In our 7/29/15 note, Ceramic Proppant Is Going to Zero, we argued that the decline in CARBO Ceramics Inc.'s (NYSE:CRR) sales volume was secular rather than cyclical. As of 2Q15, CRR's ceramic proppant volumes had already fallen 58% YoY as the price of oil had fallen ~50%.
With CRR's 2Q16 report on 7/28/16, we now know that while the price of oil only fell another 21% from the average of 2Q15 ($58 per barrel) to 2Q16 ($46 per barrel), CRR's ceramic sales volume fell another 58% YoY to only 71MM pounds (down 87% from the peak in 3Q13). The quarter-over-quarter comparison is also compelling: The price of WTI oil rose 36% from 1Q16 to 2Q16 (from $34 to $46 per barrel), yet CRR's sales volume fell 41% quarter over quarter in what is typically a seasonally stronger second quarter.
In our 3/18/16 note, Carbo Ceramics: Will You Take the Bait?, we presented a detailed operator level analysis of how CRR's largest E&P customers were structurally shifting away from ceramics to sand-based completions. We further warned investors not to be fooled by the prospect for an equity offering when CRR's long-term earnings power would be a decidedly negative number.
With the 2Q16 earnings release, investors were greeted to another large negative FCF quarter and a $75MM at-the-money (ATM) equity offering courtesy of the Cowen Group (NASDAQ:COWN). COWN will earn 3% of gross sales proceeds, or a sum of up to $2.25MM, for placing CRR shares with investors.
On the 1Q16 earnings call, Gary Kolstad provided the following 2Q16 cash flow guidance:
Although we plan to reduce costs further, the operating outlook may likely deteriorate in the coming quarters and, as such, our near-term quarterly cash burn could remain in the low- to mid-teens.
However, actual cash burn in 2Q16 was $20.1MM. If we assume the mid-point of "low-to mid-teens" means $14MM, the actual cash burn figure represents a miss of 44% versus the midpoint of CRR guidance provided just a few months prior. CRR missed its cash burn target while spending only $351k of capex in the quarter, calling into question management's visibility on its own business fundamentals.
Investors may not have realized the cash burn was so high since the 2Q16 cash balance was obfuscated by two one-time items: (1) a $37.5MM tax refund and (2) a $25MM infusion from a related party loan. Absent those two one-time items, we believe CRR would already be in violation of its minimum liquidity covenant and potentially already commencing bankruptcy proceedings.
On the 2Q16 earnings call, Mr. Kolstad made the following statement:
In the second quarter, we appeared to have hit the low point at the beginning of the quarter and saw monthly improvements as the quarter progressed, following a stabilizing rig count, modest commodity price increases and modest client activity increases in completions. In addition, with the increasing commodity price, we have received increasing customer inquiries about procuring ceramic proppant for completions in the second half of 2016.
Despite claims for improving cadence of monthly sales volumes, ceramic proppant volume fell 41% quarter-over-quarter in 2Q16 despite the fact that the price of WTI crude increased 34% in 2Q16 on average, and the second quarter is historically a seasonally stronger quarter than the first quarter. Furthermore, with respect to the third quarter, we believe "customer inquiries" is a non-meaningful metric and does not correlate to actual sales or even backlog.
CRR sold such a small amount of ceramics in the second quarter that volumes almost have to increase sequentially in the third quarter, which is typically the seasonally strongest quarter anyways. The tone of the CRR 2Q16 earnings call and press release would have one believing the ceramic proppant market is experiencing green shoots and the market recovery is just around the corner. This is decidedly not the case in our view.
In "Carbo Ceramics: Will You Take the Bait?", we argued that it appeared that CRR's two largest customers, COP and CLR, were in the process of removing ceramics all together from their completion budgets in 2016. At the time of the report, COP had reported its latest two wells were all sand while CLR was still using ceramics to complete its Bakken wells.
Our updated analysis using FracFocus reporting suggests that all COP and CLR wells reported since our last report are indeed all sand, which supports our claim that these E&Ps, up until now CRR's two largest E&P customers based on FracFocus data, have switched to all sand-based completions.
E&Ps like COP and CLR do not necessarily buy proppant directly, but rather procure it through their oil field service suppliers. From the above table, we can see that Halliburton (NYSE:HAL) has the largest market share of COP's recent Bakken well completions, and thus HAL itself would likely be the company buying ceramic proppant from CRR on behalf of COP. According to CRR's 2Q16 10-Q, HAL and CRR have now terminated their long-term contractual arrangement, which supports our FracFocus analysis that COP and CLR have now stopped utilizing ceramics.
According to CRR's 2015 10-K:
Our largest customers are participants in the hydraulic fracturing industry. Specifically, Halliburton Energy Services, Inc. and Schlumberger Limited each accounted for more than 10% of our 2015 and 2014 revenues.
Also attached to the CRR 2015 10-K, we see that HAL and CRR entered into Amendment No. 5 to their 8/28/08 long-term Proppant Supply Agreement. HAL and CRR amended the term of the agreement but blocked out the termination date with an "XXX" so that investors could not see when expiration would occur:
According to the amended agreement, after the "XXX" date, HAL no longer has any obligation to CRR. With the 2Q16 10-Q, we find Amendment No. 6 to the HAL/CRR Proppant Supply Agreement, which was amended on June 30th, 2016. Amendment No. 6 states the following:
Thus, we believe the disclosure indicates that HAL has terminated its long-term proppant supply agreement. We believe that the decision to terminate the agreement was likely made at the beginning of the year, when COP and CLR appear to have made the budgeting decision to remove ceramic proppant from their Bakken well completions.
The collapse in CRR's proppant sales volume in 1H16 (sales volume down 48% YoY in 1H16 vs. 1H15), COP and CLR's switch to 100% sand well completions per FracFocus disclosure as of the beginning of the year, and the termination of the HAL Proppant Supply Agreement all support our belief that CRR lost its two largest remaining E&P customers at the beginning of the year.
So if CRR may have lost its biggest remaining customers, how does the company see an improvement in 3Q16? We believe it's likely just a dead cat bounce based on seasonal tailing volumes.
Mr. Kolstad made the following claim on the 2Q16 earnings call:
We're becoming more optimistic on the industry operating environment given that the oil and natural gas commodity prices appear to have firmed. Communications with our clients lead us to believe third quarter 2016 ceramic proppant sales will increase from the second quarter.
Let's take a look at the Baker Hughes (BHI) oil rig count, which is reported weekly. While it is true that the rig count has risen off the trough in May, the US average rig count was up 17% in the third quarter (390 rigs) versus the average for 2Q16 (334) and is still down 12% versus the 1Q16 average (442). More meaningfully for CRR, in the Bakken, the third quarter rig count is up only 2.6 rigs to 27.6 versus the 25 rig average in 2Q16 and is still down 29% from the 1Q16 average of 39 rigs. As we can see from the below chart, most of the US rig count growth has come from the Permian basin where ceramics are only used minimally for the occasional tailing.
Source: BHI Investor Relations
Furthermore, given the large scale customer loss that we have documented extensively in our research on CRR, we believe that a structural shift away from ceramics has occurred in the industry. Why else would HAL terminate its long-term supply agreement? Thus, in our view, the rig count is no longer a meaningful indicator of ceramics demand anyways since the vast majority of oil and gas completions are utilizing all sand (or occasionally, resin coated sand).
We wonder how CRR can make the claim that "communications with our clients lead us to believe 2016 ceramic proppant sales will increase from the second quarter" with a straight face. 2Q16 proppant sales were so slow that there indeed should be a positive sequential improvement from seasonal demand in 3Q16, especially since we believe COP and CLR, CRR's two largest customers, were already out of the ceramic proppant business by the beginning of 2016 and thus sales to these customers would not be in the 2Q16 volume figures.
Since, according to FracFocus, there are no longer any major E&Ps completing oil wells with ceramics in the Bakken, the small amounts of ceramic volume we continue to see from CRR relate to tailings and should thus be seasonal along with typical industry completion patterns.
John Daniel of Simmons & Company attempted to shed some light on this issue in the Q&A of the 2Q16 call:
Gary, we had a lot of chatter over the last couple years of people switching away from ceramic, trying to take the lower-cost options. Some would say that they were getting good results with the lower, with sand, et cetera. Have you had any customers in the last several months or so come back to you and say, look, we're going to start returning to ceramic, because we've now found that the value proposition is better? Can you provide some color along that topic?
Gary Kolstad from CRR responded:
I think what's interesting for me is some of those clients that up until the last day held out and they were still tailing in with ceramics until, let's say, the fourth quarter of last year, the first quarter of this year, we see some of those folks starting up activity and still tailing in with ceramic. And I think that's your best testament.
Mr. Kolstad's is saying that some customers are "still tailing in with ceramics," but cannot point to any customers truly returning to ceramics for full well completions like they historically did in the Bakken.
However, even if volumes increased sequentially by 200%, the total volume figure would be below 3Q15 volume of 236MM pounds, and total CRR gross profit was still negative $4.6MM in 3Q15, and yes, that figure includes the highly touted "technology" sales. Of course, since 2015, CRR appears to have lost both COP and CLR as customers, so we find it highly unlikely CRR will register anything near 236MM pounds of sales in the future.
CRR is once again guiding to unrealistic cash burn guidance
In the 3Q16 earnings call Q&A, Ernesto Bautista provided the following guidance:
So for the quarter, it's probably, it's in the $20 million range. That was directly impacted by some severance payments that we made that totaled about say, $4 million. So you're really back to that mid-teens range that we had talked about prior quarter. Going forward and maybe focusing on second quarter, I am sorry, third quarter, first of all, I think as we look forward, our cash burn picture would be something in the low-double digits. We've done a lot to reduce costs. So we're benefiting from that.
While CRR appears to have mastered the art of adding back seemingly recurring one-time expenses, the company is admitting that 3Q16 was in the $20MM range nonetheless but expects an improvement to "mid-teens" range in 3Q16. To calculate "mid teens," CRR starts with $20MM of burn and adds back $4MM for one-time severance cash costs in the quarter plus some benefit from cost reduction and seasonal volume uptick.
However, while above we outlined that CRR expects volumes to improve in 3Q16, we do not believe that it is possible for cash burn to be reduced if volumes are going to increase, since then CRR's working capital requirements will once again increase. Receivable days have blown out to 97 days, up 40% sequentially and 13% YoY to an all-time high, and net receivables have fallen to only $21.9MM.
Indeed, it appears the opposite of what CRR predicts is about to happen: we believe that free cash flow burn may be about to accelerate. Since 4Q14, CRR has experienced a $111MM cumulative free cash flow benefit from liquidating its receivable balance. If activity actually increases, receivables will again become a usage-rather than a source- of cash, unless CRR is able to extract better payment terms from its customers even though today CRR's customers are taking record long amounts of time to make payment. CRR described this conundrum in its 2Q16 10-Q:
We generally supply our domestic pumping service customers with products on a just- in- time basis, with transactions governed by individual purchase orders and/or a master supply agreement. Because of their purchasing power, our key customers may have greater bargaining leverage than us with respect to the negotiation of prices and other terms of sale in their supply contracts with us, which could adversely affect our profit margins associated with those contracts.
CRR makes the claim that ceramic pricing "has likely bottomed for base ceramic." This claim may indeed be true. The problem is that pricing has settled at a level where CRR is decidedly unprofitable. CRR generated gross profit of negative $20MM in 2Q16. We think that CRR's addback of "slowing and idling production" costs to its negative gross profit figure is misleading, as the underutilization of its asset base due to collapsing demand is not a "one-time" expense.
On top of negative gross profit, CRR of course has SG&A expense. CRR has cut SG&A to the bone but it remains a $10MM expense.
Next comes interest expense. As of 7/28/16 per the 10-Q, $58.9MM remains outstanding on CRR's amended revolving credit facility at Libor + 7.00%. As of May 2016, CRR also has $25MM of subordinated related party debt at a 7% cost. The combined debt balance represents a ~$1.5MM cash expense each quarter.
Also, because CRR is losing so much money, the company is currently getting a ~35% cash tax benefit on its losses. For conservatism, let's assume that CRR is able to continue to get future tax refunds from the government as the company achieved in 2Q16.
Finally, while CRR cut capex to only $356k in 2Q16, in the 2Q16 10-Q, CRR tells us that it expects to need to spend an additional $3MM in 2H16, which equates to $1.5MM per quarter.
Therefore, we project CRR will burn ~$15-23MM per quarter absent any working capital impact. If CRR does see an increase in activity, the company will have to build receivables in advance of collecting its bills, which will cause cash burn to accelerate in the short term as the working capital unwind reverses. The average accounts receivable cash flow benefit CRR received from 4Q14 to 2Q16 was $16MM per quarter. Thus, we find CRR's claim that cash burn will be "mid teens" while activity will increase at the same time to be difficult to believe since if activity actually increased, near-term cash burn could easily be double what CRR projects due to an increase in working capital requirements. Remember, on the 1Q16 earnings call, CRR said cash burn going forward would be "low to mid teens" and instead in 2Q16 cash burn was $20.4MM.
We have time for one more Kolstad quote from 2Q16:
The good news is that the industry has been working through the large inventories of low-quality Chinese ceramic proppant that made its way into the States prior to the severe downturn. The industry will recover, and today, we believe the U.S. ceramic proppant suppliers can adequately produce and supply the domestic ceramic proppant market with no need for the low-quality Chinese imports.
We believe the inverse of his statement is also true: The domestic ceramic proppant industry has no need for domestic producers like CRR. The entire remaining industry, which appears to be just a few million pounds of tailing demand, can be serviced by Chinese imports. CRR has been talking about low quality Chinese imports for years, yet the recurring inventory problem has never gone away, and we do not believe that it will go away now that demand appears to have secularly tanked, meaning that the Chinese ceramic proppant industry itself is even more oversupplied.
With the apparent loss of COP and CLR at the beginning of 2016, we believe that all major Bakken E&Ps have officially ended ceramics completions.
As we argued on 3/18/16, absent an equity deal, we believed CRR would be in violation of its covenants and potentially bankrupt before the end of 2016. We previously modeled a $75MM primary equity raise and argued that even after completing this, CRR might still be insolvent in 2017 or 2018.
Since our March note, CRR has amended its asset coverage covenant and replaced it with a $40MM minimum liquidity covenant (so that the lenders are made whole prior to CRR burning their remaining cash) and announced a $75MM ATM courtesy of Cowen Group and a $25MM related party loan.
We believe that Cowen's efforts can only serve to only temporarily delay the inevitable. As such, depending on the amount of equity raised via Cowen, we are now predicting a bankruptcy filing in 4-8 quarters, with the range of outcomes dependent upon (1) the amount of quarterly FCF burn (we forecast $14.8-22.6MM per quarter and (2) the success (or lack thereof) of the Cowen ATM.
Disclosure: I am/we are short CRR.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.