Here total return = spot return + collateral return + roll return.

If the forwrad curve in in backwardation (negative), who will have greater roll return- the long or the short position? assuming everything else is the same?

Here total return = spot return + collateral return + roll return.

If the forwrad curve in in backwardation (negative), who will have greater roll return- the long or the short position? assuming everything else is the same?

In backwardation, roll return is positive for the long, negative for the short.

Can you elaborate?

In backwardation the futures curve is upward slowing and a long position would roll backwards down the curve buy selling short dated futures at a higher price and buying longer dated futures at lower price. If you are short position then the opposite is true and you are losing money on the roll back…

I think this is correct.

backwardation: F(0) < S(0)

At T: F(T) = S(T)

Roll-Return: = (F(T) – F(0)) – (S(T) – S(0)) > 0

>>> Profit for the Long Position

I would not say that this is per se correct. See my post here: http://www.analystforum.com/forums/cfa-forums/cfa-level-iii-forum/91340123

However, I am still confused, too.

That’s better.

It is correct; it’s the definition of backwardation: forward (futures) price is lower than spot price: the forward curve slopes downward.

S2000, I have a sticking point on this backwardation vs contango… I have looked at some graphs of the curves and can see that the futures curve is upward sloping for backwardation…but you (and books) saythat it is downward sloping… . I am obviously missing something fundamental.

Is the graph below not showing the futures curve? And is the backwardation not upward sloping?..

thanks!

S2000magician:

JJ1337: UpUpDownDown:backwardation: F(0) < S(0)

I would not say that this is per se correct. See my post here: http://www.analystforum.com/forums/cfa-forums/cfa-level-iii-forum/91340123

However, I am still confused, too.

Jep, I also think that it is more or less correct. However, I don’t get this one:

JJ1337:

However, when it comes to currency (I refer in the following to the base currency and price to base nomenclature) we speak of backwardation if the spot exchange rate is below the forwards exchange rate (i.e., S_0

pokhim:

S2000, I have a sticking point on this backwardation vs contango… I have looked at some graphs of the curves and can see that the futures curve is upward sloping for backwardation…but you (and books) saythat it is downward sloping… . I am obviously missing something fundamental.

Is the graph below not showing the futures curve? And is the backwardation not upward sloping?..

thanks!

There are a couple of points here.

First, the green curve is not defined as backwardation; it’s _ **normal** _ backwardation, which is different.

Second, the curves here do not depict today’s forward curves; they depict what happens to forward prices as we move into the future. In contango, the forward price is above the spot price (as shown by the “Today” points). As we move into the future, the forward price and the spot price have to converge, so, assuming that the spot price remains unchanged, the forward price will have to decrease. The opposite is true for backwardation.

It’s a lousy graph; the upper curve should be described as _ **normal** _ contango, not contango.

In any case, it’s not a graph showing how forward prices today compare to today’s spot price.

pokhim:

S2000, I have a sticking point on this backwardation vs contango… I have looked at some graphs of the curves and can see that the futures curve is upward sloping for backwardation…but you (and books) saythat it is downward sloping… . I am obviously missing something fundamental.

Is the graph below not showing the futures curve? And is the backwardation not upward sloping?..

thanks!

This is important. What the graph appears to be showing is what happens to the price of the contract as time goes by if the futures market is stagnant in contango or backwardation with no change in spot. Just my guess. Not clearly presented. Expect to see graphs presented in none standard terms. And questions that require you to mentally shift the curves.

S2000magician:

pokhim:Is the graph below not showing the futures curve? And is the backwardation not upward sloping?..

thanks!

There are a couple of points here.

First, the green curve is not defined as backwardation; it’s _

normal_ backwardation, which is different.Second, the curves here do not depict today’s forward curves; they depict what happens to forward prices as we move into the future. In contango, the forward price is above the spot price (as shown by the “Today” points). As we move into the future, the forward price and the spot price have to converge, so, assuming that the spot price remains unchanged, the forward price will have to decrease. The opposite is true for backwardation.

It’s a lousy graph; the upper curve should be described as _

normal_ contango, not contango.In any case, it’s not a graph showing how forward prices today compare to today’s spot price.

How do you define Normal Backwardation vs. Backwardation?

Only definition i’ve been able to find is during cases where backwardation is evident due to natural market conditions where they (the author) would call this market structure “Normal”, not due to shortage of the product.

Roll Return/ yield: Roll yield can be earned by rolling long futures positions forward through time.

Backwardation: It occurs when longer maturity futures contracts have lower price i.e. downward sloping term structure of futures prices. In this situation, positive return can be earned through buy-and-hold strategy i.e. when futures price < spot price, it increases over time (converges to spot price) as it gets closer to maturity and generates positive roll yield.

Don:t think this has been asked on this forum yet but I amn having alot of issues reconciling the two (from CFAI): Price return derives from changes in commodity futures prices, which comes from the changes in the underlying spot prices via the cost-of-carry model. In other words, when the **spot price goes up (down), so does the futures price, giving rise to a positive (negative) return to a long futures position.** VERSUS Roll return or roll yield arises from rolling long futures positions forward through time and may capture a **positive return when the term structure of futures prices is downward sloping.** Can someone please help me understand why price return and roll return do not contradict each other? My read of it is, yes, futures px converge to the spot px, earning you a positive futures position return when markets are in contango. However, at the same time, the roll return is negative if the futures price is in contango. Something is fundamentally amiss here, please help! is it because roll return is totally separate from price return? if so, how do these two soure sof returns interact? do they balance each other out if one is negative and one positive, as my aforementioned logic (?) states? Thank you

bump…

I think that the best way to get contango and backwardation straight in your mind is to do an example with quotes in table format. Ignore all of those charts because they are inconsistent and confusing, if you see them in prep material go ahead and rip them out and throw them out. There is a good example with tables in the cfai book, do that.

contango: you pay more today to ensure that you get an asset in the future. So you overpay for the forward contract to make sure you get an asset.

backwardation: It’s cheaper to buy the asset by ordering ahead of the time you need it. Think about the pricing of concert tickets, if you order early you can get them cheaper than if you try to buy them an hour before the concert.

In both cases the prices of the forwards get closer and closer to the spot rate as they get closer to settle. You can think of it like amortizing an accounting premium or discount over the lifetime of an asset.

Thanks kikentai. I fully understand contango and backwardation and what each implies for the roll return whether you are long or short the forward contract. However, my question was how to reconcile a positive spot return with a negative roll return when we are in a contango environment. Think i figured it out though - total return is sum of spot, collateral, and roll return. Therefore, spot and roll are completely distinct sources of return. So my initial struggle around reconciling the two has been resolved.

Galli:

S2000magician: pokhim:Is the graph below not showing the futures curve? And is the backwardation not upward sloping?..

thanks!

snip.

There are a couple of points here.

First, the green curve is not defined as backwardation; it’s _

normal_ backwardation, which is different.Second, the curves here do not depict today’s forward curves; they depict what happens to forward prices as we move into the future. In contango, the forward price is above the spot price (as shown by the “Today” points). As we move into the future, the forward price and the spot price have to converge, so, assuming that the spot price remains unchanged, the forward price will have to decrease. The opposite is true for backwardation.

It’s a lousy graph; the upper curve should be described as _

normal_ contango, not contango.In any case, it’s not a graph showing how forward prices today compare to today’s spot price.

How do you define Normal Backwardation vs. Backwardation?Only definition i’ve been able to find is during cases where backwardation is evident due to natural market conditions where they (the author) would call this market structure “Normal”, not due to shortage of the product.

futures price < **curent** spot price - backwardation.

if futures price < **expected** spot price - normal backwardation.

Add EXPECTED in spot price and it becomes “normal”

Magician, can you confirm?