Divestment Options under Tacit and Incomplete Information

Transcription

1 Divestment Options under Tacit and Incomplete Information Qing Ma and Susheng Wang 1 Department of Economics Hong Kong University of Science and Technology July 2016 Abstract: While the literature has mainly focused on why a firm decides to divest a subsidiary, we investigate theoretically what the best divestment option is for a firm to divest a subsidiary. The firm chooses among the four most popular divestment options in practice: sell-offs, spinoffs, carve-outs, and management buyouts. In an infinite-period growth model, where divestiture is completed in the first two periods, the owners of a parent firm divest a subsidiary for the best value. The information possessed by the owners, subsidiary managers and outside buyers about the subsidiary s profitability may be incomplete and this information may be explicit or tacit (the nature of information). We investigate how the nature of information, the incompleteness of information, risk aversion and discount on future performance determine the best divestment option. Keywords: divestitures, tacit information, explicit information JEL classification: G34 1 Address for both: Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong. qmaaa@ust.hk and s.wang@ust.hk. Phone:

2 1. Introduction Corporate divestitures, especially management buyouts (MBOs), spin-offs and carve-outs, have gained much popularity recently. There is hence an increasing interest in academic research on corporate divestiture. This paper offers a theoretical analysis of the four most popular divestment options. For convenience, we refer to divestiture as when a firm divests one of its wholly owned subsidiaries. Divestiture is very different from sales of non-productive financial assets. Since the subsidiary being divested is a productive asset, the timing of sale, information and synergy with the acquirer are important considerations. In particular, information may play an important role. For example, a parent company may choose to divest a subsidiary via an MBO due to the subsidiary managers information advantage. Welch (1989) and Nanda (1991) find that a key characteristic of going-public decisions is that the owners of the firm have an informational advantage over outside buyers. Slovin et al. (1995) find that the owners opt for carve-outs if the subsidiary is overpriced in the market. According to Pham (2012), a parent firm s predivestiture level of asymmetric information is related to its choice of divesting via a sell-off. Schipper and Smith (1986) attribute a gain in price of the parent stock to the parent s benefiting from more publicly available information about the subsidiary after a carve-out. Further to prior literature on the role of incomplete information, we argue that the nature of information (explicit versus tacit) should also be considered. This is particularly so in our model in which divestiture is a staged divestment of a subsidiary over two periods. People need time and experience to digest information, just as it takes many hours of practice to really learn to swim. When implications from a piece of information are immediately known to an economic agent, we say that this information is explicit to the agent; when the implications take time (one period in our model) to become known to the agent, we say that this information is tacit to the agent. The firm owners, subsidiary managers, and outside buyers are all economic agents in our model. Divestment options affect the revelation of profitability information to economic agents. This information may be symmetric or asymmetric, and it may be explicit or tacit. To our knowledge, we are the first to develop a theory of divestiture that takes into account the nature of information. Prior literature on divestiture does provide some empirical evidence regarding the nature of information. For example, Bergh and Lim (2008) find that experience with corporate restructuring (sell-offs, spin-offs) influence subsequent restructuring and financial performance. They find that cumulative and repetitive experience with sell-offs is related to the subsequent adoption of sell-offs and to better financial performance, while short-term and contemporaneous experience with spin-offs is related to the subsequent use of spin-offs and to better financial performance. This finding suggests that divestiture information gleaned from sell-offs is tacit while that from spin-offs is explicit. 2/24

3 We develop an infinitely repeated growth model in which the owners, subsidiary managers, and outside buyers may have incomplete and tacit information about the subsidiary s profitability. In our model, private information is not given per se; in some cases there is no private information, but asymmetric information can result from the activity of divestiture depending on whether the divestiture is public or private and whether the information is tacit or explicit. Information can be asymmetric due to the choice of divestment option under tacit information. Divestiture can be considered as a two-stage process. In the first stage, the parent company decides whether or not to divest a subsidiary. In the second stage, the parent company evaluates available divestment options and then selects the best one. Much of the prior literature on corporate divestiture focuses on the first stage and the main concern is usually whether the divestiture creates value. Some empirical studies take into account the difference among divestment options and identify its impact on empirical results. However, these studies often cannot explain why a parent company chooses a particular divestment option. We address the second stage the selection of divestment options, taking as given the decision to divest a subsidiary. Once the firm has made the decision to divest its subsidiary, the owners assess the four most popular divestment options sell-offs, spin-offs, carve-outs, and MBOs and choose the best one. That is, while the literature focuses on why a firm divests a subsidiary, our study focuses on how the firm divests the subsidiary. As Eckbo and Thorburn (2008) have elaborated, In this survey, we have focused on the individual transactions and their associated empirical evidence. This is also how most of the literature progresses. A major drawback of this approach is the resulting lack of analysis of alternatives. That is, when a company self-selects a divestiture, what were reasonable alternative strategies? In what sense was divestiture superior to, say, a spinoff or an equity carveout? Ideally, one would use a theoretical model to structure the answers to these types of questions. Perhaps the greatest challenge to the restructuring literature is to achieve a modicum of integration of the analysis across transaction types. We expect these issues to be resolved as both theories and data become more readily available in the future. Empirical comparisons of divestment options abound in the literature (Michaely and Shaw, 1995; Slovin et al., 1995; Maydew et al., 1999; Nixon et al., 2000; Frank and Harden, 2001; Powers, 2001; Chen and Guo, 2005; Bergh et al., 2008; Bergh and Lim, 2008; Damaraju, 2008; Jain et al., 2011; Bergh and Sharp, 2012; Pham, 2012), but theoretical comparisons are few and far between (Khan and Mehta, 1996; Choi and Merville, 1998; Chemmanur and Liu, 2011). Empirical comparisons are restricted to observable and measurable factors, while theoretical comparisons are not. Factors such as risk aversion, incentives and asymmetric information are hard to observe and measure in empirical studies. Theoretical comparisons generally focus on the asymmetry of information between the owners and outside buyers. Choi and Merville (1998) investigate corporate acquisitions and divestitures in a unified agency frame- 3/24

4 work where an agency parameter and a synergy parameter interact to determine the optimal organizational structure. They assume two separate production processes for the parent firm and a subsidiary, where the subsidiary s production process includes a synergy parameter. Chemmanur and Liu (2011) emphasize insiders private information about firm value. They find that insiders with the most favorable private information implement spin-offs; those with less favorable private information implement carve-outs; those with even less favorable private information implement tracking stock issues; and those with unfavorable private information remain consolidated. One crucial difference between our model and Chemmanur and Liu s is that they assume that the owners have perfect knowledge of the subsidiary s profitability while we do not. In fact, we think that one of the reasons for the owners to divest a subsidiary is so that they could discover the subsidiary s market value. This idea is consistent with that of Perotti and Rossetto (2007) who model carve-outs as a way for the parent firm to obtain information from the market about the value of the subsidiary. Bergh and Lim (2008) discuss the role of explicit and tacit information in their empirical study. For them, explicit and tacit information about the operating procedures of divestment options derives from past divesting experiences and is indirectly associated with the adoption of a particular divestment option. In our model, explicit and tacit information about the subsidiary s profitability is exogenous and is directly related to the choice of divestment option. Our study is the first to consider all four popular options. In our model, it takes two periods to complete a divestiture and the value of the subsidiary is based on the expected value of future incomes after separation. We assume that the subsidiary managers have complete information about the subsidiary s profitability while the owners and outside buyers may have incomplete information. In addition, we assume that information about the subsidiary s profitability may be explicit or tacit to the owners and outside buyers at the time of divestment. Further, although prior literature such as Chemmanur and Liu (2011) often assumes risk neutrality for the parties involved, risk aversion plays a role in the choice of divestment option in our model. Our model setting supports the view (Cusatis et al., 1993; Allen, 2001) that firms choose public divestitures (spin-offs and carve-outs) to establish the market value of a subsidiary prior to selling it off to an outside firm or to the public. It also supports the situation where the owners want to find out about the market value of the subsidiary first before taking the next step; in particular, if information is tacit, it may take the owners some time to discover the subsidiary s market value. This delay in discovering the information may affect the owners choice of divestment option. We identify the owners best divestment option from four of the most popular divestment options sell-offs, spin-offs, carve-outs, and MBOs depending on the nature of information, the incompleteness of information, the risk attitude, and the time discount on future income. We find that 4/24

5 Selling off a subsidiary privately to a single outside buyer is always better than selling it off publicly. Expected profit growth has no effect on the owners choice of divestment option. Sell-offs are likely to be the best divestment option if synergy between the subsidiary and outside buyers is sufficiently large; or if outside buyers are substantially less risk averse than the owners; or if information is tacit and the owners are highly risk averse. Spin-offs are likely to be the best divestment option if the owners are not very risk averse; or if information is explicit; or if information is tacit and the owners have complete information while outside buyers do not. Carve-outs are likely to be the best divestment option if information is explicit, the owners and outside buyers are equally risk averse, and the subsidiary s synergy with acquirers is sufficiently small; or if information is tacit and outside buyers are much less risk averse than the owners. MBOs are more likely to be the best divestment option if the subsidiary managers are less risk averse; MBOs are better than spin-offs if information is tacit and all parties are equally risk averse or if information is explicit and future discount is small; MBOs are better than sell-offs if all parties are equally risk averse. These findings complement prior literature on corporate divestiture by taking into account the nature of information, the incompleteness of information, risk aversion, and discount on future profits. This paper proceeds as follows. Section 2 presents the model. Section 3 derives the payoffs from divesting a subsidiary by the four divestment options sell-offs, spin-offs, carve-outs, and MBOs under tacit and incomplete information. Section 4 derives the best divestment option under various conditions. Section 6 concludes the paper. 2. The Model Sequence of Events Consider a firm that has decided to divest its subsidiary. Time is discrete and goes all the way from to infinity. Divestiture is initiated at time and completed at time, except for sell-offs and MBOs which are initiated and completed at time Three parties are involved: the owners (the seller), the subsidiary managers (the inside buyer), and outside buyers, who are indicated respectively by subscripts and in relevant variables. 5/24

6 We have an infinite-period growth model, in which the owners can divest the subsidiary at time or or both. That is, the owners can sell the subsidiary right away at time or separate it first at time manage it for one period, and then sell it at time ; or sell a portion of it at time and sell the rest at time More specifically, we consider four options for the owners to divest the subsidiary: Option 1 (sell-off): Sell the subsidiary in whole at time to another firm. Option 2 (spin-off): Separate it from the parent company at time, manage it for one period, and then sell it in whole at time. Option 3 (carve-out): Sell a portion of it to the public in shares, manage the rest for one period, and then sell that too to the public at time. Options 4 (MBO): Sell it to the subsidiary managers at time We want to find out under what circumstances the owners will choose a particular divestment option. Separation, Initial Sale Manage the Subsidiary Completion of Divestiture 0 1 Figure 1. Timeline of Divestment. Performance Let the profit of a subsidiary that becomes an independent firm after separation be at time This profit includes all possible effects of separation, such as potential improvements in incentives and the management of the subsidiary. Assume that besides a fixed growth component, the subsidiary s profit is subject by a white-noise random shock. Specifically, its profit at is where is the expected growth rate, is a random shock at time and is the expected profit of the subsidiary. Assume that is a white noise, with and Further, if the subsidiary is acquired by a company, it may provide additional value such as synergy to the acquirer. If so, its expected value may be higher. Hence, if the subsidiary is acquired by a company, we assume that its expected profit is with where stands for synergy. Specifically, if the subsidiary is acquired by an outside company, instead of becoming an independent firm, its profit at time is 6/24

7 (2) Information However, the owners and outside buyers may not have complete information about the subsidiary s profitability. We use the Bayesian approach to model this incomplete information, by which economic agents form beliefs when they face incomplete information. If the owners have incomplete information, they will form the belief that the subsidiary s profit is (3) where represents the extra uncertainty about the subsidiary s profitability after separation due to incomplete information. Similarly, if outside buyers have incomplete information, they will form the belief that the profit is (4) where represents the extra uncertainty about the subsidiary s profitability due to incomplete information. We assume that the subsidiary managers have complete information and know the distribution of perfectly, i.e.,. 2 The notion of uncertainty assumes the complete knowledge of the distribution function of the random profit, while incomplete information refers to knowledge about the distribution function possessed by economic agents that is incomplete. We assume that and have zero mean, implying that individual assessment of incomplete information is unbiased. The Bayesian approach imposes restrictions on beliefs. Requiring unbiased beliefs is a weak version of Bayesian consistency. We further assume that the expected profit and the expected growth rate are common knowledge, and and are independent variables with for and where and are known as variance risk in the literature on preference ambiguity. If agent has complete information, we have otherwise. There are two types of divestiture: private and public divestitures. Sell-offs are mostly private divestitures in practice; we hence treat them as such in our model. MBOs are also private divestitures. In contrast, spin-offs and carve-outs are public divestitures. Public divestitures create public companies (publicly traded companies). We assume that public divestitures reveal profitability information while private divestitures do not. This assumption is supported by Bergh et al. s (2008) empirical study. Comparing spin-offs and sell-offs, they find that spin-offs most effectively and profitably reduce information asymmetries and increase the 2 We can easily extend our model to allow the subsidiary managers to have incomplete information. The subsidiary managers incomplete information would then be represented by a random variable with and 7/24

8 transparency of the parent firm, while sell-offs best mitigate asymmetries in productive assets across industries. There are two types of information: explicit and tacit. We refer this as the nature of information. If information is explicit, it is immediately understood; but if information is tacit, it takes time (one period in our model) for economic agents to understand. We assume that divestiture results in the complete revelation of information at time but at time, depending on the type of divestiture and the nature of information, information may be incomplete even after divestiture. The importance of the nature of information in practice is supported by Bergh and Lim s (2008) empirical findings as mentioned earlier. In our model, divestiture may reveal profitability information, depending on the type of divestiture and the nature of information. A public divestiture at time reveals profitability information to the public, while a private divestiture does not. This information may be explicit or tacit. If the information is tacit, an agent will need one period to find out about the distribution function of profit upon observing a public divestiture. If information is explicit, the agent will immediately know the distribution function of profit at time It is well known among practitioners and empirical researchers alike that public divestitures help analysts and investors better understand the value of the divestitures. For example, Chemmanur and Liu (2011) confirm that spin-offs and carve-outs play a role in information revelation. We assume that the nature of information is the same for all agents. For example, if profitability information of a subsidiary revealed through a spin-off is explicit to outside buyers, it is also explicit to the owners. We also assume that the nature of information is the same across different divestitures. For example, if profitability information from spin-offs is tacit, then that from carve-outs is also tacit. We further assume that the owners divesting plan is made and the divestiture is carried out at the same precise moment of time. Hence, profitability information from a public divestiture is useful to the owners in making the divesting plan at time in particular, if profitability information is explicit, then information becomes complete at time so that Information may be asymmetric at time when divestiture occurs. Depending on the type of divestiture and the nature of information, the trading parties may have asymmetric information about the subsidiary s profitability at the time of divestiture. Each of the four options of divestiture has its own unique features. Sell-offs offer synergy to outside buyers; spin-offs reveal profitability information; carve-outs balance between early and late sales; and MBOs allow the subsidiary managers to profit from their information advantage. 8/24

9 Payoff Let be the utility function of agent, where utility functions of the form. We will use mean-variance (5) where is an information set, represents economic agent s risk aversion, is the expected value of conditional on information set, and is the variance of conditional on information set, for agent and We often shorten to for convenience. At time let be the set of information available at time. Assume and for all Across periods, utility values are discounted by a common discount factor 3. Divestment Options After a decision to divest a subsidiary is made, the owners focus their attention on choosing a divestment option. There are four popular divestment options in practice: sell-offs, spinoffs, carve-outs, and MBOs. We evaluate each of them in this section Sell-Offs In practice, sell-offs involve subsidiaries being acquired by other companies. Hence, we assume that for a sell-off, the subsidiary is sold off directly to another firm (an outside buyer) in a private sale at time. Assume that outside buyers are perfectly competitive. Then, the sale price of the subsidiary is defined by (6) Here at time when the buyers are to pay for the subsidiary, they not only face uncertainty about profitability but also have incomplete information about profitability. This incompleteness of information affects the buyers assessment at time of current and future profits. The pricing formula (6) is based on the assumption that the demand is perfectly elastic. This is a typical assumption in asset pricing. The justification is that the subsidiary is a small competitive firm in the stock market. Since sell-offs involve subsidiaries being acquired by other companies, we will use the profit process in (2). By the mean-variance utility function in (5), (6) becomes 9/24

10 At time, the owners receive. Hence, the payoff from selling off the subsidiary is (7) where the following formula is used: 3.2. Spin-Offs For a spin-off, the entire ownership of the subsidiary is distributed pro rata to the owners at time. Assume that after this distribution the owners hold these shares for one period and then these shares become tradable in the stock market at time. We normalize the total number of shares to At time, since information is complete, between the public and the subsidiary managers, the owners will choose the public to sell their shares to since outside buyers are competitive but the subsidiary managers are monopsonists. Given a realized value of time is the selling price of the subsidiary to the public at (8) Since at time information is complete to the buyers, we set. Given a fixed, at time the buyers have complete information but are still uncertain about profitability Since spin-offs usually become independent firms instead of being acquired by other companies (Desai and Jain, 1999; Allen, 2001; McConnell et al., 2001), we use the profit process in (1). Then, (8) becomes However, at time besides uncertainty, the owners may have incomplete information. Denote the time- price at time by. By (9), this price to the owners is 10/24

11 The owners receive income from the subsidiary at time and a payment for the subsidiary from the public at time Hence, the expected income to the owners at time satisfies (11) where the value of is at time We have (12) 3.3. Carve-Outs For a carve-out, a fraction of the subsidiary s shares is issued to the public at time. Let be the fraction of shares sold to the stock market at time where. Here, as a partial sale, a carve-out is a mixed strategy of early and late sales. Since carve-outs typically involve subsidiaries becoming independent firms after separation instead of being acquired by other companies (Vijh, 2002; Hulburt, 2003; Otsubo, 2009), we use the profit process in (1). According to (10), Besides the income from selling or holding shares, the owners also receive a share of the subsidiary s profit from their holding of shares. Then the owners problem is Here is known to the public at time is The first-order condition (FOC) of the above problem implying With the profit process in (1), we have in the pricing formula of. Then, by (7), 11/24

12 Then, We always have But it is possible to have Hence, with condition the optimal solution is (13) When the owners expected income at time is determined by where the value of is at time The owners expected income is thus Using the FOC, we have 3.4. Management Buyouts For an MBO, the subsidiary is purchased privately by its managers at time. This is a bilateral trade, where the owners as a group are monopoly sellers and the subsidiary managers as a group are monopsony buyers. In this case, they bargain with each other and try to settle on the selling price. They do so under asymmetric information. The subsidiary managers can benefit from their information advantage, while the owners can reduce risk by selling the subsidiary early. Assume that they try to settle on the selling price by the Nash bargaining solution. If no settlement is reached, the owners keep the subsidiary and the owners welfare is equal to the value of the subsidiary: 12/24

13 The managers get nothing, implying that is the status quo social welfare. Here the owners may have incomplete information. But if a settlement is reached, the value of the subsidiary to the subsidiary managers is implying that the new social welfare is Here, as always, the subsidiary managers are assumed to have complete information at time i.e., Hence, according to the Nash bargaining solution, the payoffs of the owners and subsidiary managers are respectively These payoffs imply that the selling price is by which the subsidiary managers payoff is We have and which imply that the owners obtain the following value for the subsidiary: (14) 4. The Optimal Divestment Strategy In this section, we identify the optimal divestment options under various circumstances. Prior literature suggests that the choice of a divestment option depends on the actual circumstances. Bergh et al. (2008a) find that the influence of corporate restructuring on financial performance is determined in part through how the restructuring is implemented. Bergh and Lim s (2008b) empirical findings suggest that different kinds of restructuring experiences were associated with different modes of restructuring and performance records. We show that the owners choice of a divestment option depends on the nature of information, the incompleteness of information, risk aversion, and discount on future income. 13/24

14 As derived above, the owners payoffs from the four alternative divestment options are By comparing these payoffs, the owners decide on the best divestment option Sell-Off as the Best Option Under what conditions is a sell-off the best divestment option? A sell-off is better than an MBO if i.e., Since sell-offs and MBOs are private divestitures, the nature of information is irrelevant. If profitability information from public divestitures (spin-offs and carve-outs) is tacit, i.e., and for spin-offs and carve-outs at time, a sell-off is better than a carve-out if i.e., That is, if synergy with an outside buyer is large enough, a sell-off is better than a carve-out. Since a spin-off is a special case of a carve-out when, this condition also ensures that a sell-off is better than a spin-off. Hence, if profitability information is tacit, under conditions (15) and (16), a sell-off is the best divestment option. But if profitability information from public divestitures is explicit, i.e., for spin-offs and carve-outs at time, a sell-off is better than a carve-out or a spin-off if, i.e., Hence, if profitability information is explicit, under conditions (15) and (17), a sell-off is the best divestment option. 14/24

15 Conditions (16) and (17) have an interesting implication. In our derivation, we allow the subsidiary to be sold off to the public at time (with ), by which. Since conditions (16) and (17) require they imply that the parent company will sell off the subsidiary at time only if it is to a company, in which case synergy is possible (. If the subsidiary is not acquired by a company, there is no synergy ( in which case conditions (16) and (17) fail and a sell-off is not the best option. This result explains why in practice sell-offs typically involve subsidiaries being acquired by other companies instead of becoming separate firms (Hulburt, 2003; Otsubo, 2009; Thompson, 2010). Result 1. Selling off a subsidiary at time privately to a single outside buyer is always better than selling it off publicly. That is, a private sell-off is always better than a public selloff at time. Since sell-offs involve subsidiaries being acquired by outside buyers, potential synergy can substantially increase the value of a subsidiary. Indeed, if synergy is sufficiently large, conditions (15)-(17) are satisfied, indicating that a sell-off is the best option. Prior literature offers ample empirical support for this finding. For example, John and Ofek (1995) find evidence that some of the seller s gains result from a better fit between the divested asset and the buyer. If outside buyers are risk neutral ( since the price in (7) is fair (i.e. in line with the buyers expectation), it is efficient to sell the subsidiary off to such a buyer at time and let that buyer take all the risk. Indeed, if outside buyers are risk neutral, conditions (15)-(17) are satisfied and a sell-off is the best divestment option irrespective of the nature of information. Common sense suggests that the owners would likely choose an early sale if they are highly risk averse. Indeed, if the owners are sufficiently risk averse ( is large) and profitability information from divestiture is tacit, conditions (15) and (16) are satisfied and a sell-off is the best divestment option. Result 2. Sell-offs are likely to be the best divestment option if synergy between the subsidiary and outside buyers is sufficiently large; or outside buyers are substantially less risk averse than the owners, especially when outside buyers are risk neutral; or profitability information is tacit and the owners are highly risk averse. Khan and Mehta (1996) compare sell-offs with spin-offs. Their main theoretical and empirical finding is that a candidate-for-divestiture division with high operating risk is divested through spin-off and the one with low operating risk is divested through sell-off. This result is confirmed by our theory. Condition (17) is more likely to fail if is large, implying that a 15/24

16 spin-off is more likely to be a better divestment option than a sell-off if the subsidiary s profit is more uncertain. Moreover, the empirical results of Damaraju (2008) reveal that high information asymmetry about the true value of a business unit increases the probability of spin-offs or carveouts but not sell-offs. Our theory confirms this result. Condition (17) is more likely to fail if is large, implying that a spin-off or a carve-out is more likely to be a better divestment option than a sell-off if outside buyers have high incomplete information about the subsidiary s profitability Spin-Off as the Best Option i.e., If profitability information is tacit, by (13) a spin-off is better than a carve-out if (18) By (12) and (14), a spin-off is better than an MBO if i.e., (19) By (7) and (12), a spin-off is better than a sell-off if i.e., Hence, if profitability information is tacit, under conditions (18)-(20), a spin-off is the best divestment option. But if profitability information is explicit, by (18) a spin-off is better than a carve-out if i.e., (21) By (12) and (14), a spin-off is better than an MBO if i.e., By (7) and (12), a spin-off is better than a sell-off if i.e., Hence, if profitability information is explicit, under conditions (21)-(23), a spin-off is the best divestment option. 16/24

17 Since a spin-off is a late sale, it is preferred by owners who are not very risk averse. Indeed, independent of the nature of information, by conditions (18)-(23), a spin-off is the best option if the owners have low risk aversion. A spin-off gives the market one period of time to evaluate the subsidiary. 3 This is useful if profitability information is explicit. Indeed, by comparing (19) with (22) and (20) with (23), one will find that a spin-off is more likely to be the best option if profitability information is explicit. A natural conjecture is that when the owners have complete information about the subsidiary while outside buyers do not, if profitability information is tacit, a spin-off may be the optimal option. Indeed, conditions (18)-(20) are more likely to be satisfied if and. The explanation is that since outside buyers have incomplete information, they are not willing to pay a high price. This result is consistent with Krishnaswami and Subramaniam s (1999) main empirical finding that firms that engage in spin-offs have higher levels of information asymmetry compared to their industry and size matched counterparts and the information problems decrease significantly after the spin-off. With tacit information, it is better for the owners to take a wait and see approach. This is again consistent with Krishnaswami and Subramaniam s empirical finding that firms mitigate information asymmetry before approaching the capital market for funds. If discount on future income is small ( conditions (18)-(23) would all fail, implying that a spin-off is not a good option irrespective of the nature of information. The explanation is that, with a small discount on future income, the demand for the subsidiary would be high and buyers would be willing to offer a high price for the subsidiary. Hence, the owners should sell it early, instead of selling it late as a spin-off. If profitability information is explicit and outside buyers and the owners are equally risk averse ( then condition (23) would fail and a spin-off is not the best option. The reason is that when outside buyers can gain complete information from divestiture at time and they are willing to offer a higher price under complete information, a late sale is not an optimal option for the owners. Result 3. A spin-off is likely to be the best divestment option if the owners have low risk aversion; or 3 In practice, major parent shareholders generally do not sell their shares immediately after the distribution. They hold the stock for some time and then sell it when appropriate. Only disperse minority parent shareholders may sell their shares immediately. However, the fraction sold is very small relative to the fraction held by major parent shareholders. Therefore, we here ignore the selling behavior of disperse minority parent shareholders by assuming that parent shareholders hold their shares for one period. 17/24

18 profitability information is explicit; or profitability information is tacit and the owners have complete information while outside buyers do not. On the other hand, a spin-off is not the best divestment option if discount on future income is small; or profitability information is explicit, and outside buyers and the owners are equally risk averse. Riskier, more leveraged and less profitable firms are expected to be highly risk averse. Comparing spin-offs with carve-outs, Michaely and Shaw (1995) show empirically that such firms tend to divest their subsidiaries through spin-offs. Our theory confirms this. By (18) and (21), independent of the nature of information, if outside buyers are substantially more risk averse than the owners, divestitures are likely to be completed through spin-offs instead of carve-outs Carve-Out as the Best Option i.e., If profitability information is tacit, by (13) a carve-out is better than a spin-off if By (16), a carve-out is better than a sell-off if A carve-out is better than an MBO if i.e., Hence, if profitability information is tacit, under conditions (24)-(26), a carve-out is the best option. But if profitability information is explicit, by (21) a carve-out is better than a spin-off if i.e., By (17), a carve-out is better than a sell-off if 18/24

19 A carve-out is better than an MBO if, i.e., Hence, if profitability information is explicit, under conditions (27)-(29), a carve-out is the best option. Carve-outs let the owners find out about the potential of the subsidiary early, which may explain their popularity. Our theory indeed supports this argument. If profitability information is explicit, after selling a portion of the shares to the market, outside buyers and the owners will gain complete information ( If outside buyers and the owners are equally risk averse (, then by (27) a carve-out is better than a spin-off, by (29) a carveout is better than an MBO, and by (28) a carve-out is also better than a sell-off provided the subsidiary s synergy with acquirers is sufficiently small. A second advantage of carve-outs is that they allow the owners to diversify risk across periods. Hence, besides information revelation, risk aversion may be a reason for choosing carve-outs. Given tacit profitability information, if outside buyers are substantially less risk averse than the owners, especially when outside buyers are risk neutral, conditions (24)-(26) are satisfied, implying that a carve-out is the best option. On the other hand, if the owners are not very risk averse, the advantage in diversifying risk disappears so that a carve-out may no longer be the best option. Indeed, if the owners are risk neutral, conditions (24)-(25) and (27)- (28) indicate that a carve-out is not the best option. Differences in information may be another reason for choosing carve-outs. Given tacit profitability information, if outside buyers have complete information ( but the owners have rather incomplete information ( is large), conditions (24)-(26) are satisfied, implying that a carve-out is the best option. Result 4. A carve-out is likely to be the best divestment option if profitability information is explicit, outside buyers and the owners are equally risk averse, and the subsidiary s synergy with acquirers is sufficiently small; or profitability information is tacit and outside buyers are much less risk averse than the owners, especially when outside buyers are risk neutral; or profitability information is tacit and outside buyers have complete information but the owners have rather incomplete information. On the other hand, a carve-out is not the best option if the owners are risk neutral. 19/24

20 4.4. MBO as the Best Option MBOs are becoming popular recently. Prior literature has rarely discussed this divestment option. Our study considers this option and compares it with the other three popular divestment options. This makes our study a rare work indeed. By (15), an MBO is better than a sell-off if Since both MBOs and sell-offs are private divestitures, the nature of information is irrelevant. If profitability information is tacit, by (19) an MBO is better than a spin-off if By (26), an MBO is better than a carve-out if Hence, if profitability information is tacit, under conditions (30)-(32), an MBO is the best option. But if profitability information is explicit, by (22) an MBO is better than a spin-off if By (29), an MBO is better than a carve-out if Hence, if profitability information is explicit, under conditions (30), (33) and (34), an MBO is the best option. If profitability information is tacit and all parties are equally risk averse, due to the subsidiary managers information advantage, by (31) an MBO is better than a spin-off. However, by (33), if information is explicit, spin-offs have the advantage of revealing information, so that they are actually better than MBOs if the owners have substantial incomplete information. MBOs and sell-offs are both private divestitures, but the subsidiary managers have an information advantage over outside buyers. Hence, if risk aversion does not play a role (i.e. 20/24

21 assuming equal risk aversion, and outside buyers are much less knowledgeable about the subsidiary s profitability than the owners, by (30) MBOs are better than sell-offs. If future discount is small condition (33) is satisfied, i.e., MBOs are better than spin-offs if profitability information is explicit. The interpretation is that when future payoffs are the main consideration, with complete information, an early sale to the subsidiary managers (an MBO) is better than a late sale to the public (a spin-off). Intuitively, if the subsidiary managers are less risk averse (smaller, especially if they are risk neutral, MBOs are more likely to be the best choice. Indeed, conditions (30)-(34) are more likely to be satisfied if is smaller. One explanation for the fact that MBOs are becoming popular recently is that, as financial markets are becoming more complete, the risk appetite of subsidiary managers grows, and hence by our theory MBOs become the optimal divestment option. Since MBOs are private divestitures, they offer no informational benefit. With explicit information, a carve-out reveals the same amount of information as an MBO, but a carve-out has the advantage of asset diversification across periods. Hence, if risk aversion does not play a role (i.e. assuming equal risk aversion, when profitability information is explicit, an MBO is inferior to a carve-out, which is indeed confirmed by (34). Result 5. MBOs are better than spin-offs if profitability information is tacit and all parties are equally risk averse. MBOs are better than sell-offs if all parties are equally risk averse and outside buyers are much less knowledgeable about the subsidiary s profitability than the owners. MBOs are better than spin-offs if information is explicit and future discount is small. MBOs are more likely to be the best option if the subsidiary managers are less risk averse. On the other hand, MBOs are worse than spin-offs if profitability information is explicit and the owners have substantially incomplete information. MBOs are worse than carve-outs if information is explicit and all parties are equally risk averse. Finally, among all four options, the expected rate of profit growth is irrelevant. That is, expected profit growth has no effect on the owners choice of divestment option. Result 6. Expected profit growth has no effect on the owners choice of divestment option. 21/24

22 5. Concluding Remarks To our knowledge, we are the first to develop a theory of divestiture that takes into account the nature of information. We emphasize factors such as tacit information, risk aversion and discount on future performance, which have rarely been discussed in the divestiture literature. While the literature has mainly focused on why a firm decides to divest a subsidiary, we investigate a firm s reason for choosing a particular divestment option to divest a subsidiary. Once the firm has made the decision to divest one of its subsidiaries, the owners assess and choose the best divestment option from the four most popular divestment options: sell-offs, spin-offs, carve-outs, and MBOs. There are extensive empirical comparisons of divestment options in prior literature but only limited theoretical comparisons. We offer a theoretical study on how various divestment options weigh against each other. Theoretical comparisons are not restricted to observable and measurable factors as empirical comparisons are. Our study is the first to consider all four popular divestment options. MBOs in particular have rarely been discussed in prior literature. However, MBOs have grown in popularity recently. To our knowledge, we are the first to offer a theoretical analysis of MBOs and to compare them with other popular divestment options. In our model, it takes two periods to complete divestiture and the value of the subsidiary is based on the expected value of an infinite stream of future incomes after separation. A twoperiod divestment process allows us to discuss the impact of the nature of information on the choice of divestment option. References Allen, J.W., Private information and spin-off performance. Journal of Business, 74, Allen, J.W., McConnell, J.J., Equity carve-outs and managerial discretion. Journal of Finance, 53, /24

23 Bergh, D.D., Johnson, R.A., Dewitt, R.L., Restructuring through spin-off or sell-off: Transforming information asymmetries into financial gain. Strategic Management Journal, 29, Bergh, D.D., Lim, E.N.K., Learning how to restructure: Absorptive capacity and improvisational views of restructuring actions and performance. Strategic Management Journal, 29, Bergh, D.D., Sharp, B.M., How far do owners reach into the divestiture process? Blockholders and the choice between spin-off and sell-off. Journal of Management. Chemmanur, T.J., Liu, M.H., Institutional trading, information production, and the choice between spin-offs, carve-outs, and tracking stock issues. Journal of Corporate Finance, 17, Chen, H.L., Guo, R.J., On corporate divestiture. Review of Quantitative Finance and Accounting, 24, Choi, Y.K., Merville, L.J., A unified model of corporate acquisitions and divestitures: An incentive perspective. Review of Quantitative Finance and Accounting, 10, Cusatis, P.J., Miles, J.A., Woolridge, J.R., Restructuring through spin-offs: The stock market evidence. Journal of Financial Economics, 33, Damaraju, N.L., Why and how do firms divest? Dissertation. The Ohio State University. Desai, H., Jain, P.C., Firm performance and focus: Long-run stock market performance following spinoffs. Journal of Financial Economics, 54, Eckbo, B.E., Thorburn, K.S., Corporate restructuring: Breakups and LBOs. In Eckbo, B.E., Handbook of Corporate Finance, Vol.2, Chapter 16, , North-Holland. Frank, K.E., Harden, J.W., Corporate restructurings: A comparison of equity carveouts and spin-offs. Journal of Business Finance and Accounting, 28, X. Hulburt, H.M., Equity carve-outs and changes in corporate control. Journal of Applied Business Research, 19, Jain, B.A., Kini, O., Shenoy, J., Vertical divestitures through equity carve-outs and spin-offs: A product markets perspective. Journal of Financial Economics, 100, John, K., Ofek, E., Asset sales and increase in focus. Journal of Financial Economics, 37, Khan, A.Q., Mehta, D.R., Voluntary divestitures and the choice between sell-offs and spin-offs. Financial Review, 31, /24

24 Krishnaswami, S., Subramaniam, V., Information asymmetry, valuation, and the corporate spin-off decision. Journal of Financial Economics, 53, Maydew, E.L., Schipper, K., Vincent, L., The impact of taxes on the choice of divestiture method. Journal of Accounting and Economics, 28, McConnell, J.J., Ozbilgin, M., Wahal, S., Spin-offs, ex ante. Journal of Business, 74, Michaely, R., Shaw, W.H., The choice of going public: Spin-offs vs. carve-outs. Financial Management, 24, Nanda, V., On the good news in equity carve-outs. Journal of Finance, 46, Nixon, T.D., Roenfeldt, R.L., Sicherman, N.W., The choice between spin-offs and sell-offs. Review of Quantitative Finance and Accounting, 14, Otsubo, M., Gains from equity carve-outs and subsequent events. Journal of Business Research, 62, Perotti, E., Rossetto, S., Unlocking value: Equity carve outs as strategic real options. Journal of Corporate Finance, 13, Pham, D.A., Two essays on corporate restructuring. Graduate school theses and dissertations. University of South Florida. Powers, E.A., Spinoffs, selloffs and equity carveouts: An analysis of divestiture method choice. Working paper. University of South Carolina. Schipper, K., Smith, A., A comparison of equity carve-outs and seasoned equity offerings: Share price effects and corporate restructuring. Journal of Financial Economics, 15, Slovin, M.B., Sushka, M.E., Ferraro, S.R., A comparison of the information conveyed by equity carve-outs, spin-offs, and asset sell-offs. Journal of Financial Economics, 37, Thompson, T.H., Partial price adjustments and equity carve-outs. Financial Review, 45, Vijh, A.M., The positive announcement-period returns of equity carveouts: Asymmetric information or divestiture gains?. Journal of Business, 75, Welch, I., Seasoned offerings, imitation costs, and the underpricing of initial public offerings. Journal of Finance, 44, /24