Influence of investments in assets on IPO valuation
Master Thesis
Finalized on January 25, 2013
Bertjan van den Berg
University of Groningen: S1681834 Uppsala University: 890823-P130 Email: bertjanvdberg@gmail.com
Supervisor: Dr. J.H. von Eije Assessor: Dr. W. Westerman
MSc International Financial Management Faculty of Economics and Business
1 ABSTRACT
In this paper I investigate the differences in initial public offering (IPO) valuation between countries. By using a sample of social media IPOs, I try to find the differences in valuing investments in assets on the IPO value between professional underwriters and investors from the US compared to their foreign counterparts. Since there are no differences in experience in the industry it is expected that underwriters value investments the same because they act rationally, while investors might assign different values to investments because of their national culture. This results in the hypothesis that underwriters in both the US and elsewhere will evaluate investments in assets by the IPO firms similarly. This hypothesis cannot be rejected. I also assume that the long-term orientation outside of the US will make investors outside of the US more interested in investments in assets by the IPO firms. This hypothesis is rejected. No significant relationship has been found in differences between the valuations of investments in assets prior to an IPO.
2 1. Introduction
The initial public offering (IPO) process of a firm takes place when a firm wants to raise money and sells its equity at the stock market for the first time. In previous literature on the valuation of IPOs several anomalies are present. The most discussed anomaly is the phenomenon of underpricing. Underpricing occurs when the first-day return of a stock is positive which means that the offer price is set too low since investors already value the stock higher after one day of trading.
Since there is a large focus on IPO underpricing this involves that the offer price set by the underwriter might be incorrect. On average, investors believe that the stock has a higher value for which it eventually goes public. This involves that firms are leaving money on the table since they could have set a higher offer price and receive more money doing so. Possible explanations for it are information asymmetry (Rock, 1986), investor compensation (Benveniste and Spindt, 1989), the signaling theory (Welch, 1989), prospect theory (Loughran and Ritter, 2002) and the marketing role of IPOs (Demers and Lewellen, 2002). The focus of previous studies on IPO valuation and IPO underpricing is on different samples of public offerings from different countries. However, the issue of the difference in valuations between actors in different geographical regions has never been discussed.
3 To see whether professional underwriters and investors value companies differently than their foreign counterparts I use a sample of IPOs from the social media industry. The social media industry is a global industry with companies dispersed all over the world and with operations which crosses borders of countries. Furthermore, it is a relatively young industry in which the know-how and expertise of
professional underwriters and investors in several countries will not be far greater than those in others.
Since IPOs are mainly performed by relatively young firms it is interesting to study to what extent investments in assets of the firm prior to the IPO will be valued. Since these firms are in their startup phase, investments in assets are very important for the firm’s future. Furthermore, some of these young firms are hardly making profits and are therefore not profitable in the short run when they go public. By using investments in assets in their analysis, investors can see whether the firm is aiming at the long haul or not. Since the sample contains social media IPOs divided between IPOs from firms from the US, Asia and several other countries I will compare the differences in valuation between IPOs from the US and outside of the US. Therefore, my research question is:
Are investments in assets prior to an IPO differently valued in the US than outside of the US?
To determine the differences between underwriters and investors I will use different dependent variables for firm value. For the value determined by underwriters I will use the offer price of the IPOs since these are set by the underwriters before the shares go public. Investors have hardly any influence on this price. For the valuation by investors I will use the first-day closing price and the first-day return to determine if investors differ from their foreign counterparts.
The paper is organized as follows. In the next section I will describe previous literature on IPOs and the valuation of IPOs. Section 3 is the methodology part which contains the theoretical framework including the hypotheses followed by an explanation of the variables used. The next section contains the
4 section containing the robustness tests. The seventh section gives the conclusion followed by
suggestions for further research.
2. Literature Review
In this section I will review previous literature on various topics. First, I will explain what an IPO and IPO underpricing are. Moreover, I will elaborate on the valuation of IPOs and the differences in valuation.
2.1 Initial Public Offerings
For a company an IPO means that it will leave the startup phase and that it transforms to a larger corporation that is valued by the public. The going public process is a way for the company to attract new funds. Although the offer of new equity is last in the pecking-order theory (Myers, 1984), since debt is preferred over equity and firms are even more inclined to do investments with internal rather than external funds, it is a perfect opportunity for a company to expand its equity base.
When a company decides to perform an IPO it searches an investment banker or a combination of multiple investment bankers to underwrite it. This is followed by creating a prospectus which is a document that the firm uses to communicate with outside investors. This prospectus contains the terms of the offer and the financial and non-financial information about the company. The time between the filing of the prospectus and the final offer date is the so-called waiting period. This time is used by the underwriters to see if there is interest in the stock (Bartov et al., 2002). When there is a high demand, the offer price will be set higher and vice versa.
2.2 Underpricing
5 Underpricing is a phenomenon that the first-day return of a stock is positive which means that the offer price to the market is set too low since investors already value the stock higher after one day of trading (Ritter, 1991). Moreover, it is one of the anomalies in the valuation of IPOs of young firms (Ibbotson and Ritter, 1995).
Underpricing is explained in the literature in two different ways. The first one is on the supply side by showing that underpricing is used to compensate outside investors for information asymmetry (Rock, 1986; Benveniste and Spindt, 1989) or to signal the quality of an IPO firm (Welch, 1989). In the other way it is explained on the demand side stating that underpricing is caused by overly optimistic investors. This was assumed to be the case when Google went public (Berg et al., 2009).
According to Loughran and Ritter (2004), underpricing and its underlying reasons changed over time. In their paper, Why Has IPO Underpricing Changed Over Time?, they try to find possible reasons for the change in underpricing. According to them underpricing changed because of changing risk composition, a realignment of incentives and a changing issuer objective function.
6 believe it is an interesting investment opportunity which might explain the problem of underpricing from the supply side since the underwriter’s prestige gives a quality signal. These findings are strengthened by Ellul and Pagano (2006), who find information asymmetry and risk to be the main drivers for IPO
underpricing. Investors demand a higher return for more risk that their stocks cannot be sold to the secondary market.
However, this last reason is not the same for the cooperation between an internet firm and an investment bank. According to Jain et al. (2007) the prestige of an investment should lead to a higher valuation in theory due to the signaling theory (Welch, 1989). However, the results in their research appeared to be negative which means that an internet firm that cooperates with a high prestige
investment bank has a lower IPO value than other firms. Therefore, there is a debate whether the supply side story on IPO underpricing is really true.
2.3 Valuation of IPOs
IPOs are performed by mainly young firms that experience high growth and it will leave the startup phase to transform to a larger corporation that is valued by the public (Loughran and Ritter, 2004). The valuation of IPOs of young growth companies is influenced by two other anomalies according to Ibbotson and Ritter (1995). The first is that there are cycles in both the volume of new stock issues and the size of first-day returns. The last anomaly is the long-run underperformance of young growth firms. These anomalies are much stronger for younger and smaller firms than their older and larger counterparts. They state that young firms go public when there is periodic over optimism which creates an opportunity for them to raise capital. However, long-term investors receive disappointing returns since the shares cannot live up to the initial expectations held.
7 for which it is difficult to make forecasts on their future cash flows. Therefore, the discounted cash flow (DCF) analysis is very imprecise and another method for valuation is needed. This was found in using the relative valuation approach, in which firm specific financial and non-financial variables are used. With this method IPOs could be compared to offerings in the same industry, which meant a better way for explaining differences in prices, underpricing and the valuation of companies (Aggarwal et al., 2005).
The first one to use the relative valuation approach was Klein (1996). She investigated the relation between IPO price and a set of variables for a sample of 193 IPOs from 1980-1991. The restriction was that they all should have positive income prior to the year they went public. She found that the price per share is positively related to earnings and the book value of equity.
Kim and Ritter (1999) adopted this approach as well. The sample consisted of 190 US IPOs from 1992 and 1993 with positive earnings per share (EPS) and positive book value per share (BPS). Variables used to explain the IPO price in their research were EPS, BPS, sales for the last 12 months and the price-to-earnings (P/E) multiple. They found that when historical accounting information was used, the enterprise value-to-sales ratio works reasonably well for both young and old firms taking leverage effects into account. However, the rest of the historical information did not explain a lot of the variation in the IPO price.
Partially in contradiction to the conclusions of Kim and Ritter (1999), Beatty et al. (2000) found that the “accounting book value, earnings and revenue, in conjunction with several other firm and market characteristics, explain a large portion of IPO offer prices” With an R2 of 80% they found that historical accounting information explains the variation in the filing price, offer price and the first-day return. This was also the case for the difference between the offer and first-day return price. Elaborating on the issue of underpricing, Purnanandam and Swaminathan (2004) used the relative valuation approach to
8 to find whether there is a relationship. They used several price multiples, such as price-to-earnings before interest, taxes, depreciation and amortization (EBITDA), price-to-sales and price-to-earnings, and found significant results for IPO underpricing.
Aggarwal et al. (2009) finally included IPOs of firms with negative earnings in their valuation of IPOs. They found that these were valued different than IPOs of firms with positive earnings. Furthermore, a shift in the valuation of fundamentals was documented between the periods 1986-1990 and 1997-2001.
Shu et al. (2012) tried to explain the offer price and the first-day price by using the relative valuation approach too. They used earnings management and managerial optimism to show that investors value these characteristics in the offer and first-day price. Since these variables are about managers there could be a large bias in measuring the differences in earnings management and managerial optimism since it is measured differently for each manager. The measurements are subjective and not precise. Therefore, it is hard for investors outside the firm to evaluate these practices, while Shu et al. (2012) were not able to find a relationship between these two assumptions.
2.4 Valuation of internet firms
The going public process of social media firms is a recent phenomenon. Therefore, there is hardly any research on the valuation of these firms. Since it is a special kind of Internet firm, I expect that the valuation process of social media firms can be linked to the valuation of Internet firms. Therefore, it is important to know what is written on the topic of internet firms since this might also be relevant for social media firms.
There are at least two distinct problems in valuing Internet firms according to Trueman et al. (2000). Firstly, the industry and the firms in it are so young that there is hardly any historical financial
9 as a prediction of the future. This is in line with the underpricing anomaly found by Ibbotson and Ritter (1995). Since investors are not able to value the IPOs correctly they pay within one day of trading more for the shares than the offer price.
Trueman et al. (2000) find that gross profits are the only financial influence on a firm’s market value after breaking down the net income variable. Unique visitors and page views are the non-financial influences on the stock price. From these results they conclude that investors value Internet stocks rationally in the sense that they value them relative to other Internet stocks.
Schwartz and Moon (2000) were the first that tried to develop a model to value internet stocks. This model was used to value Internet IPOs in which they applied real-options theory and capital-budgeting techniques. They demonstrated that uncertainty about key variables plays a distinct role in valuing high growth Internet companies. However, their initial model was not very practical to use and was tested with only one firm. Therefore, they improved their model to make it more practical (Schwartz and Moon, 2001). The stochastic variables used in the revised model are revenues, the growth rates in revenues and the variable costs. The path-dependent variables are the amount of cash available, the
loss-carry-forward and the accumulated property, plant and equipment.
10 Berg et al. (2009) evaluated the IPO of Google Inc. to see if it was valued correctly. In the initial
prospectus Google stated that the offer price reflects the market valuation estimated by the underwriters and that it was set to avoid “boom-bust cycles”. When the stock went public it encountered underpricing because of the excess demand for the stock. However, information
asymmetry was not present since insiders made all information public. This might be the result of the marketing role of an IPO described by Demers and Lewellen (2002), which also confirms underpricing from the demand side.
3. Methodology
According to previous literature on IPO valuation there is a large focus on different financial and non-financial variables that influence the price of public offerings. However, no focus has been on the
different valuations by different actors and their counterparts from other regions. One might expect that people from different countries have different procedures in valuing equity. Since practices in valuing equity are a result from education on this subject in a specific way. Since education is not the same all over the world, one would expect that there are differences in valuing practices as well.
11 did not include a rate for investments other than R&D while there are many other departments of a firm in which could be invested. Moreover, investments in assets in the firm will signal investors that the firm has a long-term view and wants to grow substantially. Since most companies that go public are young firms which have an uncertain future, the investments in assets will show the new shareholders that the firm’s managers are long-term oriented.
I will investigate this for three different types of valuation. The first will be on the offer price set by professional underwriters. The second will be the difference between investors from the US and
investors from outside the US. These will be measured using two different measures. The first measure is the first-day closing price which reflects the market value according to investors. The second measure will be the first-day return value which is the difference between the value set by professional
underwriters and investors in the firm. I want to find if there are differences between the actors from different countries which in this case is the difference between the US and mainly Asia.
3.1 Professionals
IPOs are underwritten by professional investment bankers that have experience in the field of equity offering. They guide the managers of the firm to sell shares on the stock exchange for the first time. During this process they screen the firm and create a prospectus in which financial and non-financial information is reported. This prospectus has the purpose to inform investors so they can use it to value the firm’s shares. In this prospectus they communicate how they came up with the offer value and their valuation of the firm. This is reflected in the offer range and eventually in the offer price for which the firm goes public.
12 social media industry is relatively young. The firms are thus also relatively young when they go public. Therefore, underwriters from one country do not have more experience than from another one, because the industry and the firms within it have not been around for that long. Moreover, the number of firms active in the social media industry that went public is not large. Professionals should act in an ethical and professional way when underwriting an IPO and are assumed here not to differ between nationalities. Therefore, the first hypothesis is:
Hypothesis 1
Investments in assets by social media firms prior to an IPO will be valued the same by professionals in the US as outside of the US
3.2 Investors
The investors that invest in stocks on the day that they go public differ from professionals in several ways. Investors see a good investment opportunity and may value the stock at a higher price than for which it goes public. They base their analysis on experience in investing in other stocks, information on the industry and the information reflected in the prospectus created by the underwriters. After
consulting these sources investors decide if it is worth it to invest in the IPO or not. One of the anomalies in the going public process is underpricing, which means that on average a stock that goes public is sold at a higher price after one day of trading. This means that investors give equity a relatively higher value than the price set by the professional underwriters.
13 Cultural differences will distinguish firms and investors from one country to another. The assumption and measurements of a national culture was made by Hofstede (1980). He defines culture as the
collective programming of the human mind that differentiates people belonging to one group from those of another. Basically, culture is a combination of collectively held values. This definition applies to the culture of a nation. In this way investing may also be different for people from other countries. Hofstede distinguished five dimensions of national culture, respectively Power Distance; Masculinity;
Individualism; Uncertainty Avoidance; and Confucianism which is also known as Long-term Orientation. These dimensions influence the collectively held values and define a country’s culture. A country has a score from zero to hundred on these dimensions with zero being the lowest and one hundred the highest.
Since the sample of social media firms consist of mainly US and Asian IPOs (Figure 1) it is interesting to see if these cultural differences influence investors valuing investments in assets prior to an IPO. According to Barkema and Vermeulen (1997), uncertainty avoidance and long-term orientation are important economic dimensions than the other three. Devine et al. (2000) specified Hofstede’s dimensions in more detail. According to them, uncertainty avoidance shows the level of a nation´s culture feeling uncomfortable with an uncertain future. When a country avoids uncertainty it will stay with their old trusted habits. Secondly, the long-term orientation also known as the Confucian dynamism depends on the extent by which society is looking into the future or has a short-term view.
According to these dimensions we can see a difference between the US and the rest of the world. In the figure below the numbers for uncertainty avoidance and long-term orientation are represented for the countries in which social media firms went public.
14 determine whether the firm has a long-term orientation. The score for long-term orientation is not available for Russia. The US has the lowest score on the LTO dimension of 29 while the two other large countries in our sample, China and Japan, have values of 118 and 80 respectively. The initial scale had a maximum of 100, but China surpassed this since it has an even longer orientation than the initially designed scales by Hofstede.
Figure 1 Scores for Hofstede’s cultural dimensions1 and the number of IPOs per country
The scores from figure 1 tell that investors outside of the US are more long-term orientated than their US counterparts. Because of this and because investments in assets are a variable that gives return in the long term, I assume that investors outside the US value investments in assets higher than investors in the US. Therefore, the second hypothesis is:
1 Scores obtained from http://www.geert-hofstede.com/countries.html
15
Hypothesis 2
Investments in assets by social media firms prior to an IPO will be valued higher by investors outside of the US than in the US
This can be measured using the first-day closing value as well as the underpricing value, since these values are both influenced by the investors after one day of trading. The difference between the two measures is that the first is the absolute value of the first-day closing value while the underpricing value is the difference between the offer value and first-day closing value.
3.3 Dependent and independent variables
In this paper I examine the differences in valuation by different nationalities. The dependent variables are the values of the firm set by different kind of actors. The first one is the value set by the
professionals. The underwriters influence the offer price, so the relative professional value of a firm is calculated as follows:
Equation 1
16 be issued from the overallotment option. The overallotment or greenshoe option gives underwriters the option to short-sell shares for the offer price on a designated date in the future. However, they will only be exercised if the price is higher at the exercise date which lies ahead of the IPO date. Therefore, this is not taken into account by the initial investors in the firm when they value it since these shares are sold later in the future at a lower price.
Secondly, I examine how investors from different countries value investments in assets in social media firms differently. Therefore, I use the market value as a dependent variable which is calculated as follows:
Equation 2
The market value of the firm will be calculated using the first-day closing price times the number of outstanding shares. I choose to use the first-day closing price since it is in my opinion the best proxy for the market value of an IPO. When the stock is publicly sold, the market will set its price by supply and demand and therefore reflects the true market value. Moreover, I want to see if there might be a difference in underpricing between nationalities. This should namely also result if underwriters' valuations do not differ, but investor valuations do differ. Therefore, the third dependent variable is calculated as follows:
Equation 3
17 Investments in assets
By using the assets growth of a firm, investors will see whether the firm has a long-term orientation. To measure investments in assets growth in the firm I will use the change in assets in the year prior to the IPO. This is calculated as follows:
Equation 4
I expect the investments in assets to be of a greater difference in the valuation by investors since they are less educated and do not act in the same rational way as professional underwriters. By taking the relative change in assets prior to the going public process I can see whether a firm is more long-term or short-term focused. By taking the relative change I will follow the relative valuation approach.
3.4 Control Variables
Earnings
The net income of the firm shows the profitability of the firm. In most investment opportunities valuation techniques like DCF analysis are used. In these techniques the free cash flows are discounted to come up with a net present value. In valuing IPOs this method is very imprecise and combined with the fact that the sample consists of firms active in the social media industry which is a relatively young industry, another approach is needed. This was found by using the relative valuation approach, in which firm specific variables are used.
18 positive net earnings (Klein, 1996; Kim and Ritter, 1999). By using gross income, only two firms end up with a negative value for income. The value for gross income will be the last number available before the firm went public. This is because investors make their decisions on information in the prospectus which is available prior to the IPO. With respect to the relative valuation approach the gross income will be calculated as follows:
Equation 5
Debt ratio
To control for variation in the market value I will include the debt ratio of the firm prior to the IPO. This is because young firms have a relatively high leverage ratio since they are growing firms in their start-up phase. The debt ratio will be calculated as follows:
Equation 6
Including all these variables the regression model will give the following equation:
Equation 7
19
Equation 8
The relative firm value will be the firm values calculated by the different actors specified before. These are the relative professional value (RPV) done by professional underwriters, the relative marketing value (RMV) done by investors and the relative underpricing value (RUV) which is the difference between the valuation done by the underwriters and investors.
4. Data
The sample used in this paper is from firms that are listed in the Solactive Social Media Total Return Index provided by the company Structured Solutions. This includes firms that were formerly represented in the index, but are not included nowadays since they are not publicly traded or not operating anymore. The firms represented in this index “are active in the social media industry, including companies that provide social networking, file sharing, and other web-based media applications.” These firms can be found in appendix A. The firms come from China, Germany, India, Japan, The Netherlands, Russia, Taiwan and the United States of America.
20 has no financial numbers prior to the IPO. This will leave me with 27 social media IPOs of which the data are collected.
The financial data are collected from the Thomson One Banker financial database. In this database the financial variables are derived from the firms’ prospectuses and annual reports. When several of these numbers were not available in this database, prospectuses of the IPOs were manually collected and consulted. When the first-day returns were not available in the financial database I used Datastream to retrieve historical prices and the first-day closing price. All financial numbers are in thousands of dollars including the offer price, first-day closing price and the underpricing value. This is done to make a good relative comparison since the firms come from multiple countries with different currencies. In appendix B the descriptive statistics are reported.
In the table the range of the different firm values is presented. There is a large range for all three dependent variables. We can see that on average there is underpricing, since the mean of the relative underpricing value is 0.3580 which is in thousands of dollars. From the 27 IPOs, only 4 reported a negative first-day return. Netease Inc., a Chinese firm, had the largest negative return which skewed the average rather negatively.
21 Investments in assets tend to be very large in the sample as a whole. Only one firm has a negative value (Demand Media Inc.) which means it reports disinvestments. The two firms reporting the highest change in assets prior to their IPOs are US firms Facebook and Meet Me/Quepasa, which means they were high growth firms before they went public. This might be a problem since Sky-Mobi Ltd. is the only firm that has a higher debt ratio than 1 (2.07). This means that it has a negative value for equity and is highly leveraged. This is something which should be taken into account, since it might scare investors and is therefore not a good representative of a social media IPO. This will be corrected for in another robustness test.
5. Analysis
22
Table 1 Regressions of the total sample for professionals, investors and the difference between them from 1999 – 2012
The dependent variables are RPV, RMV and RUV. RPV is the relative professional value of the firm calculated by underwriters. RMV is the relative market value of the firm calculated by the investors after one day of trading. RUV is the relative underpricing value which is the difference between the valuation by investors and underwriters. The independent variables are RIA, RGI and TDR. RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. The latter two variables are used as control variables. RGI is the relative gross income in the year prior to an IPO. TDR is the total debt ratio. Constant is the intercept of the regression equation. The
Adjusted R2 is the adjusted R-square for the regression. F-statistic is the overall F-statistic of the regression. Prob.
indicates the significance of the regression. Observ. is the number of observations used in the regressions.
RPV RMV RUV RIA -0.014 -0.011 0.003 (0.665) (0.665) (0.735) RGI -10.779 -7.068 3.711 (0.627) (0.693) (0.502) TDR -33.108 -23.384 9.724 (0.415) (0.475) (0.337) Constant 46.204 39.131 -7.072 (0.107) (0.092) (0.313) Adjusted R2 -0.075 -0.088 -0.05 F-statistic 0.393 0.297 0.589 Prob. (0.759) (0.827) (0.628) Observ. 27 27 27
23
Table 2 Regressions of the total sample including slope dummies for professionals, investors and the difference between them from 1999 – 2012
The dependent variables are RPV, RMV and RUV. RPV is the relative professional value of the firm calculated by underwriters. RMV is the relative market value of the firm calculated by the investors after one day of trading. RUV is the relative underpricing value which is the difference between the valuation by investors and underwriters. The independent variables are RIA, the control variables RGI and TDR, USD and interaction terms. RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. RGI is the relative gross income in the year prior to an IPO. TDR is the total debt ratio. USD is a dummy variable which has the value 1 for IPOs performed by firms in the US. USD*RIA, USD*RGI and USD*TDR are interaction terms between the US dummy and
other independent variables. Constant is the intercept of the regression equation. The Adjusted R2 is the adjusted
R-square for the regression. F-statistic is the overall F-statistic of the regression. Prob. indicates the significance of the regression. Observ. is the number of observations used in the regressions.
RPV RMV RUV RIA 6.803 5.202 -1.601 (0.000) (0.000) (0.000) RGI 3.501 3.650 0.149 (0.499) (0.614) (0.959) TDR -0.731 1.198 1.929 (0.938) (0.927) (0.711) USD 6.601 -0.206 -6.807 (0.681) (0.993) (0.450) USD*RIA -6.800 -5.199 1.601 (0.000) (0.000) (0.000) USD*RGI -2.514 -3.431 -0.918 (0.843) (0.848) (0.897) USD*TDR 1.648 -1.470 -3.118 (0.946) (0.965) (0.817) Constant -5.901 2.170 8.071 (0.435) (0.837) (0.066) Adjusted R2 0.954 0.858 0.773 F-statistic 77.439 23.478 13.627 Prob. (0.000) (0.000) (0.000) Observ. 27 27 27
24 way to value equity. However, according to this sample there is a difference between US professionals and their foreign counterparts. The US dummy is insignificant. Since it is insignificant there is not a significant difference between IPOs performed in the US and outside of the US. Investments in assets for non-US countries have a positive influence to the professional value of a firm. It is significant at the 1% level. However, the variable for investments in assets multiplied by the slope dummy variable tends to be negative at the 1% significance level which means that US professionals value investments in assets negatively. This leads to rejection of the first hypothesis, since it was expected that investors all over the world act rationally and would value equity in the same way. The other variables are insignificant.
For investors almost the same results follow from the valuation done by professionals. This means that investors outside of the US value investments in assets higher than their US counterparts. This is significant at the 1% level which gives support for the second hypothesis which is that investors outside the US value investments in assets higher than their US counterparts. This leads to the acceptance of hypothesis 2. The control variables RGI and TDR are insignificant.
Since professionals in the US do not value investments in assets the same as their foreign counterparts, it means they take something else into account. This could mean that underwriters take the desires of local long-term investors into account when setting the offer price. In this case professionals do not consider the real value to be relevant in setting the offer price, but the perceived local value to perform an IPO.
25 However, there is one remarkable fact on the variables investments in assets and the interaction term with investments in assets. They almost have the same coefficient with the interaction term being negative and the variable for the whole sample having a positive variable. This might be the result of a misspecification of the model and the existence of multicollinearity. Therefore, I will check the
correlations between the independent variables. The results of this can be found in appendix C.
The maximum desired correlation between two independent variables is 0.75. However, the variables investments in assets and the interaction term of investments in assets are highly correlated (0.9998). This is because several firms for the US have very high values for investments in assets, while non-US firms have a relatively low value for investments in assets. Therefore, several observations have to be left out of the regression to make sure that the problem of multicollinearity will disappear. In total, 5 observations have to be left out to get the correlation coefficient below the desired value of 0.75. This is done by setting a limit on the value that investments in assets can have. The limit for this number is 10 which excludes the American firms Angie’s List, Facebook, Groupon, Meet Me Inc. and the non-American firm Netease, and the regression then contains 22 observations. In the table below the three regressions are presented including the slope dummies and excluding the five observations which caused
26
Table 3 Regressions for firms with RIA lower than 10, including slope dummies for professionals, investors and the difference between them from 1999 – 2012
The restriction in this sample is that the RIA is lower than 10 to leave out multicollinearity. The dependent variables are RPV, RMV and RUV. RPV is the relative professional value of the firm calculated by underwriters. RMV is the relative market value of the firm calculated by the investors after one day of trading. RUV is the relative
underpricing value which is the difference between the valuation by investors and underwriters. The independent variables are RIA, the control variables RGI and TDR, USD and interaction terms. RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. RGI is the relative gross income in the year prior to an IPO. TDR is the total debt ratio. USD is a dummy variable which has the value 1 for IPOs performed by firms in the US. USD*RIA, USD*RGI and USD*TDR are interaction terms between the US dummy and other independent
variables. Constant is the intercept of the regression equation. The Adjusted R2 is the adjusted R-square for the
regression. F-statistic is the overall F-statistic of the regression. Prob. indicates the significance of the regression. Observ. is the number of observations used in the regression.
RPV RMV RUV RIA -0.472 -0.919 -0.447 (0.842) (0.819) (0.793) RGI 3.146 3.351 0.205 (0.497) (0.667) (0.950) TDR 3.547 4.798 1.251 (0.675) (0.737) (0.836) USD -8.188 -13.471 -5.282 (0.694) (0.702) (0.724) USD*RIA 0.620 0.991 0.371 (0.870) (0.877) (0.891) USD*RGI -0.484 -0.833 -0.349 (0.971) (0.971) (0.971) USD*TDR 0.081 -1.997 -2.078 (0.998) (0.963) (0.910) Constant 6.446 12.558 6.112 (0.414) (0.348) (0.285) Adjusted R2 -0.285 -0.318 -0.348 F-statistic 0.334 0.276 0.225 Prob. (0.925) (0.953) (0.973) Observ. 22 22 22
27 6. Robustness tests
Next, it might be interesting to do several robustness tests which were described earlier on. Although there are no significant results, it is interesting to see if the firms that influence such a small sample have a large impact one the regression as a whole. One of the issues several researchers had in valuing IPOs was the inclusion of firms with a negative income (Klein, 1996). In this sample 2 firms have a negative gross income. Following this method the 2 negative gross incomes will be excluded. One firm that was excluded for the high rate of investments in assets causing multicollinearity had a negative gross income so that the sample is reduced by only one. After performing the regressions there are again no
differences in significance and none of the coefficients changed from a negative to a positive coefficient or vice versa.
28
Table 4 Regressions for firms with RIA lower than 10 and TDR lower than 1, including the LTO variable for professionals, investors and the difference between them from 1999 – 2012
The restriction in this sample is that the RIA is lower than 10 and TDR lower than 1. The dependent variables are RPV, RMV and RUV. RPV is the relative professional value of the firm calculated by underwriters. RMV is the relative market value of the firm calculated by the investors after one day of trading. RUV is the relative
underpricing value which is the difference between the valuation by investors and underwriters. The independent variables are RIA, the control variables RGI and TDR, LTO and interaction terms. RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. RGI is the control variable relative gross income in the year prior to an IPO. TDR is the total debt ratio which is used as a control variable. LTO is the long-term orientation dimension of national culture, defined by Hofstede (1980). LTO*RIA, LTO*RGI and LTO*TDR are interaction terms between the country specific orientation and other independent variables. Constant is the intercept of the
regression equation. The Adjusted R2 is the adjusted R-square for the regression. F-statistic is the overall F-statistic
of the regression. Prob. indicates the significance of the regression. Observ. is the number of observations used in the regressions. RPV RMV RUV RIA 0.476 1.424 0.948 (0.626) (0.454) (0.378) GIA 1.228 -2.059 -3.287 (0.622) (0.669) (0.237) TDR 1.955 -4.133 -6.088 (0.779) (0.760) (0.428) LTO 893.946 1442.740 548.795 (0.000) (0.000) (0.000) LTO*RIA -0.012 -0.039 -0.027 (0.478) (0.225) (0.135) LTO*GIA 0.011 0.082 0.071 (0.793) (0.323) (0.138) LTO*TDR 0.065 0.257 0.192 (0.573) (0.261) (0.143) Constant -0.950 -0.643 0.307 (0.692) (0.889) (0.906) Adjusted R2 0.909 0.878 0.782 F-statistic 28.247 20.559 10.729 Prob. (0.000) (0.000) (0.000) Observ. 20 20 20
This regression shows a satisfying result for the differences in valuation practices. Because the LTO variable is significant at the 1% level, this means that countries with a higher score on long-term
29 the valuations are not significantly affected by the investments in assets. Therefore, hypothesis 1 cannot be rejected, but hypothesis 2 is rejected.
7. Conclusion
In this paper I tried to find if there is a difference in the valuations of IPOs done in different countries. To make sure there was no advantage in experience and expertise the group of social media IPOs was used to determine whether professional underwriters and investors value investments in assets prior to an IPO differently than their foreign counterparts.
This research consisted of two hypotheses. The first was that investments in assets by social media firms prior to an IPO will be valued the same by professionals in the US as outside of the US. The second was that investments in assets by social media firms prior to an IPO will be valued higher by investors outside of the US than in the US. The US was chosen since a large part of the sample of social media firms consisted of US firms, while firms from the rest of the world have a large concentration in China and Japan.
To distinguish between the valuations different dependent variables where used. For the professional underwriters the offer value was used, while the valuation of investors was determined by the first-day closing value. To determine the difference between US actors and their foreign counterparts, a dummy variable together with interaction terms were included. This resulted in several findings.
30 actors from different countries. However, this was not the case for investments in assets and therefore hypothesis 1 could not be accepted and hypothesis 2 is rejected.
Although both hypotheses are rejected, it is interesting to see that the valuation done by professionals is close to that of the investors. This might be that the definition of valuation done by professionals is another one than specified before. Professionals might value the firm with the valuation of investors in mind which is different from the assumption that professionals are rational and use a fair valuation.
8. Suggestions for further research
In this paper I tried to prove that investments in assets prior to an IPO are differently valued by actors in the US and their foreign counterparts. However, the research is based on a small sample of social media IPOs. A suggestion for further research is to expand this research and test this on a different and
especially larger sample to draw conclusions for further generalization. Moreover, this is the first paper that uses a sample of social media IPOs. It is an interesting industry with young high growth firms which is likely to expand in the future. Therefore, when the industry consists of more firms it is interesting to see if the results will show significant relationships.
31 9. References
Aggarwal, R., Bhagat, S. and Rangan, S., 2009. “The Impact of Fundamentals on IPO Valuation.” Financial Management, 38: 253–284
Barkema, H. and Vermeulen, F., 1997. “What differences in the cultural backgrounds of partners are determinant for international joint ventures.” Journal of International Business Studies, 28: 845-864
Bartov, E., Mohanram, P. and Seethamraju, C., 2002. “Valuation of Internet Stocks-An IPO Perspective.” Journal of Accounting Research, 40: 321-346
Beatty, R. and Ritter, J. R., 1986. “Investment Banking, Reputation, and the Underpricing of Initial Public Offerings.” Journal of Financial Economics, 15: 213-32
Beatty, R., Riffe, S. and Thompson, R., 2000. “IPO pricing with accounting information.” Southern Methodist University, Working Paper
Benveniste, L. and Spindt, P., 1989. “How investment bankers determine the offer price and allocation of new issues.” Journal of Financial Economics, 66: 105-137
Berg, J. E., Neumann, G. R. and Rietz, T. A., 2009. “Searching for Google's Value: Using Prediction Markets to Forecast Market Capitalization Prior to an Initial Public Offering.” Management Science, 55: 348-361
Demers, E. and Lewellen, K., 2003. “The marketing role of IPOs: evidence from internet stocks.” Journal of Financial Economics, 68: 413-437
Devine, K., O’Clock, P. and Rooney, C. P., 2000. “Implications of culture on the development of control systems.” CPA Journal, 59: 37-41
32 Hofstede, G.H., 1980. “Culture and Organization.” International Studies of Management & Organization, 10: 15-41
Ibbotson, R. G. and Ritter, J. R., 1995. “Chapter 30 Initial public offerings, R.A. Jarrow, V. Maksimovic and W.T. Ziemba.” Handbooks in Operations Research and Management Science, 9: 993-1016
Jain, B. A., Jayaraman, N. and Kini, O., 2007. “The path-to-profitability of Internet IPO firms.” Journal of Business Venturing, 23: 165-194
Kim, M. and Ritter, J. R., 1999. “Valuing IPOs.” Journal of Financial Economics, 53: 409–437
Klein, A., 1996. “Can Investors Use the Prospectus to Price Initial Public Offerings?” The Journal of
Financial Statement Analysis, 2: 23–39
Ljunqvist, A. and Wilhelm, W. J., 2003. “IPO Pricing in the Dot-com Bubble.” The Journal of Finance, 58: 723-752
Loughran, T. and Ritter, J. R., 2002. “Why don’t issuers get upset about leaving money on the table in IPOs?” Review of Financial Studies, 15: 413-444
Loughran, T. and Ritter, J. R., 2004. “Why Has IPO Underpricing Changed Over Time?” Financial Management, 33: 5-37
Myers, S. C., 1984. “Capital Structure Puzzle.” Journal of Finance, 39: 575-592
Purnanandam, A. K. and Swaminathan, B., 2004. “Are IPOs Really Underpriced?” The Review of Financial Studies, 17: 811-848
Ritter, J.R., 1991. “The Long-Run Performance of Initial Public Offerings.” The Journal of Finance, 46: 3-27
33 Welch, L., 1989. “Seasoned offerings, imitation costs, and the underpricing of initial public offerings.” The Journal of Finance, 44: 421-449
Schwartz, E. S. and Moon, M., 2000. “Rational Pricing of Internet Companies.” Financial Analyst Journal, 56: 62–75
Schwartz, E. S. and Moon, M., 2001. “Rational Pricing of Internet Companies Revisited.” Financial Review,
6: 7-26
Shu, P. G., Chiang, S. J. and Lin, H. Y., 2012. “Earnings Management, Managerial Optimism, and IPO
Valuation.” Journal of Behavioral Finance, 13: 147-161
34 10. Appendix
A. Sample of social media firms
Company name Country IPO date
ANGIE'S LIST, INC. United States of America 11/16/2011
CHANGYOU.COM LIMITED China 4/2/2009
DEMAND MEDIA INC. United States of America 1/26/2011
DENA CO LTD Japan 2/4/2005
FACEBOOK, INC. United States of America 5/18/2012 GOOGLE INC. United States of America 8/19/2004
GREE INC Japan 12/8/2008
GROUPON, INC. United States of America 11/4/2011 JIVE SOFTWARE INC United States of America 12/12/2011 LINKEDIN CORPORATION United States of America 5/19/2011
MAIL.RU GROUP LIMITED Russia 11/5/2010
MEETME, INC. / QUEPASA United States of America 6/24/1999
MIXI INC. Japan 9/15/2006
NETEASE, INC. China 6/30/2000
NEXON CO LTD Japan 12/5/2011
NUTRISYSTEM, INC. United States of America 6/13/1981 PANDORA MEDIA, INC. United States of America 6/14/2011
PCHOME ONLINE INC. Taiwan 1/24/2005
REDIFF.COM INDIA LIMITED India 6/14/2000
RENREN INC. China 5/3/2011
SINA CORPORATION China 4/12/2000
SKY-MOBI LTD. China 12/9/2010
TENCENT HOLDINGS LIMITED China 6/16/2004
UNITED ONLINE INC United States of America 9/23/1999
XING AG Germany 12/7/2006
YANDEX N.V. Netherlands 5/24/2011
YELP INC. United States of America 3/2/2012
YOUKU.COM INC. China 12/8/2010
35 B. Descriptive Statistics of the sample (all values in 1000s of US Dollars)
The dependent variables are RPV, RMV and RUV. RPV is the relative professional value of the firm calculated by underwriters. RMV is the relative market value of the firm calculated by the investors after one day of trading. RUV is the relative underpricing value which is the difference between the valuation by investors and underwriters. The independent variables are RIA, RGI, RNI and TDR. RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. The latter two variables are used as control variables. RGI is the relative gross income in the year prior to an IPO. RNI is the relative net income in the year prior to an IPO TDR is the total debt ratio. Mean is the mean of the variables. Median is the middle observation of the variables. Max is the maximum value of the variable. Min is the minimum value of the variable. Std. Dev. is the standard variation of the variable. Observ. is the number of observations of the variables.
RPV RMV RUV RIA RGI RNI TDR
Mean 21.392 21.750 0.358 150.780 0.636 -0.114 0.480 Median 1.644 1.962 0.242 1.962 0.379 -0.011 0.407 Max 401.798 314.44 31.350 2116.391 2.941 0.980 2.068 Min 0.191 0.220 -87.358 -0.091 -0.149 -1.412 0.053 Std. Dev. 76.940 61.735 19.385 521.207 0.742 0.496 0.410 Observ. 27 27 27 27 27 27 27
C. Correlation matrix of the independent variables including US dummy
RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. RGI is the relative gross income in the year prior to an IPO. TDR is the total debt ratio. USD is a dummy variable which has the value 1 for IPOs performed by firms in the US. USD*RIA, USD*RGI and USD*TDR are interaction terms between the US dummy and other independent variables.
RIA GIA TDR USD USD*RIA USD*GIA USD*TDR
36 D. Correlation matrix of the independent variables including LTO
RIA is relative investments in assets which is the relative asset growth in the year prior to an IPO. RGI is the relative gross income in the year prior to an IPO. TDR is the total debt ratio. LTO is the long-term orientation dimension of national culture, defined by Hofstede (1980). LTO*RIA, LTO*RGI and LTO*TDR are interaction terms between the country specific orientation and other independent variables.
RIA GIA TDR LTO LTO*RIA LTO*GIA LTO*TDR
