In North America and Europe, we tend to be quite good at measuring certain things, such as natural resource production and consumption. This is a characteristic of market-based economies where transparency is critical for properly functioning capital markets. People maintain great faith in a statistical ability to extrapolate from historical data.
This habit becomes a problem when we extend that assumption of accuracy to the other parts of the world. We forget (or ignore) that many countries have neither the desire nor the capability for full statistical transparency. Yet despite this, it appears markets routinely overreact to minute changes in global energy statistics that fall way beyond the margin of error of measurement.
As a case in point, energy industry markets react strongly to weekly storage reports documenting a rise or fall in US crude inventory levels, or drop in rig counts. A change of a few million barrels (out of total US inventories of 500 million barrels) or a few rigs (out of a total of 500) can send oil prices up or down 5 percent.
Meanwhile, on the other side of the world, there are OPEC countries that produce more than thirty percent of global production with the majority of market participants having no clue what’s going on within their borders. Some countries’ production levels can only be estimated in the short term counting the number of ships loaded at their ports.
How much oil does Iraq produce? Iran? Libya? Afghanistan? Russia? China? Even in North America, a jurisdiction with some of the most transparent measurement systems in the world, market analysts have trouble figuring out exact production numbers. Yet somehow, production estimates seem to be taken as gospel from regions where entrepreneurial types riddle oil pipelines with holes in the dead of night (hello Nigeria) or clandestinely load jalopies from half-bombed meter-less loading facilities (any war zone in the Middle East, i.e., much of it). Of this collection, even where there are functioning governments, what are the odds that a truly accurate physical count is being reported? And what about corrupt or lazy reporting – how much oil is stolen globally and goes through unofficial reporting channels? How much oil is never produced at all, yet is reported as part of government efforts to “stabilize” trends, or paint a certain desired picture for the world’s media?
These aren’t crazy questions. Even the International Energy Agency, a global powerhouse in forecasting, has to plug its numbers to make everything work. The IEA includes a “miscellaneous to balance” line item in its monthly Oil Market Report. This is merely a sophisticated variant of simply shrugging and saying “beats me.” Recently, the miscellaneous to balance number was 800,000 barrels per day, which is an enormous error bar in a market that is supposedly over supplied by a million barrels per day.
The problem isn’t just that it’s hard to make the numbers balance. The real problem is that there is no real concrete source for many of the numbers. From the same glossary under OPEC crude production: “Estimates of OPEC crude production are based on information from a wide range of sources with tanker tracking information being particularly useful.”
Compare that level of guesswork with this bizarrely in-depth analysis (in the comments section, no less) of a Texas monthly industry publication. These are amateur analysts meticulously going through well-level production data for inconsistencies, and finding some (imagine what the pros are doing). This is taking place, mind you, in one of the most open and dedicated reporting systems in the world.
Note the difference: a bunch of well-meaning zealots poring over detailed well-level information has trouble pinning down exactly how much oil is produced in Texas, while 30 million barrels per day of OPEC production is so opaque that tanker tracking information is “particularly useful.” That’s like saying the number of Spanish speaking people going through a McDonalds drive-through in Palm Springs is particularly useful in estimating the number of illegal immigrants in the US.
We should train ourselves to become much more discerning in our evaluation of sensational headlines that move markets, particularly in the context of the big global numbers. It takes no more than a few seconds to find an example such as this story about how prices jumped because crude stock builds were 700,000 barrels less than expected (or 100,000 barrels per day, a number 1/8th the size of the IEA plug number).
In a world of imperfect information, it is perfectly acceptable and even a good thing to provide estimates of these yardsticks. They are extremely useful for macro planning and international scheming. But to pretend that we can glean anything useful out of minute changes is of no value to anyone except volatility-loving traders who like to stir up markets.
Read more insightful analysis from Terry Etam here