Thursday, March 5, 2009 3 comments ++[ CLICK TO COMMENT ]++

A hint on the marginal cost of production for oil

One of the big "mysteries" in the oil market right now is the marginal cost of production. According to some theories that I follow, commodities should trade near their marginal cost of production in the long run. Anything can happen in the short run but the prices should be slightly above the marginal cost of production.

To figure out if a commodity is irrationally low, one can look at marginal cost of production for an idea. Prices can drop below but it will bankrupt some producers and when that happens, it will rise back up to the marginal cost. This thinking doesn't take potential increases in demand or decline in supply but it does give a very conservative estimate.

So is the current oil price below the marginal cost of production? If so, it would imply that it is undervalued and will rise for certain--although timing will be uncertain.

Well there was a hint of the cost of oil production, at least in the Canadian oil sands. I ran across an obscure story from the Globe & Mail of some Americans complaining that Canadian oil companies are dumping oil in their market. They charge that the oil is being sold below cost in America. The relevant point for our discussion is the response by the Canadian Association of Petroleum Producers, the industry lobby group, claiming that the cost of production in the oil sands does not exceed $35:

The prospect of a U.S. state investigating Canadian oil dumping raised the ire of producers in Alberta and prompted the Canadian Association of Petroleum Producers to bring out numbers showing that the current cost of producing oil sands crude does not exceed $35 a barrel, which is below current oil prices.

Assuming CAPP is correct with their numbers, it goes to show what the marginal cost of production for oil may be. One often hears that the oil sands, which is supposed to be high cost, has a cost of production of $80. Well, it clearly isn't $80 (it was around that figure a few years ago due to high input costs, labour, etc but that's not the case right now.) One could argue that $80 is required to significantly increase oil sands production but that has nothing to do with the marginal cost of production.

None of this is to say that oil won't go to $150; all it implies is that a conservative investor should value oil closer to $35 than to $80 or $150.


3 Response to A hint on the marginal cost of production for oil

March 6, 2009 at 10:33 AM

I don't have access to that Merrill Lynch report but my guess is that costs are closer to $30 or $40 than $80. I'm just guessing here but here is why...
I remember 3 or 4 years ago, when oil was closer to $40 or $50 or whatever it was, oil sands producers said their cost was in the high 20's or 30's. I don't remember the exact numbers but it was something like that. Suncor, the pure play oil sands company, was actually making a profit (although its free cash flow was negative since capex was high.)
But I also remember reading last year that oil sands companies have costs over $70. What happened is that input costs, labour costs, etc skyrocketed in the last few years. Even raw land leases with no proof of oil skyrocketed to high levels.
I think if you assume that the high input costs, labour, etc will stay high, then oil sands likely has a cost closer to $80 per barrel. I suspect that's what the Merrill Lynch report and various other analysts are assuming. Analysts in any industry tend to lag the reality.
However, if you assume that commodity prices of inputs (steel, aluminum, natural gas, etc,) or labour costs or whatever else, will not go back to what they were last year, I think the marginal cost will be closer to $30 or $40.
So, it's correct to say that no new projects will be undertaken. This makes sense since there is a glut of oil supply and existing projects were undertaken with much higher projections. But this argument doesn't prove that the marginal cost of production for oil is $80.

March 6, 2009 at 10:24 PM

DOes it include Finding and Developmentscost, or is it just cash cost. They could be substantially different as it takes a lot of capital to develop oil producing facility in the tar sands.

March 7, 2009 at 3:36 PM

Good question and I don't know the answer. I'm not sure what the cited figures are. It's also never clear to me if the costs ever include a minimum return on investment that investors need. Analyst reports generaly detail what their numbers represent but the figures commonly cited in the media or in investing articles are never clear and I have no idea. Even the $80 that is quoted is not clear.
Anyway, oil sands tend to have very low finding costs but they have very high capex costs so I would imagine that a decline in steel or natural gas (a big input for in-situ oil extraction) will have a bigger impact.

Post a Comment