Serious Money: A long time ago, I was peripherally involved in a deal that involved a purchase of an American investment bank. Those of us on the equity side of the business were invited to New York to meet up with our new US colleagues. It was, as these things often are, a bit of a boondoggle, essentially amounting to dinner and a few beers with the new guys, writes Chris Johns

I vividly recall the reaction of the European equity analysts to what, for many of them, was their first direct exposure to Wall Street types. Several of them remarked along the following lines: "I always thought the stories about these people were myths, but they are true. They may all have Harvard MBAs, but US equity analysts have a three-month time horizon at most, and are exclusively focused on the next quarterly earnings report."

While transatlantic rivalry may have played a part in this harsh assessment, it wasn't wholly inaccurate. Investors and analysts alike all seemed, at that time, to be interested in one thing only - whether or not a company would meet, miss or beat earnings forecasts for the next three months. In some ways, this was a product of the mania that led to the tech bubble and many of us had hoped that short-termism on this scale had dissipated as Eliot Spitzer waged his war on some of Wall Street's dodgier research practices.

But I was reminded of all of this when Microsoft's share price fell by 11 per cent last week, on the day that it surprised the markets with an announcement that it was going to be spending more money than expected on things like product development.

Just prior to the Microsoft surprise, the shares had been trading roughly 8 per cent higher than a year ago - a respectable, if unexciting, performance for the software giant.

The shares have been tipped by Serious Money: the bigger picture is that they haven't gone anywhere since the second half of 2000, a fact not altered by the recent large fall - the last time anything so dramatic happened was in November 2000.

I have thought for some time that, while Microsoft will take a long time to return to its 1999 peak of nearly $60 (€47) - the price is just over $24 at the time of writing - the market was wrong to think that Gates and Ballmer will not be able to mount an effective response to the threats posed by Google, in particular, and other competitors in general.

But the market has been sceptical: the shares have been steadily derated to the point where, according to Bloomberg data, the price earnings ratio is now close to its all-time low, last seen in 1989. At around 18 times earnings that may not, nevertheless, stand out as obviously cheap in an absolute sense. But, as ever, it depends on whether this company can resume earnings growth. A related, but more interesting, question is whether Microsoft can maintain its current very high level of free cashflow generation.

The competitive headwinds facing Microsoft come from a variety of sources. The move to open-source software threatens its core revenues from operating systems and sales of programs like Word and Excel. Newer sources of revenues - essentially web-based advertising - are, of course, being grabbed by Google. Regulators on both sides of the Atlantic have chipped away at Microsoft's profitability, successfully arguing that the company has followed monopolistic practices. Customers have been frustrated by the repeated delays to the launch of the next-generation operating system. Put like this, it is hardly surprising that the share price has suffered: the gloom has been pretty relentless.

But at some point, all of the concerns are priced in and, given the way markets typically behave, too much pessimism is factored in. I think we are now at that point.

The world's leading tech analyst (and lots of others) lowered his rating on Microsoft in the wake of its profit warning, arguing that "it sounds like Microsoft is building a Google or Yahoo inside the company". I'm not sure that I would have put it this way: any company threatening to replicate the Google experience should, presumably, see its share price rise to the stratosphere. If Microsoft is about to mount a credible challenge to Google, then its shares are a screaming buy.

In this regard, it has been interesting to note that Google is shouting rather loudly about the "unfair" practices being adopted by Microsoft with regard to aspects of its new browser, currently in beta testing. When the competition is moaning about you, then you must be doing something right.

The issue, of course, comes down to the returns that we can expect Microsoft to earn on the extra investment that so displeased the analysts. Extra capital spending is, of course, only a bad thing when it doesn't earn an acceptable return. If this is what the analysts are arguing, then they were right to downgrade the stock. But they seem merely to have been miffed that the company delivered such a large surprise (rather than leading the analysts by the nose - the more familiar way of doing things). There was also a knee-jerk "all big capital spending programmes are bad" response: this is a legacy of the instant gratification climate of the last few years, which has seen the markets demanding cash now, rather than vague promises of future returns.

This short termism is an understandable response to the bubble years (when there was no cash today, everything was a promise about tomorrow), but it has gone too far. Companies do have to invest for the future and this always, by definition, comes at the expense of today's cashflow. Too many smart people seem to have forgotten this.

The numbers are fascinating. Put simply, if we assume Microsoft's extra spending does nothing to alter the profile of its cashflows, the right share price is $24, roughly today's level. In other words, the market is assuming that the money is going to be wasted.

But if we give Ballmer some credit and allow the company's current rate of return on capital to be merely maintained (not increased) for three years - rather than fading immediately as previously implicitly assumed - as a result of the extra spending, the correct share price is $30. (Full disclosure: all calculations based on in-house modelling technology at Collins Stewart.) Three years is not terribly demanding and only gives the company partial credit for achieving its objectives. That's the role of the dice: will the company succeed? I'll take that bet.

Chris Johns is an investment strategist with Collins Stewart. All opinions are personal