Gasoline Sulfur and Benzene
In 2006, EPA’s Tier 2 rule limited gasoline sulfur to 30 parts per million. One pound of gasoline could not contain more than 30 millionths of a pound of sulfur. In 2017, the Tier 3 rule lowered the standard to 10 ppm for all but the smallest refineries, which must also reach 10 ppm by 2020. The trading arrangement for gasoline sulfur credits is straightforward. If gasoline contains less sulfur than the standard, credits are earned. If more than the standard, credits must be purchased. A refiner producing one billion gallons of 11 ppm gasoline must pay fines or purchase and retire one billion ppm-gallons of sulfur credits (1 ppm excess sulfur times one billion gallons of gasoline). Likewise, a refiner producing two billion gallons of 8 ppm gasoline may sell or bank four billion ppm-gallons of sulfur credits (2 ppm below the standard times two billion gallons).
In 2011, the gasoline benzene standard was set at 0.62 volume per cent. That is, every 100 gallons of gasoline may not contain more than 0.62 gallons of benzene on average. For example, one million gallons of gasoline may contain 6,200 gallons of benzene. A refiner producing one million gallons of gasoline containing 6,100 gallons of benzene earns 100 gallons of benzene reduction credits. Likewise, for 1,000,000 gallons containing 6,300 gallons of benzene, 100 gallons too much, the refiner must pay a fine or purchase 100 gallons of credits.
As they matured, the sulfur and benzene programs succeeded in reducing both pollution and its control costs. Tier 2 gasoline sulfur credits first traded in 2006 in the range of 150 $/Mppm-gals. As controls came on line, prices drifted lower to a range of 15 – 30 $/Mppm-gals in 2011 and 2012. As the much stricter Tier 3 standards phased in, Tier 2 credits, valid for early Tier 3 compliance, rose to about $150. Recently, Clear Energy Brokerage reported a December 18, 2017 credit trade at $2,000 for final Tier 3 credits, reflecting the tougher 10 ppm standard and the phasing out of Tier 2 credits. We shall see whether Tier 3 prices drop. I believe they will as credit prices have in the past. Benzene credits are now trading in the 25¢ to 30¢ per gallon range after starting out near $1.70 in 2012 and peaking around $3.10 in 2013. The following charts track the history of the sulfur and benzene credit prices through early May 2017.
These two patterns are similar. Initially, a low price holds until the compliance deadline nears. The price then peaks and falls as more control equipment comes online. During this industry buildout, facilities with control equipment will sell surplus credits for less than the total cost of producing them. The total cost involves both capital and operating components. Selling surplus credits for more than the marginal operating cost provides additional revenue for capital recovery. The credit producer does not need to recover its full cost to benefit from surplus credits. As more control equipment is built, more sellers emerge, and the price drifts lower.
The equipment required to comply with these two standards is not mass manufactured like automobiles or even airplanes. For example, high pressure steel reactor vessels with walls a foot or more thick are sometimes needed. Often, there are only a handful of qualified manufacturers worldwide. In this situation, converting all refineries to lower sulfur or benzene standards over a short period of time is not possible. Credit trading allows facilities with the largest projects tying up equipment vendors and contractors for several years to sell credits to smaller refiners waiting for vendors and contractors to be available. The net result is the environmental goal was accomplished, and all refiners were generally able to comply on schedule at the lowest overall cost.
Compare the relatively smooth tracks of these sulfur and benzene credit price charts with ethanol RIN prices in the chart below. The sulfur and benzene charts appear to reflect steady and orderly progress driven by the compliance obligation and competition to supply credits at an acceptable price. Ethanol RIN prices, on the other hand, have jumped all over the board. I suspect this erratic price behavior is caused by RIN suppliers having no obligation to comply and a lack of supplier competition. Buyers are under legal duress to comply and have few alternatives for acquiring RINs other than buying.
What criteria make the gasoline benzene and sulfur programs so successful?
First, compliance credits are created on the basis of measurable characteristics. If the amount of sulfur or benzene in gasoline is not measured, the number of credits for reducing sulfur and benzene cannot be determined, and credits cannot be traded. In short, the credits must be fungible, acceptable to all market participants.
Second, the standard is framed as a concentration, ratio or factor, which, when multiplied times an activity, results in a quantity limit. Both the standard and the activity must be applicable to all regulated parties. For example, a standard of 30 parts per million pounds of gasoline would be meaningless for an electric power plant. For sulfur and benzene, the quantity allowed and credits sold and bought are determined by a standard relevant to gasoline and are proportional to the gasoline volume.
Third, all buyers are similarly situated to each other as are all sellers. No refiner buying credits is competing with another receiving free credits. Every seller sells credits earned only by removing more sulfur and benzene than the law requires. No seller may sell credits for removing merely what the law requires, let alone something less.
What’s missing in the RFS credit trading program?
RFS misses the mark and fails miserably on the third criteria. On the first, the gallons of biofuel blended and gasoline produced are easily measured as are credits. On the second, EPA’s national percentage standards are framed as a percent of refinery production, a concept applicable to all obligated parties.
On the third, however, all RIN buyers are not on the same footing. For some refiners, ethanol RINs are free. For others, they are not. See, “Big Oil’s Free RINs – The Root Cause”. For that matter, all RIN sellers are dissimilarly situated, too. Every exempt marketer/blender may sell RINs just for blending anything, even if the corn ethanol standard is never met. A refiner sells only surplus RINs because compliance comes first. With the competition among buyers and among sellers so corrupted, the problems between buyers and sellers have become unmanageable.
The next chart illustrates the tilted playing field. Prior to 2013, the RFS implied E-Value and actual blending results were in line with each other and both were below the E10 or 10% ethanol blendwall. I calculated the E-Value from EPA’s mandated annual volumes. For example, the conventional biofuel or corn ethanol mandate for 2018 is 15 billion gallons. The projected total gasoline consumption is 143.22 billion gallons. 15 ÷ 143.22 = 10.47% or E10.47. What blenders actually did was calculated from EIA tables for “U.S. Refinery and Blender Net Input of Fuel Ethanol (Thousand Barrels)” and “U.S. Product Supplied of Finished Motor Gasoline (Thousand Barrels)”. OPIS RIN prices were obtained from Weaver LLC before I retired and have not been kept current.
In 2013, EPA increased the implied E-Value above the blendwall from E9.76 to E10.40. Blender response was inadequate, and refiners did not or could not install more infrastructure to increase ethanol blending in time for compliance. Average ethanol blending stayed at about 9.62% even though the standard was higher. RINs were supplied at the rate of 9.62 for every 100 gallons of gasoline while 10.40 RINs were required for industry-wide compliance. There was a severe RIN shortage, and prices increased by 2,000 % or more as refiners bid up prices attempting to comply.
This conventional biofuel gap continues to this day. For 2018, EPA projects 14.71 billion gallons of conventional renewable fuel or ethanol will be blended into gasoline, but the implied volume mandate is 15 billion gallons, creating a shortfall of 290 million gallons and RINs. On the other hand, the projected biodiesel volume is 2.53 billion gallons while the mandated volume is 2.10 billion gallons creating a 430 million gallon surplus. At 1½ ethanol RINs for every biodiesel gallon, biodiesel blending can generate the equivalent of an extra 645 million ethanol RINs, more than enough to cover the projected conventional biofuel blending gap. Thus, although EPA annually raises the standard in the face of the ethanol blendwall, the gap is filled with RINs made available through a biodiesel volume mandate set below actual blending projections.
From EPA’s viewpoint, this strategy makes sense. Setting corn ethanol above projections at 15 billion gallons complies with the statutory volume requirements. It has the appearance, perhaps more, of encouraging higher ethanol blending rates. For blenders, it’s a perfect world. Blending below the standard allows blenders to eschew E85 and sell popular E10, creates a conventional RIN shortage, keeps conventional RIN prices at the higher biodiesel RIN levels and gives blenders a Goldilocks existence. Blending levels are not too high, and RIN prices are not too low. Everything is just right courtesy of the structurally tilted footing among RIN buyers, among RIN sellers and between sellers and buyers.
This state of affairs will not disappear until the 15 billion gallon corn ethanol statutory volume drops below 10% of consumption. That is, gasoline consumption needs to be 150 billion gallons or more. At that point, selling only E10 or less will be sufficient to consume 15 billion gallons of ethanol, the blendwall will no longer be an issue and the conventional biofuel compliance gap will disappear. Increasing petroleum fuel consumption to rescue a statute designed to displace petroleum fuels is more than a bit ironic.
The compliance gap, however, may never disappear. 150 billion gallons of annual gasoline consumption is, in oil terms, 9.78 million barrels per day (Mbpd). The Energy Information Administration’s 2018 Annual Energy Outlook predicts gasoline consumption through 2050 will peak at 9.36 Mbpd in 2018, this year. See next chart. The more recent Short Term Energy Outlook predicts 2019 gasoline consumption will be 9.39 Mbpd. Both projections fall short of consumption required to wedge 15 billion gallons of corn ethanol inside the blendwall. We may never outgrow the blendwall.
Using biodiesel RINs to fill the gap certainly cures ethanol shortfalls, but it would not be critical to RFS if rack blenders were obligated. As in the sulfur and benzene programs, credit sellers and buyers – blenders all – would weigh the relative costs of complying, over-complying and under-complying. RIN prices would rise until supply and demand converged making E85 economic for marketers and consumers. In fact, since refineries would not be obligated, bulk fuel prices would not reflect any part of the RIN value. 100% of every surplus RIN sale could be invested in E85 infrastructure upgrades or consumer discounts.
Credit markets will find ways to fill the conventional biofuel compliance gap in the most economic manner: sell E85 or use biodiesel RINs, one more option than exists now. Blenders’ reluctance to produce E85 would no longer leave ethanol and biodiesel RINs, both at biodiesel prices, as refiners’ only option for filling the conventional biofuel compliance gap.
The only way for RFS RIN trading to match the spectacular success of gasoline sulfur and benzene is to ensure equal footings for all RIN buyers, and further ensure buyers and sellers alike weigh compliance, over-compliance and under-compliance in their business decisions. The RFS obligation and the ability to comply must converge by moving the point of obligation. After all, RFS is not a commandment to refine. It is a commandment to blend.
Best,
Bob