AFM 371 Winter 2008 Chapter 14 - Efficient Capital Markets 1 / 24
Outline Background What Is Market Efficiency? Different Levels Of Efficiency Empirical Evidence Implications Of Market Efficiency For Corporate Financial Management 2 / 24
Background AFM 271 largely focussed on spending money (capital budgeting), the l.h.s. of the balance sheet now we consider raising money, the r.h.s. of the balance sheet we will hold capital budgeting/corporate investment decisions constant (for now) and concentrate on determining the best financing strategy what generates positive NPV? typical financing decisions: timing/size of security issuance when (or if) to pay dividends compared to corporate investments, financing decisions are easier to reverse corporate investments: projects can have positive NPV because the firm is not typically facing a perfectly competitive market 3 / 24
Background Cont d financing: if securities issued by the firm are fairly priced, then the financing activities have NPV of zero key question: are securities fairly priced? example: you have a project which pays $1 per year in perpetuity discount rate is 10%, required initial capital is $5 what is the NPV of this project? if you contribute all of the equity, how much of the NPV do you have a claim to? suppose you only have $3 of capital, so you sell a 40% equity stake for $2 is the equity fairly priced? does this enhance NPV? 4 / 24
What Is Market Efficiency? an efficient capital market is one in which prices fully reflect available information efficiency here is informational efficiency a market in which information is widely and cheaply available to investors and all relevant information is impounded into security prices prices are right at any time any new information is quickly (and correctly) absorbed into prices 5 / 24
What Is Market Efficiency? (Cont d) example: reaction of stock price to new information that is good news Stock price over-reaction with reversion efficient market response to good news under-reaction -30-20 -10 0 +10 +20 +30 Days before (-) and after (+) announcement 6 / 24
What Is Market Efficiency? (Cont d) implications of the efficient market hypothesis: investors may hope for superior returns but all they can rationally expect in an efficient market is to obtain a return that is just sufficient to compensate them for the time value of money and for the risks that they bear firms expect fair value for the securities that they sell from The Economist, December 5, 1992: Because prices are efficient they reflect all available facts. Future prices differ from current prices only if buyers or sellers get new information. This, by definition, is random. But why should prices be efficient? Put simply, if they are not, it means the market is ignoring price-sensitive information. But this gives whoever has that information a chance to make big profits by trading on it. As soon as he does so, the overlooked information is incorporated in the price. This will make it efficient. 7 / 24
What Is Market Efficiency? (Cont d) example: triangular arbitrage in foreign currency markets suppose we can trade at the following prices: 1 e = 1.50 USD 1 USD = 1.01 CAD 1 CAD = 0.68 e how can you make arbitrage profits? a recent Oxford University study (D. Fenn, M. McDonald, S. Williams, S. Howison, and N.F. Johnson, Triangular Arbitrage in the Spot Foreign Exchange Market, October 2007) explored arbitrage opportunities between euros, USD, and Swiss francs and between euros, USD, and Japanese yen in 2005 how long do these opportunities last? what is their magnitude? 8 / 24
Different Levels of Efficiency 1 Weak form: security prices fully reflect all information contained in past prices changes in security prices are independent of what happened in the past 2 Semi-strong form: security prices fully reflect all publicly available information publicly available information includes historical prices, volume, accounting statements, etc. 3 Strong form: security prices fully reflect all information, public or private 9 / 24
Weak Form Efficiency and Random Walk the debate about market efficiency started with the discovery that security (and commodity) prices seemed to follow a random walk at any moment, it was not possible to tell what prices would be a moment later at the daily frequency, the best predictor of today s price is yesterday s price: P t = P t 1 + e t where e t is a random error term with E(e t ) = 0 over longer time periods (e.g. a year): P t = P t 1 + expected return + e t we can basically think of weak form efficiency and the random walk concept as being synonymous 10 / 24
Technical Analysis and Weak Form Efficiency if markets are weak form efficient, technical analysis is doomed to fail (since it presumes that patterns of past security prices can be used to predict future security prices) any predictable patterns will self-destruct Stock price Sell Sell Buy Buy Time 11 / 24
Semi-Strong Form Efficiency prices reflect all publicly available information: historical prices, trading volume, published accounting statements, news releases, etc. fundamental analysts study firm/industry fundamentals and try to judge whether a stock is under- or over-valued if a market is semi-strong form efficient: it is impossible to earn consistently superior returns by fundamental analysis markets and stock prices react exceptionally quickly (and correctly) to the release of information 12 / 24
Strong Form Efficiency prices reflect all information, public or private anything pertinent to the stock and known to at least one investor is already incorporated into the security s price if a market is strong form efficient, it is impossible to earn consistently superior returns from insider information note that strong form efficiency implies semi-strong form efficiency which in turn implies weak form efficiency 13 / 24
Common Misconceptions About the Efficient Market Hypothesis investors can throw darts to select stocks in efficient markets, on average investors will earn fair returns given the risk they take on investors need to consider their tolerance for risk when making portfolio decisions price fluctuations: people are sometimes skeptical about market efficiency because prices frequently change price changes are driven by new information, which arrives randomly (and very frequently) 14 / 24
Empirical Evidence there is an enormous amount of evidence about market efficiency, and it is mostly (but definitely not always) consistent with efficiency in general, there are three broad categories of tests: correlation tests are price changes random? Are there profitable trading rules? event studies does the market react quickly and efficiently to new information? performance of mutual fund managers the first one above relates to weak form efficiency, the second and third to semi-strong it is hard to test strong form, but what evidence there is suggests that it does not hold (insider trading is profitable) 15 / 24
Weak Form Tests - Serial Correlation basic idea is to examine the statistical correlation between past returns and current returns (on a single security) a significantly positive correlation would indicate momentum (e.g. positive returns tend to be followed by positive returns) a significantly negative correlation would indicate reversals (e.g. positive returns tend to be followed by negative returns) these tests have been done many times, in lots of markets, and they almost all show that markets are weak form efficient (reported correlations are almost zero) one caveat: these tests are typically done at the daily frequency, the evidence is much more mixed over longer periods (e.g. annual) also note that pyschological experiments show that people really cannot tell what is random they will see patterns in random noise 16 / 24
Semi-Strong Form Tests - Event Studies basic idea is to examine returns around the arrival of new information and look for evidence of under-reaction, over-reaction, delayed reaction, etc. returns are adjusted to determine if they are abnormal by taking into account what the rest of the market did on the same day one way to measure abnormal returns is to subtract market return R M : AR = R R M another way is to use the market model : AR = R (α + βr M ) the cumulative abnormal return is the sum of abnormal returns over some time period around the event date announcement: CAR = AR t 17 / 24
Semi-Strong Form Tests - Event Studies Cont d example: dividend omissions (S.H. Szewczyk, G.P. Tsetsekos and Z. Zandout, Journal of Investing, Spring 1997) Day AR CAR -7 0.146 0.146-6 -0.038 0.108-5 -0.180-0.072-4 0.218 0.146-3 -0.390-0.244-2 -0.239-0.483-1 -3.136-3.619 0-1.393-5.012 1-0.399-5.411 2 0.228-5.183 3 0.285-4.898 4 0.335-4.563 5-0.184-4.747 6 0.062-4.685 7 0.195-4.490 18 / 24
Semi-Strong Form Tests - Event Studies Cont d graphically: 0-1 CAR (%) -2-3 -4-5 -7-6 -5-4 -3-2 -1 0 1 2 3 4 5 6 7 Days relative to announcement 19 / 24
Semi-Strong Form Tests - Event Studies Cont d event study methodology has been applied to many events including: dividend changes earnings announcements merger announcements new issues of securities the studies generally conclude that there aren t profit opportunities available, i.e. markets are semi-strong form efficient in fact, many studies conclude that news tends to leak out in advance of public announcement dates some exceptions: continuing positive returns after good earnings announcements, IPOs tend to underperform over long periods of time 20 / 24
Semi-Strong Form Tests - Mutual Fund Performance expert money managers do not outperform the market consistently annual return performance of U.S. mutual funds relative to market index (1963 1998): All funds -2.13% Small company growth funds -8.45% Other aggressive growth funds -5.41% Growth funds -2.17% Income funds -0.39% Growth and income funds -0.51% Maximum capital gains funds -2.29% Sector funds -1.06% there is very little evidence of persistence in ability mutual funds do provide fairly cheap diversification and useful services such as record keeping 21 / 24
Semi-Strong Form Tests - Some Counter Evidence market crashes: the NYSE fell 22.6% on Oct. 19, 1987, after a weekend in which little information was released 3Com s divestiture of Palm: on March 2, 2000, 3Com sold a fraction of its stake in Palm to the public via an IPO for Palm 3Com retained 95% ownership, and announced (subject to IRS approval) that it would spin off the remaining shares of Palm to 3Com s shareholders before the end of the year 3Com shareholders would receive about 1.5 shares of Palm for every share of 3Com that they owned this gave two ways to buy Palm: for example, buy 150 shares of Palm directly, or buy 100 shares of 3Com (giving a claim to 150 shares of Palm plus a portion of 3Com s other assets) implication is that 3Com s share price had to be at least 1.5 times that of Palm s the day prior to the IPO, 3Com s share price closed at $104.13 the day of the IPO, Palm s price went from the offer price of $38 to a close of $95.06, but 3Com s price fell to $81.81 (it should have been at least $145!) note, however, that to exploit this, an arbitrageur would have to buy 3Com and short Palm (but it s difficult to take short positions right after an IPO) 22 / 24
Semi-Strong Form Tests - Some Counter Evidence Cont d pricing anomalies are statistical regularities that occur over long periods of time and across different countries and which cannot be explained by existing financial models some examples: size effect: small cap stocks tend to outperform large cap stocks January effect: most of the difference between small and large cap performance happens in January weekend effect: returns tend to be lower on Mondays, higher on Fridays (no longer true for large cap Cdn stocks), and especially high before trading holidays value vs. growth: stocks with high book-to-market equity ratios and/or high earnings-price ratios tend to outperform long term price reversals: the most extreme losers (winners) over the past 3-5 years tend to have very high (low) returns relative to the market during the following years 23 / 24
Implications of Market Efficiency if markets are efficient, then the following three implications arise: accounting and efficient markets: don t attempt to fool investors by cooking the books, they will see through it timing of security issuance: market efficiency implies that prices are correct, so that managers cannot deliberately sell over-valued securities (using publicly available information) price pressure effects: in an efficient market, firms can sell large amounts of securities without depressing prices the actual evidence on all three of these is somewhat mixed another important lesson: trust market prices. They reflect a lot of information. One recent example is the current ABCP crisis in Canada. For several years, yield spreads were substantially higher in Canada than in other markets for ABCP. This was a big warning signal the market was pricing ABCP as if it was substantially riskier than government bonds. 24 / 24