
BRONWYN H. HALL University of California, Berkeley ROBERT E. HALL Stanford University The Value and Performance of U.S. Corporations THE VALUE OF A CORPORATION is known from hour to hourin the stock market.The performanceof a corporation,from the shareholders'per- spective, is measuredby the corporation'sability to pay dividends, now and in the indefinitefuture. Our research investigates the relation be- tween value and performance. We use modern finance theory as a benchmarkfor valuation. Finance theory holds that, on average, the currentvalue of a share is the discounted value of the future dividends the share earns. The theory is explicit about the discountrate. If, on av- erage, over firmsand over time, shares sell for less than the discounted value of the dividendsthe shares ultimatelypay, it means that the stock marketundervalues those shares; investors requirea higherrate of re- turnthan theory suggests they should. Ourmotivation for this researchis the persistentcriticism that Ameri- can capitalism,with its focus on stock prices determinedby myopic in- vestors, diverts managersfrom efficient, long-terminvestments toward the style of managementmost pleasing to the stock market. We ask if certainmanagerial decisions or firmcharacteristics result in stock prices thatare higheror lower thanthe benchmarkprovided by financetheory. Is the marketsystematically shortsightedwith respect to all activities, placing too little value on deferred payoffs? Did this problem worsen We are gratefulto Chris Hall for assistance and to the EconometricsLaboratory at Universityof California,Berkeley, for massive computations.This research was sup- portedin part by the National Science Foundation.We benefitedenormously from the commentsof the discussantsand othersat the BrookingsPanel meeting. I 2 Brookings Papers on Economic Activity, 1:1993 duringthe 1980s?Does the marketfavor higher current accounting earn- ings? Does it put a highervalue on firmsthat invest in plant and equip- ment? Does it put a lower value on firms that invest in research and development (R&D), advertising, and other forms of intangible, diffi- cult-to-valuecapital? Does it put a highervalue on firms that pay high dividends?All of these are claims madeby the critics of Americancapi- talism. If the answer is yes to any of these questions, there is an unexploited opportunityfor arbitragein the U.S. stock market.For example, if the marketundervalues firms that follow a Japanese-stylestrategy of high investmentin productdesign and marketpenetration, then an investor can beat the marketby investingin these firmsfor the long term, deriving highnet value afterten or more years of holdingthe sharesuntil the pay- off becomes apparentto otherinvestors. Centralto the critiqueof Amer- ican capitalismis the absence of patient arbitrageurswith decade-long buy-and-holdstrategies. The whole focus of the professional money managerswho dominate shareholdingsin the United States is on arbi- trage strategieswith payoffs in minutes, hours, or at most months, say the critics. Our findingsgive strong but partialsupport to the critique. We find statisticallyunambiguous evidence of importantarbitrage profits from long-termstrategies. Three of our findingsfavor the critique. First, the stock marketis systematicallyshortsighted; it favors poli- cies that generate near-termdividends over those that requirewaiting. All investors who place theirfunds in the stock marketrather than in the bond marketearn large extra rewards over time after considerationof the relative riskiness of stocks and bonds. This findingconfirms earlier well-known results on the equity premiumpuzzle.' Second, although the bias againstdeferred payoffs lowers the incentive for investment of all types, the bias is smallerfor investmentin plant and equipmentthan for investmentsin intangibles.Third, the marketdisfavors intangible in- vestmentin advertising. On the other hand, we make three findings unfavorableto the cri- tique. First, the marketputs a lower value on firms with higher book earnings, after standardizingfor actual subsequent performance. In other words, the patient arbitrageurcan make money by buying firms 1. Grossmanand Shiller(1981) and Mehraand Prescott(1985). Bronwyn H. Hall and Robert E. Hall 3 with unusually high reported current earnings and holding them until subsequentperformance shows that the marketerred by placing such a low value on these firms. Second, the market puts a higher value on R&Dinvestment than is warrantedby subsequentperformance. The pa- tient arbitrageurcomes out ahead by avoiding R&D-intensive firms. (Thisfinding is at a lower level of statisticalconfidence than the previous ones.) And third,excess discountsfell in the 1980srelative to the 1970s. Our work looks at the values of the shares of a sample containing about half the publicly traded U.S. manufacturingcorporations from 1964through 1991. Within the frameworkof modernfinance theory, we study the relationbetween share value and actual subsequentpayouts to shareholders.Our approach is an applicationof the generalprinciples developedin RobertHall's workwith Steven N. Durlauf.2The approach permitsus to makerigorous statements about departures of shareprices from the level mandatedby valuationtheory and to associate those de- partureswith particularcharacteristics of firms.Although our approach puts a predictedfundamental value on each firmin our samplefor each year, these valuationsare quite noisy. We reach strongerconclusions by lookingat statisticalaverages of the differencebetween stock prices and fundamentalvalues over many stocks and manyyears, which eliminates most of the noise. Ourwork is a departurefrom the abundantrecent literatureon valua- tion anomalies.3That literatureshows that such a thing as an underval- ued firmexists. The findingsresult from a searchfor the most successful currentvariables for forecastinglater performance. The best forecasting variablesare invariablyratios with the currentstock pricein the denomi- nator.Thus the characterof the findingsis thatinvestment in stocks with high earnings-priceratios, high dividend-priceratios, or high book val- ue-priceratios will earnabnormally high returns. The researchersin this traditionadvocate value strategies and have impressive evidence that such strategiesearn high returnswhen appliedin the real world. In con- trast, we take as given that such a thing as an undervaluedfirm exists. We are interested in describingthe association of undervaluationwith the choices made by the firm's managers. For example, we are inter- 2. Durlaufand R. Hall (1990). 3. Basu (1977), De Bondt and Thaler(1985), Fama and French (1988), and Lakoni- shok, Shleifer,and Vishny (1993)are leadingcontributions. Scott (1985)shows the same thingin the frameworkof Shiller(1981) and LeRoy and Porter(1981). 4 Brookings Papers on Economic Activity, 1:1993 ested in measuringthe undervaluationof firmswith policies of heavy in- vestmentin plantand equipment.To achieve this objective, we must ex- clude the currentstock price from our right-handvariables. It would be uninterestingin our frameworkto conclude that firms with low stock prices suffer high discounting in the stock market, even though the stock-pickingrule that tells the investor to look for low stock prices gen- erates the highest expected returns. Our work is not a contributionto the financeliterature showing that valuationanomalies exist. Rather,we apply the methods of financein a new way to consider the issues raised by the critics of the stock market. A second importantwarning to the readeris that our research deals with the external valuation of the firm. We can comment on how the stock market responds to the observable variables as they are deter- mined by the firm's managers.We cannot comment on the internalre- sponses to the valuationerrors made by the stock market.For example, we show that the marketis shortsightedwith respect to investment; it puts a discounthigher than the one meritedby financetheory on the sub- sequent earnings from a capital project. We presume, but we do not show, that managersrespond by launchingtoo few capitalprojects with deferredpayoffs. Ourwork deals with stock marketmyopia, not corpo- rate myopia. Of the two majorelements of the case that capitalismis shortsighted,we consider only one. The restriction to issues of external valuation brings clarity to our work, we believe. Otherapproaches have to deal with conflictinginter- nal and external influences. For example, Michael Jensen has argued that the tendency for a firm's share price to jump when the firm an- nounces an investmentproject is a sign that the stock marketis not myo- pic, and so managersdriven by stock marketincentives should not be- have myopically.4But Jeremy Stein observed that the findingis hardly dispositive.5In his model, managersbehave myopicallyin equilibrium. They set a hurdlerate above the market'sdiscount rate, so the adoption of a projectgenerates a positive gain for the shareholdersprecisely be- cause of myopia. Our approachin this paper is complementaryto the approachtaken by Bronwyn Hall in previous work.6 She has studied the relation be- 4. Jensen(1988). He cites the event study of McConnelland Muscarella(1985). 5. Stein(1989). 6. B. Hall (1992, 1993a). Bronwyn H. Hall and Robert E. Hall 5 tween the currentreproduction cost of a firm's hardwareand software capital (plant and equipment, inventories, ownership of other firms, R&D capital, and advertisingcapital) and the total value of the firm's debt and equity. She estimatesan overallTobin's
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages49 Page
-
File Size-