What China Knows About The Coronavirus That Oil Traders Don't Know

Yesterday, world-wide, one hundred-times more people died from malaria than died from COVID-19/coronavirus. But the economic cost from contagion is not human suffering; it is the restrictions on travel put in place to contain it.

Over nine months SARS infected 8,098 people, 774 died (9.6% fatality). Over the first few months of the COVID-19 contagion, 78,635 people were infected and 2,459 died (3.1% fatality). IEA is predicting a collapse in demand for oil.

What’s the cure?

Easy; stop flying, and don’t take public transport. Drive, with the windows closed and the ventilation on re-cycle. Don’t; whatever you do, book your dream cruise; and above all, stay out of jail.

I looked it up on Wikipedia. Apparently, in 2010, flying burned 2,691 British Thermal Units per passenger-mile, driving burned 4,218; 43% more.

This is what happened in China during the SARS travel lockdown which was in place for all of 2003 and 2004.

(Click on image to enlarge)

No wonder, today, when the whole world is cutting back oil deliveries, China has started to stockpile.


  • There was a slowing in GDP growth from 2003 to 2004 (full year). But that was not “statistically significant” comparing to a dummy variable SARS-lockdown = 1, not = 0; either for annual data, or quarterly data (Prob>[t]=0.59).
  • There was a huge jump in oil consumption, 11% year-on-year in 2003 and 16.5% in 2004: 30% over two years. That was borderline “statistically significant” for SARS-lockdown (1, 0) as an explanatory variable (Prob>[t]=0.0634).
  • The reason for the huge growth, even when GDP growth was flat, was a massive drop in GDP-per-Barrel of oil consumed, by 1% in 2003 (from a trend of plus 3% per year), and by 5.5% in 2004, that was highly “statistically significant” (Prob>[t]<0.0001*).
  • GDP per barrel is a measure of how efficiently oil is used. The travel disruptions compromised the efficiency of how oil was used in China to move people around.

An aside, the words “statistically significant” are, technically, meaningless; but everyone uses them. The correct wording is “null hypothesis, the probability that the drop in GDP-per-barrel in 2003 and 2004 was not linked to SARS is less than 10,000 to one”, which is a bit of a mouthful.

The analysis does not prove SARS, or more precisely, the travel-lockdowns to contain it, caused the drop in GDP/bbl, it simply says that the probability it wasn’t connected, is infinitesimally small.

Looking at the numbers a different way:

(Click on image to enlarge)

I never saw that chart before; I’ve been meaning to plot the numbers out for ten years. Glad I did, finally.

I have a theory that the starting point for most economic activity is burning oil, then there is trickle-down, so GDP/bbl is important for figuring the value of oil (another subject). Put that another way, imagine if there was no oil to burn, we would be heading back to live in caves.

What surprised me was how straight the lines are, and the steady growth; these days we get about double the amount of economic benefit from slowly destroying the planet, as we did forty-years-ago.

Another thing is that price doesn’t seem to affect the trajectory of the lines.

1 2 3 4
View single page >> |

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. 

The above is a matter of opinion provided ...

How did you like this article? Let us know so we can better customize your reading experience.


Leave a comment to automatically be entered into our contest to win a free Echo Show.