23 March 2012

Japan, Innovation and Electronics

Writing in the WSJ, Richard Katz has an insightful commentary on the decline of the Japanese electronics industry.  Here is an except:
The plunge in the past decade is nothing short of breathtaking. From 2000-2010, Japan's electronics production plummeted 41%, exports tumbled 27%, and the sectoral trade surplus dived 68%. Counting only exports by high-income OECD countries (to avoid the impact of China), Japan's global market share of electronics goods and services exports fell by nearly half, to 10% in 2009, from 19% in 1996. In the same period, Germany's share rose to 11% from 8% and Korea's rose to 9.2% from 6%.

The immediate cause of the problem is bad product strategy. Japanese firms and the government failed to heed two big lessons taught by Harvard Prof. Michael Porter. First, as countries mature, their sources of competitive advantage change. At one point, abundant skilled labor, cheap capital and price are keys to competitiveness. Later on, innovation in products and processes becomes pivotal. Secondly, strategy is not just about what products to offer, it's also about what products not to offer.

Rejecting these lessons, Japanese firms tried to compete with newcomers like Samsung on cheap capital and manufacturing prowess instead of product innovation. They kept producing formerly world-beating products that now lose money year after year. Forty percent of Japan's electronics output still consists of consumer audio-video products and semiconductors.
Why does this matter? Katz explains:
Seventy-seven percent of Japan's entire electronics output now consists of parts and components that often go into other firms' products. Yet a cost breakdown of Apple's iPod or iPad or Samsung's Android smartphone shows that the real money does not go to the parts producers but to the product inventors. Japanese firms are competing against Samsung when they should be competing against Apple, Intel and Microsoft.
Knowing when to hold'em and when to fold'em is essential to effective innovation, for businesses and policy alike.


  1. I saw it first hand working for Intel then for a Japanese electronics company '80 to '04.

    "At one point, abundant skilled labor, cheap capital and price are keys to competitiveness."

    These advantages allowed Japan to take over the memory market from Intel et al in the 80's. Unfortunately these advantages moved to Korea in the 90's [there we stories of Japanese engineers moonlighting in Korea]. China now has the lowest cost labor and capital.

    Japanese companies tried very hard to innovate. They were not blind to what was happening. However, successful innovation is not easy (see Apple vs Intel).

    By the way, Prof Porter's class was my favorite in grad school.

  2. Sony should be bouncing heads against Apple, but it’s not. The one-time champion of take-your-music-with-your-Walkman lost focus, lost its way. Apple’s success in mobile-music was not just in selling non-volatile storage and simple controls, but in creating a storefront where folks could easily and legally download their favorites at a reasonable price.

    Here’s the irony: Sony owns the rights to more entertainment -- audio and video -- than most, yet has not been able to capitalize on it in the way that Apple has. Apple developed more than just a smart-looking delivery device, it offered a delivery ecosystem that guaranteed it a revenue stream separate from the hardware device that generated a one-time sale.

    I suspect, however, that Apple is at least mildly concerned about the success of its new super-duper resolution iPad because 3G and 4G wireless customers are finding that the price of hi-rez is not cheap. Acolytes of George Gilder are right that spectrum should be cheap, but the FCC’s approach to date has made bandwidth expensive, enabling AT&T, Verizon, and others to find out what the market will bear.

    Stay tuned, somebody is bound to get smarter.

  3. Cannot see any mention of Japan's postion in the electronic games industry.

  4. Roger,
    I'm almost entirely ignorant on the subject, but if you don't have the new products and processes, what else can you do but keep going with what you do have.

    The question might better be put, "where is the innovation in Japan?" If it is there but not funded, or enabled in other ways, what are the constraints?

    If it is not there, why not?

    I mean by innovation the invention and bringing to market products and processes which are widely desired and heretofore either unavailable, or were far too costly to provoke wide demand.

  5. I never thought of Japan as very innovative. They seemed very good at getting it right- quality, etc. Apple is more about marketing genius (Jobs).Japan has a fairly rigid labor market. That makes it hard to adjust quickly. Corporate structure is pretty bureaucratic. That makes it hard to challenge the status quo.

  6. Roger, I read that article as well and felt it meshed with what I've seen in the industry. Here's my reading of the semiconductor industry (which I work in) in Austin (where I live).

    The top semiconductor/electronics employers in Austin are IBM, Freescale, AMD, Intel, and Samsung.

    AMD spun off the chip manufacturing five or so years ago into Global Foundries. GF has facilities in Germany, and in Singapore (formerly Chartered). In this sense, Global Foundries is specializing on the manufacturing portion, which on the leading edge is very difficult. Global Foundries now makes chips for a variety of fabless companies, not just AMD.

    IBM does not have any fabs in Austin. In fact they don't manufacture much these days. They are more of an IP company. They are part of a high-end chip manufacturing consortium up in New York, that includes GF and two Japanese companies (Renesas and Toshiba).

    Samsung has a memory plant here in Austin. This is to my knowledge the only leading edge chip manufacturing plant in Austin. They seem to be taking advantage in the market of people with engineering degrees. Most of the IP and R&D is carried out in South Korea. This R&D is assisted with software tools built in the US, by companies such as Synopsis, Cadence, and Mentor Graphics.

    Freescale was spun off from Motorola 6 or 7 years ago. They make a lot of different kinds of chips, including chips which have redundant capability for automotive brakes. The manufacturing facilities they still have in Austin are 10 years old or so. They're newer generation chips are manufactured in Asia, by companies such as TSMC, in Taiwan. In other words, Freescale is going fabless.

    Finally Intel is an island unto itself. They do their own design and their own manufacturing. I think most of their manufacturing is in Arizona in Portland.

    This is all to say that companies here seem to be shuffling their capital (human, intellectual, etc.) into the places where it is more efficient.