Oil and gold looking a little bit more "normal" as 2017 draws to a close

A few weeks ago in London, at a pre-conference dinner before the Platts Digital Commodities Summit, the conversation naturally turned to bitcoin and other cryptocurrencies, and the rejection - and disdain - that many supporters of the entire crypto movement have for what are known as "fiat currencies." Most people know fiat currencies as everything from afghanis (Afghanistan) to zlotys (Poland), with dollars, yen and euros in-between.

Crypto currencies have been likened to gold: not controlled by a central government, a supply that rises only incrementally (and in the case of bitcoin, will eventually be capped), and therefore not inflationary. One of the conversation’s participants in the restaurant’s cigar bar, a backer of bitcoin, said to another person: "Look at that suit you’re wearing. I’ll bet you 200 years ago, if you had bought the equivalent suit and paid for it in gold, it would cost you the same amount in gold then as it would today." Gold isn’t volatile against the dollar, he said. The dollar is volatile against gold.

S&P Global Platts has a price database of many things. It doesn’t have one of men’s suits, so it is tough to confirm the relationship. But every year at this time, we take a look at the price of oil through the prism of how much gold it would take to buy a barrel of oil.

Last year, we suggested that changes in the fundamentals of oil may have moved the normal relationship between the two commodities to a higher number. Our 34-year comparison of gold and oil — which commences in 1984 with WTI, as that data goes back further than that of Brent — averaged 16.65 through 2016. That is somewhat higher than the conventional wisdom that the ratio of oil to gold—the number of barrels of oil needed to buy an ounce of gold—has a “norm” of 15. From the start of 2003, the average is much closer to 15, standing at 15.41.

The problem is that within that norm, there are vast swings. Our data shows the ratio dropping to as low as 7 (2008 oil price surge), and as high as 40 (oil price crash, February 2016).

Given that in the price crash of ’98-’99, gold was crashing right along with oil and the ratio never got much above 20, could it be that the significant change in drilling technologies over the past few years means that the 15 number is no longer valid, as evidenced by it blowing out to 40 at its most extreme? We suggested that after the average for 2016 hit 29.54, more than 5 “points” above the 2015 level. The writer of last year’s blog — OK, it was me — declared that it was possible that the shale revolution had moved oil to a permanent wider discount to gold.

There has been no hint of that “normal” ratio since the price of oil began falling in 2014. The average ratio in 2015 was 24.18; in 2016, it was 29.54. The average for 2017 through December 14 is slightly more than 25, so not that much different than it was two years ago. Halfway through the year, the six-month average was just less than 25.

It is true as the year comes to a close that the ratio has moved closer toward that theoretical norm, but it still has a ways to go. Even with the price of WTI tacking on $12 from its midyear level, the ratio at the end of the year was just under 22. Part of the reason it didn’t fall more is that gold has risen from the middle of the year as well, adding about $30 to a level of about $1,250 on December 14.

Even if gold stayed flat and WTI went to $70, the ratio would decline to only a bit under 18. And such a situation is almost impossible to fathom; if oil were to go up $15 from here, gold is not just going to sit unchanged.

It’s clear that the blowout numbers of 2016 in this ratio were an aberration. It also seems true that there are going to need to be some big unexpected shift in oil markets, financial markets, or both, to get this spread down toward normal. And maybe that means there’s a new normal.