When a vision for acquisitive growth overrides rational analysis, it may be vulnerable to cognitive drawbacks that can undermine objective decision-making. Thras Moraitis, Executive General Manager, Group Strategy and Corporate Affairs at Xstrata plc and Han Smit, a Criticaleye Thought Leader and Professor of Corporate Finance at Erasmus University, investigate how to reduce the risks when embarking on a journey of ambitious M&A

www.criticaleye.net 01 The strategy pursued by during the late 1990s and early 2000s was a classic example of global consolidation. After a successful marketing campaign had secured it a dominant position in its UK home market, its acquisitions in the US, Europe, China and Japan gave it a platform to create a global telecom brand.

Vodafone gained a strong position in the US when it outbid Bell Atlantic to buy The strategic option AirTouch for $62bn in January 1999. The merger created one of the world’s largest international mobile telecoms companies, perspective can promote with an extensive presence that covered most of Europe and the US. Vodafone rational decisions and considered a number of subsequent options in the US, including making a move to combine the US interests help identify takeover of Vodafone AirTouch and Bell Atlantic.

In 2000, the company strengthened its premiums for certain position in Europe when it launched a $205bn bid for Germany’s Mannesmann platform acquisitions telecoms group in a battle that remains one of the largest hostile takeovers in history. This move secured its place as a refusing to abandon bidding contests or to not in a conservative mode and framed key European operator, and created further divest assets. For instance, Vodafone was its strategy much more broadly, targeting acquisition opportunities in a European accused of excessive bidding in repeatedly multiple or parallel platform acquisitions in mobile market that was fragmented, raising its offer (up to a 70 per cent premium) the US, Europe, China and Japan to position but rapidly beginning to consolidate. to win its hostile takeover battle against itself to be able to make follow-on asset Mannesmann’s management. Similarly, when acquisitions across multiple geographies. In China, Vodafone made a strategic alliance a market entry decision is considered at a loss with China Mobile with an investment of relative to a reference point – for instance, Pitfall 2. Judgment biases result in $2.5bn in October 2000, which gave it where the price to be paid for a platform overinvestment in an acquisition strategy a foothold in the world’s largest mobile acquisition exceeds its standalone value – market, while China Mobile gained access management may choose to continue making While the first strategy pitfall can result to Vodafone’s marketing and technical risky investments in an unattractive region from the way executives perceive risk, Pitfall expertise, particularly in third-generation hoping to gamble their way out, rather than to 2 results from the way they estimate risk, mobile technology. Meanwhile, the group’s divest the platform at a lower but certain loss. and other critical factors. A management initial investment in Japan followed from the team may envision a well crafted long-term merger with AirTouch, and comprised taking A CEO may also ‘look at’ a strategic platform consolidating strategy that involves the equity stakes in nine regional mobile telecoms acquisition in too narrow a way. Experimental selection of a series of targets to secure its companies. This created the option for further evidence shows that, when people are strategic position and to capitalise on – or investment which led to J-Phone Vodafone offered a new gamble, they tend to evaluate even catalyse – industry consolidation. and the acquisition of Japan Telecom. it in isolation, separately from other risks. However, judgment biases reinforced This tendency is called ‘narrow framing’ by ‘group think’ and the disregard of In this article, we first review such pitfalls and can occur in platform acquisitions in warning signals, can mean that the based on a study of behavioural economics particular because the payoff distribution of strategy they envision is based on overly literature, and categorise and illustrate them an investment is much easier to understand optimistic expectations of synergies and based on Vodafone’s serial acquisitions in isolation than it is in the overall context the illusion of ‘control’ over the targets strategy. We draw on option-game models of a serial acquisition strategy that involves concerned, that may not really reflect to develop a structural quantitative toolkit interdependent follow-on investments the realities of the uncertain world. to minimise the risk of failure due to and/or acquisitions in other regions. irrational decisions and then we guide Thus a company considering a platform One such judgment bias that CEOs may objective managerial judgment to deal acquisition as a ‘one shot’ deal will tend to face is overconfidence. Extensive experimental more appropriately with rivals’ irrational underestimate its importance and behave evidence has shown that people tend to behavior or mispricing in financial markets. too conservatively. Clearly, too narrow be overconfident, and this factor can be framing was one bias that did not affect particularly true – and particularly tempting Pitfall 1: Over-investment or over- Vodafone’s former CEO, Sir Christopher – in the kind of environment that typically conservatism in consolidating industries Gent. As the company was capturing its surrounds highly successful executives who leading position, some of its rivals remained may have already executed a string of value- Acquisition are surrounded by uncertainty. conservative, unable to appreciate the new accretive transactions. Such managerial hubris, A CEO may avoid taking a ‘certain loss’ by growth opportunities. But Vodafone was the unrealistic belief held by bidding managers www.criticaleye.net 02 that they can manage the assets of a target firm hypotheses critically. Seeing only what we want more efficiently than its current managers and to see is labelled confirmation bias: experiments that they can unlock unique synergies or create show we tend to seek confirmatory data a proprietary competitive position, goes hand- and opinions, which leads on to systematic in-hand with excessively optimistic expectations. overconfidence in favoured hypotheses. This type of general bias is revealed in the Table 1. provides an overview of simple ‘Confirmed!’ card experiment described in ‘boardroom’ experiments that can reveal senior Table 1. Confirmation and overconfidence executives’ sensitivity to cognitive drawbacks in strategy can also partly result from (or be that may introduce judgmental biases in their reinforced by) other biases, such as group think. ability to envision their acquisition strategy Just as in the experiment in Table 1., participants rationally. While experiments such as ‘Feeling choose cards that support their hypothesis. Confident?’ ought to cause executives to look Executives, supported by their business again at the rationality of their own belief development teams, consultants and investment in their abilities, it is worth remembering bankers (who are, after all, incentivised to that overconfidence is a general self-serving complete, rather than abort, transactions), human bias, and one that, sadly, appears to rarely seek or are presented with data that be only minimally affected by feedback! contradicts a transaction’s value proposition.

In particular group think and team decisions Pitfall 3. Entrapment in an tend to accentuate the risk of unwarranted escalating bidding contest confidence in the acquisition decision and lead to the illusion of control. CEOs Consider the classic thought experiment ‘War of and top management teams often credit Attrition’ (described in Table 2.) that shows how their own actions when they succeed, two rivals engaged in a bidding war can end up but blame ‘bad luck’ when they fail. paying much more than the object is worth to justify their initial bidding expenses, a tendency A related judgment bias in strategy design that is exacerbated by the competitive lies in a subconscious resistance to testing instinct typical of behaviour in an auction.

TABLE 1. BIASES THAT AFFECT COMPANY STRATEGY Pitfall Bias/Heuristic Board Room Experiment

Pitfall 2. Judgment cognitive biases: Experiment: ‘Feeling confident?’ The most common response to the question overconfidence over optimism, illusion of (Response 1) is the combination of the ‘E’ control and confirmation bias to a business ‘Rank your driving skills (on a scale of 1 to 5) and ‘2’ cards. ‘E’ confirms the statement model enforced by board room group think compared to the average in this group (=3)’ by having ‘4’ on its reverse (an uneven number would refute the statement). Subjects are asked to write down in a However, turning over the 2 does not sealed envelope their estimate of their provide new information. A consonant competence as drivers in relation to the (e.g., ‘K’), on the other side would not group. When the average result is calculated actually say anything about the statement, it is almost invariably greater than 3, a while a vowel would appear to confirm statistical impossibility. The majority of the statement, but could not make it subjects regarded themselves as more necessarily true for all the cards. So turning skilful and less prone to taking risks the ‘2’ over cannot test the assertion. than the average driver in the group. The correct answer in this experiment is Experiment: ‘Confirmed!’ Response 2: the combination of ‘E’ and Four double-sided cards are laid out on ‘3’. Turing over the ‘3’ choice could prove the table, each with a number on one that the statement is false if the other side and a letter on the other. Suppose side reveals a vowel (as it does: i.e. A). that we can see only the upper face of This is a simple example of actively the four cards, as shown below. seeking data that could contradict a ‘Which pair would you choose given the opportunity hypothesis or assertion. Not reading the to flip over just two cards to test the assertion statement carefully enough – in fact, ‘If a card has a vowel on one side, then there reading an unjustified assumption into must be an even number on the other side’?’ it - participants tend to choose the ‘E’ and ‘2’ cards because they see them as problem E 2 C 3 being capable of providing evidence that confirms the assertion. They tend not to response 1 4 K choose the ‘3’, even though it is the only one that could disprove the statement. response 2 4 A

www.criticaleye.net 03 In a serial acquisition strategy, the premium by the mean of their distribution, of $2bn. As paid for a platform acquisition that can be the auction rules ensure both companies are viewed as a precursor to the current target can sold to the highest bidders, it is likely that also be seen as a ‘sunk cost’ – thus acting as a company A’s acquirer will have paid somewhat further incentive to overpay once more. Worse less than $2bn. But the acquirer of the uncertain still, powerful arguments can be made that the target B is the one that most over-estimated its entire serial acquisition strategy will stall or value and this ‘cursed winner’ will pay $3bn. even fail unless the current target is acquired (See again Table 2. for an ‘experiment’ which at all costs. Previous expenditures, such as can illustrate this to boards and executives.) on due diligence, consultancy, financing or even on a previous high premium platform Even when a good valuation method is available, acquisition, should never be used as arguments uncertainty will remain. A rational bidder should by themselves to continue bidding. They take account of the danger of the winner’s should be ignored, unless their nature and curse and submit a bid that is lower than the any associated learning provide real insights value estimate – and the more uncertain the into whether it makes sense to continue. valuation, the lower the bid should be. A seller should recognise that bidders’ awareness of Pitfall 4. The ‘winner’s curse’ – successful the winner’s curse may lead them to bid more bidding on inaccurate information cautiously. Indeed, sellers at private equity auctions can be observed doing everything Bidders in auctions are not only vulnerable they can to limit uncertainty for buyers, eg, by due to irrational escalated bidding: inaccurate providing strategy reports from consultants information can also lead auction bids and approved leverage from financiers. to exceed the value of the acquisition. At the time of Vodafone’s expansion, ‘the winner’s curse’ was quite common in the TABLE 2. BIASES THAT AFFECT BIDDING telecoms industry, in auctions for licences and bidding contests for acquisitions. Pitfall Bias/Heuristic Board Room Experiment

When several bidders contemplate the Pitfall 3. Escalated commitment Experiment: ‘War of attrition’ acquisition of the same target company, they may not know the target’s exact value when A £5 note is auctioned between two rivals they bid. Where ‘good’ information as to the in an English (i.e. ascending price) auction. target’s value is difficult to come by, or just Bids increase by £1 per round until one uncertain, bidders are obliged to fall back on bidder gives up and the other wins the £5 trying to estimate its value independently. When note. However, in this format the loser and the company is worth (roughly) the same to all the winner still both pay their highest bid. bidders, the only thing which distinguishes them In practice, the winning bid often will be their respective valuation estimates. The exceeds £5, since neither rival wants to winner will thus be the one that makes the give up and pay their bid unnecessarily. highest estimate. But if, in fact, the average bid The above example is in some sense is accurate, then the highest bidder will have similar to a company auction, with overestimated the target’s value – so (almost by cost of advisors that are sunk. definition) the winner is likely to have overpaid. Pitfall 4. Winner’s curse Thought experiment: ‘The winner’s curse’ Consider a simple numerical example of two (can result from bias and different companies being auctioned that can information asymmetry) An undisclosed amount of small change illustrate the phenomenon. One with a well- in a wallet (taken randomly from a defined value deriving from ascertainable cash group member) is auctioned among flows (A) and the other (B) with an uncertain value the group. Bids are sealed and the highest wins the unknown amount. based on growth options. Suppose the value of company A is $2bn: as this value is known, ‘What would you bid for an asset bidders will make an offer just under that level. with an unknown value?’ It’s high enough to outbid rival bidders, but low enough to leave the transaction profitable. The Participants will want to bid a margin range of bids will, therefore, be quite restricted, below their estimate (which is inevitably since they will be based on ‘good’ information. uncertain, and made on guesswork rather than evidence) to allow for profit, but high But, in the company B case, where the asset is of enough to win the auction. However the dynamic created by the combination of the (comparatively) unknown value, the increased uncertain value and selection of the highest uncertainty will mean a much wider distribution bid often results in the winner being the of bids, since they are all based (more or less bidder who most overestimates the target’s entirely) on estimates. Suppose the bids vary value – which they subsequently regret. from $1bn to $3bn, with an average, measured www.criticaleye.net 04 Pitfall 5. Relative valuation with Executives need to be constantly aware of insufficient adjustment their potential biases if they are to protect themselves from making suboptimal During normal decision making, individuals decisions in their strategy, valuations and anchor, or overly rely on, specific information bidding decisions. But awareness alone or a specific value and then adjust to that may not be sufficient to ensure a successful value to account for other elements. But, consolidation strategy: psychologists once the anchor is set, there is usually a bias have demonstrated that recognising our towards that value. Take, for example, an errors and biases does not necessarily executive deciding to set an offer price on a lead us to be able to correct them and go target. They may start from the basis of prices on to more rational decision-making. paid for similar companies, and then use ‘multiples’ (eg, Enterprise Value to EBITDA We argue that quantification based on real and Price/Earnings multiples) as their basis for options and game theory may offer further refining their evaluation of the focal company, help in mitigating the effects of several rather than considering how well the target cognitive biases. The advantage of integrating company and its strategy fits into the bidder’s game theory into the options approach is own strategy. The valuation may suffer from related biases, eg, ‘representitiveness’ – where the valuation is overly dependent on relative TABLE 3. BIASES THAT AFFECT Company VALUATION valuation compared to its peers or to a rival bid. Again, insufficient adjustment to an early Pitfall Bias/Heuristic Board Room Experiment ‘anchor’ figure can lead an acquirer to underbid or to overpay. The boardroom experiment Pitfall 5. Biases in valuation Experiment: ‘Anchor and Adjustment’ ‘Anchor and Adjustment’ (in Table 3.) allows analysis: Anchoring and executives to illustrate the dangers of such insufficient adjustment, This experiment shows how we cognitive drawbacks in the valuation analysis representative, available anchor our estimates. A valuation and affect heuristics. team is asked two questions: of target companies for – and to – themselves. ‘Is the number of countries in Africa more or less Pitfall 6. Mispricing in financial markets: than [a high or low random number]?’ and beauty is in the eye of a beholder ‘How many countries do you believe there are in Africa?’ The analysis becomes more complex when (already biased) bidders observe rival bidders Participants given a low random number and attempt to account for their actions in the formulation of the first question and reactions. The ‘Biased Beauty Contest’ tend to significantly underestimate the number of countries when answering game (see again Table 3) can be a metaphor the second question, while those given for thinking about multiple valuations in a high random number in tend to financial markets. Sometimes the price overestimate their second answer. seems to be determined by the ‘perception of beauty’ of others who, in their turn, might be Pitfall 6. Excess pricing Experiment: ‘Biased beauty contest’ basing their perceptions on their view of how in financial markets: others perceive the ‘beauty’ of the target. Biased beauty contest. Participants are asked to write down a number between 0 and 100 such that Figure 1. shows the market value of Vodafone their guess will be as close as possible to over a 15-year period from 1994 – 2009. 2/3 of the average guess in the group. Financial markets were at the peak of their irrational exuberance at the turn of the Behavioural economists use this experiment millennium and, with ‘new economy’ firms to show the bounds of rationality in game in hot demand into 2001, may have greatly theory. A player might think that the average participant is not involved and overvalued the telecoms industry’s growth that their average guess will be 50. So opportunities. Serial acquirers sometimes they guess 33, which is 2/3 of 50. The next choose to finance their acquisitions with order of sophistication in ‘thinking how shares, both to take advantage of relative others think’ incorporates the view that the appreciation within the sector and, at the same other guessers will understand the same time, to hedge against whole sector mispricing first level of thinking, and so assuming by financial markets. Vodafone countered the that they will guess 33 on average, and anxieties expressed about the costs of its serial therefore selecting 22 (as 2/3 of 33). acquisition strategy in this period by paying Game theorists search for an equilibrium with its own shares, so as to benefit from the - that is, a number which, if everyone general re-rating of the telecoms industry at guessed it, no one would have any the time – in effect, they served as a useful incentive to change their guess: in this ‘currency’ helping to mitigate the potential instance, that number would be 0. for overpayment for its ‘fully priced’ targets. www.criticaleye.net 05 that it explicitly recognises value erosion is to map alternative acquisition paths – that and the interaction of rival bidders playing is, to adopt a ‘lattice’ approach, mapping a the game of ‘musical chairs’ for available long-term contingent series of acquisitions acquisitions under conditions of uncertainty. and organic investments against the key uncertainties that, as they are resolved, In this view, bidding on a platform could change the available options. acquisition should be based on an expanded NPV (net present value) criterion: Figure 2. presents a simplified illustration of the staged decision structure of Expanded NPV = Vodafone’s strategy, and of the synergistic (stand-alone value of platform price) opportunities arising from its acquisitions + added value of future (shared) synergistic opportunities or alliances. The sequential acquisition strategy stages are represented as branches This starts with the platform’s standalone value on a tree. Exercising its UK acquisition and expected synergies, but also uses real options gave Vodafone a dominating First, the option characteristics of different options to take into account the potential for home market position that enabled it to types of potential acquisition targets must follow-on acquisitions, as well as using game pursue a worldwide network expansion be considered. Acquisitions that realise theory to predict the strategic impact of rival strategy and leverage the competences their benefits primarily through synergies bids. When a target is offered at an auction, it gained into follow-on acquisitions, or through expected earnings streams (or or when multiple bidders make an offer for a as well as enhancing its cash flows via operating cash-flows) are classified here publicly traded firm, value can be eroded if various cost and marketing efficiencies. as asset acquisition options. For instance, other buyers seem likely to bid up the price. Vodafone pursued a strategy of acquiring To avoid overpayment, executives should use The various types of acquisition options are the follow-on options in existing operations, financial market valuations as a top-down building blocks of a serial acquisition strategy. sequentially increasing its shareholdings in check to find an ‘outside view’ based on A real options analysis requires the team to Vodafone Libertel (Netherlands), Vodafone- objective empirical evidence of growth-option identify the various types of options explicitly Panafon Hellenic (Greece), Vodafone value from financial markets that confirms (eg, platform investment or asset or divestment Telecel-Comunicacoes Pessoais SA (Portugal) the added value of consolidation, rather options) and to use option valuation methods and Europolitan Vodafone AB (Sweden). than subjective management information. (eg, the binomial model) to value the set of The fundamental rationale for this was real options embedded in the organisation and simply to deliver growth. A minority stake The successful way to structure a list of its businesses. (Several of these acquisition that creates an opportunity for the holder to target options into a consolidation strategy options types are described below in Figure 2.) increase its stake in a target can be viewed

FIGURE 1. MARKET VALUE OF VODAFONE GROUP VS. FTSE INDEX (3/1/94-31/12/2009)

Nov 2000 Jul 2000 Price Communications Airtel SA Wireless Spain United States 100% $2708 min 100% $1700 min

Nov 2000 Jan 2000 Mobile AG Airtel SA Switzerland Spain 25% $2545 min 65% $14365 min

Dec 2000 Oct 2007 Nov 1999 Japan Telecom Co Ltd Italia SpA Mannesmann AG Japan Italy Germany 15% $2211 min May 2004 100% $1096 min 1000% 100% $202785 min Vodaphone KK Japan Dec 2000 100% $4229 min Dec 2005 Jul 2007 Jun 2008 900% Eircell Ltd Telsim Mobil Rural Cellular Corp Group Ireland-Rep Jan 1999 Oct 2002 Telekomunikasyon United States (Pty) Ltd 100% $4155 min AirTouch Cegetel Turkey 100% $733 min South Africa 800% Communications Inc France 100% $4557 min 65% $2413 min United States May 2001 30% $2270 min Feb 2007 100% $60287 min Japan Telecom J-Phone Hutchison Jun 2008 700% Japan Essar Ltd Alitel Corp 20% $5486 min Nov 2002 Nov 2005 United States Airtel SA VenFin Ltd India 67% $12748 min 100% $28100 min May 2001 Spain South Africa 600% Airtel SA 100% $2193 min 100% $2913 min Spain Feb 2009 92% $1577 min Hutchison Au- 500% Feb 2001 Telecommun Bus Japan Telecom Australia Co Ltd 100% $2456 min Japan 400% 25% $1350 min

Sep 2001 300% Japan Telecom Co Ltd Japan Nov 2004 67% $2684 min Nextwave Telecom Inc 200% United States 100% $3000 min 100% 0% 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09

Vodaphone FTSE 1oo

Source: Thomson Financial www.criticaleye.net 06 as a ‘call option’ on the company value (its underlying asset), with the ‘bid’ as the exercise price and time to maturity equal to the horizon of the acquisition opportunity.

A platform acquisition option, however, involves a higher growth option value than an asset acquisition option, since it involves an option on the underlying call options. Whether it relates to accessing an emerging market or a joint venture where firms combine resources, a platform acquisition can be a critical component in a serial acquisition strategy. Such an acquisition derives a significant part of its value by creating a new path for follow-on investment General market context made opportunities, so altering both the position of the enterprise and its strategic course. telecoms assets difficult to

Vodafone executed a series of opportunistic platform acquisitions to position itself for value, making auction bidders multiple follow-on acquisitions using the buy-and-build principle. Serial acquisitions vulnerable to the ‘winners curse’ represent conditional platform trajectories that establish specific targets and have intersections at several junctures in time, as well as defining a limited number of FIGURE 2. STAGED DECISIONS IN VODAFONE’S BUY-AND-BUILD ACQUISITION STRATEGY possible outcomes for an industry endgame. Vodafone’s platform acquisitions in the US,

Europe, China and Japan each provided FOLLOW-UP ACQUISITIONS Success a basis for growth in that region. (Simple option) Platform US GROWTH POTENTIAL US airtouch Failure Defer or Abandon Success CONCLUSIONS FOLLOW-UP ACQUISITION Platform EUROPE GROWTH POTENTIAL EUROPE Mannesmann Failure The various behavioral biases at play can Success Defer or Abandon Success make competitive interactions in the Start FOLLOW-UP ACQUISITION GEOGRAPHICAL Platform China GROWTH POTENTIAL CHINA acquisition game increasingly complicated EXPANSION Failure Defer or Abandon Success Start Success and sophisticated. The existing ‘toolkit’ FOLLOW-UP ACQUISITION Expansion Platform japan GROWTH POTENTIAL JAPAN for company valuation and acquisition J-Phone Failure FOLLOW-UP ACQUISITIONS strategy – present value and traditional Platform Failure (Simple option) strategy approaches – falls short in Defer or Failure Abandon resolving biases. We argue that using Defer or Abandon GEOGRAPHICAL more sophisticated analysis, applying real EXPANSION options, with extensions based on game theory and modified market valuations, can represent a strategy that is at once more rational and more dynamic. CATEGORY DESCRIPTION OPTION DEFINITION Vodafone’s acquisition history provides Stage I: Multiple or Parallel Platform businesses with domestic A platform acquisition is a compound Platform Options (Series of (UK) reputations and competencies option - an option on options. The an example of the behavioural dilemma Compound Growth Option) provided the basis to leverage core platform can be viewed as a call option competencies and marketing capabilities on follow-on investment opportunities involved in consolidation acquisitions. into a broader geographical base. (i.e. as an underlying call option), with the exercise price as the follow-on target Although Vodafone’s CEO was sometimes price and the time to maturity being accused of managerial hubris (Pitfall 2), equal to the horizon. Valuing a series of parallel compound options is more he framed Vodafone’s strategy consistent complex, since it may involve a collection of interacting sequential options, which has with a real option view (and much more an additional operating value derived from broadly as some of his conservative – Pitfall the opportunity to benefit from uncertainty through the coordination of subsidiaries 1 – competitors). From this perspective, that may be geographically dispersed Vodafone’s Mannesmann acquisition (e.g., see Kogut and Kulatilaka, 1994). gained it embedded platforms in two of Stage 2: Timing of Asset Acquisitions Consolidation, building up scale in a Each opportunity for an acquirer to bid (Simple Timing Option) fragmented market. Vodafone continued to for or increase its stake in a target can be Europe’s most important mobile telecoms consolidate - or increase stakes in existing viewed as a call option on the company markets (Germany and Italy) as part of a participations - as long as the net synergies value for the bidder (i.e. an underlying of the acquisition and growth options at asset), with the initial bid as the exercise global strategy targeting multiple platform that time were sufficiently valuable. price and time to maturity being equal to the horizon of the acquisition opportunity. acquisitions across multiple geographies. This resembles a call option on the value of synergistic benefits where the Second, the option game approach is not only exercise price is the cost of the merger. about option values, but also about price and www.criticaleye.net 07 the bidding game. Even though Vodafone’s of their irrational exuberance at the turn of the Mannesmann acquisition provided a powerful millennium, and may have greatly overvalued step into new markets, to justify its final the telecoms industry’s growth opportunities bid for Mannesmann effectively estimated (Pitfall 6). Following the insights from the the value of the strategic growth options top-down or market method, the excess involved at $82bn – a figure so high one pricing of Vodafone’s own shares as result of would consider it unrealistic from an option the general overrating of the telecoms industry game perspective to bid this high. One could at the time, in effect gave it the potential to surmise that Vodafone’s bid might have been overpay for its ‘fully priced’ targets. ‘New motivated in part by its desire to win, (Pitfall economy’ firms were in hot demand in 2001, 3) rather than fairly reflecting the target’s and it would be going too far to hold Vodafone fundamental real option value. The general responsible for the general overvaluation market context made telecoms assets difficult of the telecoms industry – but the cost of its in their industries by increasing their to value, making auction bidders vulnerable to serial acquisition strategy and of building the capabilities to value and bid for target the winners curse (Pitfall 4) and often overly company to be the world leader it is today companies successfully. Financial market dependent on relative valuation compared to were born by those investors who bought analysts can also benefit from an improved the target’s peers or on rival bids (Pitfall 5). Vodafone shares at the top of the cycle in 2001. understanding permitted by these new tools for identifying potential targets and valuing Third, top-down (PVGO) valuation requires This academic framework can assist companies in consolidating industries. When that company valuation and financing takes executives of acquiring companies by properly applied, we believe the strategic into account consideration of the (biased) helping to discipline their strategy and option perspective can promote rational appreciation of growth options in financial thus improve their ability to participate decisions and help identify when takeover markets. Financial markets were at the peak successfully in consolidation trends premiums for certain platform acquisitions are justified. Moreover, it can clarify the relationships between the restructuring FIGURE 3. VODAFONE’S CONCENTRATION OF DEALS IN 1999 -2001 of fragmented markets and the strategies and market values of specific companies. Panel A. Deal value over time This is an abbreviated version of an article that first Total deal value ($billion) appeared in the California Management Review 300 © Criticaleye 2011 250 200 Thras Moraitis Executive General Manager 150 Group Strategy and Corporate Affairs, Xstrata plc 100 Xstrata is a major global diversified mining 50 group, listed on the London and Swiss stock 0 exchanges and headquartered in Switzerland. Thras joined in 2003 and is responsible 1991 1992 1993 1994 1995 1996 1997 1998 1999

2001 for the Group’s strategic development, 2000 2002 2003 2004 2005 2006 2007 2008 2009 post-acquisition integration, leadership development, external affairs and investor Panel B. Number of deals over time relations as well as the Group’s technology businesses. Thras has postgraduate degrees in Electrical Engineering and Computer Science No. of deals and an MBA from the University of the 16 Witwatersrand, South Africa. He now lives 14 in London with his wife and two children. Han Smit 12 Professor of Corporate Finance, 10 Erasmus University Rotterdam 8 Han’s work focuses on private equity, 6 valuation and serial acquisition strategies. Recently, he published a book: Strategic 4 Investment: Options and Games – synthesises 2 valuation and strategy. He has been a Visiting Fellow at NIAS (Netherlands Institute of 0 Advanced Studies in the Humanities and Social Sciences) and Harvard University, 1991 1992 1993 1994 1995 1996 1997 1998 1999 2009 2001

2000 2002 2003 2004 2005 2006 2007 2008 Columbia University and UC Berkeley.

Contact the authors through www.criticaleye.net Source: Thomson Financial www.criticaleye.net 08 HOW THE OPTION GAME VALUATION OF ACQUISITION OPPORTUNITIES WORKS

Consider the case of two comparable bidders (A and B) and $0.5bn > $0bn). Thus, bidder A has a dominant strategy to each having the opportunity to acquire a target. pursue the acquisition. Given this, firm B will not pursue the acquisition (since $0bn > -0.5bn). The Nash equilibrium (marked An option games valuation works backwards *) outcome of this game is given by the down left cell (1, 0), based on the following steps: where bidder A acquires the company against firm B‘s value.

(A) calculate the payoff values of the target for bidder In a down cycle we can make a similar analysis. Here we assume A and bidder B in a 2 x 2 matrix (sub-game); the payoff in the right hand 2 x 2 matrix below where neither firms can generate net synergies from the consolidation - (B) determine dominant bidding strategies and the neither will make an offer resulting in a (0,0) payoff. Nash equilibrium in each 2 x 2 sub-game; and C. Backward option valuation of Growth (C) use option valuation to work backward in the Opportunities (PVGO) along the binomial tree binomial tree using the equilibrium values from the two sub-games in the up and down cycle states.

The current value of the company for bidder A equals up cycle, p = 0.56 down cycle, p = 0.44 $4bn and for bidder B $3.2bn. Uncertainty in the cycle Bidder B Bidder B and company value is modeled by a lattice approach WAIT BID WAIT BID where the value can either move up (u = 1.25) or down (d = 0.80) depending on industry product demand. iv iii iv iii WAIT WAIT A. Payoffs in the 2 x 2 matrix (bidding sub-game) (0.75, 0.25) (0, 0) (0, 0)* (0, -0.8) Making a bid corresponds to taking or increasing a stake in a target. Exercising the option early may pre-empt the rival’s Bidder A Bidder A acquisition opportunity, but deferring a bid is consistent with a cautious strategy that is less expensive, more flexible and ii i ii i BID BID safer in case of unfavorable economic developments. The (1, 0)* (0.5, -0.5) (-1, 0) (-0.8, -1) figure below shows the payoffs in this economic uncertainty dilemma. Consider first the up-cycle scenario. The left 2 x 2 table summarises the payoffs (bidder A, bidder B) in four investment- timing scenarios at an upward moving market situation: Note: Payoff in each cell is for (Bidder A, Bidder B); (i) when both firms enter into a contest and bid (simultaneously) Strategies of firm A: wait (upper row) or bid (lower row); the total expected synergies are reflected in the price. Suppose Strategies of firm B: wait (left column) or bid (right column). that in the up cycle the target value for bidder A is $5bn (V+=uV) and for bidder B $4bn. The bidding cost is equal at $0.5bn. The The risk neutral probability equals NPV for bidder A equals the target private value for the bidder – price – bidding cost, that is 0.5 = 5 - 4 - 0.5. Since bidder A assigns a higher value to the target (due to its complementary (1+r)-d (1+0.05)-0.80 assets), bidder B will be outbid and its NPV equals the bidding P= = = 0.56 cost – $0.5bn, resulting in a (0.5, -0.5) payoff for each bidder, u-d 1.25-0.80 with bidder A capturing the target against B’s value; For the up and down cycle situations, the sub-games have (different) (ii)/(iii) when one investor bids while the other makes Nash equilibriums in pure strategies. As we have seen above (see also no offer, the bidder pre-empts the target for $3.5bn and left 2 x 2 matrix), in the up cycle the advantaged bidder captures the captures a larger part of the value creation for themselves opportunity (with Nash equilibrium values being ($1bn) in box (ii). (5 - 3.5 - 0.5 = 1 for bidder A), resulting in a payoff of (1, 0) But in the down cycle, the Nash equilibrium is for both bidders to wait if Bidder A bids and B does not or (0,0) respectively if B bids instead (thus receiving $0 in box (iv) in the right hand 2 x 2 matrix and bidder A does not enter (4 - 3.5 - 0.5 = 0 for bidder B) above). The present value of the growth opportunity for bidder A to acquire the target (PVGO) is the discounted weighted average of the and (iv) when both firms decide to wait, they share Nash equilibrium outcomes across the up and down cycle states. the total additional value created by the timing flexibility equally, cushions their downward losses. Backward induction using the sub-games represented by the above Suppose that the payoff equals (0.75, 0.25). two 2 x 2 matrices (under up vs. down cycle) results in a (sub- game perfect) equilibrium trajectory. The process moves backward B. Dominant strategies and Nash equilibrium over random demand moves using the associated risk-adjusted The next step is to identify dominant strategies in the above probabilities to calculate values at the beginning of the binomial tree. sub-game and determine the resulting Nash equilibrium. In determining the growth option value the average is computed as Dominant strategies are those actions that always the risk-adjusted expectation of the two equilibrium outcomes (1 give a player a higher payoff value than any alternative for high cycle, 0 for low cycle) using the risk-adjusted probabilities action, whatever the other player decides to do. (p = 0.56 for an up-cycle trend and 0.44 for a down-cycle trend, with discounting performed at the ‘riskless’ interest rate of r = 0.05). Consider the equilibrium implications of an asymmetric payoff structure as in the bidding game shown below. 0.56(1) + 0.44(0) Suppose that bidder A’s payoff for pursuing the acquisition * 0.53bn. (lower row) exceeds the payoff of letting it go (upper row), PVGO A= = 1.05 no matter which strategy Bidder B chooses ($1bn > $0.75bn

www.criticaleye.net 09