The way they get that number is something like this: imagine a small economy with one oil well, one gas station, and one commuter, and one speculator. The oil well owner pre-sells $1000 worth of oil to the gas station (that's $1000 in notional derivative value). The gas station owner pre-sells the same amount of oil to the commuter (that's another $1000). The commuter decides he'd rather walk, after all, and sells the same value of oil to the speculator ($1000). The speculator decides he'd rather play the soybean market anyway, and looks for someone who wants to trade oil. Alas, the only person who feels good about oil is the oil well owner, who is willing to buy $1000 worth of oil from the speculator.
Since nobody has canceled any of their speculative contracts, the notional value of the derivatives is the sum of the value of all of the trades: $5000. And yet, all but one of the participants made a positive bet (buying $1000 worth of oil) followed by an exactly-offsetting negative bet (selling the same amount of oil). So they all net out to zero, except the oil-well owner; we're all back where we started.
The problem is credit risk. If the gas station is destroyed in an earthquake, the commuter may not be able to collect on his bet with the station owner, and may thus be unable to pay the speculator, who may thus be unable to pay the oil well owner, forcing everyone to cut their spending or possibly even default. But in this case, it takes an outside catalyst, and the problem is not the derivatives themselves. The problem is credit risk. Also, earthquakes.
I still don't get it. If people are buying and the value is not there, they will loose money, right? Speculators can also bid on a lower price (derivatives), right? So, who are all those crazy people bidding only in 1 direction (higher prices)?
Note: That wall o' text makes it hard to pick out your key points.
> The "speculator" was placed into the mix with a limited role for providing liquidity
This makes no sense. Who did the placing? Every market has speculators, and even end-user-buyers base their decisions in some part on speculation.
The speculator has no altruistic diktat to provide liquidity. The speculator is there to make money, and providing liquidity is a side-effect. i.e. the speculator wants to take on a position, but can only do this if someone else wants to exit a position -- thus making a trade. The seller's interests are served just as much as the buyer's.
Now, there is a clear issue with cornering the market. However, there is a natural check to this behavior, in that taking delivery of millions of barrels of oil is expensive. And if you have no intrinsic need to consume a commodity, then it is very expensive to sit on. If supply is indeed artificially constricted by hoarding speculators, they will need to sell off their supply at some point.
There has been a lot of noise (in Washington, in the media, on reddit, and everywhere else) about speculators being responsible for the rising price of oil. Of course, it is true that speculators help set the prices. But, I cannot imagine how the world's consumers can be tricked into overpaying for the liquid supply.
There is one obvious way this could be accomplished: but, it doesn't seem practical for speculators to take physical delivery of a meaningful amount of oil and drive up the price by restricting supply.
So, as far as I can tell, if speculators drive up the price of an oil contract beyond the natural price point they lose -- not the consumers.
Just because there are a large number of market participants involved purely for speculation does not mean we can just ignore those who are there for genuine commercial transactions. What percent of people who buy oil futures contracts intend to take delivery of thousands of gallons of oil?
My understanding is the "speculators" create liquidity for the risk managers, thereby assuming their risk.
Why does the money leaking via commissions necessarily make the game sub-zero sum? Is it because we aren't looking at the big picture where everyone wins?
It seems like the risk managers must have a positive incentive to sell their risk on the marketplace instead of assuming it themselves. I.e. hedging another investment like someone else said already.
There are tens of thousands of consumers and tens of thousands of producers for many commodities. Each of them have different cash flow demands and different tolerances for risk. All of them operate in a world that creates risk.
Sure, products could still be moved without all the market's actors. But they'd move at inferior prices, jackpotting some producers and bankrupting others by sheer chance with little opportunity for smart producers to mitigate (or even profit from) risks.
Even pure speculators serve a purpose in the market; if nobody wants to hedge a risk or lock in a price by dealing with them, they can't succeed; if people do want to do that, the speculator is providing a service to them.
This, by the way, isn't any great insight on my part; it's the first chapter in any book that explains futures and options.
The interesting thing is that speculators are an essential part of the functioning of financial markets.
A futures contract is essentially an apportionment of risk. When a farmer and a speculator write a futures contract, the farmer is selling the risk that the price of corn will be lower than is profitable, in exchange for a guaranteed profit that is likely to be less than the true expected value of the corn harvest (hence, "fees and expensive derivative products").
If there's a bumper crop of corn and the true market price is much lower than expected, then the person who is wiped out is the speculator, not the farmer. It's much better for society for commodity traders who invested in corn futures to go bankrupt than for all of its farmers to go bankrupt, because farmers have specialized knowledge that is necessary for the future production of corn. In exchange for taking on this risk, they get a profit, which they may choose to invest elsewhere to hedge the risk.
It's basically the inverse of insurance. The farmer may also choose to take out an insurance contract that says "In the event of bad weather that causes my harvest to fail, the insurance company will pay me $X, enough for me to make good my futures contract with the commodity trader and have enough profit that I stay in business next year." As a result, he is completely insulated from events outside of his control, and can focus on what he does best, farming.
Now, I probably agree that there are (or were, pre-2008) too many people in the financial markets, and that we got some chaotic behavior that had to be bailed out by the taxpayers. I also personally chose not to go into the industry - I started my career in financial software, but decided I liked making things more than underwriting other people making things. But it's worth understanding the financial industry's social purpose before condemning it.
Speculators are transmitting information about future prices to the present, which is a useful thing to do.
For instance, after Katrina and various other oil shocks, speculators sent oil prices up in anticipation of shortages. This sent consumers the message: "please use less oil right now."
Usually, this is a good thing, though collective insanity ("house prices never go down!" or "it's a new economy, earnings don't matter!") can cause problems.
Ok, that sounds like a more convincing argument than the original one.
Speculators are risking their own money re-aligning badly-allocated capital in the economy and correct prices. If they are right they earn a nice commission.
Speculators get a bad rap, but actually play a pretty important role in futures markets by aiding price discovery and providing a more efficient transfer of risk between hedgers. To make money they need to try and accurately forecast where prices will move, and to do so will incorporate any and all available information - it is literally their job to try and know everything there is to know about what will effect commodity prices.
There has also been quite a bit of academic research on the effect of speculation in futures markets, and by and large the conclusion seems to be that they help, rather than hurt these markets and rises in commodity prices have been driven mostly by external factors [1].
And back when investing in such derivatives was limited to, e.g. buyers of such commodities, it worked fine.
Now the average commodity "investor" just sees a differently-labeled slot machine and cares not at all about the underlying commodity, and never intends to accept delivery of same. For speculators, increasing (not decreasing) the boom-bust cycle is beneficial.
Also, an important sidenote, that is unintuitive and very hard to internalize:
When quantifiably making profit via speculation, someone else is strictly losing value. This can be a major contributor for inflation.
Aren't the "speculators" a bit of a boogeyman? Nobody complains when the speculators provide economic benefit (which studies say they do).
Besides the general market efficiencies that speculation has been shown to provide, sometimes the speculators drive prices down very sharply in anticipation of commodity devaluation. No one ever seems to complain about that.
I'm not against speculators. They are a useful part of the market.
But let's not pretend they can't have a negative effect on society. People are anticipating water moving onto the commodities market. If water somehow surges, that's not good for the people at the bottom. Super high oil prices are bad for the world economy. It's advantageous for oil market traders but bad for everyone else.
True but most derivatives are based on an asset of some sort with real world value. It is not speculation based solely on the speculation that others will continue to speculate.
Since nobody has canceled any of their speculative contracts, the notional value of the derivatives is the sum of the value of all of the trades: $5000. And yet, all but one of the participants made a positive bet (buying $1000 worth of oil) followed by an exactly-offsetting negative bet (selling the same amount of oil). So they all net out to zero, except the oil-well owner; we're all back where we started.
The problem is credit risk. If the gas station is destroyed in an earthquake, the commuter may not be able to collect on his bet with the station owner, and may thus be unable to pay the speculator, who may thus be unable to pay the oil well owner, forcing everyone to cut their spending or possibly even default. But in this case, it takes an outside catalyst, and the problem is not the derivatives themselves. The problem is credit risk. Also, earthquakes.
reply