Seeking Alpha

Macro Strategist's  Instablog

Macro Strategist
Send Message
Max Verline is an investment analyst based in New York.
My company:
My blog:
Emerging Markets Macro Strategy
View Macro Strategist's Instablogs on:
  • The Next Big Short Strategy: Rise And Fall Of Fixed Income ETFs

    The Next Big Short Strategy: Rise and Fall of Fixed Income ETFs. Select Fixed Income ETF Performance Forecast

    Part 1: Summary - Fixed Income ETF Slaughterhouse

    The beginning of the new interest rate cycle has created a new wealth creation opportunity. This opportunity is a fixed income ETF short-selling strategy that aims to take advantage of the U.S. government rates policy and structural imbalances inherent in the majority of the fixed income ETFs currently traded on the U.S. market. Given historically low overbought benchmark rates, fixed income ETFs with high level of exposure to the U.S. Treasury will likely lose significant value in the next 24 months. Loss of value with occur along the entire benchmark curve - from short to long duration.

    Despite warning signs, many investors still perceive fixed income ETFs as a safe investment as those ETFs appear to trade like stocks, often times with lower volatility than the market. And most of the retail investors and financial advisors still do not understand the fundamental risk drivers behind the fixed income ETFs price performance and potential dangers of the Fixed Income ETFs as a product class. To demonstrate that things are not what they seem, this article will attempt to show fixed income ETF risks that might make fixed income ETF investing less attractive, especially to retail investors.

    Surface - Ignorance is Bliss

    The ETF business has been one of a few successful attempts to make post-crisis investing appealing to an average retail investor (low risk, diversification, liquidity, "smart beta", etc). ETF conferences, ETF specialty companies and a growing number of new ETF products are all elements of the ETF sales frenzy which is similar to the hype that was created to sell structured-products as "safe investments" back in the day. The financial history tends to repeat itself through a series of booms and busts. Each time Wall Street finds new ways around regulations by presenting potentially hazardous ideas as safe investments. Seven years later the new perfect short candidate has arrived in the form of fixed income ETFs.

    Structure -Fixed Income ETFs: Underperformance by Design

    For the next rates tightening cycle, fixed income ETFs are a poor investment choice, both from an interest rate as well as credit spread perspective.

    The fixed income ETFs were created as directional interest rate bets after the completion of the most recent fed rate tightening that happened in 2004-2006, and, more recently, back in 2009 with a purpose of riding the QE wave. At the time, fixed income portfolio managers could easily deduct that the quantitative easing programs would continue pushing the bond yields to decades' lows. The majority of the long only fixed income ETFs have never been meant to last longer than until the end of the current easing period (end of 2015 - mid 2016).

    Most of the fixed income ETFs are index-based products. Their risk is not actively managed by credit research analysts who are actually familiar with the corporate and sovereign credits that form a bond portfolio.

    Ability to conduct expert research and properly manage bond portfolio credit and interest rate risk will be essential when Fed will start tightening the rates.

    Part 2: Forecast

    Fixed Income ETFs: Forecast

    Many fixed income ETFs might implode when the Fed starts tightening interest rates in the next year or so. There are 3 main bond price decline drivers:

    • Benchmark interest rates will rise. Longer duration U.S. treasury, corporate and emerging markets ETF prices are very sensitive to yield change
    • Liquidity risk - High yield ETFs could lose additional 7-9% of their value due to extremely illiquid market in case of a major bond market sell-off. Many ETFs will come under double selling pressure - first, because of their low daily trading volumes, and secondly, because of the absence of the bids and asks in the secondary bond market. Liquidity risk has become the single most dangerous risk factor in the corporate bond market. Wall Street banks are no longer willing to provide cushion and absorb bond trading losses as an opportunity cost of quickly flipping the bond to another buy-side buyer later on. Given lack of liquidity and prevalence of manual bond execution, sell-side desks are no longer sure how long it will take sell the bond, hence their unwillingness to buy this bond for their own book. Additionally, electronic trading exchanges like MarketAxess and Tradeweb are still relatively small and typically do not have enough liquidity for large size orders. For example, let's imagine an extreme scenario that an investor wants to sell $5 million of bond X. An investor is not able to get a market level bid from a sell-side desk, and trades scraps of the bond away on Tradeweb or MarketAccess at 100bps discount. The bond market immediately shifts to the new equilibrium, the bond price drops again and sell-off spiral continues with the new bond price propagating from small order sizes on MarketAxess and Tradeweb to large manual orders in bond salesmen's Bloomberg chat rooms to dealer runs to autoexecution platforms like Deutsche Bank's Autobahn, etc.
    • Little idiosyncratic value that many ETFs add to the investors' portolios: High correlation of corporate credit ETFs to benchmark rate moves, render those corporate credit ETFs overpriced relative to already overbought U.S. treasury bonds.

    Strategy Summary

    • I recommend taking short positions across fixed income ETFs, including derivative trades where possible - calendar spreads, etc. This short strategy has a good risk/reward probability as its payout is dependent on the U.S. Fed rates policy.
    • Sample Portfolio 1: U.S. Treasury-based ETFs: There is a total of 62 U.S.-listed U.S.Treasury ETFs with assets around $45 billion. Out of those, I recommend shorting the largest and the most liquid ETFs that also have full options chains available for trading: (click to enlarge)

    I also recommend shorting leveraged Treasury ETFs because of their natural decay / beta-slippage factor. Calendar puts on ultrashort leveraged ETFs could also represent a good risk/reward opportunity.

    • Sample Portfolio 2: U.S. Corporate Credit ETFs: U.S.-listed U.S. Credit-focused ETFs are a larger segment of the U.S. ETF market. There is a total of 74 ETFs with over $99.5 billion in assets. The majority of the U.S. Corporate Credit ETFs are index market-cap products - 57 tickers with $97.5 billion in assets. There is one fundamental ETFs with $569 million and 2 actively managed ETFs that share $718 million. There are 6 specialty ETFs that share $771 million. Prevalence of the market-cap index ETFs simplifies short portfolio selection strategy as I do not have to spend time analyzing benefits of the actively managed ETF strategies. For corporate credit, I recommend taking short positions on the top 20 U.S. Corporate Credit ETFs by assets. Together these top 20 ETFs manage $89.9 billion in U.S. Corporate High Grade and High Yield Credit, or 90% of total category assets. High concentration of assets in top 20 tickers provides for additional liquidity and opportunity to take out sizable short positions (click to enlarge):
    • Corporate Credit ETFs and concept of "Credit Alpha" - Particular attention should be paid to the fixed income ETFs that have produced very little credit alpha in the last five years, the Fed Fund easing period. The rationale here is the following: Given near zero Fed Fund rate policy, the last six years (2009-2014) have in general been extremely positive for the fixed income market, for the U.S. high yield and Emerging Markets credit in particular. The obvious exception here would be summer 2013 turbulence in emerging markets credit. The fixed income ETFs with very little credit alpha are those whose annualized CAGR is, to a large extent, driven by the performance of their benchmark treasury bond of the corresponding duration. For example, if fixed income ETF's 5 year CAGR is 10% and the corresponding benchmark U.S. Treasury bond is 9%, the credit alpha is only 1% (that is unadjusted for yield volatility / risk). That 1% would imply that a corporate credit ETF with very little credit alpha is overpriced relative to similar duration Treasury: only 10% of the ETF return can be attributed to asset picking talents / investment strategy of an ETF manager, and 90% of return to general improvement in the market. Hence, an improvement in credit alpha can be driven by multiple qualitative and quantitative factors, such as ETF structuring manager's ability to create a profitable bond selection/filtering strategy, picking select promising credits during rebalancing periods, or, for example, getting extra yield from appreciation of local currencies in case of local currency bond ETFs.
    • Credit Alpha can be either an indication of manager's talent of developing a successful investment strategy or creating a smart bond index ETF methodology, or pure market-driven luck, i.e. a passive corporate credit ETF where a big chunk of return has been driven by a credit spread compression regardless of direction of the benchmark rates. Thefore Credit Alpha does not measure how smart (or mediocre) ETF managers are. Presence of a large Credit Alpha simply shows that a particular ETF might have a better relative performance during the fed fund rate tightening period
    • Credit Alpha is a much better indicator of positive idiosynchratic risk / alpha presence in the Fixed Income ETFs compared to traditional ratios such as Sharpe Ratio. For example, Bloomberg calculates Sharpe Ratio as a measure of ETF performance as difference of annualized mean return and 3 month risk free rate divided by annualized standard deviation ofr returns. Understandably, bonds with the higher Sharpe ratio would be those that also have had very high level of correlation to the U.S. Treasury since 2009. Therefore, for fixed income ETF analysis, the Sharpe ratio does not represent an objective indicator of "historical risk adjusted performance" as the Sharpe ratio does not calculate the risk adjusted performance of the ETF's idiosynchratic risk component (that's because the return of the interest rate benchmark is included into the Sharpe ratio calculation as a part of the ETF's return)
    • You get what you pay for: Extending the observation further, a prudent investor would then question sanity of why he or she would pay additional 0.50% in ETF manager fees for such a low return, as zero or low credit alpha ETFs would already be relatively more expensive compared to similar duration Treasury
    • While shorting fixed income ETFs with little credit alpha is a simple well-defined strategy (see the above bullet for definition), a more elaborate investment approach should be taken when analyzing corporate credit ETFs with better quality credit alpha, where a larger part of the returns is driven by underlying bonds' credit spread performance. This performance could be a result of a better portfolio rebalancing effort, index composition, credit market conditions (i.e. outperforming HY spread compression, for example), etc. In case of these ETFs, I will need to estimate U.S. HY, U.S. HG and EM credit spread performance for the next tightening phase. I will then use historical analysis to estimate returns of those in my base case scenario

    Estimated Price Impact of the Fed Interest Rate Increase on Major Fixed Income ETF Prices: Our Strategy

    Our forecast of the impact of the Fed Rate increase on the fixed income ETFs is determined by the three factors:

    • Estimating the scale and velocity of the Fed Fund rate increase
    • Projecting correlations, timeline and scale of the U.S. Treasury bond yield changes given the Fed Fund rate tightening
    • Determining correlations of the Fixed Income ETFs to the corresponding benchmark interest rates. Performing proper attribution analysis and return forecast, i.e. separating the fixed income ETFs performance into the benchmark interest rate component and credit spread, its idiosyncratic risk component
    • I am also mindful of the fact that U.S. Treasury action is in part driven by its spread to the sovereign debt of developed nations

    Let's consider the most recent 2004-2006 Fed rate tightening cycle as our base case scenario. Our methodology is as follows:

    • I assume that the U.S. Treasury Yields will rise to the Fed-tightening levels of 2005-2006:

    A. Many strategists and hedge fund managers expect that the long-term rates in 2015-2017 will be significantly lower the 2004-2006 levels thanks to the inflow of external funds into the U.S.. Supporters of the low rate view typically cite such reasons as the doom and gloom situation in Europe, necessity to further prop up the U.S. economy, deflationary fears in the U.S., etc. My position is contrarian. I think that the effects of even small Fed Fund increases will be very pronounced and exacerbated by lack of liqudity and rising volatility effects

    B. The overbought short-end of the U.S. Treasury curve will experience a precipitous shift upward following the FFR (Fed Fund Rate) increases

    C. While the longer end of the U.S. Treasury curve, less susceptible to the short-term rate increases, will nevertheless revert back to more sustainable recent historical levels

    • Given the current shape and level of the U.S. Treasury curve, we forecast that both short-end and long-end of the U.S. treasury curve will widen with biggest sell-off happening in the first year of the fed fund rate increases
    • The Treasury Bond price decrease on a particular Fed Fund Rate date is estimated as (click to enlarge)
    • (where Yield Change is the U.S. Treasury Bond's current yield minus the base period U.S. Treasury bond yield around the dates of the 2004-2006 announcement FFR rate increases)
    • The FFR tightening policy price impact on the Fixed Income ETF of a corresponding duration is then estimated by multiplying the U.S. Treasury bond estimated price change by the Fixed Income ETF's CAGR performance correlation to that particular U.S. Treasury bond. It is easy to see that Fixed Income ETFs with little credit alpha are those most susceptible to the FFR rate increases
    • Additionally, by design, fixed income ETFs can be sold short which makes these products more volatile and tend to produce lower risk-adjusted returns compared to bond funds for example. The higher volatility of the bond ETFs make them attractive for implied volatility trading strategies. Additionally, given scarce liquidity during bond market sell-offs, fixed income ETFs will likely trade substantially below their NAV as market making desks would be slow to bring ETFs back to NAV, thus enhancing short-selling returns

    Federal Reserve Monetary Cycles (2004-2006, 2007-2014)

    Please refer to the below diagrams for Fed Fund rate changes and their impact on the U.S. Treasury curve yields (click to enlarge):

    Click to see the extended version:

    Forecast - US Treasury Target Yield Levels, 2015-2017

    Please click to see our U.S. Treasury curve yield forecast for the 2015-2017 FFR tightening phase (click to enlarge):

    Extended version:

    Part 3: Select Fixed Income ETF Analysis

    Sample Skeletons in the Closet

    Here's a list of select 2015-2016 fixed income short opportunities: HYG, JNK, LQD, CIU, VCIT, VCLT, EMB, EMLC, PCY, ELD, LEMB, PHB, BSJF, HYLD, HYEM

    Estimated Price Impact of the Fed Interest Rate Increase on Major Fixed Income ETF Prices: Our Short Picks (click to enlarge):

    (click to enlarge)

    The above ETFs represent a good snapshot of the U.S. listed fixed income ETF types. We have selected the 3 non-U.S. Treasury fixed income ETF investment themes - total of 15 ETFs:

    • USD-denominated U.S. High Yield corporate bonds (Below BBB- rating)
    • USD-denominated U.S. High Grade corporate bonds (BBB- and higher)
    • Local Currency Emerging Markets Sovereign

    The selected ETFs are also ones of the largest in terms of their assets and trading volume. Our rationale here was that analysis of the largest ETFs would be more objective compared to smaller ETFs, as the smaller ETFs' performance is often affected by lack of liquidity and other non-market factors.

    The Select ETFs Performance Forecast

    The performance of the above 15 ETFs was then analyzed under the baseline 2015-2017 FFR Tightening / U.S. Treasury curve scenario. According to the baseline analysis, the best ETF performers, from the credit alpha perspective, will be the ETFs that either track or actively manage portfolios of the USD-denominated U.S. High Yield Corporate bonds. The top performing HY Corp ETFs - HYG and JNK have ~ 4.4 year duration and have returned average of 7.9% annually since 2010, beating their benchmark treasury performance by 4.3% annually. The benchmark treasury is represented by IEI 3-7 year Treasury Index ETF.

    We favor BlackRock's HYG ETF that is designed ans sampled after the USD High Yield Corporate Index (Markit iBoxx USD Liquid HY Index). While the HYG index product charges a higher a fee of 0.50% vs. JNK's 0.40%, HYG has a lower correlation of its CAGR return to the benchmark yield (HYG 45% vs. JNK's 47%). Please refer to ETF's performance chart for detailed analysis.

    A good quality credit alpha of HYG and JNK can be explained by the overall high yield credit spread compression in the last 5 years due to the easy financing climate in the U.S. - ability of the high yield issuers to sell large amounts of the high yield debt to investors searching for extra yield. This observation is also confirmed by that fact that each of these index ETFs has a very large number of bonds in its portfolio (986 for HYG and 748 for JNK). Fixed income indices in general are created to reflect general market returns using market-weighted or fundamental index construction approaches.

    Near the bottom of our list are the U.S. Investment Grade Corporate bond ETFs that are primarily represented by ETF index products - VCLT and LQD ETFs that mirror long duration Barclays U.S. 10+ Year Corporate Index and Markit iBoxx USD Liquid Investment Grade Index, respectively. The Index ETFs are those that mirror performance of the Bond Indices via various sampling techniques. As of October 16, 2014, VCLT had 1,379 and LQD had 1,231 U.S. high grade corporate bonds in their portfolios. These are truly large numbers of bonds which imply that these ETFs will closely reflect performance of the total US High Grade segment.

    Both VCLT and LQD credit alpha is quite low 0.9% and 1%, respectively. The main reason for that is VCLT and LQD's very high correlation to the returns of their benchmark treasury securities that stand at 85% and 87%, respectively. Given low credit risk, and hence credit spread, the U.S. Investment Grade bond yields and returns tend to have a very high correlation to their benchmarks. Given their high correlation to benchmarks, VCLT and LQD have returned impressive CAGRs of 9.2% and 6.7%, and total period returns of 55% and 38% for period of 2010-2014, respectively. VCLT's higher returns were driven by its longer duration of 13.6 years vs. 7.9 years for LQD.

    High correlation of returns to U.S. Treasury, makes VCLT and LQD vulnerable to the new tightening cycle. We project that VCLT will lose around 20% of its value and LQD will lose around 14% of its value once the Fed will start tightening. These numbers can be either higher or lower depending on the ETFs' credit spreads' performance.

    On the bottom of our list are underperforming ETFs - EMLC, LEMB and ELD. EMLC and LEMB are index-based ETFs managed by Van Eck and BlackRock, respectively. ELD is managed by WisdomTree. All three ETFs invest in local currency emerging markets sovereign bonds (Local Currency EM Sov). The rationale of investing in local currency bonds is to realize extra credit alpha due to improvement in an emerging economy's credit and appreciating of its currency. A good example of a local currency sovereign bond would be OFZs, Ruble-denominated Russian government bonds that at a time enjoyed spectacular post-crisis performance given economic growth in Russia, rising oil exports and strengthening local currency. Most recently the emerging markets credit has been a turbulent place to invest funds and many investors have pulled their money out of the EM credit funds and EM credit focused ETFs which resulted is smaller funds under management. Currently, LEMB as $600 million and ELD has $800 million capitalization. We project further market capitalization reduction for this ETF class.

    Our top short picks are Local Currency Emerging Markets ETFs - index-based EMLC and LEMB and actively managed ELD. EMLC, LEMB and ELD belong to a relatively small niche part of the fixed income ETF universe - Emerging Markets Fixed Income. The total amount of assets in this ETF category stood at USD$24 billion as of October 28, 2014:

    Our short Emerging Markets Fixed Income ETF picks (EMLC, LEMB, and ELD) have underperformed the markets and USD-denominated U.S. corporate ETFs significantly, both on a relative and absolute value basis. The underperformance of EMLC, LEMB and ELD is reflected by their negative 5 year credit alphas of -2.8%, -3% and 3.3%, respectively. On a relative basis, EMLC, LEMB and ELD CAGR is lower than the CAGR of the benchmark U.S. Treasurys of similar duration which is reflected in their negative credit alpha. The negative credit alpha means that the returns of these ETFs have been negatively affected by both widening sovereign yields as well as depreciation of local currencies, while most of the appreciation value of these ETFs was due to the declining benchmark yields of the last 6 years. On an absolute basis, LEMB and ELD have returned 2% and 5% annually since 2012 and 2010, respectively.

    WisdomTree's ELD ETF has experienced significant asset outflow in 2014, losing $408 million, or 34.4% of its total assets in the last 10 months. This is a very significant outflow compared to iShares' LEMB losing $18 million and Market Vectors EMLC adding $45 million in assets in the same time period.

    Additionally, ELD ETF is a fairly illiquid product given amount of its assets. On average ELD trades 133,000 shares a day which is rather a low number for $780 million ETF. Typically, low volume ETFs and stocks tend to be more volatile. The less liquid ETFs could potentially perform worse than more liquid ETFs in a distress, because of additional selling pressure on those ETF by short-sellers as well as real money funds and advisors reducing their exposure. Once selling begins, more predator short-sellers start feeding on an illiquid ETF, thus prompting real-money and theme hedge funds to continue exiting this particular ETF. For example, $394 million or 52% of ELD ETF is owned by top 10% holders. Many large ETF holders tend to move in and out of their positions by selling large amount of shares. According to Bloomberg's ETF Holders' October 29, 2014 screen, Creative Planning reduced its ELD holdings by 608,385 shares (3.6% of total ELD assets), while Envestnet increased ELD by 282,914 shares (1.7%) and United Capital by 276,935 shares (1.6%). That shows that 7% of the $800 million ELD ETF was recently rebalanced under fairly illiquid conditions*:

    For example, if top 10 ELD shareholders decided to liquidate 10% of their ELD holdings, that could cause additional pricing pressure given additional selling volume and price imbalances. if ELD holders decided to sell 10% of ELD or $40 million in 5 days, that would add additional $7.7 million in daily sell orders, or additional 172k shares to daily 133K volume:

    To sum it up, we project that EMLC, LEMB and ELD underperformance will continue. We are particularly concerned with ELD's asset outflows that in our opinion demonstrate investors' uncertainty about effectiveness of ELD ETF's actively managed strategy. ELD's liquidity is also a concern.

    We forecast that these three ETFs (EMLC, LEMB and ELD) could lose between 9.5% and 13% of their value given expected decline in value of the US Treasury in the first year of Fed tightening. Additional losses in EMLC, LEMB and ELD might be further exacerbated by depreciation of the emerging markets currencies, rising sovereign yields, and reduction of EMLC, LEMB and ELD assets.

    Part 4: Fixed Income ETF Industry Risk Factors

    Liquidity and ETF Size

    ETFs are now the second biggest holders of the U.S. and foreign bonds and by far the biggest risk factor. Let's imagine that investors have decided to offload $6bln or 10% of HYG ($12bln), JNK ($9bln) and LQD ($17bln), VCIT ($4bln), EMB ($4.4bln), PCY ($2.3bln) and ELD ($0.8bln). Together these select ETFs hold $60bln of corporate and sovereign bonds in their underlying portfolios. A steep bond market sell-off would cause sell-side banks and broker dealers scramble for liquidity. The example above shows that a mini paralysis of the corporate bond market liquidity could be possible by a forced sell-off of just 7 large fixed income ETFs. There are many more fixed income ETFs out there, so the fat tail event and a potential liquidity crunch could be possible even in case of the much milder fixed income ETF sell-off. As of October 28, 2014, the total number of fixed income ETFs stood at 968 with USD$468 billion in assets. 88% of assets were allocated to market-cap weighted index ETFs, 3% to actively managed ETFs with proprietary strategies, and 10% to other types of ETFs.

    Despite a great number of different fixed income ETFs, the fixed income ETF allocation is highly concentrated and U.S. product centric:

    • Top 10 U.S. ETFs by market-cap accounted for 32% of world's fixed income ETF assets
    • Top 20 U.S. ETFs by market-cap accounted for 44% of world's fixed income ETF assets
    • U.S. traded fixed income ETFs accounted for 66% of world's ETFs' assets (click to enlarge):
    • The U.S.-listed and U.S.-focused fixed income ETFs represent 60% of all fixed income ETFs with assets over $279bln. Out of those U.S. Treasury ETFs have around $45 billion of assets and U.S. corporates-focused ETFs have around $100 billion in assets. Together U.S. Treasury ETFs and U.S. Corporate Credit ETFs represent two great shorting opportunities (click to enlarge):

    It is also worth to mention that in case of a major bond market-selloff it might take weeks to sell ETFs' underlying assets - illiquid and high yield bonds, and even investment grade bonds. The gap between the ETF price and underlying assets will put additional pressure on the market.

    Corporate Bond Liquidity and ETFs: Absence of Lockup Periods Drives Volatility

    Just like bond funds, fixed income ETFs have loaded up on the corporate and sovereign bonds that are hard to sell. If a client wants to withdraw the money ("redemption"), a mutual fund manager often times needs to sell bonds to satisfy redemption request, or they need to keep a cash reserve. In the ETF case, an investor will simply sell ETF, like a stock. A mass exit from ETF positions creates an enormous stress on the ETF administrators by forcing them to reduce their underlying portfolios at the time of market sell-off - something they have to do. In case of index ETFs, this will further exacerbate market imbalances and liquidity, as the trading desk would have to offload hundreds of bonds to satisfy rebalancing procedure. The consequences can be very severe.

    Corporate Bond Inventory on Wall Street: Risk Factors

    According to Bloomberg, the Wall Street banks have around $6 billion of corporate bonds on their books. The banks call those bonds an inventory. This "inventory" is typically managed by a sell-side trader. Officially, banks do not have a mandate to speculate and prop trade using the bond "inventory" intended for resale. On the other hand, banks can't put up meaningful inventory hedges as this would mean taking profit out of their own pocket - remember, in theory, they do not plan on keeping these bonds on the books for too long. In times of bond market sell-offs, banks and other sell-side institutions could hit hard as they still do not have a meaningful risk strategy of what to do with those trading books, besides selling those bonds.

    Part 5: Idiosyncratic Risk Analysis

    Corporate High Yield Spreads: Sideways Action

    We are neutral on the U.S. corporate credit spreads because of the extremely limited spread compression upside as well as limited downside thanks to the expanding U.S. economy. The same observation is also true for the non-U.S. credit markets that will follow the U.S. closely.

    In our analysis, we used Markit CDX North America High Yield Index credit default swap contract as a proxy for corporate credit spread risk. This CDX contract comprises 100 BB/B-rated U.S. corporate bonds and typically has duration of about 4.5 years.


    The NA HY CDX contract traded at 346 bps as of October 28 2014. The contract traded at similar level (350sbps) in the end of December 05 / January 06 (FFR@4.25). By March 2006, despite the continuing Fed tightening, NA HY CDX compressed impressive 100bps to 251 bps, and reached all-time low of 184bps on February 21 2007. Starting Spring 2007, NA HY CDX started to widen diverging from the equity markets that were still rising. Finally, the credit crunch pushed to spread to all time high of 1933 bps on March 9, 2009.


    Historically low cost of funding helps high yield issuers in many ways - from getting cheap financing to reducing probability of not paying the coupon to investors. Even when Fed raises rates, high yield issuers will still be able to sell bonds at historically low yields as long as the U.S. economy and macroeconomic factors will stay at current levels or better. We are neutral on the U.S. credit spread compression both for high yield and high grade. Credit spread compression is a much bigger source of idiosyncratic risk and alpha in high yield. However, at the current spread levels, the high yield spread compression upside seems to be very limited for majority of the high bonds. Here's a very good chart in the research paper called "The Case For High-Yield Corporate Bonds Investing" by Guggenheim Investments (published June 30 2014) . The chart shows the OAS spread between the Barclays U.S. Corporate High Yield Index (around ~4.5 year duration) and comparable duration Treasury (click to enlarge):

    The current historically low credit spreads actually prompt money managers to buy longer duration and less liquid bonds in the portfolio, which in turn prepares fertile ground for a new fat tail event. Widening of the credit spreads will largely depend on the ability of the U.S. economy to continue expanding, i.e. if economy does not expand and benchmark rates are already high, the rising probabilities of default will prompt credit spreads to widen. Longer term this will probably be the case as the current high yield default rate of slightly above two percent is less than a half of historic average of 4.66% between 1983 and 2014.

    Markit NA HY CDX Historical Spread (click to enlarge):

    Markit CDX NA HY vs. U.S. 10 year note (click to enlarge):

    Markit CDX NA HY vs. UST 10 year note vs. SPX Index vs. EURUSD:

    Part 6: Some Additional Stylized Facts about Fixed Income ETF Methodology

    • There are currently two ETF construction methodologies: #1 passive / index based with monthly or quarterly rebalancing and #2 active management where ETF portfolio managers use their proprietary strategies to invest in a particular fixed income product category. The actively managed ETFs have never become the mainstream product. Globally, only 4% (42) out of total 968 fixed income ETFs are classified as proprietary / actively managed according to Bloomberg. Currently, ETF managers are supposed to disclose their actively managed holdings daily which is not appealing to most of them. To solve this problem, there have recently been attempts by mutual funds to get approval to create listed / exchange-traded non-transparent active ETFs based on the mutual funds's propritetary portfolio strategies. Instead of a full daily disclosure, a non-transparent ETF manager would provide investors and market makers with a jack in the box - a proxy portfolio to approximate an ETF nav. Even if these products do get approved, this initiave will likely fail to get widespread appeal as majority of the mutual funds would not be willing to disclose their proprietary fixed income strategies which in turn will make it rather complicated for market makers to manage nav on non-transparent active ETFs and for institutional investors to invest in the non-transparent jack in the box ETFs. For extensive write-up on this topic, please refer to "Blow to asset managers in ETF arena", October 30, 2014, Financial Times,
    • The index-based/passive strategies can be divided into market-cap and fundamental allocation approaches. To create and rebalance these types of ETFs, ETF managers use filters to attempt to qualitatively estimate expected returns of the fixed income instruments. These filters are used in addition to the sampling techniques that help portfolio managers replicate certain bond index in terms of duration and other factors by investing in a smaller number of bonds
    • Human Factor: ETF Fees and Research Talent. It could be the case, that successful bond portfolio managers are simply not interested in the low fees of 0.50% that ETF administrators typically charge. It would be fair to assume that it is a rather complicated matter for an ETF administrator to hire a truly talented bond fund manager to run an actively managed ETF and still be able to charge 0.50% fee. ETF administrators also incur bond bid ask spread expenses each time the underlying portfolio needs to be rebalanced.
    • Law of Large Numbers kills smart beta. Given portfolio size/diversification, it is statistically impossible that an ETF manager would know well every single bond and issuer that is a part of the ETF portfolio. Hence, the entire ETF "smart beta" filtering process is futile and statistically insignificant unless the number of ETF portfolio constituents could be decreased significantly
    • A typical passive ETF construction methodology calls for sorting bonds into quantiles [for further portfolio allocation] by corporate finance and other performance indicators, an approach similar to the filtration process used in creation of the equity index ETFs. These filters or their variation are applied in the process of portfolio allocation when using the two most popular methodologies - market-weighted and fundamental. The market-weighted approach allocates bonds based on their market cap, hence companies with more debt get more weight in the portfolio. The fundamental allocation approach ranks bonds based on their corporate finance parameters, such as cash flow, sales, and book value-based allocation weights. This methodology is called Fundamental Index methodology

    Part 7: Active Credit ETF Management in The New Rates Cycle

    • It is never enough "risk management" in order to get maximum alpha
    • Dynamically manage benchmark rates risk
    • Take a proactive quantitative approach to fixed income ETF to ensure maximum return
    • The allocation process must include optimization and rebalancing of the ETF portfolio's duration/yield curve as well as ensuring proper convexity of the components
    • Abandon traditional fixed income indices once highly volatile interest rate markets arrive
    • Now, that we have established that a credit ETF is not a pure investment product (given its marginal added value) but rather a trading and a theme product, our recommendation is that an ETF manager concentrate on optimizing a credit ETF return characteristics and its liquidity leaving it up to the market to decide how to utilize this product class
    • More importantly, fixed income ETF investors must do their homework to monitor which sectors of economy and issuers are doing well in the current rate cycle

    This leads us to the final question that this article is trying to answer - knowing all the fixed income and macro idiosychrasies, which bonds and sectors should comprise a high quality credit ETF that has a good potential for idiosychratic risk improvement that would result in credit risk compression regardless of the move in the benchmark interest rates?

    History does not necessarily repeat when Fed tightens the rates. Historically, different sectors ended up as winners and losers across each monetary tightening and easing cycle. Often times, historical intermarket and intersector relationships do not mean-revert as it would be expected in the beginning of a new secular interest rate cycle. That is why the comparable sector and issuer allocation strategies based on their spread performance in the prior cycle will typically underperform the market.

    The below sectors performed relatively well during the previous tightening cycles:

    • Technology
    • Healthcare
    • Finance
    • Basic Materials
    • Emerging Markets

    So there are quite a few investment options and there are plenty of the Machiavellian reasons to justify why one sector would outperform the other. Banks - due to the rising interest charges, Materials - price growth driven by inflation, etc.

    What are the reasons that there are no set rules why some sectors perform better than others during the tightening cycle? One explanation could be the ever increasing pace of the technological innovation that has been affecting and driving economic cycles on an unprecented scale. Companies and governments compete for the future. Every significant economic breakthrough in the U.S. has been driven by innovation and creative iterative use of the new technology. For example, the most recent periods of the U.S. economic growth have been fueled by the effective use of Internet and social media technologies (invented long time ago).
    Still, for an average investor back in 2004-2006, it was rather a challenge to imagine the new economy that has been created just a few years later. Could we imagine that WhatsAp, Instagram, Groupon, Alibaba and other new generation internet firms would have so much weight in the new economy? Things change and evolve. What does the future hold and how will technologies affect the rate cycle? Instant language translation, humanoid robotics, bio and nano technologies could all be those new things.

    Source: Federal Reserve Bank

    Part 8: Additional Sample Fixed Income ETF Short Portfolios

    Sample Portfolio 1: U.S. IG Long Term ETFs

    This is a sample list of large, small and micro caps

    (click to enlarge)

    Sample Portfolio 2

    This is a list of large, liquid global ETFs that will likely be affected by Fed's rate tightening policy (click to enlarge)

    References: Bloomberg, Reuters, Barclays, Guggenheim Investments, Invesco, Federal Reserve, Treasury

    * Please note that according to Wisdom Tree's website the total number of outstanding ELD ETF shares is 16,800,000. ( We calculated the current number of ELD shares based on Bloomberg's ELD shareholder information and current ETF prices, so there is a small basis discrepancy of 879 shares between WisdomTree and our calculations

    Oct 15 12:57 AM | Link | Comment!
  • Commodity Strategy: Model Trade of the Week Update [Short Euribor and German Bonds vs Long U.S. Equity Futures]


    Trade Open: August 12, 2011
    Trade Close: August 15, 2011

    Net absolute dv01 (if all positions were at profit) = -$7.3k, since they are not perfect long/short combinations.

    Trade parameters:

    $280k margin cash

    $29 mln notional + fx trade

    Stop and limit orders, alternatively OCO (one cancels/initiate another) to cut losses quicker.

    3 Day P&L: $50K

    Cash Margin Used: $280K

    3 Day Net Return: 18%

    Strategy Description – Market (August 12) is set to stabilize today:



    ·         Short Euribor 3 month future – expectation that liquidity will improve

    ·         Short USD, Long Euro – on receding volatility, U.S. is relatively better than Europe

    ·         Short Euro Schatz futures (Here’s thinking that European markets should decrease appetite for short-duration (1.75-2.25 years) German Sovereign debt on positive equity markets in Europe

    ·         Short 10 year UST



    ·         Long SPX and Dow Jones Spet 11 futures (U.S. is relatively better than Europe)


    Out of these bets, 10 year UST lost some cash as a directional “short on top” bet but still is an effective hedge. We still expect that USTs should sell off after record high last week.

    Strategy P&L:
    Long Eur/Short USD @1.4268 Close @1.4300, $10 mln notional, @$32K profit
    Long Emini SPX @1,174 Close @1,184, 10 Contracts, @$5K profit
    Long Dow Jones Emini @11,143
    Close@11,314, 10 contracts, @$8k profit

    Short Euribor 3 Mo @98.76 Close @98.715, 100 contracts @$14k profit
    Short Euro Schatz  @109.33 Close @109.345, 100 contracts @-$4 loss
    Short 10 year UST @129'11 Close @129'16, 30 contracts @-$5k loss



    Emini SPX 59K notional x 10 contracts = $590k notional

    Euribor EUR246K x 100 = EUR24  mln notional = $16.6 mln

    Euro Schatz Bonds Eur109K c 100 contracts = eur 10.9 mln notional = $7.6 mln.

    Dow Jones future $56k x 10 =contracts = $560K notional

    10 YR UST $129.5K x 30 = $3.9mln notional

    Total: $29 mln notional


    Aug 16 7:40 AM | Link | Comment!
  • Commodity Strategy Update: Trade of the Day on U.S. Market Open [Long CHF and 10 year Treasury]
    August 10, 2011 - Trade of the Day:
    10AM UPDATE - Analysis proved correct. Market is 440 bps down. Model portfolio returned healthy 10% leveraged P&L.

    Notional weighted-average trade:
    Long CHF / Sell weighted basket of AUD, JPY, SEK, TRY, USD, and ZAR.
    Long 10 Year U.S. Treasury
    Short S&P 500 E-mini.
    Short Euro / Long USD
    Short Ruble / Long USD

    Enter 8:00 AM ET - Exit 10:30AM ET, August 10, 2011

    8AM EST
    From technical analysis point of view:
    Markets move in waves + flight to safe haven in absence of CHF direct intervention. Looks like market is capable of absorbing additional CHF liquidity rather quickly. CHF basket in combination with long 2/10 is a good high prob intraday trade this morning on S&P emini already down 2.3%  intraday on profit taking and Uk growth news.
    From fundamental point of view:
    Euro markets did not follow through as high as it was expected – hence initial nervous move into USD, CHF and gold, out of WTI/Brent . Along with CHF, 10 year is following emini intraday, up 0.30% (As well as Short Eur/Long USD @1.43617 for 50bps +ve p&l) .  Btw, OIS spread on eurolibor is highest in 2 years.

    Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
    Aug 10 10:27 AM | Link | Comment!
Full index of posts »
Latest Followers


More »
Posts by Themes
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.