Metallgeschaft - Contango / Backwardation

[dennis_cmpe]

I think I got the MG question right in the 2008 FRM exam, but I confuse myself when thinking about why positions lose value in contango or backwardation.

For the short position in long-term forward contracts during backwardation, these gain value because it locks in a higher price in the future for the seller? During backwardation, the forward price will be lower compared to the agreed price in the forward contract, creating profits?

For the long position in short-term futures contracts during contango, these lose value because it locks in a higher price in the future for the buyer? Each rolling hedge will require MG to buy at a higher price?

 

[Stefan]

From my point of view, your analysis and conclusion are correct.
Rgds
Stefan

 

[David Harper CFA FRM]

Dennis,

I too agree with your analysis. I would add:

In regard to long-term short forwards during backwardation: IMO, these are the simpler case. These are MG's contracts with customers. I don't think contango/backwardation is the essence here; rather, these short forwards gain when spot prices declines and lose when the spot increases (i.e., for this leg, MG is just like a commodity hedger - as you say, locking in a sales price - but, because MG needs to buy the deliverables, they profit on the short when the price goes down). The backwardation enters here because: this historical backwardation had tended to associate with stable or declining spot prices, but not increasing spot prices. But I say backwardation is not essential here: in MG's case the spot price declined yet the backwardation shifted to contango (i.e., the forward declined too but less than the spot) and these forward-contracts-with-customers are still okay (temporarily). The role of backwardation here is that it tends to signal that spot prices are steady or declining. So, okay, this is the simpler case: shorting is profitable when the spot is going down.

The other leg, IMO, is more difficult. These are the short-term long futures that are "the hedging instruments." Here the key perspective, arguably, is that we expect spot and futures to converge. So, under backwardation, for example, if spot is $100, forward one month = $90, forward two months = $80 (i.e., backwardation). Now, assume stable spot price at $100, and further assume forward must converge to spot. Now, I go long the 2-month forward @ $80 today. Now go forward in time one month. Spot is $100 again, my forward contract price has increased +$10 to $90 because it is converging with the spot. This is my "rollover" profit that i am virtually assured if (i) backwardation persists and (ii) the future converges to the spot (zero basis at maturity).

In MG's case, the shift to contango involved: forward prices falling (with stable spot, notice this by itself implies backwardation) but spot prices falling even more. In this way, the forward curve shift from inverted to "normal" while generally drifting down. Now this is is a bad deal. The rollover above is losing; the forward is converging but it is converging to a spot that is dropping. It may help to think of the forward as "chasing the spot" over time. In MG's contango, the forward started to chase the spot down - so the forward price is dropping.

The terrible irony in MG's case is that contango did not hurt the underlying contracts. The underlying contracts were winning (these forward just could not be booked). The hedges broke MG. Ironically, without the hedges they were fine !!??

What i've said IMO is still consistent with (not in opposition to) what you've said.

David

 

[shazrul]

IMO easiest way to understand how positions lose/make money in these two term structures is to imagine holding two forward positions.

In a backwardation, if you are long front to back i.e. long near term and short long term, and the structure goes into contango you will lose money as the front has lost more relative to any possible gains at the back. The steeper the contango, the more you lose.

 

[frm_daniel]

HI David,

Could it be possible to present the case in excel ? Numbers tell thousand words. I am sure many many students will benefit from it.

Daniel

 

[David Harper CFA FRM]

Hi Daniel,

Yes I agree, I will add to the "to do" list for XLS spreadsheets. Thanks for the good idea (I think the essence of such an XLS is to illustrate the loss on roll return with shift to contango). I can't promise timing just b/c there is so much content to focus on....David

 

[[email protected]]

In MG's case, the shift to contango involved: forward prices falling (with stable spot, notice this by itself implies backwardation) but spot prices falling even more. In this way, the forward curve shift from inverted to "normal" while generally drifting down. Now this is is a bad deal. The rollover above is losing; the forward is converging but it is converging to a spot that is dropping. It may help to think of the forward as "chasing the spot" over time. In MG's contango, the forward started to chase the spot down - so the forward price is dropping.​

Hi David,

Sorry for replying to an old thread, but the example is still relevant. Perhaps this gets to the debate of whether MG was hedging or speculating, but why didn't they just enter into longer-term futures contracts? The 6 month contracts used forced MG to rollover under unfavourable terms (contango) - was this strategy employed just to maximize profit?

Thanks!

 

[David Harper CFA FRM]

Hi Roberto,

I think it does relate to the debate, as i understand it: our previously assigned case (Culp) asserted that MG was simply not hedging. I do tend to agree with you, that if the goal was to hedge (i.e., to minimize the basis risk), then matching the maturities best achieves this. More simply, the STRIP hedge is generally more appropriate than the STACK (stack and roll) so-called hedge. I think we might even re-phrase the question, why would anyone stack hedge when they can strip hedge?

McDonald (Chapter 6 is FRM assigned, but maybe not this section) gives two reasons:
1. more trading volume and liquidity in the near term contracts, and
2. speculate on the shape of the short-term curve (note: this motive isn't hedging! this dilutes the hedging motive...)

but also, I think you must be correct, that their motive was to book profits in addition to hedging: as i recall when i looked at the numbers, they were booking profits on the stack futures for a while (as they should have been under backwardation). It makes total sense this influenced them into a weak hedge that proved to not really be a hedge at all

Thanks for the good point!
David