MERGERS AND ACQUISITIONS

MERGERS AND ACQUISITIONS Free Riding and Large Shareholders (The Grossman and Hart, 1980, and Schleifer and Vishny, 1986, models) Master in Finance (also Master in Accounting, in Economics and Management of Cities and in International Business as elective course) • Summary 1. Takeovers and Free Riding (Grossman and Hart, 1980 model) 2. Takeovers and Large Shareholders (Schleifer and Vishny, 1986 model) 2 Takeovers and Free Riding Takeovers and Large Shareholders • Assumption: Shareholder structure of target is widely dispersed, in the sense that a large number of shareholders exist, each one of them having a very small stake which is not able to exert any meaningful individual influence on the firm or market price • Let us consider the following: A  Space of all possible actions taken by managers f (a)  Firm market value as a function of action a  A, as the result of managerial "ability", implying that the maximum possible firm value will be max f (a ) , where aA f (a*)  max f (a) aA p  price offered in a takeover ("tender price") by a raider, p  f (a*) for the acquisition of all oustanding shares 3 Takeovers and Free Riding Takeovers and Large Shareholders • Suppose that a raider intends to acquire the company, possessing different managerial ability than the current management, with the possibility of introducing firm value improvements Z>0 such that v  max f (a)  Z a A • Suppose also that Z and v are well known by all shareholders 4 Takeovers • If v > p > f(a*), the offer will be attractive but it will not be successful because according to rational expectations, each shareholder that knows about the offer will prefer to earn v rather than just p! and Free Riding Takeovers and Large Shareholders • Therefore, nobody will make any bid for the shares as it would be necessary that p  v , which would translate into a loss for the raider (particularly in the presence of relevant transaction costs such as fees paid to investment bankers or research costs). • Thus, there is no incentive to launch a takeover bid! 5 • When could then a takeover take place? • A possibility could be in the presence of asymmetric information Takeovers and Free Riding Takeovers and Large Shareholders • In that case if the firm value perception by shareholder were v s  v and assuming takeover transaction costs of c, the raider’s profit will be  r  v  vS  c • Another possibility would be the dilution of minority shareholders’s rights by the raider by a factor of   v  vS 6 Takeovers and Free Riding Takeovers and Large Shareholders • Note that if the current management brings a market value of q, shareholders will not be willing to receive less than q and of course they will not also accept less than vS  v   • Thus, in order to maximize the raider’s profit, we will need to have : p  max(v S , q)  max(v   , q) • And, accordingly, the raiders’ profit will be:  r  v  p  c  min( , v  q)  c 7 Takeovers • Note that the choice of current managerial action a0 on the part of current managers that corresponds to market value q=f(a0) will depend on the level of dilution  and Free Riding Takeovers and Large Shareholders 8 Takeovers • Suppose that the current manager has an utility U(q) which is a function of market value q in the case when takeovers do not take place and an utility of zero if the takeover happens and Free Riding Takeovers and Large Shareholders • Being v and c random variables in what concerns the manager (we´ll call these v and c ), let us consider the probability of a takeover taking place:  ( , q)  Pr min( , v~  q)  c~ 9 • The expected utility for the manager arising from a market value of q will be Takeovers and Free W (q)  U (q) 1   ( , q) Riding Takeovers and Large Shareholders • The randomness of v and c is essential in order to conclude that takeovers will take place. • Without such randomness, takeovers will occur with a likelihood of 1 or 0, and since  ( , q) is a decreasing function of q, the manager would choose a sufficiently high q so that the takeover will not actually take place 10 • How will shareholders choose the desired level of dilution? Takeovers and Free Riding Takeovers and Large Shareholders • Let us consider the expected return for shareholders r( ) r ( )  q 1   ( , q)  E max(v~   , q) min( , v~  q)  c~  ( , q) • Note that the impact of an increase in  on return is ambiguous since the price offered in the takeover will be reduced (which is bad for returns) but the likelihood of the takeover taking place will increase as well as the current market value q (which is good for returns) 11 • Grossman e Hart (1980) demonstrate that if shareholders know the takeover cost c, and if they determine that   c , the raider will see this cost covered in full Takeovers and Free Riding Takeovers and Large • Therefore, in a model of shareholder dispersion, the free-riding problem is overcome through a dilution mechanism! Shareholders 12 • Alternatives for solving the free riding problem: • Two-Tier offers Takeovers and Free Riding Takeovers and Large Shareholders • Offer in cash in a first phase over a limited number of shares, with a substantial premium • If the raider achieves control in the first phase then in a second one a lower price is offered for the remaining shares and eventually a merger between target and acquirer follows (see Comment and Jarrell, 1985) • Note however that • Target shareholders may cooperate between themselves and refuse the offer • There may be competition with other acquirers or between the acquirer and current managers 13 Takeovers and Free Riding Takeovers and Large Shareholders • Suppose that, in the previous model, there is a large (but minority) shareholder (which we will designate by L) that owns a certain percentage of the shares  lower than 50%, ie 1 0  2 • By assumption, L could introduce a value improvement Z in the firm (for instance, because it has an access to better technologies); • Such improvement Z is extracted from a probability distribution F(Z) with values within an interval (0, Zmax) where I is the probability of a positive improvement Z • Such possibility occurs with an associated cost c(I) satisfying the conditions c´>0 and c´´>0 14 • The offer price p must be higher than q, that is, Takeovers and Free Riding Takeovers and Large Shareholders  p  q  • Where must respect the following condition for the takeover to be profitable for the raider (who will aim at achieving firm control by holding 50% of the shares and tendering for 50% -  : 1 Z  (0.5   )  c  0 2 • note that L may well bid a price higher than the true value after the takeover (why?) • For the small shareholders the gain following a takeover will be: E Z 0.5 Z  (0.5   )  c  0 15 • Under such conditions, the best strategy is to accept the takeover as long as:   E Z Z  (1  2 )  2c  0  0 Takeovers and Free Riding Takeovers and Large Shareholders • The role of L will be to determine the premium  that will minimize the left hand of the previous equation. Note that : • the optimal value which we will term  * ( ) depends on  , and on the expected value of the improvement contingent on the existence of a profitable takeover • Shleifer and Vishny (1986) prove that such value  * ( ) is unique and is a decreasing function of  • If   0 , then we end up in the widespread shareholder dispersion model where Z   , and no-one will launch a takeover 16 • Shleifer and Vishny (1986) show that the larger the proportion of shares owned by L, Takeovers and Free Riding Takeovers and Large Shareholders • The more L will be willing to pay (eg., in research expenses) to increase the likelihood of a positive Z following a takeover, being I * ( ) the optimal probability of extracting a Z>0, which increases with  • The lower will be the premium that L will be willing to pay to the remaining shareholders • The lower will be the minimum improvement Z that will be needed for an indifference situation regarding the launching of a takeover, i.e., Z c ( ) • The greater the likelihood of a takeover 17 • Impact of an increase in  on the market value of a firm: V ( , q)  q  I * ( ) 1  F Z c ( ) E Z Z  Z c ( ) o Takeovers and Free Riding Takeovers and Large Shareholders t • In other words, the value of the firm increases as a function of the product of three factors (and each of these in turn will depend on  ): • The optimal likelihood of achieving a Z>0, i.e., I * ( ) • The expected value of the improvement in the market value of the firm by substituting an inefficient management team (contingent on the investment being a profitable one by exceeding a certain critical c threshold, i.e., E Z Z  Z ( ) ) • The likelihood that such improvement exceeds the c critical threshold , ie 1  F Z ( ) • Shleifer e Vishny (1986) prove that the value of the firm as above increases with  18 • Shleifer and Vishny (1986) propositions: Takeovers and Free Riding • “An increase in the proportion of shares owned by L will decrease the takeover premium but at the same time will increase the market value of the firm” Takeovers and Large Shareholders • “An increase in the takeover cost c will increase the premium paid but at the same time will reduce the market value of the firm” 19 • Alternatives to a managerial change through a takeover :  will not change but the profit for L will now be • Proxy fights (with a cost cp<c) Takeovers and Free Riding Takeovers and Large Shareholders • Z  c p • Jawboning (friendly negotiations) • L cannot impose a radical change but only a partial one • Being 0    1 the takeover will be a preferable solution only if 0.5Z  (0,5   )  c   Z  0 or 0.5   1  Z c 0.5   0.5   20 • Jawboning might be preferable in some cases when • Z is sufficiently low Takeovers and Free Riding • Cost c is sufficiently high Takeovers and Large Shareholders •  is sufficiently high 21