NBER WORKING PAPER SERIES STOCK ISSUES AND INVESTMENT POLICY WHEN FIRMS HAVE INFORMATION THAT INVESTORS DO NOT HAVE Stewart C. Myers Nicholas S. Majiuf Working Paper No. 884 NATIONALBUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge MA 02138 April 1982 Research support from the Office of Naval Research is gratefully acknowledged. The research reported hereispartofthe NBER's research program inTaxation.Anyopinionsexpressed arethoseof the authors andnotthose of the National Bureau of Economic Research. NBER Working Paper No. 884 April1982 STOCK ISSUES AND INVEST{ENT POLICY WHEN FIRMS HAVE INFORMATION THAT INVESTORS DO NOT HAVE Abs tract This paper describes corporate investment and financing decisions when managers have inside information about the value ofthe firm's existing invest- ment and growth opportunities, but cannot convey that information to investors. Capital markets are otherwise perfect and efficient. In these circumstances, he firm may forego a valuable investment opportunity rather than issuestock to finance it. Thedecision to issue cannot fully Lnformation. If stock is issued, stock convey the managers' special price falls. Liquid assets or financial slack are valuable if they reduce the The probability or extent of stock issues. paper alsosuggests explanations for some aspects of dividend choiceof capital structure. policy and Stewart C.Myers SloanSchool of Management MIT/E52—243F 50Memorial Drive Cambridge, MA 02139 (617)253—6696 Nicholas J. Majiuf Universidad Catolica de Chile Escuela de Itngenieria Casilla 114—D Santiago, CHILE STOCK ISSUES AND INVEST1ENT POLICYIHEN FIRNS HAVEINFORMATIONTHAT INVESTORS DO NOT HAVE Stewart C. Myersand Nicholas S. Majluf-' Consider a firm thathas assets in place and alsoa valuable real investment opportunity. However, it has to issue common sharesto raise part or all of the cash required to undertake the investment project. If it does not launch the project promptly the opportunity willevaporate. There are no taxes, transaction costs or other capital market imperfec- tions. Finance theory would advise this firmtoevaluate this investment opportunity as if it already had plenty ofcash on hand. In an efficient capital market, securities can always be sold at a fair price; thenet present value of selling securities is always zero because the cash raised exactly balances the present value of the liability created. Thus, the decision rule is: take every positiveNpV project, regardless of whether internal or external funds are used to pay for it. iThat if the firm's managers know more about the value of itsassets and Opportunities than outside investors do? As we willshow, nothing fundamental is changed so long as managers alwa follow the decision rule just noted. The shares investors buy will becorrectly priced on average, although a particular issue will beover or underpriced. The manager's inside information creates a side bet between old and new Stockholders but the equilibrium issue price is unaffected. However, if managers have inside information there must be Some cases in which that information is so favorable that management, if it acts In the interest of the old stockholders, will refuse to Issue shares —2— That is, the even if it means passing up a goodinvestment opportunitY. the project's NPV. cost of issuing shares at a bargainprice may outweigh investors, aware of their Thispossibilitymakes the problem jerestiflg: that a decision not toissue shares relative ignorance, will, reason bad or at least signals "good news." The newsconveyed by an issue is investors are willing to payfor the less good. This affects the price decision. issue, which in turn,affects the issue_investment conditional The problem is to figure outthe equilibrium share price and also a decision, assuming rational investors, on the issue_investment issue—investment decision on theprice it rational firm which bases the that problan, and solves it underreasonable faces. This paper addresses simplifying assumptions. This The assumptions are set outand discussed in Section 1. formulation section also contains twonumerical examples. A general 3. The last section describesex- and solution is given in Section tensions of our model andsummarizes its implications. introductory review of the literatureuntil the We defer the customary assumptions have been more fullyexplained. end of Section 2, after our 1. ASSUMPTIONS ANDEXAMPLES that investors We assume the firm (i.e.,its managers) has information and investors realize this.We take do not have, and that both managers as given——a fact oflife. We side—step the this information differential except to note question of how muchinformation managers should release, ransmittiflg information is costly. Our problem the underlying assumption that information to costlesSly convey their special disappears if managers can the market. —3— Thefirm has one existing asset and one opportunity requiring investment i. The investment can be financed by issuing stock, drawing down te firm's cash balance or selling marketable securities. Thesum of cash on hand and marketable securities will be referred toas financial slack (S). Financial slack should also include "debt capacity," definedas the amount of default—risk free debt the firm can issue. (Discussion of risky debt is deferred to Section 3.) However, it's simpler forour purposes to let the firm use risk—free borrowing to reduce the required investmentI. We may thus interpret I as required equity investment. The investment opportunity evaporates if the firm doesnot go ahead at time t 0. If S < I, going ahead requires a stock issue of E =I—S.Also, the project is "all or nothing"——the firmcan't take partof it. Weassume capital markets are perfect and efficient with respect to publicly available information. Thereare no transaction costs in issuing stock. We also assume that market value of the firm's sharesequals their expected future value conditional on whatever information the market has. The future values could be discouned for the timevalue of money without changing anything essential..?] Discounting for risk isunnecessary, because the only uncertainty important in this problemstems from managers' special informaion Investors at time t =0do not know whether the firm's stock price will goup or down when that special information is revealed at t =lHowever, this risk is likely to bediversifjable?] We can now give a detailed statement of who knows what when. —4— A Three—Date 1odel 1. There are three dates, t=1,0 ad+1. At t —1 the 0, management market has the same informationthe manageeflt does. At t ;alue of the firm's asset—ifl receivesadditional infoatiOn about the and upda:es their valuesaccordingly. place and investment opportunity, this informatOfl until t =+1. The market does not receive —l is A E(); the 2. The value of the asset—in—place values at distribution of Xrepresentsthe asset's :ossible (updated) =0is a. That is, a is t =0.Management's updated estimate at t the realization of —1 of the investment oppor- 3. The net present value (NPV)at t represents the asset's tunity is B =E(B). The distribution of possible updated NPVs at t=0. Manageme:-'S updated estimate at o is b, the realization of 4. Negative values for a andb are ruled out. This makes because of limited liability. It makes sense for the asset_in—place is opportunity because the opportunity sense for the investment 0. In other discarded if it turns out to have anegative NPV at t words, the distribution of is truncated at zero. 5. Management acts in the interestof the "old" shareholders, those =V(a,b,E) owning shares at t —1. That is, they maximize will not however, the market valueof the old stockholders' shares void. Let P be the marke: value. P reflects the generally equal issue shares or not. also managem2mt'S decision to distribution of and and Let —L — = P' marketvalue at t=0of old stockholders' shares if stock is issued. P = marketvalue at t=0if stock is not issued. 6. Slack, S, is fixed and known by bothmanagers and the market. The information available to management and the marketissummarized below: Date: t—1 t0 t+l Information available to: Managers Disributons of A and ;S a,b;S a,b; remaining S, ifany Market Distributions Distributions a,b; remaining S, if any of and ;S of and also E, either E 0 or E= I —S Two Examples The following two examples should give a betterunderstanding of the problem just posed and the steps required to solve it. In thefirst example, the firm always issues stock and goes ahead .ithapositive NPV opportunity. In the second example itmay not. First example. There are two equally probablestates of nature. The true state is revealed tomanagement at t=0and to investors at t=+1.Assetvalues are: State 1 State 2 Asset—in—place a =150 a=50 InvestmentOpportunity (NPV) b 100 b - 10 —0— (S =0).The investment Thefirm has no cash or marketable securities stock to raise E =100if opportunity requires I =100,so the firm must issue it goes ahead. We now examine a trial solution assumingthe firm issues stock of whether the favorable or unfavorable and undertakes the project regardless state occurs. In that case P' =155because A + B =155. In state 1, the true value of the firm,including 100 raised from The market value the stock issue, is 350. That is v vold + new =350. is P', the new shares' E). Thus is P' + E (the old shares' market value old PT 155 = V = P'+Ev = . 350 212.75 E = = . 350 137.25 E In state 2, v = V' =160 void = . 160=97.25 = . 160=62.75 Note that both old and newshares are correctly priced to investors, who regard the twostatesas equally probable. p '= -(2l2.75+ 97.25) =155 E' = + 62.75) =100 -(137.25 —7— Becausethe firm issues stock in both states, thedecisionto issue tells investors nothing about thetrue state. This trial solution is the equilibrju solution, becauseissuing stock and going ahead with theproject leaves the old stockholders better off regardless of thetrue state: yoff Issue and Do nothing invest (E =100) CE =0) void in 212.75 state 1 150 void in 97.25 state 2 50 In this example the firm has no use for financial slack.