From Monday February 3, 2020, Fastmarkets will publish its price assessments for standard and alloy-grade cobalt, in-whs Rotterdam, on a daily basis.
Here, Fastmarkets summarizes the feedback received as part of an extensive market consultation and details how that feedback informed the conclusion to increase the publication frequency of its benchmark price assessments.
Purely in number, responses showed a slight preference in favor of a shift to daily pricing compared with the alternative (maintaining the current twice-weekly pricing structure).
Some feedback acknowledged potential advantages and disadvantages of daily pricing but showed indifference in terms of increased frequency or otherwise.
Some responses expressed concern that one particular group or side of the market was pushing for daily pricing. We grouped responses according to the reporting party’s primary exposure to the cobalt market (and therefor to our price assessments); we found there was generally a prevailing preference among producers and consumers for daily pricing. Traders and distributors had a preference to maintain the twice weekly structure.
We also received feedback from those with no current, direct or rolling exposure to the cobalt market, such as banks or future physical market participants, included above within the “other” group. These parties were broadly in favor of daily pricing and the additional market transparency it affords, provided it could be done in a robust way.
In some cases, feedback was received from more than one person at a company - each company providing feedback was only considered once, no matter how many individuals provided comments.
The conclusion to increase pricing frequency was not drawn on the basis of a preference for daily pricing in number of responses alone. Since common themes were identified in consideration of the benefits and drawbacks of daily pricing, we considered the value or validity of those points and how they should inform next steps.
Most responses gave multiple reasons for their position; some gave no context.
That feedback was treated as detailed below. Arguments highlighted were considered material in informing the conclusion and shaping our approach to cobalt pricing:
Arguments in favor of daily cobalt pricing (in order or frequency of occurrence)
Provide additional market transparency (material): Daily pricing gives a more up-to-date view of the market; additional transparency was sought by most sections of the market to varying degrees. Price reporting agencies such as Fastmarkets exist to provide such transparency. We consider market liquidity, the amount of data we can expect to collect on a regular basis and market demand for such transparency when deciding how frequently it is appropriate to publish price assessments. Market liquidity is discussed in more detail below but the interest in additional cobalt market transparency was key to our consultation conclusions.
More responsive to market moves (material): As above, when markets are moving quickly, as is often the case in cobalt, daily prices are more responsive. At the same time, if the market is stable, that will be reflected in the daily price assessment. Daily pricing gives the capacity to respond to moves in prices in a timelier manner but a daily price move is not a requirement.
Prerequisite for active futures trade or hedging opportunities: Volatility in the cobalt market, including near-10-year highs in 2018, has meant increased appetite for a hedging mechanism that reflects the realities of the physical cobalt market. The London Metal Exchange launched a futures contract to be cash-settled against Fastmarkets’ standard-grade cobalt price in March this year; these contracts often function better if the underlying price is assessed daily (as in iron ore, alumina and aluminium premiums, for example). But since an active futures contract is not a foregone conclusion should cobalt prices be assessed daily, we did not feel compelled to use this argument to justify an increase in frequency.
Price assessment ranges will tighten: When prices are moving quickly, the ranges we publish often widen to reflect the spread of business that has been concluded within the assessment window (one example: a range of $12.25-13.95 per lb on the day Glencore announced it was closing its Mutanda mine). In cases like this, day-to-day pricing would likely have shown tighter ranges, since the extent of the price increase would have been shown incrementally.
In this case, the market was also very liquid, which is not always the case. During times of low spot liquidity, daily pricing could reflect two extremes of trades since market activity will be split across five days instead of two. Price reporters consider all the data received during a given pricing session in a critical way but it is not guaranteed that assessment ranges will tighten if cobalt prices are assessed daily.
Underpin use and adoption of hydroxide assessments: Fastmarkets launched two reference prices for cobalt hydroxide in February this year. Our assessment of cobalt hydroxide payables - the percentage of the standard-grade metal price paid for intermediates - has been proposed as the basis point for multiple 2020 hydroxide deals; it has been written into physical market contracts in some cases. Some feedback suggested the increase in frequency of publication of the underlying metal price would give longer validity to the payable assessment, which is published weekly. But most feedback (collected outside this consultation) has suggested that any hesitation in use of the hydroxide assessment lies in the need to gain familiarity with a new pricing mechanism rather than concerns about the frequency of the publication of the underlying metal price.
Arguments opposed to daily cobalt pricing (in order of frequency of occurrence)
Liquidity is insufficient to support daily pricing (material): The consultation notice originally published in October said an internal review of traded volumes, market activity and pool of contributors had established that prices could be assessed daily without compromising their robustness.
We have compared the data informing other Fastmarkets price assessments each time they are published with that collected for our standard-grade and alloy-grade assessments. As it stands, the amount of data informing daily cobalt price assessments would be akin to that feeding into other benchmark prices already assessed daily (more than some daily prices, less than others, if current data collection were maintained and split across five days rather than two).
But we acknowledge that liquidity was the primary concern cited for not introducing daily price assessments. The following point of clarification will therefore be added to the methodology for the cobalt price assessments:
“During times of low-liquidity or limited meaningful market activity (i.e. that which is considered to have taken place in competitive market conditions), the price assessments will reflect the relative stability of the market by being stable or little changed.”
Pricing rationales clearly state the treatment of data points collected by Fastmarkets during a pricing session.
Greater risk of manipulation attempts, or lack of balance from different contributors or sides of the market:
It is expected that data will be collected from both buyers and sellers over the course of a pricing session, whether daily, weekly, twice-weekly, etc. This is one of the factors considered during the process of prices being signed off, which involves two members of the editorial team not directly involved in that day’s price discovery.
We are also familiar with anti-trust or compliance considerations relating to a company’s own weight of influence on a particular price. We can provide transparency if required.
Concerns were raised about the impact of small-tonnage deals dictating price. Deals of 1-100 tonnes are considered for Fastmarkets’ price assessment. A review of deals reported in the first three quarters of 2019 shows that smaller deals are not concluded at a meaningfully higher or lower price than higher tonnage deals.
Concerns were also raised in relation to an over-dependence on inter-trade business, though broad support for daily pricing among producers and consumers suggests this is not a shared concern.
As ever, price reporters will consider both the veracity of a deal having taking place and the significance of it having been concluded. It is expected that reported business is concluded in a free-market and competitive environment; a price reporter will seek context to determine that this is the case.
Limited benefit vs cost or required investment: Feedback here centered on the sense that market events that prompt significant day-to-day price moves are the exception rather than the rule. Respondents said the burden of reporting data on a more regular basis outweighed the occasional advantage of more responsive pricing. While market events like the closure of a mine are unusual, their effects on market activity are felt for considerably longer. We therefore considered that there was greater value to daily pricing than merely on a few days in one year.
Other (similar) feedback focused on the need to submit data or speak to a reporter on a more regular, potentially daily, basis, depending on market activity. We are keen to discuss automated data submission, back-office data submission or to sign data submitter agreements to ease the process of providing data to our price assessments (none of these options precludes a data submitter from also discussing the market and other contextual information with a price reporter, in our view).
Prices should be reflective of a trend: Several respondents said they felt the twice-weekly price assessment enabled a price reporter to smooth over day-to-day volatility. This is not the intention of our price assessments, which is to reflect market moves in a timely manner. Instead, monthly or weekly averages can be used as an indicator of market trend over time.
Some were concerned that liquidity split over five days, rather than two, would make it harder to identify outliers (addressed above).
On a similar note, some feedback indicated an expectation that daily pricing would suppress some of the volatility that has been a feature of the cobalt market in recent week, or asked whether that was Fastmarkets' intention in introducing daily pricing. That is also not the intention of daily cobalt pricing: our price assessments exist to reflect market activity and provide transparency; the purpose is not to lead the market or indicate where the market should be trading.
Disruptive to market flow: Fastmarkets’ cobalt price assessments are watched closely as an indicator of market strength. Market activity often reacts to our published prices; some said it could generate additional activity on the Wuxi night market. Others said it would limit the validity of their offers; some said it meant additional workload in having to mark to market their book each day. We considered the support for market transparency in a volatile market to outweigh these concerns. Similar concerns were voiced in the reverse: That twice-weekly pricing stifles market activity because prices are quickly outdated. While we are mindful of the far-reaching impact a change in the benchmark price can have on market sentiment and activity, these effects are largely beyond our control. The requirement to provide timely and accurate price assessments is our priority.
Timetable for January launch too tight (material): Our original consultation notice said we were open to a January launch, suggested on the basis of feedback to previous consultations, which requested a January implementation date for any substantial changes to specifications. Some feedback to this consultation requested a later implementation data, thereby suggesting a January implementation for any changes is not actually essential.
Fastmarkets will typically give 30 days’ notice ahead of any specification changes so a February 3 launch for daily pricing goes beyond that standard. Where publication frequency is concerned, prior knowledge of a publication schedule is crucial for those using our prices, since contracts are typically signed on the basis of an average of quotes published within a given period.
We acknowledge that a daily cobalt price was not the preference of all market participants responding to our consultation and that this stemmed largely from concerns about liquidity or data submission. As a result, daily pricing will begin in February to allow Fastmarkets and the market more time to discuss the practicalities described above.
To this end, Fastmarkets will host a web seminar on Thursday January 16 at 3pm London time to discuss the results of this consultation in further detail, as well as some of the practicalities of daily cobalt pricing. Click here to sign up.