ERX ETF is a powerful hedging tool in the arsenal of an Emerging Markets portfolio manager. ERX as a hedge for an Emerging Markets Energy / Russia portfolio. 0 comments

This article was originally written in April 2011 and edited in mid June 2011. This highly successful strategy returned superior returns during the market sell-off between July 25 and August 8, 2011

ERX has 73% correlation with RTS/Micex. Systemic ERX option trade:

This strategy is a short-term trade that protects a basic Emerging Markets Energy or Russia long/short portfolio during downtrending market, in addition to NDF and CDS hedges.

Why it makes sense:

We have been experiencing a very choppy market in 2011, that is looking for a trend direction. Technically, SPX is oversold, touching 20-indicator that is usually a leading indicator that a market will bounce back short-term. On the other hand, macro uncertainty can easily send SPY (SPX ETF) another 4-5% down - below $130 support level to $125 (200 MA).

Hedging systemic risk of the Russian Stock and Bond market:

Periods of large losses are of systemic nature and are correlated with either SPX, Volatility spikes, Brent and/or combination of other macro sell-off drivers. One could think of a part of the EM portfolio losses as basically a beta-adjusted tail risk between 1 and 2 sigmas on the SPY chart below. Even though we cannot anticipate and hedge the entire risk, we still can hedge a part of the systemic risk that has a high probability of occurring and each time taking a big chunk from P&L. The basic idea is to hedge a part of the larger than average losses occurring during the losing days using plain vanilla ERX put options (or a combination of spread puts, subject to optimization). I think, 3x leveraged ERX could be a good short candidate to hedge EM and Russian market exposure and much more powerful than EEM or any other plain vanilla ETF hedge, (NYSE:A) due to the instrument’s propensity to lose value over time due to negative compounding effect, and (NYSE:B) because ERX is 87% correlated to SPX as well as Crude (74%) and Brent (73%) prices. Shorting ERX is basically employing four hedges simultaneously (oil, SPX, volatility effect, and negative compounding/option price path dependency due to scalar nature of its volatility) due to the nature of the instrument. (“The Dynamics of Leveraged and Inverse ETFs” by Barclays Global Investors, April 8, 2009, free on the web).

ERX vs Vix Volatility Index, Source: Yahoo Finance

a.How much to trade/hedge: If we are to hedge 50% of estimated 2% tail loss of rolling hypothetic $100mln P&L (can be any P&L number), in April-May 2011, we were to hedge avg. $1mln. with ATM PUT, i.e. if hedge amount is $1mln., buy $1mln/put option price $5 (nearest ATM put price, or 2,000 contracts (200,000 shares). The base option hedge for 1 month would be 200,000 * 5 = $1mln., probability adjusted option price would be $2 per share or $400,000 a month (see notes below for math). This number seems high at first, but given its payoff structure, this hedge makes sense when the market is choppy or trending down.

b.P&L: Given positive convexity of ERX long put returns due to the 3X negative compounding effect, this hedge’s reward to risk ratio is almost 2x fold and grows exponentially, i.e. 10% drop in underlying almost doubles the cost of the put option, while 10% increase in underlying makes the option lose only 50% of its value. For example, 3.2% decrease in underlying from $82 to $79.4, cause May 16 ATM jump from $5.6 to $7.6 (35% increase) that in our case scenario would pay off the hedge cost and make $400,000 profit, etc. Given increased volatility during sell-off periods, effect of Theta will be less pronounced. Bloomberg option scenario analysis screen does not take into consideration a substantial volatility jump during sell-off periods that leads to put option price increases higher than model estimates, i.e. if ERX were to fall 10%, higher vols would put the ATM option price higher. In the model portfolio below, the $2mln. hedge returned over $5mln. when ERX fell over 18% over a course of 1 month, as well as our P&L (that’s an ideal case scenario of course.). To model 3X leveraged options, one would need Garch-based Black to account for path-dependency of 3X vols.

c. Payoff and breakeven probability: Breakeven probability = 40% (See the notes for payoff probability extrapolation and other math)

1.Assuming total loss of option value (i.e. expire worthless): around 6% of underlying for 1 month

2. 15% +-ve increase of underlying: 3% of underlying or 50% option value loss

Given 40% breakeven probability, there is a 60% probability that the option lose some of its value. If we assume worst case scenario (100% loss of option value), probability adjusted average option cost would be: $5 – 5*0.6 = $2.

2.Main point: Hedge only takes place when the “market” (our independent variables – SPX, WTI/Brent, Europe Debt, etc.) is trending down. It was obviously not be appropriate to initiate a bearish strategy in 2010 when energy was trending up and ERX returned almost 200%.

3.Triggers for entrance: (A) Buying nearest ATM put on market rally reduces cost and increases risk/reward payoff in the down-trending/choppy market. Therefore dynamic hedge rebalancing will be required following technical signals. -> any time P&L adds 3-5% and market (SPX, Brent, ERX) rallies over 1.5/2%.

4.Triggers for closing of this position: close/reduce position & take profit when (A) there is a systemic market selloff (SPX/Brent/ERX down 2%+) (B) Market Turns to a steady uptrend mode, i.e. 10 day MA above 20 day MA for over a week or longer. This kind of entry/exit strategy would send signals to initiate hedge when we officially got into the choppy market – March 12-13 when MA10 crossed under MA20. The market has been range bound even since - last 3 months. The selloff was also confirmed by RSI that did not follow thru on second SPX peak.

5.Static Hedge: Hedge static notional per “Strategy Cost above” – subject to excessive cost.

6.In the March-May 2011 test period, this kind of strategy could potentially realize approximately $5.4 million, with maximum -$821K drawdown. Maximum capital usage for the 3 month period is $2.3 million.

Sample Period Trade Sequence:

We back-tested our trading strategy on two datasets - January - May 2011 and July 25 - August 8, 2011. The model performed well. For example, in period of January-May 2011, the model was used to hedge notional of $100 mln. hypothetical long/short portfolio of Russian stocks and bonds. For the period, the model, generated $5mln positive HPR, with maximum drawdown of $821K and maximum capital used at $2.3mln. The model generated total of 16 trades - 12 short trades (buying ATM puts) and 4 long exit trades. Graphically:

Chart 1, 2, 3: (1) ERX is a convenient hedging tool (2) Strategy is employed when markets move sideways (3) S&P500 enters into a bearish territory in May-June 2011.

ERX Strategy Cost Example:

On June 7 we buy a July 16’11 put on 1 share of ERX with $69 strike for $6.3. ERX closing price is $68.76.

Trade Breakeven Probability:

1.As of 6/8/2011, ERX closing price was $68.58. June 18 (11 days to expiration) $69 Strike price = $3.35 (avg bid-ask), July 16 (39 days) K=$69, PRICE = $5.90. Therefore, those option prices were worth 4.85% and 8.55% of total strike for June and July 2011 expiration dates, respectively.

2.For June 18 expiration breakeven was @-4.89% and for July 16@ -8%

Underlying ERX ETF volatility:

3.Trade Breakeven: For June 18 option, delta = -48.62%. $ Option Delta = $68.58 Underlying * -48.62% Delta = $33.34

$ Option Premium = - $Delta * % [Underlying should move to break even] =>

Non-path dependent probability of breaking even on this option trade:

% [Underlying should move to break even] / Monthly Underlying Volatility = 4%/10% = 4/10 = 40%, which is close to Bloomberg’s breakeven calculation:

Historic Return/ Breakeven Probabilities:

May 2011 (4./25-5/23):

Avg. Trading Days: 20

Days Underlying Dropped more than 1 sigma: 4 (-5.5% to -8.5% daily) = 25%

Days Dropped over 1%: 3 (-1.5% tp -4.5% daily)

Total Days with significant negative returns: 7 = 7/20 = 35%

May-June 2011 (5/23-6/6):

Avg. Trading Days: 10

Days Underlying Dropped more than 1 sigma: 3 (-4.5% to -6.88% daily) or 3/10 = 30%

March-April 2011 (3/25-4/21)

Avg. Trading Days: 20

Days Underlying Dropped more than 1 sigma: 4 (-3.5% to -8.5% daily) or 4/20 = 25%

Here's the March-April 2011 HRH - Historic Return Chart for ERX:

Correlation Matrix:

To select a "perfect" hedge for an energy-heavy portfolio of Russian bonds (Ruble Corporates, Ruble Goverment - "OFZ", and Eurobonds) and Russian stocks, we have analyzed weekly and daily correlations ("CORR" Bloomberg screen) between major indicators of the Russia-related markets (RTS, Micex, ADR/GDR and Russian Bond market) and portfolio of 200+ instruments that are viewed as possible hedges.

In particular, we have reviewed instruments that ranged from Russia and EM-related CDS, WTI/Brent, EM and Russia-related ETFs, FX hedges, Volatility hedges, etc. We have found ERX to have 73% correlation with Russia's Micex Index, 71.2% with RTS Index, 90% correlation to SPX, 73% correlation to Brent and 75% to Crude, and 50-62% correlation to major Russian OFZ bond prices. Additionally, we think that weekly pricing correlations are a more correct representation of correlation between assets that trade in different time zones and markets. Daily volatility generated too much noise and was less usefull than weekly closing price comparisons.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha
community. Instablog posts are not selected, edited or screened by Seeking Alpha editors,
in contrast to contributors' articles.

Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.

## ERX ETF is a powerful hedging tool in the arsenal of an Emerging Markets portfolio manager. ERX as a hedge for an Emerging Markets Energy / Russia portfolio. 0 comments

This article was originally written in April 2011 and edited in mid June 2011. This highly successful strategy returned superior returns during the market sell-off between July 25 and August 8, 2011ERX has 73% correlation with RTS/Micex. Systemic ERX option trade:Why it makes sense:Hedging systemic risk of the Russian Stock and Bond market:systemic risk that has a high probability of occurringand each time taking a big chunk from P&L. The basic idea is to hedge a part of the larger than average losses occurring during the losing days using plain vanilla ERX put options (or a combination of spread puts, subject to optimization). I think, 3x leveraged ERX could be a good short candidate to hedge EM and Russian market exposure and much more powerful than EEM or any other plain vanilla ETF hedge, (NYSE:A) due to the instrument’s propensity to lose value over time due to negative compounding effect, and (NYSE:B) because ERX is 87% correlated to SPX as well as Crude (74%) and Brent (73%) prices. Shorting ERX is basicallyemploying four hedgessimultaneously (oil, SPX, volatility effect, and negative compounding/option price path dependency due to scalar nature of its volatility) due to the nature of the instrument. (“The Dynamics of Leveraged and Inverse ETFs” by Barclays Global Investors, April 8, 2009, free on the web).ERX vs Vix Volatility Index, Source: Yahoo Finance

If we are to hedge 50% of estimated 2% tail loss of rolling hypothetic $100mln P&L (can be any P&L number), in April-May 2011, we were to hedge avg. $1mln. with ATM PUT, i.e. if hedge amount is $1mln., buy $1mln/put option price $5 (nearest ATM put price, or 2,000 contracts (200,000 shares). The base option hedge for 1 month would be 200,000 * 5 = $1mln., probability adjusted option price would be $2 per share or $400,000 a month (see notes below for math). This number seems high at first, but given its payoff structure, this hedge makes sense when the market is choppy or trending down.How much to trade/hedge:Given positive convexity of ERX long put returns due to the 3X negative compounding effect, this hedge’s reward to risk ratio is almost 2x fold and grows exponentially, i.e. 10% drop in underlying almost doubles the cost of the put option, while 10% increase in underlying makes the option lose only 50% of its value. For example, 3.2% decrease in underlying from $82 to $79.4, cause May 16 ATM jump from $5.6 to $7.6 (35% increase) that in our case scenario would pay off the hedge cost and make $400,000 profit, etc. Given increased volatility during sell-off periods, effect of Theta will be less pronounced. Bloomberg option scenario analysis screen does not take into consideration a substantial volatility jump during sell-off periods that leads to put option price increases higher than model estimates, i.e. if ERX were to fall 10%, higher vols would put the ATM option price higher. In the model portfolio below, the $2mln. hedge returned over $5mln. when ERX fell over 18% over a course of 1 month, as well as our P&L (that’s an ideal case scenario of course.). To model 3X leveraged options, one would need Garch-based Black to account for path-dependency of 3X vols.P&L:Breakeven probability =Payoff and breakeven probability:40%(See the notes for payoff probability extrapolation and other math)around 6% of underlying for 1 month. If we assume worst case scenario (100% loss of option value), probability adjusted average option cost would be: $5 – 5*0.6 = $2.there is a 60% probability that the option lose some of its value(A) Buying nearest ATM put on market rally reduces cost and increases risk/reward payoff in the down-trending/choppy market. Therefore dynamic hedge rebalancing will be required following technical signals. -> any time P&L adds 3-5% and market (SPX, Brent, ERX) rallies over 1.5/2%.Triggers for entrance:close/reduce position & take profit when (A) there is a systemic market selloff (SPX/Brent/ERX down 2%+) (B) Market Turns to a steady uptrend mode, i.e. 10 day MA above 20 day MA for over a week or longer. This kind of entry/exit strategy would send signals to initiate hedge when we officially got into the choppy market – March 12-13 when MA10 crossed under MA20. The market has been range bound even since - last 3 months. The selloff was also confirmed by RSI that did not follow thru on second SPX peak.Triggers for closing of this position:We back-tested our trading strategy on two datasets - January - May 2011 and July 25 - August 8, 2011. The model performed well. For example, in period of January-May 2011, the model was used to hedge notional of $100 mln. hypothetical long/short portfolio of Russian stocks and bonds. For the period, the model, generated $5mln positive HPR, with maximum drawdown of $821K and maximum capital used at $2.3mln. The model generated total of 16 trades - 12 short trades (buying ATM puts) and 4 long exit trades. Graphically:

Chart 1, 2, 3: (1) ERX is a convenient hedging tool (2) Strategy is employed when markets move sideways (3) S&P500 enters into a bearish territory in May-June 2011.

ERX Strategy Cost Example:Trade Breakeven Probability:Trade Breakeven:For June 18 option, delta = -48.62%. $ Option Delta = $68.58 Underlying * -48.62% Delta = $33.34Non-path dependent probability of breaking even on this option trade:% [Underlying should move to break even] / Monthly Underlying Volatility = 4%/10% = 4/10 = 40%, which is close to Bloomberg’s breakeven calculation:Historic Return/ Breakeven Probabilities:Correlation Matrix:In particular, we have reviewed instruments that ranged from Russia and EM-related CDS, WTI/Brent, EM and Russia-related ETFs, FX hedges, Volatility hedges, etc. We have found ERX to have 73% correlation with Russia's Micex Index, 71.2% with RTS Index, 90% correlation to SPX, 73% correlation to Brent and 75% to Crude, and 50-62% correlation to major Russian OFZ bond prices. Additionally, we think that weekly pricing correlations are a more correct representation of correlation between assets that trade in different time zones and markets. Daily volatility generated too much noise and was less usefull than weekly closing price comparisons.

Disclosure:I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.Instablogsare blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.## Share this Instablog with a colleague

Comments(0)