1. Mr. Madoff’s Amazing Returns: An Analysis of the Split-Strike Conversion Strategy Ken Jeong Written by Carole Bernard and Phelim Boyle

2. Agenda l Summary of Paper l Bernard Madoff l Ponzi Scheme l Investment return of Fairfield Sentry l GISW method l Theoretical Analysis of the Split-strike conversion strategy l Numerical Analysis l Conclusion

3. The summary of this paper l The authors of this paper analyze the Madoff’s previous investment strategy, which is called split- strike strategy in general and derive expressions for the expected return, standard deviation, Sharpe ratio, and correlation with the market of this strategy. l They find that Madoff’s returns lie well outside their theoretical bounds and should have raised suspicions about Madoff’s performance eariler than 2008. l In this paper, we will specifically see the investment return of Fairfield Sentry Ltd hedge fund, which was one of the Madoff’s feeder funds.

4. Bernard Madoff l He is the former non-executive Chairman of the NASDAQ stock market, and the admitted operator of a Ponzi scheme that is considered to be the largest fraud in U.S. history. l Madoff had attracted a wide following because he delivered consistently high returns with very low volatility over a long period. l He claimed to use a split-strike conversion strategy to obtain these low-risk return. l Madoff pled guilty to 11 federal felonies and was sentenced to 150 years in prison in 2009.

5. Ponzi Scheme l A ponzi scheme is an investment fraud wherein the operator promises high financial returns or dividends that are not available through traditional investments. Instead of investing victim’s funds, the operator pays “dividends” to initial investors using the principle amounts “Invested” by subsequent investors. l Ponzi schemes are named for Charles Ponzi, who deceived thousands of New England residents into investing in postage stamps back in the 1920s. Ponzi thought he could take advantage of differences between U.S. and foreign currencies used to buy and sell international mail coupons. Ponzi told investors that he could provide a 40- 50% return in just 90days compared to the interest rates for bank savings accounts, stocks, and bonds.

6. 1. Fairfield Sentry Monthly Returns from December 1990 to October 2008 l The annual average return of FS is 10.59% with annual volatility of 2.45% over this period. l The annual average return from investing in the S&P has been 9.64% with a standard deviation of 14.28% over this period. l Investors clearly put a very high value on this combination of high returns and low volatility.

7. 2. Fairfield Sentry Monthly Returns from December 1990 to October 2008

8. Hypothetical Split-strike strategy l Long position in S&P 500 Index, short call, and long put. Both options are European and have one month maturity. l Call price will exceed the put price and the proceeds are invested for one month at risk-free rate. l At the end of month, the option positions are settled in cash. l All available monies are invested in the index at the start of the second month and the same option strategy is implemented and repeated every month for 215 months. l One month U.S. T-bill rates to proxy risk-free rate. l Use VIX to proxy implied volatility. l Receive monthly dividend yield on the Index

9. Hypothetical Split-strike strategy l Hypothetical Split-strike conversion strategy has a higher expected return than the FS strategy, However, the returns are more highly correlated with the S&P Index and have much higher volatility than the FS strategy.

10. Hypothetical Split-strike strategy l Volatility skew dramatically reduces the returns on the split-strike strategy by about 5% of annual return. l Severe volatility skew will reduce the attractiveness of the split-strike strategy.

11. Summary Performance Statistics for Four Strategies for the Period December 1990 to October 2008

12. Manipulation-proof performance measures l Recently Goetzmann, Ingersoll, Spiegel, and Welch have developed a manipulation-free portfolio performance measure. l Even though their measure was not designed to detect outright fraud, it can provide valuable insights on the nature of the split- strike strategy and the Fairfield Sentry returns. l This equation measures the portfolio’s implied excess return after adjusting for risk.

13. Manipulation-proof performance measures l For the risk-free portfolio, Θ = 0. l If one had a portfolio that earned exactly 50 bps above the risk- free rate every month with no variation, this portfolio would have Θ = 0.06. l If the portfolio is risky, then Θ will decrease if a more risk-averse investor is considered.

14. Manipulation-proof performance measures l This table shows the value of Θ corresponding to Different levels of ρ(risk aversion) for three investment strategies. l Fairfield Sentry portfolio outperformed the other strategies based on this measure. l If the investor becomes more risk averse, the value of Θ declines rapidly for the investment in the S&P and the split-strike strategy. l Based on this measurement, The Fairfield Sentry returns outperformed the other strategies and gained extra 6% per year above the risk-free rate for all investors.

15. Theoretical Analysis of the Split-Strike Strategy l Mr. Madoff claimed to use a Split-Strike conversion strategy to obtain high-return with low-risk. l Long position in Equities together with a short call and a long put on an equities. l The primary purpose of the long put is to limit the market risk of the stock basket at the strike price of the long puts. l The primary purpose of the short call is to largely finance the cost of the put hedge and increase the stand-still rate of return

16. Theoretical Analysis of the Split-Strike Strategy l At time 0, Buy one share of the index and simultaneously sells a call option at the premium C 0 and buy a put option at the premium P 0. l The call option has a strike price K c and the put option has a strike price K p. l Both options have the same time to maturity and are priced by the Black-Scholes formula. l The value of the portfolio at time 0 is

17. Theoretical Analysis of the Split-Strike Strategy l The amount of (c 0 -p 0 ) accumulates to the end of the period at the risk-free rate and the value of the call option and put option at time h are denoted by C h and P h l The value of the portfolio at time h is l The value of call option and put option

18. Theoretical Analysis of the Split-Strike Strategy l When K c = K p = K, the pay off of the strategy is equal to based on put-call parity. l However, we are interested in Sharpe ratio of this portfolio when the strike of the put option is different from the strike of the call option.

19. Theoretical Analysis of the Split-Strike Strategy l Expectation payoff the strategy is equal to l The expected return from the strategy is an increasing function Kc and a decreasing function Kp. l The split-strike strategy has a lower expected return and a lower variance than a direct investment in the index. l The correlation of the strategy with the index S can be written as

20. Numerical Analysis l Consider a split-strike strategy with maturity T=h. We assume the strikes of the call option and the put option are given by K c = S 0 + b and K p = S 0 – b, where b ∊ [0, S 0 ]. l We study the case when Kp and Kc are chosen independently and discuss the optimal choices of the two strikes.

21. 7. Expectation and Standard deviation of the Split- strike strategy

22. 8. Sharpe ratio of the Split-strike conversion strategy compared to the Sharpe ratio of Investing in the index

23. 9. Correlation between S h and V h and Beta of the strategy

24. 10. Sharpe Ratio of the Strategy versus Sharpe Ratio of Investing in the index

25. 11.Choice of the Strikes Kc and Kp to Maximize the Sharpe Ratio

26. Conclusion l Investment performance of Madoff is now known that these results were based on a giant Ponzi scheme which flourished for a long time despite several red flags and the highly suspicious nature of the returns. l The Sharpe ratio based on the paper’s split-strike strategy was very much lower than Fairfield Sentry’s Sharpe ratio over the same period. In addition, the correlation between the split-strike strategy and the market was more than twice the corresponding correlation for the Madoff’s strategy. l One of the most unbelievable statistics of Madoff’s performance is the very low volatility. l The theoretical analysis of this paper reaches that there is a theoretical maximum Sharpe ratio that can be obtained using options. On the other hand, the performance statistics reported by Fairfield Sentry lie well outside their theoretical bounds. l In summary, there are some simple quantitative diagnostic that should have raised suspicions about Madoff’s performance.