market folly

The following is a Guest post on MarketFolly.com.

'tradefast' is the nickname of an independent equity trader who has more than 20 years of market experience at a major financial institution and 2 hedge funds. He now trades for a private investment fund, using a combination of both fundamentals and technicals. (He's our kinda guy).

Last week, he sat down to explain how contango affects the crude oil ETF's and ETN's that many investors and traders usually play, including USO, OIL, & DBO. This next piece is a follow-up post to that topic. So, before beginning, make sure you check out: How Contango Affects Crude Oil ETFs & ETNs.

Next, he takes a look at how to play crude oil using those same ETFs & ETNs. He writes,

Objective

This article provides some straightforward insight as to how a retail trader/investor can implement a directional play on the price of crude oil. Included is a discussion of the manner in which the forward market for crude oil can cause crude oil ETF returns to deviate from spot market returns. This article is not intended to be authoritative, comprehensive, or highly technical. It is simply a compilation of previous discussions on the topic (with some added elbow grease and my version of common sense). Readers should be aware that much of the material in this article has been discussed previously here and here. Certain elements of this article are pulled directly from these sources.

Note

When I first wrote about the effects of forward curves on crude oil ETFs, the crude market was in steep contango and the discussion attracted widespread attention. More recently, the crude curve has flattened somewhat and it may appear to some individuals that the curve shape has become less of an issue to retail speculators. I have a different view. I believe that sharp volatility in the shape of the curve makes it imperative that retail crude speculators understand how curve adjustments can affect their positions. This topic is not dying, but rather garnering added importance.

Introduction - Retail Investors Cannot Trade Spot Crude Oil

I have a friend with fairly extensive stock trading experience who generates most of his technical market analysis using the S&P 500 futures contract. When it comes time to transact, however, he will swing over to the cash market and trade an ETF such as SPY (S&P 500) or a leveraged ETF such as SSO (2x the S&P 500) or SDS (2x the inverse of the S&P 500 - a double short). Although he and I share the same trading objectives (to capture a directional movement in the S&P 500), I choose to generate my technical analysis using the precise instrument which I expect to trade. In my case, SPY (or SSO or SDS if I desire leverage). I am confident in my approach because I know the instrument I am trading exhibits an extremely tight relationship to movement in the S&P 500 cash market. Simply put, I trade SPY because is a highly dependable proxy for replicating the spot market of the S&P 500.

Unfortunately, traders or investors wishing to implement a directional play on crude oil lack access to a tradeable instrument which tracks spot crude oil in the same manner that SPY tracks the S&P 500. Instead, we must choose from an array of instruments which are structured with the intent of tracking crude prices, but with flaws relating to the fact that crude oil (unlike the S&P 500) is a physical commodity for which spot trading is limited to those who can transport, store, or produce crude oil.

Crude Oil Futures – The Curve

Fortunately, despite our lack of access to the crude oil spot market, there is a highly liquid market for futures contracts which reference crude oil. And, the price of these contracts exhibits volatility which generally resembles the movements in the spot market. Many traders transact directly in crude oil futures contracts, electronically or in the futures pit. These traders rely on the fact that they will never have to physically handle the commodity itself. They can simply close out their futures positions prior to expiration and net out the difference between their entry and exit price.

The fact is that trading in crude oil futures (which expire monthly) is spread out over several months, even years. And, the price of crude for delivery at future expiration often varies substantially from month to month. At times, prices in future delivery months are progressively higher than in the nearest delivery month (contango). And, more often than not, the opposite is true (backwardation). Many futures traders have specialized knowledge of the day-to-day shifts in supply and demand fundamentals and they are comfortable projecting price movements at specific points along the forward curve. The typical retail investors probably lacks the skill-set needed to profitably exploit forward curves in a sophisticated manner.

Introducing Crude Oil ETFs/ETNs

For the retail crude oil speculator who is incapable of trading crude oil futures, there are tradeable ETFs (Exchange Traded Funds) and ETNs (Exchange Traded Notes) which employ futures contracts in pursuit of the general objective of “tracking the price of crude oil”. A typical crude oil ETF will hold long positions in WTI (West Texas Intermediate) crude oil futures contracts. As with most futures traders, these funds employ leverage, putting up a small portion of the capital to buy the contracts. The rest of the fund’s assets are invested in money market instruments which generate a modest amount of interest income for the fund.

ETFs – The Roll

In theory, the existence of crude oil ETFs enable individuals to implement a single equity trade to express a view that crude oil prices will either rise or fall in the future. Unfortunately, this objective is compromised by the existence of a forward curve in the crude oil futures market.

Because of the forward Curve, any ETF referencing crude oil cannot simply rely on ownership of the existing front month (closest to expiration) futures contract. To remain invested at all times, it must periodically sell its existing futures holdings and roll its exposure to a futures contract expiring in a more distant month.

ETFs – Return Components

With crude oil ETFs, the technical result of utilizing crude oil futures for the NAV (net asset value) return is dependent on three variables: 1) changes in the spot price of crude oil, 2) interest earned on un-invested cash, and 3) the ‘roll yield’ – which is a function of the spread between the price of the contract being sold and the price of the contract being entered. In contango markets, the roll yield will be negative because the fund must pay up to enter the more distant contract, and the opposite is true in backwardated markets. Furthermore, the precise timing of the forward roll can have a material impact due to the propensity of the expiring contract to experience high volatility in the days immediately prior to expiration.

Roll Yield – ETF Return Illustration

Let us consider the case of a hypothetical crude oil ETF which provides exposure to the front month crude oil futures contract, with the exposures rolled forward as expiration approaches.

To illustrate the concept of the roll yield, assume that the 2009 spot return on crude oil is +20%. But, assume that persistent contango in the market results in a cumulative roll yield of -15 %. In these circumstances, the combined return of a crude oil based ETF might be in the ballpark of +5%, a far cry from the +20% return generated in the crude oil spot market.

As illustrated, contango in the crude oil market may cause ETF returns to lag spot market returns. Not surprisingly, a flat/stable forward curve would result in a minimal roll effect. On the other hand, a backwardated curve may cause the ETF to outperform the spot market.

Roll Yields – A Source of Tracking Error

The variability of roll yields coupled with the shifting slope of the forward curve should dispel any notion that the return on crude oil ETFs will track the spot market of crude oil in a predictable manner. The managers of the crude oil ETFs and ETNs are fully aware of this issue and they make no claims regarding their ability to replicate spot crude returns. They merely claim to attempt to track a return benchmark that is comprised of crude oil futures contracts, thereby providing traders/investors with some ability to participate in directional movements in the price of crude oil. (Note: This situation is analogous to an issue which exists with certain leveraged (non-oil) ETFs which have proven to be extremely deficient in terms of tracking their benchmark indices in volatile markets. These instruments have strictly adhered to their stated strategies and objectives, but have failed to achieve the imaginary objectives of many careless traders.)

ETFs/ETNs – Examining the Fine Print

Included below is some language directly out of the prospectuses of some of the more popular crude oil ETFs and ETNs. Notice that the managers willingly acknowledge the issues related to forward curvature and the roll yield impacts on returns. This segment of analysis will focus on the three most popular crude oil instruments in existence today: an ETN (OIL) and two ETFs (USO & DBO). Many of the issues raised herein are applicable to all crude oil ETFs and ETNs, although leveraged ETNs (such as DXO) involve complications which are not analyzed here.

ETF versus ETN – Counterparty Risk

No discussion of crude oil ETFs would be complete without some mention of the important difference between an ETF (Exchange Traded Fund) and an ETN (Exchange Traded Note). With an ETF, holders are secured by the assets of the fund, so the credit worthiness of the fund manager firm is not a relevant consideration. If the manager collapsed into insolvency, the ETF would be unaffected, except for the possibility of a change in management (which is not an important consideration in the case of a fund which is passively managed to match a specific benchmark). In contrast, owners of an ETN are unsecured creditors who receive a mere ‘promise to pay’ equivalent to the value of the underlying assets. Let's simplify this distinction: If a given ETF and a given ETN have the same exact net asset value (NAV), it is conceivable that the ETN could be worth less than the ETF if the manager of the ETN asset pool experienced a level of financial distress which resulted in a material downgrade of the credit quality of the manager’s bonds. Given that most ETN managers are financial institutions with challenging balance sheets, this risk is worthy of consideration.

In a worst case scenario, the ETN manager could face an abrupt insolvency and default on the ETN. This risk, often referred to as counter party risk, is similar to the risk that credit default swap (CDS) holders faced when Lehman Brothers and AIG encountered insolvency. We, as rational individuals, do not buy insurance from high risk insurers. And, as such, we should think similarly about owning ETNs issued by high risk managers.

US Oil Fund (USO)

USO is a standard crude oil tracking ETF that utilizes a strategy resembling the hypothetical ETF analyzed earlier in this article. Accordingly, USO entails all of the risk factors related to the use of crude oil futures as a tracking mechanism for crude oil prices. As with all ETFs, the objectives, the portfolio structure, and the major risk factors are clearly disclosed in the prospectus.

From the ‘risk factors’ section of the USO prospectus,

"in the event of a crude oil futures market where near month contracts trade at a lower price than next month contracts, a situation described as ‘‘contango’’ in the futures market, then absent the impact of the overall movement in crude oil prices the value of the benchmark contract would tend to decline as it approaches expiration. As a result the total return of the Benchmark Oil Futures Contract would tend to track lower. When compared to total return of other price indices, such as the spot price of crude oil, the impact of backwardation and contango may lead the total return of USOF’s NAV to vary significantly. In the event of a prolonged period of contango, and absent the impact of rising or falling oil prices, this could have a significant negative impact on USOF’s NAV and total return."

Notice how they warn that USO may experience a negative roll yield which may cause the NAV of USO to deviate significantly from the spot price of crude oil. Is there historical precedence for USO deviating from spot oil by a material amount? As it turns out, the answer is ‘yes.'

"During the past two years, including 2006, these markets have experienced contango. This has impacted the total return on an investment in USOF units during the past year relative to a hypothetical direct investment in crude oil. For example an investment made in USOF units on April 10 and held to December 31, 2006 decreased, based upon the changes in the closing market prices for USOF units on those days, by 23.03%, while the spot price of crude oil for immediate delivery during the same period decreased 11.18%."

The only logical conclusion is that USO is not a direct play on the spot price of crude oil. It is, instead, a play on the spot price, forward prices, and the relationship between spot and forward (or, the slope of the futures curve).

Power Shares DB Oil Fund (DBO)

DBO is different from USO in that its managers utilize specialized strategies intended to mitigate the effect of roll yields on returns. As with USO, the prospectus for DBO directly addresses the issue of roll risk. In the case of DBO, however, the manager is not passive about accepting a negative roll yield in a contango market. In the words of the manager,

"Rather than select a new futures contract based on a predetermined schedule (e.g., monthly), each Index Commodity rolls to the futures contract which generates the best possible ‘implied roll yield.’... [The manager] is able to potentially maximize the roll benefits in backwardated markets and minimize the losses from rolling in contangoed markets."

I think it is fair to point out that the active approach being utilized by DBO presents both opportunities and risks in relation to the more passive rule-based approach used by USO. It is certainly conceivable that, by employing their optimization model, DBO will exhibit improved roll yields and better returns than USO. But, as with all things financial, the DBO model may backfire due to faultiness of imbedded assumptions. If DBO’s approach were full-proof, it is probably fair to conclude that an arbitrage opportunity might exist whereby profits could be generated by pairing a long position in DBO with a short position in USO. At this juncture, I am extremely hesitant to advocate such a strategy. Advanced readers wishing to gain a better understanding of DBO’s optimization approach may benefit from this link.

IPath S&P GSCI Crude Total Return ETN (OIL)

OIL is structured as an ETN issued as an uncollateralized obligation by Barclays Bank PLC. It is a financial institution that can be regarded as vulnerable to rising default risk in the current environment. As with USO and DBO, OIL is managed with the objective of tracking WTI crude oil prices by trading in crude oil futures (rather than the physical commodity).

OIL uses a specific benchmark to guide its futures trading activity: a crude oil index devised by Goldman Sachs. As with most index funds, OIL’s objective is to minimize the performance tracking error in relation to the index. OIL is not intended to perform better or worse than the index. The composition of the index, by design, can include any of the crude oil futures contracts which expire within three months. At present, however, the only contract used to calculate the index is the front month contract (which expires three business days prior to the 25^th of the next calendar month).

OIL’s roll strategy is formulated and it is unique from that of DBO and USO. But, the differences in relation to USO probably do not have a material economic effect. In particular, holdings of the front month futures contract are rolled over a five day period commencing on the fifth business day of the month. Essentially, this means that OIL will have completed its roll approximately two weeks before front month expiration. (Note: USO rolls two weeks prior to expiration).

In essence, OIL is more similar to USO because the roll strategy is formulaic, and not intended to minimize the effects of negative carry in a contango futures market (or maximize the benefits of backwardation). But, this similarity is also offset by the fact that OIL carries material counterparty risk since it is an ETN, while USO and DBO do not (since they are ETFs).

USO Versus DBO Versus OIL - Expenses and Liquidity

The volatility of these instruments is so high that expenses have a relatively minimal impact. But, for frugal and/or longer term investors, the following information may be relevant:

Annualized expense ratios:

• USO 0.86%

• DBO 0.54%
  

• OIL 0.75%
  

Empirical Data – Historical Price Returns

It is unwise to draw any specific conclusions from this information, but a quick examination of the recent market returns of USO, DBO, and OIL reveal the following facts:

Year-to-date price returns (thru 1/24/09):

• USO -2.9%

• DBO +1.1%
  

• OIL – 7.9%

Although inconclusive, the material disparity between the return of USO and OIL is potentially due to the fact that OIL is an ETN issued by Barclays – a Bank that has suffered from extensive credit quality impairment in recent days.

Returns, 2008 peak to current:

• USO -72%

• DBO -65%
  

• OIL -75%

Further assessment of the relative returns of these instruments can be found here.

Conclusion

1. Counterparty risk
2. Futures roll strategy (and roll yield)
3. Liquidity and expenses