March was a challenging month for refiners. Three weeks ago in The Energy Letter I discussed the issue of the looming ethanol blend wall, which is leading to soaring costs for refiners as they attempt to comply with federal ethanol mandates. Refiners saw their share prices beaten up a bit as a result.

Then, at the end of March, came news that the US Environmental Protection Agency (EPA) is proposing to lower the limit on sulfur in gasoline from 30 parts per million (ppm) down to 10 ppm. The cost of complying with the new regulations has been estimated in the range of $10 billion in capital for adding new hydrotreater capacity to the refineries. Annual operating costs are expected to increase by $3.4 billion by 2030, according to the EPA.

The EPA sought to downplay the cost issue:

“The program would cost about a penny per gallon of gasoline, and about $130 per vehicle. The annual cost of the overall program in 2030 would be approximately $3.4  billion; however, EPA estimates that in 2030, the annual monetized health benefits of the proposed Tier 3 standards would be between $8 and $23 billion.”

The EPA cited the following justification for the tighter standards:

“The new standards would annually prevent:

• Between 820 and 2,400 premature deaths

• 3,200 hospital admissions and asthma-related emergency room visits

• 22,000 asthma exacerbations

• 23,000 upper and lower respiratory symptoms in children

• 1.8 million lost school days, work days and minor-restricted activities”

I won’t dissect the EPA’s estimates on the health benefits, because I simply don’t know if they are reasonable. But the agency’s estimate of a penny per gallon in increased costs isn’t realistic.

US demand for gasoline is now at 133 billion gallons per year. Some of that gasoline is imported, and some is already at the lower sulfur level. But even if you spread the EPA’s own estimate of $3.4 billion in new operating costs across all 133 billion gallons of US gasoline consumption, that alone comes to more than 2.5 cents per gallon. That is 2.5 times the EPA’s estimate of 1 cent per gallon in increased costs even before accounting for the additional capital expenditures and the other factors I discuss below. Even if we assume that US gasoline consumption again begins to grow, there is no way that $3.4 billion in new annual operating costs translates into only 1 cent per gallon.

I was working at the ConocoPhillips refinery in Billings, Montana when the ultra-low-sulfur-diesel (ULSD) specifications were enacted. Thus, I got a firsthand look at the situation before and after these regulations were implemented. Again, I am not arguing that the new specs weren’t needed, but instead simply making observations of what transpired. (For the record, I support tough environmental standards, but as an investor think it’s also important to properly calculate their costs.)

First, ConocoPhillips invested a great deal of capital across the company to make sure that we could comply with the new regulations — as did many other refiners. I am not sure if the amount was made public, but these hydrotreater projects cost hundreds of millions of dollars. Following the conversion we experienced ongoing costs to the hydrotreaters that were required to reduce the sulfur content. These included maintenance expenses as well as the cost of the additional hydrogen.

Those costs are well-known, but the cost of ULSD after the new specs came into effect was higher than many anticipated. This is because there were at least three other impacts that weren’t widely anticipated.

First, the additional level of processing reduces the overall product yield. When hydrocarbons are fed to a hydrotreater some of what would have ended up as diesel is converted into light gases that end up being burned in the refinery as fuel gas. A lower liquid product yield from a barrel of oil will necessarily increase the cost of the product.

Second, these changes increase the complexity of the refinery, which is something I pointed out in a 2007 column. The more failure points within a refinery, the more potential for an outage. While I wouldn’t attribute it entirely to the new ULSD regulations, according to the Energy Information Administration’s US Percent Utilization of Refinery Operable Capacity, the average refinery utilization in the five years before the new regulations (2001-2005) was 91.9 percent. Starting in 2006 — when the regulations came into effect — refinery utilization never again reached 90 percent. Utilization declined for four straight years, averaging 86.6 percent in the five years beginning with the implementation of the ULSD specifications (2006-2010).

Finally, some refiners decided that it was more economical not to invest the money, and instead simply export the product into markets that could still accept the higher sulfur diesel. In the five years prior to the new regulations, exports of higher sulfur diesel averaged 43,000 barrels per day. By 2008 those exports had skyrocketed to nearly 500,000 barrels per day, and today remain about four times higher than they were prior to 2006.

These factors resulted in higher production costs and significantly higher diesel prices. Again it is instructive to look at the five-year time periods just before and after the regulations took effect. From 2001 to 2005, the average retail price of diesel was 4 cents per gallon cheaper than gasoline. From 2006 to 2010, that relationship was flipped as retail diesel was 17 cents per gallon more expensive than gasoline.

Whether the more restrictive sulfur specifications are money well spent is an entirely different issue, and it isn’t my intent to argue that these specs should not be implemented. However, the EPA’s estimate that the incremental cost added to a gallon of gasoline will only be a penny per gallon is not credible.

I don’t believe the new specs will have a huge long-term effect on refiners, but they will cut into profits. Companies will be forced to allocate capital, which will mean diverting it from other projects that could have added to profitability. In light of this, the increased cost of complying with the ethanol mandates, the recent softening in the price of gasoline (which has cut into margins), and the fact that refiners have had such a tremendous run over the past year, conservative investors would probably be wise to lighten up on the sector.

For those with a longer-term horizon, it is certainly frustrating to see the recent run of strong performance partially derailed by the government, but the refining sector will survive. If you are investing for the next five years, you could do a lot worse than sticking with US refiners.

Around the Portfolios

Marathon Petroleum (NYSE: MPC) has been one of my better bull-market picks. I’ve been recommending it from time to time since shortly after its spinoff from Marathon Oil (NYSE: MRO) in 2011, and the stock has rewarded the faith with a triple-digit move.

It’s also the first pick I made for The Energy Strategist, back in late January, and it proceeded to rise 25 percent over the next 50 calendar days. Then came last week, which sliced that gain nearly in half and pushed the stock below our buy point.

I wish this were a bargain to be seized. Instead, I’m changing our rating on the stock to Hold and urging you to defer any additional buying until we better understand the effect of shrinking refining margins on the industry’s earnings prospects.

I’m convinced the margin squeeze was the main reason the refinery stocks cracked last week. Since Feb. 22, the WTI Cushing Oil 3-2-1 Crack Spread, a rough measure of the profitability of refining a barrel of West Texas Intermediate crude from the Oklahoma hub into two-thirds of a barrel of gasoline and one-third of a barrel of heating oil, has shrunk from $36 to less than $20.

Among the factors in play is the ramping up of refinery runs after recent maintenance and the continued slow erosion in US gasoline demand, recently down 1.2 percent year-over-year and 8 percent from its 2007 peak.


Source: EIA

But — more importantly, I believe — production from the recently developed oil fields in the US interior and the tar sands in Western Canada is increasingly finding its way to refineries further and further afield, whittling the discounts that previously obtained closer to the wellhead. Pipelines new and old, expanded and reversed are now delivering more crude from west Texas, Oklahoma and North Dakota to the powerful refineries on the US Gulf Coast. Crude is also increasingly getting shipped out of the North American interior to the Gulf Coast and the East Coast by rail.

Meanwhile, generous spending on refinery upgrades has eroded the industry’s market leverage. For instance, Marathon’s Detroit refinery, recently upgraded to process the heavy Canadian crude, will soon have to compete for it with BP’s refurbished Whiting, Indiana plant, which will have much more heavy crude capacity next year. And indeed, the discount on heavy Canadian crude relative to the WTI has narrowed by more than half since the beginning of the year.

Other differentials are shrinking as well. Bakken crude, which traded at a $10 discount to the WTI as recently as late October, now fetches a slight premium. And the WTI itself is now just $12 a barrel cheaper than the Brent, half of the spread just two months ago. Not only is the WTI more accessible than ever (and therefore more valuable)  these days but the Brent hasn’t been as scarce as once feared and as a result its price has been in a downtrend. The bottom line for the refining industry is more competition and less bargain-basement crude for the best-placed refineries.

How hard will all this prove on Marathon? Perhaps not very. As a major refinery operator both in the Midwest and on the Gulf Coast it’s hedged its bets and is thus in a position to get hold of some of the cheapest crude one way or another. Analysts’ consensus estimates for the current quarter and year have slipped by only a few cents a share in the last week and remain higher than they were two months ago. On the other hand, margin shrinkage this dramatic cannot be ignored, and its effect on the bottom line will become much clearer after Marathon’s April 30 earnings announcement. Given the stock’s rapid retreat from recent highs and the recent margin trends, the near-term rewards of bottom-fishing here likely don’t justify the risks. Let’s wait for hard numbers before taking the plunge.

I can promise you one thing: if the margins hold up OK despite the stiffer competition, the stricter sulfur limits will prove chump change. If anything, the EPA’s cost estimates as quoted  above may overstate things. The consultant commissioned by the refining industry itself last year estimated annual compliance costs at $2.4 billion, or about 1.9 cents per gallon. Marathon’s gross margin was nearly 25 cents per gallon last year. If it can’t pass on less than 2 cents of increased cost to consumers it will have much bigger problems than the EPA.

— Igor Greenwald