.S. statistics do not capture this drop in price, the savings by the U.S. factory can show up as a gain in output and productivity.

Changes like these aren’t an increase in output due to innovation and greater productivity per worker. These are savings, which are certainly great for that factory, and certainly mean better profit margins, but which aren’t productivity increases that we get to point to as the silver-lining of our inky manufacturing cloud.

That’s one element. The other element sits very near and dear to our heart here at EMSI. There’s a very basic theory at work in our impact studies and I/O model, and it’s that tight supply chains are good. The goal of any regional economy should be to develop complete supply chains. The less that an industry has to bring goods and services into the region from outside, the greater effects we see from increases in those industries. In other words, “Great job, China.” But this leaves the US at a disadvantage. Yes, low-cost Chinese manufacturing has facilitated growth in the US economy. However, this has happened at the expense of the US developing these complete supply chains.

Maybe If We Ask Real Nice

Now, this is an extremely complicated issue, and things have developed as they have for a reason. However, when we ask the question, “what would it take to manufacture iPhones in the US?” — a question tantamount to “what would it take to get all of those manufacturing jobs over here?” — if the supply chain is the deciding factor, then the answer might be something along the lines of “too late.” At this point, manufacturing supply chains are so thoroughly concentrated in China that bringing them back to US soil seems more or less impossible. There may be little we can do about manufacturing jobs leaving the US.

The Data

However, instead of simply offering some sound — albeit pessimistic — reasoning on the issue, let’s look at some data on manufacturing. Since jobs are in the spotlight, and productivity measures have received some heat, we’ll simplify the issue a little and just look at job growth in manufacturing.

In terms of industry performance, 74 manufacturing industries saw growth from 2001-2006; 65 saw growth from 2006-2009; and 210 saw growth from 2009-2011. This is, of course, growth of any amount. This indicates a slight, recent uptick in the variety of industries experiencing some amount of growth.

Typically, we look at net growth for industries. We combine growth and decline to show the overall growth and decline. Here’s net decline for manufacturing for three different periods:

Job loss seems to slow in that 2009-2011 period.

We’ll look at one more series of data, before we wrap up. What if we look at decline and growth separately instead of simply the net of both? This will allow us a discrete view of whether decline has slowed or growth has quickened, or perhaps a combination of both.

First, decline:

Manufacturing decline considered on its own appears to have slowed in the past few years.

Now, growth:

Gain in the manufacturing sector seems to have picked up pretty noticeably.

This analysis is not nuanced. As other sources have made plain, this is a complex issue. However, it remains true here that job growth in manufacturing has picked up somewhat, and that decline has slowed. We’re still seeing net losses, so this isn’t evidence of recovery. However, it does mean that not absolutely every manufacturing job is moving offshore. At least not yet.

Not to mention that there’s some support for the idea that we’re seeing some “reshoring” — manufacturing jobs coming back to the US. A recent report from The Boston Consulting Group claims that by the end of the decade we could see a reshoring of 600,000 to 1 million jobs. The report explores the seven broad industry sectors “most likely to reach a ‘tipping point’ over the next five years—a point at which China’s shrinking cost advantage should prompt companies to rethink where they produce certain goods meant for sale in North America.” Are we seeing the beginnings of this in our more recent industry data? Or do we stand to lose more manufacturing jobs to China? We’ll have to wait and see.

Stevenson has worked at EMSI, an Idaho-based economics firm, since 2006. He's a regular contributor to the EMSI blog. Contact him at josh@economicmodeling.com.