Why Pundits Like Dennis Gartman Are Dead Wrong About Oil

This piece originally appeared on Oilprice.com.

I follow oil pretty closely given our exposure. As such, I get frustrated with many press and news show accounts of the commodity. It gets worse when the pundits and writers should know better. Frequently inexact terminology leads to misconceptions and sometimes I see outright falsehoods that completely distort the truth.

As a former oil analyst and professional energy investor, I feel compelled to take those to task. As a realist, I see that all markets require a difference of opinion and all investors talk their “book”. For this reason, when Jeff Currie at Goldman Sachs Commodities Group gets on CNBC and opines about future price movements, I give little notice. Jeff is posturing for his customers’ and GSs’ positions. Jeff can spin the story either way and chooses his statistics accordingly…That’s what he is paid very well to do.

Last week (March 28, 2016), I heard Dennis Gartman of the Gartman Letter, a trader and investor that I respect and have learned much from, spout an outright falsehood on CNBC. Everyone can have a bad day, but I’ve been hearing various versions of this for months. Dennis said in essence that oil prices could not rise very much because of “all the capped wells that could be brought on line very rapidly”. He predicted no more than $42/bbl this year. He estimated that at current strip pricing, you could lock in $45/bbl in 12 months, making large numbers of these “capped” wells profitable. The implication being that at current prices, the market would be rapidly flooded with new oil.

I’ll take the over on price, the under on production and bet all my capital that I’m right. (Oh, I already did that…). Dennis should know better. For fun though, I thought I’d like to take apart his thesis.

First, there are no “capped” wells in the U.S. To my knowledge not one well has been capped due to low prices, especially relatively young horizontal shale wells. Older wells are capped all the time when production is no longer sufficient to pay operating expenses for the well. Generally, onshore wells may cost something in the order of only $2,000 per month to operate. At $40 dollar oil, 3 barrels per day of production (gross) should cover operating costs.

What Dennis is likely referring to is the “Drilled Uncompleted” or DUC well inventory in the various shale plays. Some estimates have shown as many as 4,000 of these DUCs exist and the numbers are rising. Many pundits cite these DUCs as an effective ceiling on oil prices.

However, a DUC is very different from Gartman’s implied “capped” well. There are many reasons why a producer would drill and not complete a well. They may have had a rig under contract, they may want to beat competitors, retain their or their service companies’ good employees, they may be able to hold expiring acreage, they may just want to see what the rocks look like in a particular area. However, the most likely reason is that the completion costs of these wells can amount to over 60 percent of well cost maybe – $3 to $4 million per well. As such, this investment is very difficult to recoup if a well’s flush initial production is sold at low prices. This is compounded when whole well pads are completed at the same time to increase efficiency. If you don’t like the price one well gets, six wells coming on line at the same time is worse.

This also flows into the other reasons why this production will not flood the market, namely the intersection of costs, timing and decline rates.

• Costs – 4,000 wells at even $3 million per well is $12 billion dollars. Given the upheaval among producers, where does Dennis suppose the $12 billion will come from to “instantly” “uncap” these wells and increase production? Not from the banks, the high yield market is tight, equity investors have stepped up for some Permian and Eagle Ford producers, but $12 billion is a lot of money.

• Time – Let’s say that oil prices above $40/bbl equals a green light for energy producers to attack their DUCs. (There appears to be no factual basis for this, but let’s pretend.) A quick look at C&J energy services, which controls the country’s third largest frac fleet as well as other completion services, tells part of the story. Today, just over 50 percent of the companies’ fleet is working and the rest is “stacked” or to be retired. The people were laid off months ago. Clearly, when they get the signal that their customers want more completion services, they will begin to reactivate some of this idle iron – one frac fleet at a time. The problem is the C&Js stock price is $1.46 and they have close to $1.2 billion in debt. Where will the money come from to rehire people, and reactivate idle equipment? After that, will the people return? Yes, but slowly and at a high cost. What about Baker and Schlumberger? Both are in better financial shape but their fleets have been stacked also and at this time, investors are in no mood to hear a company talk about adding capacity. When these companies return fleets to active status, they will be competing to hire a smaller pool of laid off workers.

• Decline rates – Wells producing from tight rock or shale (wells that must be fracked) exhibit steep decline curves on the order of 75 percent during the first year of production. The implication is that producers are on a never ending treadmill in order to maintain or grow production volumes. That is, they must complete new wells in order replace the natural declines from existing wells. There are two critical points associated with these steep decline curves that pundits like Gartman don’t appear to grasp. The first is that based on current data, the four key liquids rich shale plays have declined by over 600,000 bopd since their peak of production in March, 2015. This production is gone. These wells have depleted. They can’t be turned back on. The only way to increase production again is new completions and new wells – in other words massive new reinvestment. This is very different from past cycles when OPEC dialed back production by idling a major field or two until demand rebounded. These OPEC giant and super giant fields are a totally different animal. It’s all about the infrastructure, not the productivity of a single well. The entire complex can be shut down, reworked, maintenance performed, etc. Then turned back on…more akin to a refinery than typical single or multiple well fields. But that’s another story. Bottom line – that 600,000 bopd is not magically coming back. It took the onshore industry something like 12 months running flat out to add those volumes. Given oil prices, it will be quite a while and it will take higher prices before the industry even gets back to a steady walk, much less a flat run.

Another key thing to understand about decline curves is that they are continuous and right now declines are accelerating. However for example purposes, let’s look at the Eagle Ford. There are some 10,000 wells in the Eagle Ford producing today, and they are all in decline. The EIA estimates the average Eagle Ford well adds 800 bopd in its first month of production. Last month, Eagle Ford production is estimated to have declined by 60,000 bopd. That implies that 75 new wells per month must be drilled and completed to just replace this 60,000 bopd. Assuming it takes 15 days to drill a well, that implies around 38 rigs drilling and around 25 frac fleets running above what is running today! Today, there are 42 rigs drilling for oil and we estimate 10 – 15 frac fleets running in the Eagle Ford…so just to replace production, the industry would have to increase rigs running by nearly 100 percent and frac fleets by 150 – 200 percent. This would require a massive mobilization of capital and manpower. During this whole mobilization process, production from existing wells is declining, month after month. Don’t get me wrong, I believe this will happen. However, I know this won’t happen quickly and won’t happen at $40/bbl oil, making Gartman’s thesis and pricing argument completely false.

Production data, or lack thereof, is a primary hindrance to clear and transparent oil fundamentals. The mechanics of the above discussion would be more obvious if we could measure field production in real time. In fact, production data in Texas takes some three months to even estimate, and these estimates are often revised. The same goes for well completion data. The EIA tries to model this through its “Drilling Productivity Report”. However, there are no similar efforts for the rest of the global oil industry, in fact, OPEC publications use third party reporting not internal or “real” data from the companies themselves.

In Saudi Arabia, production statistics are a state secret. Not surprisingly, many countries distort the data to suit their own needs. That’s why the IEAs look at G7 storage data is an important industry statistic. It is widely recognized that both global demand and supply data is inaccurate, but changes in storage inventories should reflect supply and demand changes. The only problem with this approach is they only get data for around 2/3 of the global storage capacity. This is what led to the recent headlines “800,000 bopd of oil is missing”. Supply estimates exceeded demand estimates by 800,000 bopd during the quarter, yet storage didn’t build, leaving the question of where did the oil go? The answer is that there never was this extra oil…if it existed, it was burned. More than likely, both supply and demand estimates were off by that amount.

Third parties like “Drilling Info”, BTU Analytics, CERA, etc. Provide their looks at the market for very high prices, and as such are much more granular than those from government data providers. As much as they try, they are still limited by the availability of international data and reporting time lags domestically, not to mention their own biases.

Generally it takes 18 months before the world has a decent picture of supply and demand. This is little consolation to those trying to do real time analysis on the direction of prices. That is why I can say categorically “the fix is in”. In other words, fields are declining, meaning investment is far below levels required just to replace production. The only thing that will change the vector of these declines is more spending, lots more spending, and the only thing will spur lots more spending is higher prices. Significantly higher than $40/bbl.

In conclusion, we have a typical commodity price cycle. Prices have dropped to levels destroying capital, bankrupting businesses, idling massive amounts of equipment and manpower. The cycle is reversing now. The weekly EIA numbers are showing steady declines in production (this is a balancing item – not real production estimates) and also increasing demand—In the United States. The IEA is showing the same thing in their monthly report that has a decent look at the G7 countries and attempts to look at the G20. Between these two, there is a large world with little accurate measurement. China for instance jailed a Platts reporter for espionage when he tried to put together a fundamental energy statistics database.

Inevitably, we will have another price shock—or at minimum an upside surprise. It’s unavoidable at this point. Oil never transitions smoothly. Just like all the oil bulls had to be run out during the declining price stage, all the price bears, like Dennis Gartman, will be run out when fundamentals hit them over the head. Gartman, to his credit, will change his tune 180 degrees when he sees the actual data shaping up. That’s how he has survived so long and profitably as a trader.

But by then it will be too late, the world will want incremental supplies immediately—yet the industry cannot scale in real time. In order to motivate producers to get busy and provide incremental supplies, prices must increase sharply from current levels. My prediction – $80/bbl in 18 months, but it won’t last very long. I think $60-$70/bbl is a healthy range.