Technology stocks - Bubble or opportunity?

The investment case for technology goes something like this: "Technology remains uniquely positioned at the sweet spot of the deflationary cycle. Not only is the industry one of the primary drivers of deflation in the global economy it is also its primary beneficiary as companies, faced with diminishing profitability as a result of the loss of pricing power, are forced to invest aggressively in technology in an attempt to restructure their cost base. And, as Microsoft boss Bill Gates observed, "everything changes with the Internet" and there are not too many other industries on the right side of that change".

Valuing Internet stocks presents the analyst with challenges almost unlike anything that has gone before : how does one put a value on a business model which didn’t exist a couple of years ago, which will almost certainly be fundamentally different in a couple of yeas time and which at best will not produce any profits until 3, 4 and 5 years down the track?

Most conventional valuation methodologies rely on capitalising profits or cashflow to derive value but the technology or Internet analyst has no clear idea of cashflows let alone profits so he or she must make key assumptions about sales and margins well into the future when not only are the actions of competitors, costs and markets uncertain but the very product that the company is selling is also constantly evolving. Furthermore the Internet compresses distance permitting global competitors to converge on a specific opportunity – last week Commonwealth Bank of Australia announced that it would shortly offer online banking in two European countries. How many profit forecasts for Deutsche Bank would have anticipated this increased competition?

The advent of tech stocks has not changed the basic fundamentals of investment analysis, ie: that the value of any stock is simply the net present value of future cash flows. What it has done however is to make estimating the cashflows extremely difficult and getting it wrong, more dangerous.

One of the US’s top Internet analysts, at Morgan Stanley Dean Witter, told me that her team values Internet stocks using DCF and discounted price to earnings multiples. They use market capitalisation to revenue or market cap to gross profits ratios to determine relative value in the sector.

There is nothing particularly clever or different about this, indeed I used the same discounted cash flow technique to value paper machines some 20 years ago. However the reason these people are amongst the highest paid analysts on Wall Street is the expertise they bring to bear in determining the assumptions underlying the analysis – which company’s business model is likely to succeed, how will business to business e-commerce develop, what rates of subscriber growth and gross margins are appropriate, is the company’s business model scaleable ie can it be expanded internationally at low cost, etc.

Deutsche Asset Management who manage the UK and NZ listed investment trust, Anglo & Overseas, recently prepared a paper for their key fund managers outlining their thoughts on how to analyse Internet stocks. They begin with something of a disclaimer; a quote from the chairman of Intel who, when asked what return on investment he expected from Intel’s e-commerce efforts, said "Are you crazy? This is Columbus in the New World. What was his ROI?"

Deutsche believe that qualitative factors are particularly critical. They look closely at brand, scale, alliances and acquisitions, innovation and customer service. As far as quantitative analysis goes – running the numbers – DB’s US arm (BT Alex Brown) have developed a technique they call Theoretical Earning Multiples Analysis whereby the analyst forecasts the revenues of the business over five or so years and estimates the long term operating margin of the business. They then apply that operating margin to their revenue forecasts, tax it and create a theoretical earnings number which can be compared with other publicly listed companies. Other approaches used are revenue multiples, registered user multiples, and unique user multiples. What all the above distils down to is:

1. Forecasting is difficult and imprecise due to the Internet volatility of the industry and the long lead times between investment and positive cash flows.
2. Valuations are used for comparative purposes to other companies in the same industry as much as anything else.
3. PE multiples only fall back into double figures after five or so years of stratospheric growth. Such growth is forecast not only because the market is growing rapidly but because much of the typical Internet company’s cost base is fixed giving it very high operating leverage. Furthermore the early years of an Internet companies life are often characterised by the aggressive pursuit of market share with attendant high marketing costs. When the marketing spend falls margins will skyrocket, at least that is the theory.
4. As investment pioneers Graham and Dodds wrote in 1934 "the buyer of (new companies in new sectors) such securities is not making an investment but a bet on a new technology, a new market, a new service".
5. Every "buy" recommendation for the sector devolves from the belief that "the net changes everything".

One way perhaps of understanding what logic is behind the compound shareprice growth of 300 per over three years for a stock like Yahoo is to look at the fundamentals : at a recent price of $397 per share it is trading at 827 times 1999 earnings. Crazy – perhaps. However, analysts believe that Yahoo’s earnings will grow at a rate of 75 per cent annually for the next five years based on consensus sales and margin forecasts (Investors generally belief that the tech sector is so strong the ups and downs of the economic cycle are irrelevant). That means its PE in five years time will be 50x, not too outrageous.

Now five years is a very long time and 75 per cent pa compound growth is rather rapid but Yahoo’s profits rose almost 10 fold in 1999 and if you are a fund manager with the responsibility for 20 million individuals retirement resting on your shoulders can you take the risk and ignore what people are calling the biggest change since the industrial revolution? I don’t think so. And the same goes for the average New Zealander who has responsibility for his or her own retirement. Maybe ignoring the tech sector is more risky than buying into it.

The above discussion does not argue that technology stocks are necessarily fairly priced although the extent of the resource devoted to researching the sector suggests that it probably is. They may be overpriced, they may even be cheap. The implication is however that current prices are based on a set of assumptions and a standard methodology by a large number of market participants highly incentivised to "get it right".

Valuations are not merely the result of mania, speculation and greed by naive individuals. The main reason tech stocks are as high as they are can be summed up in one word "growth". It may be appropriate however to leave the last word on investing in the new world to one of the doyens of investing in the old. Warren Buffet reportedly suggested that all business school students should be asked to do a paper justifying Internet company valuations and that any who attempted it should be failed.

Brent Sheather is a stockbroker in Whakatane.

Commenting is closed