I am looking at USO portfolio holdings and see that it contains CL March futures contracts and treasurys. Holdings - United States Oil Fund
This appears to be in the later category (ie Buy Prompt, Sell Forward, Roll Futures every month.) which you claim will not make profits long term. Yet looking at the long term price action of this etf it seems to go in unison with the spot price of WTI crude which it claims to be its objective. So I am wondering if there is some huge hidden cost to owning this etf? If there is no huge hidden cost, then how is USO profiting from its futures contracts?
For sake of this argument, assume the the price of crude will return to $100/barrell in 10 years, and USO will rise from $10 to $40/share (the price it was when oil was 100). As I understand you, the etf gains will be eaten away over the 10 years with hidden costs, (roll cost/yield) not reflected in its share price, so that there will be no net gain.
The reason they hold so many treasuries is because of the 7.6:1 leverage (30.39 * 1000 / 4000) currently offered by futures. Hence for every dollar (or $30,390 using CLH6 contract value Friday night) invested in USO they only need to use 13c ($4,000) to buy futures, leaving 87c ($26,390) in cash that they put into treasuries. It's actually not that simple when you include maintenance margin but the example holds for initial margin. The interest received from the treasuries should give the ETF's performance an uplift but unfortunately with near zero interest rates that uplift is insignificant.
Back in the 90's when interest rates where a lot higher and equity volatility was a lot lower than modern day a popular investment was a product that guaranteed you the greater of x% of the stock market or your money back. The way the product worked was that they invested a percentage of your initial capital into zero coupon bonds and the balance in long dated equity call options. Basically the interest received was paying for the call options.
I'm not saying its impossible, or that you won't make money if the market goes up, I'm just saying that it's inefficient. In a contango market you have negative roll yield on a long USO position which means that you will under perform the futures.
For example the prompt crude close peaked on 20th June 2014 at $107.26 at that time USO closed at $39.32 and CLG6 closed at $92.94.
On Friday USO closed at $8.79 which is down 77.6% while CLG6/prompt CL closed at $29.42 which is down 72.6% for Prompt CL and down 67.1% for CLG6.
Alternatively consider USL which is an ETF that invests in the Prompt 12 months rather than just the prompt month. On 20th June it closed at $47.31 and on Friday at $14.04 which is down 70.3%. ie better than USO or prompt crude but not as good as CLG6. The reason behind this is because by spreading the position out over 12 months rather than 1 month, you have approximately 1/12th of the negative roll yield that USO has.
Note that in a backwardated market roll yield actually becomes positive, unfortunately we are a long way away from that currently.
When a futures market is in contango, the price of the commodity for future delivery is higher than the spot price (longer-dated futures prices are higher than near-dated futures). A chart plotting the price of futures contracts over time is upward-sloping.
When a futures market is in backwardation, the opposite occurs (far-dated futures are lower than the spot or near-term futures price). A chart plotting the price of futures contracts over time is downward-sloping.
To recap, an investor buying a commodity futures tracker must reinvest continually from expiring nearer-dated contracts into further-from-expiry longer-dated contracts.
When the market is in contango, this means selling out of futures at lower prices and reinvesting at higher prices, a policy that generates a negative roll yield.
When the commodity market is in backwardation, a futures investor earns a positive roll yield by selling out of expiring contracts at higher prices and reinvesting at lower prices.
The following 4 users say Thank You to SMCJB for this post:
Thank you SMCJB for your informative reply.
I have placed your example data in a table for discussion sake:
My observations of the table data:
The prompt crude gain does not reflect the delivery and holding costs of a barrel of oil, and is not practical option for most of us investors, but none the less, it serves as a benchmark for comparison.
CLG6 expires Feb 2016. The prices shown is for a long term (1.5 year) contract purchased on 6/20/14. This option has best gain, and I assume would be the better performer (vs. any of the etf's) even if one had to roll it over for a multi year hold.
USO is the poorest performer.
The above example is in backwardation (futures prices falling), so there would be a positive roll yield. So I was wondering why USO, which uses short term monthly contracts, does not have a better gain than USL which used 12 month contracts? It would seem that the roll yield is negligible.
Is an effective long term strategy, if one is expecting the market to fall, to use short term contracts to take advantage of positive roll yield and if the market is expected to rise use long term contracts to minimize the effects of negative roll yield? Or is it always better to minimize the number of rollovers?
I believe USO is currently invested in CLH6 and will have to roll that position CLJ6. On Friday H6 settled 3039 and J6 settled 3138 so they will have to pay 99c more for April which will reduce their purchasing power by 3.2%
USL on the other hand is currently invested in CLH6-CLG7 and will have to roll the H6 position to CLH7. On Friday H6 settled 3039 and H7 settled 3743 so they will have to pay 704c more for Mar17 than they receive for Apr16 BUT they only have to roll 1/12th of their position, hence 704c/12 = 58.6c which is only 40% of the roll cost of USO.
If your expecting the market to fall why would you be buying it?
Under/Over performance is all about roll yield which is all a function of backwardation/contango.
Lets say you buy $10,000 worth of USO. On Friday night that would have meant you were long approximately 329 barrels of oil. If USO rolls from March to April at the 99c differential illustrated above, they will sell your 329 barrels of oil equivalent ("BOE") and buy 319 barrels of April. Your barrels are still worth $10000 but now you have 3.15% less. May is currently worth 93c more than April. Let's assume that oil prices stay unchanged for the next month and when they roll from April you will sell your 319 BOE in April and buy 310 BOE in May which is 2.9% less again. You still own $10000 worth of USO but now you own 310 BOE rather than your initial 329 BOE. Hence if Oil now goes up $1/barrel you only make $310 as opposed to the $329 you would have made if it went up 2 months prior. The two month wait/negative roll yield cost you 5.9% of your investment gain!
The following 2 users say Thank You to SMCJB for this post:
One small caution, on the less popular oil ETF's like "usl", their daily volume is much lower than USO, OIL etc, so be sure to use limit orders and not get caught with those high bid-ask spreads getting in and out..
I swing trade the "oil" etf on occasion but would love an etf without the contango problem for long term investing.. Something like "SPY" but for Crude...
The following user says Thank You to sptrader for this post:
Re something more SPY like, the reason that SPY works is because it actually buys the underlining equities in the SPX index where as USO is buying futures and not spot oil. If USO bought physical oil and held it, the roll issues would be eliminated, unfortunately physical oil has a significant holding costs, while equities in the SPX do not.
The following user says Thank You to SMCJB for this post:
I am trying to understand how to discern contango from backwardation. Looking at the chart line 6/20/14, the futures price is selling at 92.94 and spot price is 107.26. For the period between 6/20/14 and 1/15/16, why is this not backwardation, (far-dated futures are lower than the spot or near-term futures price)?