1	Stock Volatility: Past, Present & Future G. William Schwert Financial Management Association Keynote Address October 21, 2009 (Data updated through 8/31/2009) http://schwert.ssb.rochester.edu/fma_2009.htm
2	What is market volatility? Standard deviation of rates of return to broad market indexes Following plots show: Changes in Dow Jones Industrial Average from 1893-2009 Affected by growth in the level of the index Percent changes in DJIA (rates of return, ignoring dividends) from 1893-2009 Rolling annualized standard deviations of rates of return to DJIA from 1893-2009
3	Looking at the absolute scale of stock indexes is very misleading . . . The sixty largest changes in the DJIA have been within the last 12 years The only exception among these sixty days is Oct 19, 1987
4
5
6	Looking at the percent change of stock indexes is relevant . . . This measures the rate of return on the investment i.e., how many more dollars you would have at the end of the day if you invested $100 at the beginning of the day The sixty largest percent changes in the DJIA (or the S&P 500) have been before the last 12 years The only exceptions among these sixty days are after 9/11/2001, and nine days in 2008-2009
7
8
9	How to lie with statistics . . . - focus on very recent history Newspapers often focus on the last few years in discussing current conditions On this basis, people would think stock volatility is unbelievably high in the past year or so . . . This is misleading when viewed from the perspective on the longer history we have available to us Compare the plots of rolling standard deviations from 2004-2009 versus the plot from 1893-2009 . . . Good news is that things seem to have settled down a bit now (compared to 12 months ago)
10
11
12	Similar conclusions from recent intra-day data: 15-minute returns to the S&P 500 and S&P futures These estimates of annualized volatility are independent from day-to-day Not overlapping Volatility was very high in late 2008, but now looks fairly “normal” Focusing on the post-2004 period is misleading
13
14
15	Stylized Facts/Questions: A very long-term view! Market-level volatility has been remarkably stable over time Data back to 1802, covers many wars, financial crises, depressions/recessions Also, major changes in the composition of the US economy Mainly banks, insurance companies, canals in early 1800s Railroads started being important after 1834 Great Depression is the most notable period of prolonged high volatility
16
17
18
19	Implied Volatility: S&P vs. Nasdaq Next figure shows the implied volatility series published by CBOE with ticker symbols VIX (S&P) and VXN (Nasdaq) VXN is much higher, especially in 2000-2002; similar since mid-2007 These measures represent forecasts of future volatility (covering the span of the underlying index options, usually about a month)
20
21
22	Term structure of implied volatility: Things will settle down . . . Recently the CBOE has started to report implied volatility for the S&P 500 for horizons longer then 30 days Looking at the term structures from 2000-2009, until very recently they were pretty flat (i.e., similar forecasts for all horizons) The big spikes in volatility starting last Fall have led to a sharply declining forecast of future volatility Things are now back to more normal patterns
23
24	The CBOE now trades a futures contract on VIX Looking at the longest future maturity (typically about 9-10 months) Futures value of VIX peaks at 43% in mid-December 2008, and now has returned to around 30% Even at the worst of the liquidity crisis, traders were not expecting the spike in volatility to last long or be as bad as it was briefly in the Fall of 2008
25
26	Prior Issues: Technology “Bubble”? 2000-2002 was a period of high volatility for Nasdaq/technology stocks This seems to have returned to more normal levels in the last couple of years High volatility was primarily in technology stocks, independent of firm size, exchange listing, or age of the firm Not limited to “DOT.COM” stocks See Schwert (JME 2002) for more details
27	Technology Portfolios Next figure shows historical volatility for: S&P Technology portfolio, Nasdaq Computer, Biotech, and Telecom, and the CBOE Technology portfolios They all move together, increasing substantially since mid-1998 Decreasing in 2003 Not increasing as much lately
28
29	Was the 2008 Credit Crisis a Finance “Bubble”? Next figure shows historical volatility for: Nasdaq Financial portfolio, the S&P 1200 Financial portfolio, and the Datastream US Financial portfolio, since 1973 They all move together, increasing modestly during the Technology “bubble” from 1998-2002 Increased substantially in 2008-2009 during the liquidity crisis
30
31	Foreign Markets – FTSE (UK) Volatility now is similar to late 90’s early 2000’s, and similar to US levels Also similar to 1973-74 (first OPEC crisis) Exchange rate volatility is higher recently, but small compared with stock volatility
32
33	Foreign Markets – Nikkei 225 and Topix (Japan) Volatility now is similar to 1989-2003, and similar to US levels Also similar to 1973-74 (first OPEC crisis) Exchange rate volatility is higher recently, but small compared with stock volatility
34
35	Summary Market-level volatility often rises after prices fall Recent poor performance of the market is consistent with the higher levels of volatility [counter-cyclical] Inflation of index levels exaggerate perceptions of increased volatility
36	Should Someone Try to Lower Volatility? If So, How? Margin requirements? Regulation of trading? Taxes on Trading?