THE IPO SENTIMENT INDICATOR

Transcription

1 124 Annex Ann-Kristin Achleitner 1 THE IPO SENTIMENT INDICATOR An innovative method for forecasting IPO activity while allowing for the underpricing effect. The mood in the securities markets is of considerable significance for all market participants as well as for economic development in general. However, that mood is difficult to grasp without measurement tools, any conclusions drawn in this regard are inevitably vague. The Center for Entrepreneurial and Financial Studies (CEFS) at the Munich University of Technology wanted to find a remedy for that. Under the guidance of professor Christoph Kaserer, it applied the methods of the still relatively young field of behaviourrelated capital market theory (behavioural finance) 2 to develop the IPO Sentiment Indicator (ISI) 3 The ISI is an innovative instrument for gauging the current climate in the marketplace and ultimately providing support in terms of IPO activity. Design features and theoretic foundation While most other indicators for measuring market sentiment are limited exclusively to surveys of market participants, the ISI consists of two equally weighted components: the Underpricing Sentiment and the IPO Climate. The major innovation of ISI is the fact that it takes into account data from the secondary market as a part of determining the Underpricing Sentiment. Underpricing Sentiment The choice of a proper point in time plays a crucial role in an initial public offering. In literature, two differing explanations for this are to be found in the life cycle theories and the market-timing theories. While the former go on the assumption that there are particularly favourable phases or events in the life cycle of a company that lend themselves to an initial public offering, the latter emphasise the enormous significance of capital market conditions. 4 In the timing hypothesis, it is the level of underpricing, e.g. the difference between the issue price of shares and their first publicly traded price, 5 that is of great importance. For example, there exist findings from empirical capital market research, according to which IPO activity and underpricing exhibit a cyclical tendency. 6

2 Annex 125 The concrete process underlying this interrelationship evolves as follows: Investors draw conclusions on the price they are willing to pay for an IPO by, among other things, observing the price developments of recent IPOs. The greater the underpricing of previous new issues, the more investors expect positive returns from future issues, thus they are prepared to subscribe to additional IPOs. It follows that investors are inclined to project recent performance into the future. 7 In consequence, when investors perceive favourable underpricing in the market, it increases their willingness to participate in the next IPO (positive feedback loop). Issuers exploit this upbeat mood in the capital market by placing additional IPOs or increasing the size of previously planned new issues. 8 Because the issue prices are only partially adjusted to reflect investors greater willingness to subscribe, there results a self-reinforcing process in which more aggressive underpricing goes hand in glove with an increase in IPO activity. Naturally this process will keel over at some point, for example when it is determined that altered general economic conditions have led to an IPO being overvalued. At that point, a similar type of downtrend in activity can ensue. On the basis of this cursorily described interrelationship, it is inevitable that there is a delayed correlation between underpricing and IPO activity, something that ultimately constitutes the theoretic foundation of the ISI. It follows that the perceived underpricing, which is referred to here as the Underpricing Sentiment (USI), is a key com- 1 Sincere gratitude to Markus Ampenberger of CEFS for his support in the composition of this treatise. 2 The theoretic decision-making dimension of behavioural finance is mainly based on the Prospect Theory of Kahneman / Tversky (1979). For a good overview of the status of research in this field, see e. g. Stracca (2004) and van der Sar (2004). 3 Deutsche Börse AG, which assisted in the development of the ISI, has been publishing on its website since 1 January 2006 the current level of the ISI as it stands on the second Monday of each new calendar quarter. 4 See Ritter / Welch (2002). 5 Empirical research has demonstrated that, on average, the issue price is significantly lower than the price paid on the first day of public trading in the given shares. Depending on the specific country and period under observation, the first publicly traded price lies between 10 and 20 % above the issue price, and during shorter market phases the reading can even increase to 50 %. For the USA, see Ritter / Welch (2002), and for Germany, Kaserer / Kempf (1995) or Löffler et al. (2005). 6 Lowry / Schwert (2002) for example demonstrate that IPO activity and underpricing both follow cyclical patterns. 7 An initial application of such a positive feedback loop as a capital market phenomenon can be found in de Long et al. (1990). 8 See in this regard Oehler et al. (2004), Lowry / Schwert (2002) and Lerner (1994), who demonstrate that the degree of underpricing increases the probability of an IPO, although with a certain time lag.

3 126 Annex Kapiteltitel ponent in the calculation of the IPO Sentiment Indicator. It is mathematically defined as follows: Whereby: K = standardisation constant (100) P0iq = closing price on the first trading day of the IPO (i) in calendar quarter (q) OPiq = mean bookbuilding range or, as the case may be, the issue price of a fixed-price deal for the IPO (i) in calendar quarter (q) q = moving-average index for the given quarter, whereas q = 1 previous quarter, q = 2 next to last quarter, etc. γiq = weighting factor for the IPO (i) in calendar quarter (q), whereas this is immediately set at 1 if P0iq > = OPiq and otherwise at 1.5 Nq = number of IPOs in a given calendar quarter (q) κ = temporal weighting parameter The make-up of the Underpricing Sentiment can therefore be described as follows 9 : First, the level of underpricing for each new issue is calculated. However, if bookbuilding was involved, then contrary to the normal definition of underpricing, it is not the relative variance to the initial publicly traded price that is applied, but instead the relative variance between the initially paid price and the mean bookbuilding range. The reason for this is to be found in the fact that the bookbuilding range is communicated in the capital market for a considerably longer period of time than the final fixed issue price. It is therefore implicit that investors compare the initial price paid in the marketplace with the bookbuilding range, which normally centres on the middle of the pricing corridor. This numerical value for the level of underpricing is then multiplied by two factors. The first is the γiq factor, which ensures that the 1.5x weighting of a negative underpricing is consistent with a positive underpricing in arriving at the mean value. The theoretic rationale for this unequal treatment of negative and positive underpricing is attributable to loss aversion, which is a central element of the Kahneman/ Tversky Prospect Theory (1979). According to that theory, investors intuitively attach considerably greater significance to the negative consequences that could result from their investment decisions than they do to the positive consequences. In addition, the availability bias is based on the assumption that more easily accessible information regardless of its actual merit plays a greater role in the investment decision than less easily accessible information. For that reason, a second, time-related weighting factor q-κ is applied so that IPOs launched further in the past are given less weight in calculating the mean value. In total, all IPOs brought to market in the preceding eight quarters are included in the calculation. Ultimately, the first factor in the formula ensures that the weightings add up to 1, and the perceived underpricing is expressed as a percentage. The addition of the standardisation constant of 100 at the end

4 SWX Group Annual Report Annex Developments in the IPO Sentiment Indicator, IPO Climate and Underpricing Sentiment since March 2005 IPO Sentiment Indicator IPO Climate Underpricing Sentiment March June September December March June Source: Deutsche Börse AG of the formula is then introduced so the USI can be interpreted as a percentage-based underpricing factor. For example, a reading of 125 means that the perceived underpricing amounts to 25 %. IPO Climate Apart from the perceived underpricing element, the second component of the ISI is derived from a survey of three groups the syndicate banks, active investors (institutional as well as large private investors) and issuers (in their stead, venture capital and private equity firms are interviewed because as a general rule their portfolios contain a number of IPO candidates). In this process, two of the five questions relate to the attractiveness of the IPO market, two to an assessment of the current valuation levels in the capital market, and a final question to the future development of IPO activity. Questions 1 and 2 are then evaluated en bloc to compare the assessment of market attractiveness (question 1) with the actual dealings of the interviewees (question 2). The greater the consistency between the two answers, the greater the contextual reliability of the responses. 10 All questions are based on a five-level scale; thus overall the IPO Climate can range between 12 and 60. The IPO Sentiment Indicator In calculating the ISI, the Underpricing Sentiment (USI) and IPO Climate are linked with each other multiplicatively: In so doing, it is always ensured that a positive (negative) change in either of the two subcomponents leads to a positive (negative) change in the overall reading. In the case of counter-directional changes, the result is undetermined, whereas the multiplicative link in fact establishes a substitutive relationship that is dependent on the level of the given subcomponent, which ultimately cannot be substantiated. Predictive capability and function of the IPO Sentiment Indicator The table on top reflects the way the IPO Sentiment Indic a tor has developed on the Frankfurt Exchange since March For example, the March 2005 value of the ISI is derived from the multiplicative link between the IPO Climate and the Underpricing Sentiment readings (32.48 x / 100 = 32.67). In Germany, an increase in IPO activity was in fact observed during the second half of 2005, as correctly predicted by the marked increase in the ISI. Following the decline in IPO activity in the first quarter of 2006, which was signalled already in late 2005 by a 9 For a detailed presentation of the theoretic foundation and method of calculating the Underpricing Sentiment, see Kaserer (2006), whose article serves as the primary source for this treatise. 10 For reasons of space, a detailed catalogue of questions cannot be addressed in this treatise. However, further enquiries regarding the ISI can be directed at any time to CEFS (

5 128 Annex Ann-Kristin Achleitner, who studied, graduated and was promoted to professor at the University of St Gallen, has held the KfW Foundation chair in entrepreneurial finance since 2001 and is also science director of the Center for Entrepreneurial and Financial Studies (CEFS) at the Munich University of Technology. Among other things, she is presently a member of the commission of experts Future Bavaria 2020 Megatrends and Their Better Exploitation for Economy and Science, Prosperity and Jobs in Bavaria as well as the middle-class advisory committee of the Federal Ministry of Economics and Labour, and is also deputy chair of the specialist advisory committee of the Federal Office for Financial Services Supervision. retreat in the ISI, the second quarter of 2006 witnessed again as correctly forecast by a rising ISI in Q a total of 25 new issues with an overall value in excess of EUR 3 billion: a clear increase in the number of IPOs. Despite the relatively short time frame in which the ISI has been applied, it would appear that its predictive capability due also to the application of real market data via the underpricing component is indeed reliable. 11 Even though an initial public offering generally requires a longer-term preparatory phase of at least one year, a reliable assessment of the future market climate by means of a sentiment indicator facilitates the short-term adjustment of issuance volume, pricing ranges and timing. As a result, an improved method for measuring conditions in the capital markets through allowance for the perceived degree of underpricing contributes to more certainty for all parties involved, namely the syndicate banks, issuers and investors. Hence the ISI in addition to many other elements can represent a small building block in enhancing the attractiveness of the public capital market for the financing of medium-sized companies, traditionally the backbone of the German as well as the Swiss economy. 11 For a more precise analysis of the predictive capability of the ISI by means of a statistical approach, see Kaserer (2006). References: De Long, J. Bradford / Shleifer, Andrej / Summers, Lawrence H. / Waldmann, Robert J. (1990): Noise Trader Risk in Financial Markets, in: Journal of Political Economy, 98, p Kahneman, Daniel / Tversky, Amos (1979): Prospect Theory: An Analysis of Decision under Risk, in: Econometrica, 47, p Kaserer, Christoph / Kempf, Volker (1 995): Das Underpricing-Phänomen am deutschen Kapitalmarkt und seine Ursachen Eine empirische Analyse für den Zeitraum , in: Zeitschrift für Bankrecht und Bankwirtschaft, 7, p Kaserer, Christoph (2006): Der IPO-Sentiment-Indikator Eine Innovation für den Primärmarkt der Deutschen Börse AG, unpublished CEFS Working Paper 2006 Lerner, Josh (1994): Venture Capitalists and the Decision to Go Public, in: Journal of Financial Economics, 35, p Löffler, Gunter / Panther, Patrick F. / Theissen, Erik (2005): Who Knows What When? The Information Content of Pre-IPO Market Prices, in: Journal of Financial Intermediation, 14, p Lowry, Michelle / Schwert, G. William (2002): IPO Market Cycles: Bubbles or Sequential Learning, in: Journal of Finance, 47, p Oehler, Andreas / Rummer, Marco / Smith, Peter N. (2004): IPO Pricing and the Relative Importance of Investor Sentiment Evidence from Germany, unpublished manuscript at the Otto-Friedrich University Bamberg, 2004 Ritter, Jay R. / Welch, Ivo (2002): A Review of IPO Activity, Pricing, and Allocations, in: Journal of Finance, 57, p Stracca, Livio (2004): Behavioral Finance and Asset Prices: Where Do We Stand?, in: Journal of Economic Psychology, 25, p Van der Sar, Nico L. (2004): Behavioral Finance: How Matters Stand, in: Journal of Economic Psychology, 25, p