Notes for commodity analysts

Our readers are sophisticated investors and their advisors as well as corporations with a direct stake in commodity markets and prices, rather than the general public. For the most part, they read because they have to and their money and livelihoods are at risk rather than for pleasure or interest.

In an information-rich environment, most have access to several sources of news (Reuters, Bloomberg, WSJ, FT) and a range of more analytical content (consultancy studies, bank research, publications by official agencies, as well as commentaries published in the mainstream media and on wire services).

Our analysis needs to stand out and add-value by being more authoritative, forward-looking and innovative than clients read elsewhere. In a crowded marketplace, it will succeed or fail by its objectivity, willingness to challenge the conventional wisdom and examine stories or aspects that have not been covered elsewhere.


We are not forecasters or investment advisers and should be careful not to get into the business of offering investment or hedging recommendations. If clients want that they can turn to their own professional advisers, brokers and consultants, who know their specific circumstances and are paid to provide such expertise.

In any event, most clients are highly experienced decision-makers. They are paid to take responsibility for their investment decisions and recommending hedging strategies. So they are not looking to us to take decisions for them.

Instead most are looking for new angles and approaches to test their ideas and thinking. It does not matter whether they agree with us. They are free to accept our analysis or reject it. They can incorporate it into their own thinking; reject it as flawed and sharpen their own thinking in the process; or accept parts and reject others. We want to be part of a dialogue.

Given that the future is unknowable, analysts should try to avoid deterministic forecasts and think in terms of probabilities and the range of possible outcomes — identifying the balance of risks and the factors that make one outcome more likely than another or that could change the market’s direction.

Commodity markets are dominated by uncertainty, feedback loops and indeterminate outcomes. For almost all the interesting questions there are no definite right or wrong answers. Outcomes are context-dependent and timing is everything. The future is never obvious. If it was the market would have fully anticipated it already.

It is crucially important that analysts focus on offering analysis and analytical comment rather than opinion. Everyone from Ben Bernanke to the doormen at the Fed is entitled to their own opinion; they all are equally valid as opinions. But analysis is a matter of expertise and judgement and it stands or falls on how well it fits the evidence and helps predict what will happen next.

Opinion is cheap; clients won’t pay for it. In contrast, insightful analysis is a scarce, valuable commodity for which clients pay a substantial premium. We are interested in what will happen next (analysis) rather than what should happen (opinion).

In terms of tone, robust arguments and conclusions are fine. There is nothing duller than articles arguing indecisively “on the one hand … on the other”. But they must be supported by the evidence, and where arguments are controversial, we should be generous in acknowledging the existence of other points of view.

Let’s be courteous and avoid the heated rhetoric and personal attacks that have marred the blogosphere and make so many postings tedious. Ad hominem attacks are the refuge of those who lack better arguments.


We should assume clients are already familiar with the headline news and at least some of the analysis published by other institutions. There is no need to repeat huge amounts of news detail. Simple references with links to the main news story should be sufficient in most cases.

We are writing for a specialist audience rather than general readers. We should assume readers have a fair amount of financial sophistication (in most cases exceeding the level of the average journalist and analyst). There is no need to explain basic market concepts like contango, implied volatility or the relationship between bond prices and yields. It is however useful to explain specialist terms familiar only to users of a particular market.

There is a lot of value in writing both quick reactive pieces in response to major news events and pieces of market-moving data, as well as more stand-back reflective and strategic pieces identifying and analysing medium and long-term trends. The balance depends among other things on news flow, author expertise and where we feel we can add most distinctive value, and is likely to vary over time and across subject areas.

But it is probably worth thinking in terms of a decay curve or trade off between speed and depth. Rapid reactive pieces need not offer profound insight provided they are quick off the mark (traders measure time in minutes or at most a couple of hours). If we cannot offer a reaction in that sort of time frame, we need to add far more depth and distinctiveness otherwise we are not adding much to the debate.

News is important and (sometimes) moves markets. But markets are forward-looking and investors and corporate strategists are more concerned about what will (may) happen next. Identifying issues no one has yet thought about and which are not in the headlines at the moment but will be next week or next month or next year is often where we can add most value.

There is no standard template for a column. Short reactive pieces keyed off a major development or specific piece of data can happily co-exist with longer articles pulling together data from various sources to assemble a picture or high-level overviews of big trends.

Columns are units of thought rather than length. We have to draw the line somewhere for technical reasons and so as not to overtax readers (many of whom have short-ish attention spans). In practice the maximum length is 1300 words, while the minimum is probably around 300. Most articles probably come in at 600-1200 words.

Journalists think largely in terms of text but most of our clients think extensively in terms of graphs, charts, numbers and maths. Graphics and data are therefore integral to what we do. We should make the most of them — not just as after-thoughts and pretty illustrations to a well-written story but to help carry and prove the argument.

Analysts have their own areas of expertise. Even individual columns will vary. Some will be more focused on physical supply-demand fundamentals, others on issues about market pricing. But at a minimum anyone writing about market-pricing, prices and price trends should make sure they are familiar with the theoretical basics of how futures and options are priced, as well as implied volatility, liquidity, the factors shaping the forward price curve and if you are really ambitious volatility smiles and surfaces.

This stuff is central to the decisions our clients make every day. It shapes every single investment and hedging strategy and determines whether trades are expected to be profitable or not. Offering a view on likely price movements or hedging/investing strategies without a thorough understanding of the basics is perilous — a quick way to lose credibility with readers. There are lots of books on commodity derivative pricing (some good some bad) or I am always willing to help.

John Kemp

3 May 2011

Published by

John Kemp

Energy analyst, public policy specialist, amateur historian