Valuation of Internet Companies

Published on May 2016 | Categories: Topics, Art & Design | Downloads: 67 | Comments: 0 | Views: 666
of 12
Download PDF   Embed   Report

Comments

Content

Valuation of Internet companies: A Survey of the Evidence

Pim Jansen Rabobank International, London

Enrico Perotti University of Amsterdam and CEPR

Abstract1: In this article, we review the recent academic research on the valuation of Internet companies. In particular, we focus on the valuation method(s) which were said to be suitable for new economy start ups in the boom years 1998-2000, and conclude that they were neither novel nor very accurate. Since the downturn in the sector, the valuation focus has returned to advanced fundamental valuation methods such as real options valuation.

1

This article is an abbreviated version of a paper, written as part of the Master in International Finance Program of the University of Amsterdam

“The silliest claim was that in this New World of rapid technological change, old methods of share valuation had become irrelevant. Profits were for wimps, it claimed. Falls of 90-100% in the share prices of loss-making dot.com firms show that profits do matter after all” (The Economist, May 12th 2001)

At the height of the Internet craze, flotations of Internet companies have created many millionaires. It was argued at that time that new rules of valuation had to be applied in the New Economy. Revenues and growth potential, rather than cash flows, were to be the new foundation of value. After the abrupt fall of the NASDAQ index in the spring of 2000, reality set in. Many Internet companies continued to burn cash and to generate significant losses, and many have since gone bankrupt. Furthermore, in the financial press, among economists and among respected academics around the world, a broad discussion has finally developed on the value creation of Internet companies. In this article, we review the recent academic research on the valuation of Internet companies. In particular, we investigate the valuation method(s) which were said to be suitable for new economy start ups. S pecific characters of Internet Firms Internet firms are often strikingly different from other firms. Consider the results from a study by Hand [2000a] on a sample of 274 Internet firms, 274 non-Internet firms and 213 IPO-matched non-Internet firms. An Internet firm has over half its revenues from Internet. The median Internet firm has ten times the market capitalisation yet employs only 40% of the number of people from a median firm. Relative to the median firm, the median Internet firm also has more than three times the beta risk (2.55 versus 0.78), one third of its stock held by institutions (8% versus 27%), half as much of its issued shares in public float (31% versus 62%), a public float turnover that is 6.5 times faster (once every 19 versus 143 trading days) and five times as much of its public float sold short (5% versus 1%). This pattern holds also when Internet firms are compared with IPO-matched firms. Another striking feature is the underpricing of the median Internet firm: the Internet firm is four times as underpriced at its IPO as the median IPO matched non-Internet firm (37% versus 9%) with the mean underpricing for Internet firms reaches 69% to be compared with an average underpricing for all US IPOs over the period 19601996 of 16%. These differences are unusual but not unique historically. A study by Amir and Lev [AL, 1996] for the ten years beginning 1984, reported that 69% of quarterly EPS of the 14 independent cellular telephone companies they examined were negative. AL also report that t he corresponding figure for 44 biotechnology companies over the same period was 72%. This compares to Hand’s result of 77% of Internet firms reporting negative EPS over the period 1997:Q1-1999:Q2, suggesting that Internet firms may be no more unprofitable than other groups of firms in earlier technology-based, high growth industries which went through an IPO cycle. Classification of Internet firms We start by a general classification of the numerous types of Internet companies that have emerged in the last few years. The main distinction in use is between Business to Business (B2B) and Business to Consumer (B2C) Internet firms (e.g. Bowen, Davis and Rajgopal [2000], Demers & Lev, Davis [2001]). Across these categories, the most widely used in the USA (from www.internet.com) distinguishes: 1) e -tailers and e-commerce, (2) Software, (3) Enablers, (4) Security, (5) Content & portals, (6) High speed and infrastructure and (7) ISP and access. Trueman Wong and Zhang [2000] especially focus on the difference between e-tailers (producing revenues by attracting visitors to their web sites and selling products) and the p/c firms (portal- and content community firms, who depend for their revenues largely on advertising). Keating, Lys and Magee [2001] make a distinction between firms that market or sell primarily via the Internet (“direct” firms, or B2C firms) and firms that provide Internet infrastructure (“support” firms). Perotti and Rossetto [2000] make a further distinction within the content and portal category, distinguishing between vertical portals, who specialise in the sales of one product (e.g. Amazon focusing originally on selling books vi Internet) and horizontal portals, who sell all sorts of products or give access to multiple services.

2

Financial Value Drivers We now move to review common and less common financial measures used for valuing Internet firms and present some evidence on the value drivers of stock valuation. Is accounting information useful for In ternet valuation? Hand [2000a] assesses the main accounting measures listed Internet firms operations and relates them to their stock market valuations. He finds evidence that basic accounting data is value relevant, albeit in a non-linear manner. Overall, he concludes that Internet firms’ log-transformed market values are neatly linear in both logtransformed book-equity and log-transformed net income. This suggests a remarkable dispersion. Keating, Lys and Magee [2001] conclude that annual report information indeed explains a significant portion of the cross-sectional variation in valuations at both March and May 2000. They find that, for support firms, the annual report provides 20.5 times the explanatory power on valuation relative to bits of information provided by analysts’ forecast and earnings surprise information (the new information) combined2. Furthermore, the contribution of the annual report information to explaining valuations was significantly higher than new information. Perhaps the more succesfull Internet firms were expected to spend more on customer and product development than was generated in revenue during 2000, resulting in a lower forecasted earnings for the year. The earnings expectations appear to have changed by the end of May. The annual information provides relatively 21.3 times the information provided by analysts forecasts and earnings surprises combined, significant at the 0.01 level. Further they observe that the coefficient on Gross Profit is significantly positive, while the coefficients on R&D and Other Expenses are significantly negative. They conclude that firms earning greater profits in 1999 experience a relatively smaller stock price decline in spring 2000, while firms that spend more on research and development in 1999 experience a relatively larger stock price decline. Adjusted accounting information Accounting measures often need to be adjusted for the purpose of valuation. A primary adjustment to earnings is to subtract the cost of committed capital. This requires a precise measure of capital. Economic Value Added (EVA 3) considers R&D, marketing and advertising expenses as investment rather than expenses. Using EVA, a greater percentage of the value appears in the earlier years, where forecasting is easier. When initial costs (R&D, marketing and advertising) are capitalised as investments, this creates higher earnings in the earlier years, where forecasting is more practical. When a considerable part of a company’s value is in future growth opportunities, a better u nderstanding of intrinsic value may be gained by applying real options’ techniques. We will discuss this issue later in the article.

Price-to-sales ratio Demers & Lev [2000] examine the value relevance of two categories of Internet companies’ expenditures related to the acquisition of intangible assets: (i) marketing expenses and (ii) product development and R&D expenses. They hypothesise that the market will positively value both of these variables as long as it views them as positive NPV investments. DL also examine the value relevance of the income statement components cost of goods/services sold (COGS), R&D and marketing expenses. With the exception of COGS, the income statement components are significantly value relevant in 1999. This is consistent with studies on start-up industries (e.g., Amir and Lev [1996]) and R&D intensive firms (e.g., Lev and Sougiannis [1996]). Their findings suggest that early in the Internet boom, the market viewed B2C Internet companies’ material expenditures directed towards customer acquisitions and product development as investments rather than expenses. Following the major crash of NASDAQ in early 2000, the market appears to have shifted in valuation approach. In the year 2000, COGS is negatively and significantly associated with P/S-ratio but Marketing and Product Development expenses no longer explain Internet companies price-to-sales ratio. These findings suggest that in the year 2000, the market is no longer willing to implicitly capitalise expenditures on intangible assets in valuing Internet stocks. DL conclude that ‘the market continued to capitalise these expenditures during the first quarter of 2000 (i.e. until the bursting of the bubble) but then stopped capitalising them and/or began treating them as expenses in the second quarter of 2000’.

2 3

Based on the relative information content provided by Theil (1987) EVA is calculated as [adjusted operating profit after tax-/-(cost of capital x adjusted capital employed)] 3

Using aggressive accounting practices: barter trade & grossed up revenues Stimulated by concern raised by the regulatory bodies Securities and Exchange Committee (SEC) and the Federal Accounting Standard Board (FASB), Angela Davis [May 2000, AD] studies the value relevance of revenues for Internet firms. In particular, she studies whether reporting grossed-up or barter revenues has any impact on their valuation between January 1998 and November 2000. Consistent with prior studies she finds a positive association between market value of equity and revenue announcements. This provides evidence for the notion that Internet managers had incentives to increase reported revenue, possibly through the use of grossed-up and/or barter revenue. Davis concludes that use of these aggressive accounting techniques appears to be concentrated in only a few Internet sectors, particularly e-tailers, content/community firms and portals. There is a positive and highly significant relation between quarterly value and revenue for these firms in both the pre- and post-crash periods. She finds that there is no statistically significant positive relation between market capitalisation and earnings in the pre-crash period but it becomes positive and statistically significant in the post-crash period. She further concludes that the market’s valuation of revenue declines from the pre- to the post-crash period for firms reporting grossed-up revenue and/or barter revenue while this remains virtually unchanged for firms not reporting grossed-up or barter revenue. Additionally, she shows that the larger the number of individual investors following an Internet company, the less value is attributed to reporting grossed-up revenue and barter revenue. This suggests a changing investor behaviour in the post crash period. In a more recent study, Bowen, Davis and Rajgopal [BDR 2001] focus on economic motives underlying management’s accounting choices of Internet companies. First, BDR document the extent to which Internet companies use allegedly aggressive revenue recognition policies and confirm the results of Davis [2001]. Second, they examine the economic incentives that potentially influence firms to choose aggressive revenue accounting practices . Where Davis [2001] and Bagnoli et al [2001] find that the market responds to revenue surprises, BDR document that revenue levels are strongly associated with market values of Internet firms. This leads them to conclude that managers of Internet firms have economic incentives to report high levels of revenue. Since this strategy cannot be sustained for long, the question is why Internet managers seek to fool the markets. Specific incentives to influence stock prices according to BDR include (1) stock option compensation and (2) access to (and cost of) equity capital. Furthermore, Internet managers also likely have incentives to influence third parties (e.g. capital providers other than stock market investors, suppliers and customers by reporting higher revenues). Similarly, they find a positive association between the use of both barter and grossed-up revenue and the extent of cash burn, suggesting that pressure to seek external financing influences the aggressive revenue recognition choices of Internet firm managers.

Cash-burn ratio Demers and Lev [DL, 2000] construct a proxy for B2C companies’ ability to sustain their current rate of cash burn. Their proxy for cash burn is cash on hand divided by current period’s cash flows from operations. They find that this proxy is a significant value-driver in both 1999 and 2000. Their proxy is defined as [cash on hand]/[current period’s cash flows from operations]. For both 1999 and 2000, this ratio is significantly associated with the priceto-sales ratio of Internet companies. For companies with negative cash flows form operations, the ratio is negative, and therefore the significantly positive coefficient results in a reduction to overall firm value. The intuition for this might be, that free cash flows (i.e. significant cash stores and limited cash burn) provide these Internet companies with greater option value for growth. Thus cash provides these companies with the flexibility to adapt to rapidly changing market conditions and to react to emerging opportunities.

4

“Non-financial” value drivers In this paragraph we discuss studies analysing non-financial value drivers. Web metrics will be discussed extensively, together with managerial actions, strategic alliances and stock options. Web site usage: indicator of potential future demand Trueman, Wong and Zang [TWZ, 2000] consider the role of non-financial data in the valuation of Internet stocks. They provide insight into valuation of Internet stocks by examining the impact of both accounting information and measures of Internet usage. First they relate Internet stock prices with accounting information. To that extend they decompose the firm’s earnings in three components: gross profits, operating expenses and non-operating expenses. They assume future gross profits to be positively and linearly related to a current period’s gross profit, operating expenses and web site usage. This latter is based on the assumption that current period Website usage reflects potential future demand for the companies products and affects the rates a firm can charge for advertising on the company’s Web sites. Web usage is measured alternatively by the number of unique visitors to the firm’s Website and by the number of page views at its site. Consistent with those who claim that financial statement information is of very limited use in the valuation of Internet stocks (such as Stern Stewart, Bagnoli et al), TWZ are unable to detect a significant positive association between bottom-line net income and their sample of market prices. In fact, the association is actually negative. However, when decomposing the net income into its components, they find that book value and gross profits are significantly and positively related to stock price. Market values of web businesses in relation to web traffic Rajgopal, Kotha and Venkatachalan [RKV, 2000a] analyse the importance of web traffic as a non-financial indicator of the market values of B2C Internet firms. They examine a sample of 92 firms (compared to ‘only’ 56 firms for TWZ) over the five quarters beginning with the first quarter of 1999. They find that the number of unique monthly visitors to a site is positively associated with stock prices and adds significant incremental explanatory power to a valuation regression including just financial statement information. Internet companies use many strategies to generate traffic, such as an alliance with America Online (AOL), extensive marketing expenses and media visibility. Consistent with TWZ [2000b], RKV find that web traffic levels predict one- and two -quarter ahead sales. However, traffic has no incremental information about future revenues once the predictive information in past sales is controlled for. Hence, the market does not appear to value traffic merely because it predicts future sales. Finally, RKV find that the market values of web businesses are positively associated with the number of unique visitors to the firm’s website. While RKV offer important insights to the literature, the direct link with the valuation of Internet companies is weak.

Log linear regression on web traffic data Hand [2000b] claims to clarify the previous research [i.e. RKV 2000a, TWZ 2000a, Hand 2000a] by investigating the value-relevance of a more comprehensive set of web traffic measures. He assesses the value-relevance of unique visitors, household reach, page views, the number of hours spent on the page views and the gender, age and income levels of the visitors. This reveals the value relevance of supply and demand forces on Internet stocks. He regresses the log-tra nsformed market value of equity on log transformations of three groups of factors: economic fundamentals, web traffic and supply and demand forces. Economic fundamentals are captured by current book equity, forecasted one-year earnings and forecasted long-run growth in earnings. Contrary to popular perception, he finds that the prices of Internet stocks are not driven by web traffic as much as by economic fundamentals. Current book equity, one-year-ahead forecasted earnings and long-run-forecasted earnings growth dominate in explaining cross-sectional variation in Net stock prices. Only marginally are Internet firms’ market values related to only one of the three measures of web traffic: the number of unique visitors to the firm’s web site. Supplementary tests also indicate that Internet firms’ equity market values are unrelated to the average income, age or gender of Website visitors. Third, after controlling for 5

economic fundamentals and web traffic, Internet firms’ equity market values are negatively correlated with their public float, indicating that Internet firms stock prices are higher the lower is the fraction of total shares available for trading. In contracts, Internet firms’ equity values are positively correlated with their short interest and institutional ownership, suggesting that Internet firms benefit from the stability and/or reduced irrational trading risk provided by greater institutional ownership. In summary, Hand concludes that pricing of US Internet stocks is dominated by expectations of near- and long term profitability, although they are also uniquely impacted by some non-traditional value-drivers such as web metrics. Using lagged web-traffic metrics Demers and Lev [DL, 2000] expand upon the prior web metric studies by examining the value-relevance of three web performance factors both before and after the Internet market correction in March-April of 2000: (i) reach (attraction of new visitors to a Website), (ii) stickiness (retention of visitors on the site) and (iii) customer loyalty (ability to generate repeat visits from surfers who have been attracted to the site in the past). They find that both contemporaneous levels and changes in various web traffic metrics are significantly correlated with monthly stock return in each of 1999 and 2000, but the significance levels decrease in the later time period, suggesting a reduction in the implicit valuation of web traffic. More strikingly, the one-month lag in web traffic levels is significantly correlated with monthly stock returns in 1999, but the significance disappears in 2000. Thus, while investors generally appear to react promptly to the release of traffic measures, the results suggest that there may have been some delayed reaction by the market in the earlier stages of the Internet economy. DL’s findings of significance for the year 2000 on reach and stickiness confirm that web traffic measures remain important after the burst of the bubble in 2000. Thus, even as the Internet sector begins to mature and B2C companies develop longer operating histories (so that a longer time series of financial valuation variables becoming available), web traffic metrics that were relevant during the bubble period of the market continue to be significant determinants of Internet companies’ price-to-sales ratios after the Internet shakeout, at least throughout 2000. The impact of financial analysts Lui and Song [LS, 2000] study the role of financial analysts in the Internet stock bubble in spring 2000. They try to test if financial analysts as a whole overestimated the prospect of Internet companies. To this extent they examine the change of financial analysts’ earnings forecasts around the crash in April 2000. They conclude that: ? Financial analysts as a whole were too optimistic about the Internet stocks before the crash. The percentage of analysts who were optimistic declined significantly after the crash (from 41,5% before the 33.3% during the crash quarter and 35.6% one quarter later). Analysts revised down their forecasts significantly after the burst of the bubble; ? Although financial analysts seemed to change their forecasting significantly for the quarterly earnings, they remained quite upbeat on the long-term prospects of the Internet firms; ? Forecasts of analysts whose affiliated brokerage firm had done initial public offerings (IPO) for the Internet companies are more negative in their quarterly earnings forecasts relative to those of non-affiliated analysts. Especially on the quarter after the crash, the financial analysts had 19% of the optimistic forecasts, while non -affiliated analysts had 35%. This is important evidence that some investment banks oversold Internet IPO’s. Bagnoli, Kallapur and Watts [BKW, 2001] perform a similar study. They examine the characteristics and relative information content of revenue and earnings news for Internet firms during and after the bubble. In line with TWZ [2000b] they find a tendency for analysts to underestimate revenues. Keating, Lys and Magee [KLM, 2001] examine the impact of analysts fo recasts as well as new disclosures on Internet stock prices. Financial analysts’ average buy/sell recommendation became less favourable for pure Internet firms and more favourable for support firms. While statistically significant, the changes were however small and do not suggest a major systematic shift in analyst’ attitudes towards Internet stocks. KLM conclude that analysts recommendations were mixed: less favourable recommendations for pure firms and more favourable recommendations for support firms. The only significant adverse changes were seen in fiscal year 2000 earnings forecasts which were lowered significantly.

6

Managerial actions: rose.com Cooper, Dimitrov and Rau [CDR, 2000] analyse the effect of a corporate name change (i.e. taking a name that ends on .com or .net). In the popular press, CDR came across several articles that mention extremely large returns earned by these companies. These articles suggest the large returns being an effect of irrational day traders searching stock chat sites on the Internet. Companies that announce dot.com name changes between June 1, 1998 and July 31, 1999 gave a mean return of 142% above that of similar companies during the period between 15 days before changing their name and 15 days after changing their name . The return is 122% for Internet companies and 203% for companies that have no relation with Internet. The change in value does not appear to be temporary! CDR do not find evidence of a negative post-announcement drift even when removing the more extreme observations. CDR question whether it is a rational response for investors to apply a ‘premium’ to a dot.com stock. They find that in the shorter horizons, market participants appear to apply a similar positive price premium across all companies changing their names to dot.com names, regardless of a company’s level of involvement with the Internet. In the longer horizon, and with the caveat that the sample size of the category not Internet related is very small, firms that have less involvement with the Internet have the greatest returns following a dot.com. Overall a mere association with the Internet seems enough to provide a firm with a large and permanent value increase. CDR conclude that their results indicate irrational exhuberance by investors to be associated with the Internet at all costs. However, they acknowledge that only time will prove whether investors are rational in pricing large expectations of future earnings from the Internet into the stock price. The value relevance of strategic alliances Analysts’ reports and anecdotal evidence suggest that strategic alliances are potentially important value drivers for Internet stocks. Demers & Lev [2000] find support for this argument after examining whether these are still positively valued by the market in 2000. A strategic alliance with AOL has a positive relation to P/S ratio of B2C stocks in 1999, which complements the view of RKV [2000] who document a weak positive association between alliances with AOL and web traffic measure for reach during 1999. In 2000, they find that AOL alliances are not adding value to B2C stocks. They also find that the cumulative sum of alliances entered into is significantly negatively associated with the P/S ratio in both 1999 and 2000. Stock option plans Keating, Lys and Magee [2001] examine the valuation effect of Internet companies granting employee stock options. Stock grants are a form of employee compensation that many regard as an expense. Most companies grant options with strike price equal to the current market price which under current rules results in no expense recognition. In addition, changes in market conditions which make stock options less attractive for employees will force firms to offer more cash to attract and retain them, so in this case there is a risk that the company will not be able to fulfil investor’s growth expectations. KLM argue that if these effects dominate investors’ valuation of Internet firms, then the coefficient on stock option grants in the valuation regression should be negative. KLM construct as stock option variable the fair market value of options granted as disclosed during 1999 in the annual report footnotes. They expect this variable to change from March 2000 regression to the May 2000 regression in both a valuation model and a returns model. They conclude that investors positively view stock-option grants, most likely due to their effect on employee retention.

7

Using the real option approach to value growth opportunities The evidence on Internet firm valuation suggests a frequent use of non -financial information, which suggests that the market is attempting to use proxies for growth opportunities. We review the leading approach for their valuation: the real option approach. Using real options Real options represent the most recent capital budgetting techniques to value growth opportunities. Schwartz and Moon [SM, 2000] and Perotti and Rossetto [2000] argue that Internet companies have call-option characteristics since they have large potential upside and limited downside potential (i.e. bankruptcy). Furthermore, Internet companies have volatile revenues, rates of growth in revenues, costs, cash flows and stock prices. Since options are often claims on highly uncertain outcomes, these valuation techniques may be used to value Internet companies. Many strategic investments create opportunities that eventually will or will not be taken. In these cases, the investment opportunity can be viewed as a stream of cash flows plus a set of options. The real options approach involves options on real (non-financial, non-traded) assets. Uncertainty is defined as the unavoidable randomness of the external environment. Exposure to uncertainty is by consequence defined by a multiple of factors and can be only partially controlled by managerial actions. Risk is thus the potentially adverse consequence of a firm’s exposure. The traditional valuation models are not appropriate to deal with these uncertainties since they normally use a single expected value of the future cash flows and it is difficult to find the appropriate discount rate for the options present (e.g. exit option) methods. SM [2000] argue, using a conventional DCF model with stochastic growth rates, that depending on the parameters chosen and given high enough growth rates of revenues, that the value of Internet stocks can be rationalised. Even when the chance that a company may go bankrupt is real, if the initial growth rates are sufficiently high and if there is enough volatility in this growth over time, valuations can be what would otherwise appear to be unbelievably high. In addition, they find the valuation has great sensitivity to initial conditions and exact specification of the parameters. This finding is consistent with observations that the returns of Internet stocks have been strikingly volatile. In 2001, SM apply the model to price the Internet company Ebay. Its market price at the time of measuring (11 April 2000) was USD 39.17, 75% above the model price. The reasons could be that the market is implying higher margins, growth rate expectations and growth rate volatility and/or the model is missing some key aspects of value. (Alternatively Ebay was overvalued at that date; the more recent stock price information suggests the latter). Even though real option valuation may not justify Internet valuation, it deals explicitly with option-like characteristics of Internet companies (i.e. asymmetric payoffs, large uncertainty in sales and sales growth and cost uncertainty)

Internet portals as portfolios of entry options Perotti and Rossetto [PR, 2001] investigate the valuation of platform investment, such as a software operating system or an Internet portal webpage. They define a platform investment as a creation of an innovative distribution and production infrastructure which increases access to customers; as a result it reduces entry costs in related products. A portal is a point of entry connected with relevant information or access to products. It lowers the cost of reaching customers and thus the cost to entry new market segments. PR make a distinction between horizontal and vertical internet portals. They relate the platform investment to the strategic real option literature. They build a model of the value of associated strategic advantages and compare the relative valuation of such ‘platform stocks’ versus conventional producers (old economy). In the comparative statistics they focus on the effects of uncertainty and show that it produces the greatest difference between option based and NPV based models of valuation. The effect of uncertainty on the strategic advantage and thus of the platform value is not obvious as there are two future countervailing effects: the higher the uncertainty the higher the value to wait, but at the same time the higher the expected profits in case of immediate cross entry.

8

The important result is that in a context of increased uncertainty, the relative value of platform to traditional strategies increases; the value of waiting to invest rises, but the value of platform increases even more. In some cases, platforms can reduce entry by making a parallel monopoly sustainable. PR conclude that relative to conventional producers, portal related firms enjoy enhanced entry options in uncertain market segments, which may be valuable under some market conditions on internet usage. Their advantage lies in the ability to exercise entry options at the optimal strategic timing. Conclusions Based on the overview that has been given of recent academic research on the v aluation of Internet firms, we draw the following conclusions: I. Traditional accounting data remains important for valuing Internet companies yet the link between accounting numbers and Internet valuation is tenuous at best. ? Large marketing expenses, R&D and selling expenses seem to be activated as intangible assets or valuable investments; ? Internet firms which were profitable in 1999 are valued more realistically after the burst of the bubble (April 2000), when bottom line profit does matter much more; ? Aggressive accounting techniques (grossed-up revenue and/or barter revenue) may have responded to exaggerate the focus on revenues. However, revenues are only seen as main value driver before the NASDAQ crash. Thereafter the relationship seems to break down . ? Managers have economic incentives (stock option compensation and access to/cost of equity capital) to report higher revenues. Web traffic is not a major value driver for Internet companies. Web traffic is only value relevant for e-tailers; more precisely it appears in fact to be used as a good predictor for future sales; ? The significance of web traffic as a driver of monthly stock return disappears after the burst of the bubble; ? If any, than only number of unique visitors is a useful web traffic variable in relation to valuation of Internet stocks.
?

II.

III.

Financial Analysts stimulated the overvaluation of Internet stocks: Analysts were too optimistic in their forecast of quarterly earnings before the market crash, revenues were significantly downgraded after the burst of the bubble; ? Long term prospects remained upbeat after the burst of the bubble; ? Focus during the boom shifted from profit to earnings; ? Support firms were revised upwards after the burst, B2C revised downwards, thus more value is attributed to infrastructure providers.
?

IV.

New valuation factors deem unsustainable: Corporate name changes into dot.com, strategic alliances and stock option plans in the long term are not adding value to Internet stocks, only before the burst they have an impact on Internet stocks. New approaches to valuing internet stocks are not new: a. Volatility of growth in revenues is a key value driver for Internet companies, this option variable rationalises the high valuation of Internet stocks; b. Internet platform related firms enjoy enhanced entry options in uncertain market segments, to be exercised at the optimal strategic timing. The better the timing, the higher the value.

V.

All in all, we find no convincing evidence on new valuation techniques and/or value drivers for Internet stocks. Web traffic has important effects on predicting future sales and thus contributes to expectations on near- and long-term profitability but has no consistent direct impact on the valuation of Internet stock valuation. For strategic alliances, granting employee stock options and changing the company name into dot.com, only the latter impacted on the valuation of Internet stocks. This suggests that it is rather irrational exuberance before the crash than extreme optimism. Extreme optimism did play a role though. It is generally agreed upon that Internet stocks have many uncertainties -e.g. volatile (rates of growth) of revenues, cash flows and earnings- and it is a business that has little historical track record. However, after the burst of the bubble it has become obvious that the 9

expectations (even those from financial analysts) were too optimistic and that this was the main reason for the overvaluation of Internet stocks. Traditional valuation techniques have not lost any of its relevance. It is just a matter of using the correct parameters and presenting them in a correct matter in the financial accounts to make the valuation of Internet stocks a ‘mission possible’!

10

REFERENCES Bagnoli, M., Kallapur, S., Watts, S., “Top line and bottom line forecasts: a comparison of Internet firms during and after the bubble”, Krannert Graduate Schoool of Management, Purdue University Brealey, R.A. and S.C. Myers, “Principles of Corporate Finance, sixth edition, 2000, Irwin McGra w-Hill Cooper, M.J., Dimitrov, O., Raghavendra Rau, P., “A rose.com by any other name”, November 2000, Purdue University Davis, A.K., “The value relevance of revenue for Internet firms: does reporting grossed-up or barter revenue make a difference?”, May 2001, Journal of Accounting Research, conference version, Washington University, St. Louis; Demers, E., Lev, B., “A Rude Awakening: Internet Shakeout in 2000”, September 2000, William E. Simon Graduate School of Business Administration, University of Rochester Financial Accounting Standard Board, EITF Abstracts 99-17, Accounting for Advertising barter transactions, January 2000 Financial Accounting Standard Board, EITF Abstracts 99-19, Reporting revenue gross as a principal versus net as an agent, May 2000 Greene, W.H., “Econometric Analysis”, second edition, 1993, Prentice-Hall Inc. Englewood Cliffs, New Jersey, p.702 Hand, J.R.M., (2000a), “Profit, Lo sses, and the Non -linear Pricing of Internet Stocks”, draft January 2000, KenanFlagler Business School, UNC Chapel Hill Hand, J.R.M. (2000b), “The role of economic fundamentals , web traffic, and supply and demand in the pricing of US Internet Stocks”, April 2000, Kenan-Flagler Business School, UNC Chapel Hill Ho, N., Hui, N. and Li, L., “Does EVA © beat earnings? Round 2: evidence from Internet companies”, XXX 2000, University of New South Wales Keating, E., Lys, T., Magee, R., “The Internet downturn: finding valuation factors in spring 2000”, January 2001, Kellog graduate School of Business, Northwestern University Kozbeg, A., “The value drivers of Internet stocks: A business models approach”, April 2001, Stern School of Business New York University Liu, Q., Song, F., “The rise and fall of Internet stocks: should financial analysts be blamed?, March 2001, University of Hong Kong Milano, G., Stern, E., Fencl, T., Piza, N., “Internet valuation, why are the values so high?”, February 2000, Stern Stewart Europe Limited Perotti, E., Rossetto, S., “Internet portals as portfolios of entry options”, Preliminary draft November 8, 2000, University of Amsterdam and CEPR Rajgopal S., Kotha S. and Venkatachalam M. (2000a), “The relevance of web traffic for Internet Stock Prices”, October 2000, University of Washington/Stanford University Rajgopal S., Kotha S. and Venkatachalam M. (2000b), “Managerial actions, stock returns and earnings: the case of business to business Internet firms”, October 2000, University of Washington/Stanford University Roon, F. , “Valuing Dot.com”, September 2000, Erasmus University Rotterdam

11

Schwartz, E., Moon, M.. “Rational pricing of internet companies” in Financial Analysts Journal Volume 56, no. 3, May/June 2000, Association for Investment Management and Research Schwartz, E., Moon, M. September 2000, revised April 2001, “Rational pricing of Internet companies revisited”, Anderson School at University of California, Los Angelas. Shapiro, C., Varian, H., Information Rules, 1999, Boston, MA: Harvard Business School Press The Industry Standard, May 1, 2000, “Portals Start to feel the heat”, p63 Trueman, B. , Franco Wong, M.H. ,Zhang, X-J. (2000a), “The eyeballs have it: searching for the value of Internet stocks”, January 2000, Haas School of Business, University of California Berkeley, Berkeley CA Trueman, B. , Franco Wong, M.H. ,Zhang, X-J. (2000b), “Back to basics: forecasting the revenues of Internet firms, April 2000, Haas School of Business, University of California Berkeley, Berkeley CA Trueman, B. , Franco Wong, M.H. ,Zhang, X-J., “Anomalous stock returns around internet firms’ earnings announcements, April 2001, Haas School of Business, University of California Berkeley, Berkeley CA UBS Warburg Global Equity Research, May 2000, Navigating the I-valuation Jungle (www.ubswarburg.com/researchweb) Vuong, Q., 1989, "Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses," Econometrica, 57, p307-334. Wall Street Journal, Wingfield, N. , November 22, 1999, “The tricky task of tracking web users”, p. C1

12

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close