All,
Please note that there is a call tomorrow to discuss the attached.
AGENDA:
1. Review outstanding actions
2. Continue with Commodity Swaps Discussion Paper.
Dial in details:
US: 1 888 481 3032
UK: 0 800 904 7961
Intl: 1 617 801 9600
Participant Code: 52016709
Note that the minutes to the last call are included below.
Regards,
Hans Ellis
*************************************************************
* Present
Hugh Brunswick, EFET
Alistair Cross, GS
Hans Ellis, SWIFT (Co-Chair)
Piers Evans, Markit Wire (Co-Chair)
Jared Getz, Glencore
Marc Gratacos, ISDA
Raphael Iyageh, Goldman Sachs
Chito Jovellanos, Forward Look
Owen King, Markit Wire
Brian Lynn, GEM
Bulent Ozkan, Triple Point
Ali Peera, GS
Richard Rigby, Glencore
John Solder, UBS
Chuck Witter, MS
* Apologies
Luis Fierro, Deutsche Bank
* New Members of WG
Jared Getz of Glencore introduced himself to the group. He is a business
analyst with prior experience in industry automation and standardisation
initiatives via the Leadership for Energy Automated Processing (LEAP)
organisation.
Richard Rigby, operations manager at Glencore, also joined the call.
Two new representatives from Goldman Sachs joined the call in place of
Esther Canosa, who has left the bank. Ali Peera is the GS commodity
operations manager and Alistair Cross is also from operations.
* Review actions from last meeting 1500 LDN 6th June 2008
>> In the light of feedback from GS, inclusion of deliveryLocation may
no longer be necessary . Does the group still feel that deliveryLocation
needs to be included in the light of this feedback?
Group agreed that this is not required. PE to remove from model.
>> PE to review how timeZone and holidayCalendar could be modelled.
PE asked if these fields were required given GS feedback.
AP stated that the holiday calendar would not be required as it is
inferred from the publication / exchange.
PE suggested timeZone might be obvious from the Commodity Reference
Price, but noted that if defining this without the use of the
ISDA-defined CRPs, it would be useful to have in the schema.
HB thought that if timeZone and holidayCalendar were included, they
should be mandatory but that the group should decide if they were
required or not.
Group to review how important they felt these fields were.
>> BO said he would provide a list of commonly-traded units to the
group.
The list was received.
JG questioned the inclusion of 'LOT'. Group agreed that this was too
generic.
PE will review and make recommendation to the group.
>> PE to follow up with the banks on effective date / termination date
position.
PE stated that the feedback received was that the effective and
termination dates would always be common to the legs.
Group agreed that these dates should be modelled at the trade level, as
is the case in the EFET schema. (See also section 1.2.2 of minutes)
EFET use the concept of delivery periods versus calculation periods to
distinguish between pricing periods and settlement periods.
>> PE to follow up with HB with regards to an ISDA FX / rounding
convention mentioned on the last call and with banks.
HB explained that EFET only provided one rounding container at the trade
level as there was often disagreement on how to specify rounding for a
confirmation.
Section 6.1 of the ISDA definitions is often cited as the answer to this
question:
"Fixed Amount = Notional Quantity per Calculation Period * Fixed Price"
JG said that if different units were in use, there was still confusion
as to whether the conversion should be applied to the notional quantity
before or after the multiplication by the Fixed Price.
AP pointed out that a lot of work was done on this in the LEAP working
group and suggested that Johanna Coehlo, the GS representative on the
LEAP working group, could circulate something to explain where things
have got to.
Group to take a view once this information has been received - the FpML
call is not the forum to go over operational issues being decided by
other working groups.
>> PE to look again at frontline (including / excluding) contract for
delivery date in the light of examples provided by HB.
AP said that the future always rolls to the next month, but could be the
sixth nearby or another month and not just the first nearby.
PE suggested that a type called deliveryDateRollConvention could be
created that would allow users to specify a roll to the next future a
certain number of days prior to the last trading day.
* Minutes
1. The group continued with the Commodity Swap Discussion Paper.
1.1 Fixed Leg
1.1.1 Leg Dates
As above, Group agreed to put dates at trade level rather than at leg
level.
1.1.2 Fixed Price Unit
Group agreed that it is possible to price with a different unit to the
one in which the Commodity Reference Price is quoted and this should be
provided for by the model.
1.1.3 Fixed Price
HB pointed out that the Fixed Price can vary per period. This is why
EFET attach the price to delivery dates and do not model it as a one-off
price on the fixed leg.
1.2 Floating Leg
1.2.1 Rename certain elements
Group agreed that this could be renamed to calculation to tie in more
closely with ISDA Commodity Definitions terminology.
Group also agreed that notionalQuantity should be totalNotionalQuantity
- PE to review inclusion of a notionalQuantityPerCalculationPeriod
element.
RR said that the model must support the ability to specify an amount of
commodity per day.
1.2.2 FX
HB explained the EFET approach - the FX to the settlement currency is at
the trade level. In other places where FX might be required, users may
optionally specify a contextual FX convention.
HB said the same is true of EFET's rounding approach.
AP said usually only a rate source is specified, but sometimes a time is
given too - particularly in metals trading.
HB pointed out that the EFET model allows for averaging of FX rates in a
similar fashion to the underlyer model.
Group agreed that the model should support a reference price source and
an optional reference price time as well as FX averaging.
1.2.3 Notional Resetting
PE asked if these variable notional trades vary in a predictable fashion
in each period.
AP said that these deals are rare but do trade - e.g. a refinery might
want to put a hedge in place that reflects projected volume declines in
the summer.
HB explained that the world scale multiplier can also be set per period
in the EFET schema.
* Decisions
deliveryLocation is not required in the underlyer model.
'Lot' is not to be included in the price-quote-units scheme.
effectiveDate and terminationDate will be at the trade level.
Rounding will follow the conventions determined in the LEAP working
group. The FpML call is not the forum to go over this again.
The ability to specify a roll convention for the nearby future on the
last trading day should be included - regardless of which nearby
contract is in use.
Model needs to support the price being in a different unit to the unit
of the Commodity Reference Price.
It should be possible to vary the fixed price / notional quantity per
calculation period.
Notional quantity should support x units per day as well as per
calculation period.
Certain elements will be renamed:
valuation --> calculation
totalQuantity --> totalNotionalQuantity
FX component should allow a reference price source to be specified, but
also support a reference price time and allow averaging.
* Actions
PE to remove deliveryLocation from the underlyer model.
Group to decide on importance of timeZone / holidayCalendar. Please
consider whether they would be needed in the event that no pre-defined
commodity reference price exists.
PE to review BO's list of units and make a recommendation to the group.
PE to update underlyer model for frontline future (including /
excluding).
Goldman Sachs to provide an update to the group on rounding.
* Next meeting 1500 LDN Fri 20th June 2008