FIN 355 (24) - Market efficiency and anomalies (module 7).pptx

FIN 355 Principles of Investments Efficient Markets, Predicting Returns, Anomalies Lecture Preview Big-Picture Motivation: There are millions of investors that are interested in making money via stock investments. As discussed earlier in the semester, prices can be modeled as the present value of...

FIN 355 Principles of Investments Efficient Markets, Predicting Returns, Anomalies Lecture Preview Big-Picture Motivation: There are millions of investors that are interested in making money via stock investments. As discussed earlier in the semester, prices can be modeled as the present value of expected future cash flows. These expectations are based on currently available information. The overarching discussion topic for this lecture is what information is (already) incorporated into these expectations? The potential upside to thinking about this question is that if we can identify information that is not fully incorporated into current prices then we could potentially trade on this information. This discussion has broad implications for active versus passive approaches to investing. Discussion Outline: • What information is incorporated into prices? • Are markets “efficient”? 3 types of market efficiency • Anomalies, examples where prices seem to incorporate the wrong information From earlier in the semester • The price of a security can be thought of as the present value of all the cash flows tied to that security. • The cash flow expectations are based on all sorts of information (employees, customers, firm-level, industry-level, economy-wide, etc.) • Earlier in the semester we talked about specific examples of price changes that occurred when new managers, mergers, and fraud were announced. What types of information can affect expected cash flows?  Management Changes March 14, 2005: Autozone announced a long-time employee of the firm would be the next CEO. The stock dropped more than 12% (the S&P500 went up 0.5% on the same day). Aug 20, 2012: Best Buy announced that Hubert Joly, an outsider, would be the new CEO. The stock price dropped more than 8% (the S&P500 went up 0.19% on the same day). July 25th, 2012: Symentec announced a new CEO. The stock price rose 13% (the S&P500 went down 0.04% on the same day). Important Intuition: The current stock price incorporates investors’ opinions (good and bad) about the talent of current management. What types of news can affect prices?  Fraud March 2003: $1.3 billion fraud announced at Healthsouth. The stock price lost almost all of its value when the fraud was announced. Important Intuition: The current stock price incorporates investors’ opinions and fears about fraud. Current stock price adjustments are often far larger than the dollar value of the announced fraud. What type of news can affect prices?  M&A Jan 27, 2011: Verizon announced its intent to acquire Terremark for $19.00 per share. Terremark’s share price jumped from $14.05 to $18.92 per share. On the same day that Terremark went from $14.05 a share to 18.92 a share, Savvis’ (a competitor of Terremark) share price jumped from $26.57 to $30.26. Important Intuition: The current stock price incorporates investors’ expectations for how acquisition activity might affect future cash flows. These expectations are broad enough to incorporate even the likelihood that a firm will become a target in an acquisition even before a specific announcement has been made about that firm. Cumulative stock returns around M&A announcement 0.7 0.6 R e t u r n s Terremark Savvis S&P500 0.5 0.4 0.3 0.2 0.1 0 1- 6- 9- 14 17 22 28 31 5- 10 13 19 24 27 1- 4- 9- 14 17 23 28 De De De -D -D -D -D -D Ja -J -J -J -J -J Fe Fe Fe -F -F -F -F c- c- c- ec ec ec ec ec n-1 an an an an an b- b- b- eb eb eb eb 10 10 10 -1 -1 -1 -1 -1 1 -1 -1 -1 -1 -1 11 11 11 -1 -1 -1 -1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 If prices already incorporate all this information, what information is left that can be used to predict future returns? Efficient market hypothesis (EMH): Security prices are “efficient” in that they already reflect information. “An informationally efficient market is one in which information is rapidly disseminated and reflected in prices.” If prices are efficient, then it is difficult to predict future price changes (i.e., future returns) using current information because the current information is already, to some extent, reflected in the current price. Important questions: • Which information is already in the price? Public? Private? • How does the information get into the price? • How quickly does information affect the price? Which information is already in the price? Public? Private? People sometimes use 3 categories to talk about what information might already be in the price. Weak-form market efficiency: Past public trading information is already in the price (price trends, volume, shorting, etc.) • Implication: Technical analysis should not work Technical analysis “Technical analysis attempts to exploit recurring and predictable patterns in stock prices to generate superior investment performance. Technicians do not deny the value of fundamental information, but they believe that prices only gradually close in on intrinsic value.” Technical traders have developed various measures to assist in trying to identify patterns. These measures focus on different price and volume trends and can use various calculations including moving averages, relative strength, sentiment indicators, etc. Many of these patterns are identified using charts. Example technical analysis charts Which information is already in the price? Public? Private? People sometimes use 3 categories to talk about what information might already be in the price. Weak-form market efficiency: Past public trading information is already in the price (price trends, volume, shorting, etc.) • Implication: Technical analysis should not work Semistrong-form market efficiency: All public information is reflected in the price • Implication: No past or present publicly available information should predict future price changes. Fundamental analysis should not work. Fundamental analysis “Fundamental analysis uses earnings and dividend prospects of the firm, expectations of future interest rates, and risk evaluation of the firm to determine proper stock prices. Ultimately, it represents an attempt to determine the present value of all the payments a stockholder will receive from each share of stock. If that “intrinsic value” exceeds the stock price, the fundamental analyst would recommend purchasing the stock.” Fundamental analysis would include a study of past and current financial statements, management guidance, analyst reports, industry analysis, etc. Example of an online article based on fundamental analysis 14 Value = With a two stage model: Dividends are assumed to grow at a high rate (g1) for the first N periods and then to grow at a lower stable rate (g2) after N periods. Price0 Price0 DN is the “N”th dividend. DN*(1+g2) is the “N+1” dividend. is the PV of a constant growth model in year N dollars. What were analysts saying about Crocs in early 2023? 75-$175 per share These summaries were from • MarketWatch (top left) • Benziga (top right) • Finviz (bottom) Links shown in notes 16 Which information is already in the price? Public? Private? People sometimes use 3 categories to talk about what information might already be in the price. Weak-form market efficiency: Past public trading information is already in the price (price trends, volume, shorting, etc.) • Implication: Technical analysis should not work Semistrong-form market efficiency: All public information is reflected in the price • Implication: No past or present publicly available information should predict future price changes. Fundamental analysis should not work. Strong-form market efficiency: All public and private information is reflected in the price • Implication: There should not be price reactions to announcements of previously private information Cumulative returns around M&A announcement 0.7 0.6 0.5 R e t u r n s Terremark Savvis 0.4 S&P500 0.3 0.2 0.1 0 1- 3- 7- 9- 13 15 17 21 23 28 30 3- 5- 7- 11 13 18 20 24 26 28 1- 3- 7- 9- 11 15 17 22 24 28 D D D D -D -D -D -D -D -D -D Ja Ja Ja -J -J -J -J -J -J -J Fe Fe Fe Fe -F -F -F -F -F -F ec ec ec ec e e e e e e e n n n a a a a a a a b b b b e e e e e e -1 -1 -1 -1 c- c- c- c- c- c- c- -11 -11 -11 n-1 n-1 n-1 n-1 n-1 n-1 n-1 -1 -1 -1 -1 b- b- b- b- b- b0 0 0 0 10 10 10 10 10 10 10 1 1 1 1 1 1 1 1 1 1 1 11 11 11 11 11 11 Important Intuition: The fact that Terremark’s stock price surged when Verizon announced its intent to acquire Terremark shows that prices can’t be completely strong form efficient. Another way to say this is that we would not see strong price reactions when private information is made public if the private information was already fully incorporated in the price. Forms of market efficiency Forms of market efficiency Weak form Semistrong form Strong form* Market prices reflect: Past market Public data information X X X X X Private information X If markets are weak form efficient then we should not be able to use past market data involving past price changes, volume, etc. to predict future price changes. If markets are semistrong form efficient then we should not be able to use any publicly available information (market data or fundamental) to be able to predict future price changes. If markets are strong form efficient then we should not be able to use public or private information to predict price changes. How does the information get into the price? Why do prices react to news (new information)? • Once information becomes available, market participants analyze it looking for profit opportunities. • Competition for profits assures that prices reflect information Important Intuition: It is active competition by many investors that causes the various investors’ insights to all be part of the current price. Prices, in competitive markets, are “information rich”. The active competition for profits by a large group of people ensures that prices reflect a comprehensive information set. Competition is the mechanism that leads to efficient prices. Note that no single individual or group needs to know all the information. It is the aggregate trading that causes the aggregate information to be incorporated into the current price. How quickly does information affect the price? It depends. As discussed in your textbook it seems to take longer to affect small firms, firms with little or no analyst following, and private firms. • Small firm effect • Neglected firm effect • Public vs private How quickly does information affect the price? Even 20+ years ago studies suggested that market reactions occurred within minutes. Evidence of quick price reactions Figure 11.2 from your textbook. This figure shows stock price reactions to CNBC reports. The plot shows the average return reaction to the news coverage. Video showing speed of market reaction 20+ years ago Background information for video • Time zero is when the ticker symbol appears onscreen. • Transaction prices are plotted in red. Mid-quotes (average of the bid and ask prices) are plotted in white. • There were 29 trades from -15 minutes to time 0. • There were 125 trades from 0 to 15 minutes (29 in the first minute) • The average trade size is roughly $70,000. • $1.7 million traded from time -15 to 0. • $9 million traded from 0 to 15 minutes (with $2.4 million in the first minute). Link to Video: https://www.youtube.com/watch?v=FC7kd6dTw9c Random walk and the EMH Changes in stock prices are often described as a “random walk with drift”. In other words, stock price changes appear almost random and not predictable. If markets are “efficient” then tomorrow’s price change is independent of what happened to today’s prices. For a brief video that explains the “random see the Investopedia link in the notes. walk” Random walk and the EMH Changes in stock prices are often described as a “random walk with drift”. In other words, stock price changes appear almost random and not predictable. If markets are “efficient” then tomorrow’s price change is independent of what happened to today’s prices. Important Intuition: Why would future price changes be “random”? Many intelligent investors constantly compete from moment-to-moment to discover and invest using relevant information. If current prices are efficient (i.e., they already account for public information) and the new information flow in the future is random then future price changes will also be random. If today’s price already incorporates all the current information, then tomorrow’s price change is not because of today’s news… The above statement implies that you cannot predict the future movements of prices using the history of prices or past information – because past information is already impounded in the price today. Hence the idea of efficient markets implies that future price changes are independent from today’s price changes. But some markets, and some securities’ prices, are likely more efficient than others. This suggests that some of the best profit opportunities might be in those markets, or with those assets, whose prices are less efficient than others. Summary of evidence for market efficiency • It is hard to predict future returns • We observe strong price reactions to announcements that often quickly stabilize • Prices often move in advance of public announcements suggesting that private information is affecting the price even before it is made public • Random walk nature of price changes Implication from EMH arguments: • You can’t use past and/or current information to consistently predict future price changes • In the 1970s Princeton professor Burton Malkiel claimed in his famous book “A Random Walk Down Wall Street” that “A blindfolded monkey throwing darts at a newspaper’s financial pages could select a portfolio that would do just as well as one carefully selected by experts.” • The WSJ tested this for several years by having WSJ staff throw darts at a board and then observe whether their stock picks did as well as investment professionals. How does the idea of efficient markets relate to…? • Fundamental analysis • Technical analysis • Information signals (i.e. insider transactions, share repurchases, earnings surprises, etc.) • “Common sense” strategies for identifying firms that are undervalued in some way. For example: relative to peers, 52-week lows, P/E ratios, dividends, size, etc. • Warren Buffet What evidence is there of (predictable) non-random price changes? Tests of EMH: • Weak form: Do past price movements predict future price movements? • Semi-strong form: Does public information predict future price movements? Do we observe prices moving at the time new information is made public? • Strong form: Can people with private information predict future price movements? Weak form tests – some evidence on predictive power of past price movements Serial correlation in returns? • Some evidence for very small/weak positive serial correlation in short-term returns on indices. Stronger evidence of serial correlation at the 3 -12 month horizon. This result is sensitive to what sample period is used. • Small negative serial correlation in returns on individual securities in some samples but this result also depends on what sample period is used. • Some evidence for reversal effects over certain horizons after major events. • Some evidence for momentum in the worst and/or best performing portfolios over certain horizons in some years. Whatever short-term correlations/patterns exist in stock price movements are so small that they don’t typically lead to trading profits. This is generally consistent with weak form efficiency predictions. The longer-term correlations (and reversals) suggest that some patterns may exist that are inconsistent with weak form efficiency predictions, but these patterns seem to disappear in some time periods making it hard to draw strong conclusions. 31 Semistrong tests – evidence of predictive power of publicly available information (looking for market anomalies) What do market anomalies look like? • Persistent patterns in returns not explained by risk but related to publicly available information. What does it mean to be “explained by risk”? What is an “abnormal return”? • We have models of what the risk-return relation looks like. So after we measure the amount of risk in an investment the model can tell us what the “expected return” should be for that level of risk. An abnormal return is the observed return above and beyond (or below and beyond) the return predicted by the risk-return model. Anomalies are persistent patterns in returns not explained by risk. 32 Examples of common “market anomalies” • P/E Effect: Low P/E firms tend in some years to have better long-term performance than high P/E firms. • Small Firm Effect (January Effect?): Small firms tended to have superior returns movements in January. Neglected firm? Liquidity? • Value versus growth stocks: Firms with low market-to-book ratios perform well on average in many years. • Neglected Firm: Neglected firms are those that receive less attention from large institutional investors. January effect strongest amongst neglected firms. • Post-earnings announcement drift: Post-announcement drift in returns. (see figure) *** Note that these anomalies are not always consistent. Some appear intermittently over different years. Some (e.g., the small firm effect) change after they receive increased attention. Figure 11.5 in your textbook. The plot shows how stocks tend to drift in the same direction as their earnings surprises over time. Day 0 is the date that the quarterly earnings announcement is made. The figure shows that the firm’s stock price tends to drift in the direction that the earnings surprise will be (i.e., positive drift before positive earnings, negative drift before negative earnings) and then the price tends to continue drifting in the same direction after the announcement. Stock market reactions to announcements of dividend changes, earnings, and mergers can also be viewed as being consistent with semistrong form efficiency and against strong form market efficiency. • Announcements of dividend changes • Earnings • Mergers and acquisition activity There are examples of over and under reaction to these types of announcements, but the market overreacts as often as it underreacts and on average get it seems to get it right. “Strong form” efficiency Evidence for (& against) strong form efficiency: • (For) Run-up in short interest prior to fraud revelation • (For) Run-up in prices prior to merger announcement • Corporate insider sales signal declines in prices (some samples) • Corporate insider buys signal increases in prices (some samples) • (Against) Investors with inside information (like politicians) appear to earn abnormally good returns. • (Against) Strong price reactions to important announcements about earnings, guidance, law suits, mergers, etc. 36 Price increases of target firms’ stock prices before the acquisition announcement… Figure 11.1 from your textbook. This plot shows the surge in prices (positive returns) at target firms. Day 0 is the public announcement date. The run-up in price in the days before the official announcement may be evidence of private information leaking to a small subset of people before the private information is released to the general market. Summary of efficiency discussion It appears the market generally is not strong form efficient. This conclusion follows because if you have inside information you can earn abnormal returns on your investments and stocks do react to announcements. I personally would describe the market as generally being semistrong efficient but with occasional obvious (at least in hindsight) examples of misvaluation. Markets are not so efficient that there are not valuable opportunities for some lucky/sophisticated investors. Active traders would argue that markets are not efficient. 38 Prices in public markets are generally efficient but there are exceptions… • Portfolios of the best and worst performing stocks often exhibit momentum over 3 -12 month periods. • Prices react dramatically to news announcements (e.g., earnings announcements, merger announcements, etc.) but in some cases the prices continue to drift in the direction of the announcement over subsequent days. In some cases the prices drift in the direction of the announcement in the days leading up to the announcement. • Other market anomalies exist • Some investors earn incredible returns over time • Investors with inside information appear to earn abnormally good returns in some studies. • Why do senators’ portfolios significantly outperform the average investors’ portfolio? ( “Abnormal Returns from the Common Stock Investments of the US Senate” by Ziobrowski, Cheng, Boyd, and Ziobrowski in JFQA). • Also see “Lawmakers’ committee assignments and industry investment overlap” (2010 Washingtonpost) Evidence that doesn’t support complete price efficiency – perhaps “limits to attention”? • • • • • Repeated news announcements Friday announcements High news days Bubbles Hard-to-explain price changes and market reactions where investors trade on news with the wrong company • GameStop • Twitter vs Tweeter Oct 2014 (Tweeter increased ~700%) • Graco Inc vs Graco Children’s Product Inc • Nest Labs vs NEST • Castle Convertible Fund (ticker CVF) dropped 32% in 1 day due to Financial Times article about Czech Value Fund abbreviated CVF being investigated for fraud. • Other anomalies • Time of day matters (see next slide) Bubbles and market efficiency Looking back across history in retrospect there appear to be many examples of “bubbles” in valuation. A few famous examples include: • Tulip bubble in Holland 17th century • South Sea bubble in England 18th century • Railway bubble 19th century • US stock market in the 1920’s • Japanese stock and assets in the 1980’s • Dot-com bubble • Housing bubble Do security prices represent rational, unbiased assessments of intrinsic value during bubbles? Most bubbles only become obvious in retrospect. They are difficult to recognize in real time in the presence of new technology/paradigm shifts. Even if you suspect that prices are too high it is hard to time the end of the bubble. Evidence that doesn’t support complete price efficiency – perhaps “limits to attention”? GameStop Twitter vs Tweeter Oct 2014 • Tweeter increased ~700%) Graco Inc vs Graco Children’s Product Inc • one company’s announcement to recall products caused the other to decrease in value. Nest Labs vs NEST • NEST went from a penny to ~ 10 cents after Google said it would buy Nest Labs Castle Convertible Fund (ticker CVF) dropped 32% in 1 day due to Financial Times article about Czech Value Fund abbreviated CVF being investigated for fraud. Source: WSJ May 28 2015 “Stock-Market Traders Pile In at the Close” by Dan Strumpf “More than one in six trades in S&P 500-listed stocks took place between 3:30 and the 4 p.m. closing bell last year... For shares of smaller companies, 19.3% of trades were in the final 30 minutes…. That is important because it can often be difficult for investors to trade smaller-company stocks without pushing the price up or down.” 44 Alternative and non-liquid asset classes • The evidence suggests that small, neglected, and less transparent firms all exhibit evidence of information affecting prices in a slower manner, sometimes, than larger, more famous, and more transparent firms. • This suggests that there might be profitable investments to be found among alternative and non-liquid asset classes. • This conclusion is reinforced by the observation that competition among many players appears to drive measures of efficiency among investments in public equity. So if the alternative assets and/or the less liquid assets have fewer investing groups this may mean that there are still profit opportunities to be found. What are the implications for portfolio management? Question: If the EMH along with its characterization of stock price changes as being random (due to intense competition) is (mostly?) true, does this suggest that there is no place for portfolio managers? Answer: No! Portfolio managers and personal wealth advisors can add value… • Achieving diversified positions with targeted outcomes specific to our risk preferences • Help investors make strategic asset allocation choices • With advice on how best to deal with tax considerations • Providing information about how retirement considerations may relate with investment decisions • Helping investors understand their options What are the implications for stock market analysts? If markets are (mostly?) semistrong efficient, do stock analysts add value on average? • A Womack study found that positive changes in buy recommendations are associated with increased stock prices of about 5%. Negative changes result in average price decreases of 11%. • Are price changes due to analysts’ information or due to pressure brought on by the recommendations themselves? • Circular logic: Without analysts prices would become less efficient creating a demand for analysts…  Stock analysts help find and summarize key information for investors. Their activities help prices be more efficient. Concept Checks 1. Why might stock price changes appear “random”? 2. Does this imply that price levels are “random”? 3. Are markets strong form efficient? How do you know? 4. What does it mean if prices move in non-random ways before a major announcement? 5. Do investors show evidence of limits to attention? What evidence shows this? 6. When are most trades enacted? For next time.. Review the concepts and terminology introduced in today’s lecture and check the online schedule for the assigned reading and upcoming assignments. Terminology You should be able to give a 1-2 sentence description of each of these terms. • • • • • • Market efficiency 3 forms of efficiency Market anomalies Evidence for and against efficient markets Fundamental analysis Technical analysis • What advantages might private assets offer in a broader portfolio? • Are markets efficient? • What are the implications of market efficiency arguments for technical analysis? Fundamental analysis? Alternative asset classes? Thank You