Intel – The hard road pt. II

This is Intel’s make or break moment.

• Intel’s determination to return to manufacturing supremacy is existential because if Geisinger fails, the vultures will break-up the company and sell off the pieces sending another American titan to the graveyard to rub shoulders with Motorola.
• The market’s reaction is indicative of just how short-term obsessed it is as there is absolutely no consideration of how valuable this company could become should it return to its former glory.
• Pat Gelsinger, the new CEO, pretty much committed Intel to this direction on the results call and it will not be without considerable difficulties and risks.
• However, I think that this is the right decision as:
  • First, Nothing else: Without cutting edge manufacturing delivering best-in-class performance, Intel does not amount to very much.
  • This is being ably demonstrated by the resurgence of AMD which has been outperforming Intel on almost every level for the last couple of years.
  • Consequently, for Intel to survive in the long-term, I think it has to return to its former glory.
  • The alternative is for the company to outsource manufacturing leading to a slow decline into a break-up and oblivion.
  • The risk is that failing to affect a turnaround will result in a break-up that yields a lot less for investors than simply going directly in this direction right now.
  • Second, Technology rivalry: If Intel cedes manufacturing to TSMC and Samsung, then the USA will have created a critical technology dependency on Taiwan and Korea.
  • RFM and Alavan Independent have identified Taiwan as a potential flash-point as China has committed itself to re-unification while the population and political parties in Taiwan have become far more pro-independence.
  • Hence, the only realistic possibility for China to reunify with Taiwan at the moment is through the use of military force.
  • There are already signs of this heating up with China making 2 incursions with military aircraft into Taiwanese airspace in recent days.
• • This would massively disrupt TSMC’s operations and lead to a large supply shortage of cutting edge semiconductors.
  • Korea is also not without its own risks of this nature with its erratic and economically challenged neighbour to the north.
  • Hence, not only does Intel need to return to cutting edge manufacturing for security reasons but it also creates an opportunity for Intel to offer foundry services for TSMC and Samsung customers concerned about rising geopolitical risks.
  • GlobalFoundries also has this opportunity and is once again considering cutting edge manufacturing.
  • Third, competition: Without cutting edge manufacturing, I think that Intel will struggle to compete against the fabless semiconductor makers like AMD and Qualcomm.
  • It will find itself way behind as a result of the transition, resulting in an erosion of gross margins, profits and cash flow.
  • This is a deficit Intel is unlikely to make-up as it will have given up the one thing that the company really knows how to do.
  • Manufacturing is Intel’s DNA and without it, I am not convinced that the company amounts to very much.
  • Intel has had problems in the past and recovered from them, but the hurdles that Intel must overcome are by far the greatest it has ever faced:
  • First, Arm: Until recently, the ability of Am-based processors to challenge x86 in personal computers and servers has been a dream rather than a reality.
  • Arm-powered PCs have been disappointing while the legacy software problem has kept Arm largely out of the server market.
  • However, the M series of processors from Apple have demonstrated what is possible with the right design skills.
  • This combined with Qualcomm’s acquisition of NUVIA and the collective sigh of relief heaved by 22 of Qualcomm’s partners (see here) is a sign of what is to come.
  • This raises the question as to whether the x86 architecture is now obsolete which Pat Gelsinger and Intel must now emphatically answer.
  • Second: TSMC & Samsung: Both of these two are significantly ahead of Intel when it comes to cutting edge manufacturing.
  • Furthermore, there is a good possibility that the superiority of the silicon that they are currently producing is due to their relatively new 5nm processes.
  • By the time that Intel fixes its processes, these companies will be implementing 3nm and beyond meaning that not only does Intel have to fix its processes, but also be faster in developing new ones so it can catch up and overtake the Asians.
  • This will be no small feat and will test Intel’s management, morale and culture to the very limit.
  • Third, AMD: has been beating Intel at its own game pretty much ever since Intel fell off the manufacturing wagon.
  • Intel has to catch and surpass AMD if it wants to see its gross margins return to the very high levels that it has historically enjoyed.
  • Again, this will not be easy as Lisa Su and her team are a force to be reckoned with, having affected, one of the great turnarounds in the industry of recent times.
  • Intel has not sunk to anything like AMD’s level yet but the challenges that it faces from the industry are greatest that they have ever been.
  • I think that Pat Gelsinger needs to surpass Lisa Su’s achievements.
• The good news is that liquidity is not a problem and Intel will have access to all of the resources that it will need to execute its strategy.
• Its recent approach has been to bet on all technologies in order to avoid a repeat of its failure in mobile phone chips, but this widening may have contributed to losing its way on its core competence.
• Hence, I can see Intel re-focusing on its core competences and making a series of very focused bets on what it thinks is going to be important over the next 10 to 20 years.
• The way ahead is going to be difficult and there is the possibility that the recent rally driven by a short-term outsourcing theme completely unwinds.
• This will create a very large opportunity for those that believe that Intel will be successful in its turnaround.
• Big question is will I chicken out again should I be presented with another chance to buy Intel at $44?