Chapter 13, Market Efficiency

1
Chapter 13, Market Efficiency
2
Financing Decisions/Efficient Markets

• Up until now, we have slipped in an assumption
  that Markets correctly price investments. If
  markets correctly price investments, then markets
  correctly price risk, and thus we can use the
  markets to know what premium should be associated
  with accepting risky investments.
• This leads to the idea of Efficient Markets. A
  market is efficient if all information is
  incorporated into the price of an asset. We will
  see three possible sets of information that will
  be considered, leading to three different forms
  of efficiency.

3
Three forms of efficiency

• Weak Form
• Prices reflect only past prices (and dividends)
• Semi-Strong
• Prices reflect all publicly knowable information
• Strong
• Prices reflect all information
• Hard to argue with Weak Form, hard to accept
  Strong form

4
Arbitrage (or a similar idea)

• One way that the markets will work to be
  efficient is through arbitrage. If someone finds
  a profit-making opportunity, they will take it
  (if its profitable)
• This arbitrage will move the market in a way that
  will incorporate the knowledgeable arbitragers
  information into the stock price
• Plott, Information experiment.
• These also depend on how rational investors are.

5
An aside about Random Walks

• A random walk is a term used in statistics to
  denote a certain process of generating subsequent
  terms in a sequence (in this case, daily stock
  prices). Each term in the sequence is equal to
  the previous term plus an error term that is
  statistically independent of any other error
  term. These error terms have an expected value of
  0 (and are independent of any other term)
• Example Flipping a coin, and counting the number
  of heads minus the number of tails.

6
Implications of Random Walks

• The Efficient Market Hypothesis implies that the
  error term ej1 is unpredictable. If we had
  information to predict ej1 we could (possibly)
  profit, but if we didnt have extra information,
  there would be no way to systematically profit.

7
Weak Form Efficiency

• Consider a stock of Ford which pays dividends of
  1, 2, 1, 2, ... starting one year hence. We will
  assume a discount factor of 10 throughout. What
  is this stocks price?
• This is the sum of two perpetuities, one that
  pays 1 every year, and one that pays 1 every
  other year. The value of the first perpetuity is
  1/0.1 10. The second perpetuity is worth
  1/0.21 4.76.

8
Without considering past prices

• Thus, if the fluctuations are known from past
  data, we will price the stock at 14.76. (Note,
  if we did this calculation beginning one year
  later, we would obtain 15.24 as the price).
• Weak Form efficiency says that the past cycles
  are taken into account, and thus when we see a 1
  dividend, we dont price the stock at just 10.
  The market then has taken the cyclical nature of
  the dividends into account, and thus profit cant
  be obtained from knowledge of the cycles.

9
Tests of Weak Form Efficiency

• Trading rules These are rules that make
  decisions based upon the past pricesif a stock
  rises by x buy it until it has lost x from its
  subsequent high. Fama and Blume (1966) showed
  that no value of x could beat a buy and hold
  strategy (after commissions). Some rules can be
  much more elaborate (almost an art form), but
  these generally dont work.

10
Tests of Weak Form II

• Serial Correlationsthe correlation between Rt
  and Rt1 can be calculated. If the Weak Form
  Efficiency Hypothesis holds, the correlation
  should be 0.
• Studies of daily returns usually show a small
  positive correlation, but it is too small of
  which to take advantage.
• The conclusion is that the market is efficient in
  the weak form.

11
Some people dont agree

• Some people dont seem to believe even in the
  weak form efficiency hypothesis. There are
  chartists and technical analysts who try to see
  where the market is going based upon past prices.
  They will see patterns in the (past) prices to
  determine which way they think the stock price
  will go.
• But I dont know many successful day traders or
  chartists! In fact, a major trading house (Smith
  Barney) recently (a year ago) released its
  entire technical analysis division

12
Semi-Strong Efficiency

• Tests for this form efficiency look at timing of
  stock price adjustments with respect to events.
• For instance, when earnings are announced, new
  information is available about the price of the
  stock. Almost the full adjustment (90 for
  increased earnings, 76 for decreased earnings)
  takes place in the first three hours of the
  announcement. Also, the Chinese yuan
• RWJ has more tests of this sort.

13
Mutual Funds

• Mutual funds do not outperform the market
  consistently, even though they have immense
  resources to look through all publicly available
  information. This supports semi-strong form
  efficiency.
• So why do we hear about (mostly) good mutual
  funds? Those that do bad lose money and have
  investors leave, so cease to exist. Also, if a
  company starts lots of small funds, some of those
  are going to do well by chance. Those funds can
  then be touted, while the others can be folded
  into the successful fund.

14
Sell Ratings on Stocks
WSJ, Monday, April 11 2005, C1
15
Strong Form Efficiency

• Strong Form Efficiency implies that ALL possible
  information about a stock is reflected in its
  price.
• This is too extreme to hold in practice.

16
Implications of Efficient Markets

• Markets have no memory
• There are no financial illusions
• Trust market prices
• Cant time markets
• This means that trying to time issuing debt or
  IPOs cannot create value.

17
Some (small?) anomalies

• Size (Market Cap)
• Value (Book versus Market)
• Seasonal Patterns
• January Effect
• Day of the week Effect