Capital Market Efficiency The Empirics 4 basic traits

Capital Market Efficiency The Empirics

4 basic traits of efficiency • An efficient market exhibits certain behavioral traits. We can examine the real market to see if it conforms to these traits. If it doesn’t, we can conclude that the market is inefficient. 1. 2. 3. 4. What if? Act to new information quickly and accurately Price movement is unpredictable (memory-less) No trading strategy consistently beat the market Investment professionals not that professional Definitions Implications Price Empirics

Empirical Strategies • Look at the historical data. See if they conform with the 4 traits What if? Definitions Implications Price Empirics

Stock price ($) 1 st trait: reaction to news Early Reaction Delayed Reaction Days relative to announcement day -t 0 +t The timing for a positive news What if? Definitions Implications Price Empirics

1 st trait: reaction to news Event study “one type of test of the semi-strong form of market efficiency to see if prices reflect all publicly available information or not. ” • To test this, event studies examine prices and returns over time (particularly around the arrival of new information. ) • Test for evidence of [1] under reaction, [2] overreaction, [3] early reaction, [4] delayed reaction around the event. • If market is “semi-strong-form efficient”, the effects of an event will be reflected immediately in security prices. Thus a measure of the event’s economic impact can be constructed using security prices observed over a relatively short time period. • Some examples of events include mergers and acquisitions, earnings announcements, issues of new debt or equity, and announcements of macroeconomic variables such as the trade figures. What if? Definitions Implications Price Empirics

1 st trait: reaction to news Procedures of a event study: • Identify the event of interest. (e. g. , Merger Acquistion) • Define the event period? (e. g. , -10 days and +10 days) • Select the sample (e. g. , firms which have in common the incidence of the event of interest) • Measure the impact by defining abnormal return • Estimate the parameters needed to calculate expected returns. • Calculate cumulative abnormal returns What if? Definitions Implications Price Empirics

1 st trait: reaction to news • The event period should start before you think the event has an effect on the stock price. As an example, for merger announcements, a typical choice is from 25 trading days before the announcement day to 25 trading days after the announcement day • The estimation period should be a period right before the event period. For merger announcements, a typical choice is 100 trading days before the start of the event period -125 -25 Estimation Period What if? Definitions Implications 0 +25 Event Period Price Empirics

1 st trait: reaction to news • Abnormal return = Actual realized return – Expected return E. g. , E(rj|r. M, t) = a 0 + ajr. M, t (Return on security j conditional on the return on market) εj, t = rj, t – E((rj|r. M, t) • Cumulative abnormal return: CARj, t = ∑-Tt εj, t • (Aggregate abnormal returns from –T to t) Average cumulative abnormal return over a sample of securities: Average CARt = (∑j CARj, t)/J (where J = no. of securities in the sample) • Plot the graph, examine the pattern. Of course, perform hypothesis testing as well. What if? Definitions Implications Price Empirics

1 st trait: reaction to news Efficient market response to “bad news” Source: Szewczyk, Tsetsekos and Santout (1997) What if? Definitions Implications Price Empirics

1 st trait: reaction to news -29 0 30 How stock splits affect value? Source: Fama, Fisher, Jensen & Roll (1969) What if? Definitions Implications Price Empirics

1 st trait: reaction to news Announcement Date What if? Definitions Implications Price Empirics

1 st trait: reaction to news Average Cumulative abnormal return Announcement Date for quarterly earnings reports Days relative to Announcement Date Source: Remdleman, Jones and Latane (1982) What if? Definitions Implications Price Empirics

1 st trait: reaction to news • Event study methodology has been applied to a large number of events including: – – – Dividend increases and decreases Earnings announcements Mergers Capital Spending New Issues of Stock • The studies generally support the view that the market is semi-strong from efficient. • In fact, the studies suggest that markets may even have some foresight into the future—in other words, news tends to leak out in advance of public announcements. What if? Definitions Implications Price Empirics

2 nd trait: Random price movements • Studies of serial correlation • Studies of seasonality – Day of the week effect – January effect What if? Definitions Implications Price Empirics

2 nd trait: Random price movements Studies of serial correlation H 0: Cov(ΔPt, ΔPt-i) is significantly different from zero or not, for i ≠ 0 Alternatively, H 0: Cov(Δrt, Δrt-i) is significantly different from zero or not, for i ≠ 0 • Plot the following types of graph. • Note: Statistically significant ≠ Economically significant – If you are aware of the correlation, and attempt to trade on the basis of it, brokerage commissions may make your expected profits negative. What if? Definitions Implications Price Empirics

Return on day t+1 (in %) 2 nd trait: Random price movements Return on day t (in %) What if? Definitions Implications Price Empirics

2 nd trait: Random price movements Return on week t+1 (in %) FTSE 100 (correlation = -0. 08) Nikkei 500 (correlation = -0. 06) DAX 30 (Correlation = -0. 03) S & P Composite (correlation = -0. 07) Return on week t (in %) What if? Definitions Implications Price Empirics

2 nd trait: Random price movements Studies of seasonality – Day of the week effect – French (1980) and Gibbons & Hess (1981) – Using S&P 500 index to proxy returns of stocks for each of the 5 trading days of the week. – Found Monday returns are on average lower than returns on other days. – If transaction costs are taken into account, however, trading rule based on this pattern fails to generate abnormal returns consistently. – But you may consider this effect in timing your own purchases and sales. What if? Definitions Implications Price Empirics

2 nd trait: Random price movements Studies of seasonality – – The January effect Keim (1983) and Roll (1983) The most mystifying seasonal effect. Stock returns, especially returns on small stocks, are on average higher in January than in other months. – Moreover, much of the higher January return on small stocks comes on the last trading day in December and the first 5 trading days in January. What if? Definitions Implications Price Empirics

3 rd trait: Superior trading strategy • Caveat - Be careful here!!! It’s in the interest of those who find such rules to hide them rather than publicize them. • Price-to-earning ratio. (P/E Ratios) • Size effect What if? Definitions Implications Price Empirics

3 rd trait: Superior trading strategy • Price-to-earning ratio. (P/E Ratios) • The trading rule of “buying stocks that have low price-to-earning ratios , and avoiding stocks with high price-to-earning ratios” seems to consistently outperform the market. • Question: – 1) what does it mean by low P/E ratio? – 2) Survivorship bias? What if? Definitions Implications Price Empirics

3 rd trait: Superior trading strategy • Size Effect. (Banz (1981)) • Small firms tend to have higher returns as compared to larger firms. • The trading rule of “buying stocks of smaller firms” seems to consistently outperform the market. • Question: – 1) Is there any inherent risks of small firms not captured by risk measures? – 2) Is it because transaction cost of smaller firms’ stocks are more expensive (due to thinner market)? What if? Definitions Implications Price Empirics

4 th trait: professional investors? • If the market is semi-strong form efficient, then no matter what publicly available information mutual-fund managers rely on to pick stocks, their average returns should be the same as those of the average investor in the market as a whole. • We can test efficiency by comparing the performance of professionally managed mutual funds with the performance of a market index. Evaluating mutual funds performance. (Jensen (1969)) • Managers of mutual funds are usually highly trained and have access to broad sources of investment information. • Thus, if their managed mutual funds consistently outperform the market, then we conclude that such evidence is against the market efficiency hypothesis What if? Definitions Implications Price Empirics

4 th trait: professional investors? • Using S & P 500 as proxy for the market, estimate the security market line. • Estimate the beta for each mutual funds. • Plot the mutual funds on the security market line graph (NOTE: net of all expenses) What if? Definitions Implications Price Empirics