Entrepreneurs of the World, Unite in Eco-systems!

By: Jørn Bang Andersen

Clientelism, Wintelism and innovation eco-systems: moments of truth within the global semiconductor industry. Europe and Japan tried for years to challenge Intel by building national seminconductor champions, yet wihout success. Today, a small British company named ARM has qua its eco-system strategy managed to launch a more serious go at Intel.

The story begins with that of failed European champions, Intel’s global monopoly in microchips or semiconductors for PCs and servers and ARM, an aspiring game changer in the global semiconductor industry. The three questions raised in this article are:

First, will the global semiconductor industry be the next industry to see radical changes as a consequence of the ‘modus vivendi’ of the global mobile phone industry and the organizing principle around open innovation eco-systems?

Second,  what can both smaller and large companies learn in terms of strategy and threats to their business?

Third,  what  policy lessons to take from the case and the history of the European semiconductor industry?

Europe and the global semiconductor industry

When the American company Intel Corp. was founded in California in 1968 the seeds were sown for one of history’s great industrial transformations. Intel started out producing semiconductors and the world’s first microprocessor in 1971.  A few years later in 1975 Microsoft and in 1976 Apple was founded and the USA had already the global champion IBM within the new electronics industry.  For Europe the 1970 was gennerally a lost decade which since then have been labelled ‘Euroschlerosis’ due to Europe’s decreasing economic comeptitiveness, oil crisis, high inflation, high unemployment and inability to foster companies within the new industries of information technology and biotech.

At the beginning of the 1980s it became clear that the semiconductors and microprocessors were going to be the ‘crude oil’ of the bugeoining information technology revolution.

At the beginning of the 1980s it became clear that the semiconductors and microprocessors were going to be the ‘crude oil’ of the bugeoining information technology revolution. Moreover, it was clear that Europe had not only lost out to the US companies but also had to realize that especially Japan and to some extent the Asian tigers of South Korea, Taiwan and Singapore had surpassed Europe within the semiconductor industry and information technology in general.

For European policy makers this development was not only alarming in commercial terms but also in respect to security and defence issues. This was so because semiconductors and information technology came to define the boundaries for new generation of weapon systems such as guided missiles etc.

Clientelism vs Wintelism

As a response the largest European countries (notably Germany and France) and leading European national champions, Siemens, Bull and Philips, at the time the European companies with the  best chances within the electronics industry, were incentivized to form a global European champion within the semiconductor industry. At European level the idea of creating a European champion was backed by the launch of the technology support programmes, primarily Esprit.  The first Esprit programme was launched in 1983. Yet, the European attempts of creating a European champion within the semiconductor industry was from the beginning weakneded by notably the French government’s decision to nationalize Bull in 1982 and merge the company with most of the rest of the French computer industry. In similar ways Germany tried to make Siemens a German champion and the Netherlands with Philips.

Whilst Europe really wasn’t getting anywhere within the global technology and market race within the semiconductor industry the United States government decided in 1987 to form the partnership SEMATECH. This was a partnership between the US government and 14 US based semiconductor manufacturers to pool know-how and regain the global market lead which had been lost to Japanese manufacturers. The US government invested $500 million over five years in SEMATECH.

The European attempt to build one or more semiconductor champions is most telling with the example of Bull.  Between 1983 and 1990 the French government fed more than $1 billion in capital into the company. Yet, Bull never proved able to compete within the electronics or semiconductor markets and in 1991 the company had additional 10 billion FFr in a rescue package. Neither Siemens nor Philips managed to establish anything remotely close to a global semiconductor market on par with Intel. And in the 1990s it was clear that Europe’s policy of ‘clientelism’ had lost the technology race within semiconductors. Moreover, Europe had failed to developed players within the rapid developing PC market and during the 1990s Microsoft and Intel controlled the global market for PC software and microprocessors with more than 90% of all sales and the position became known as Wintelism.  From Asia Japan recede its leading position to newcomers such as Samsung and LG from South Korea and Taiwanese manufacturers.

Innovation eco-systems – black holes in the universe

Until very recently most people had almost forgotten about the European semiconductor industry or considered it an irrelevant issue.  The Economists briefing on the semicondutor industry, 7^th January 2012 edition, brings however the issue back to the fore and makes it  highly relevant again although for new reasons. In short, the Economist informs that Intel has no rivals in the global market for PCs and servers. In 2011 Intel had a revenue of $40 billion and $13 billion in pre-tax profit. The British company ARM is ant sized with a revenue of $588 and pre-tax profit of $172 in 2011.  ARM has the global edge within the market for mobile phones and tablets.  Intel is a traditional hierarchical organization and controls both its design and manufacturing of chips and supply the market from its 10 global semiconductor fabs. According to Intel’s management this control gives Intel a competitive advantage over its rivals who use external firms to make their chips. This is so because Intel can bring its chips faster to market and with fewer faults than its rivals.

ARM’s business model means that it is kind or organized as a federation with several hundred chip makers, designers of chipmaking tools etc. , whereas Intel is organized more like a feudal kingdom involving very few outside players.

ARM’s chips are on the other hand designed to compete on low energy use rather than processing power and ARM’s has no production of its own. ARM’s business model is based on selling licensing of its semiconductor designs.  In essence what provides is a platform on which others can build and the costs are shared. ARM’s business model means that it is kind or organized as a federation with several hundred chip makers, designers of chipmaking, tools, etc. , whereas Intel is organized more like a feudal kingdom involving very few outside players.  Hitherto the two companies have not been in direct competition due to different focus markets PC and servers vs mobile chips. However, with a shrinking market for PCs and growing market for servers and mobiles Intel and ARM are inevitable to entering into each other’s market domains.  In this regard ARM has the upper hand because Intel’s organization and business model will have difficulties to match the relationships and rivalry characterizing in the mobile phone industry.

As  Mr Brown ARM’s president puts it ‘the complex reciprocal relationships that make up ARM’s eco-system, are probably our biggest barrier to entry’. Moreover, ARM’s low energy and greener chips becomes increasingly interesting to the server farms needed for running mobile traffic, the cloud and other data because of the steadily increasing costs of electricity.

Open innovation ecosystems – a new chance for Europe?

Although it is too early to tell what will be the outcome of the small European company ARM taking on the global leader Intel in the global competition, the holds interesting perspectives.  Around 15-20 years ago Europe’s failed to establish national or European champions within the semiconductor industry. The strategy back then was overwhelming technology driven and costed billions of dollars.  The fact that a smaller British seminductor manufacturer like ARM that never received state funding or alike and already seems closer to the European goal than ever before is primarily due to its business model around a decentralized global open eco-system. The  success stories from Google’s android, Apple’s iphone, Nokia’s fall and ARM challenging Intel all point in one direction, namely that busines models organized around a more or less open innovation ecosystem are becoming the winning recipe within so far the global information and communication technology industries.

In view of the last 20-30 years of evolution within ICT it seems it seems that there could be a new avenue open for European companies and support policies for having a second go at becoming serious players in the global ICT market. That go should for starters focus on better understanding both the advantages and pitfalls of systematically developing mangement and organizational know-how around open innovation eco-systems.  Innovation policies could support European companies by analyzing  the pros and cons for open innovation eco-systems within particular industries, and make the information freely available to companies and branch organizations. It might be that some technologies or markets are more conducive and likely to succeed with this type of organization than others. For example, it is hard to see how Airbus 380 could have been build and coordinated in an open innovation system like that of ARM’s.

Eco-systems seem more like ‘black holes in the universe’. Once formed they can continue to grow by absorbing additional matter.

Finally, it seems that traditional industrial, innovation and cluster based policies have had a tendency to drift towards a zero-sum game primarily because of their delimited geographical focus, whereas open innovation eco-systems seem more like ‘black holes in the universe’. Once formed they can continue to grow by absorbing additional matter.

Lessons learned

The case told here holds some lessons for both company managers as well as policy makers.

• Most companies still focus their innovation on product or process.  But more examples from different industries (mobile telephone, semiconductors etc.) show that innovation based on partnerships or eco-systems can be a very potent strategy and represents a possbile launch-pad for how start-ups or smaller companies can attack even the global leaders within an industry.

• Once the innovation ecosystem reaches a certain critical mass it becomes in itself a barrier to entry and a strategic asset for those involved.

• Larger companies should be on guard regarding the potential implications of being attacked by network or ecosystem based competitors – especially within business areas that might seem periphal. This is so because, it is often in areas perceived periphal  that the large company is most vulnerable to new entrants.

• Policy makers should at all times avoid trying to pick ‘winners’.

• Innovation policy should try to nuture the evolvement of open innovation eco-systems and act as catalyst for their development, as they seem to offer a strong platform for entrepreneurs and for the general climate of competition.

By Jørn Bang Andersen