The note is entitled “Independent Refiners: The Crack Ate My RINs – – Policy and Profit Implications”. The first sentence reads, “In what may be a surprise to some, most Independent Refiners now enjoy a net benefit from Renewable Identification Numbers (RINs), based on our analysis.” Needless to say, I found this statement surprising and, after reflection, believe it is wrong.
The note identifies three independent/merchant refiner trends relevant to the RFS/RIN/point-of-obligation discussion. One is the increasing “build out” of blending facilities, i.e., vertical integration, by independent and merchant refiners. The note states, “The increase in integration reflects how the RINs market has incentivized adding blending capacity (i.e., capture the RINs value / minimize RINs costs) since 2013.” It would appear Sir Bacon was clairvoyant: if the point of obligation will not move to the blending rack, the blending rack must move to the point of obligation. The note observes this trend is happening but offers no detail on its size or strength. The “build out” may very well be occurring with refiners Wells Fargo covers, the subject of the note. I do not believe, however, significant increased vertical integration is happening everywhere in the refining industry. If it were, the second trend would not exist.
The second trend is increased pass through of refinery RIN expenses in bulk fuel prices. The note identifies in some detail a correlation between spot/bulk/wholesale prices and RIN prices: “Significant changes in RINs prices were directionally matched to a meaningful portion of the changes in Gulf Coast gasoline prices.” This relationship was not seen in 2013. The note correlated the direction of price changes but not the magnitude. Without the latter, it is difficult to say just how much of refiners’ RIN expenses are really passed through to spot/bulk and wholesale prices. Nevertheless, a directional correlation previously absent now exists. The crack spread, i.e., the gross margin between the purchase cost of a barrel of crude oil and the spot price of a barrel of refinery product, has indeed eaten some, but probably not all in my view, of refiners’ RIN costs.
The note’s third trend is, “[B]ottom line performance appears positive for most of the Independent Refiners across our coverage universe as the vast majority of the cost of RINs is embedded in the crack spread. Consumers now bear the majority of RINs costs – like a tax.” (My underlining.) I believe this observation is an incorrect interpretation of the note’s own data and analysis. While it appears the crack spread is indeed covering part of refiners’ RIN expenses and that crack spread is passed through to consumers, it is not happening to the extent displayed in the note’s tabular data.
The remainder of this post is my explanation of why a significant first trend cannot coexist with the second trend, why the second trend is occurring and why the note’s analysis does not show, “Refiners now enjoy a net benefit from Renewable Identification Numbers (RINs) . . . .”
Significant vertical integration “build out” is inconsistent with greater RIN pass through. According to the note, refiners are “building out” more wholesale, i.e., blending, infrastructure. Much of this new infrastructure will appear at pipeline terminals where wholesale operations already exist and not at upstream refineries since gasoline containing ethanol cannot be shipped by pipeline. As a result, new RIN sales at downstream marketing locations will offset more upstream refinery RIN expenses. Refiners will sell more RINs, buy fewer of them and more “free RIN” blend stock will enter the bulk/spot fuel market. See my November 19, 2017 post, “Big Oil’s Free RINs – The Root Cause”.
There cannot be more “free RIN” blend stock in the bulk/spot market and, at the same time, greater bulk and wholesale response to RIN prices. More blend stock free of RIN expenses should dampen RIN influence on bulk and wholesale prices, not amplify it. Something else must be going on here. Perhaps, one of Wells Fargo’s hypotheses, i.e., RINs in the crack spread or expanded blending and wholesale activity, is overstated or wrong.
Why are RINs showing up in bulk prices and the crack spread? Well, if there is anything better for a refiner than free RINs, it just might be no RINs at all. Exports are exempt from the RFS and have zero RIN obligation.
Big Oil owns about 45% of Gulf Coast refining capacity with merchants and small refiners owning the rest. The Gulf Coast, one of the world’s major blend stock exporting regions, also supplies and competes in domestic markets. Combined, Big Oil and merchants over supply those markets as demonstrated by Gulf Coast exports. Separately, however, neither can totally capture those markets. Both are necessary to fill demand, but neither is sufficient. As a result, both compete in domestic markets.
What if, rather than merchants significantly expanding wholesale blending infrastructure, Big Oil exported more of its “free RIN” blend stock? In the domestic bulk market, “free RIN” blend stock would have less market share and pricing power while merchants’ “paid RIN” blend stock would have more.
To be sure, exporting offers several benefits for Big Oil. Exports shrink the RFS compliance obligation but not marketing operations. Big Oil exporters would purchase fewer RINs at the refinery and sell the same number at the wholesale rack. In addition, a larger market share for merchants means a larger RFS obligation. It seems to be the ideal situation for Big Oil: shift the compliance obligation to the competition but sell the same number of RINs. “Paid RIN” blend stock’s resulting larger share of domestic spot markets will increase RIN demand, push RIN prices up and benefit all marketing operations. Big Oil’s one offsetting risk is a greater pass through of refinery RIN expenses increasing marketing’s supply costs.
One thing is clear: Gulf Coast exports of blend stock are increasing significantly. I strongly suspect Big Oil is behind this trend. Its refineries are often located close to maritime transport, and they produce volumes by the boat load. The following chart was developed from Energy Information Administration (EIA) data available online. Exports’ share of Gulf Coast production increased from 11.2% in 2013 to 16.3% through Q3 of 2017 – a 46% increase. By absolute volume, exports increased 55% between 2013 and the end of 2016. (Full year 2017 data are not yet published.)
One thing, however, is certain. Increased pass through is not the result of refiners “building out” more marketing and blending infrastructure. The note does not argue such although it might be inferred. Such “build out” introduces more “free RIN” blend stock into the market and inhibits rather than promotes RIN pass through. The more likely cause is blend stock export growth that partially abandons the domestic market giving “paid RIN” blend stock more market share and pricing power. While it is not certain just who is exporting, Big Oil has “free RIN” blend stock and is in the best position with significant incentive to do so.
A majority is not 100% of RIN costs. The note includes this pro forma analysis of three refineries blending BOB (“blend stock for oxygenate blending”) with ethanol at 105%, 65% and 15% of refinery production. I have annotated the table to aid in interpretation.
Note the “Gross margin, net RINs” is identical for all refiners. Refiners are differentiated only by the extent of their wholesale or blending operations. For each refiner, the RIN expense is $991,667 per day leaving a gross margin, net of RINs, of $2,976,249 per day. Adding to this margin is the offset or “Wholesale/retail recapture” of selling RINs at the downstream marketing operation. This offset increases as the blending assumption increases from 15% to 105% of production as do the all-in “Gross margin, adjusted net” and the “Gross margin expansion vs. 2013”. The table then displays this margin increase as passed on to consumers. For example, the 105% integrated refiner margin increase of $823,667 divided by 10,500,000 gallons is presented as a gain of 0.078 ¢/gal obtained from the consumer.
This conclusion about increased margins coming from consumers is wrong. The data in the table demonstrate it is wrong. In another table not reproduced here, the note assumes a refiner’s gross margin in 2013, when RINs were 59¢ and pass through in spot prices was assumed to be 10%, was $2,505,499 for producing 10,500,000 gallons per day. The gross margin in 2017, when RINs were 85¢ and pass through is assumed to be 85%, is $2,976,249 per day, an increase of $470,750. This increase is indeed passed through to consumers; if the refiner is receiving this increase through crack spread and spot prices, the increase is paid by the wholesale marketer and will be passed on to the consumer.
The hypothetical refiner, however, also expands his marketing operations from 100% to 105% of refinery production. So, instead of selling 1,666,667 RINs per day at 59¢ ($688,333), the refiner sells 1,225,000 RINs per day at 85¢ ($1,041,250). This is a gain of $352,917 per day from selling more RINs at a higher price. The total gain from an increased crack spread and increased RIN sales is $470,750 + $352,917 or $823,667 per day as displayed in the table above.
Only part of that increase, $450,750, is paid by the consumer due to increased crack spreads and spot market prices. The remainder of $352,917 originates with RIN sales to other refiners and is paid by the consumer only if it shows up in bulk market prices for all refiners. The consumer is paying twice for RINs if he/she pays for a higher crack spread and, also, is paying for RINs sold by wholesale marketers to refiners. Simply put, refiners, not consumers, are buying RINs. Consumers will pay what refiners can pass on, no more and no less.
So, using the note’s assumptions, the bottom line is consumers are paying about 57% of the refiner’s margin expansion. 43% is borne by refiners buying RINs and not recouping all of that cost in their crack spread. Whether the assumed 85% of RIN expense is truly in that crack spread is really unknown. The note found a directional correlation between RIN and bulk fuel price movements. The magnitude of those movements was not correlated. Even so, if a merchant refiner is recovering the assumed 85% of RIN expenses and eating the other 15%, that is a 1.4 ¢/gal cost the merchant bears but Big Oil does not. At 10,500,000 gallons per day, it starts to look like real money. Finally, please observe from the table’s right column, a refiner blending 15% of its production and, presumably, those blending less, did worse in 2017 than in 2013. See my November 27, 2017 post, “Let Them Eat Cake” to understand why these refiners do not blend more. Clearly, RINs are not a “net benefit” to independent refiners.
Bob
A closing note. I communicated with Wells Fargo Securities to obtain some clarity regarding my questions on the note. Several emails were exchanged in both directions. However, there has been no response to my last two emails, dated December 19, 2017 and January 17, 2018.