The Myers and Majluf (1984) Model of Capital Structure
Kamis, 08 Mei 2014
Reading for Topic 3:
The Myers and Majluf (1984) Model of Capital Structure
The Myers and Majluf (1984) model of capital structure choice was the first fully articulated model of the impact of information asymmetry alone on the debt-equity mix. The model elaborates the arguments of Myers (1984), and refines a similar model developed by Miller and Rock (1985), which does not differentiate between debt and equity financing. Previously, Modigliani and Miller (1958, 1961) had developed a set of theorems based on ideal conditions and denying the financing role any capacity to add value, except for cases of unequal corporate and personal tax rates (see Miller (1977)). When financial distress costs induced by high leverage are admitted, the Modigliani and Miller model is represented by the static trade-off theorem, which proposes that any tax subsidy of debt (that is, the tax deductibility of interest) is offset by increasing distress costs, implying an optimal leverage choice.
Myers and Majluf present a model where capital structure is driven by managers’ desire to avoid underpricing of the firm’s outstanding securities by the capital market. A consequence of underpricing is undervaluation of the firm. Undervaluation is induced by adverse selection costs imposed by investors who are aware of managers’ information advantage[1]. Being irredeemable, equity issues are the most sensitive to underpricing, while at the other extreme internal financing ensures no undervaluation. Maintaining an internal pool of funds to fund investment activity reduces adverse selection costs. Given positive information asymmetry costs and that managers act in existing shareholders’ interest, managers rationally will underinvest whenever underpricing of an equity issue exceeds the net present value of the investment.
The Myers and Majluf model is the progenitor of several extensions and alternative formulations, reviewed in Harris and Raviv (1991). Extensions are Krasker (1986), who specifies the size of new investment, and Narayanan (1988) and Heinkel and Zechner (1990) who focus on overinvestment induced by information asymmetry with respect only to the new investment. Timing arguments on equity issues are examined by Korajczyk, Lucas and McDonald (1990) and Lucas and McDonald (1990). Brennan and Kraus (1987), Noe (1988), and Constantinides and Grundy (1989) demonstrate that the underinvestment problem can often be resolved through debt signalling in the context of a richer set of financing options (for example, convertible debt or stock repurchases)). Thus, the Myers and Majluf model is the focal point of a controversial literature. An even larger empirical literature has been spawned to discriminate these models’ explanatory power.
In Myers and Majluf’s model, informational asymmetry relates to firm value irrespective of the proportions of assets-in-place and growth opportunities, which are implicitly assumed to have the same expected variance of returns. The degree of information asymmetry declines with the variance of returns on the firm’s assets[2],[3]. Myers and Majluf further assume that (a) managers act in existing shareholders’ interest, and (b) existing shareholders are passive in that they do not trade on information released by debt or equity issues[4]. As a consequence of (b), all equity issues are made to new shareholders, who may therefore purchase stock at a premium or discount relative to its (true) value. When equity is sold at a premium (discount), the firm is over- (under-) valued, and existing shareholders gain (lose) at the expense of the new shareholders. Given information asymmetry, new shareholders discount the probability of overvaluation in their pricing of equity issues, and this is the cost of adverse selection.
When a firm’s investment opportunities require financing that exceeds operating cash flows, liquid assets and the ability to issue safe debt[5], it may be in existing shareholders’ interest for the firm to forego positive net present value opportunities rather than issue risky securities. The adverse selection cost for equity is higher than for debt owing to the risk of a permanent wealth transfer from existing to new shareholders, which occurs when the security issued is highly sensitive to the private information held by managers. Thus, highly-rated debt with a fairly certain payoff stream is issued in preference to equity because the latter is difficult to price without knowing the exact value of the firm’s assets-in-place and investment opportunities. On the other hand, debt is easier to value when there is sufficient collateral because investors need only value the collateral and not the whole firm as in the case of equity.
Myers and Majluf’s solution to the underinvestment problem is for firms to carry sufficient financial slack to undertake all profitable opportunities (on average) as they arise, equity issues are never necessary. Financial slack consists of cash, liquid securities and the capacity to issue safe debt, and is known to the market. The amount of financial slack is increasing in (a) the expected size of investment opportunities, and (b) the degree of uncertainty about the value of the firm’s investment opportunities (which are usually tied to the value of the firm’s assets-in-place). Hence, Myers and Majluf always predict negative stock price responses to equity issues. Alternatively, if safe debt is issued, the stock price will not fall, and may even increase if the announcement conveys information of profitable opportunities. They argue that it is costly to cut dividends to generate financial slack because dividends also signal operating cash flows. Equity issues therefore always incur adverse selection costs. The cost of information asymmetry is lower for debtholders than for shareholders because, unlike equity claims, debt claims are usually tied to specific (or collateralised) assets about which debtholders have more information[6]. To minimise undervaluation firms therefore maintain financial slack.
The preceding arguments generate the well-known pecking order theory[7]. Debt issues or leverage increases are expected to be negatively related to internally-generated cash flow[8], and positively related to the need for external funds to finance growth[9]. Further, operating cash flows are expected to dominate external financing as a determinant of investment or capital expenditure, whereas the static trade-off theory posits rebalancing of external funding as firms move toward shifting debt-equity target ratios[10]. Importantly, they imply that leverage increases until a target ratio is achieved irrespective of the level of informational asymmetry implying that leverage increases with the degree of informational asymmetry. A second group of propositions follows when the degree of information asymmetry is allowed to vary through time. Myers and Majluf themselves recognise managers’ incentive to time external financing visits when information asymmetry is known by investors to be small[11].
The Stulz (1990) Model of Capital Structure
In the Myers and Majluf’s (1984) model, financial slack is more valuable than risky debt, while in the agency models of capital structure debt has a broader role. Jensen and Meckling (1976) argue that debt effectively lessens the divergence of interests between managers and shareholders, and reduces the residual loss caused by this conflict. Offsetting these benefits are the agency costs of debt, which expose debtholders to subsequent opportunism by managers. Jensen and Meckling propose an optimal capital structure equating the marginal costs and benefits of debt and equity. Since debt commits managers to pay out cash flows in whatever state occurs, Jensen (1986) advocates a disciplinary role for debt in reducing the amount of free cash flow in the hands of self-interested managers[12].
Stulz (1990) extends the Jensen (1986) proposition by developing a capital structure model driven by information asymmetry and agency costs, in which investors are risk-neutral. Shareholders are assumed to observe neither operating cash flow nor the firm’s investment opportunity set[13], both of which are exogenous. In contrast to Myers and Majluf (1984), Stulz argues that managers value having more resources under their control than shareholders prefer, so they have a propensity to overinvest[14]. To eliminate this cost, shareholders need to contract ex ante with managers to payout any free cash flow, but this is assumed not possible because collective action by shareholders is costly[15].
Stulz (1990) argues that debt can be used to optimise the level of firm investment across cash flow states. Optimal leverage depends on the probability distribution of operating cash flow and on the firm’s investment opportunity set. When operating cash flow is higher than expected (given the firm’s investment opportunities), managers overinvest in negative net present value projects rather than pay out a cash dividend. Higher leverage at the outset diminishes this risk because a schedule of higher debt payments restricts managers’ capacity to overinvest. Conversely, lower leverage is optimal when operating cash flow is lower than expected for a given investment opportunity set. The predictions reverse when investment opportunities vary for a given cash flow: higher (lower) leverage is optimal when investment opportunities are lower (higher) than expected. Since debt and equity issues decrease one cost of managerial discretion and increase the other, there is a unique solution to the firm’s capital structure. Debt arguments are shown to dominate liquidation for firms with free cash flow as long as there are sufficiently valuable investment opportunities[16].
The argument is now paraphrased in terms of the expected variance of operating cash flows, which is likely correlated with the volatility of current operating cash flows. When the volatility of the firm’s operating cash flow is low, the risks of over- and under-investment are both low because operating cash flow is largely predictable and an optimal debt-equity ratio can be set to minimise the over- and under-investment costs. On the other hand, when operating cash flows are volatile, larger corrections are required to the target debt-equity mix, which is set on the basis of expected operating cash flows and investment opportunities, seen by managers and investors alike. In the case of a cash windfall increased leverage is predicated, but debt is harder to sell when operating cash flows are more volatile. Stulz (p. 17 and the Appendix) argues in this case that the costs of managerial discretion are reduced by diversification[17]. The reasoning is as follows. Deviations from the target debt-equity mix are costly to shareholders, because managers overinvest (underinvest) when more (less) funding is available than needed. As operating cash flow volatility falls, the risk of both overinvestment and underinvestment diminishes, so the value of diversification to shareholders is increasing in the expected volatility of the firm’s operating cash flow.
Diversified firms are therefore argued to have lower costs of managerial discretion than non-diversified firms (because their operating cash flows are more predictable). A firm’s investment opportunities may be seen as real growth options, whose value is increasing in the volatility of expected operating cash flows. Thus, if investment opportunities decline unexpectedly, operating cash flows become less volatile and diversification is less valuable for reducing managerial discretion. However, diversification is ineffective in reducing overinvestment because, unlike debt, diversification does not force payout of free cash flow. Despite these insights, Stulz does not elaborate upon the relation between leverage, diversification and investment opportunities.
External financing arguments relate to debt arguments[18]. When investment opportunities turn out as expected, no changes in leverage or dividends are required because the costs of managerial discretion are minimised. Alternatively, if investment opportunities turn out worse than expected, then overinvestment results but no external financing is necessary because not enough debt was issued ex ante to curtail overinvestment. Then again, if investment opportunities turn out better than expected, underinvestment ensues because too much debt was issued ex ante, when with hindsight equity should have been issued. To mitigate underinvestment, a current equity issue is required, but this is unlikely to take place because managers cannot credibly signal shareholders that the investment opportunities are genuinely valuable. Hence, firms that issue equity must have been successful in signalling this information to the capital market[19].
[1] Adverse selection costs are recognised in Akerlof (1970) who shows how markets can break down when potential buyers cannot verify the quality of the product they are offered. Faced with the risk of buying a lemon, the buyer will demand a discount, which in turn discourages the potential sellers who do not have lemons. Myers and Majluf (1984, p. 196) differ in that the seller is not offering a single good, but a partial claim on two, the assets-in-place and the new project.
[2] See Myers and Majluf (1984, p.206).
[3] As Dybvig and Zender (1991) point out, this and similar models assume that optimal contracts cannot be written with managers to eliminate over- or under-investment.
[4] The model is sensitive to changes in assumptions, if for example (i) managers can costlessly signal their private information, (ii) existing shareholders can rebalance their investment portfolios in response to securities issues, or (iii) firms can sell assets-in-place without affecting the value of growth opportunities. Equity visits become good news if the information asymmetry is limited to the volatility of returns. Giammarino and Neave (1992) and Viswanath (1993) present models in which the managers of undervalued firms issue stock even when cash is available.
[5] Debt is safe (as opposed to risky) when resolution of the information asymmetry (no matter how bad the news) does not alter the value of the debt. For example, the first tranche of a small amount of debt issued by a well-diversified firm would approximate this property.
[6] At the individual contract level, debt holders are even more informed than in aggregate because collateral is identified in the contracting process.
[7] This is articulated in Myers (1984).
[8] Leverage is the ratio of total debt to total assets, conventionally measured at book.
[9] Supporting evidence is documented by Chua and Woodward (1993), who also argue that a negative relation is expected between leverage and the liquidity of the firm’s assets. This assumes that asset sales avoid the adverse selection costs of equity issues.
[10] For empirical evidence, see Fazzari, Hubbard and Petersen (1988).
[11] Myers and Majluf (1984, p. 220). Klein and Belt (1993) report supporting evidence.
[12] Free cash flow is defined in Jensen (1986, p.323) as cash flow in excess of that required to fund all projects that have positive net present values when discounted at the relevant cost of capital.
[13] After the initialising date.
[14] Other important assumptions are (i) borrowings do not signal investment opportunities, and (ii) overinvestment is not controlled by management compensation contracts, or by capital markets when debt or equity visits are planned (see Myers (1977)).
[15] Jensen and Murphy (1990) provide evidence that compensation contracts are only weakly related to managerial performance.
[16] Conversely, high debt levels (which increase the probability of liquidation when cash flows fall in a “bad” year) are optimal if the value of investment opportunities lost through liquidation is small. This can be seen more clearly when growth opportunities attach to assets-in-place and are effectively onsold with the underlying asset.
[17] A change in cash flow volatility has no impact on firm value if managers are maximising shareholder wealth in the first place.
[18] Jung, Kim and Stulz (1996) extend Stulz (1990) to explain the prevalence of equity issues.
[19] See Stulz (1990, p. 14).