db Currency Returns

1 Db Currency ReturnsGLOBAL MARKETS| FOREIGN EXCHANGEC arrynOne of the most widely known andprofitable strategies in currencymarkets are carry trades, where onesystematically sells low interest ratecurrencies and buys high interestrate a strategy exploits whatacademics call forward-rate bias or the forward premium puzzle ,that is, the forward rate is not anunbiased estimate of future spot. Put another way, contrary to classicalnotions of efficient markets, carrytrades have made money over believe the reason this is possible is that investors whoemploy the carry trade exposethemselves to Currency taking this risk arerewarded by positive Returns over : Bond Indices Investors can earn a higher term premia by increasing durationMomentumnA widely observed feature ofcurrency markets is that manyexchange rates trend on a multi-yearbasis.

2 Therefore, a strategy thatfollows the trend, typically makespositive Returns over segmentation of currencymarket participants with some acting quickly on news while othersrespond more slowly is one reasonwhy trends emerge and can : Equity Indices Investing in amarket capitalisation weighted equityindex is effectively employing a longmomentum strategyValuationnIn the long-run, currencies tend tomove towards their fair value .Consequently, systematically buying undervalued currencies and selling overvalued currencies is profitablein the of the strongest conclusions inacademia is that fundamentals tendnot to work for currencies in theshort-to medium-term, yet they dolong-term.

3 One of the oldestmeasures of fair value , purchasingpower parity, has been shown towork in the long-run. Parallel: Equity Indices Similar toincorporating a fundamental metricsuch as earnings or revenuesDBCR Component StrategiesDBCR Historical ReturnsnThere are 20 years of data on freely floating currencies (since the end of Bretton Woods)by which to assess the performance of systematic strategies for investing in currenciesn3 strategies reflect the most widely used FX investment styles - Carry, Momentum and Valuation nThese form the core investment approach of many FX only funds and are supported bydecades of academic work*nLong term.

4 There is a low correlation between the Returns of employing these strategiesand the Returns gained from investing in more traditional asset classes such as equitiesand bondsnDBCR provides benchmark exposure to these systematic strategiesPerformance of the DBCR Excess Returns from June 1989 February 2009 Jun-89 Jun-91 Jun-93 Jun-95 Jun-97 Jun-99 Jun-01 Jun-03 Jun-05 Jun-07 Jun-0980100120140160180200220240260280 CarryMomentumValuationDBCRDBCR Performance StatisticsMomentumCarryAnnualised Total return in USDA nnualised VolatilityAnnualised Excess ReturnSharpe RatioCorrelation Daily Returns to DBCR% Down months% Up monthsBiggest Monthly GainMean Monthly ReturnBiggest Monthly LossLongest Profitable Streak (Months)Longest Losing Streak (Months)Maximum DrawdownFirst StartEnd Run-upFirst StartEnd : Deutsche Bank, June 1989 to February 2009 Source: Deutsche Bank* Meese and Rogoff (1983), Empirical Exchange Rate Models of the Seventies: Do They Fit Out of Sample?

5 Cheung, Chinn, Pascual (2003), Empirical Exchange RateModels of the Nineties: Are Any Fit to Survive? Rogoff (1996), The Purchasing Power Parity Puzzle New Palgrave Dictionary of Economics, Purchasing Power Pairty), Taylor, Peel, Sarno (2001), Non-Linear Mean-Reversion in Real Effective Exchange Rates: Towards a Solution to the Purchasing Power Parity Puzzles Chaboud and Wright (2002), Uncovered Interest Parity: It Works, But Not For Long Engel (1995), The Forward Discount Anomaly and the RiskPremium: A Survey of Recent Evidence Sarno and Taylor (2002), The Economics of Exchange and Menkhoff (2003), Technical Analysis in ForeignExchange The Workhorse Gains Further Ground.

6 Park and Irwin (2006), What Do We Know About theProfitability of Technical Analysis? Neely, Weller, Ulrich (2006): The Adaptive MarketsHypothesis: Evidence from the Foreign Exchange Market , Qi and Wu (2006), Technical Trading-Rule Profitability, Data Snooping and Reality Check: Evidence from the Foreign Exchange Market , Park and Irwin (2005), A Reality Check on Technical TradingRule Profits in US Futures Markets Menkhoff and Taylor (2006), The Obstinate Passion of foreignExchange Professionals: Technical Analysis , DB Currency Carry IndexnRe-balance every 3 months, with the Roll Dates set as the thirdWednesday of March, June,September and DecembernRank each Currency by its 3m LiborRate on the Observation Date (1 week before the quarterly IMM date)

7 NAllocate a 1/3 long position to eachof the 3 currencies with the highest3m Libor ratenAllocate a 1/3 short position to eachof the 3 currencies with the lowest3m Libor ratenTransact Forwards in each currencyto the next Roll DateDB Currency Momentum IndexnRe-balance every month, with the Roll Dates set as the thirdWednesday of each month nRank each Currency by its 12 monthSpot return vs. the USD on theObservation Date (1 week before the monthly IMM date)n12 month Spot return vs. the USDis defined as : Current spot level ofthe Currency versus the USD [inCCY/USD terms] divided by the spotlevel for the Currency versus theUSD 12 months ago [in CCY/USDterms], minus 1nAllocate a 1/3 long position to eachof the 3 currencies with the highest12 month Spot return vs.

8 The USDnAllocate a 1/3 short position to eachof the 3 currencies with the lowest12 month Spot return vs. the USDnTransact Forwards in each currencyto the next Roll DateDB Currency Valuation IndexnRe-balance every 3 months, with the Roll Dates set as the thirdWednesday of March, June,September and DecembernRank each Currency by its Valuationon the Observation Date (1 weekbefore the quarterly IMM date). The most undervalued Currency isrepresented by the lowest ValuationnValuation is defined as : Averagespot level of the Currency versus theUSD [in CCY/USD terms] over thelast 3 months divided by the latestOECD Purchasing Power Parityfigure for that Currency versus theUSD [in CCY/USD terms]nAllocate a 1/3 long position to eachof the 3 currencies with the lowestValuationnAllocate a 1/3 short position to eachof the 3 currencies with the highestValuationnTransact Forwards in each currencyto the next Roll DateDBCR Index ConstructionThe DBCR is an investable index that captures the long term systematic returnsavailable by investing in the world's Currency markets.

9 It replicates the three strategies most widely employed in the FX market and wraps them all into a single non-discretionary index with daily Currency Returns IndexnThe DBCR index is quoted in EUR or USD notional and in excess return terms,representing the Returns of an unfunded investmentnDBCR invests in one third of each of the following 3 indices on a daily basis:nDB Currency Carry Index, nDB Currency Momentum Index,nDB Currency Valuation IndexnThe pool of currencies eligible for inclusion in each of these indices is: USD, EUR, JPY, GBP, CHF, AUD, NZD, CAD, NOK, SEKnDaily index closing levels are published to DBIQ and BloombergTimeline for re-balancingObservation Date(IMM 5 days)Currencies are rankedTwo Roll Dateswithin the Roll Window are determinedRoll WindowThe two chosen RollDates are made publicWedThuFriMonTueWedThu FriMonIMMDBCR Historical CompositionOn the right is a table outlining thepercentage of time each Currency hasspent at each allocation level.

10 At any pointin time a Currency s allocation in the DBCRcan range from being long one third of theindex notional when it is selected as a Long Currency in all three componentstrategies, to being short one third of the index notional when it is selected as a Short Currency in all threecomponent strategies. Equities:S&P500100%EuroStoxx50100%Nikkei 100%Interest Rates:US 2 Yr Swap100%US 10 Yr Swap100%Commodities:DBLCI-OY100%Currenci es:DXY Index100%DBCR100%S&P500 EuroStoxx50 NikkeiUS 2 Yr SwapUS 10 Yr SwapDBLCI-OYDXY IndexDBCR 48%11%42%41%26%-7%5%48%46%18%21%36%-28%2 0%11%46%-2%6%25%-27%16%42%18%-2%83%11%15 %9%41%21%6%83%19%11%8%26%36%25%11%19%-46 %4%-7%-28%-27%15%11%-46%12%5%20%16%9%8%4 %12%DBCR Correlation to Other Asset ClassesCorrelations between January 2007 and February 2009 Source: Deutsche BankSource.