Master Thesis
Is there a systematic competitive advantage for platform firms? A quantitative analysis based on investment analyst evaluations.
By Rob Lode
S3273962
Msc Business Administration - track: Change Management
Word count: 6.255 Semester 1 2019-2020
Supervisor: dr. J.D.R. Oehmichen Co-assessor: dr. J. Surroca
University of Groningen Faculty of Economics and Business
2
Abstract: With the emergence of digital platform firms, firm environment is changing. However, yet unclear is of this leads to a systematic competitive advantage for platform firms. A way to investigate such a situation is by looking at analyst evaluations in more detail. In this study I analyze the relation between analyst recommendations on platform firms and the moderating effect of size (market
capitalization). I suggest that platform firms get higher recommendations compared with traditional firms because they have more intangible assets in the form of network effects. Often platform firms operate in winner-take-all markets. Therefore I propose that size will have a positive effect on the relationship between analyst recommendation and platform firms. I test the hypothesis on a sample of firms from MSCI Europe. Financial data is retrieved from Thomson Financial DataStream and analyst data from the I/B/E/S database. The findings confirm that platform firms are on average better recommended compared with traditional firms. No support, however, was found for the proposed moderation effect.
Keywords: platform, network-effect, investment analyst, analyst recommendation, competiveness
INTRODUCTION
Platform firms are no longer in their infancy. These days, platform mediated businesses such as Amazon, Uber, Airbnb and Booking.com are a major theme in business strategy discussions, political debates, numerous publications and research papers. Many business executives are eager to explore the new opportunities to create value, triggered by plentiful examples of illustrious platform companies. Recent research shows that more than 80% of executives believe that platforms will be the glue that brings together large groups of users in the digital economy (Tas & Weinelt, 2015) and that platforms are
‘indisputably the leading form of organizing modern digital markets’ (Kilhoffer, Beblavý, & Lenaerts, 2017). With the emergence of digital platforms, firm environment is changing. Nonetheless it is still unclear whether this leads to a systematic competitive advantage for platform firms. An approach to investigate such a situation is by looking at analyst evaluations in more detail.
This study aims to find evidence for that platform firms have a systematic competitive advantage over traditional firms based on investment analyst evaluations. Investment analysts are hereby described as security specialists, typically employed by investment banks and brokerage firms, who analyze the performance and future prospects of a company by gathering and processing information about the firm from published reports as well as directly from management through quarterly earnings conference calls (Wiersema & Zhang, 2011). Their research and recommendations of the firm’s stock (strong buy, buy, hold, underperform, and sell) are consequential for investors decisions (Barber et al., 2001; Womack, 1996). It can be made clear that they collect, process, and disseminate valuable information through their stock recommendation differently for platform firms. Consider the multiple examples of Initial Public Offerings (IPO) from platform mediated businesses where companies were valued extremely high, unless they reported losses or had no revenues or profits at all (Govindarajan, Rajgopal & Srivatava, 2018). I
3
argue that investment analysts are looking beyond the traditional financials and metrics when calculating a platform firm’s worth and potential. This can be clarified by the network effect which characterizes and defines platform mediated businesses (Evans, 2003; Rochet & Tirole, 2006; Rysman, 2009). The effect simply means that the value of the platform to an individual user increases with each additional user (Hidding, Williams & Sviokla, 2013). Network economics theory on platform firms (e.g., Armstrong, 2006; Caillaud & Jullien, 2003; Evans, 2003; Hagiu, 2005; Rochet and Tirole, 2003, 2006) claim that growth in the installed user base is one of the main mechanisms driving a platform’s value and market share. The user base is by some of Resource Based View’s (RBV) seminal papers described as a valuable resource (e.g., Barney, 1986; Dierickx & Cool, 1989). Consistent with this view, due to network effects, a platform that accumulates a larger user base will deliver greater value (Peteraf & Bergen, 2003). As traditional firms often do not have these network effects I expect platform firms to get higher
recommendations because they have more valuable intangible assets in the form of network effects.
However, I expect that size differences as moderating effects modify the relation between investment analyst evaluations and platform firms. Specifically, I propose that a high market capitalization, which is used as a parameter for size in the investment community, will have a positive effect on the relationship between analyst recommendations and platform firms, in such a way that analysts will be more positively biased in their recommendations towards platform firms when market capitalization is high.
This work contributes to the current research in several ways. First, it seems that no research has yet been conducted on assessing the significance of a more positively biased recommendation given by investment analysts for platform firms. Second, the theory and findings will add to the understanding of how platforms are evaluated in the global economy. Third, I contribute to the literature on platform firms by providing research which analyzes the competitiveness of platform firms. Moreover, by applying RBV’s precepts to platform firms I add to an important stream of literature.
THEORY AND HYPOTHESES DEVELOPMENT Competitive advantage for platform firms
The emergence of platform firms is inherent connected with the emergence of wireless and internet technologies (Eckhardt et al., 2018; Iansiti & Lakhani, 2017; McIntyre & Srinivasan, 2017; Parker & Van Alstyne, 2018; Teece, 2018). Firms like Uber and Airbnb have embraced unfamiliar ways of structuring firms and industry boundaries by relocating organizational design elsewhere from selling products towards the abetment of economic exchanges between two or more (related) user groups. Such platform mediated businesses mediate user interactions and therefore contradict with other firms that control a linear series of activities as well as from manufacturing platforms that orchestrate a network of suppliers to build a web of related products (Gawer & Cusumano, 2014; Thomas et al., 2014)
4
In the last two decades platforms, received significant attention from industrial organization (I/O) economics, technology management, and strategy perspectives (McIntyre & Srinivasan, 2017). While these three streams of research have made substantial gains in our understanding of the emergence of platforms and the dynamics of platform mediated networks, many studies across these perspectives have been limited to single-industry settings or narrative cases, thus limiting more robust and generalizable implications. The question addressed in this study, namely “is there a competitive advantage for platform firms” relates to the strength of the network effect as this is seen as a driver of platform value and success (Armstrong, 2006; Caillaud & Jullien, 2003; Evans, 2003; Hagiu, 2005; Rochet and Tirole, 2003, 2006) This area of research holds a I/O economics and strategic management perspective. Due to a holistic and quantitative approach I strive to add more robustness and generalizability to the current available research about platform firms.
When firms offer greater value to customers at a lower cost than the chance of competitive advantage arises (Besanko, Dranove & Shanley, 1999; Hoopes, Madsen &Walker, 2003; Peteraf &
Barney, 2003; Porter, 1985). According to the RBV, resource heterogeneity and various ‘isolating mechanisms’ (Rumelt, 1984) grant firms with high-caliber resources to maintain their resource advantage and sustain their competitive advantage (Barney, 1991, 1997; Diericks & Cool, 1989; Ghemawat, 1986;
Lippman & Rumelt, 1982). Firms gain advantage by controlling valuable, rare, imperfectly imitable and non-substitutable assets (Barney, 1991). In a world different from platform firms those tangible assets include for example mines and real estate and intangible assets like intellectual property (Alstyne, Parker
& Choudary, 2016). For platforms, the intangible assets that are troublesome to imitate are the user base and the resources its users own and contribute, be they knowledge or rooms or work or cars. Alstyne et al., (2016) calls this the network of producers and consumers and argue that it is the most important resource for platform companies. This is similar to the work of Sun and Tse (2009) who argue that participants of a platform, which are usually thought of as the customers, are the critical resources of the network. According to Peteraf and Barney (2003), critical resources are: ‘those factors that have
significant positive effect on either the economic costs or perceived benefits associated with an enterprise’s products’. They are not only essential to the firm’s effort to generate differentially greater value, but also scarce either temporarily or permanently. For platforms apply that existing customers are factors that create a positive impact on the perceived benefits of the network platform, due to the positive cross-group network effects (Sun & Tse, 2009). Thinking about customers as a critical asset is
unconventional as compared to those discussed in the RBV literature. As most researchers view resources as a firm’s internal possession, the study by Sun and Tse (2009) showed that customer relationships, which are commonly viewed as external to the firm, became critical resources (Peteraf & Barney, 2003) for platform mediated businesses.
5
Where traditional firms create value by optimizing an entire chain of product activities, from sales and service to outsourcing materials, platform firms create value by facilitating interactions between external producers and consumers. This external focus often sheds variable costs of production (Alstyne et al., 2016). Furthermore, platform firms pursue maximizing the value of an increasing ecosystem in a circular, iterative, feedback-driven process (Alstyne et al., 2016). By contrast, traditional firms seek to maximize the lifetime value of individual stakeholders who, in sit at the end of a linear process.
It is now exemplified how platform firms differ from traditional firms. However, this does not yet justify why I propose they are better recommended by investment analysts and thus gain a systematic competitive advantage as they are more attractive to investors. Prior research showed that analysts are consequential to investors decisions (Barber et al., 2001; Womack, 1996). A more recent study (Barber, Lehavy & Trueman, 2010) found that analyst recommendations of a firm’s stock and changes in their recommendations, envision a company’s forthcoming earnings and the associated market reaction. This suggests they provide valuable information for new investors. Analysts seem to collect, process, and disseminate valuable information through their stock recommendations differently for platform firms.
Consider that when the New York Times disclosed that Uber is planning an IPO, Uber’s value was estimated around 48$ and $70 billion, unless reporting losses over the last two years. Twitter commanded a valuation of $24 billion on its IPO date in 2013, unless it reported a loss of $79 million before its IPO.
In the four years that followed, it continued to report losses. Equivalently, Microsoft paid $26 billion for loss-making LinkedIn in 2016, and Facebook paid $19 billion for WhatsApp in 2014 when it had no revenues or profits at all. In contrast, industrial giant GE’s stock price has dropped by 44% last year as rumor emerged about its first losses in last 50 years. Clearly investors and analysts are looking beyond the traditional financials and metrics when calculating the firm's worth and potential. It seems that the
financial value of the user base and their network effects cannot be captured by our current financial accounting model. The critical assets for platforms are intangible in nature, and may have ecosystems that extend beyond the company’s boundaries. Therefore I propose that platform firms get higher
recommendations because they have more valuable intangible assets in the form of network effects.
Collectively, the gathered arguments forge the following hypothesis:
Hypothesis 1: There is a positive linear relationship between platform firms and the average recommendation by analysts
The moderating role of size
Whether firms get on average a higher recommendation depends not only on the type of firm (platform VS. traditional), but also on its size. As said earlier, one way to express size is by market capitalization, which refers to the total dollar market value of a company’s outstanding shares of stock and is often used
6
within the investment community (Chen, 2019). Thus, market capitalization measures a company’s worth, as well as the market’s impression of its future prospects. It reflects what investors are willing to pay for its stock. Companies with a large market capitalization commonly have a valuation of 10 billion or more.
This makes them usually the dominant players within specific industries. As a result, investments in large capitalization stocks can be considered as conservative cause they pose less risk. By contrast, small capitalization companies are generally young companies that serve in emerging industries. They are considered as more aggressive and speculative. This can potentially harm them, but on the other hand small capitalization stocks can offer compelling growth potential for long-term investors. Additionally, they have relatively limited resources resulting in a possible more vulnerable position against competitors.
With regard to platform firms, I have mentioned that the user base with their cross-group network effects (Sun & Tse, 2009) are an important asset. Large capitalization platform firms are dominant and therefore have in all likelihood a larger user base in comparison with small capitalization platform firms. As it is difficult to determine the financial value of the user base and its network effect, I reason that analysts are more generous in their recommendations towards platform companies with a large user base and thus a large market capitalization. This due to more certainty of strong network effects.
Furthermore, platform firms often follow a winner-take-all (WTA) ideology which is driven by network effects (McIntyre & Srinivasan, 2017). The ideology means that the platform with the largest number of users will ‘tip the market’ in its favor (e.g., Besen & Farrel; Caillaud & Jullien, 2003; Katz &
Shapiro, 1994; Shapiro & Varian, 1999). When following this ideology it implicates that consumers will always choose for the platform with the largest user base within a certain market. The platforms who certainly have tipped the market in their favor are US-based Apple, Amazon, Microsoft, Google,
Facebook and China-based Alibaba and Tencent. Together these firms represent 69% of the total value of the platform economy (Hottenhuis et al., 2018). These large platforms generally bring better value to users compared with smaller platforms and this turn bolsters to attract even more users and become even more valuable at an exponential rate. This effect can be understood by analyzing the average annual increase in market value of the top-seven platforms and lower valued platforms. Hottenhuis et al (2018) showed in their report that while lower valued platform companies have added an average of $460 million to their market value every year since their foundation, a top-seven platform has added an average of
$31.6 billion in market value each year since their launch. This is 69 times more. The outcome reveals that it is more relevant to fuel the growth of a few very large platforms than to create numerous small ones.
I reason that analysts will recommend platforms even better when they have proven to be the winner in their market. Those 7 platforms have clearly won the battle for users resulting in a high valuation. Additionally, data has proven that large platform companies show a much stronger increase in
7
valuation, which presumably makes them more attractive for investors. Therefore, I submit that size will moderate the effect between platform firms and average analyst recommendation in such a way that a high market capitalization will lead to a more positively biased recommendation given by investment analysts:
Hypothesis 2: The relationship between platform firms and average analyst recommendation is positively moderated by the size of the company
DATA AND METHOD Sample
This study examined companies listed on MSCI Europe index for the period 2008-2017. The choice for this sample was prescribed by the call for firms that have extensively traded stocks and are, thus,
extensively monitored by the investment community. Firm data is used from Compustat, Financial data is obtained from Thomson Financial DataStream and analyst data from the Institutional Brokers Estimate System I/B/E/S database. The sample contains companies around the world. About 36% of firms are based in Asia, 30% in North America, 20% in Europe, 7% from Middle or South America. African, Australian and Middle Eastern make up the rest of the sample. The initial dataset consisted of 3.816 firms.
Further, because data on investment analyst stock recommendations were not available for some firms I dropped with 468 firms. This resulted in a final dataset of 3.348 firms with 21.277 unique firm-year observations, of which 369 for platform firms and 20881 for traditional firms.
Dependent variable
Average analyst recommendation
This study uses one measure of investment evaluations: average analyst recommendation. This measure is calculated from data gathered from the I/B/E/S database. I/B/E/S uses a five-point scale, with a
recommendation of 1 meaning ‘strong buy,’ 2 meaning ‘buy,’ 3 meaning ‘hold,’ 4 meaning
‘underperform,’ and 5 meaning ‘sell.’ This means that in the I/B/E/S’s scale, higher ratings mean lower recommendations. To simplify the analysis, I reverse coded the mean recommendation by subtracting it from six. This method is also used in the article by Wiersema and Zhang (2011). The average analyst recommendation is measured as the mean analyst recommendation for all investment analysts who cover a firm in one year. For the total sample, the average analyst recommendation ranges from 1 to 5, with a mean of 3.44.
8 Independent variable
Type of firm
The goal of this study is to find evidence for that platform firms have a systematic competitive advantage in the form of a higher average recommendation by investment analysts. Therefore, all the companies in the dataset are identified if whether the firm can be labeled as a platform firm. The source used to identify platform firms is the WisdomTree Modern Tech Platform Index. This index encompasses mid- and large- cap companies that are listed and generate substantial revenue from a modern platform business
(Wisdomtree, 2013). It uses a classification based on customer relationships, producer relationship, Value Created by Producer, Network Effect, Network Ownership, Platform Revenue and Platform Revenue Percentage. To confirm this identifying process manual checks have been performed by analyzing every platform based market if firms were not missed out. This approach has led to the identification of 71 platform mediated businesses. Of these 71 businesses, 31 were present in the sample. Together these firms account for 369 unique firm-year observations, within the 21277 observation in the sample. A dummy variable is made to signify if it is an observation for a platform firm. A ‘1’ is given for a platform firm and a ‘0’ is given for any other firm, which I label as traditional firm.
Moderator variable
The measure of market capitalization is retrieved from Thomson Financial DataStream. To facilitate interpretation I standardized this variable.
Controls
The following variables are included as controls: performance as industry-adjusted return on assets, and firm size as the reasonable log of total assets. Additionally, I controlled for the number of analysts who provide research for a firm. This in consideration of that a more extensive investment analyst coverage presumably leads to greater investor scrutiny. Analyst coverage is measured by the number of analysts who cover a firm per year as reported in the I/B/E/S database. Additionally, I controlled for the variation across analyst recommendations of a firm since this arguably has an effect on the average analyst recommendation. Models include industry-, year-, and country dummies.
To confirm results are not driven by outliers all financial variables are winsorized at the 0,5 percentile.
Analysis
The hypotheses are tested through quantitative analysis. Analysis have been performed using Stata, the syntax is recorded and saved, and available upon request. The dataset contains observations of multiple phenomena obtained over multiple time periods for the same firms. This structure suggests the use of
9
panel estimation techniques. The relations in both H1 and H2 are tested using multiple linear regression.
For all the calculations the control variables mentioned above have been included.
I estimate my model in a panel-data sample that includes 21.277 firm-year observations. In a sample like this unobserved heterogeneity is a potential problem. It could be that analyst
recommendations are affected by systematic differences in the financial environment across countries and years or that differences in risk and performance across industries exist. By not accounting for this it could lead to bias in my estimations. An approach to address this issue is to insert additional firm-specific error terms that are fixed over time for each firm in my dataset (Sayrs, 1989). To reduce the potential problems, a set of dummy variables, one for each recommendation year, industry and country are included as regressors.
Table 1a. Descriptive statistics
Variable N mean sd min max
Average recommendation 21.277 3,44 0,441 1 5
Platform firm 21.277 0,02 0,14 0 1
total assets 21.200 25300000 40500000 1145509,00 162000000
return on assets 20.835 6,40 5,25 -0,63 18.98
market capitalization 21.174 13800000 2890000000 2741 738000000
analyst coverage 21.277 17,01 11,13 1 106
analyst consensus 21.000 0,90 0,25 0,000 2,828
year 21.277 2012,55 2,89 2008 2017
Country 21.277 30,71 15,78 1 49
Industry 21.242 3,95 1,90 0 8
Table 1b. Correlations
Variable 1 2 3 4 5 6 7 8 9 10
1. Average recommendation 1,00
2. Platform firm 0,03 1,00
3. total assets 0,01 0,03 1,00
4. return on assets 0,07 0,06 -0,33 1,00
5. market capitalization 0,09 0,18 0,48 0,08 1,00
6. analyst coverage -0,07 0,03 0,15 0,04 0,10 1,00 7. analyst consensus -0,16 -0,01 0,00 0,08 -0,01 0,32 1,00
8. year -0,01 0,02 0,09 -0,02 0,14 -0,24 -0,15 1,00
9. Country 0,03 0,01 -0,01 0,09 0,12 -0,07 -0,02 -0,01 1,00 10. Industry -0,02 0,16 0,23 -0,11 -0,01 -0,08 -0,03 0,03 0,05 1,00
10 RESULTS
Descriptive statistics for the complete dataset are presented in Table 1ab. Table 2 presents the first regression results with all controls and the main effect of the relationship between platform firm and the average analyst recommendation. This table is used to test hypotheses 1.
Table 2. Linear regression results of the relation between platform firm and average analyst recommendation*
n 20,564
R-squered 0,1312
Prob > F 0,0000
DIRECT EFFECT
Coef. Std. Err. t P>|t|
[95% Conf.
Interval]
Average recommendation
Platform firm 0,045 0,021 2,110 0,034 0,003 0,087
Total assets 0,000 0,000 10,890 0,000 0,000 0,000
return on assets 0,006 0,000 10,750 0,000 0,005 0,007
analyst coverage -0,003 0,000 -11,180 0,000 -0,004 -0,003
analyst consensus -0,237 0,013 -18,260 0,000 -0,263 -0,211
year * * * * * *
country * * * * * *
Industry * * * * * *
* model include year, country and industry dummies.
Table 2 shows that the analysis is consistent with the argument that platform firms are on average better recommended than traditional firms (b=0.045, p<0.05). The model predicts a very small advantage in average analyst recommendation for platform mediated businesses. Given that the mean of the total sample is 3,44, ranging on a scale from 1 to 5, one can argue that upward bias in recommendations is prominent. For platform firms the upward bias in recommendations is estimated slightly stronger. Among the control variables the performance and size measure are both significant. Indicating that companies who have more assets and more return on assets relatively have better analyst recommendations. Together these results provide important insights into the existence of a systematic competitive advantage for platform firms.
To test the moderating hypotheses that the effect of platform firms on the average analyst recommendation would be greater when the company’s size is high (Hypotheses 2), I add market
capitalization, the expressed figure for size in this study, as an interaction term. The results of the analysis are shown in Table 3. What is interesting about the figures in this table is that my expected moderation effect is negative instead of positive (b=-0.011, p<0.1). This suggests that market capitalization has an opposite effect than I expected. The effect is remarkable, yet not very convincing as the model is
11
significant only on the level of p<0.1. Moreover, the main effect (platform firm) is statistically
insignificant (b=0.156; t = 0.69). This suggests there is a cross-over interaction. The effect of platform firm is opposite, depending on the value of market capitalization. The results indicate that when market capitalization is small, this increases the average analyst recommendation, but, when market
capitalization is large this downgrades the average analyst recommendation for platform firms.
Table 3. Moderating effect analysis of firm size on the relation between platform firms and average analyst recommendation*
n 20,542
R-squered 0,1348
Prob > F 0,0000
INTERACTION TERM
Coef. Std. Err. t P>|t|
[95% Conf.
Interval]
Average recommendation
Platform firm 0,156 0,023 0,69 0,488 -0,029 0,06
Market capitalization 0,038 0,004 8,67 0,000 0,029 0,047
Platform firm#Market capitalization
Platform firm -0,011 0,006 -1,68 0,094 -0,024 0,002
Total assets 0,000 0,000 3,32 0,001 0,000 0,000
return on assets 0,005 0,001 8,37 0,000 0,004 0,007
analyst coverage -0,004 0,000 -11,55 0,000 -0,004 -0,003
analyst consensus -0,236 0,013 -18,22 0,000 -0,261 -0,210
Year * * * * * *
Country * * * * * *
Industry * * * * * *
* model include year, country and industry dummies.
DISCUSSION
The main goal of the current study was to determine if platform firms have a systematic advantage over traditional firms based on a higher average analyst recommendation. I argued that platform firms have an advantage because they possess a strong intangible asset which is difficult to measure, but can have an excessive impact. The user base with its network effect. Empirically, I found support for the above statement. There seems to be a clear relation between the type of firm (platform vs. traditional) and the average analyst recommendation. Also, there are signs of a mitigating role for market capitalization on the relation between platform firms and the average analyst recommendation. This result is however modest. The study’s findings is the first to show that platform mediated businesses are on average better recommended compared with traditional firms. In addition, I provide new insight into the systematic
12
competitive advantage for platform firms by integrating how firm size may moderates the relationship between analyst recommendations and firm type.
The findings of this paper contribute to the literature on platform firms by examining whether a competitive advantage exists for these kind of businesses. As this study is pioneering in analyzing the competitive advantage of platform firms based on analyst evaluations it is rather difficult to relate the outcomes to existing literature. In this study I tried to base my explanation for the proposed direct effect on the precepts of the RBV. The majority of research on the RBV did not consider the user base with its network effect as one of the critical resources to gain and sustain competitive advantage. This is not completely surprising since RBV shifts the focus of strategy research from external (market or industry) factors to firm’s internal resources (Wernerfelt, 1984; Barney, 1991). And as most researchers view resources as a company’s internal possession, it is rather unconventional to argue for something that is normally viewed as external to the firm, the customer relationships, as a crucial asset which gives them an advantage over traditional firms. Several scholars (Alstyne,2018; Sun & Tse, 2009) have claimed that the user base and it’s network effect count as a critical resource for platform mediated businesses. The outcome seems to be in line with my reasoning. However, it is hard to make undeniable statements.
Especially since in these kind of studies it is impossible to find the exact reason why analysts recommend platform firms on average slightly better. This study purely tries to provide empirical evidence for a systematic competitive advantage for platform firms and uses the current available theory to develop the propositions.
Further research has to delve deeper into the exact reasons for investment analysts why they treat platform firms differently. What for example remains unclear is how the financial value of the user base and it’s network effect can be or is determined. More surprising, the analysis did not illuminate
convincing evidence for size, as moderating effect on the relationship between firm type and average analyst recommendation. This is presumably attributable to the relatively small group of platform firms that are present in the sample. Only 1,86% of the total sample are indicated as platform firms. Also, as no entries have been deleted, a wide variety of organizations and industries have been used, with vastly different business models. All these firms are labeled as traditional firms, however they undoubtedly will differ from each other resulting in different evaluations by analysts. The result, nonetheless helps make for interesting discussion.
The results of this study has limitations that, if discussed, might provide worthwhile avenues for future research. The general assumption that a systematic competitive advantage arises when your average analyst recommendation is higher can be challenged. Despite that analyst recommendations are consequential for investors decisions (Barber et al., 2001; Womack, 1996), this does not imply that when your average analyst recommendation is higher you will generate more value compared with
13
competitors. However, being more attractive for investors assumable generate more financial resources.
Another limitation relates to the number of platform firms in the sample used in this study. The identification of platform mediated businesses was based on an external index and not by objective measures from financial data. If new studies can link certain financial data to platform firms than a replication of this study can possibly generate more robustness to the current outcomes. Also, it can be expected that in the coming years the number of ‘listed’ platform mediated businesses will increase.
Therefore it would be interesting to repeat this study in the next decade to see how results differ.
While I have examined the moderating role of firm size, as expressed in market capitalization, on the relation between platform firms and the average analyst recommendation, additional research that examines alternative factors may also prove insightful. Since I theoretically propose that analysts will recommend platforms better compared with traditional firms and this recommendation will even be better when they have a large market capitalization, it might be interesting to use different expressions for size.
In addition, Hottenhuis et al. (2018) showed in their report figures about the global platform economy by size. They categorized platforms based on valuation: Super platform +$250.000 million, Elite Unicorn +$25.000 million, Unicorn +$1.000 million and Scale up >$100 million. As shown in figure 1
(Hottenhuis et al., 2018, p. 10) , the number of platform firms is relatively low for firms valued higher than $250.000. Figure 2 (Hottenhuis et al., 2018, p. 11) indicates that super platforms are for 100%
public. In contrast, the Scale-Up platforms are for 89% privately owned.
Figure 1. Original caption of the figure. Adapted from “Unlocking the value of the platform economy,” by S. Hottenhuis, F. Blom, R. Boekhout, B. Knibbe, D. Lemstra, R. Miesen and P. Zijlema. Dutch Transformation forum. p. 10. November 2018.
14
Figure 2. Original caption of the figure. Adapted from “Unlocking the value of the platform economy,” by S. Hottenhuis, F. Blom, R. Boekhout, B. Knibbe, D. Lemstra, R. Miesen and P. Zijlema. Dutch Transformation forum. p. 11. November 2018.
This might explain the unexpected result of the moderation effect. What can be clearly assumed based on this figure is that the small market capitalization platform firms are less covered by analysts since they are for the most the most part privately owned and thus have less outstanding stocks. This less coverage might have an influence on the average analyst recommendation, especially since analysts exhibit
‘herding’ behavior (Hong, Kubik and Solomon, 2000; Hong and Kubik, 2003; Trueman, 1994). Meaning that analysts follow the consensus with regard to the recommendations they issue on a particular stock.
And as there clearly is a trend visible for platform firms who launch their stocks to the market being valued extremely high it could be that especially for those scale-up platforms herding behavior is prominent. Also, for younger platform firms the network effect and its value is rather speculative and difficult to measure, but their stocks can offer compelling growth potential. Future research could add a classification for platform firms in order to identify if differences in competitive advantage based on analyst evaluations exist for platform mediated businesses.
Furthermore, I acknowledge that outcomes of this study result from a first version of my developed regression model. Further development and possibly unforeseen necessary changes could change outcomes.
Based on existing literature practitioners can learn that platform firms possess the network effect leading to several advantages. This thesis serves as proof that there is a systematic competitive advantage for platform firms due to this network effect. Also, a new factor in understanding the relation between platform firms and analyst recommendations has been identified in the form of size as expressed in market capitalization. Managers who are eager to explore new opportunities to create value benefit from these findings in being sure that transforming their businesses towards a platform mediated business model will lead to a more attractive position towards investors. Platform firms are on average higher
15
recommended and therefore have a systematic competitive advantage. However, I do not advise managers to immediately change their businesses to a platform mediated business model as results only show a marginal difference between platform firms and traditional firms and the number of platform firms in the sample is rather limited. Apart from this managers should not base their choice of changing their business model solely on their position towards investors.
In spite of its limitations, the study certainly adds to our understanding of how platform companies are evaluated in the global economy. I conclude that a systematic competitive advantage for platform firms exist. No clear support has been found for the effect of size, as expressed in market capitalization, for the relation between platform firms and average recommendation. There are likely multiple moderating mechanisms that have an effect on the relation between firm type and average analyst recommendation. This study has raised an important question about platform mediated businesses.
I hope that researchers will further investigate the competitiveness of these kind of companies.
16 REFERENCES
Alstyne, M. W., Parker, G. A. & Choudary, S. P. (2016). Pipelines, Platforms, and the New Rules of Strategy. Harvard Business Review, April 2016.
Armstrong, M. (2006). Competition in two-sided markets. Journal of Economics 37,3, 668–691.
Barber, B., Lehavy, R., McNichols M., & Trueman, B. (2001). Can investors profit from the prophets? Security analyst recommendations and stock returns. Journal of Finance, 56(2), 531–563.
Barber, B. M., Lehavy, R., & Trueman B. (2010). Ratings changes, ratings levels, and the predictive value of analysts’ recommendations. Financial Management 39(1), 533–553.
Barney, J. (1986). Strategic factor markets: expectations, luck, and business strategy. Management Science 42, 1231-1241.
Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17, 99–120.
Besen ,S. M., & Farrell, J. (1994). Choosing how to compete: strategies and tactics in standardization.
Journal of Economic Perspectives, 8(2), 117–131.
Caillaud, B., & Jullien, B. (2003). Chicken-and-egg: competition among intermediation service providers. Journal of Economics 34, 309–328.
Chen. J. (2019). Market Capitalization. Retrieved from
https://www.investopedia.com/terms/m/marketcapitalization.asp
Dierickx, I., & Cool, K. (1989). Asset stock accumulation and sustainability of competitive advantage.
Management Science 35, 1504-1514.
Evans, D.S. (2003). Some empirical aspects of multi-sided platform industries. Review of Network Economics, 2, 191–209.
Ghemawat, P. (1986). Sustainable advantage. Harvard Business Review, 64, 53–8.
17
Govindarajan, V., Rajgopal, S., & Srivastava, A. (2018). Why Financial Statements Don’t Work for Digital Companies. Retrieved from https://hbr.org/2018/02/why-financial- statements- dont-work-for-digital-companies
Hagiu, A. (2005). Pricing and commitment by two-sided platforms. Journal of Economics, 37,720–
737.
Hong, H., & Kubik, J. D. (2003). Analyzing the analysts: career concerns and biased earnings forecasts. Journal of Finance, 58(1), 313–351.
Hong, H., Kubik, J. D., & Solomon, A. (2000). Security analysts’ career concerns and herding of earnings forecasts. Journal of Economics 31(1), 121–144.
Hottenhuis, S., Blom, F., Boekhout, R., Knibbe, B., Lemstra, D. Miesen, R., & Zijlema, P. (2018).
Unlocking the value of the platform economy. Dutch Transformation Forum. April 2018.
Hidding, G.J., Williams, J. & Sviokla, J. (2010). The IT platform principle: the first shall not be first.
Wall Street Journal, January, 25, R4.
McIntyre, D.P., & Srinivasan, A. (2017). Networks, platforms and strategy: emerging views and next steps. Strategic Management Journal, 38 (1), 141–160.
Katz, M.L., & Shapiro, C. (1994). Systems competition and network effects. Journal of Economic Perspectives, 8, 93–115.
, Dutch Transformation Forum. (2018). Unlocking the value of the platform economy.
Retrieved from https://dutchitchannel.nl/612528/dutch-transformation-platform- economy- paper-kpmg.pdf.
Kilhoffer, Z., Beblavý , M., & Lenaerts, K. (2017). An overview of European Platforms: Scope and Business Models. Brussels: European Commission
Lippman, S. A. & Rumelt, R. P. (1982). Uncertain imitability: an analysis of interfirm differences in efficiency under competition. The Bell Journal of Economics, 13, 418– 38.
18
Peteraf, M. A. & Barney, J. B. (2003). Unraveling the resource-based tangle. Managerial and Decision Economics, 24, 309–23.
Rochet JC, Tirole J. (2003). Platform competition in two- sided markets. Journal of the European Economic Association, 1, 990–1029.
Rochet JC, Tirole J. (2006). Two-sided markets: a progress report. Journal of Economics, 37, 645–
667.
Rysman M. (2009). The economics of two-sided markets. Journal of Economic Perspectives 23, 125–
143.
Shapiro, C., & Varian, H. R. (1999). The art of standards wars. California Management Review 41(2), 8–32.
Sun, M., & Tsee, E. (2009). The Resource-Based View of Competitive Advantage in Two-Sided Markets. Journal of Management Studies, 46(1).
Tas, J., & Weinelt, B. (2015). Digital Transformation Initiative: Unlocking B2B Platform Value.
Geneva: World Economic Forum.
Trueman, B. (1994). Analyst forecasts and herding behavior. Review of Financial Studies, 7(1), 97–
124.
Wiersema, M. F., & Zhang, Y. (2011). CEO dismissal: The role of investment analysts. Strategic Management Journal, 32(11), 1161–1182.
Womack K. 1996. Do brokerage analysts’ recommen- dations have investment value? Journal of Finance, 51(1), 137–167.
