The Barrel, Platts' oil blog on platts.com, last week published two recent entries discussing the level of the highest real-time price of crude. This subject has been under much discussion and speculation in recent weeks as the current price of WTI touched $90/b.
At least one newspaper, Newsday, said the highest real-time price ever had been crossed. The Wall Street Journal declared that current prices would need to be in excess of $101 to set a new record.
Below are excerpts from Platts' analysis of "the highest ever." We've tweaked our analysis from what appeared originally.
What (is wanted by the media) is a precise apples-to-apples comparison of the current NYMEX price to an inflation-adjusted price for the first three or four months of 1980 or 1981, when it is widely agreed that real crude prices reached their all-time high.
The problem, of course, is that you can't easily make that comparison. Crude prices were deregulated in steps in the late 1970s, and the final shackles came off in the first few days of Ronald Reagan's presidency, in January 1981.
So there were no daily spot crude prices until both the launch of the NYMEX and the time at which pricing services such as Platts began assessing them. At Platts, the spot price of WTI was launched with the first trading day of 1984, and we assessed $29.60-$29.70. That's not the inflation-adjusted high.
Platts' crude data prior to that consisted of dutifully recording crude oil postings, which are the prices that a company will pay at the wellhead for a certain type of crude. But the WTI posting is not identical to the spot price of WTI. There are costs involved in gathering from the field and transporting it to a delivery point such as Cushing. So we've found that some media that have contacted Platts want to take the posted price, adjust it for inflation, and declare that to be the highest inflation-adjusted price ever. That's too simplistic.
So the approach we will take here is somewhat different. The highest crude posting we can find during the period in question was a $39.50/b price in effect during March and April of 1980. So we need to calculate what would be the spot price for WTI at Cushing if all we know are two things: the structure of today's market and the posted price for WTI purchased at the wellhead back in 1980.
The structure of today's market changes every day, which makes our number somewhat of a moving target. But here's the most recent approach we've taken.
We'll use October 18 prices, since the first month-second relationship Oct. 19 was skewed by the fact that the front month expired that day.
On October 18, front-month WTI settled at $89.47/b. But crude oil postings are not pegged off front month prices. They are pegged off second month prices, since crude lifted at the wellhead today or tomorrow will not make it to Cushing until a month hence, which is why the second month price is more relevant.
On October 18, the second-month settlement was $88.04, and at the end of the day, after the day's changes were taken into account, postings stood at $86-$86.25. Postings vary significantly, because of different terms and conditions placed by different companies, but we’ll use the higher of the $86-$86.25 range, which were the most frequent postings for the day.
So the second month spot price is approximately 98.4% of the first month, and the posting was 98% of the second month.
We are going to make two assumptions for our model. The first, which is a bit of a leap, is that the market in April 1980 was in backwardation. This is a leap because there was not much of a spot market then, much less a curve. But given that inventories then were tight, resulting in the highest real-time price era ever, we will assume a backwardation.
If your preference is not to assume a backwardation -- and there are people here at Platts who believe a backwardation should not be assumed -- you can do this exercise just based on the posting, and assume a relationship between the posting and the Cushing price. What you can't do is take the posted price and simply assume that as a Cushing price; it ignores the costs of getting from the wellhead to Cushing.
That's our second assumption: that the relationship between postings and spot prices today are the same as in 1980, which is also a leap in logic. Transportation and logistics in almost everything is more efficient now than almost 28 years ago. But we'll go with the current relationship, conceding its imperfections.
So going back to 1980, assuming these relationships, a posting of $39.50 would equal a second month price of $40.32
And if the assumed second month of $40.32 is 98.4% of the first month, the first month price back in 1980 would be $40.97.
Adjusting for inflation, that $40.97 price is $103.69/b.
There are old news stories floating around in various archives, reporting spot transactions from 1981 at levels as high as $42. But we'll go with the highest undisputed published number on record, and that's the WTI posting number for March and April of 1980. And by our calculations, we've still got a ways to go to surpass it.

John Kingston offers a great analysis that answers lots of questions about the all-time, inflation adjusted price of crude. thanks John.
OK, Ben, we at Platts take back everything nasty we ever said about OPIS.
Thanks John for such a thorough analysis. What about international oil price benchmarks? I notice that the BP Statistical Workbook uses Arabian Light to 1983 and this reached a peak of US$36.83 per barrell (annual average) in 1980 or US$87.65 in 2004 prices. Would Arabian Light in 2007 be comparable with Arabian Light in 1980?
I think it would, but the problem with any of the Saudi crudes is that they were never allowed to be traded freely in the open market, not then, and not now. So it's not realistic to look upon them as a benchmark of a freely-operating market. You buy it directly from the Saudis, and you aren't allowed to resell it.