Alan Steel said, "It's probably not a bad idea to nibble away at one or two technology funds if you haven't already done so."
Headline writers have become fed up writing about Bird Flu and SARS, oil shocks are old hat and it's too early for the hurricane season so their latest big fear for investors is that old misery - inflation. You know the script. Inflation fears means the Bank has to push up interest rates. In the US the Fed keeps rates on hold because of inflation fears. Just like they did back in 1994 for those with good memories.
You won't be surprised to learn some experts suggest Baby Boom Demographics were a major cause of inflation through the '70s and the '80s, and probably the main reason why inflation is benign right now. There's also downward pressures on prices from increases in global competition. But few talk about advances in technology as being a major reason why inflation is under control, and likely to continue being so for some time to come.
Those who know me well will find it surprising I'm attempting a piece about technology. Being naturally technologically dyslexic, I've long had a problem working out how electricity works. But in our research I came across a fascinating piece by Dave Turek who's in charge of supercomputer development for IBM.
Turek reminds us it's only ten years since a supercomputer managed to beat a world chess champion for the first time. Yes, ten years ago on 11 May a supercomputer called Deep Blue became the first machine to defeat a reigning world champion chess master under tournament rules. The eight day, six game match with Gary Kasparov drew the world's attention. But things have changed so much in the last ten years that today Microsoft's X Box 360 game console is actually more powerful than Deep Blue.
Funnily enough, Deep Blue wasn't the fastest computer in the world in 1997 (it was 388th) but it had 32 processors that could handle 11 billion calculations a second. Now, 4 processor generations later, Blue Gene, today's fastest computer, uses 131,000 processors to handle 280 trillion operations every second!
The biggest advance however isn't the technology in itself. A decade after Deep Blue, the business world is embracing commercial supercomputing. And that means the real revolution is still to come.
Supercomputing has moved rapidly away from its traditional place in academic and Government research to tackle commercial applications. Of the top 500 supercomputers in 1997, less than a third were employed in business. Today, more than half sit in commercial enterprises.
It seems two essential economic forces are propelling the movement. The first is the ever increasing expense of building and testing physical prototype designs for everything from nappies to cars. The second is the rapidly decreasing cost of modelling and testing a virtual product on a supercomputer.
Ten years ago the price for the capability of processing one million operations per second was roughly £25. At that rate, Deep Blue's 11 billion calculations cost about £275,000 a second to win a chess game. Today, one million operations per second cost less than 5p. It seems as a general rule supercomputer power increases, and prices fall, about 40% each year. The fastest supercomputer in the world in 1997, a multi million pound capital expenditure, is today's £100,000 starter kit.
Even Proctor & Gamble has made use of supercomputers to make nappies more eco-friendly and absorbent and to tweak the design of Pringles so they can fly more efficiently through a production line. Use of supercomputers in the old days would be found in the province of aerospace engineers.
The energy industry is increasing use of supercomputers in terms of modelling of oil and gas reserves and given the estimated cost of drilling a single well in deep water approaches £17 - £18 million, we can expect this progress to continue. And as prices continue to fall and user ability improves, even smaller companies will be able to tap the increased efficiency made possible by supercomputers.
The last silicon revolution saw personal computers invade the office, and automation take over factory floors. From the late '70s to 2004, in the US manufacturing output doubled in real terms. While productivity increased 53%, manufacturing productivity rose over 100%. It was thanks to computers at the time. Now it seems we're about to see a far more potent business revolution powered by supercomputers. In this new silicon development it really is chips with everything.
There's more good news. We'll continue to see low inflation, increased globalisation, increased productivity, increased profits, and very likely, increased returns from world stockmarkets. Indeed, one of the most hated investment areas of recent times, quite understandably, is the technology sector. So it's probably not a bad idea to nibble away at one or two technology funds if you haven't already done so.
This letter is the personal view of Alan Steel. Please check the appropriateness to your individual position with your adviser before taking or refraining from any action.