Finance and Economics Discussion Series Divisions of Research & Statistics and Monetary Aairs Federal Reserve Board,

Washington, D.C.

Are Leveraged and Inverse ETFs the New Portfolio Insurers?

Tugkan Tuzun
2013-48

NOTE: Sta working papers in the Finance and Economics Discussion Series (FEDS) are preliminary materials circulated to stimulate discussion and critical comment. The analysis and conclusions set forth are those of the authors and do not indicate concurrence by other members of the research sta or the Board of Governors. References in publications to the Finance and Economics Discussion Series (other than acknowledgement) should be cleared with the author(s) to protect the tentative character of these papers.

Are Leveraged and Inverse ETFs the New Portfolio Insurers?

Tugkan Tuzun Board of Governors of the Federal Reserve System First Draft: December 18, 2012 This Draft: July 16, 2013

ABSTRACT This paper studies Leveraged and Inverse Exchange Traded Funds (LETFs) from a nancial stability perspective. Mechanical positive-feedback rebalancing of LETFs resembles the portfolio insurance strategies, which contributed to the stock market crash of October 19, 1987 (Brady Report, 1988). I show that a 1% increase in broad stock-market indexes induces LETFs to originate rebalancing ows equivalent to $1.04 billion worth of stock. Price-insensitive and concentrated trading of LETFs results in price reaction and extra volatility in underlying stocks. Implied price impact calculations and empirical results suggest that they contributed to the stock market volatility in the 2008-2009 nancial crisis and in the second half of 2011 when the European sovereign debt crisis came to the forefront. Although LETFs are not as large as portfolio insurers of the 1980s and have not been proven to disrupt stock market activity, their large and concentrated trading could be destabilizing during periods of high volatility.

Keywords: ETFs, Price Impact, Financial Stability, Stock Market Crashes

Tugkan Tuzun (tugkan.tuzun@frb.gov) is with the Federal Reserve Board. Address: 20th and C St. NW, MS 114, Washington, DC 20551. This paper beneted from discussions with Pete Kyle and his suggestions. I am grateful to Celso Brunetti, Eric Engstrom, Hayne Leland, Michael Palumbo, Steve Sharpe and Jeremy Stein for their useful comments. I also thank the seminar participants at the CFTC, Federal Reserve Board and the Oce of Financial Research. All errors are my own. Joost Bottenbley and Suzanne Chang provided excellent research assistance. The views expressed in this paper are my own and do not represent the views of the Federal Reserve Board, Federal Reserve System or their sta.

I. Introduction
The complex structure and behavior of Leveraged and Inverse Exchange Traded Funds (LETFs) have raised important questions about their implications for nancial stability. LETFs are exchange-traded products that typically promise multiples of daily index returns. Generating multiples of daily index returns gives rise to two important characteristics of LETFs that are similar to the portfolio insurance strategies that are thought to have contibuted to the stock market crash of October 19, 1987 (Brady Report, 1988). (1) LETFs rebalance their portfolios daily by trading in the same direction as the changes in the underlying index, buying when the index increases and selling when the index decreases. (2) This rebalancing requirement of LETFs is predictable and may attract anticipatory trading. Portfolio insurance strategies were commonly used by asset managers in the 1980s and their use reportedly declined after the stock market crash of 1987. Portfolio insurance is a dynamic trading strategy, which synthetically replicates options to protect investor portfolios. Synthetic replication of options requires buying in a rising market and selling in a declining market. Rather than buying and selling stocks as the market moves, portfolio insurers generally traded index futures. The Brady Report (1988) suggests that portfolio insurance related selling accounted for a signicant fraction of the selling volume on October 19, 1987. The report also notes that aggressiveoriented institutions sold in anticipation of the portfolio insurance trades. This selling, in turn, stimulated further reactive selling by portfolio insurers. Price-insensitive and predictable trading of portfolio insurers contributed to the price decline of 29% in S&P 500 futures through a selling cascade. This paper studies the impact of equity LETFs on various stock categories while emphasizing their implications for nancial stability and similarities with portfolio insurance strategies. Promising a certain multiple of daily index return induces LETFs to

rebalance their portfolios daily to maintain their target stock-to-cash ratios. Rebalancing demand of a LETF can be derived from a simple formula, which dictates the LETF to buy when its target index goes up and sell when its target index goes down. Although their rebalancing formulas are more complex, portfolio insurers also trade in the same direction as the market to maintain their stock-to-cash ratios constant. Anectodal evidence suggests that LETFs commonly use swaps and futures contracts to rebalance their portfolios. Swap counterparties of LETFs are likely to hedge their positions in equity spot or futures markets. If LETFs use index futures, index arbitraguers transfer the price pressure from the futures market to the stock market. If the LETFs account for a signicant fraction of trading, their rebalancing activity should leave its imprint on the stock indexes they follow. The size of LETF rebalancing demand varies with their net assets and the multiples of daily index return they promise. Based on total net asset value of $20.14 billion as of December 15, 2011, when broad stock-market indexes change by 1%, LETFs rebalancing demand totals $1.04 billion worth of stock. This is roughly 0.84% of daily stock-market volume (excluding the volume of the ETFs and the Depository Receipts) in the United States. Kyle and Obizhaeva (2012) calculate that the portfolio insurers in 1987 would sell $4 billion (4% of stock-plus-derivatives volume) in response to a 4% price decline in the Dow Jones Industrial Average. Although LETFs are not as large as the portfolio insurers were in 1987, LETF rebalancing in response to a large market move, especially in periods of high volatility, could still pose risks. For example, the Flash Crash of May 6, 2010 was triggered by a $4.1 billion (75,000 contracts of E-Mini S&P 500 Futures) sell order, which is equivalent to only 3% of the E-Mini S&P 500 Futures daily volume (CFTC-SEC Report, 2010). With a large market move, such as 4%, the total rebalancing ows of LETFs would be equivalent to this Flash Crash order. Rebalancing in the last hour of trading could, in fact, reduce the possibility of a signicant price dislocation 3

since the market close could serve as a prolonged circuit breaker. On the other hand, executing orders within a short period of time, such as the last hour of trading, may cause disproportionate price changes. A signicant price dislocation at the market close may also impair investor condence. If the market closes with depressed prices, the stock market could experience large investor outows overnight. Naturally, LETFs follow dierent stock-market indexes and the size of their rebalancing diers across stock categories. LETF rebalancing is an important fraction of daily volume, especially in nancial and small stocks. For instance, if the Russell 1000 Financial Services Index increases by 1%, the rebalancing demand of LETFs totals roughly 2% of the daily volume for an average nancial stock. Furthermore, academic studies (Cheng and Madhavan, 2009; Bai et al., 2012) and anectodal reports12 suggest that LETFs rebalance their portfolio in the last hour of trading. Therefore, a large market move could make these stocks vulnerable near the market close, or even before to the extent that opportunistic traders react in anticipation of subsequent LETF rebalancing. Although LETF activity is relatively small in some stock categories, LETF rebalancing in the last hour of trading leads to price reaction and extra volatility in all stock categories. For instance, if the S&P 500 index goes up by 1%, LETF rebalancing demand results in a 6.9 basis-point increase in price and a 22.7 basis-point increase in daily volatility in an average large-cap stock. The size of the price reaction of LETF Flows inferred from empirical specications could be an underestimate if prices at 3:00 pm already reect to some extent investors anticipation of subsequent LETF rebalancing. By directly implementing Kyle and Obizhaeva (2011a,b) measure of price impact, I show that the implied price impact of LETF rebalancing on nancial markets was notable especially during the nancial crisis of 2008Jason Zweig, Will Leveraged ETFs Put Cracks in Market Close?, Wall Street Journal, April 18, 2009 2 Andrew Ross Sorkin, Volatility, Thy Name is E.T.F., New York Times, October 10, 2011
1

2009 and at the height of the European sovereign debt crisis. If LETF rebalancing exerts signicant price pressure, this price pressure could amplify the market movements. Although LETFs have not been proven to disrupt stock markets, it is plausible that during periods of high volatility, their impact in response to a large market move could reach a tipping point for acascade reaction.

II. Literature Review

In the United States, Exchange Traded Funds (ETFs3 ) have grown rapidly and hold nearly $1.3 trillion in net assets as of September 2012. The growing size of the ETF industry has prompted researchers to analyze recent trends. Agapova (2012) and Huang and Guedj (2009) suggest that ETFs are close but not perfect substitutes for index mutual funds. Poterba and Shoven (2002) show that the largest ETF (the SPDR trust) oers similar pre-tax and after-tax returns as the largest index fund (the Vanguard Index 500). Buetow and Henderson (2012) argue that, on average, ETFs closely track their target indexes. Commodity ETFs typically roll contacts in the futures markets as they expire. Bessembinder et al. (2012) estimate the transaction costs of these ETF roll trades in the crude oil futures market and cannot nd compelling evidence for predatory trading. Another line of literature studies the arbitrage relationship between ETFs and their underlying stocks. Ben-David et al. (2012) argue that ETFs propogate liquidity shocks to the securities in their baskets. Analyzing the Flash Crash of May 6, 2010, Madhavan (2011) argues that ETFs are vulnerable in market disruptions because pricing of individual component securities becomes more dicult. Several other academics have studied the complex return structure of LETFs. Cheng and Madhavan (2009), Jarrow (2010), Avellanda and Zhang (2009) show that LETF returns could be signicantly dierent
Includes Exhange Traded Notes (ETN). ETNs are Uncollateralized Debt Instruments and investors do not own the investment assets. ETNs are roughly 2% of the ETF industry.
3

than their multiple of target index returns for holding periods longer than one day. The long-horizon return structure of LETFs that rebalance daily is alone not suitable for the investment horizons of many investors4 . The daily rebalancing of LETFs has also stimulated academic research. Trainor (2010) cannot nd evidence that suggests Leveraged ETFs increase volatility. Focusing on the S&P 500 index returns and aggregate LETF rebalancing demands, Cheng and Madhavan (2009) argue that aggregate LETF rebalancing demand has price pressure on the end-ofday S&P 500 index returns. Similarly, Bai et al. (2012) examine the impact of 6 LETFs on 63 real estate sector stocks and nd evidence for both end-of-day LETF price pressure and extra volatility. My study contributes to this literature by quantifying the implied price eects of LETF rebalancing through examining the impact of all US-listed equity LETFs on several stock categories. Several papers studied the role of portfolio insurers in the 1987 stock market crash. Contrary to the Brady Report (1988), Brennan and Schwartz (1989) suggest that the effect of portfolio insurance strategies is too small to explain the 1987 crash. Gennotte and Leland (1990) argue that informational changes, rather than the selling by portfolio insurers, are needed to explain the 1987 crash. They argue that if mistaken by informed trading, portfolio insurance strategies could have a much greater price impact-an impact of magnitute similar to what was observed in 1987. With the Flash Crash of May 6, 2010, the focus on market disruptions and large orders has been renewed. CFTC-SEC Report (2010) concludes that rapid execution of a large sell order triggered the Flash Crash5 . Kyle and Obizhaeva (2012) examine the ve stock market crashes, including the 1987 crash, with documented large sells during those events. They show that price declines in those events are similar in magnitude to price impact of large sales suggested by market
4 5

Appendix explores the drivers of the investor demand for these products. See Kirilenko et al. (2011) for a detailed examination of dierent trader behaviors on May 6, 2010

microstructure invariance (Kyle and Obizhaeva, 2011a,b). My study extends their work by computing the price impact estimates of LETFs implied by market microstructure invariance.

III. LETF Rebalancing

LETFs typically promise a certain multiple of a daily index return. Producing multiples of daily returns forces LETFs to rebalance their portfolios in response to price movements. Daily rebalancing ensures that LETFs maintain their stock-to-cash ratios at market close. The mechanics of LETF rebalancing can be illustrated in a simple asset allocation setting.6 An asset allocation problem can be written in the following way:

Wt = St + Bt Asset managers generally invest a certain fraction, , of their net assets, Wt , into the risky asset (underlying equity index), St and the rest,(1- ) into the bond market.

Wt = Wt + (1 ) Wt
St Bt

LETFs choose consistent with their prospectuses. For example, Leveraged ETFs have =2 or 3 while Inverse ETFs have =-1,-2 or -3. Assuming interest rate is zero, if the underlying index changes by r , then the investment on the index becomes (1 + r )Wt and this change is reected in the size of total portfolio.
6

See Cheng and Madhavan (2009) for their derivation of the same rebalancing formula.

Wt (1 + r ) = (1 + r )Wt +(1 )Wt

Wt+1 St

Since the LETF is promising over the daily return on the index, Wt+1 should be invested on the index to maintain a constant stock-to-cash ratio.

St+1 = Wt+1 = (1 + r ) Wt The rebalancing amount in response to the size of change r in the index can be calculated as

St+1 St = r ( 1) Wt It is important to note that when [0,1], the rebalancing amount has the same sign as r , suggesting that both Inverse and Leveraged ETFs rebalance in the same direction as their target indexes and their rebalancing do not cancel each other out. Furthermore, this formula is a function of only the target index change, not its level, making LETF rebalancing insensitive to the price level. In practice, LETFs do not have to directly trade in the stock market to rebalance their portfolios. The use of derivatives, especially futures and swaps, is believed to be common among LETFs (Cheng and Madhavan, 2009). If they trade futures contracts, index arbitraguers will transfer this eect from the futures market to the stock market. If they enter into a swap aggrement, their counterparty is likely to hedge its exposure and trade in either the futures or the spot market. As a result, regardless of the contracts LETFs trade, their portfolio rebalancing should leave an imprint on the stock indexes they target. Brady Report (1988) notes that portfolio insurers commonly used futures contracts and index arbitraguers propagated these shocks to the stock market, suggesting 8

that the markets for stocks and stock index futures behave as a single integrated market. More recently, Kirilenko et al. (2011) explain that although the Flash Crash of May 6, 2010 was triggered in the futures market, index arbitraguers quickly transmitted this price shock to the stock market.

IV. Data
Data for this study is collected from multiple sources. LETF information is obtained from Morningstar Direct, which provides total net assets, net asset value, shares outstanding and category type for ETFs. After I identify an ETF as a LETF, I check its prospectus to identify both its target index and the multiple it promises. Only US-listed equity LETFs promising multiples of daily index returns are included in the sample. I also use the daily index return series (S&P 500 Index, Russell 1000 Financial Services, Russell 2000, Russell Mid Cap, Nasdaq 100) from Morningstar Direct. Volume and return variables for the stocks are calculated from the NYSE Trades and Quotes (TAQ) dataset. Membership history, monthly index weights and oat factors are courteously provided by Russell Indexes for Russell 1000 Financial Services, Russell 2000 and Russell MidCap Indexes7 . Membership history of S&P500 is obtained from the Center for Research in Security Prices (CRSP). Membership history of NASDAQ 100 is obtained from Bloomberg and index weights are calculated with CRSP market capitalization information. The sample goes from June 19, 2006, when the rst equity LETF was oered in the US, to December 31, 2011.
7 Float is the number of shares available for trading and many stock indexes calculate stock weights from oat-adjusted market capitalizations.

V. LETFs and Target Stock-Market Indexes

Table I summarizes the distribution of LETFs over stock-market index categories as of December 15, 2011. Total net assets, number of LETFs, target stock index and LETF rebalancing ows in response to a 1% change in the target index are reported. The biggest LETF category is the large stock LETFs and the majority of them follow the S&P 500. In this category, there are 19 LETFs with $8.7 billion in net assets. A 1% increase in large stock indexes generates roughly $400 million of rebalancing ow from these LETFs. The net assets of 5 LETFs, which follow nancial stock indexes, total $3.5 billion and their rebalancing ow is $ 226.8 million in response to a 1% increase in nancial stock indexes. 14 LETFs exist in the small stock category with $3 billion in net assets. A 1% increase in small stock indexes induces these LETFs to demand $193 million worth of stocks. The mid-cap stock category is the smallest in size with roughly $ 299 million net assets in total managed by 11 LETFs. Their rebalancing ow is only $ 13 million when there is a 1% increase in mid cap stock indexes. There are 9 LETFs with $2.2 billion in net assets following technology stock indexes and their rebalancing ow totals $ 100 million if technology stock indexes go up by 1%. In total, there are 108 equity LETFs in the US with roughly $ 20 billion in net assets. These LETFs originate $ 1.04 billion in rebalancing ows if the broad stock-market indexes go up by 1%. Daily volume of the stock market (excluding ETFs and Depository Receipts) averaged $131 billion in December 2011. Hence, LETF rebalancing ows in response to a 1% increase in stock prices are equivalent to 0.79% of stock-market volume. Because LETFs also use swap and futures contracts, LETF rebalancing is less than 0.79% of the total volume of the stock, futures and swaps markets combined. Although their size appears to be smaller than portfolio insurers of 1987, it is essential to explore the relative size and eect of LETF rebalacing in various stock categories.

10

A. LETF Performance and Rebalancing

LETFs are forced to rebalance daily to avoid tracking errors by maintaining constant stock-to-cash ratios. Low tracking errors can be interpreted as successful portfolio rebalancing. Because LETFs promise certain multiples of daily target index returns, I assess their performance at daily frequency by using the following regression.

Index Rj,t = + (j Rj,t ) + j,t

(1)

The S&P 500, Russell 1000 Financial Services, Russell 2000, Russell MidCap and NASDAQ 100 are used as target indexes for the respective LETF categories dened in Table I. j is the multiple LETF j promises. The variables are winsorized at the 1% and 99% levels to remove the eect of outliers. The regression is run for each LETF individually and Table II reports the summary statistics of the regression coecients and Adj-R2 s computed within each category. The asset-weighted and equally-weighted means and medians of are all close to zero. In absolute value, equally-weighted mean of range from 0.01 basis points for the large category to 3.2 basis points for the nancial category. Asset-weighted means of are lower in absolute value and range from 0.01 for the large category to 1.6 basis points for the small category. Asset-weighted means of are close to 1 and range from 0.9 for the technology category to 1.03 for the mid-cap category. Similarly, asset-weighted means of R2 are quite high. These results suggest that LETFs, especially the ones with large net assets, rebalance regularly since they are, on average, successful in delivering multiples of their target indexes at daily frequency.

B. LETF Rebalancing Flows and Underlying Stocks

LETF categories dened in Table I are collapsed into one target stock index in each category. For simplicity, I assume that the LETFs following large stock indexes aim to 11

target the S&P 500. The Russell 1000 Financial Services and NASDAQ 100 indexes are chosen for nancial and technology categories. Small and mid-cap LETFs are assumed to follow the Russell 2000 and Russell Mid-Cap Stock Indexes, respectively. Panel A of Table III reports the summary statistics for various stock categories from June 2006 to December 2011. The nancial category includes the members of the Russell 1000 Financial Services Index. Float-adjusted and unadjusted market capitalizations of an average nancial stock are $10.1 and $ 10.9 billion, respectively. Daily volatility- the standard deviation of the previous 20 days returns- of an average nancial stock is 2.9%. There are 279 dierent stocks in this category with $125 million in average daily volume. The large stock category, which consists of the members of the S&P 500 index, has $21.3 and $22.3 billion in oat-unadjusted and -adjusted market capitalizations, respectively. There are 647 dierent stocks in the large stock category with an average daily volatility of 2.3% and volume of $203 million. Going from the mid-cap stock category (Russell Mid-Cap) to small stock category (Russell 2000), market capitalizations and daily volume decrease. The daily volume of an average mid-cap stock and small stock are $ 60 million and $7 million, respectively. For an average small stock, oat-adjusted market capitalization, $ 550.3 milion, is considerably dierent then unadjusted market capitalization, $ 652.4 million. Technology stocks, which are the members of the NASDAQ 100 index, have $ 22 billion in average market capitalization, roughly equal to that of large stocks8 . Yet, the daily volume of an average technology stock ($303 million) is higher, suggesting that they are traded more actively. Panel B and Panel C of Table III report the same variables for 2006-2009 and 2010-2011 periods. Higher volatility in all stock categories, especially among nancials, is notable in the earlier sample period due to the 2008-2009 nancial crisis. Daily volatility of an average nancial stock is 3.5% during
Stock-weights for NASDAQ 100 are computed from oat-unadjusted market capitalizations since oat variable could not be obtained.
8

12

2006-2009 period and 2.0% during 2010-2011 period. Other stock categories have 70 to 80 basis points higher daily volatility in the earlier sample period. B.1. Size of Hypothetical LETF Rebalancing Flows Figure 1 presents the share of total hypothetical rebalancing ows of LETFs for an average stock as a fraction of its daily volume for each category. Aggregate hypothetical LETF ows are calculated as the sum of all LETF ows in that category in response to a 1% increase in the target index and allocated to stocks based on their weights in the target index. For each stock, its share of hypothetical LETF rebalancing ows is scaled by its average daily volume. Hypothetical LETF rebalancing ows start in June 2009, when the rst equity LETF was introduced in the US. These ows are less than 1% of daily volume for an average stock in large, mid-cap and technology categories. Hypothetical LETF rebalacing ows are considerable for an average stock in nancial and small stock categories. In the beginning of 2009, hypothetical LETF rebalancing ows become larger than 1% of daily volume in an average nancial stock and uctuate around 2% of daily volume starting in mid 2009. Hypothetical LETF rebalancing ows become larger than 2% of daily volume in an average small stock in 2009 and reaches levels as high as 6% of daily volume in 2011. Many studies, such as Cheng and Madhavan (2009); Bai et al. (2012), and news articles mention that LETFs carry out their rebalancing in the last hour of trading. Figure 2 shows the aggregate hypothetical rebalancing ows of LETFs as a fraction of volume in the last hour of trading. These ows are 2-3% of volume in the last hour for average stocks in large, mid-cap and technology categories. For an average small stock, these ows can be as large as 18% of volume in the closing hour. Similarly, the ratio of hypothetical LETF ows to the last hours volume is signicant for an average nancial stock. Hypothetical LETF rebalancing ows for an average nancial stock uctuate between 4 and 8% after 13

2008 and are equivalent to roughly 6% of volume in the last hour in December 2011.

C. End-of-Day Price Eects of LETF Flows

C.1. Price Reaction of Underlying Stocks To identify pressure points across stock categories, it is crucial to understand the impact of LETF rebalancing ows on underlying stocks. LETF rebalancing could result in endof-day price reaction and extra volatility in those stocks. Other market participants could trade in anticipation of LETF ows, making it impossible to estimate the isolated price impact of LETF ows. However, the end-of-day net price reaction to LETF rebalancing and anticipatory trades can be estimated. To estimate the net price reaction to LETF ows, I implement the regression specication of Obizhaeva (2009) and allow for control variables. log (P(i,15:00) ) log (Pi,c) LET F F lowi = 0 + 1 + 2 + i i ADVi i

(2)

The left hand side of the regression is constructed as a ratio of two variables. log (Pi , c) is the return on a stock i between 15:00pm(ET) and 16:00pm (ET) and i is the daily volatility. Explanatory variables are also contructed as ratios of two variables. LET F F lowi , is the share of stock i from the estimated total LETF rebalancing ow-calculated as a function of the target index return between the previous days close and 15:00pm-implied by its index weight9 . LET F F lowi is scaled by ADVi , which is the past 20-day average dollar volume of stock i. The variables used in this regression are winsorized at the 1% and 99% levels to remove the eect of outliers. Panel A of Table IV reports the results of the regression for dierent categories from 2006 to 2011. The
Intraday target index returns are calculated from the intraday returns of their constituents. Unreported results verify that close-to-close return of target indexes are statistically equal to the daily returns of those indexes obtained from the Morningstar Direct.
9

14

standard errors are clustered daily. For all categories, the coecients on LETF ows are positive and statistically signicant, ranging from 0.83 for small stocks to 4.16 for mid cap stocks. The coecient estimates and regression R2 s are little changed when the stock return between the previous days close and 15:00pm,
log (P(i,15:00) ) ,

is controlled.

Regression R2 s in this specication range from 0.76% for the technology category to 6.33% for the nancial category. Panel B of Table IV reports the results of the regression for two subperiods, 2006-2009 and 2010-2011. The coecients on LETF Flows and R2 s are higher for most of the categories in the earlier sample period. The coecients on LETF Flows go from 0.78 for small stocks to 4.85 for mid-cap stocks for 2006-2009 and they go from 0.88 for small stocks to 2.74 for nancial stocks in the post-nancial crisis period. The coecients on LETF Flows can be interpreted as a change in price as a percent of daily volatility in response to LETF ows equivalent to 1% of the volume. For example, LETF ows equivalent to 1% of stock volume, increases the price of a nancial stock by 3.09% of its daily volatility. As of December 2011, when the nancial indexes increase by 1%, the share of LETF rebalancing ows is equivalent to 2.1% of volume in an average nancial stock with 2% daily volatility. If the nancial stock indexes increase by 1%, price reaction of an average nancial stock in response to LETF rebalancing is 13 basis points (3.09 2.1% 2%). With the LETF ows in response to a 1% change in the target indexes and the volatility of average stocks in December 2011, the same calculation can be done for other categories. The end-of-day price reaction is 6.9 basis points (4.32 0.8% 2%) in an average large stock, 5.5 basis points (4.28 0.6% 2%) in an average mid cap stock, 12.7 basis points (0.86 5% 2.9%) in an average small stock and 6.3 basis points (3.83 0.75% 2.2%) in an average technology stock. Although the largest price reactions to LETF ows occur in nancial and small stock categories, all other categories show price reactions to LETF portfolio rebalancing, suggesting that 15

LETFs and anticipatory traders in the same direction are stronger than the traders on the opposite side. Without the traders taking the other side of the LETF rebalancing activity, the end-of-day eect of the LETF rebalancing could be destabilizing. Rebalancing in the last hour could also amplify the eect of LETFs since executing orders within a short period of time may move the prices disproportionately. In contrast, rebalancing in the last hour has the advantage of limiting the possibility of a price dislocation because the market close may serve as a prolonged circuit breaker. Yet, if there are enough traders following the LETF rebalancing, they may start trading well before the market close and produce similar consequences. Table V reports the results of the same regression for Market Ups (Positive LETF Flows) and Market Downs (Negative LETF Flows) seperately. The standard errors are clustered daily. Coecients of positive LETF ows range from 0.63 for small stocks to 4.66 for technology stocks and coecients of negative LETF ows range from 0.80 for small stocks to 3.60 for large stocks. The coecient on the LETF Flows in Market Ups are higher than Market Downs for all but small stocks, indicating that market response to positive LETF ows is slightly stronger. One explanation for this could be the shortsale constraints. Market participants who trade in anticipation of LETF ows could be constrained by short-selling and cannot implement their strategy in Market Downs as well as in Market Ups. C.2. Price Reversals In resilient markets, prices revert back after an order is executed especially if the order does not carry information about the fundamental value. The resilence of the market could counteract the late-day price reaction of the LETF rebalancing ows and other anticipatory trades. I implement the following regression to test if prices revert back the next day after the portfolio rebalancing of LETFs. 16

log (Pi,15:00 ) = 0 + 1 i

LET F F lowi ADVi

+ 2
t1

log (Pi ) i

+ i
t1

(3)

The next days return of stock is dened as the return from todays market close to 15:00 next day, scaled by its daily volaility. Explanatory variables are LETF rebalancing ows and the previous days returns. Panel A of Table VI reports the results of this regression for the full sample period. The standard errors are clustered daily. The coecients on LETF Flow are negative and signicant for all stock categories, ranging from -0.35 for small stocks to -4.24 for technology stocks. Compared with the results from Table IV, these coecients are similar in magnitude, suggesting that prices revert back the next day after LETF rebalancing. Panel B and Panel C of Table VI present the results of the regression for 2006-2009 and 2010-2011 seperately. The coecients for LETF rebalancing ows are negative for all categories in both sample periods. They range from -0.49 for small stocks to -4.77 for large stocks in the earlier period and go from -0.31 for small stocks to -5.23 for large stocks in the later period. C.3. End-of-Day Volatility and LETF Rebalancing LETF rebalancing and trades in anticipation of LETF rebalancing may aect the stock volatility in the last hour of trading. To estimate the volatility eects of LETF rebalancing, I use the following regression specication. log (Pi,c ) i
2

= 0 + 1
2

log (Pi,(15:00) ) LET F F lowi + 2 ADVi i

+ i

(4)

The left hand side, return variance.

log (Pi,c ) i

, is the square of the return at the close scaled by daily

LET F F lowi ADVi

is the absolute value of LETF Flows scaled by the average

17

daily volume and

log (Pi,(16:0015:00) ) 2

is the square of daily return until 15:00pm scaled by

daily return variance. The variables used in this regression are winsorized at the 1% and 99% levels to remove the eect of outliers. Panel A of Table VII reports the results of this regression for the entire sample period. The standard errors are clustered daily. When introduced into the regression,
log (Pi,(16:0015:00) ) 2

has a positive coecient and increases

the adjusted R2 for all categories, indicating that high intraday volatility persists through end-of-day. The LETF Flow variable is positive and signicant in all stock categories and ranges from 0.59 for small stocks to 2.57 for technology stocks. The coecient on the LETF Flow variable can be interpreted as the change in return variance as a percent of daily variance in response to LETF Flows equivalent to 1% of volume. For example, LETF Flows of 1% of the daily volume increases the return variance, (log (Pi,c ))2 , by 1.30 % of the return variance, 2 , in a nancial stock. Point estimates can be computed for average stocks in each category with December 2011 volume and volatility averages. The end-of-day extra return volatility in response to 1% change in stock indexes is 31.1 basis points ( ( (1.3 2.1% 2%) in an average nancial stock, 22.7 basis points

(1.72 0.8% 2%)in an average large stock, 23 basis points ( (1.96 0.8% 2%) (0.59 5% 2.9%) in an average

in an average mid-cap stock, 50.5 basis points ( small stock and 30.73 basis points (

(0.6 0.75% 2.2%)) in an average technology

stock. These results suggest that LETF rebalancing and possible anticipatory trades of other market participants account for extra end-of-day volatility for all stock categories. Panel B of VII reports the results for two sample periods, 2006-2009 and 2010-2011. The coecient on the LETF Flow is positive and signicant for all categories in both sample periods. It ranges from 0.64 for small stocks to 5.35 for technology stocks for 2006-2009. In the post-crisis period, the coecient is smaller and goes from 0.71 for small stocks to 1.34 for mid-cap stocks.

18

C.4. Intraday Lead-Lag Relations: A Robustness Check Lo and Mackinlay (1990) document lead-lag patterns in stocks returns. More specically, they show that large stock returns lead small stock returns. To control for possible intraday lead-lag patterns accross and within stock categories, the S&P 500 index and underlying index returns are included in the end-of-day price reaction and volatility regressions. As mentioned previously, the underlying index is the Russell 1000 Financial Services Index for nancial stocks, the S&P 500 index for large stocks, the Russell MidCap Index for midcap stocks, the Russell 2000 for small stocks and the Nasdaq 100 for technology stocks.10 Panel A of Table VIII reports the results for end-of-day price reaction. The coecient for the LETF ow is only slighly lower than the previous regressions in Table IV and it continues to be positive and statistically signicant for all stock categories. The coecient ranges from 0.545 for the small stock category to 3.9 for the midcap stock category. Panel B of Table VIII reports the results of the regression of end-of-day volatility. As in Table VII, the coecient for the absolute value of LETF Flow continues to be positive and signicant after controlling for the intraday volatility of the underlying index and S&P 500 index returns. These results suggest that the eect of LETF rebalancing on the end-of-day price and volatility is not driven by the intra-day lead-lag relationship between small and large stocks.

VI. LETFs and Integrated Markets

The size of the price reaction of LETF Flows inferred from empirical specications could be an underestimate if prices at 3:00 pm already reect to some extent investors anticipation of subsequent LETF rebalancing. A more direct way to quantify the price impact
Because the S&P 500 index is the underlying index returns for large stock, the S&P 500 index returns are included in the regressions for large stocks only once.
10

19

of LETF rebalancing ows is to implement the implied price impact measure developed and calibrated by Kyle and Obizhaeva (2011b,a). Kyle and Obizhaeva (2012) use this measure to estimate the size of ve crash events implied by market microstructure invariance and conclude that estimates are close to the observed price declines. Aggregate stock-market segments rather than individual stocks may provide more accurate price impact estimates if the markets for individual stocks are integrated. Integration of markets may result from various factors. For instance, arbitraguers operate in multiple markets and exploit arbitrage opportunities by taking opposite positions in these markets. Hence, they transmit shocks from one market to another. As a result, market integration leads to faster and more eective transmission of shocks. Moreover, price shocks to individual stocks can be transmitted to other stocks through leveraged investor portfolios. Initial losses could lead to margin calls whereby speculators are forced to deleverage by selling assets in their portfolios, hence leading broader asset price declines (Brunnermeier and Pedersen, 2009). The expected price impact of LETF rebalancing ows in a stock-market index with daily dollar trading volume, ADV, and daily volatility, , is given by ADV 40 106
1/3

log (P ) = /10 .

0.02

4/3

LET F F low 0.01 ADV

(5)

Kyle and Obizhaeva (2011a) estimate their price impact formula in portfolio transitions data and nd equal to 5.78. This formula is implemented for 5 dierent stockmarket indexes; Large, Mid-Cap, Small, Financial and Technology Stock indexes. As in the previous sections, I use S&P 500 rms for the large category, Russell 2000 rms for the small category, Russell Mid Cap rms for the mid-cap category, Russell 1000 Financial Services Index rms for the nancial category and NASDAQ 100 index stocks for the technology category. ADV is calculated as the total daily volume of member

20

stocks averaged over the previous 20 days. Volatility is estimated as the standard deviation of the daily index returns in the past 20 days. Finally, LETF Flow is the dollar amount of LETF rebalancing in response to a 1% change in the target stock-market index. Measuring the LETF Flows in response to a constant change in the target index (such as 1%) allows for historical comparison of the impact of LETF rebalancing and critical assessment for their current size. Table IX reports the averages of the variables used, and the implied price impact estimates for 5 stock-market indexes from 2007 to 2011. In 2008 and 2009, volatility is higher for each category due to the nancial crisis. Not suprisingly, the nancial stock index experiences the highest daily volatility, 3.7% in 2008 and 3.4% in 2009. The growth in net assets of nancial LETFs increases the scaled LETF rebalancing ow from 0.4% in 2008 to 1.02% in 2009. Higher volatility and the growth in nancial LETFs leads to an average implied price impact of 0.97 % in 2009. If total LETF rebalancing leads to a price impact equal to or greater than the change in the target index level, it could signicantly amplify the target index moves and force LETFs to carry out further rebalancing. As a result, the implied price impact of 1% can be considered a critical level for this analysis. Figure 3 plots the implied price impact of LETF rebalancing in response to a 1% change in the target stock index for ve categories from June 19, 2006 to December 31, 2011. The implied price impact for the nancial category goes above the 1% level in the summer of 2008 and reaches a level of 1.5% due to over 5% daily volatility after the collapse of Lehman Brothers. The implied price impacts for large, small and technology stock categories also rise and approach to the 1% level in late 2008. Markets reacted negatively to events following the Financial Stability Plan announcement on Feb 10, 2009 and the S&P 500 reached a record low level on March 9, 2009. The implied price impacts for the nancial category skyrockets and almost reaches to the 2.5% level in the rst half of 2009 because of over 21

5% daily volatility and the growth in nancial LETFs. After the rst half of 2009, all implied price impacts of LETFs remain below the 1% level until August 2011 when the daily volatility for the nancial category increases to 4% due to the concerns about European soveriegn debt11 . The implied price impact estimates for the small and the nancial categories stay above the 1% level through late August and early September 2011. Kyle and Obizhaeva (2012) compute the implied price impact of portfolio insurers in October 1987 and nd that it ranges from 11.13% to 15.75% over the four days surrounding the stock-market crash of October 19. Although the implied price impact estimates of LETFs are not as high as that of portfolio insurers of the 1980s, the implied price impact of LETFs becomes substantial when daily volatility surges. Implied price impact results indicate that the imprint of LETF rebalancing left on the nancial stock index should be visible during the 2008-2009 nancial crisis and again in 2011 when the European sovereign debt crisis came to the forefront. Figure 4 plots the frequency of a 1% price move of the nancial stock index in the last hour of trading given that the index has already moved 1% in the same direction by 3:00pm. This frequency is calculated from a months trading days in which the nancial stock index moved by 1% by 3:00pm
12

. The frequency of a large price move in the last hour of

trading is zero until 2007 when the rst nancial LETF is launched. Consistent with the implied price impact results, the frequency of a large price move is elevated when the price volatility is high, reaching 0.8 in the 2008-2009 nancial crisis and 0.6 in the second half of 2011. These results, combined with the implied price impact estimates, suggest that LETF rebalancing contributed to the stock market volatility in the 20082009 nancial crisis and in the second half of 2011. Although LETF rebalancing has not
To calm the markets, G7 and ECB held an emergengy meeting on August 8, 2011 after S&P downgraded the long-term credit rating of US. 12 Although Russell 1000 Financial Services Index was launched in 1996, my sample starts in 2003.
11

22

been proven to disrupt stock market activity, their large and concentrated trading could pose risks. The rebalancing of LETFs in the last hour could amplify their impact since executing orders within a short period of time may move prices disproportionately and trigger a cascade reaction. In contrast, rebalancing in the last hour has the advantage of limiting the possibility of a price dislocation because the market close may serve as a prolonged circuit breaker. However, if there are enough traders following the LETF rebalancing, they may start trading well before the market close and produce similar consequences.

VII. Conclusion
Contrary to plain vanilla ETFs, LETFs typically rebalance their portfolios daily to maintain their stock-to-cash ratios. Maintaining constant stock-to-cash ratios forces LETFs to rebalance in the same direction as target index moves, selling in a declining market and buying in a rising market. Similar to portfolio insurance strategies, mechanical rebalancing of LETFs is predictable and could attract opportunistic traders, who may originate orders in anticipation of LETF ows. Although the LETFs are not as large as portfolio insurance strategies of the 1980s in terms of size and impact, daily LETF rebalancing leaves its imprint on all stock categories. The implied price impact estimates of LETFs on broad stock-market indexes become signicant during periods of high volatility, especially for the stocks of nancial rms. LETF rebalancing in response to a large market move could amplify the move and force them to further rebalance which may trigger a cascade reaction. Rebalancing in the last hour of trading could, in fact, reduce the possibility of a price dislocation since the market close could serve as a prolonged circuit breaker. On the other hand, executing orders within a short period of time, such as the last hour of trading, may 23

cause disproportionate price changes. A signicant price reduction at market close may also impair investor condence. If the market closes with depressed prices, the stock market could experience large investor outows overnight. Although the long-horizon return structure of LETFs that rebalance daily is alone not suitable for the investment horizons of many investors, LETFs delivering multiples of daily index returns are greater in size and number than the ones delivering multiples of monthly and quarterly index returns13 . Further research is needed to better understand the drivers of the demand for LETFs delivering multiples of daily index returns.

There are 7 LETFs which target to deliver multiples of monthly index returns and their total size is $ 184.1 million.

13

24

References
Agapova, A. (2012). Conventional mutual index funds versus exchange traded funds. Journal of Financial Markets, 14(2):323343. Avellanda, M. and Zhang, S. J. (2009). Path-dependence of leveraged etf returns. Working Paper. Bai, Q., Bond, S. A., and Hatch, B. (2012). The impact of leveraged and inverse etfs on underlying stock returns. University of Cincinnati Working Paper. Ben-David, I., Franzoni, F. A., and Moussawi, R. (2012). Etfs, arbitrage, and shock propagation. Working Paper. Bessembinder, H., Carrion, A., Venkataraman, K., and Tuttle, L. A. (2012). Predatory or sunshine trading? evidence from crude oil etf rolls. Working Paper. Brady Report (1988). Report of the presidential task force on market mechanisms. Brennan, M. J. and Schwartz, E. S. (1989). Portfolio insurance and nancial market equilibrium. The Journal of Business, 62(4):455472. Brunnermeier, M. K. and Pedersen, L. H. (2009). Market liquidity and funding liquidity. The Review of Financial Studies, 22(6):22012238. Buetow, G. and Henderson, B. (2012). An empirical analysis of exchange traded funds. Journal of Portfolio Management. CFTC-SEC Report (2010). Findings regarding the market events of may 6, 2010. Cheng, M. and Madhavan, A. (2009). Dynamics of leveraged and inverse exchangetraded funds. Journal of Investment Management. Gennotte, G. and Leland, H. (1990). Market liquidity, hedging, and crashes. The American Economic Review, 80(5):9991021. Hill, J. and Teller, S. (2010). Hedging with inverse etfs. Journal of Indexes. Huang, J. C. and Guedj, I. (2009). Are etfs replacing index mutual funds? Working Paper. Jarrow, R. A. (2010). Understanding the risk of leveraged etfs. Finance Research Letters, 7:135139. Kirilenko, A., Kyle, A. S., Samadi, M., and Tuzun, T. (2011). The ash crash: The impact of high frequency trading on an electronic market. Working Paper.

25

Kyle, A. S. and Obizhaeva, A. (2011a). Market microstructure invariants: Empirical evidence from portfolio transitions. University of Maryland Working Paper. Kyle, A. S. and Obizhaeva, A. (2011b). Market microstructure invariants: Theory and implications of calibration. University of Maryland Working Paper. Kyle, A. S. and Obizhaeva, A. (2012). Large bets and stock market crashes. University of Maryland Working Paper. Lo, A. W. and Mackinlay, A. C. (1990). When are contrarian prots due to stock market underreaction? The Review of Financial Studies, 3(2):175205. Madhavan, A. (2011). Exchange-traded funds, market structure and the ash crash. Working Paper. Obizhaeva, A. (2009). Portfolio transitions and stock price dynamics. University of Maryland Working Paper. Poterba, J. M. and Shoven, J. B. (2002). Exchange traded funds: A new investment option for taxable investors. MIT Department of Economics Working Paper. Trainor, W. J. (2010). Do leveraged etfs increase volatility? Technology and Investment.

26

Appendix: Investor Flows into LETFs

Since the inception of the rst LETF on June 19, 2006, equity LETFs have gained popularity among investors and currently have $20B in net assets in the US. Because the size and growth of LETFs could be important from a nancial stability point of view, examining the drivers of the LETF inows could explain the intent of their use. There are several hypotheses which explain the purpose of investing in LETFs. One explanation is that LETFs provide an easy access to leverage and short-selling. LETFs can be used by retail investors to easily short stocks or build leverage. Yet, this hypothesis cannot explain why LETFs are promising a certain multiple of daily stock index returns. Generating a multiple of daily index return requires LETFs to sell when the market is down and buy when the market is up, resulting in transaction costs. Several studies (Jarrow (2010); Cheng and Madhavan (2009); Huang and Guedj (2009)) show that returns of LETFs could be signicantly dierent that their multiple of target index returns for holding periods longer than one day. The investment horizon of a retail trader is typically longer than a day, suggesting that LETFs may not be suitable for many retail investors. Although there are 7 LETFs which target to deliver multiples of monthly index returns, their total size is very small ($ 184.1 million). Another explanation is that LETFs are used for hedging or portfolio insurance purposes. By adding a LETF into his portfolio, an investor may aim to reduce his overall risk exposure. Hill and Teller (2010) explain that Inverse ETFs could be used by investors for hedging purposes. Hedging a position involves rebalancing the hedge amount to maintain the desired risk exposure. To explore the hedging hypothesis for LETF usage among investors, I examine the drivers of investor inows into the LETFs. The dollar amount of investor ow is calculated as the daily change in LETF shares outstanding multiplied by its net asset value. The variable of interest, Investor Flow, is dened as the dollar amount of investor ow into an LETF scaled by lagged total net assets. Explanatory variables considered for this variable are daily lagged Investor Flows, daily lagged LETF returns and daily changes in VIX. Lagged LETF returns are included to explore the dynamic hedging hypothesis. If an investor is using a LETF for hedging purposes, he dynamically adjusts his hedge position, by selling his shares when his LETF position grows and by buying new shares when his LETF position shrinks. As a result, investors should redeem their shares when LETFs are generating positive returns and buy new shares when LETFs are giving negative returns. Yet, there are other explanations such as the disposition eect for negative feedback investments. The disposition eect is the tendency of investors to sell assets whose prices have appreciated while keeping assets whose prices have declined. Table X reports the results of the regression for Investor Flows.14 The standard errors are adjusted for heteroskedasticity. When only the lagged Investor Flows are used as explanatory variables, Adj R2 is 3.23%. The coecients on lagged Investor Flows
14

The results are robust to calculating the investor ows as [Net Assets - (1+r) Lagged Net Assets]

27

range from 0.10 to 0.004 and all but one are statistically signicant. When the lagged LETF returns are added, Adj R2 increases to 4.66%. The coecients on the lagged LETF returns are negative, statistically and economically signicant. For instance, if a LETF delivers a 1% return, it experiences outows equivalent to 0.07% (0.07 1%) of its net assets the next day and 0.04% of its net assets on the following day. When the lagged changes of VIX are introduced in the regression, Adj R2 increases slightly and becomes 4.72%. The disposition eect implies that LETFs should have outows when they deliver positive returns. Hence, outows should be more sensitive to recent returns than inows. Table X includes the regression results for Inows and Outows. Inows and Outows respond dierently to lagged Investor Flows and lagged changes in VIX. The coecients on Investor Flows are positive for Inows whereas they are generally negative for Outows, suggesting that Inows are persistent. More interestingly, changes in VIX have positive coecients for Inows and negative coecients for Outows. Compared to Inows, Outows respond strongly to past LETF returns as the disposition eect suggests. Although these results are consistent with the hedging hypothesis, disposition eect cannot be ruled out.

28

Table I: Leveraged and Inverse ETFs

# of LETFs Large Cap Stock Indexes S&P 500 TR DJ Industrial Average TR USD Russell 1000 TR USD Total Financial Stock Indexes DJ US Financial TR USD Russell 1000/Financial Services TR USD Total Small Stock Indexes Russell 2000 TR USD Russell 2000 Growth TR USD Russell 2000 Value TR USD S&P SmallCap 600 TR Total Mid Cap Stock Indexes Russell Mid Cap TR USD Russell Mid Cap Growth TR USD Russell Mid Cap Value TR USD S&P MidCap 400 TR Total Technology Stock Indexes NASDAQ 100 TR USD DJ US Technology TR USD Russell 1000/Technology TR USD Total Other Stock Indexes Alerian MLP Infrastructure TR USD DJ US Basic Materials TR USD DJ US Consumer Goods TR USD DJ US Consumer Svcs TR USD DJ US Health Care TR USD DJ US Industrials TR USD DJ US Oil&Gas TR USD DJ US Real Estate TR USD DJ US Select Telecom TR USD DJ US Semiconductors TR USD DJ US Telecom TR USD DJ US Utilities TR USD ISE Revere Natural Gas PR USD KBW Regional Banking TR USD MSCI US REIT GR USD NASDAQ Biotechnology TR USD NYSE Arca Gold Miners Index PHLX Semiconductor Russell 1000 Growth TR USD Russell 1000 Value TR USD Russell 1000/Energy TR USD Russell 1000/Retail TR USD Russell 3000 TR USD S&P Healthcare Select Sector S&P Materials Select Sector Total Aggregate Total 1 3 2 2 2 2 3 3 1 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 50 108 10,835,796 287,603,995 16,075,166 20,273,116 37,383,013 30,662,049 352,738,685 534,686,005 2,777,478 56,446,452 3,070,287 22,803,601 12,099,179 14,480,240 141,508,860 14,441,624 89,170,194 116,726,382 20,007,416 7,927,918 487,500,000 7,702,681 8,917,945 6,608,294 7,878,608 2,310,324,984 20,125,854,794 216,716 8,557,523 521,042 917,055 881,799 919,148 11,362,774 16,733,720 166,649 1,790,918 61,406 592,072 931,071 289,605 10,357,063 361,157 5,744,884 8,189,627 637,600 250,571 36,072,000 585,902 263,524 603,729 705,510 107,713,064 1,040,629,262 5 2 2 9 1,939,900,000 99,360,379 203,460,169 2,242,720,548 83,582,000 2,545,379 14,257,220 100,384,599 2 2 2 5 11 69,633,676 14,624,142 8,669,870 206,730,479 299,658,167 5,296,984 405,294 252,112 7,204,583 13,158,973 7 2 2 3 14 2,875,507,378 35,733,226 18,493,210 84,805,694 3,014,539,508 188,800,885 1,471,907 619,596 2,483,508 193,375,896 3 2 5 1,223,900,000 2,287,500,000 3,511,400,000 39,438,000 187,380,000 226,818,000 12 5 2 19 7,251,240,249 1,042,571,338 453,400,000 8,747,211,587 316,399,014 45,075,716 37,704,000 399,178,730 Total Net Assets ($) Rebalancing Flow (r=1%)

This table presents the descriptive variables of equity LETFs across dierent stock index categories. Total net assets are in dollars. Rebalancing ow is the rebalancing amount in dollars if the target index changes by 1%.

29

Table II: Daily Performance of LETFs

LETF Category # of LETFs Statistics Adj R2 Adj R2 Adj R2 Adj R2 Adj R2 Mean -0.014% 1.001 0.986 -0.002% 0.993 0.967 -0.004% 1.052 0.981 -0.006% 0.985 0.989 -0.004% 0.951 0.891 Weighted Mean 0.028% 1.005 0.995 0.0004% 0.985 0.991 0.013% 1.055 0.978 -0.008% 0.993 0.996 0.002% 0.950 0.900 Median -0.051% 1.001 0.996 0.002% 0.993 0.993 -0.012% 0.999 0.975 -0.020% 0.995 0.989 -0.009% 0.947 0.902 Min -0.130% 0.975 0.947 -0.074% 0.885 0.730 -0.049% 0.977 0.967 -0.086% 0.945 0.978 -0.054% 0.926 0.788 Max 0.108% 1.014 0.998 0.041% 1.382 1.000 0.033% 1.561 1.000 0.064% 1.035 1.000 0.034% 0.997 0.999 Std 0.099% 0.015 0.022 0.029% 0.124 0.062 0.031% 0.170 0.011 0.049% 0.030 0.008 0.028% 0.027 0.076

Financial

Large Cap

19

MidCap

11

Small

14

Technology

This table presents the descriptive statistics of the following regression run individually for each LETF.
Index Rj,t = + (j Rj,t ) + j,t

is the multiple of the daily target index return LETF promises to deliver.Rj,t is Index is the daily target index return on the daily return of LETF j on day t and Rj,t day t. The Russell 1000 Financial Services index is used for the nancial LETFs, the S&P 500 index is used for the large LETFs, the Russell Mid-Cap is used for mid-cap LETFs, the Russell 2000 is used for the small LETFs and the Nasdaq 100 is used for the technology LETFs. Asset-weighted means are calculated by weighting the regression statistics with the LETF net assets.

30

Table III: Summary Statistics

Financials Market Cap Float-Adjusted Market Cap ADV Daily Volatility Number of Dierent Stocks $10,910,080,286 $10,120,855,466 $125,745,187 0.029 279 Panel A: All Large $22,337,193,312 $21,234,404,621 $203,309,377 0.023 647 Panel B: 2006-2009 Large $22,106,336,850 $21,281,551,450 $203,798,955 0.026 612 Panel C: 2010-2011 Large $22,069,126,399 $21,144,825,291 $202,610,171 0.019 527 MidCap $4,897,070,865 $4,502,949,167 $60,552,960 0.026 1098 Small $652,436,240 $550,360,313 $7,312,225 0.034 2876 Technology $22,027,084,145 $303,993,868 0.024 142

Financials Market Cap Float-Adjusted Market Cap ADV Daily Volatility Number of Dierent Stocks $11,010,250,462 $10,278,857,828 $137,302,091 0.035 263

MidCap $4,803,095,818 $4,404,055,011 $59,599,684 0.028 989

Small $647,800,903 $546,768,728 $7,794,372 0.038 2489

Technology $20,608,669,582 $295,870,892 0.027 129

Financials Market Cap Float-Adjusted Market Cap ADV Daily Volatility Number of Dierent Stocks $10,751,588,913 $9,870,860,785 $107,459,608 0.020 212

MidCap $5,058,046,367 $4,671,959,829 $62,185,885 0.021 876

Small $658,682,773 $555,203,498 $6,662,489 0.030 2382

Technology $24,421,343,488 $317,705,312 0.020 112

This table presents the mean variables of the stocks in the sample. Sample runs from June 19, 2006 to December 31, 2011. Market Capitalization is calculated as the product of the number of shares outstanding and the stock price. Floatadjusted market capitalization is the product of shares available for investment and the stock price. ADV is the 20 day average dollar volume of the stock. Daily volatility is calculated as the the standard deviation of the previous 20 day returns. Financials are the members of the Russell 1000 Financial Services Index. Large Stocks are the members of the S&P 500 index. Mid-cap stocks are the members of the Russell Mid Cap stock index. Small stocks are the members of the Russell 2000. Technology stocks are the members of the Nasdaq 100.

31

Table IV: Net Price Reaction to LETF Rebalancing I

Panel A Financials Intercept LETF Flow/ADV
log (P(15:00) )

Large Cap 0.005 0.006 3.80 0.64

All Mid Cap 0.008 0.006 4.16 0.42

Small Cap 0.008 0.007 0.83 0.08

Technology 0.005 0.006 2.72 0.86

Financials 0.009 0.007 3.09 0.23 0.002 0.006

Large Cap 0.005 0.006 4.32 0.57 -0.011 0.003 684,869 1.36%

All Mid Cap 0.008 0.006 4.28 0.36 -0.004 0.004 1,025,127 1.78%

Small Cap 0.008 0.007 0.86 0.08 -0.006 0.005 2,220,771 2.14%

Technology 0.005 0.006 3.83 0.83 -0.018 0.003 129,800 0.76%

0.009 0.007 3.13 0.25

# of Obs. Adj R2

250,367 6.32%

684,869 1.26%

1,025,127 1.76%

2,220,771 2.12%

129,800 0.52% Panel B

250,366 6.33%

Financials Intercept LETF Flow/ADV

log (P(15:00) )

Large Cap 0.005 0.008 4.76 0.85 -0.010 0.004 437,059 1.43%

2006-2009 Mid Cap 0.011 0.007 4.85 0.54 -0.006 0.005 646,198 1.88%

Small Cap 0.007 0.009 0.78 0.13 -0.009 0.006 1,274,792 0.82%

Technology 0.006 0.008 4.02 1.14 -0.017 0.003 81,511 0.82%

Financials 0.004 0.010 2.74 0.31 -0.008 0.006 96,957 5.48%

Large Cap 0.005 0.010 3.78 0.69 -0.013 0.005 247,810 1.30%

2010-2011 Mid Cap 0.004 0.009 3.57 0.42 -0.001 0.006 378,929 1.71%

Small Cap 0.008 0.011 0.88 0.09 -0.001 0.006 945,979 4.09%

Technology 0.003 0.009 3.48 1.08 -0.018 0.004 48,289 0.66%

0.013 0.010 3.39 0.32 0.010 0.008 153,409 7.04%

32

# of Obs. Adj R2

This table presents the results of the following regression. log (P(i,15:00) ) LET F F lowi log(Pi,c ) = 0 + 1 + 2 + i i ADVi i is the return of stock i between 15:00pm(ET) and 16:00pm(ET) scaled by its daily volatility, i - the standard deviation of previous 20 days returns. LET F F lowi is the share of stock i from the total LETF rebalancing ow calculated as a function of the target index return between the previous days close and 15:00pm. ADVi is the past log (P(i,15:00) ) 20 day average dollar trading volume of stock i. is the return of stock i from the previous days close to i 15:00pm scaled by its daily volatility. Standard errors are clustered daily and reported below the coecient estimates.
log (Pi ,c) i

Table V: Net Price Reaction to LETF Rebalancing: Market UP vs Market DOWN Panel A: Market Up Financials Large Cap Mid Cap Small Cap Intercept LETF Flow/ADV
log (P(i,15:00) )

Technology 0.001 0.008 4.66 1.18 -0.018 0.003 69,825 0.92%

0.021 0.011 3.06 0.34 -0.005 0.006 124,625 4.36%

0.009 0.007 4.46 0.67 -0.014 0.003 364,725 1.33%

0.020 0.006 4.36 0.40 -0.010 0.003 543,413 1.74%

0.041 0.009 0.63 0.08 -0.019 0.004 1,123,013 1.20%

# of Obs. Adj R2

Panel B: Market Down Financials Large Cap Mid Cap Small Cap Intercept LETF Flow/ADV
log (P(i,15:00) )

Technology -0.001 0.011 2.93 1.49 -0.017 0.005 59,975 0.37%

-0.009 0.012 2.60 0.44 0.001 0.008 125,741 2.54%

-0.003 0.010 3.60 0.96 -0.011 0.006 320,144 0.69%

-0.008 0.008 3.36 0.52 -0.009 0.006 481,051 0.87%

-0.018 0.011 0.80 0.14 -0.012 0.008 1,097,758 1.43%

# of Obs. Adj R2

This table presents the results of the following regression for Market Ups and Market Downs. log(P(i,15:00) ) log (Pi,c ) LET F F lowi = 0 + 1 + 2 + i i ADVi i is the return of stock i between 15:00pm(ET) and 16:00pm(ET) scaled by its daily volatility, i - the standard deviation of previous 20 days returns. LET F F lowi is the share of stock i from the total LETF rebalancing ow calculated as a function of the target index return between the previous days close and 15:00pm. ADVi is the past 20 day average dollar trading volume of stock log (P(i,15:00) ) is the return of stock i from the previous days close to 15:00pm i. i scaled by its daily volatility.. Standard errors are clustered daily and reported below the coecient estimates.
log (Pi ,c) i

33

Table VI: Next Day Price Reversals

Panel A: All Large Cap Mid Cap Small Cap 0.007 0.016 -3.52 1.08 0.009 0.008 684,233 0.09% 0.003 0.015 -2.52 0.68 0.019 0.008 999,363 0.08% -0.012 0.015 -0.35 0.10 -0.015 0.007 2,217,895 0.13%

Financials Intercept
LET F F low ADV t1 log (P ) t1

Technology 0.008 0.016 -4.24 1.65 0.001 0.008 129,658 0.12%

-0.024 0.018 -1.74 0.43 0.006 0.011

# of Obs. Adj R2

250,088 0.25%

Financials Intercept
LET F F low ADV t1 log (P ) t1

Panel B: 2006-2009 Large Cap Mid Cap Small Cap -0.003 0.019 -4.77 3.89 0.008 0.009 436,987 0.03% -0.005 0.018 -2.66 0.94 0.023 0.009 631,468 0.09% -0.021 0.019 -0.49 0.19 -0.019 0.010 1,274,298 0.13%

Technology 0.004 0.019 -4.62 2.09 0.008 0.009 81,481 0.12%

-0.046 0.022 -2.00 0.54 0.012 0.015

# of Obs. Adj R2

153,342 0.30%

Financials Intercept
LET F F low ADV t1 log (P ) t1

Panel C: 2010-2011 Large Cap Mid Cap Small Cap 0.013 0.029 -5.23 4.66 -0.001 0.013 247,296 0.06% 0.015 0.028 -2.31 0.96 0.013 0.014 367,895 0.06% -0.001 0.025 -0.31 0.11 -0.010 0.012 943,597 0.14%

Technology 0.016 0.029 -3.33 2.65 -0.014 0.014 48,177 0.13%

0.011 0.028 -1.28 0.70 -0.008 0.016

# of Obs. Adj R2

96,746 0.21%

This table presents the results of the following regression. log (Pi,15:00 ) i LET F F lowi ADVi
t1

= 0 + 1

+ 2

log (Pi ) i
t1

+ i

log(Pi,15:00 ) is the return of stock i between the previous days close and 15:00pm scaled by its daily volatility, i i

the standard deviation of its previous 20 days returns. LET F F lowi,t1 is the one-day lagged share of stock i from the total LETF rebalancing ow calculated as a function of the target index return between the previous days close and 15:00pm. ADVi,t1 is the one-day lagged past 20 day average dollar trading volume of stock i.
log(Pi ) i t1

is

one-day lagged return of stock i scaled by its daily volatility. Standard errors are clustered daily and reported below the coecient estimates.

34

Table VII: End-of-day Volatility and LETF Rebalancing

Panel A Financials Intercept
LET F F lowi ADVi log (Pi,(15:00) ) 2

Large Cap 0.093 0.003 2.33 0.38

All Mid Cap 0.097 0.003 2.22 0.22

Small Cap 0.157 0.011 0.73 0.21

Technology 0.079 0.003 3.49 0.63

Financials 0.089 0.005 1.30 0.20 0.028 0.002

Large Cap 0.075 0.003 1.72 0.34 0.021 0.001 684,869 4.15%

All Mid Cap 0.076 0.003 1.96 0.21 0.022 0.001 1,025,127 4.86% 2010-2011 Mid Cap 0.070 0.005 1.34 0.24 0.019 0.003 378,929 4.16%

Small Cap 0.114 0.034 0.59 0.24 0.045 0.038 2,220,771 8.63%

Technology 0.067 0.004 2.57 0.60 0.016 0.001 129,800 3.90%

0.109 0.005 1.58 0.21

# of Obs. Adj- R 2

250,367 2.14%

684,869 0.98%

1,025,127 1.22% 2006-2009 Mid Cap 0.079 0.003 2.54 0.31 0.024 0.002 646,198 5.45%

2,220,771 0.26%

129,800 1.43% Panel B Technology 0.077 0.005 5.35 1.60 0.022 0.002 81,511 4.67%

250,366 5.11%

Financials Intercept
LET F F lowi ADVi log (Pi,(15:00) ) 2

Large Cap 0.078 0.004 2.46 0.51 0.023 0.002 437,059 4.72%

Small Cap 0.045 0.060 0.64 0.25 0.122 0.064 1,274,792 26.05%

Financials 0.077 0.008 1.03 0.24 0.022 0.004 96,957 4.11%

Large Cap 0.069 0.005 1.00 0.39 0.017 0.002 247,810 3.57%

Small Cap 0.136 0.020 0.71 0.29 0.003 0.002 945,979 1.14%

Technology 0.064 0.006 0.86 0.71 0.014 0.001 48,289 3.07%

0.095 0.006 1.58 0.27 0.032 0.002 153,409 6.03%

35

# of Obs. Adj- R 2

This table presents the results of the following regression. log(Pi,c ) i

log (Pi,c ) i 2

= 0 + 1

log (Pi,(15:00) ) LET F F lowi + 2 ADVi

+ i

is the return of stock i between 15:00pm(ET) and 16:00pm(ET) scaled by its daily volatility, i -the standard deviation of previous 20 days returns. |LET F F lowi | is the absolute value of the share of stock i from the total LETF rebalancing ow calculated as a function of the target index return between the previous days close and log (P(i,15:00) ) 15:00pm. ADVi is the past 20 day average dollar trading volume of stock i. is the return of stock i from i the previous days close to 15:00pm. Standard errors are clustered daily and reported below the coecient estimates.

Table VIII: Robustness Results

Panel A: End-of-Day Return Financials Intercept LETF Flow/ADV
log (P15:00 )
Index ) log (P15:00 Index S &P 500 ) log (P15:00 S &P

Large 0.005 0.006 3.356 0.598 -0.018 0.002 0.015 0.005

MidCap 0.008 0.006 3.909 0.371 -0.016 0.002 -0.015 0.015 0.034 0.016

Small 0.008 0.007 0.545 0.069 -0.028 0.003 -0.024 0.016 0.072 0.018 2,220,766 3.11% Small 0.112 0.028 0.561 0.156 0.045 0.038 0.005 0.005 -0.005 0.013 2,220,766 8.65%

Technology 0.005 0.006 3.564 1.075 -0.019 0.002 0.022 0.011 -0.019 0.011 129,799 0.93% Technology 0.068 0.004 2.999 0.641 0.016 0.001 0.002 0.002 -0.004 0.002 129,799 4.02%

0.009 0.007 2.528 0.329 -0.008 0.006 0.003 0.015 0.021 0.020 250,365 6.61% Financials

# of Obs. Adj R2

684,868 1.54% Large 0.075 0.004 1.670 0.338 0.020 0.001 0.001 0.001

1,025,122 2.11% MidCap 0.073 0.003 2.037 0.217 0.021 0.001 -0.003 0.001 0.005 0.003

Panel B: End-of-Day Volatility Intercept |LET F F low/ADV |

log (P15:00 ) 2
Index ) 2 log (P15:00 Index S &P 500 ) 2 log (P15:00 S &P

0.089 0.005 1.303 0.197 0.028 0.002 0.0004 0.001 -0.0003 0.0005

# of Obs. Adj R2

250,365 5.12%

684,868 4.16%

1,025,122 4.90%

This table presents the regression results for robustness checks.

log(Pi ,c) , the dependent variable in Panel A, is the i

return of stock i between 15:00pm(ET) and 16:00pm(ET) scaled by its daily volatility, i - the standard deviation of previous 20 days returns. LET F F lowi is the share of stock i from the total LETF rebalancing ow calculated as a function of the target index return between the previous days close and 15:00pm. ADVi is the past 20 day average dollar trading volume of stock i. its daily volatility.
log(P(i,15:00) ) i Index ) log(P15:00 is the return of the underlying index from the previous days close to 15:00pm scaled Index S &P ) log(P15:00 by its daily volatility. is the return of the S&P 500 index from the previous days close to 15:00pm scaled S & P log(Pi,c ) 2 is the dependent variable in Panel B. |LET F F lowi /ADV | is the absolute value by its daily volatility. i

is the return of stock i from the previous days close to 15:00pm scaled by

of the share of stock i from the total LETF rebalancing ow scaled by its average dollar trading volume. Standard errors are clustered daily and reported below the coecient estimates.

36

Table IX: Stock Market Indexes and LETFs

Category Variables ADV (in billions) Volatility (LET F F low)/(0.01 ADV ) log (P ) (in bps) ADV (in billions) Volatility (LET F F low)/(0.01 ADV ) log (P ) (in bps) ADV (in billions) Volatility (LET F F low)/(0.01 ADV ) log (P ) (in bps) ADV (in billions) Volatility (LET F F low)/(0.01 ADV ) log (P ) (in bps) ADV (in billions) Volatility (LET F F low)/(0.01 ADV ) log (P ) (in bps) 2007 $29.325 0.013 0.073 3.353 $112.906 0.009 0.093 3.121 $55.021 0.010 0.027 0.672 $20.748 0.012 0.129 3.818 $34.865 0.011 0.309 7.303 2008 $36.687 0.037 0.424 53.506 $132.555 0.022 0.192 22.405 $59.520 0.023 0.026 2.041 $18.103 0.025 0.414 25.974 $36.699 0.023 0.307 20.768 2009 $28.628 0.034 1.020 96.992 $99.618 0.016 0.394 23.221 $43.432 0.019 0.025 1.436 $11.306 0.022 0.776 31.963 $24.999 0.016 0.365 13.110 2010 $24.433 0.015 1.214 38.548 $107.613 0.011 0.385 13.656 $49.387 0.012 0.025 0.814 $13.311 0.015 1.127 31.323 $31.692 0.011 0.328 8.358 2011 $22.516 0.018 1.077 45.083 $111.514 0.013 0.323 15.578 $54.263 0.015 0.025 1.124 $15.512 0.019 1.212 51.189 $35.318 0.014 0.297 9.800

Financial Stock Index

Large Cap Stock Index

Mid-Cap Stock Index

37
Small Stock Index Technology Stock Index

This table presents the daily averages of the variables used in implied price impact computation (Kyle and Obizhaeva, 2011a,b). log (P ) = /104 . ADV 40 106
1/3

0.02

4/3

LET F F low 0.01 ADV

LETF Flow is the total rebalancing ow of LETFs in response to a 1% change in the target stock index. ADV is the total daily dollar volume of the member stocks. Volatility () is dened as the standard deviation of the previous 20 day index returns. is 5.78.

Table X: Investor Flows into the Leveraged and Inverse ETFs

Investor Flow Parameter Intercept InvestorF lowt1 InvestorF lowt2 InvestorF lowt3 InvestorF lowt4 InvestorF lowt5 InvestorF lowt6 InvestorF lowt7 InvestorF lowt8 InvestorF lowt9 InvestorF lowt10 LET F Returnt1 LET F Returnt2 LET F Returnt3 LET F Returnt4 LET F Returnt5 LET F Returnt6 LET F Returnt7 LET F Returnt8 LET F Returnt9 LET F Returnt10 V IXt1 V IXt2 V IXt3 V IXt4 V IXt5 V IXt6 V IXt7 V IXt8 V IXt9 V IXt10 Adj R2 # of Obs 0.14 0.10 0.06 0.04 0.04 0.04 0.004 0.02 0.02 0.04 0.02 t-statistic 17.25 16.90 10.56 6.91 7.31 6.63 0.73 4.05 3.46 6.71 4.08 Investor Flow Parameter 0.14 0.09 0.06 0.04 0.04 0.04 0.00 0.02 0.02 0.04 0.02 -0.08 -0.05 -0.04 -0.02 -0.02 -0.01 0.000 -0.001 -0.001 -0.005 t-statistic 18.20 14.87 9.42 6.39 6.84 6.65 0.78 4.12 3.36 6.54 4.10 -24.15 -15.34 -10.79 -5.10 -5.72 -2.96 0.16 -0.19 -0.44 -1.47 Investor Flow Parameter 0.14 0.09 0.06 0.04 0.04 0.04 0.00 0.02 0.02 0.04 0.02 -0.08 -0.05 -0.04 -0.02 -0.02 -0.01 0.000 -0.001 -0.001 -0.005 0.01 0.02 0.02 0.02 -0.002 0.01 0.01 0.002 0.003 0.003 5.11% 103,807 t-statistic 18.16 14.81 9.37 6.33 6.85 6.63 0.78 4.18 3.36 6.58 4.13 -24.30 -15.49 -10.98 -5.33 -5.68 -3.03 0.08 -0.17 -0.46 -1.48 2.96 3.32 3.80 5.23 -0.52 1.63 2.34 0.36 0.70 0.81 Investor Net Inow Parameter 1.96 0.17 0.13 0.08 0.09 0.08 0.02 0.06 0.06 0.08 0.06 -0.11 -0.04 -0.02 0.03 0.00 0.02 0.036 0.026 0.034 0.026 0.05 0.08 0.10 0.12 0.04 0.08 0.09 0.02 0.05 0.05 9.77% 22,027 t-statistic 65.94 10.73 8.78 5.08 6.25 5.55 1.18 4.33 4.10 5.25 4.05 -11.33 -4.25 -1.93 2.59 0.12 1.92 3.61 2.56 3.47 2.71 3.43 5.00 6.40 7.90 2.55 5.02 5.45 1.54 2.98 3.05 Investor Net Outow Parameter -1.33 0.02 -0.02 -0.02 -0.02 0.01 -0.03 -0.02 -0.03 0.00 0.00 -0.12 -0.08 -0.07 -0.06 -0.06 -0.04 -0.028 -0.026 -0.030 -0.024 -0.06 -0.06 -0.05 -0.04 -0.07 -0.07 -0.07 -0.03 -0.06 -0.04 7.39% 20,254 t-statistic -72.14 1.94 -2.04 -1.65 -1.64 0.57 -2.33 -1.81 -2.63 0.34 0.10 -16.81 -12.62 -10.66 -8.75 -8.50 -6.73 -4.10 -3.94 -4.39 -3.76 -5.58 -5.05 -5.04 -3.90 -7.02 -6.26 -5.66 -2.76 -5.78 -3.56

38

3.24% 103,807

5.05% 103,807

This table presents the results of the regression of Investor Flows into the LETFs. Dependent variables, Investor Flow, is the product of daily shares outstanding change and net asset value of LETF scaled by its lagged net assets. Explanatory variables are lagged daily Investor Flows, lagged daily LETF returns and lagged changes in VIX. All variables are in percentages. t-statistics are adjusted for heteroskedasticity and reported next to the coecient estimates. Net Inows are positive Investor Flows and Net Outows are negative Investor Flows.

Figure 1: Hypothetical LETF Rebalancing Flows as a Fraction of Daily Stock Volume

7%

6%

5%

4%

3%

2%

1%

0%

Large Cap

Financials

Mid Cap

Small Stocks

Technology

This gure illustrates the share of total hypothetical rebalancing ows for an average stock as a fraction of its daily volume for each category. Total LETF ows are calculated as the sum of all LETFs in that category in response to a 1% increase in the target index and allocated into stocks based on their weights in the target index. Daily volume of a stock is dened as the previous 20 day average of the stocks dollar volume.

39

Figure 2: Hypothetical LETF Rebalancing Flows as a Fraction of Stock Volume in the Last Hour

20%

18%

16%

14%

12%

10%

8%

6%

4%

2%

0%

Large Cap

Financials

Mid Cap

Small Stocks

Technology

This gure illustrates the share of total hypothetical LETF rebalancing ows for an average stock as a fraction of its volume between 15:00pm and 16:00pm for each category. Total LETF ows are calculated as the sum of all LETFs in that category in response to a 1% increase in the target index and allocated into stocks based on their weights in the target index. Volume of a stock is dened as the previous 20 day average of the stocks dollar volume between 15:00pm and 16:00pm.

40

Figure 3: Implied Price Impact of LETF Rebalancing

2.6

2.4

2.2

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

Percent

ECB-G7 Emergency Meeting

0.0 2006 2007 2008 2009 2010 2011

Financial Stability Plan Announcement

This gure plots the price impact estimates implied by market microstructure invariance(Kyle and Obizhaeva, 2011a,b). ADV 40 106
1/3

Daily

Large Cap Finance Mid Cap Small Cap Technology

Failure of Lehman Brothers

log (P ) = /10 .

0.02

4/3

LET F F low 0.01 ADV

ADV is the total daily dollar volume of the member stocks. is the standard deviation of the previous 20 days index returns. is 5.78. LETF Flow is the total LETF rebalancing ow in that category in response to a 1% change in the target index.

41

Figure 4: Frequency of A Late-day Large Price Move in the Financial Stock Index

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

Probability

ECB-G7 Emergency Meeting

0.0 2003 2004 2005 2006 2007 2008 2009 2010 2011

Failure of Lehman Brothers

This gure plots the frequency of a 1% price move in the Russell 1000 Financial Services Index between 3:00pm and 4:00pm given that the index has already moved 1% by 3:00pm in the same direction. This measure is calculated from a months trading days in which the index moved 1% by 3:00pm. More formally this frequency is:
n

i=0 n

{Amplified Return} {|R3pm |1%}

where

Monthly

Launch of First Financial LETF

Financial Stability Plan Announcement

{Amplified Return}

i=0

1 1 0

R34pm 1%, R3pm 1% R34pm 1%, R3pm 1% Otherwise

42