In this chapter, we consider the interaction ofinvestors in a secur ties market. Thetheory of efficient securities markets predicts that the security rices that resultfrom this interaction have some appealing properties. In esse ce, these prices"properly reflect" the collective knowledge and information-pro essing ability ofinvestors. The process by which prices do this is quite compl x and not fullyunderstood. Nevertheless, the general outlines of the process ar easy to see, andwe shall concentrate on these.Securities market efficiency has important implicatio s for financialaccounting. One implication is that it leads directly to the cone pt of full disclosure.Efficiency implies that it is the information content of di closure, not theform of disclosure itself, that is valued by the market. Thus, in rmation can bereleased as easily in notes and supplementary disclosures as in th financial statementsproper. The theory also affects how the accountant sh uld think aboutreporting on firm risk.In efficient markets theory, accounting is viewed as bein in competitionwith other information sources such as news media, financial alysts, and evenmarket price itself. As a vehicle for informing investors, accoun ing will surviveonly if it is relevant, reliable, timely, and cost-effective, relative to other sources.Efficient securities market theory also alerts us to what is t e primary theoreticalreason for the existence of accounting, namely informa ion asymmetry.When some market participants know more than others, press re arises to findmechanisms whereby the better informed, who wish to do so, c n credibly communicatetheir information to others, and whereby those with in ormation disadvantagecan protect themselves from possible exploitation by the etter informed.Insider trading is an example of such exploitation.92 Chapter4We can then think of accounting as a mechanism to ena~le communicationof relevant information from inside the firm to outside. In addition to enablingbetter investor decisions, this has social benefits through impr ving the operationof securities markets.As mentioned in Section 1.2, accounting theorists be an to realize theimportance of securities market efficiency in the late 1960s. ince that time, thetheory has guided much accounting research and has had major implications foraccounting practice. By and large, financial accounting stan ard-setting bodieshave accepted the full disclosure and decision usefulness impli ations of securitiesmarket efficiency. To illustrate this, we will examine two i portant standardsfrom an informational perspective.Finally, it should be emphasized that efficiency is a mode of how a securitiesmarket operates. Like any model, it does not capture the full c mplexity of such amarket. Indeed, recent years have seen an increasing number of questions aboutwhether investors are as rational as the model assumes. We ill explain some ofthese questions in Chapter 6, and evaluate their implica ions for financialaccounting. The real question, however, is whether the efficie t securities marketmodel captures enough of the real market that accountants c n bf; guided by itspredictions. In Chapter 6 we conclude that it does. We also c nclude that to theextent securities markets are not fully efficient, this increases the importance offinancial reporting.4.2.1 THE MEAMNG OF EFFICIENCYIn Chapter 3 we studied the optimal investment decisions of ra onal .investors. Nowconsider what happens when a large number of rational indi duals interact in asecurities market. Our interest is in the characteristics of them ketprices of securitiestraded in the market, and how these prices are affected by new information.If information was free, it is apparent that investors ould want to takeadvantage of it. For instance, under the ideal conditions of Ex mple 2.2, investorswould want to know which state of nature was realized, si ce this affects thefuture cash flows and dividends of the firm. By assumption, information is freeunder ideal conditions since state realization is publicly o serv~ble. Thus, allinvestors would use this information, and the process of arbitr ge ensures that themarket value of the firm then adjusts to reflect the revised cas flow expectationsthat result, as illustrated in Example 2.2.Unfortunately, information is not free under non-ideal c nditions. Investorshave to form their own subjective estimates of firms' futur prdfitability, cashflows, or dividends. Furthermore, these estimates will need re sioili as new informationcomes along. Each investor then faces a cost-benefit t adeqff with respectEfficient Securities Markets 93'to how much information to acquire. There is a variety of re~evant informationsources-the financial press, tips from friends and associate , changes in economicconditions, advice from analysts and brokers, etc. We ca think ofinvestorsas continuously revising their subjective state probabilities as s ch information isreceived. From our standpoint, of course, a major source of c st-effective informationis firms' annual reports. Probability revision arising fr m financial statementinformation was illustrated in Example 3.1.At least some investors spend considerable time and m ney to use theseinformation sources to guide their investment decisions. Such nvestors are calledinformed. Bill Cautious, in Example 3.1, is an example of sue an investor.It should be apparent that informed investors will want to ove quickly uponreceipt of new information. If they do not, other investors will get there first andthe market value of the security in question will adjust so as to educe or eliminatethe benefit of the new information.When a sufficient number of investors behave this way, t e market becomesefficient. There are several definitions of an efficient securities market. The definitionthat we shall use here is the semi-strong form.An efficient securities market is one where the prices of secu ities traded onthat market at all times "properly reflect" all information t at is publiclyknown about those securities.Three points are particularly noteworthy. First, market rices are efficientwith respect to publicly known information. Thus, the definitio does not rule outthe possibility of inside information. Persons who possess insi e information, ineffect, know more than the market. If they wish to take advan age of their insideinformation, insiders may be able to earn excess profits on thei inv~stments. Thisis because the market prices of these investments, reflecting o y oqtside or publiclyavailable information, do not incorporate the knowledge th t in$iders possess.A second, related, point is that market efficiency is a rel tive iconcept. Themarket is efficient relative to a stock of publicly available infi rmation. There isnothing in the definition to suggest that the market is omnisc ent and that marketprices always reflect real underlying firm value. Market pric s can certainly bewrong in the presence of inside information, for example.The definition does imply, however, that once new or cor ected informationbecomes publicly available, the market price will quickly adjust to this new information.This adjustment occurs because rational investors will scramble to revisetheir beliefs about future returns as soon as new informati n, from whateversource, becomes known. As a result, the expected returns and ri k of their existingportfolios will change and they will enter the market to res ore their optimalrisk/return tradeoffs. The resulting buy-and-sell decisions 11 quickly changesecurity prices to reflect the new information.Third, investing is fair game if the market is efficient. This eans ,that investorscannot expect to earn excess returns on a security, or portfolio of ecudties, over and94 Chapter4above the normal expected return on that security or portfoli~, where the normalexpected return allows for risk. One way to establish a normal 1 etum benchmark isby means of a capital asset pricing model, as will be illustrated n Section 4 .5.An implication of securities market efficiency is that a se urity's market priceshould fluctuate randomly over time. That is, there should be no serial correlationof share returns. Thus, if a firm reports GN today, its share price should rise toreflect this news the same day. If, in the absence of any furthe news, its price continuesto rise during succeeding days, this is evidence of ine iciency. The reasonfor random fluctuation of market price is that anything abo t a firm that can beexpected, such as the seasonal nature of its business, the re rement of its chiefexecutive, or the expected profit on a major new contra t, will be properlyreflected in its security price by the efficient market as soon s the expectation isformed. That is, the market's expectation of the effect of su events on the valueof the firm is on average unbiased. The only reason that pri es will change is ifsome relevant, but unexpected, information comes along. y definition, unexpectedevents occur randomly. For example, an accident may hange the expectedprofit on a contract, and share price will quickly respond to reflect this randomevent. Thus, if we examine the time series formed by th sequence of pricechanges for a particular security, this series should fluctuate randomly over timeaccording to market efficiency theory. A time series that e "bits such seriallyuncorrelated behaviour is sometimes called a random walk.14.2.2 HOW DO MARKET PRICES PROP 'RIJYREFLECT ALL AVAILABLE INFO . TION?We now consider how market prices properly reflect all a ailable information.This process is by no means obvious or transparent. As escribed previously,rational, informed investors will demand information about securities. However,there is no guarantee that all individuals will react identic to the same information.For example, they may have different prior beliefs. S me imay have superiorability to analyze financial statement information. In sense, the decisiontheory model is like an automobile. It provides a vehicle to rocess information,but nothing guarantees that everyone's driving habits are ide tical or that they alltake the same route to a destination.As a result, it is quite likely that different investors w 1 re~ct to the sameinformation differently, even though they all proceed rati nally. Yet, investorsinteract in a market, each making buy/ sell decisions about v ous ~ecurities. Sincethe market price of a security is the result of the demand for a d supply of the securityby investors, how can the market price properly reflect all vail4ble informationwhen the individuals making the demand and supply decisio s arei different?An interesting insight into this question can be gained fro~ an example inBeaver (1989, p. 150, Table 6-1). The example relates to fore asti~g the results offootball games. The Chicago Daily News, during 1966-68, print d wtekly the predic'IIIIEjjicient:Securities Markets 95Itions of each of its sports staff as to who would win that week~nd' s college footballgames. Table 4.1, taken from Beaver, summarizes the outcome 'of these predictions.Note the following points from Table 4.1. First, there wer a number of differentforecasters (15-16) and a large number of forecasts were made (619 over thethree years). Second, no one individual forecaster dominated n terms of forecastingability. The best forecasters in 1966 were well down the lis in subsequent years,and vice versa. Third, note the consistent performance of th consensus forecast.The consensus forecast was also published weekly by the Chi ago Daily News and,for each game, consisted of the team favoured to win by the ajority of those forecasting.It is clear that the consensus forecast has a quality tha transcends the forecastingability of the individual forecasters from which the co sensus is derived.1966 1967 1968Total forecasters (including consensus) 15 15 16Total forecasts made per forecaster 180 220 219Rank of consensus* 1 (tie) 2 2Median rank of forecasters 8 8 8.5Rank of best forecasters:J. Carmichael (1966) 1 (tie) 8 16D. Nightingale (1966) 1 (tie) 11 5A. Biondo (1967) 7 1 6H. Duck (1968) 8 10 1*When all three years are combined, the consensus outperforms every ne of the forecasters(that is, ranks first).SOURCE: William H. Beaver, Financial Reporting: An Accounting Rev lution 1981, p.162, Table 6-1. Reprinted by permission of Prentice-Hall Inc., Upper S ddle,River, NewJersey. Data are from "Here's How Our Staff Picks 'Em," Chicago Dai y News, November25, 1966 (p. 43), November 24, 1967 (p. 38), and November 29, 1968 ( . 43). Reprinted withspecial permission from the Chicago Sun-Times 1999.To translate the example into a securities market conte , we can think of theforecasters as investors in a security and the forecasts as eir various buy/selldecisions. The consensus forecast is analogous to the mar et price, since it is atype of average of the various individual forecasting decisio s.The rationale behind the example is not hard to see. It appears that the differencesin forecasting ability ofindividual forecasters tend t cancel out when theconsensus is formed, leaving a "market price" that outperfo ms the ability of anyof the market participants. ,Of course, just because a consensus forecast outperfo ms :individual forecastersof football games does not by itself mean that the sa e phenomenon car96Chapter4Iries over to security prices. Essentially, what is required is hat investors' estimatesof security values must on average be unbiased. That s, the market doesnot systematically misinterpret the valuation implications of stock of information,but rather puts a valuation on securities that is on average correct orunbiased. As mentioned, this does not mean that any indi dual investor willnecessarily be correct, but it does mean that on average the arket uses all availableinformation.To see how the market may behave this way, recall our ar ment above, whichcomes from Fama (1970), that when a "sufficient number" o investors can fullyexploit available information, this is enough to generate effi iency. For example,financial analysts and institutional investors may be sufficient! adept at evaluatingsecurity value that, when other investors follow their recomme dations, the resultingprices properly reflect available information about these se urities.It should be emphasized that the above argument assu es that individualdecisions are independent, so that individual differences cane 1 out in their effecton price. If this is not the case, efficiency arguments break down. Thus, if ourfootball forecasters got together to work out and agree on a consensus forecast,their forecasts would not be independent if they reflected the ews of, say, a dominantand persuasive member of the group. Similarly, if inves rs display a collectivebias in their reaction to new information about a firm, the resulting shareprice will be biased. For example, a firm may have reported a attern of increasingearnings. If investors expect future earnings growth to contin e simply because ofgrowth in the past, share price momentum may develop. Th n, share prices maybe " too high," driven by past price increases rather than by r tional evaluation ofinformation by independent investors. We will return to this point in Chapter 6,where we discuss whether securities markets are fully efficien .4.2.3 SUMMARYIn an efficient securities market, prices properly reflect all a ilable information,and the price changes on such a market will behave randomly ver time. Efficiencyis defined relative to a stock of information. If this stock of in ormation is incomplete,say because of inside information, or wrong, security p ices will be wrong.Thus, market efficiency does not guarantee that security pri es ac;curately reflectunderlying firm value. It does suggest, however, that prices are unbiased relative topublicly available information and will react quickly to new or evised information.The quantity and quality of publicly available information will be enhanced byprompt and full reporting. However, individual investors may have different priorbeliefs and/or may interpret the same information differ ntly. Nevertheless,roughly speaking, we can think of these differences as "averag ng qut," so that themarket price has superior quality to the quality of the informaf on pfocessing of theindividuals trading on the market. This argument assumes, ho evd, that investors,or at least a major subset of investors, evaluate new informatio inqependently.4.3.1 IMPLICATIONS'I!IEfficient Securities Markets 97An early examination of the reporting implications of efficien securities marketsappeared in an article by W. H. Beaver, "What Should Be the 'ASB's Objectives?"(1973). Here, we will outline Beaver's arguments.According to Beaver, the first major implication is that ccounting policiesadopted by firms do not affect their security prices, as long as these policies haveno differential cash flow effects, the particular policies use are disclosed, andsufficient information is given so that the reader can conv rt across differentpolicies. Thus, Beaver would regard accounting disputes sue as a firm's choiceof amortization method, the accounting for future tax liabi ities, and the fullcostversus successful-efforts approach for oil and gas firms s, essentially, "tempestsin a teapot." Notice that a firm's choice between different accountingpolicies in each of these disputes involves only "paper" effects The policy chosenwill affect reported net income, but will not directly affect fu ure cash flows anddividends. For example, an oil and gas firm's proceeds fro sale of crude andrefined products will not depend directly on whether it uses 11-cost or successful-efforts accounting. In particular, the amount ofincome t the firm must paywill not be affected by its accounting policy choice in any of ese,three disputedareas since the tax department has its own way of calcul ting expenses andincome in each area, independent of how the firm accou ts for them on itsbooks. If investors are interested in future cash flows and ividends and theirimpact on security values, and if choosing between accounti g pJlicies does notdirectly influence these variables, the firm's choice between accounting policiesshould not matter.Thus, the efficient market argument is that as long as rms disclose theirselected policy, and any additional information needed to conv rt fr0m one methodto another, investors are able to make the necessary calculatio s to' see through tothe resulting differences in reported net income. That is, he rharket can seethrough to the ultimate cash flow and dividend implications regatdless of whichaccounting policy is actually used for reporting. Thus, the efl cient market is not"fooled" by differing accounting policies when comparing diffe ent firms' securities.This suggests that management should not care about which artici.ilar accountingpolicies they use as long as those policies have no direct cash fl w effects.We thus see that full disclosure extends to disclosure oft e fi~m's accountingpolicies. This is recognized by standard setters. For example, t e OICA Handbook,paragraph 1505.04, states:A clear and concise description of the significant accounti g po~icies of anenterprise should be included as an integral part of the financ az st(Jtements.98 Chapter4A second implication follows-namely, efficient securi es markets go handin hand with full disclosure. If a firm's management posses es relevant informationabout the firm and if this can be disclosed at little or o cost, managementshould then disclose this information on a timely basis unles it is certain that theinformation is already known to investors from other sou es. More generally,management should develop and report information about t e firm as long as thebenefits to investors exceed the costs. The reasons are twofo d. First, market efficiencyimplies that investors will use all available, relevant information as theystrive to improve their predictions of future returns, so that additional informationwill not be "wasted." Second, the more information a rm publishes aboutitself, the more information is publicly available about that firm. Consequently,investors' confidence in the securities market is enhanced.Third, market efficiency implies that firms should not be overly concernedabout the naive investor--that is, financial statement info mation need not bepresented in a manner so simple that everyone can underst nd it. The reasoninghere is actually quite subtle: if enough investors understand t e disclosed information,this is sufficient to ensure that the market price of a fir 's shares is the sameas it would be if all investors understood it. This is because e investors who dounderstand the financial information will engage in buy/sell ecisions on the basisof the disclosed information, which will move the market rice towards its efficientlevel. Also, naive investors can hire other persons (sue as financial analystsor investment fund managers) to interpret the informati n for them, or canmimic the buy/sell decisions of more knowledgeable inves ors. As a result, anyinformation advantage that the knowledgeable investors ma have is quickly dissipated.In other words, the naive investors can trust the effi ient market to pricesecurities so that they always reflect all that is publicly know about the firms thathave issued them, even though these investors may not have complete knowledgeand understanding themselves. This is referred to. as in estors being priceprotectedby the efficient market.Since Beaver's paper, accountants have recognized tha there is a variety ofreasons for trading securities. For example, some investors ay make a rationaldecision to rely on market price as a good indicator of futur payoffs, rather thanincur the costs of becoming informed. Others may trade for variety of non-portfolioreasons-perhaps an unexpected need for cash has a isen .. Consequently,"naive" may not be the best word to describe uninformed i vestors. This is consideredfurther in Section 4.4.A final implication is that accountants are in com etition with otherproviders of information, such as financial analysts, media, disclosures by companyofficials, and so on. That is, belief revision is a continuo s process, as pointedout in Section 3.3.3. Thus, if accountants do not provide seful, cost-effectiveinformation, we would expect that the usefulness of the cco1mting functionwould decline over time as other information sources tak ov~r-accountantshave no inherent right to survive in the competitive marketp ace for information.Efficie~t Securities Markets 99Empirical evidence about securities market response to ftnancial accountinginformation is reviewed in Chapter 5.Beaver's paper was published in 1973. Consequently, it predates SFAC 1(issued in 1978) and SFAC 2 (1980) by several years. Howe er, it provides a goodexample of the early enthusiasm of accounting theorists r efficient securitiesmarkets. It also highlights the type of disclosure-oriented th nking that led to theformal statement of the usefulness criterion by the FASB in SFAC 1.4.3.2 SUMMARYBeaver argues that securities market efficiency has several i plications for financialreporting. First, managers and accountants should no be concerned aboutwhich accounting policies firms use unless different accounting policies havedirect cash flow effects. Many accounting policy alter ativ~s, about whichaccountants have argued long and hard, do not have su h cash flow effects.Second, firms should disclose as much information about t em$elves as is feasible-the fact of disclosure and not the form it takes is w at is. important. Theefficient market will prefer the least costly form of disclosur , other things equal.One can argue, however, that financial statements are a co t-effective disclosuremedium. Third, firms need not be concerned about the aiv investor whenchoosing disclosure policies and formats. Such persons are price-protected,because efficient security prices properly reflect all that is ubli~ly known aboutthose securities. Furthermore, there are a variety of mediu s, such as financialanalysts and investment funds, whereby investors can take a van~age of sophisticatedinformation without needing to fully understand it th mst1lves. Finally, theefficient market is interested in relevant information from anyi source, not justaccounting reports.4.4.1 A LOGICAL INCONSISTENCYThe careful reader may have noticed an inconsistency in o discussion of efficientsecurities markets to this point. Recall that efficiency i plies that the marketprice of a security at all times properly reflects all that is ublidy known aboutthat security. What is it that drives market price to have t is "Rroperly reflects"characteristic? It is the actions of informed investors who re al:ways striving toobtain and process information so as to make good buy/sell ecislons.However, by the definition of market efficiency, all av ilablf information isalready reflected in market price. That is, the price is fully info'rmative.2 Sinceinformation acquisition is costly, and investors could not ex ect ~o beat the marketwhen the market price already reflects all publicly own information,100 Chapter 4investors would simply stop gathering information and rely o~ market price as thebest indicator of future security returns. For example, a simple decision rule wouldbe to buy and hold an investment portfolio, changing its co position only if therisk/return tradeoff of the portfolio gets out of line.The logical inconsistency, then, is that if prices fully refle t available information,there is no motivation for investors to acquire informati n; hence, prices willnot fully reflect available information. In terms of football fi recasting, the forecasterswould stop putting effort into their forecasts becaus they can't beat theconsensus forecast, but then the consensus forecast would 1 se its superior forecastingability. Technically speaking, the problem here is th t stable equilibriumprices do not exist, as shown by Grossman (1976).This has potentially serious implications for accounting t eory, since a lack ofequilibrium makes it problematic whether financial statemen information is usefulto investors. Also, it is contrary to what we observe. SFA 1 (Section 3.8) certainlyimplies that investors find financial reporting useful, fi r example.However, there is an easy way out of the inconsistency. his is to recognizethat there are other sources of demand and supply for securit es than the buy/ selldecisions of rational informed investors. For example, people ay buy or sell securitiesfor a variety of unpredictable reasons-they may decid to retire early, theymay need money to pay gambling debts, they may have rece ved a "hot tip," etc.Such persons are called liquidity traders or noise traders. heir buy/sell decisionswill affect a security's market price, but the decisions co e at random-theyare not based on a rational evaluation of relevant informationTo illustrate how market price is affected by the prese ce of noise trading,suppose that a rational investor observes a security's pric to be higher thanhe/she had expected based on all the information curren y posessed by thatinvestor. Now, our investor knows that other rational investor also have their owninformation about the security and that this information may well be morefavourable. These other investors may be buying and drivi g up the security'sprice. As a result, our investor is inclined to raise his/her exp ctation of the security'svalue. While the investor does not know what inform ion other investorshave, it is rational to believe that the information is favoura le and this may bewhat is driving up the security's price.However, our investor also knows that the higher-tha -expected securityprice may simply be due to noise trading. Perhaps someo e has temporarilyinvested a large cash windfall in a randomly chosen por folio of securities,including the security in question. If so, our investor would ot want to increasehis/her expectation of the security's value. Since each seen rio is possible, theinvestor will increase his/her expectation of the security's val e, but to an amountless than the security's current market price. That is, th ra~ional investorresponds by putting some weight on each possibility. In effe t, th~ current shareprice conveys some information about share value but not a l infprmation as inthe fully informative case. Efficient Securities Markets 101For our purposes, an important point to note is that itjvestors now have anincentive to increase the precision of their beliefs by gatheri . g more information.If they can find out which explanation is the correct one this can quickly beturned into a profitable investment opportunity. The efforts finvestors to do thiswill then drive share price towards its efficient value. Presu ably, at least some ofthis additional information will come from analysis of finan ial statements.When investors behave as just described, they are s d to have rationalexpectations. Security prices are said to be partially inform tive in the presenceof noise trading and rational expectations. Note that mark t prices are still efficientin the presence of noise trading, but in an expected va ue sense, since noisehas expectation zero. That is, the investor expects that a s curity's market pricefully reflects all publicly available information, but furth r investigation mayreveal that this is not the case.The extent to which investors gather additional infor ation depends on anumber of factors, such as how informative price is, the qu ty of financial statementinformation, and the costs of analysis and interpretati n. These factors leadto empirical predictions about how security market prices respond to financialstatement information. For example, we might expect tha price will be moreinformative for large firms, since they are more "in the ne s" than small firms,hence their market price will incorporate considerable infor ation. This reducesthe ability of financial statements to add to what is alread known about suchfirms. Thus, we would predict that security prices respond 1 ss to financial statementinformation for large firms than for small firms.Furthermore, note that firm management has an ince tive to cater to thedesire of investors to ferret out information. For example, m nagement may haveinside information that leads it to believe the firm is underv ued. To correct this,management may engage in voluntary disclosure, that is, d sclosure of informationbeyond the minimum requirement of GAAP and other reporting standards.Such disclosure can have credibility, even if unaudited, since legal liabilityimposes discipline on managers' reporting decisions. Unfor nately, there are limitationson voluntary disclosure, not only because the legal stem may be unableto completely enforce credibility but because management n not want to revealinformation that would give away competitive advantage.However, voluntary disclosure is much more complex a d subtle than simplydisclosing information. Management can signal inside info atiot;i. by its choice ofaccounting policies and, indeed, by the nature and extent o voluntary disclosureitsel This means that there are potential rewards to invest rs, and analysts, forcareful and complete analyses of firms' annual reports. Such nalyses may identifymispricing and can quickly be turned into profitable investm nt decisions.Also, an increase in the quality of financial statemen information, otherthings equal, should lead investors to increase their utilizati n of financial statementinformation relative to price. For example, the require en~. by the OntarioSecurities Commission (OSC) that firms include manage entdiscussion andanalysis (MD&A) in their annual reports and Section 4250 of the GICA102 Chapter 4Handbook relating to future-oriented financial informa on (FOFI) mayincrease market price reactions to annual reports. Annual r ports should havehigher information content with MD&A and/or FOFI relati e to the preexistinginformation content of market price. MD&A and FOFI are iscussed in Section4.8. Empirical evidence on the decision usefulness of finan al statement informationwill be considered in Chapter 5.We conclude that the term "properly reflect" in the efficie t securities marketdefinition has to be interpreted with care. It does not mean th t security prices arefully informative with respect to available information at all points in time.Indeed, if it did, this would have adverse implications for the sefulness of financialstatements. Rather, the term should be interpreted as eflecting a tensionbetween the level of informativeness allowed by noise and li uidity traders, andthe ability of investors and analysts to identify mispriced secur ties through analysisof accounting policy choice, the nature and extent of volun ary disclosure, and,indeed, of all other available information. With this interpre ation in mind, it isimportant to point out that the implications of security mark t efficiency as outlinedby Beaver in Section 4.3 continue to apply. In particula , the importance offull disclosure remains.4.4.2 SUMMARYWhile the ability of a market price to average out individual fferences in informationprocessing, as we saw in the football forecasting ex ple, is on the righttrack, the process of price formation in securities markets is uch more complexthan this. Through consideration of ways that rational investo s can become moreinformed by careful analysis of managers' disclosure decisions, and by allowing forother sources of demand and supply for securities than fro rational, informedinvestors, accountants are beginning to understand the rol of information inprice. The presence of non-rational traders does not necess rily mean that theefficient securities market concept that share prices "properly eflect" informationis invalid, but rather that this concept must be interpreted wit care.Improved understanding of the process of price formatio leads to empiricalpredictions of how security prices respond to accounting in rmation and, ultimately,to more useful financial statements.We are now in a position to formalize the relationship betwee the efficient marketprice of a security, its risk, and the expected rate of retur on a security. Weshall do so by means of the well-known Sharpe-Lintner c pital asset pricingmodel (CAPM) (Sharpe, 1964; Lintner, 1965).IEfficient Secufities Markets 103First, we need some preliminaries. Define Rjt' the net rate ;of return on theshares of firm j for time period t, as:R _pit+ Dit -Pj,t-1 = pit+ Dit1t - p. P.J, t-1 ], t-1-1where:P. is the market price of firm j's shares at the end of eriod tJ!Djt is dividends paid by firm j during period tPj, t-l is the market price of firm j's shares at the beginn ng of period tThis is the return concept used in Examples 3.2 and 3.3. It is a et rate of returngiven that the opening market price is subtracted in the numer tor. We can alsodefine a gross rate ef return as 1 + Rjt' where:p. +D.l+R= it itJt PJ. , t-1Since the only difference between the two rate of return concept is the 1, we canuse them interchangeably. In fact, to conform to common practic , we will usuallyrefer to both net and gross rates of return as simply returns.We can think of returns as either ex post or ex ante. Ex post, e are at the endof period t and looking back to calculate the return actually re ized during theperiod. Alternatively, we can stand at the beginning of period t a d think of an exante or expected return as:E(P. + D.)E(Rjt) = Jt it - 1PJ. , t-1(4.1)That is, expected return for period t is based on the expecte price at the endof the period plus any dividends expected during the period divided by thebeginning-of-period price.Now, consider an economy with a large number of inv stors like ToniDifelice (Examples 3.2 and 3.3). Recall that Toni is risk-averse nd has a meanvarianceutility function. As shown in Tables 3.5 and 3.6, Toni an calculate theexpected rates of return, the variances of return, and the covaria ces of return foreach security in the market. Assume that there is a risk-free asse in the economy,with return Rf. Assume also that security markets are efficien and transactioncosts are zero. Then, the Sharpe-Lintner CAPM shows that:(4.2)where [3j is the beta of share j and RM, is the return on the ma ket portfolio forperiod t.104 Chapter 4Note that the model is in terms of the market's expected rFturns. Equation 4.2states that at the beginning of period t the expected return fo the period equals aconstant Rf(l - (3.) plus another constant 13 times the e ected return on the J Jmarket portfolio.Strictly speaking, markets do not have expectations- ndividuals do. Oneway to think of the market's expectations is that the price of a share behaves as ifthe market holds a certain expectation about its future perfor ance. More fundamentally,the market price of a share includes a sort of averag of the expectationsof all informed investors, much like the consensus forecast i the Beaver footballexample (Section 4.2.2) includes an average expectation oft e forec.asters.It is not difficult to see the intuition of the model. Since ational investors willfully diversify when transactions costs are zero, the only ris measure in the formulais 13 .. Firm-specific risk does not affect share price bee use it disappears infully dive:sified portfolios. Also, note that the higher is 13j t e higher is expectedreturn, other things equal. This is consistent with risk aversion, since risk-averseinvestors will require a higher expected return to compensate for higher risk.Note also the role of the current market price P. t- l in e model. The returndemanded by the market on share j for period t, that fs, E(R) is a function only ofRr, RMt' and 13 .. In Equation 4.1, given expected end-of-piri d price P.t and dividendsD.t, we fee that P. t-l in the denominator will adjust s that the1right handside of Equation 4.1 eqJ~s E(Rjt). That is, a share's current pr ce will adjust so thatits expected return equals the return demanded by the market for that share.We can now see how new information affects firm j's hare price. Supposethat at time t-1 some new firm-specific information com s along that raisesinvestors' expectations of P.t (and possibly also of D. ), witho t affecting Rf, 13. orE(RMt). This will throw E~uation 4.1 out of balance!! since E R) from ( 4.2) doesnot change. Thus, Pj,t-1' the current price, must rise to rest r~ equality. This, ofcourse, is consistent with market efficiency, which states tha the market price ofa security will react immediately to new information.For our purposes, there are three main uses for the CA M formula. First, itbrings out clearly how share prices depend on investors' e ectations of futureshare price and dividends. If these expectations change (the numerator ofEquation 4.1), current pricePj,t-l (the denominator) will im ediately change toreflect these new expectations. For a given change in expect tions, and given Rfand E(RMt), the amount of the change in current price d pends only on theshare's beta. To put this another way, the larger the change in expectations, thelarger the change in price, other things equal.Second, by reverting to an ex post view of returns, the CAPM provides uswith a way of separating the realized return on a share into expected and unexpectedcomponents. To see this, consider the following ve sion of the model,where we are now at the end of period t and looking back:Rt = a. + R.RM + E J J 1-'3 t JtEfficient S:r:curities Markets 105This version of CAPM is called the market model. It states that the realizedreturn Rjt for the period is the sum of the beginning-of-pe! iod expected return(a. + [3.RMt) and the unexpected or abnormal3 return Er he expected returncohies frbm the CAPM, with a. = ~(1 - [3.). The Et caJtur s the impact on Rtof all those events during perioa t that were bot expe~ted at he beginning of ti{eperiod. By definition in an efficient market, E( Ejt) = 0, sin e new informationcomes along randomly. But, in any period t the realized val e of Et will not bezero. Thus, the market model enables an ex post separation o the r~alized returnRt into expected (a. + [3.RMt) and unexpected or abnormal E) components.J Third, the market mbdel provides a convenient way to es iinate a stock's beta,which, as we saw in Section 3.7, is an important risk measure or investors. Noticethat the market model is presented in the form of a regr ssion equation. Byobtaining past data on Rt and RMt' the coefficients of the re ession model can beestimated by least-squar~s regression. If we assume that the market is able to formaccurate expectations of RMt (so that RMt is a good proxy fc r E(RMt), which isunobservable), and if we assume that f3 is stationary over ti e, then the coefficientof RMt from least-squares regressi~n is a good estimate of f3-. Furthermore,the reasonableness of the estimation can be checked by com aring the estimatedcoefficient a. with (1 - [3-)Rf-the two should be the same.As we will see in Cha~ter 5, much empirical research in ace unting has requiredan accurate estimate of beta, and we will return to its estimafon in Section 7.6.l.For now, it is important to realize that the CAPM provides a important and usefulway to model the market's expectation of a share's returns and that the modeldepends crucially on securities market efficiency. Also, it sho s clearly how newinformation affects current share price.4.6.1In this section, we take a closer look at the notion of"publicl avail.able" informationin the efficient securities market definition. This leads directly to what isundoubtedly the most important concept of financial accou ting theory-informationasymmetry. Frequently, one type of participant in the market (sellers, forexample) will know something about the asset being traded hat another type ofparticipant (buyers) does not know. When this situation exist , the market is saidto be characterized by information asymmetry. As mentio ed in Section 1.5,there are two major types of information asymmetry-ad erse. selection andmoral hazard. We now consider these in greater detail.One effect of information asymmetry is to hamper the rop~r operation ofmarkets. In Examples 2.1 and 2.2, there was no informatio asyJ!nmetry, by the!106 Chapter 4definition of ideal conditions. Then we saw that market value$ and present valueswere equal. This is not necessarily true when information as etry is present.These effects were studied by Akerlof ( 1970). An examp e of a market characterizedby information asymmetry is the used car market. he owner of a carwill know more about its true condition, and hence its future stream of benefits,than would a potential buyer. The owner may try to take a vantage of this bybringing a "lemon" to market, hoping to get more than it is rth from an unsuspectingbuyer. However, the buyers will be aware of this te ptation and, sincethey don't have the information to distinguish between lemon and good cars, willlower the price they are willing to pay for any used car. As a res t, many cars-thegood ones-will have a market value that is less than the real alue of their futurestream of benefits. The arbitrage effect, whereby cars of sim ar service potentialmust sell for similar prices, operates less effectively when it s difficult to knowexactly what the service potential of a used car is. Thus, own rs of good cars areless likely to bring them to market. In other words, the mark for used cars doesnot work as well as it might.In extreme cases, a market may collapse completely as a r sult of informationasymmetry. To illustrate, consider the market for insurance po cies. You may wishto buy insurance against the possibility of failing to attain yo r university or collegedegree or professional accounting designation. You woul be better off withsuch a policy, at least if the cost was fair. Serious illness or ac ident, for example,may prevent your completion of the course of studies, and ou could eliminatethis risk if you had a policy that reimbursed you for your loss f the present valueof the increased future income that would follow the attainm nt 0f your degree.However, if you owned such a policy, you would probably shir yo4r studies, evenif you were perfectly healthy. Why put in all the time and effo to complete yourcourse of studies when,.by merely failing, you could receive eq ival~nt compensationfrom your insurance policy?As a result, no insurance company would sell you a poli that would reimburseyou for your full income loss if you failed to attain your egree. Essentially,the problem is one of information asymmetry. You have a ajor informationadvantage over the company, because the company can only o serv:e whether youfail, not whether your illness caused you to fail. This is call d a moral hazardproblem, for you are tempted to cheat the company by shirkin yout studies. Notethat requiring a medical certificate would not be of much use ere, because of thedifficulty in establishing that it was the illness that led to the ur.Another difficulty the insurance company would face i that people whowere sick would flock to enroll in university programs (calle an ~dverse selectionproblem, because people whose health is adverse to the i surance company'sbest interests self-select themselves to buy insurance). Then, whep their illnessled to their failure, they could collect on their policies and sti 1 enjoy the monetaryfruits of a degree. IEfficient :rcurities Markets 107Faced with information disadvantages of this magnitude, the companyresponds by not writing insurance policies of the type descri ed. Hence, no marketdevelops. Obviously, if there is no market for an asset, uch as a universitydegree, it is impossible to value this asset using market value.It is interesting to note the variety of devices that mark ts use to reduce theeffects of information asymmetry. Thus, used car markets re characterized byguarantees, safety certificates, test drives, dealers who attemp to establish a goodreputation, and so on. Insurance markets are characterized b medical examinationsfor life insurance, co-insurance and deductible clause for fire insurance,premium reductions for good driving records, and so on. Ho ever, because theyare costly, these devices do not completely eliminate the pro lem. Nevertheless,they may be sufficiently effective to at least allow the market t operate, albeit notas properly as it would in the absence of information asymme ry.The presence of risks, such as the impact of illness on arning power, thatindividuals would like to protect themselves against but cann t because a marketdoes not develop, is a consequence of incomplete markets. ecall from Section2.6 that incompleteness results when estimation problems, su h as in oil companyreserves, prevent market prices from developing. Here, we s e that informationasymmetry is another source of incompleteness.Incompleteness of markets also results when markets e st but do not workproperly. For example, despite the devices mentioned above t at enable the usedcar market to operate, a used car buyer still bears a risk of buy g a vehicle of differentquality from what he or she wants and is paying for. hat is, if price doesnot perfectly reflect the quality of a commodity, individuals e unable to buy theexact quality they want, so they bear more risk than they wo d like. As a result,they would like to see market incompleteness reduced. But because of adverseselection and the cost of devices to overcome it, it may not e cost-effective toeliminate it completely. Nevertheless, we will now argue that nancial accountinghas a role to play in improving the operation of markets, there y at least reducingthe problem ofincompleteness.One of the reasons why information asymmetry is of s ch importance toaccounting theory is that securities markets are subject to info mation asymmetryproblems. This is because of the presence of inside informati n and insider trading.Even if security market prices fully reflect all publicly av ilable information,including that which can be inferred from firms' accounting p licy and disclosuredecisions, it is still likely that insiders know more than outsi ers about the truequality of the firm. If so, they may take advantage of their i formation to earnexcess profits. This is another example of the adverse select on problem, sinceinsiders will be attracted by this opportunity, which is advers to the interests ofinvestors. Of course, investors will be aware of this possibility and will lower theamounts that they would otherwise be willing to pay for sh es, tJo reflect theirexpected losses at the hands of insiders. Just like the used car arke~, the efficientsecurities market does not work as well as it might. .108 Chapter 4' We can think of financial reporting as a device to reduce the ~dverse selectionproblem, thereby improving the operation of securities mark 'ts and reducingincompleteness.To reduce adverse selection, accountants have adopted polici s of full disclosure,to expand the set ofinformation that is publicly available. Also, meliness of reportingwill reduce the ability of insiders to profit from their info ation advantage.Of course, since financial reporting is costly, it is unlikely at the problem ofinside information can be eliminated. Nevertheless, full and ti ely disclosure willincrease the usefulness of financial reporting to investors by anding the set ofpublicly available information. This should help with Be ver's concern thataccounting is in competition with other information sources, i eluding price itsel4.6.2 SUMMARYUnder ideal conditions, the firm's market value fully refleWhen conditions are not ideal, market value fully reflectsinformation, if security markets are efficient.s all information.publicly availableThe difference between these two information sets inclu es inside information.The ability of insiders to profit from their informati n advantage is anexample of the adverse selection problem. Full and timely di closure will reducethis problem, thereby making financial reporting more use to investors andimproving the working of securities markets. Since reportin is costly, however,the inside information problem will still be present.In a capitalist economy, securities markets are the primary veh" de whereby capitalis raised and allocated to competing investment needs. Conse uently, it is sociallydesirable that these markets work properly in the sense that se urity prices shouldprovide correct values to guide the flow of investment funds We call a marketthat does this a properly working market. For example, a rm that has highexpected-value capital projects will be encouraged to invest in hem ifit receives ahigh price for its securities. Conversely, investment should be discouraged infirms that do not have high-expected-value capital projects. his will happen ifsecurity prices properly reflect underlying value. Of course, is is exactly whatsociety wants, since investment capital is in scarce supply. So ial welfare will beenhanced if scarce capital goes to the most productive alternat ves. :In the previous section, we formally faced up to the existe ce 4f informationasymmetry and, in particular, the problem of inside informat" on, ~hereby manIIEfficient surities Markets 109!agers and other insiders have an information advantage over utside investors. Itis not hard to see that this adverse selection problem operates gainst proper securitiesmarket operation, since insiders may withhold, delay, o bias the release ofrelevant information for their own advantage.It is also important to note that investors will be aware of this possibility.Then, a "lemons" phenomenon, as described in Section 4.6 1 for the used carmarket, would also come into play here. Investors would reco nize that the marketis not a "level playing field" and would either withdraw rom the market orlower the amount they are willing to pay for any security. The , firms with highqualityinvestment projects will not receive a high price for t eir securities, andthe market is not working as well as it should. If too many i vestors withdraw,the market becomes thin or, equivalently, it loses depth. A problem with thinmarkets is that investors may not be able to buy or sell all the want of a securityat the market price.Of course, developed capitalist economies have a variety of mechanisms forpromoting the proper operation of securities markets. One ch approach is toimpose penalties on the market. Thus, we witness governm nt securities commissionssuch as the OSC in Ontario and the SEC in the U "ted States. Theseagencies create and enforce regulations to, for example, con rol insider tradingand promote prompt disclosure of significant events, with pe alties for violation.However, the natural operation of a market can provide incentives for therelease ofinside information even in the absence of penalties fc r abuse. For example,a variety of mechanisms are available whereby firms with gh-quality investmentprojects can credibly communicate this to the market, thereby enhancingthe price they obtain for their securities. Signalling is one su h mechanism-forexample, insiders may retain a substantial equity position in n w projects, therebysignalling to the market their beliefs in their high-quality proj ct. The higher theproject quality, the greater the incentive to signal. Signals wi 1 be' considered ingreater detail in Chapter 12.For present purposes, a related incentive mechanism is disclosure. Firmswith high- (or low-) quality projects have an incentive to enga e in full disclosurein their financial reports.5 If such disclosures are credible, in estors will removethem from the "lemons" category and will, as a result, be wi ling to pay higherprices for their securities than they otherwise would.Obviously, penalty-based and incentive-based mechanis s are not mutuallyexclusive-we witness both in our economy. The penalty appr ach is like a "stick"and requires regulation to enforce it. The need for regulation will be reduced,however, to the extent that "carrots" are available to enable fir s that wish to doso to credibly reveal their information, thereby enabling the to receive a fairprice for their securities.We may conclude that the social benefits of properly wor ng ~ecurities marketswill be attained if the following two conditions are met:110 Chapter4I All relevant information is in the public domain, at leas up to the abilityof penalties and incentives to cost-effectively motivate e release ofinside information. Securities market prices are efficient relative to this infi rmation.4.8.1 INTRODUCTIONIn this section, we will consider two examples of accountingfirms may avail themselves to increase the quality of their finan ial disclosure. Thefirst is MD&A pursuant to OSC Policy Statement 5.10, issue in 1989. The secondis FOFI pursuant to Section 4250 of the GICA Handbook, so issued in 1989.Besides being of interest in their own right, these two standards provideimportant illustrations of how the amount of relevant inform tiort in the publicdomain can be increased. The MD&A standard is in between he carrot and stickapproaches to information release. It is required of firms to wh ch Statement 5.10applies. However, it is written in fairly general terms so that fir s have latitude inthe extent to which they release MD&A information. Also, i need only includeinformation available without undue effort or expense and that is not already clearfrom the financial statements. The second standard (FOFI) is oluntary; hence, itis a carrot approach. Section 4250 applies only if firms decide t release a forecast.The signalling implications of forecasting will be discussed in Chapter 12. Here,our interest is in its full-disclosure aspects.4.8.2 MANAGEMENT DISCUSSION ANDObjectives of MD&AStatement 5.10 requires certain firms in the OSC's jurisdic ion to prepare an''Annual Information Form (AIF) and Management's Discussi n and Analysis ofFinancial Condition and Results of Operations (MD&A)." Our coverage willconcentrate on MD&A.The MD&A requirements apply only to relatively large irms, that is, firmswith shareholders' equity greater than $10 million and with re enues greater than$10 million.According to the Introduction to Statement 5.10, the pri ary objective is to:enhance investor understanding of the issuer's business by pr viding supplementalanalysis and background material to allow a fuller u ders~anding ofthe nature of an issuer, its operations and known prospects far he fature.,,Efficient S~curities Markets 111Thus, we see that the objective of MD&A is derived from the decision usefulnessapproach introduced in Section 3.2. In particular, he emphasis is onassisting the user to assess the future prospects of the furn. ote that this is notaccomplished in Statement 5.10 through the financial state ents, nor even bydirect forecasts of future profits. Rather, the argument is:There are practical constraints on the amount of infarmation t at can be effectivelyconveyed in financial statements, which are subject tog erally acceptedaccounting and auditing standards. Important transactions, vents and conditionsare not always fully reflected in the financial stateme ts and some arenot easily expressed in dollar amounts. Additional disclosu e and analysisbeyond the financial statements is necessary to provide an a quate basis farassessment of an issuer's recent history and outlook far the fut re. This PolicyStatement requires such expanded disclosure.The "additional disclosure and analysis" referred to is oment discussion of its current financial statements and futureDisclosure RequirementsGenerally, the MD&A disclosure requirement is to disc ss those aspects ofthe financial statements and other statistical data that enhance he reader's understandingof financial condition, changes in financial condif on, and results ofoperations. More specifically, discussion is required under the ollowing aspects: Discussion of current operations and financial conditio Specific requirements are set out to disclose informatio on risks anduncertainties. This is consistent with our discussion oft e FASBConceptual Framework in Section 3.8, in particular wit SFAC l'ssecond objective of financial reporting, which includes a sessment ofthe uncertainty of prospective cash flows and dividends. Information about the nature and magnitude of "financi instruments"is required. In addition to bonds and shares, this term in udes mortgagebackedsecurities and hedging instruments. Thus, the P licy Statementrequires disclosure of financial instruments that typicall do not yetappear on financial statements proper in Canada, such as futures contracts,options, and swaps. This represents a start at bringing "o -balance-sheetfinancing" into the open. Many of the requirements are designed to help users inte pret the financialstatements. For example, "known trends" that will have a favourable orunfavourable effect on liquidity, capital resources, and con inuing operationsare to be described. Again, this assists the user to assess t e future liquidityand profitability of the firm.112 Chapter 4 The Policy Statement contains a number of provision~ to assist firms incomplying. Thus, firms whose securities are also traded in the United States(and hence must comply with the MD&A requireme ts of the SEC) cansubmit their SEC reports to the OSC in satisfaction o the OSC's MD&Arequirements. That is, they do not have to prepare the same informationtwice. Small firms do not have to comply. Firms whose MD&A disclosurescould put them at a competitive disadvantage can app for exemptionfrom reporting of"sensitive information." Also, firms eed'only includein MD&A information that is available "without und e effort or expenseand which does not clearly appear in the issuer's finan al statements."Finally, specific auditor involvement in the MD&A d sclosures is notrequired. Presumably, these conditions have been incl ded in response tomanagement concerns about the costs and sensitive n ture of compliance.Thus, StatementS.10 includes a combination of carro and stick mechanismsto promote information release, as discussed in ection 4.7.Discussion of MD&A DisclosureStrictly speaking, MD&A is not part of the financial state ents (this explainswhy MD&A is an OSC standard rather than a GICA andbook standard).Nevertheless, it is not hard to see that it is consistent wi h the spirit of theConceptual Framework, where the emphasis is on suppl ng information toenable investors to assess the future prospects of the enter rise. This forwardlookingapproach is evident in MD&A, for example in t e niquirements todescribe known trends and uncertainties.More generally, MD&A is consistent with the decision sefulness approach.The information that it provides should be helpful in bett r enabling users toassess the probabilities of future profitability, cash flows, or 'vidends from theirinvestments. Thus, the thrust of MD&A is to assist users i making their ownassessments, rather than providing these assessments directly.In addition, MD&A can improve the proper working of s curities markets tothe extent that it expands the set of relevant information avail ble to investors. Ineffect, it has the potential to reduce inside information.An Example of MD&A DisclosureExhibit 4.1 reproduces the risk and uncertainties portion of he MD&A in the2001 annual report of Mark's Work Wearhouse. The firm's s ares are traded onthe Toronto Stock Exchange; consequently, it is subject to SC requirements.The firm also provides extensive discussion of current opera ions and financialconditions, but this is not reproduced here.III'1Efficient Serities Markets 113Management's Discussion and AnalysisRISK AND UNCERTAINTIESTable 17 shows the external and internalrisk factors that affect the Company'sbusiness, and ultimately its profitability.Management's responsibility is tomitigate external risk factors to theextent possible, and to achieve an appropriatebalance among the internal riskfactors, in order to optimize profits.The consumer environment inCanada as reflected by the growth in totalretail sales and in specific segments withinthe total retail sector has been as outlinedin Table 18 over the last five years.As can be seen from Table 18, totalmen's wear sales have grown at a slowerrate than the growth in total retail sales andsales in men's clothing stores have declinedover the last five years. As well, total salesin women's clothing stores have grown at aslower rate than the growth in total retailsales and sales in shoe stores have declined.Thus, recent economic slow-downconcerns notwithstanding, Table 18 doesnot provide comfort that consumers willcontinue to purchase apparel at the ratesthey have historically. In fact, in recentRISK FACTOR Table 17ExternalConsumer environmentCompetionSeasonalityWeatherMerchandise sourcingForeign exchange ratesyears consumers have sh wn a markedpreference for bigger-ticke items such asfurniture, appliances, autos and electronics.The Company is confi ent that it hasmitigated this risk in its l ark's Divisionby having developed a stab e yet evolvingproduct offering, "On Co cept" stores,sound marketing progra s and is currentlygrowing its ladi s' wear andBusiness-to-Business sales apidly, and isdeveloping its e-Commerce sales in orderto continue growth in its ark's Divisionby increasing its market sh e in the men'swear, ladies' wear and footw ar markets inCanada. In addition, the C mpany introducedits "Corporate Store trategy" in itsWork World Division three years ago andwith eight pilot stores i testing theDOCKERS Stores conce t. The additionof the Work World D vision is alsocontributing to the Comp ny's growingmarket share in the segmen s of the retailtrade in which it operates i Canada andshould a DOCKERS Store roll out everbecome a reality, that would rovide a furthervehicle to increase th Company'smarket share in Canada.Interest ratesUnsolicited offer to purchase theCompany's outstanding Co mon. SharesSmall cap company in curren Canadiancapital marketsShare trading information114 Chapter4InternalCustomer serviceSales blendMarketing strategiesStore openings and closingsExpense rates in payroll, advertising,occupancy and systemsInventory levelsNumber and strength of franchise stores"Corporate Store Strate~ y'' in the WorkWorld Division LiabiJ :ties-to-equitylevelsThe introduction of new divisions undernew store banners, i.e., DOCKERSStores Division_ Foreign exchange expos1n-eCapital expenditure investments in stores Interest rate exposureand systemsRETAIL SALES GROWTH Table 18Percentage lncrease/(Decrease) ove;.- Prior PeriodMen's Total Wo1 inen'sTotal Clothing Men's Clo hing ShoeRetail* Stores* wear** Ste res* Stores*Year 2000 over Year 1999 6.3 0.1 3.8 2. 7*** (0.4)Year 1999 over Year 1998 5.8 (2.3) 2.9 .9 (2.6)Year 1998 over Year 1997 4/3 (0.2) 5.4 : .8 1.4Year 1997 over Year 1996 7.3 3.0 3.3 '. .6 (1.6)Year 1996 over Year 1995 2.4 (6.1) 3.9 ( .8) 0.4Year 2000 over Year 1995 31.3 (3.4) 15.4 l~.6 (7.2)* Statistics Canada** Trendex North America (includes men's wear sales in department stores, men's clothingstores and discount stores)*** Total sales growth in total ladies' wear which includes ladies' wear sales ir department stores,women's clothing stores and discount stores was 1.2% in 2000 over 1999 according toTrendex North AmericaCompetition in the men's wear apparelsector remains fierce as department stores,discount department stores, other discountstores, unisex stores, sporting goodsstores and men's specialty stores battle formarket share within this market sector.Many of these stores are large U.S.-basedretailers. Some mergers and subsequentstore consolidations are also occurringwithin the sector. Management feels thatit has mitigated this ri, k by keeping theCompany well-positioned in this marketsector by continually developing andintroducing new proc ucts to enhanceproduct selection for ts customers, byoffering products aero~ s all price pointsand by offering its custorri-ers differentgeographic shopping 101 atio~s through itsthree divisions (e.g., pc wer :Centres, stripmalls, regional malls, ~tc.)l Clearly, theCompany does not believe that it is isolatedfrom the effects of this competitionand it intends to continue to be rigorousin maintaining good relationships with itscustomers, protecting its businesses, generatingnew customers and continuing totest the introduction of new divisions withnew store banners.The Company's business remains veryseasonal with the fourth quarter of the lastthree fiscal years continuing to producebetween 37% and 39% of total systemannual sales and most of the annual profits,resulting from the general increase inconsumer spending in that period. Thesales reporting and merchandise planningmodules of the Company's informationsystem assist the Company in mitigatingthe risk and uncertainties associated withseasonal programs, but cannot removethem completely, as inventory orders,especially for a significant portion of offshorecommodities, must be placed wellahead of the season.Five years ago, approximately 33% ofthe Company's Mark's Division (theCompany's largest division) annual businesswas in seasonal commodities specificallyrelated to winter weather. Today theCompany's Mark's Division does 20% ofits annual business in seasonal commoditiesspecifically related to winter weatherand does 20% of its annual business inseasonal commodities specifically relatedto summer weather. While weatherdependency cannot he totally disassociatedfrom the Company's business, theCompany's Mark's Division has clearlyspread its winter risk between winter andsummer over the last five years. As theWork World Division matures, it will alsofollow this pattern. The DOCKERSStores Division is not a material part ofIIIEfficient Stcurities Markets 115the Company's sales a this time andbecause of the nature of ts assortments itis less weather dependen .In the area of merch dise sourcing,the Company has several ources of supplyfor most of its key comm di ties in order tobe able to provide a con inuous supply ofquality products to its c stomers. Whileshort-term interruptions could occur, theCompany continues to ork with both itsdomestic and foreign s urces, to ensurethat they have the ability nd commitmentto supply the Company s that customers'needs are met.As part of its offsho e sourcing practice,the Company advi es its importersnot to provide it with an goods producedin factories that use chil labour or unacceptablypaid or treated abour. For directimports, the Company sits and inspectseach factory it deals wit to determine ifthe factory employs chil or unacceptablypaid or treated labour. T e Company usesa comprehensive chec ist during eachinspection to ensure co pliance with itsethical sourcing policies. Nevertheless, theCompany cannot gua antee that suchactivities will not occur n the factories ofthe offshore suppliers wi h which it deals.The Company is als a member of theRetail Council of Cana a and the RetailCouncil's Executive Tra e Committee andbas adopted the volunta y code of ethicalsourcing developed by t e Retail Council.In addition, the Com any's CorporateCode of Conduct prohi its any employeefrom accepting gifts, fav urs or trips otherthan a nominal amount om anyone withwhom they deal on Co pany business.The Company's for "gn ~urrency riskis generally limited to urr~ncy fluctuationsbetween the Cana ian and U.S. dollars,as most of the Co papy's offshore116 Chapter 4suppliers conduct business in U.S. dollars.The Company has no U.S. dollar revenuesto use for the purchase of offshore commoditiesin U.S. dollars The Company'spractice is to enter into forward contractsfor over 50% of its anticipated U.S. offshorepurchases to help manage this risk.At January 27, 2001, the Company hadforeign exchange collar arrangements inplace for committed and anticipated foreignpurchases during the Company's nextfiscal year totaling $6,680,000 U.S. Underthe terms of the collars, the Companybears the exchange risk on foreign purchaseswhen the Canadian dollar tradesagainst the U.S. dollar within the rangesand for the time periods listed in Note 13to the Consolidated Financial Statements.At January 27, 2001, there were $102,204of unrealized gains on the foreignexchange collars based on the January 27,2001 exchange rate of $1.5063. See NoteslM and 13 to the Consolidated FinancialStatements.In addition, at January 27, 2001, theCompany had foreign exchange fixedcontract arrangements in place for committedand anticipated foreign merchandisepurchases during the Company's nextfiscal year totaling $14,064,500 U.S.Under the terms of the fixed contractarrangements, the Company has fixed itsexchange risk on foreign purchases at anaverage Canadian dollar to the U.S. dollarrate of $1.4738 ($20,728,260 Cdn.). AtJanuary 27, 2001, the unrealized gains onthese contracts were $456,866 based on aJanuary 27, 2001 exchange rate of$1.5063. See Note lM and 13 to theConsolidated Financial Statements.In fiscal 2001, the Company purchasedapproximately 58% of its merchandisefrom Canadian manufacturers inCanadian dollars (Marks Division 57%,Work World Division 6 % and DOCKERS Stores Division 60Jt&).The Company's inte est rate risk is aresult of its short-term oaring rate debtrequirements during pa t of every fiscalyear. Interest rate swap c ntracts are usedto hedge the interest rate risk on over 50%of the anticipated short- erm floating ratedebt requirements for th coming year. AtJanuary 27, 2001, the C mpany had fixedits borrowing rate on $20.0 million of itsanticipated short-term b rrowing requirementsat a 7.295% all-in ate and on $14.5million of its anticipate short-term borrowingrequirements at an all-in rate of6. 965% The mark-to-m rket value of theinterest rate swap cont acts is a $7,069unrecorded gain atJanu ry 27, 2001 basedon the Company's floa ing rate interestcost of 7.25% at Janua y 27, 2001. SeeNotes IM and 13 to he ConsolidatedFinancial Statements.Since the Compan is a public companywithout a manage ent control-shareblock, unsolicited offer to purchase theCompany's outstandin Common Sharescould appear from tim to time, as happenedduring fiscal 199 . This possibilitymay have a higher pro ability currently,given that institutional investors seem tobe totally disintereste in investing insmall cap stocks, and iven the earningsmultiple at which the ompany's sharesare currently trading. Se trading multiplesat the end of this sectio . While managementhas processes in lace to have theCompany's Board of irectors and nonoperationsmanageme t deal with suchmatters should they ar se; yhere is s riskthat such activities co ld distract operationsmanagement to th poi~t of affectingperformance and create xpenses which, incombination, could cause the Company tofall short ofits forecast range. See Forecast.The internal risk factors are oftentied together, and thus action taken tostimulate one factor often results in a negativeeffect on other factors: New store openings may increase sales,but, in the first year or two of operationsof a new store, the increase in payrollcosts, advertising costs, occupancy costsand interest costs may cause that storeto contribute an operating loss, until itbecomes a mature store from a sales persquare foot perspective. Additional advertising campaigns mayincrease sales, but not sufficiently in theshort term to cover the cost of theadditional advertising. Staff reductions can lower payroll costs,but may cause a loss of sales due tolower sales per customer and customerdissatisfaction with the level of salesservice and stock outages in the stores.Management believes that it is achievingan appropriate balance among the internalrisk factors in order to optimize profits.The Mark's Division franchise operationsconsisted of 25 franchise stores atJanuary 27, 2001, 88% of which meetCompany-set capitalization standards.During fiscal 2001, the Mark's Divisionpurchased four of its franchise stores andconverted them to corporate stores. Thisfranchise store purchase activity washigher than normal in fiscal 2001 as threefranchisees decided to retire during thatfiscal year and offered to sell their storesto the Mark's Division. The Mark'sDivision franchise operation is very stableand is expected to shrink a little over timewith the occasional franchisee selling hisor her store to the Corporation.IEfficient Secu1ties Markets 117IIWith a "Corporate Stdre Strategy"for new store openings (fuur in fiscal2001, five in fiscal 2000 and nine in fiscal1999) and the purchase of fr nchise storesas they become available ( ix purchasedfranchise stores in fiscal 200 , one in fiscal2000 and 31 in fiscal 1999 and the closureof non-performing fra chise stores,the Work World franchise peration hasreduced to 90 franchise sto es at January27, 2001from150 atJanu 25, 1997. AtJanuary 27, 2001, 51% of he remainingWork World franchises m et Companysetcapitalization standar s that weredeveloped after the Dece bet 1, 1996acquisition date of Work orld, as therewere no capitalization stand ds under theprevious administration. very year, theWork World Division intr duces at leasthalf a dozen or so new me chandise programsand continually see s to improveupon existing assortments i order to positivelyimpact a significa t part of themerchandise offering an , it is hoped,store sales in both the Work WorldDivision's franchise and cor orate stores.In addition, over the la t several years,the Company has put the ecessary creditcontrols in place to contr 1 the level ofmerchandise shipments an other cost riskservices provided to the Wi k World franchisees.Nevertheless, given the capitalizationlevel of many of these tores, there is arisk that more of the stor s could close,causing a loss of royalty and other revenuesand bad debt write-offs for the Company.In its purchased franchi e stores and inits new corporate stores, t e Work WorldDivision expects to generat the appropriatesales per square foot, gr ss margin rate,and expense rate to prod ce a 1 front-linecontribution higher than t e r