DB's FX Strategy Weekly (2007/04/27)

DB's FX Strategy Weekly
April 27, 2007

1. 10 FX Carry Seasonal

If we look at the each monthly return's deviation from entire sample average performance carry trade, May is the worst month in last 10 year (1997-2006), August and July are the next two. In the other side, January, November are the best two month.


2. Carry trade has a highly correlation with equities (MSCI World Index)

If we look at the correlation between G10 carry trade and MSCI World Index, there is a interesting appearance. Before Feb 2007, G10 carry trade's 1m correlation with MSCI World is low positive even negative. Since Feb 2007, They become highly positive correlation.

Compared to G10 carry trade, Global carry's correlation with MSCI World remains positive usually. But it is too highly right now.


3. Stephen's thought

The implication is that carry trade is supported by global equities' better than expected performance since Feb 2007. But, this would be not worked well all the time.

In this stage, US economics's housing problem (bad) and S&P 500 performance (good) just offsets each other's effect. Further more, this help to built up the carry trade environment.

LB's Global Fx Strategies, April 26, 2007

LB's Global Fx Strategies
April 26, 2007

I. Watch M&A trends in the G10

1. M&A is a kind of capital flow proxies. In the 1990s, the US economy was the beneficiary of the M&A boom and foreign direct investment (FDI) followed.

2. M&A activity has some leading indication for FDI.

3. In LB's research, EUR and GBP benefit from M&A activity.

4. LB's M&A pressures indices
(1) Data: LB collects the aggregate of announced deal over $500m in the US, European and Japan by target company and presents as % of global capitalisation.
(2) LB's M&A pressures index = time index * ln(deal size)
where time index = the change * (0.27 * ln(days following announcement of deal))
where the change = ??
(Stephen's thought: It should be a summation concept of M&A data flow.)


II. G10 FX Strategy

1. LB's study shows that a simple pairwise regressions over the period 2005-2007 (daily data) suggest that every 20bp convergence, EUR/USD is able to appreciate by 2.2 cents. The correlation and explanatory power the two-year spread (German / US spread) is able to exert on EUR/USD remain high (0.84 and 0.7). (p.7)



III. Taiwan Focus

1. USD/TWD Spot (QoQ change) v.s. Financial account balance has positive correlation. (see Figure 1 at p.29)

2. An increase in the Lifer's oversea investment gap from 35% to 50% of total assets generate around $35.9 bn capital outflow (February life insurance total assets at $239bn). The potential outflows over next 2 year period could average $1.5bn to $3.0bn per month.

3. LB takes a long INR/TWD position this week, TWD as a funding role.

4. The growth backdrop in Taiwan is weak, while the CBC's monetary policy tightening cycle is close to its end.(??)

LB's Global FX Strategies, April 19, 2007

LB's Global FX Strategies
April 19, 2007

I. Trading FX on an episodic risk basis, p.23

1. FX carry trades could be considered a strategy with steady positive return but with infrequent large drawdowns.

2. LB's risk filter

• Market Risk Sentiment Index, MARS
• RPI
• VIX index
• Credit Market Spread (= Moody's AAA rated corporate bond yields less 30-year Treasury yield)
• FX implied volatility (Global Hazard Index, GHI) see definition on p.25.

3. Comparing the difference among passive and activity carry trade strategies

Passive carry trade strategies (Long or Exit the carry)

• Strategy 1 (continuous-based strategy) : Equally weighted basket of Long high-yield currencies (AUD/ NZD/ SEK/ CAD) Short low-yield currencies (CHF/ JPY)
  
• Strategy 2 (discrete-based strategy) : Using MARS Index be a downside risk filter. Long carry if MARS is in the risk averse area. Exit carry if MARS is in the risk loving area.

Activity carry trade strategies : (Long or Short the carry)

• Using the value of risk filter index to make a Z-score. The Z-score of risk factors could measure risk extremes, and points out the long / short direction.
• Z-score = risk factor's 6-month rolling average / 6-month rolling standard deviation
• In the other words, the Z-score measure how far away (in number of standard deviations) the risk conditions are from the normal levels.
• A large negative Z-score signals benign risk conditions => Long carry
• A large positive Z-score signals severe risk conditions => Short carry
• Strategy 1 : Using the RPI be the risk factor
• Strategy 2 : Using the VIX index be the risk factor
• Strategy 3 : Using the Credit Market Spread be the risk factor
• Strategy 4 : Using the GHI be the risk factor
• Other decision variable, Threshold ( -1, -1.5, -2 )

The results among the strategies:

• Data from 1996-2007
• Annualised average return
• the benchmark (passive carry trade strategies (Long or Exit the carry)) is 5.04%
• Passive : discrete-based strategy is 6.21%
• Activity (Threshold sets at -2) : VIX is 11.78%, Credit Spread is 20.21%, FX Vol is 3.48%
• From max drawdown aspect, Credit Spread is 2.0% far less than Benchmark's 17.9%.
  

II. China Focus, p.22

• In the early 1990s, China hade a dual-exchange rate regime: the PBoC's official rate for all transactions through the banking system, while exporters could trade part of their retained FX earnings among themselves at market-determined rate. As of December 1993, the market rate was 8.80 CNY/USD, and official rate was 5.80.
• 1994 1/1, 正式成立外匯市場, 將官方匯率定在8.70. 稱之為"a managed-floating regime based on market supply and demand" 且匯率是由在China Foreign Exchange Trading Centre (CFETC)內的交易所決定.
• RMB gradually appreciated to 8.27 until Asian financial crisis. While most Asian currencies depreciated in 1998, RMB almost unchanged at 8.27 and stayed at this level for the next 8 years.
• 2005/07/21, "CNY revaluation on 21 July 2005.
• 2005年7月21日起，央行宣佈開始實行以市場供求 (managed-floating regime based on market supply and demand) 為基礎、參考一籃子貨幣(Top trade partners - US, Japan, Euro, South Korea, Singapore, UK, Malaysia, Russia, Australia, Thailand, Canada)進行調節、有管理的浮動匯率制度。(人民幣兌美元的匯率8.11)
• After CNY revaluation on 21 July 2005, today, spot trading of RMB through the CFETC (5%) via 22 commercial banks' OTC market (95%). Every morning, the PBC announces central parity rates based on the weighted average of quotes from the market makers. The RMB is allow to trade in a range of 0.3% on either side of the central parity rate against the USD and 3% against other currencies.