Empirical Evidence for the Efficient Market Hypothesis

Only available on StudyMode
  • Download(s) : 374
  • Published : May 26, 2013
Open Document
Text Preview

Outline various versions of Efficient Market Hypotheses. Discuss whether there is sufficient empirical support for each of these hypotheses. The efficiency of financial markets has long been a contentious issue, and as financial markets have evolved both in their breadth and complexity the question whether financial markets can effectively and efficiency allocate resources has never been more relevant. In this essay I intend to investigate the validity of the various forms of the Efficient Market Hypothesis (EMH) using empirical evidence from various studies; and attempt to determine whether any of these forms of the EMH are accurate in describing the workings of international financial markets. Traditional finance textbooks have long offered three ‘versions’ of informational efficiency of financial markets: Weak, Semi-Strong and Strong, with the definitions of these ‘versions’ relatively settled. I will firstly outline these versions and then evaluate the evidence to determine their validity. The Weak form of the EMH asserts that financial markets efficiently process all past prices of a financial asset which are reflected in its current price. Furthermore, it implies that asset prices follow a random walk process. This renders technical analysis futile as all information contained in previous prices has been efficiently priced in. Formally: ( ) ( )

The weak form of the EMH has had a substantial amount of research into testing its validity, in particular using econometric analysis. In addition, several observable phenomena have been presented as evidence against the weak form of the EMH. The ‘December Effect’ is an empirical observation that during the month of December stocks generally outperform when compared to the rest of the year, this effect has been long observed and appears to have continued to persist (Since 1950 December has been the best performing month for the S&P 500 with an average return of 1.62%). This poses a contradiction to the Weak form of EMH; this is as any past price information should have been already processed by the market eliminating this trend. Explanations for the ‘December Effect’ vary, in my opinion the most convincing is that of Tax gain selling, concluded by Chen & Singal (2003), investors are reluctant to sell winner stocks in December in order to avoid crystallising their capital gain in the current financial year, they hence hold until January to sell their outperforming shares. Also, what also must be considered is that given there are 12 months in the year, by the Weak EMH the average returns of all these months should also follow a normal distribution, making it inevitable that there would be a month that appears to ‘outperform’ the rest. Similar to ‘December Effect’ is the less prevalent ‘Monday Effect’ which claims that market returns on Mondays are statistically significantly less than that of other days of the week, by the same logic as above this also violates the Weak EMH. Another observation that appears to violate the EMH occurs weekly in Options markets. As option prices decay each day (cetaris paribus), on Fridays the implied volatility (the most often quoted ‘price’ of options) drops significantly in order to cater for the two weekend days that the option cannot be traded on, the implied volatility level will then rebound back on Monday. This creates an easily arbitrageable weekly pattern of a sell-off in option implied volatility on Fridays and the consequent rebound on Monday. Arguably the easiest way to test the validity of the Weak EMH is to check whether asset prices truly follow a random walk. Enninful & Dowling (2013) investigates both large and small capitalisation


European stocks for the period 2000-2012, finding evidence of negative serial correlation for large capitalisation stocks and positive for smaller stocks, this supports the hypothesis of random walk efficiency for large, liquid stocks. Nisar & Hanif (2011) studied stock...
tracking img