IPOs and the long-term gains from spin-offs

University of Amsterdam BSc economics and business economics Specialization Finance & Organization Thesis Finance

IPOs and the long-term gains from spin-offs

Author: Bart van der Poel Student Number: 11021942 Supervisor: Patrick Stastra Submission date: 30/06/2020

Abstract

This thesis investigates the effect of an earlier spinoff on the long-term stock performance. An advice on whether to invest in IPO stock of firms who were spun-off on beforehand is given and possible reasons why this stock could differ in performance compared to the benchmark is explained. This is done by analyzing the buy and hold returns of 95 IPO firms for the first three years after their IPO. No significant difference has been found between the buy and hold returns of spinoff firms and non-spinoff firms. A possible limitation of this thesis is the concentration on United States based firms. Because of a law that forced companies to go public when they exceed 500 employees, findings in this thesis might not be external valid to IPO markets outside the United States.

Statement of originality

This document is written by Student Bart van der Poel who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Table of contents

1. Introduction ………. 4 2. Literature review ……….. 6 2.1 Initial public offerings ………... 6 2.2 Motives to go public ……….. 6 2.3 Short-term IPO performance ……… 7 2.4 Long-term IPO performance ………. 7 2.5 Divestitures ………..……… 8 2.6 Hypothesis ………. 9 3. Data & Methodology ………. 10 3.1 Data ……….. 10 3.2 Literature on event study methodology ………. 11 3.3 Applied methodology ……….……. 12 4. Research results ……….. 14 4.1 Descriptive statistics ………. 14 4.2 Regressions ………. 14 5. Conclusion & Limitations ……….. 16 5.1 Conclusion ……….. 16 5.2 Limitations & recommendation ……….. 17 References ……… 19 Appendix ……… 21

1. Introduction

An initial public offering is the first time a private company is selling shares to the general public (Pagano, Panetta & Zingales, 1999). The main motive to go public is because it is a relative cheap way of gathering capital. However, in the process of going public a lot of direct cost are incurred so it is not always rewarding. Studies of among them Ritter (1984) showed that the first day closing price of IPO shares are significantly higher than the offer price, this is called underpricing. Even though IPO shares show positive returns after the first trading day, in the long run they tend to underperform compared to the benchmark of non-IPO shares. Ruud (1993) concluded that IPO shares underperform in at least the first 5 trading years. Divestitures are the sale of a portion of the firm (Eckbo & Thorburn, 2013). The most common form of divestitures are spinoffs. Spinoffs are the sale of a division by the parent company and thereafter, shares of the spun-off division (subsidiary) are distributed among shareholder of the parent or through an IPO. Not much research has been done about the performance of these subsidiaries compared to private companies that go public. The subsidiary got spun-off for a reason, this could be due to financial distress or because either the parent or subsidiary wants to focus more on the core business. The effect of spinoffs on the long-term performance is expected to be ambiguous. This thesis tries to contribute in the understanding of the performance of spun off subsidiaries. Hence, the main research question is: “To what extent does performance differ among spin off and non-spinoff firms in the first three year after their initial public offering?” To research this, stock performances of 95 American companies that went public in the period of 2009 till 2017 are analyzed and compared by using the three year buy and hold abnormal returns. After that, firm characteristics that could influence the buy and hold returns are taken into account and controlled for in a second regression. First results show that subsidiaries have 16.6% return over three years, while the benchmark return is 45.4%. However, after using a t-test to test whether the difference between the returns differs from

zero, no significant difference was found. This implies that the performance of spun-off firms does not differ in the first three year after an IPO compared to the benchmark. This thesis is organized as follows. Chapter 2 gives an overview of existing literature on IPO’s, motives to go public, the short and long run performance of IPO firms and the concept of divestitures is provided. Chapter 3 is divided into two parts. In the first part, the data criteria and the collection process is discussed. In the second part, literature on the methodology and the applied methodology is explained. Chapter 4 presents the descriptive statistics and research results. Finally, in chapter 5 the conclusion will be given and the research question is answered.

2. Literature review

This chapter will provide a theoretical background about concepts discussed in this thesis and discuss existing literature. Most important are the concept of initial public offerings and the concept of divestitures. 2.1 Initial public offerings The two most common types of companies are private and public companies. An initial public offering (IPO) is the first time a private company is selling stock to the general public. Through this offering, the private firm gathers capital to finance future expansions (Pagano, Panetta & Zingales, 1998). 2.2 Motives to go public The advantages of going public should outweigh the disadvantages of going public. Most of the advantages of going public are finance related, it is a relative cheap way to gather capital because companies don’t have to borrow and pay dividend. Another benefit is an increase in reputation and publicity. By going public companies are more known by the general public which may increase sales or other reputation related business operations like closing deals with clients. Going public also allows to reward the management with stock options which are a good incentive to run the company the best they can (Marcus & Myers, 2001). The main disadvantage of an IPO is that the direct costs are very high. In the process of going public, an underwriter has to be hired. The underwriter does research on the firm that wants to go public and comes up with the offer price of the new shares and the volume of the new shares. The offer price is the price of shares issued and the share volume is the quantity of the issued shares. If the direct costs are more than the cost of debt, it is not beneficial to go public. Unfortunately, the total direct costs are not known beforehand so going public always comes with risk (Draho, 1971). Another disadvantage of going public is underpricing which means that the offer price is lower than the fair value of the shares. This will be explained in the next part.

2.3 Short-term IPO performance Various research has been done by for example Jog and Riding (1987) and Ritter (1984) about the closing price after the first trading day of new issued shares. They concluded that the first day closing price is significantly higher than the offer price, which means that shares are offered for a discount. This concept is called underpricing and shows that the short-run return of IPO shares is positive. There are several theories that try to explain this phenomenon of underpricing. The first theory is the theory of adverse selection. Rock (1986) investigated two groups of investors who invest in IPO’s. One group had no information about the quality of the IPO firm and the corresponding share price, the other group had. What turned out was that the group investors that had no information invested randomly, and the informed group only invested in shares that were issued below or at their true value. Because of this, the informed group crowded out the uninformed group if profits were to be made. This resulted in negative returns for the uninformed group, after all, they only got the overpriced shares. This is also known as ‘the winner’s curse’. This experiment can be used to explain underpricing. Rational investors won’t buy IPO shares if they know they break even at best, this will lead to a drop in demand of IPO shares. So, to encourage uninformed investors to buy IPO shares, firms have to offer a discount to investors. Chemmanur (1993) came up with a second theory that could explain underpricing, namely asymmetric information. He stated that IPO firms know the fair value of a share while investors do not, this could lead to the so called ‘lemon problem’ (Akerlof, 1970). The lemon problem comes down to the fact that only low-quality firms are willing to sell their shares at an average price, which drives out the high-quality firms. As a result, rational investors are only willing to invest in IPO-firms when the shares are underpriced. 2.4 Long-term IPO performance New issued shares often have a significant higher closing price than the offer price, however in the long run they frequently underperform. Ruud (1993) and Ritter (1991) have pointed out that IPO firms have a below average performance compared to the benchmark in at least the first 5 years after the IPO. Ritter (1991) reported an underperformance of 28%. The average return of IPO firms between 1975 and 1984 was 34% in the first 3 trading years, in

that same period, the non- IPO firms had an average return of 62%. Underperformance has also been found by Levis (1993) in the United Kingdom and by Stehle et al. (2000) in Germany. The price support hypothesis tries to give an explanation for these findings. According to Rudd (1993), underwriters keep the initial trading price artificially high for the first period after the issuance to create a favorable situation for investors. After this period of interference, the stock prices slowly adjust downwards to their true market value. This theory can be linked back to underpricing because when underwriters sell the stock initially too low, they create an excess demand. Because of this excess demand, the stock price overshoots its fair price leading to a correction to the mean price in the long run (Shiller, 1988). Ibbotson (1994) showed that investors who initially invest in IPO stock are the most risk seeking investors. When the value of a share is unknown, the price risk seeking investors want to pay for a share is higher than the price risk averse investors want to pay and the mean price will probably lie between these prices. Because risk seeking investors buy IPO stock, the stock will initially be priced higher than the mean price.

2.5 Divestitures Corporate divestitures are defined as the sale of a portion of a firm and come in a variety of forms (Eckbo & Thorburn, 2013). The most common form of divestitures are spin-offs. A spin-off is characterized by the fact that a new independent firm is created and the parent continues existing as a separate entity. The divested assets are transferred to a new created firm whose shares may be distributed to the shareholders of the old firm. The new shares also could be distributed via an IPO (Hite & Owers, 1983). For distinctness, the old firm who divests its assets will be referred as ‘parent’ and the unit spun-off as ‘subsidiary’. Hite and Owers suggest that spin-offs are the opposite of mergers, since one company becomes to companies. A spin-off creates value when negative synergies between units are eliminated. This occurs when two separate companies have more value than those two together as one. A lot of research has been done about value creation for the parent through spin-offs and studies by for example Desai and Jain (1999) conclude that spin-offs often create value for shareholder of the parent. Desai and Jain (1999) state that the main motive to spin off

subsidiaries is because they are underperforming and that value creation is mainly caused because the parent can focus more on the core business. In line with this, the subsidiary is able to focus more on its core business as well which could lead to better performance. Earlier, Hamilton and Chow (1993) researched possible reasons why managers divest certain assets. They conducted a questionnaire among CEO’s of the 98 largest companies in New Zealand. The two main motives to divest where shortage of capital and lack of growth potential of the asset. One of the biggest success stories is the spin-off of Ferrari. Ferrari was a part of the American-Italian automotive maker Fiat Chrysler NV. Fiat’s CEO at the time Sergio Marchionne (Ferrari NV, 2015, p3) stated “An independent Ferrari will be better positioned to realize its potential and will have increased access to capital”. In January 2016, Fiat Chrysler spun-off Ferrari via an IPO with an offering price of 53 dollars per share. Four good performing years later, the stock price is 170 dollars per share, an increase of almost 70%. However, not much further research has been done about the performance of subsidiaries after IPO’s and how subsidiaries perform compared to non-spun-off companies.

2.6 Hypothesis With the literature taken into account, it is hard to make a predicting about the performance of IPO firms and whether spun-off firms will have nonzero abnormal return with respect to firms that were not spun-off beforehand. Often, spin-offs are done by parent companies because they want to focus on their core business or because the subsidiary has issues meeting their financial goals (Hamilton & Chow, 1993). In general, the parent company has more information than the general public (asymmetric information). When a subsidiary is spun-off, the public could think that this is because of financial issues. However, some spun-off subsidiaries have shown better than average returns in the first years after the IPO, see the Ferrari case. Since the effect is ambiguous in earlier literature, the following hypothesis is formed: Spinoff firms perform as well as non-spinoff firms in the first three year after their initial public offering.

3. Data & Methodology

This chapter provides an overview of existing literature on methodology and the methodology which is used in this thesis will be explained. 3.1 Data The data used in this thesis contains IPO’s between July 2009 and January 2017 in the United States. According to the national bureau of economic research the American recession ended in July 2009. By choosing this timeframe influences by the economic crisis are excluded. The Covid-19 outbreak and the potential influence it has on stock prices has also been accounted for by ending the timeframe in January 2017. To measure the three-year post IPO stock performance, at least three year of known stock data is needed. The last selection criterion is that the headquarters of the IPO firm is in the United States. After collecting the IPO information, the monthly stock returns of 95 companies could be retrieved. These 95 companies are from diversified industries and are both spinoff and non-spinoff companies. Table 1 shows how many IPO’s took place in every year, with a distinction between spinoff firms and non-spinoff firms. Table 1: Number of IPO’s per year Year 2009 2010 2011 2012 2013 2014 2015 2016 Total Spinoff Yes 0 1 6 3 3 8 10 15 46 No 3 5 2 7 4 12 11 5 49 Total 3 6 8 10 7 20 21 20 95 The Factset database also provided the Standard Industrial Classification numbers which could be matched with the respected industries. The 9 biggest sectors are used. Table 2 shows how many IPO’s took place per industry, again with a distinction between spinoff firms and non-spinoff firms. The service industry has the most observations (32), followed by manufacturing (15) and retail trade (12).

Table 2: Industries of subsidiaries and non-spun off firms

Spinoff No Yes Totaal

Industry Agriculture 3 0 3 Construction 2 1 3 Finance & Real estate 6 4 10 Manufacturing 4 11 15 Mining 2 3 5 Retail trade 5 7 12 Service 21 11 32 Transportation 3 5 8 Wholesale trade 3 4 7 Total 49 46 95 3.2 Literature on event study methodology The event study methodology is used by most authors, among them Mackinley (1997), when reviewing long-term IPO performance and to test whether nonzero abnormal stock returns occur because of a corporate event. In this approach, average stock returns of the IPO firm are calculated from the first trading day till the desired ending of the time window. After that, the average stock return of the benchmark is calculated and these returns are compared with each other. The event study approach can be subdivided into multiple types, the two most common are the Buy-and-Hold Abnormal Returns (BHAR) and Cumulative Abnormal Returns (CAR). The BHAR method assumes a Buy-and-Hold investment from the first trading day till the end of the time window. To use the BHAR method, compounded returns over the time period have to be calculated from the subsidiary as well as the from benchmark. After this, the two returns are subtracted from each other. Fama (1998) critiqued the BHAR method because according to him, one year with extreme results gives a biased result of the compounded returns. The BHAR approach does give a good representation of the investors point of view, because it gives the return on a stock which is held for a given period. The second method, the CAR approach does not make use of compounded returns and is therefore preferred in some research. The CAR method calculated the excess returns relative to the benchmark, and then aggregates this over time. This assumes monthly

rebalancing during the time window (Ikenberry, Lakonishok & Vermaelen, 1995). However, the CAR method has also some shortcomings. The main critique on the CAR method is that it gives a small upward bias. There are two ways of proportioning the values of firms to calculate abnormal returns, value weighted and the equally weighted. The value weighted method gives bigger firms more proportion in comparison to smaller firms in the calculation of abnormal returns. The equally weighed method does not account for firm size and gives all firms the same proportion. The value weighted method has some shortcomings, several research has shown that bigger firms cause bias because they have a bigger proportion than smaller firms. To use nonzero abnormal returns to measure whether there is an effect on IPO firm performance, two assumptions have to be true: firstly (1), the stock price reflects all the public known information about the firm and the event. And secondly (2), the benchmark model must be correct. If the first assumption does not hold, the measure of abnormal return would be incorrect, and if the second assumption does not hold, the measure is biased. 3.3 Applied methodology This thesis takes a time horizon of three years after the IPO, in line with earlier research of Ritter (1991). The BHAR method to calculate nonzero abnormal returns will be used to test whether subsidiaries have different returns than non- spun off IPO firms because this method gives a better view of the investors point of view. Following the research of Chan et all. (2004), the BHARs are calculated by the following formula: (1)

Where ÕI (1+Ri,t) is the average annualized return on a portfolio of subsidiaries and

ÕI(1+RP,t) is the average annualized return on the benchmark portfolio. The average returns

A T-test will be run to test whether the BHAR differ from 0 and thus whether subsidiary’s have different returns compared to non-spun off IPO firms. Next, the following regression with control variables will be run to try to explain the possible BHAR:

(2) BHR = b0 + b1 Spinoff+ b2 Firmsize +b3 Industry + b4 Year IPO

The hypotheses below will be used to test whether subsidiaries perform as well as non-spinoff firms in the first three year after their initial public offering: H0: b1 = 0 H1: b1 ¹ 0 BHR is the Buy and Hold returns of all individual companies in the sample, these are annualized monthly holding returns over the 3-year time horizon. The difference with the BHAR from (1) is that these returns are for every individual company instead of the average of all companies. This regression is run to explain the buy and hold returns and whether spinoffs influence the returns. The spin-off variable is a dummy variable, it shows whether the IPO firm was spun-off. The variable spin-off will be tested with a t-test to test the hypothesis formulated under (2). Firmsize, industry and Year IPO are control variables. Firmsize is measured by market capitalization at the time of the IPO. The natural logarithm will be used to account for big firm bias. Industry is a dummy variable for the firm’s industry, the 9 biggest sectors will be used. The variable year is also a dummy variable and denotes the year of the IPO.

4. Research results

This chapter discuses and analyzes the results of the regressions. Whether subsidiaries have abnormal buy and hold returns is tested first. The potential influences on the returns will be test and discussed next. 4.1 Descriptive statistics The average buy and hold return of the sample is 31.59%, this is in line with findings from Ritter (1991), whose average return was 34% after the first 3 trading years. The spinoff variable shows the percentage of spinoff firms in the sample: 48.2%. Firmsize is interpreted as the logarithm of the share capital at the issuing date. Firmsize is divided into two groups to compare the mean share capital of spinoff firms and non-spinoff firms. The dummies for issuing year and industry are omitted from this table for simplicity. Table 3: descriptive statistics

Variable Mean Std. Error Std. Dev. 95% Confidence interval

BHR .3259 .1087 1,2254 .1000 - .5317 Spinoff .4821 .0515 0 .3819 - .5866 Firmsize Spinoff Yes 17.011 .410 2,7808 16.185 - 17.8368 No 19.285 .196 1,3724 18.8909 - 19.6792 4.2 Regressions Table 4 shows the correlation matrix, (the industry and year dummy variables are omitted for simplicity). The correlation between the buy and hold returns and the dummy spinoff shows that spinoffs have some negative relation with the buy and hold returns. This table shows no strong correlation of more than 0.7.

Table 4: correlation matrix BHR Firmsize Spinoff BHR 1 Firmsize 0,0173 1 Spinoff -0,135 -0,4676 1 Table 5 compares the 3-year buy and hold returns of the subsidiaries and the benchmark firms. Following this table, we see that the mean return of subsidiaries is 16.9% and the benchmark return is 45.4%. When comparing the mean differences, subsidiaries are outperformed by 28.4%. However, the outcome of the t-test shows no significant difference (p= 0.1921, t= 1.3141) between the two samples. Table 5: t-test on BHAR

Group Obs Mean Std. Error 95% confidence interval

Spinoff Yes 49 .4538 .1751 .1018 - .8057 No 46 .1690 .1234 -.0796 - .4176 Difference .2167 -.1456 - .5317 Difference mean 1 - mean 2 t = 1.3141

H0: Diff = 0 Degrees of freedom = 93

H1: Diff ¹ 0 Pr(|T|>|t|) 0.1921 Furthermore, to account for other factors that could have effect on the buy and hold returns, control variables are added to the regression. Results of this multiple regression are below in table 6. The R-square of this model is 0.196, which means that 19.6% of the variation of the BHR is explained by the model. This is considered as very weak effect but the model does show an increase in R-square due to adding the control variables.

The variable spinoff shows no significant effect on performance (p= 0.132). This is in line with earlier findings from table 5, which implies that being spun off before has no effect on the buy and hold returns on the first three years after an IPO. For firm size, year dummies and industry dummies no significant effects on the buy and hold returns are found. Table 6: regression on buy and hold returns Variables BHR Spinoff -.4144 (.132) Firmsize .0262 Dummies: (.639) Year included Industry included Constant -.1614 (.901) Observations 95 R-squared .1960 P-values in parentheses; * p<0.05

5. Conclusion

In this chapter, the research results are summarized and the research question is answered. In the second part of this chapter, the limitations of this research are discussed and recommendations for further research are provided.

5.1 Conclusion The focus of this thesis was to research whether subsidiaries perform different than non-spun off firms in the long run. In line with earlier research of Ritter (1991) the time horizon for the long-term performance was three years. Furthermore, this thesis concentrates on 95

IPOs of which 46 were from subsidiaries, all in the United States from 2009 till 2017. This thesis builds on existing literature about IPO performance and spin-offs. From the literature, an ambiguous effect was expected because arguments could be made for both positive as negative effects. The findings of thesis show the effect is in line with the literature, there was no significant difference in returns of firms that where spun-off beforehand. Secondly, to dig deeper into potential reasons for abnormal returns of subsidiaries, a regression was run with control variables firm size and dummies for industry and year of the IPO. Spinoff firms showed an abnormal return of -41.8%. However, this finding was not significant. It can be concluded that spinoff firms perform as well as non-spinoff firms in the first three years after their initial public offering. From an investor’s point of view could be concluded that is does not matter is what kind of IPO firm is invested in. Both firms that where spun-off on beforehand and firms that were not show the same long-term returns. However, earlier research from among Ruud (1993) showed that IPO firms have below average long-term performance compared to non-IPO firms so investing in IPO firms might only be for less risk averse investors. 5.2 Limitations & recommendation There are some limitations to this study. The first one is the sample size of 95. Due to this sample size, some of the standard errors of variables were relatively high. Therefore, results are harder to get statistically significant. Another limitation is the focus on American firms. American law prescribes that if a firm exceeds 500 employees, it needs to go public. This law could influence the returns of firms that were private beforehand. This thesis did not take this law into account so, findings in this thesis might not be external valid to IPO markets outside the United States. The last limitation to this study is the use of buy and hold abnormal return. Because compounded returns are used, one year with extreme results could cause bias in the results. However, alternatives to the BHAR method also have some shortcomings such as the upward bias when using the CAR method. Further research is needed in this topic because there is not much literature about the performance of subsidiaries. It would be interesting to research how subsidiaries perform in Europe because in Europe, there is no law that forces companies to go public.

This could lead to a better performance of the firms that were private before the IPO because firms only go public when they expect to perform well. Another suggestion for further research is the use of bigger sample size and more control variables to get a more powerful result.

Reference list

Akerlof, G. (1970). The Market for "Lemons": Quality Uncertainty and the Market Mechanism. The Quarterly Journal of Economics, 84(3), 488-498. Quarterly Journal of Economics, 84(3), 488-500. Brealey, R. A., Marcus, A. J., & Myers, S. C. (2001). Fundamentals of Corporate Finance. Chan, K., Ikenberry, D., Lee, I., 2004. Economic sources of gain in stock repurchases. Journal of Financial and Quantitative Analysis 39, 461–479 Chemmanur, T. J. (1993). The Pricing of Initial Public Offerings: A Dynamic Model with Information Production. The Journal of Finance, 48(1), 285. Chow, Y. K., Hamilton, R. T. (1993). Why Managers Divest – Evidence from New Zealand’s Largest Companies. Strategic Management Journal, 14(6), 479-484. Draho, J. (1971). The IPO Decision: Why and how Companies Go Public. Eckbo, E., & Thorburn, K. (2013). Corporate Restructuring. Working paper. Fama, E. F. (1997). Market Efficiency, Long-Term Returns, and Behavioral Finance. SSRN Electronic Journal, 49(3), 283-306. Ferrari NV. (2015). Ferrari NVv2015 Annual Report. Retrieved from https://corporate.ferrari.com/sites/ferrari15ipo/files/ferrari_nv_- _2015_annual_report_feb_25_final.pdf Hite, G.L. & Owers, J.E. (1983). Security price reactions around corporate spin-off announcements. Journal of Financial Economics 12: 409–436. Ibbotson, R. G., Sindelar, J. L., & Ritter, J. R. (1994). The market's problems with the pricing of Initial Public Offerings. Journal of Applied Corporate Finance, 7(1), 65-74. Ibbotson, R. G. (1975). Price performance of common stock new issues. Journal of Financial Economics, 2(3), 235-270. Ikenberry, D., Lakonishok, J., & Vermaelen, T. (1995). Market underreaction to open market share repurchases. Journal of Financial Economics, 39, 185-190 Jog, V., & Riding, A. (1987). Underpricing in Canadian IPOs. Financial Analysts Journal, 43(6), 48-55. Levis, M. (1993). The Long-Run Performance of Initial Public Offerings: The UK Experience 1980-1988. Financial Management, 22(1), 28.

MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of economic literature, 35(1), 13-39. Pagano, M., Panetta, F., & Zingales, L. (1998). Why do companies go public? An empirical analysis. The Journal of Finance, 53(1), 27-64. Ritter, J. R. (1991). The Long-Run Performance of Initial Public Offerings. The Journal of Finance, 46(1), 3-27. Rock, K. (1986). Why new issues are underpriced. Journal of Financial Economics, 15(1-2), 187-212. Rudd, J.S. (1993). “Underwriter Price Support and the IPO Underpricing Puzzle”. Journal of Financial Economics, 34 (1), 135-148. Shiller, R. J. (1988). Initial Public Offerings: Investor Behavior and Underpricing. Working Paper No 2806, National Bureau of Economic Research, Cambridge

Appendix

Anova and regression output provided by stata. Anova regressie Source SS df MS Number of obs = 95 F (17,77) = 1.1 Model 20,6898 17 1,217 Prob > F = 0.3654 Residual 84,8413 77 1,102 R - squared = 0.1960 adj R - squared = 0.0185 Total 105,5411 94 11.223 Root MSE = 1.0497 Regression output, Dummy variables included. BHR Coef. Std. Err. t P > |t| Spinoff -.4177 .2741 -1.52 0.132 Firmsize .0262 .0556 0.47 0.639 Year 2010 .3031 .8010 0.38 0.706 2011 1,5571 .8048 1.93 0.057 2012 .5524 .7229 0.76 0.447 2013 .0405 .7661 0.05 0.958 2014 .1231 .6997 0.18 0.861 2015 .4412 .6951 0.63 0.528 2016 .8382 .7053 1.19 0.238 Industry Construction -.7952 .9473 -0.84 0.404 Finance .0089 .7445 0.01 0.991 Manufacturing -.4245 .7383 -0.57 0.456 Mining -1.6416 .8594 -1.91 0.060 Retail trade -.2803 .7450 -0.38 0.708 Service -.2626 .6791 -0.39 0.700 Transportation .2121 .7698 0.28 0.784 Wholesale trade -3377 .7710 -0.44 0.663 Constant -1613 1.2980 -0.12 0.901

Two-sided T-test, used for testing whether the mean returns between spinoff and non-spinoff firms are different.

Group Obs Mean Std. Error 95% confidence interval

Spinoff Yes 49 .4538 .1751 .1018 - .8057 No 46 .1690 .1234 -.0796 - .4176 Difference .2167 -.1456 - .5317 Difference mean 1 - mean 2 t = 1.3141

H0: Diff = 0 Degrees of freedom = 93

H1: Diff = 0