Flood of Low-Cost Supply Swamps Our Frac Sand Valuations
We've reduced our long-term pricing forecasts.
We have lowered our fair value estimates for the frac sand companies we cover to reflect recent industry developments, most notably deteriorating pricing in the face of continued low-cost sand additions. Our fair value estimate for U.S. Silica (SLCA) moves to $15.50 per share from $24.50, for Covia (CVIA) to $7 per share from $15.75, for Smart Sand (SND) to $3 per share from $6, and for Hi-Crush Partners (HCLP) to $8.50 per unit from $11.50. Our no-moat rating for all of these companies remains in place. We also increased our fair value uncertainty rating for U.S. Silica to extreme. After the revisions, each company looks approximately fairly valued, in our view.
The lower fair value estimates are the result of lower long-term frac sand pricing forecasts. Our midcycle (2022) Northern White mine-gate price is now $26 per ton, down from $28 per ton previously, reflecting a lower marginal cost of production due to a reduction in our midcycle Northern White supply forecast to 36 million tons from 40 million tons. The reduction in Northern White supply is caused by an increased forecast for lower-cost regional sand supply. In an astounding reversal of the situation several years ago, when Northern White supplied around 75% of all North American frac sand demand, we expect it to supply less than 30% of our total midcycle forecast frac sand demand of about 128 million tons.
Indeed, our base-case forecast now incorporates a view that Northern White sand will be almost certainly be pushed out entirely of the Permian, Eagle Ford, Haynesville, and Oklahoma shale plays. Northern White sand, mined in Midwestern states like Wisconsin and Minnesota, is located around 800-1,200 miles away from these shale plays and so it is at a substantial disadvantage because of higher transportation costs. The discovery of more closely located sand of suitable quality (combined with a shift to new hydraulic fracturing techniques) has enabled the disruption of Northern White.
At this point, the downside for Northern White supply seems limited, as we think the threat of regional sand disruption in the Bakken, Marcellus, Canadian, and other plays remains small. These are older plays; the rewards for finding regional sand have been in place for a nearly decade, but essentially no suitable sand has been found. Additionally, they are somewhat closer to Northern White mines than the plays subject to disruption, limiting the prospective logistics cost savings. Additionally, we see plausible upside relative to our base case for Northern White shipments to the Southwestern plays. The quality of the sand for recently announced Eagle Ford, Oklahoma, and Haynesville mines remains an open question, as almost none of these mines have yet commenced operations. Additionally, although we expect all of even the Permian’s 40/70 mesh sand (a grade that generally has higher quality requirements and which we expect to account for about one third of the Permian’s midcycle sand demand) to be supplied by non-Northern White mines, we’ve heard several operators express doubts about the feasibility of using non-Northern White sand.
In addition to our reduced Northern White mine-gate pricing, we have also lowered our expectations for our covered sand companies’ regional sand gross margins by around $5-$10 per ton. This reflects the fact that the Permian will become self-sufficient in 100 mesh sand (accounting for over 60% of our forecast sand demand), which will lead to drastically lower pricing as pricing no longer will be set by higher-cost Northern White sand. We think 40/70 pricing will hold up generally around current contracted levels, as marginal costs will probably be set by imports from other regional mines in East Texas or Oklahoma.
Other Southwestern plays like the Eagle Ford, Oklahoma, and the Haynesville have a solid chance of becoming self-sufficient, although we actually expect some imports of excess Permian 100 mesh and a small amount of Northern White 40/70 in our base case into one or more of these plays. The situation will remain fluid over the next year as around 25 million tons in gross new regional capacity is added in these plays (by our estimates). As with the Permian, for regional sand producers, self-sufficiency is a curse, not a blessing, as it means that pricing collapses to the marginal in-basin mine cost.
In addition to the pricing revisions, our updated forecasts have implications on the volume side. We’ve edged back volumes for Northern White producers in line with our overall forecasts. Additionally, we’ve curbed existing non-Permian regional sand volumes for existing producers. The older regional mines typically entail much higher production and transportation costs compared with the newer entrants and are actually more at risk of disruption than many Northern White mines. We’ve heard of several of these mines in Texas, Missouri, and Arkansas already shutting down operations; for example, both Pioneer and Covia have announced plans to shut their central Texas mines at Brady and Voca, respectively. This development actually has the strongest deleterious effect on volumes for U.S. Silica, which made the earliest push into regional sands among large frac sand companies and now has around 6.5 million tons (28% of company capacity) of at-risk regional sand capacity, by our estimates.
We increased our uncertainty rating for U.S. Silica to extreme, putting it in line with our extreme ratings for all frac sand companies. Our uncertainty regarding frac sand companies surpasses all other areas of oilfield services, owing to the wide range of potential outcomes for the long-term supply of low-cost regional sand in the Permian and elsewhere.
Preston Caldwell does not own (actual or beneficial) shares in any of the securities mentioned above. Find out about Morningstar’s editorial policies.