I Don't Get Futures Markets

[PerpetualCzech]

Here's one thing I've never understood about futures markets. Say you have oil trading today at $35 but one-year futures contracts are trading at $40. How can that futures contract ever trade at anything other than a price which represents the current price plus whatever a 1-year T-Bill is currently returning? Doesn't any other price offer an opportunity for a scalp? In the above example, why can't someone borrow a million dollars at money-market rates, go long at $35 and sell short the one-year futures contract?

Or another way of maybe putting it: if the expectation is that the price will hit $40 in a year's time, why doesn't that make the price $40 right away *today* (minus any appropriate interest)?

(Hope I explained that well ...)

[Prop Man]

one possible reason i can come up with: the cost of keeping the product in storage
if you buy 100 barrels of oil and take delivery (in order to deliver the product in a year), you'd need a whole setup in order to not only keep the oil around, but possibly to service it (just a guess, I don't know if you can just keep oil lying around or if you have to "stir" it once in a while to keep it fresh or something odd like that).

My thought is that it is possible to do those type of arbs, but the barriers to entry are high because not everyone can take delivery or deliver. So the question is : do the entities that can take delivery think there is enough profit in holding it for a year for 14.2% ROI? Or is it worth more to them to crack it and turn it into heating oil or gasoline and make their profit that way? I'm sure at some level, its worth more to hang on and make money on the time spread, but maybe not at a $5 difference.

if it was equities, it would be a different story since there is little barrier to entry as the cost of keeping a portfolio of stocks is relatively small. so you rarely see the odd differences in time spreads in equity index futures as you do in commodity futures.

you should check out the electricity futures spreads from time to time. storing electricity is near impossible (or so I was told 10 years ago), so you can't buy it today and keep it for 6 months to sell in the future.

[PerpetualCzech]

OK, well then what about something that's easier to store then, like gold?

[Prop Man]

gold is probably much easier to store. I'd assume there is a less difference in gold futures than oil futures...but probably more than in equity futures.

[PerpetualCzech]

Also, what if the 1-year future price is *less* than current price? (Does this ever happen?) In that case you are selling short the commodity today, buying back the futures contract and storage isn't an issue.

[Rudy1957]

The mechanics of physical futures delivery have to taken into the equation, too, and that's what messes up what would seemingly be a simple incremental cost calculation to separate out-month contracts.

In short, there are arbatraguers who daily try to exploit the price differences between the close months and the outlying months, but periodically their efforts are overwhelmed. Such was the case 6-9 months ago in the oil market when speculative frenzy was at its peak, and out-month futures contracts carried a huge premium, which became unarbitagable becasue there wasn't enough oil available for future delivery to support more shorting of the out months.

When the calculation is out of whack with the normal cost-to-carry for future months, the market is called in "contango" -- Wiki that one for further explaination.

Now, the oil market is in contango for a different reason: there is such a glut of near-term physical supply, that it's being dumped on the market at prices lower than those that people are willing to sell for later on. Logically, then, someone might want to buy some of that glut now, find a place to store it, and then sell it in a few months at the higher price now available for future delivery. Except for one thing -- there's no place to store it. Arbatraguers have already done this to the maximum extent physically possible -- every available tanker and storage tank is already taken, most of them biding time until they come up for delivery, or already contracted for later months.

This will continue to be the condition of the marketplace until either demand picks up to absorb supply, or supply drops to the level of current demand. What's going on is that the oil-producing countries need the money, especially because prices have dropped so precipitously, that their only tool is to pump faster to generate incremental revenue. Despite what they say in public about production plans or targets to try to manipulate the markets, they lie and cheat. So, more oil is coming out of the ground right now than is being used, and the spot price has to keep coming down until things are back in balance again.

Part of the solution is for the oil producers to sell more out-months production rather than immediate supply, but they play a balancing act in doing so because they want to prop the out-month prices so that the spot prices aren't under further pressure. Moreover, there is less trading volume in the out months, so it is harder for oil producers to unload as much future supply without moving the market.

Ultimately, a balance of supply and demand in the near month is the only cure for contango, which is why people pay close attention to the inventory numbers more so than the stated (often ficticious) production numbers.

[Rudy1957]

Also, what if the 1-year future price is *less* than current price? (Does this ever happen?) In that case you are selling short the commodity today, buying back the futures contract and storage isn't an issue.

The only time you'll periodically see a future price being lower than present price is in perishables (like ag), where there's a near-term unmet need and no way to accelerate near-term supply. It could happen briefly in other commodities if there was a major short-term supply disruption (i.e., major gas pipeline blowing up), but future supply is still in balance. It also happens sometimes in the interest rate markets, which is known as an inverted yield curve. When that happens economic trouble is invariably on the horizion.

Again, you have to remember that the shorting mechanics are different for physical assets than financial assets becasue they're not as fungible. You can't be short February natural gas, and say that you'll just buy a May contract to cover the short because gas is cheaper then. You must deliver in February or cover by buying a February contract. Sometimes, big speculators (or forward-selling producers or hedgers) will misgauge their risk, which is what can cause spiralling short squeezes. The rising price needs to go far enough to bring out more supply immediately that otherwise wouldn't have gone to market. That's largely what drove the last leg in oil last spring.

[PerpetualCzech]

OK, I get the delivery and storage problems. I wish I hadn't used oil as my example. So what about gold, which presumably does not have these problems (it should be fairly easy to deliver and store a hundred thousand dollars worth of gold, no?)

Why don't we take a specific example for gold:

What is gold current trading at for standard delivery in a few days? (I do not know standard settlement period for gold)

What are the 3-month, 6-month, 1-year, 3-year, 5-year and 10-year future contracts for gold currently trading at? (For that matter, what is a good source to get these types of quotes for future contracts, including other commodities?)

[PerpetualCzech]

You can't be short February natural gas. You must ... cover by buying a February contract.

So is that the same as saying you simply can't sell short commodities?

[Prop Man]

You can sell short commodities, but you just have to be prepared to buy it back before expiration or deliver the product. Equity futures will often settle to cash, meaning they settle to the pricing of the index reflective of all stock openings (or closings, depending on the contract specifications) on expiration day. This can be a little tricky due to program trading and expiring options as well.

Gold futures at the NYMEX: http://www.nymex.com/gol_fut_cso.aspx
Not sure if the formatting will come through, but here is the cut/paste.

2/9/2009 Session Overview



Last Open
High Open
Low High Low Most Recent
Settle Change


Feb 2009 902.30 n/a 910.00 911.00 898.00 913.90 -11.60


Mar 2009 903.40 n/a 910.10 910.10 899.00 913.80 -10.40


April 2009 903.50 n/a 913.40 913.40 898.30 914.30 -10.80


June 2009 905.00 n/a 912.50 912.50 900.50 916.20 -11.20


Aug 2009 904.70 n/a 911.00 912.50 904.70 918.10 -13.40


Oct 2009 n/a n/a n/a n/a n/a 919.90 0


I think the best number to look at is the "Most Recent Settlement" (the number second to the right). Not all futures contract trade actively, especially in the out months. So looking at last sale may be misleading. At the end of the day, the traders will mark every futures contract to be fairly in line with what they think the market is. Just looking at it briefly, there does not seem to be anything currently unusual.

adding: if you go to the "View all contract months, volume, and open interest" link, you'll see that April is the month with the greatest volume right now. That's the one most people are trading, and the two front months (Feb and March) are not nearly as active. My guess is that this is normal - that the most active month is 2/3 months ahead, and not the closest month...maybe because the closest months have issues with delivery that make them less useful to traders overall.

[Rudy1957]

OK, I get the delivery and storage problems. I wish I hadn't used oil as my example. So what about gold, which presumably does not have these problems (it should be fairly easy to deliver and store a hundred thousand dollars worth of gold, no?)

Why don't we take a specific example for gold:

What is gold current trading at for standard delivery in a few days? (I do not know standard settlement period for gold)

What are the 3-month, 6-month, 1-year, 3-year, 5-year and 10-year future contracts for gold currently trading at? (For that matter, what is a good source to get these types of quotes for future contracts, including other commodities?)


Gold will typically trade much as you initially contemplated because it's supply is fairly consistent, demand for physical gold is relatively low, and the arbatrageurs can set up their typical spreads when the prices get out of whack. The premium for out-contract gold is almost exclusively cost of money plus cost of storage. Presently, the premium for out-months is actually fairly low, owing to low cost of money, roughly a dollar an ounce per month.

The only exception to this normally rational market is when there's a dramatic/disruptive world event that temporarily will increase demand for physical gold, causing near-term contracts to trade at a premium, potentially causing short squeezes, until the market settles back down and the arbs do their thing. Similarly, the same squeeze conditions we discussed earlier can cause the front month to trade at a small premium. That's happening a little bit now because there has been higher-than-normal demand for physical gold by people who are worried about the veracity of paper currency. That's also why people who are out there buying coins, etc., are having to pay higher-than-normal premiums to spot market. Supply and demand is everything in gold.

An interesting note about gold is that the size of the world economies have dwarfed the available supply of gold. All of the gold ever produced in human history would fit in a block roughly 60 feet square. And roughly 2/3 of all possible gold has already been mined. The remaining gold is increasingly more expensive to mine.

Don't get too carried away with these stats (as some gold bugs are prone to do), as gold can't effectively serve as a true stable currency because of the limited supply. Other assets can and do serve the same function. The eternal argument is whether there's truly any inherent value to gold. I don't want to engage in that argument, because it's unresolvable. In an Armageddon scenario, it will clearly be the currency of last resort, but short of that, not so much. Always remember that the downside of gold is that it has an opportunity cost, so if you're holding it, you're effectively losing money if it doesn't go up in price. It hasn't been a good investment over a 25-year holding period.

[Rudy1957]

PC, going back to your original post, beyond the mechanical issues, far-out contracts also have a built-in speculative premium or discount, much the same way stocks do. When consensus expects prices to be rising or falling, the speculative premium rises or falls accordingly. Once you get past the cost to carry a contract, the rest is all speculative premium. That's how speculators try to generate alpha (return in excess of cost to carry), by being long- or short-biased in calendar spreads If one thinks they have a better opinion than that reflected by the market price, they position accordingly, and their trades aggregately reset the market price. Much the same as how lines get beaten into shape, but with far bigger money involved.

But be wary of thinking there are easy scalping opportunities out there. These markets are hugely efficient except during times of great upheaval. A truism that will always serve well is that the aggregate wisdom of the market is consistently greater than any individual's.

[TJMAXX]

These markets are hugely efficient except during times of great upheaval.

Lol. Basically, you're alive until you're dead. I'll leave out the fat tail jokes from the past 18 months for now.

Btw, BJ21 had some interesting posts with Don S'. view of Taleb et al.

[Prop Man]

Futures prices have nothing to do with distributions (which is what Taleb is talking about - fat tails and black swans). Options and other derivatives are the ones that deal with distributions. Futures are more akin to the mean.

[custer]

Thank you very much guys. This thread is great.

Prop man, do you think that there are many squarish thinking people who pretend to know much more than they do operating in the financial markets? Or are the people that really move the money very knowledgeable? ie, what do you think the capacity is for logical but uneducated people such as myself and pc in these markets?

[Prop Man]

I think it is possible.

The good news is that the spread is usually much tighter than it is in sports and the volume you can get down is usually much bigger. The market place likely won't be scared of your action.

The bad news is that the guys making markets are tons smarter than the bookmakers.

I know that's pretty vague, but it is tough to get specific without knowing the details of how you'd go about trying to make money...and without knowing the specific product. For example, I don't think you'd be able to do that well trading as a customer in the S&P 500 futures or options markets. I have that opinion because i know that market a bit better than I know other markets (even if it is almost 10 years removed now), and it is tough for me to think any customer has positive EV in that market. There is so much information that someone as a market maker may know that you as a customer would never know, which makes me think it would be tough for a customer to have positive EV in that environment.

But if it was a less active and understood market, I think you'd have a better chance. Just remember there are hundreds or thousands of guys just as smart as you, if not smarter, possibly looking at the same thing. There are millions of dumb guys, so that'll give you a possible edge.

Lastly, if you want to make the real bucks...sports is not it. It is just too small. Even with the reshaping of Wall Street, my opinion is that the big money is still there.

[custer]

Thanks a bunch, prop man. My question was more prompted by reading so many stupid opinions in the news and hearing some people talking at my father's posh health club than from an actual idea. But it is something we've tossed around for awhile, citing the same benefits as you mention.

[Jeff Jones]

I count not understanding Futures Markets as a huge blessing.