How much R&D is being offshored? NSF releases new numbers

My view is that we suffer from an “information deficit” rather than an “information overload.”  We know far too little about the important things, and far too much about the unimportant.

One of the notable gaps in our knowledge has been reliable statistics about R&D spending.  The economy  is driven by innovation, but we knew very little about who was doing the innovating and where.  In particular, we had absolutely no idea how much R&D was being offshored by U.S.-based companies.

In order to answer this question and many others,  the National Science Foundation got funding a few years ago for a new survey on business R&D and innovation. I wrote about it in a September 2008 BusinessWeek cover story, Can America Invent Its Way Back?: “Innovation economics” shows how smart ideas can turn into jobs and growth—and keep the U.S. competitive.  (This story is also the answer to a trivia question: What BusinessWeek cover was being delivered to the homes of subscribers when Lehman failed?).

Now the NSF has released the initial results from the new Business R&D and Innovation Survey.  For the first time, we have a good read on how much U.S. businesses are offshoring their R&D—setting up research facilities in other countries. 

It turns out that in 2008, manufacturers did about 20% of their R&D overseas. That’s actually a bit less than I would have expected. Here’s how it breaks down by some industries.

Surprisingly, the industry which has done the most to offshore its R&D is the auto industry,  with 39% of its spending done overseas.  Other industries with a high rate of R&D offshoring  include electrical equipment—that’s lighting, generators, and the like—and info tech hardware.  Pharma had less offshoring than I would have expected.

The survey was incredibly detailed, and there’s a lot more data releases coming in the future that will enable us to figure out the effectiveness of this spending for innovation, and the employment generated.  It’s great stuff…take a look.

[Note: The survey counts R&D done by U.S.-owned businesses, both home and abroad and R&D done by U.S. affiliates of foreign parents I’m pretty sure that in future data releases, we’ll get numbers that allow us to separate out the two. ]


  1. Brandon W says:

    On Deficit vs Overload, it’s like “listening” and “hearing”. The best way to phrase it may be: We have an information overload but a knowledge deficit.

    Or, it may be that we have as much relative knowledge as before, but it’s lost in the deluge of information. By way of comparison, 25 years ago we had 4 free TV channels and there were about 5 good programs. Now we have 300 channels but there are still only 5 good programs, and I have to pay $65/mo to watch them.

  2. Brandon, the fallacy of 5 good programs is easily punctured by pointing out that you could just buy those 5 programs on DVD or iTunes for much less than the $780/year you’re paying for cable. I guarantee you won’t though, because I bet you like plenty of other shows on cable too. 😉 Also, even if you like only 5 programs today, those programs are likely to be better and more specific than shows in the past. This is because when there were only 30 shows, people were more likely to choose the same 5, while with 500 shows, each person might still only like 5 shows, but there will be a lot more variation in the 5 they pick. A good example is a show like Mad Men, which only thrives on a cable channel like AMC and could never air on a broadcast network even today, yet I’m sure it makes a lot of people’s top 5. I never listened to the radio in the past because I found it all mainstream pap, yet I regularly listen to 3-4 podcasts today, which deliver the kind of extremely high-quality niche content that could never make it on a broadcast medium like radio. This is true for all content, whether video, audio, or text, so things are much better today, but don’t let that get in the way of your reactionary moaning about how “things were better.” 😉

  3. Brandon W says:

    Actually, I watch fewer than 5 programs now, if you want to get technical about it. I only have cable because my wife wants to watch TV (and she mostly just watches Seinfeld reruns!). About the only thing I ever watch are baseball games (usually the Red Sox, when they run the game in Michigan). Previous to getting cable, I went over a year without a television, and I didn’t miss it. I made my wife watch TV on Hulu and I listened to Red Sox games streamed over the Net from Boston radio. You’ll never convince me that what’s on TV now is better than 25 years ago, or that there is more *good* programming. More does not equal better.

    Having said that, you were so focused on attacking my example – which I intended with some humor – that you completely missed my actual point. Which is something you do often (including going back to the BW forum). I can’t decide if you miss it because you’re not paying attention, because you’re unable to comprehend the analogies I often use, or if you’re actively evading my point with straw-man debate tactics. Whatever reason, it’s annoying.

  4. Well, if you just want that stuff, you could buy the Seinfeld DVDs and get an MLB season pass on your computer, saving that cable TV money you were complaining about. I haven’t had a TV in 5 years, I just steal stuff online through torrents for the 2 hours or so of video I watch in an average week. Not only is the stuff today better than 25 years ago but more is better. With a lot of choice comes the ability to choose the 5 shows that are best for you: the top 5 is going to vary widely for different demographics and choice lets everyone have their own programming. However, it’s always possible that your tastes are deeply conventional and that all this choice isn’t helping you specifically. 😉 Whatever your original point, my point is that your example is flawed, possibly just like your original point. 😉 Nice try guessing the reason with your self-serving possibilities, but it’s most likely because you’re not making much of an argument. 🙂

  5. Brandon W says:

    Since you’re still fixated on the TV thing, it demonstrates that you’ve missed the actual point I was making (which, for the record was meant to support and help clarify one of Michael’s ideas). That is all.

  6. CompEng says:

    Thanks for the info, Michael. Very interesting… I’ll have to think about this a bit.

  7. Brandon Shackelford says:

    Hi Mike – I think the domestic/foreign R&D split is even more interesting when compared next to the same split for sales. The auto industry may perform 39% of its R&D overseas, but at the same time 38% of its sales are from foreign operations. Of course, the auto industry still sticks out compared to the overall statistics which indicate that the share of sales from foreign operations (32%) is relatively larger than the share of R&D performed at foreign operations (20%).

    We’ll see if you change your mind about information overload when all gazillion data points from the survey (and the requisite footnotes) are published!

  8. Yes. As a US citizen engineer I have been struggling for years. Used to have tons of ideas and studied methods for doing cutting edge things. It is just too hard to make money in engineering anymore. Go ahead. Send it overseas. I won’t teach them the basics that they need and can never teach them how to become innovative…. that comes from living in a truly free society.


  1. […] Looking for Reliable Data Mandel on offshore R&D spending. […]

  2. […] innovation and R&D? New statistics from the National Science Foundation (via Mike Mandel) certainly point in that direction.Way back in 1990, I wrote a book titled The Breakthrough […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: