Walking Down Spreads the Line

Dr. Ken Rietz

April 11, 2024

We continue with our monthly spreads issues. This time, we circle back to the crack spread and look at it more closely. We can learn about the value of a refinery, and the cautions that need to be observed. Also, we can get some indication of the demand (or lack of it) from the graphs. Finally, we give tentative thoughts about the short-term direction of the crack spread. The first order of business is the graph of the crack spread on NYMEX WTI crude oil.

Figure 1: The NYMEX WTI crack spread for May, in USD per barrel

The crack spread is evaluated by using the price of one barrel of WTI crude oil as the cost, and (using the standard assumption of 3:2:1) the price of two-thirds of a barrel (or 28 gallons) of gasoline and one-third of a barrel (or 14 gallons) of heating oil (essentially the same as diesel fuel) as the revenue. You can see that calculation in the header of the graph. This gives the gross profits for the refinery. Note also that each of the inputs to the crack spread have the same expiration month, in this case, May. This is necessary for the spread to make sense and will be used later in the commentary.

The value of the crack spread is used by the stock market to estimate the value of a company. Last Friday, 5 April 2024, an article in Seeking Alpha promoted a refinery specifically because of how healthy its crack spread looked. Two days later, 7 April 2024, another article in Seeking Alpha downgraded the same company from buy to neutral, in part because of the crack spread combined with other measurements. There are two lessons here. First, crack spreads are powerful indicators of the financial health of refineries. Second, crack spreads are not sufficiently powerful to be used alone; they must be used in combination with a range of other indicators and metrics to be accurate enough.

But there are other pieces of information that can be seen in the crack spread. It can be an indication of demand for gasoline and heating oil. For this, we need to understand the graph that FA produces. There are multiple colored curves in the graph, each color coded to a year at the bottom. The curve on the graph with the corresponding color represents the crack spread ending in May (in this case) of that year. (I deleted the graph for May 2020, because the anomalous data in April of that year, when the price of crude oil went negative one day, causes the graph to spike so much, scaling that data to fit the spike squashed the graphs of interest.) Note that the curve for previous years goes all the way to the expiration date, since we have that data, but not for the current or future years, since that data has not yet been settled. So, you can see how the demand for gasoline, and therefore its price, dropped dramatically during 2021 (yellow curve) during the COVID-19 lockdown, giving the drop in the crack spread for that year. Similarly, you can see the rebound in demand (and price) for the summer of 2022 (blue curve) as people were getting eager to get away from home.

That observation leads to some indication of the direction for the crack spread, based on the demand for heating oil and gasoline the past several months. The weather in the northeastern portion of the US was unseasonably cold during mid-January 2024, setting numerous records. Since the northeast is exceptionally dependent on heating oil, according to the EIA, a burst of demand occurred. You can see the upward blip on the graph that ensued in mid-January. A second burst in demand, this time for gasoline, has also just occurred. The total eclipse of the sun drew a very large number of people to a small band across the country on Monday. Many of the total eclipse regions declared states of emergency in advance, anticipating overwhelming numbers of visitors. This event will probably affect the total crude oil demand and increase prices some. So, I would not be surprised if the crack spread went up in the short term.