The mad science of drilling the earth with Austin Vernon

I once wandered into a room where Austin Vernon was giving a talk about his specialty, fracking, and was transfixed by how delightfully steampunk the enterprise sounded. It is also deeply misunderstood by non-specialists. Plus, supporting the substantial engineering and chemistry, there is a cottage industry of capital stack optimization. So I thought I'd have a conversation with him as part of an occasional mini-series on energy economics. (See also: Casey Handmer on solar.)

[Patrick notes: I add notes to my transcript, set out in this fashion.]

Sponsors:

Check is the leading payroll infrastructure provider and pioneer of embedded payroll. Check makes it easy for any SaaS platform to build a payroll business, and already powers 60+ popular platforms. Head to checkhq.com/complex and tell them patio11 sent you.

Building an enterprise-ready SaaS app? WorkOS has got you covered with easy-to-integrate APIs for SAML, SCIM, and more. Start now at https://bit.ly/WorkOS-Turpentine-Network

Timestamps:

(00:00) Intro
(01:38) Fracking technology and horizontal drilling
(05:12) The history and development of fracking techniques
(12:26) Communication methods in drilling operations, including mud pulse technology
(15:50) The economics of drilling operations
(17:28) Scale and cost comparisons between different types of drilling projects
(19:30) Safety considerations in onshore vs offshore drilling
(20:50) Sponsors: WorkOS | Check
(24:04) Discussion of small-scale "mom and pop" oil operations
(27:13) The impact of oil on local economies
(31:45) The lifecycle of fracked wells and their long-term production
(36:15) Financing in the oil and gas industry
(39:19) Unique aspects of US mineral rights laws and leasing practices
(42:38) The process of setting up and funding new drilling operations
(52:55) Environmental concerns and groundwater protection measures in fracking
(56:40) The physical footprint of drilling operations
(59:12) Learning curves in fracking and geothermal energy extraction
(59:56) Diamond drill bits are not quite forever
(01:03:59) Where fracking goes from here

You can Ctrl-F these headings to find them in the below transcript. I cannot hyperlink them, unfortunately, due to technical limitations.

Transcript

Patrick: Hi everybody. My name is Patrick McKenzie, better known as patio11 on the internets, and I'm here with my buddy Austin Vernon.

Austin: Hello.

Patrick: So, Austin was previously a petroleum engineer, probably also currently a petroleum engineer – I think that’s a thing one does not necessarily stop being, sort of like the former vs.  retired distinction in the military. Professions are something we carry around for the rest of our days.

[Patrick notes: A shibboleth which some people are extremely serious about: do not call people who have previously served in the military “ex-”, as this implies that they were stripped of their membership via e.g. a dishonorable discharge. A socially appropriate word in those circles is “former.” Retirement is a subset of “former” since you can exit honorably without retiring.

I think servicemembers are entitled to some amount of special deference but that many professions put an indelible stamp on people.]

Patrick: But be that as it may, we had a chat once about the miracles of modern engineering involved in fracking, and I was so intellectually interested in it that I wanted to have you on the podcast to talk about it with a wider audience. 

To set the stage a little bit, I'm from the more software side of the hardware versus software ecosystem; I think energy is the thing that comes out of the outlet on a wall. The only time I have to think about it seriously is when it's at a data center. There are scary wizards, sitting next to the other scary wizards that handle networking. They handle power and cooling. I try to avoid thinking about that as much as possible.

It is likely not the case that we can continue to avoid thinking about this with forecasted energy curves demanded for, among other things, the utilization of AI over the next decade to two decades. So I want to dig into the (bum bum bum!) Complex Systems that result in those electrons getting from several kilometers under the soil into our homes, factories, and data centers.

And with that out of the way, take it away Austin. What is fracking, for those of us who haven't followed the news for the last 20 years? 

Austin: Yeah, I think technically it’s horizontal drilling with multi-stage fracturing – essentially, you drill down about a mile or two miles, then you make a wide turn that takes about 900 feet to turn all the way. Then you drill another – it used to be that we only drilled about a few thousand feet, but now there's people that drill up to three miles horizontally.

Then you run some pipes down there and then you're ready to do your multi-stage fracturing – typically with some of these long ones, you might do as many as 50 or 60 different stages. That essentially allows you to control how the fluid goes into the rock and breaks it at various intervals, every few hundred feet.

Then with that, we also pump in sand – that usually is the prop, which holds the cracks open after you make them. That allows the oil and gas to flow out the well up to the top, and then get processed from there.

[Patrick notes: You may also hear this referred to as “proppant.”] 

Fracking technology and horizontal drilling

Patrick: The general theory here is that, over the past 100 plus years of doing petroleum and other sordid energy resources exploration, we tapped the stuff that was easily accessible – where you could either literally siphon it off of the surface of the earth, Beverly Hillbillies-style, or just drill down vertically and suck it up with a straw, essentially – and we got more technologically advanced and more creative at taking out, for example, gasses that were diffused among, I believe shale rock is the term.

[Patrick notes: For those not familiar with classic American television, come listen to the story of a man named Jed, … and then one day he was shooting at some food, and up from the ground came a-bubblin’ crude. Oil, that is.] 

Patrick: We use liquids that we pump out in the hole to engineer the working environment around these pipes that we put down, such that the gas un-diffuses from the rock, comes into the pipe that we have, and some combination of buoyancy and liquids moving through, etc., moves that to the top where we can then put it in a pipeline and send it to refineries.

Austin: Actually the shale is the source – that's where oil comes from, or gas comes from, basically old sea creatures, like plankton and stuff, that fell down onto some shelf, and then they got covered up, and then through geological processes got turned into oil. So really what we were doing before was going and finding very special features, like a place where the gas or the oil had leaked out of the shale and gotten trapped.

So, a special rock feature, where it'd be like a little hill or something underground essentially. And then you also had to have an actual reservoir there, like a lot of, say, a good sandstone or good limestone that had porosity and permeability. You could recover the oil. So essentially, the advances were going straight to the source and just getting the oil and gas where it was generated, and still often is.

Patrick: So my limited understanding of geology is that we have, essentially, pseudo-randomly generated subsurface features due to some combination of physics, chemistry, and plain old luck, and some of those subsurface features are amenable to creating oil, gas, et cetera, and then capturing them in such a way that in the last several hundred million years, it didn't just escape into the atmosphere uselessly.

And so, we explored to find which places in the world had those weird combinations of features, and certain combinations of features were exploitable by technology that we had back in the day. Certain combinations of features, we were aware they had oil and gas around them because we understood physics, but didn't have the technology to repeatedly exploit them in an economical fashion.

And as a result of technological development, we have essentially found the capability to exploit resources that were not previously economically exploitable. Do I have that roughly right in terms of science and engineering?

The history and development of fracking techniques

Austin: Yeah. And it's a very common theme as we're seeing this – we had the same thing with geothermal where we were only going for special features, and now essentially we're going for that source. Then also we'll see the same thing if geological hydrogen ever becomes a thing where, right now, you have people looking for those special features – if they even exist, we don't know that they do – but really long term, if you're probably going to make that work in a big way, you're going to be going right to that iron-rich rock and feeding it with water to do the reaction and all that kind of stuff.

Patrick: So for folks who might not have heard the term geological hydrogen before, the hypothesis is that there are some places in the world where hydrogen gas is naturally occurring as a result of superheated iron-dense rock, which causes, presumably, the breakdown of water into oxygen and hydrogen. The oxygen does something, hopefully doesn't explode on us, and then the hydrogen is available for tapping as a potent, high-density energy source – that is the rough idea, presumably.

[Patrick notes: I am not sure whether I remembered that iron catalyzes that reaction from chemistry class or from video gaming.] 

Austin: Right. And I think hydrogen is just always tough because you have so little mass that comes with a volume. Like, if you think you have a certain reservoir volume of oil, you have way, way, way more mass than you have with hydrogen. There's just not very much hydrogen in there. And so, almost implicit in this, if we're going to find a special feature where you can actually produce hydrogen, is that there's actually active generation going on, replenishing it, because it's not just going to sit there for a long time.

Patrick: And so effectively there the hydrogen becomes a replenishable transport mechanism for tapping the latent waste heat in the earth and moving that up to places where humans could actually make use of it.

Austin: Yeah. I think actually, you're talking about the oxygen – I'm pretty sure the oxygen actually gets locked up in the rock. So it's more like you're taking advantage of the earth’s latent reduction potential in its rocks, as opposed to heat there.

Patrick: Oh, okay. So, as you know (because we've previously talked about it), I've been going out a bit of a geothermal bender recently, and one of the things that I've learned – or I believe myself to have learned in the course of researching geothermal – is that heat within the crust of the earth comes from the breakdown of trace radioactive elements, rather than whatever magical energy source that I thought was providing it. I just sort of always assumed, “ehh, the center of the earth is really, really hot for random reasons, and has stayed like that forever because the earth is a good insulator.”

And it turns out like, no, actually, heat is generated over time in the same way that it radiates into space over time, and we are not simply at equilibrium; there continues to be heat energy added as radioactive decay of nuclei happens.

Austin: Yeah. And I think one thing people don't quite always get at first about geothermal is that – actually, I get this question a lot – they do the math on what that flux is and what the heat being generated is, and it’s actually very, very low; it's like, “well, why is anyone excited about geothermal?”

Really what we're doing with geothermal is mining the heat that's already in the rocks, because the rocks are dense enough and they're hot enough that they actually hold a pretty enormous amount of energy. It's all about mining heat, as opposed to taking advantage of that ongoing generation.

Patrick: And the mass of the earth is just – you know, on the one hand, we have a very good approximation for that number; on the other hand, it has so many zeros in it that it sort of boggles human comprehension. The mass of the earth relative to all human activity is very large, so we can siphon a very small percentage of heat from it for a very long time if we successfully work out the engineering kinks.

[Patrick notes: 5.97 x 10^24 kg. If a carbon atom massed 10 kg, then a mole of carbon atoms would have approximately the mass of the earth. A mole of carbon actually masses about twelve grams because carbon atoms are really tiny. Or, to put it another way, if all of the ~120 billion humans who have ever lived had equal shares in that mass, your share would be 50 trillion kg, or about the U.S.’s production of cement if we hold it constant for the next 500 years. And I certainly hope I did not flub this calculation by an order of magnitude or three in any direction, but this is what incomprehensibly large means.]

Patrick: Talking about working out the engineering kinks, if you rolled back to the popular discourse around energy economics when I was an undergrad in the early 2000s, Peak Oil was all in vogue, “we're going to run out of this,” et cetera. The peak oil thesis, for those of you who haven't heard this before, was this theory that, looking in the rear-view mirror, there was a year in which we had extracted more oil and gas than we ever would again in the future history of the earth – that the resources were depleting and, you know, they only go in one direction. And so the theory said that we had tapped the easy stuff, there was no more easy stuff, and we were locked into a future of declining energy abundance, as just a consequence of physics.

[Patrick notes: I’m being somewhat unfair to Hubbert here, given that he prognosticated the peak decades in advance using advanced-for-the-time curve fitting. But, at the point where I was in university (2000-2004), Peak Oil was considered a historical fact for the U.S. and an inevitability for the world within ~20 years.

Fast forward twenty years: “U.S. produces more crude oil than any country, ever” tells the tale very succinctly.] 

Patrick: My narrative for fracking is, we are not simply in a battle against physics – we also have human ingenuity, which allows us to bring things that were not previously in the equation into the equation as a result of engineering advances.

[Patrick notes: I award my college self a cookie for consistently telling the debate team that Peak Oil was in fact unlikely due to technological progress, despite the reams of evidence we had “cut” where credentialed experts said the opposite.] 

Patrick: But in the early days, my understanding is it was pretty rough. What did fracking look like in those first early couple of experiments, before we actually put in the effort to get good at this?

Austin: Yeah, so it's really a combination of several factors. Like we talked about, there's the horizontal drilling aspect, there's the multi stage fracturing aspect; it was really about putting it all together, in the right way, in the right place. People had tried to do horizontal drilling as long ago as a hundred years ago, but once we got computers that were practically small, that's when it became realistic to actually do underground steering of the drill bit – that would be in the late seventies, early eighties, that became a thing. 

You had horizontal wells being drilled in the eighties. Actually, there's a good example, one field I ended up drilling in: they had tried to drill just a horizontal well in the early nineties, and it wasn't very successful because they just had this big open hole – they tried to do one frack, [but] this iron law of fracturing is that the fluid goes the path of least resistance, so what ends up happening is you end up only creating one fracture along this whole long wellbore because all your fluid just goes out one place. That's when we really started getting into the multi-stage. (There was actually some multi-stage, like only one or two stages, happening in vertical wells.)

Eventually we're figuring out – with vertical wells, we rely on gravity to get our tools down mostly, but it's kind of non-trivial to get the tools down in a horizontal well. What we do is you have a plug on the end – cause usually you need a plug anyway – with the wireline, and then you get everything ready in the wireline truck. You have to call the radio and tell the pump trucks to start up; you pump it down and you're using just the velocity of the fluid to push your plug  and your perforating guns down. It's a little wild. But finally we got good at that kind of stuff.

You made a modem for WHAT

Patrick: Some of the stuff you've shared with me has whiffs of mad science to it. One of the things that I remember is that we were talking about the centrality of computers to run these operations these days, and it turns out that it's very useful to have a computer close to the drill bit, you know, way down the well, away from all of us humans who can send the computer commands.

I thought, naively, like, you put a copper wire between the surface and the computer down there to control the thing – but it turns out the well is not a hospitable place to maintain wires in good working order. What was our crazy mad science solution to that?

Austin: It’s called mud-pulse communication. It’s kind of like Morse code: it's sending sound, down from near the drill bit – where it's taking measurements about what the inclination is, where we're headed – and it sends it up to the surface. You have microphones all along the pipe that listen to it and then interpret it.

Patrick: For those members of the audience who remember the dial up internet when it came about, this is like the [PATRICK VOCALLY IMITATES DTMF CODES, POORLY] sounds literally getting communicated through the liquid that you have pumped down, the surrounding dirt, this muddy slurry, to get acoustically decoded at the bottom of the well – to get turned into reliable commands for running the equipment down there, and to take telemetry from the bottom of the well and send it up so that people and algorithms can make decisions on the next set of commands to send down. We reinvented acoustics through mud. This is wild to me. I love this.

[Patrick notes: The sound of a dial-up modem connecting is auditorily jarring and also one of my most cherished childhood memories.] 

Austin: Yeah, it's pretty slow. The survey is just a couple of numbers, really, and it can take a few minutes, though they've gotten faster. And thankfully, the algorithms have gotten much better because even, I don't know, 10 years ago, sometimes you'd have to get multiple surveys because it wasn't quite reading right.

And then, you know, the techs would be out there fiddling with the microphones, and then they'd get mad at the drilling rig company, like, “Oh, it's your pumps.” They always blame the pumps for messing up the surveys, cause you know, everything out there is time-based in how you pay – so if this survey takes 10 extra minutes cause they are fiddling around, everyone's angry because the whole operation is several thousand dollars per hour. It's kind of expensive.

The economics of drilling operations

Patrick: This has been one of my fun learnings about the economics of drilling. Apparently, in addition to raising children, it takes a village to drill a gas well – I think you quoted a number of something like 50 different vendors on site that are providing either expertise, or equipment, or some combination of the two.

A lot of that is very expensive, very specialized equipment, which costs hundreds of thousands to millions of dollars, and the cost of renting it for a day is in the five figures. When you add up the number of expensive professionals and expensive capital investment that needs to be on site, that per hour cost is, as you suggested, thousands of dollars. A two-minute delay because, “I can't get a reading from the bottom of the well because someone is making too much noise up here. Can you shut down that machine so I can take a reading?” It's like the meter is running, and we are very good at keeping an eye on that meter because that ultimately determines the profitability of the well years down the line.

Austin: Yeah – usually depending on what phase you're in and what you're doing, you're spending between thirty and fifty thousand dollars on a normal day in drilling; it's probably a little higher now, that was like ten years ago.

Patrick: We sometimes throw around crazy numbers in Silicon Valley for what it takes to bring products to market. You can get started with an AWS account and spend $5 a month on servers or whatever, but on the other hand, getting software into the hands of thousands, hundreds of thousands of users costs what it costs, and those numbers are not small.

[Patrick notes: Useful numbers to know: YC tries to be the first check into startups at $500k, a McDonalds costs about $2.5 million to open, a Series A venture funding round and a small San Francisco apartment might be about $10 million, and…]

Patrick: But round numbers here, if you have, like, “the typical oil and gas project,” what does the typical oil and gas project cost all in?

Austin: If you're just drilling the well, then it's highly dependent on depth. The ones I was drilling were relatively shallow, [and] we were spending around $3 million on those – that's pretty low-cost. The deeper horizontal wells, especially now if you're talking about doing three miles or something like that, you're looking at more like $8 to $10 to $12 million.

So they get pretty expensive, but that pales in comparison to offshore, where you're talking about hundreds of millions of dollars. Actually this is one of the major regions that, going back to that peak oil thing, like, why did everyone in the industry even believe in peak oil? So much of it was that we had depleted all the easier onshore stuff.

We really are getting just like blood from stone on the onshore US lower 48. If we could just do our technology we had in 1980 with the same ecosystem in the US, we could produce, I don't know, a crazy amount of oil. It would still be, like, $30 a barrel or something in today's prices, you know, same as in the good old days.

Safety considerations in onshore vs offshore drilling

Austin: But just because of the bad regulation, the lack of a good ecosystem and all that, it's very, very expensive to drill elsewhere – actually, we end up preferring to do these large offshore projects that are hard to steal, because you have to have the big fancy stuff to even operate them. And so the offshore stuff is such a long cycle, and it's so hard to do, there are so very few people – I've told people before, it's more like doing space.

Patrick: That was exactly the metaphor I was going to go for. It's like the sea. It's an environment that is intrinsically hostile to humans and all works of humanity, and yet somehow we made it work anyhow. You would probably know this better than I do, but that, presumably the number of lives lost, on, offshore oil rigs – both as a percentage of the energy supply and as a percentage of total lives lost in the industry – it's probably a lot more than industrial accidents that happen when we have our feet on solid ground.

[Patrick notes: We spend approximately a hundred lives a year on oil and gas directly, against about ten a year for e.g. coal. As mentioned in my conversation with Casey, the indirect death toll from coal burning is much higher than the direct death toll among miners.

But I think it is important to acknowledge our civilization depends, very directly, on asking a relatively small number of people to do dangerous work in inhospitable environments.]

Austin: Yeah. I mean, they ended up being pretty safe because everything's over-engineered, just like space. (I think the Deepwater Horizon in the Macondo Prospect would be the exception there. [Patrick notes: 11 direct deaths and very significant environmental damage.]) If you look at worker safety, worker-hours, I bet it's much safer than onshore.

[Patrick notes: Somewhat surprisingly to me, the most likely thing to kill you on an offshore drilling project is the helicopter ride.]

Patrick: Oh, wow.

Austin: Onshore is more wild west. I don't know that that's true, but that would absolutely be my guess: if you're looking at recordables per worker hour, it would be safer offshore than onshore, possibly with the exception of the helicopter ride out there.

Patrick: Yeah, I think one of the things that is vastly underestimated about the oil and gas industry from the perspective of consumers [and] everybody else in the world, is that it's heavily professionalized, very numerically-oriented; “We have geeks too. We have spreadsheets. This is a capital-intensive outpost of Capitalism Inc.; we actually do track worker safety very rigorously” – versus, I think people's image of the industry is sometimes dominated by, like, the popular conception of the workforce as being very rural with every kind of politically-freighted image that comes to mind in the United States with regards to that.

Austin: I would say there's everything in between. It ends up being a very diverse industry. At the big companies, it's absolutely highly regulated and extremely professional, but there's also a lot of mom-and-pop action that's a little bit sketchier and everything a little bit wilder.

It really just depends on where you are and everything – offshore would be the most professionalized, and then some shallow little oil field out of the way would be the least, so where it still looks much more like you might imagine it in the movies or something.

Discussion of small-scale "mom and pop" oil operations

Patrick: So I'm curious because I don't think I have good intuitions for this. What does a mom-and-pop oil or gas operation even look like? Like, you know, if you get home to the dinner table someday and say, "Hey honey, I think I want to hang out a shingle and do petroleum exploration," what is the next step there?

Austin: An operator (in the term in the industry) means you are in charge of operating an individual well or lease. I think from a business standpoint, you can think of it like a single-family home rental or something that's just a little more technical.

And so you, you absolutely have people that, say they are a pumper – which is the person that goes once a day and checks the wells – you can get it where they own the wells that they pump. Maybe they own like 20 wells, and they're like old tired wells that maybe were even just given to them because they didn't make money for someone that has more overhead.

That's how mom-and-pop it can get, where it's just someone that goes and checks the wells, and they have a few little wells they go to. Then it just goes up from there, kind of like with the distribution where there are a few companies that are very, very large – but there are limits to how large it can get in certain ways, especially onshore, just because the overhead starts getting to you at some point.

It's kind of the same thing, like, “how can single-family rental mom and pop things exist?” – it's kind of the same phenomenon there.

Patrick: So my understanding of the industrial organization of the oil and gas companies is we have some of the largest companies in the world that are called the “majors.” As you've explained, they basically have exclusivity with regards to offshore exploitation because no one else can afford a $500 million buy-in to get a deep-sea drilling platform made. 

Onshore, we have some combination of the majors providing capital or services to the industry but maybe not running each individual well themselves; just like we see in real estate, there's a spectrum of mom-and-pop operators to mid-size regional operators, up to (in the places that reward scale) the majors that come in and attempt to put billions of dollars of capital and tens or hundreds of thousands of people to work on projects.

The impact of oil on local economies

Patrick: Alberta, taken as a system, probably had hundreds of thousands of oil workers at… I don't know if it's past the peak or not, but when a particular region becomes hot for a number of projects for a number of companies, that can be very economically transformational for the region in a fast order – [that] is something I've picked up from reading the news. Does that match your expectations?

Austin: Yeah, I mean, ideally you want it to be not a huge portion of your economic growth. I did my work in Western Oklahoma – there's a sustained difference in service costs between there and, say, West Texas, with Oklahoma's being much, much lower.

The real difference there is that we have this large labor pool relative to how many workers we need available, whereas if you have North Dakota with the Bakken or West Texas, where they have just way, way more labor needed and there's not much labor pool existing there, it gets very expensive. You get these boom-bust extremes that are just not very helpful. You have an intense amount of activity that just breaks everything compared to what you were doing before for three or four years, and then that tends to be it. Once you've drilled the wells, then that's it, and the rigs have moved on to the next place.

So it's quite unpleasant. And a lot of it, you know, you don't even end up building local housing or anything, you just bring in man-camps, trailers and stuff like that – and then you try to go to Pizza Hut and there's like no one to work at Pizza Hut, and there are crazy lines, it's just really unpleasant in many ways. 

People are probably the happiest right after the drilling stops, because that's when, you know, if you're a royalty owner, you're still getting pretty big checks, but all the activity has gone. During the peak of the activity, oftentimes people are quite unhappy – especially people that aren't getting royalty checks.

Patrick: And typically speaking, there will be a lot of outsiders moving into a rural community in a fashion that, as you said, is effectively transient and heavily disruptive to local activity – prices go all out of whack for things that have been mainstays of the local economy, et cetera.

And yet, again, the lights are on and we were able to do this podcast because the energy economy works – so, you know, we take the good and the bad together.

Austin: Yeah, definitely. It’s the nature of the industry. 

There really is this aspect of fracking that actually helps that to a certain extent: when you have the big offshore, they have such long lead times that it makes the supply-side just not very elastic. Then you also have the problem that the platforms are so expensive, they plan to have flat production for years and years at very low costs.

Whereas fracking is much better for the market because you can pick up a rig and go in as little as 60 to 90 days; you can see a very quick response to the market there, as opposed to four or five years for the offshore.

[Patrick notes: Speaking of complex single-digit-million-dollar engineering projects which immediately start creating value in the world, the construction timeline for apartments in Tokyo is generally ballparked as “number of floors, plus one, in months.” Slightly longer than that to arrange financing and permits.]

Austin: Then when the depletion rate is so much faster, you produce all your oil in the first year or two, or most of it, at least, then there's this automatic balancing mechanism on the back end; if you have too much production it just quickly tails off. 

I think that ends up being super helpful, and it also combats the issue the industry has with capital costs because the capital cost for oil and gas is extremely high.

It's capital intensive, so if you didn't have that quick production that allows you to turn your capital base over very fast, then it just wouldn't work – there was no way you could ever finance it or anything like that. It'd be back to the supermajors that are doing giant projects very slowly, and then we all suffer boom and bust high prices in the interim.

The lifecycle of fracked wells and their long-term production

Patrick: So, out of curiosity, what happens to a fracked well after, you say, if we get 95 percent of the energy out in the first, two years, then maybe it gets flipped to someone who has a cost structure such that they can maintain it for the next couple of years? When it becomes totally uneconomical, what is the next stage in the life cycle of that equipment?

Austin: The operating costs once you're done are so low that even if it's producing at a much lower rate, it can go on for years and years and years. They make these charts, how much oil is being produced from wells in what decade or whatever, and there's still a pretty significant amount of oil being produced from wells drilled in the fifties and the sixties.

It will continue to be that way for fracked wells. The decline rate slows down after the first couple of years, it becomes more like eight or 10 percent a year or something – and then that's an exponential, so it just stays steady, steady, steady after a while.

Patrick: Curious, what's the physical constraint in nature that causes that?

Austin: It's the pressure; you're depleting the pressure. You’re also getting the oil close to the well bore, so now the oil is traveling from further away, and you have lower pressure – that's how you get that exponential. It can be more complicated than that, especially in fractured reservoirs, but if you're thinking about a clean reservoir, you would have more like that decline, steady over years.  

It's actually a way people get into trouble: when you're getting your bank line, you draw these curves of what you think your future production is going to be, and that's how you estimate what the well is worth; people try to get a little too aggressive on that. Or, just even if they have new wells, they try to draw it more aggressively before they have much data, and then they go drill a bunch, they have several years of production on those original wells, then it turns out that they weren't any good because they depleted faster than they had expected and all their margin for profit went away, and then they go bankrupt. That’s a common modeling error.

Patrick: The financial infrastructure geek in me loves the part about banks because it turns out that one of the reasons that the United States has as large a banking sector as we do is that banks need to specialize to some degree in what projects they're willing to loan to, underwrite, etc. – and so you can't just go to any bank of the 6,000 in the United States and say, “Hey, I've got some petroleum engineering documents to show you. Can you write me a loan against a well that doesn't exist yet?”

[Patrick notes: Actually about 4,500 banks and 4,500 credit unions. Consolidation continues to happen.] 

Patrick: Who are the banks that one typically goes to for this sort of thing?

Austin: It’s very, very specialized. There are not many banks that do it, especially not many banks that will do it for smaller companies. It tends to be regional banks that have deep roots in oil and gas – one of the biggest energy lenders would be Bank of Oklahoma, owned by George Kaiser, who also owns Kaiser Francis Oil Company, so he has a very good understanding of the oil business. 

But they also have these heuristics that help them. They will only lend you half the value of a well, and the well has to be a certain age – they won't lend you value on new wells. They're also constantly tipping the scales with other little things, like, if the oil price shoots up to a hundred, they will not allow you to use that oil price for the modeling – you'll have to use $70 or $80. 

If it seems like expenses are too low, they'll just add to it – lots of things like that to try to fight that natural human optimism (and of course, people trying to fool [with] the data a little bit). Even with all that, it's just hilarious – they may not get a full recovery if things go bad, like a price drop.

We never usually used our bank line, so I didn't spend too much time with it – and then they wouldn't even get to us to update it because they were deep in the weeds with all these other people that had their bank line maxed out. (Of course, now oil went from $80 to $40, and they have to figure out if they're going to make them sell properties, or what they're going to do to try to take care of it.)

But even with all that extra padding, it still ends up being tough. One of the times that they took away all that padding, like in the eighties, we precipitated a wider national banking crisis.

Financing in the oil and gas industry

Patrick: If I recall correctly, the savings and loan crisis was heavily tied up in commercial real estate in some oil boom states, including oil wells themselves – or was that a different financial crisis?

[Patrick notes: Their larger problem was industrial-scale control fraud. It’s a wild story.]

Austin: Yeah. I think before 2008, I think maybe the largest bank failure in the US was Continental Illinois, which was basically completely done in by oil and gas lending. One bank originated all these loans to several others – like Seafirst Bank, and there was another one in Michigan that was a big failure – and it was not too far from where I grew up, like a mile. It was just a mall parking lot bank.

[Patrick notes: Seafirst Bank was rescue-acquired by Bank of America because it was insolvent after multifaceted commercial dealings with Continental Illinois went bad after Continental’s failure.] 

They got into this loan origination and all the loans were bad. They were loaning money based on acreage – which is, like, an absolutely should not do, because acreage values go from very valuable to zero very quickly as the prices change and different tastes change. 

Then we went back to normal lending standards. There was actually a point in the late 80s where you couldn't get credit, so you could make money buying oil wells at auction with credit cards. There were lots of people in that time period that were actually taking out as many credit cards as they could and buying an oil well.

Patrick: So, presumably, there were oil wells going for, like, $25,000, which, I suppose eventually, you know, It is a numbers game. Like if you have, for the same reason people were buying like, you know, GPUs to speculate on Ethereum back in 2017 or whatever. If you can buy an asset for $20,000 that produces cash flows of $1,000 a month in the indefinite future, and you are confident about those cash flows, the math pencils; you just have this enormous system that makes it economical to continue operating assets at $1,000 a month of cash flow.

On the one hand, that's surprising; on the other hand, like, “welcome to real estate in the United States.” There are many properties that operate on like $1,000 to $2,000 a month of gross rents, and a huge portion of the economy is devoted to servicing that. 

It's one of the – I don't want to say grubby, that isn't quite the right connotation, but like, (1) it's a thing that people don't usually have their models, and (2) you don't have all that margin to play with to support the actual physical operation of that on a month to month basis. You're really close to the edge. It's a profitable business if you do everything right – and doing everything right will often be quite unpleasant for many of the people involved.

[Patrick notes: I gesture broadly at the history and present practice of landlord/tenant relations.]

Patrick: Anyhow, talking about capital stacks for this: if you can't get bank debt for new oil and gas wells, presumably then you have to fund it with equity. My expectation from background real estate knowledge is, probably you have some amount of common equity with some amount of preferred equity, where they're getting like 9% annually plus some share in whatever the well ends up producing.

[Patrick notes: In real estate that would be “participating preferred”, which is definitely not the only type of preferred equity.] 

Patrick: Then presumably after the well has some – I guess it's less relevant for fracking because you get all your money back in two years anyway, or most of your money back in two years – but for traditional projects, you would get it to the point where you have production data to show a lender and then take out a loan against the anticipated value of the well and then repay some of your equity investors, or buy them out or similar and then ride it into the future. Is that the rough model?

Austin: Well, there are many models, but probably the most traditional and on the smaller side operator, is you identify a place that's good to drill. What you do is go around – they used to call it doctor and lawyer money – doctors and lawyers, and say, "Oh, we're going to drill this here."

Because the ownership of the well is determined by who owns the mineral rights, generally, you're going to be leasing the rights from the farmer that doesn't really want to participate in an oil well; you actually assign those rights to these individual companies – it's kind of like a mini corporation. They actually own that.

Unique aspects of US mineral rights laws and leasing practices

Patrick: This is actually to my understanding a quirky bit of US law. In most countries, national resources are owned by the state until the state in its beneficence allows you to exploit them. In the United States, the historical understanding is – what's the phrase, “from the center of the earth up to the sky?”

Austin: Yeah. (chuckles) That's actually probably the main reason why the US is so dynamic in oil and gas, why we have basically invented every oil and gas technology and method since about 1860.

So what tends to happen is that those doctors and lawyers will basically pay a little extra.

Usually the old school is like the person who put together the deal keeps one-eighth, and everyone else pays for their eighth or whatever their share, and then a little more, they also pay for your eighth. So you don't have to put much money down to drill.

[Patrick notes: Many forms of entrepreneurial capital formation recapitulate this structure, at slightly different percentages and with slightly different nomenclature. It is called “the promote” in real estate, “the carry” in VC investing, and whaling agents had something similar. Broadly, passive investors compensate the coordinator for bringing them together, developing the deal, and seeing it to fruition.] 

Austin: (There are some fun scams where people sell like 12 eighths of a well and then drill a dry hole, so you just collect the money and don't have to worry about actually having people to pay afterward and them finding out.) 

So you drill the well, and everyone gets their share of the well. A lot of times, there will be a back-in after payout: once everyone earns what they put in, then maybe a slightly larger share will go to the operator – that's pretty common.

Patrick: That is quite similar (with different words involved) to the way that we do, small- to mid-scale real estate development (as someone who grew up with a dad in commercial real estate for most of his life.)

The process of setting up and funding new drilling operations

Austin: Yeah. It's very similar, just with more equity. Really for drilling, unless you're borrowing against your existing properties, you can't borrow money to drill – at least not in sane times.

As things get more advanced, we also got a hedge fund-type model. The original company was called NCAP, so people just call it NCAP now; it's like a Kleenex situation. What they would do is find someone, say they were someone that had worked at Exxon, and they were knowledgeable, an engineer – you'd get together a team of an engineer, geologist, and a land man that were all friends at Exxon.

They go out and start their own company, and NCAP gives them like a hundred million dollars. What you try to do is you try to go lease up a certain amount of land. 

This actually started working really well with fracturing because, old school, your feature may only be a few acres, but if you find a good fracking play it might be tens of thousands of acres. 

So you could go lease up some land, drill like 10 wells to prove it up, and then you sell it. That's the NCAP, more like private equity, venture capital-type model.

Patrick: So to draw out the implications for non-specialists, you've gone to farmers, you say, “we don't exactly know what is under your land right now, and so in expectation, mineral rights are worth nothing; we're going to pay you substantially more than nothing to buy those rights in the future.” Then we're going to drill and learn whether there is nothing, or something that is vastly more than nothing. 

In the case where it is vastly more than nothing, I now own the rights – where I've leased it for some period – and I will sell my lease on to someone who is going to drill not just ten exploratory wells, but fifty wells, a hundred wells, two hundred wells to actually make productive use out of that, and I will sell it to them for far closer to its “true value” now that we know its true value.

Austin: Exactly. It's very much a land-increasing value play. An interesting thing you might like about it: how do you deal with the mineral owners that don't really understand oil and gas? 

You tend to give them a lease bonus, an upfront payment that's guaranteed whether there are any well hits or not, but they also get a share of the oil revenue that aligns them if it does end up being really good. You might pay them, in a hot area, several thousand dollars an acre, and then also they might get like three sixteenths of the oil and gas value that comes out. It’s an interesting way to handle that uncertainty.

Patrick: For the four or five members of the podcast audience who have not been farmland owners in the past, what's the typical value of a random acre of farmland in Kansas or Illinois, so we can put a lease payment of a couple thousand dollars for mineral rights next to that value?

Austin: Ranch land can be as low as a thousand or $2000 an acre; good row-crop, Iowa land like five to $10,000 an acre. So a lot of times you're getting, you know, as much as your land is worth, if not more than you actually have. 

The great thing for the farmers is they don't really care how profitable the wells are – even a well that's a dud fracking-wise for the company that drilled it still gives them a lot of money. So, yeah, it's like people go from poverty to, like, winning the lottery overnight kind of thing in some of these most prolific areas.

Patrick: Interestingly, I've heard that this often helps with the political economy of this. If you have a particular town and a geologist goes out to it and says, "Good news, bad news, folks. I found a lot of oil under your town," people who are already plugged into the socio-economic infrastructure of town – the major landowners, etc. – will see the value of their land immediately skyrocket contingent on the political economy of the town embracing the fracking project.

So they pull all the levers, go out to the town hall meetings, work the PTA, etc., to say, "Hey, this is really important for the town. This is going to bring in a lot of money. People are going to spend a lot at the diner on Main Street, etc. – we need to get this done," and that results in these projects actually getting done in the United States – at least in some places in the United States – where, because of the nationalization issue and the political beneficiaries being much more diffuse in other nations, you cannot successfully win the hyperlocal politics to get projects actually permitted, approved, and started in many places in the world.

Austin: 

I would say that doesn't end up being quite right because we still can't win the hyperlocal politics – but what it does do is that you have enough interested voters at a state level that allow you to remove all those local barriers. In almost every state, the mineral domain is dominant towards the surface domain, so you can use a version of eminent domain to force a surface location to happen, like say, "I'm going to drill on your land whether you like it or not,” then you have to go through that really annoying process – but you come to an agreement anyway because you'd rather not go to court. 

Many of the small permitting things are like a completely separate domain for oil and gas – it's like oil and gas is exempted, or it's just different to fit with the oil and gas industry. There’s very little ability for local authorities to stop it; they just don't have the jurisdiction, like the state has taken it away. They even have a thing where – this is especially true in Oklahoma – like this concept called forced pooling. 

If I have a unit where I'm going to drill my well, this area – I might have dozens to hundreds of mineral owners that I have to contact, and you can see a world where one mineral owner could hold you up, some angry person could buy one of these minerals for a relatively small amount and just stop you. So we actually have a mechanism where if you can just get one acre, then you can force everyone else to make a decision and drill – they have to either sell to you or participate in the well.

So it's more about, you know, we get out the steamroller as opposed to winning some local PTA. A lot of times, there are lots of people that don't own minerals, and they're pretty angry. If you had any kind of local process, they'd make it go crazy – I've had some enemies in there, and it can get pretty extreme. 

This was thankfully not my company's lease, but some people related, they were this wolf pack that was after us. Some guy's nephew, he actually tried to burn down this oil site – he was going to stick a flaming rag in one of the tanks, but unfortunately for him, he didn't realize, you know, the vapor pressure for oil is quite low, and so he blew himself up along with the site before he even got to the tank. 

And the whole reason he was doing that is because if you're the mineral owner and you sign a lease, then, you know, as long as the well is producing, the leaseholder now has the rights, the mineral owner gets paid based on the old terms – so when this new fracking thing would come on, a lot of times you were leasing land that already had the old vertical still producing.

If there's a lease bonus, or any extra juice going on, then the old mineral owner doesn't get any of that lease bonus – it's the guy who owns the vertical well that gets all the lease bonus. So they wanted to destroy the well, so then the lease would go away and then they could renegotiate a new lease. You have like lots of crazy stuff going on like that – if you had local processes, it just wouldn’t work. 

Patrick: One thing I think folks might not appreciate about infrastructure is that infrastructure often operates on the time scale of decades, sometimes centuries, and contracts are also capable of being enforced decades and centuries later – which is not people's natural inclination about the legal system.

I'll have to have my father on the podcast one day. He was once a paralegal working for a railway, and the railway had negotiated a lease with another railroad back in – I'll make up a number which is roughly accurate – like the 1860s. This was literally negotiated over telegraph because we didn't have airplanes to send someone out to San Francisco to physically negotiate back in the day.

So in the 1980s, he went down to archives to pull out old telegraphs from more than a hundred years ago to send a demand letter on behalf of a railroad that was almost defunct, but still had some very valuable paperwork available to it.

“We negotiated a formula in good faith 120 years ago. That formula says you owe my employers $100,000 this quarter. Where the heck is our check?”

[Patrick notes: The apple did not fall far from the tree, eh. And I only learned of this story in the last six months.] 

Patrick: When the lawyers discovered that he had sent that letter, they were like, “Atta boy, that's how the legal system works,” even though that is against many people's inclination of how it should work.

Anyhow, I'll have Dad tell that story some other day, but… when you make commitments, they're kind of binding on you and your successors, it says that in bold font at the top of them – they have consequences for decades down the line.

Environmental concerns and groundwater protection measures in fracking

Patrick: Let's see, what are other fun aspects about this?

I tend to think it's overblown, but you've been educational to me on many things, so maybe you can be educational to me here: A lot of folks seem to think fracking will poison groundwater, etc. I assume we've gotten better at that story over time, but what's the current state of play?

Austin: When you're putting the fluid down into the reservoir, that's a mile or two down; it doesn't interact near the surface where we're getting well water or groundwater that we're drinking. Actually, the water that's already down there is extremely salty, oftentimes two or three times saltier than the ocean, and radioactive, and extremely nasty – a lot of old oil field tanks become hazardous waste because they are too hot to go to the junkyard. So not stuff you'd want to be messing with anyway. 

All of the groundwater protection evolves to stopping surface spills, because we don't want to spill that water on the surface, like from a tank – that would cause issues. We also don't want to have any interaction with our well, drilling through that groundwater layer for the first few hundred feet; there’s something called a surface casing and that is specifically there just to protect the groundwater. There's all sorts of testing that is done around that.

It was also a concern when we do saltwater, where we're injecting that fluid back in, so they do things like if I want a permanent saltwater well, I have to go to nearby wells and get water samples, so we have a “before,” and if there's ever a question we can go back and resample those wells and compare them to the old samples that we had – all that's stored at the state regulatory commissions. 

That's actually the biggest purpose of the state regulators, protecting groundwater – and when you have an issue, it’s fatal for most companies. It's kind of one of those things that will blow up to the size of your company. For example, BP with their offshore disaster, it cost them 20 billion or something – but if they had been a smaller company, it would have been the value of whatever the company's worth.

Patrick: So I suppose that the long story short is that since this is an existential risk for companies and there are engineering solutions to it, you attempt your best efforts to get the engineering right because you would prefer to not be penniless at the end of the day. Are there any backstops? In case the company blows up, the equity owners lose all their money – is there an insurance fund or insurance companies involved that would attempt to send out checks to the people whose surface wells are now brackish?

Austin: Yeah. Typically, what you have – each state has its own thing – but you pay a portion of your revenues into a fund that cleans up abandoned well sites and stuff because occasionally you do have something where some company goes broke. (At the end of the life, you plug the well – you put cement in there and make sure nothing can move from down at the bottom up to the top, and then you clean up the surface area around it. Occasionally, there are companies that just go broke, and you have to clean these up.) You also have to have a bond a lot of times as an operator that helps you do that. 

That's really how that kind of stuff is handled. Super fun.

The physical footprint of drilling operations

Patrick: So I’m curious, what does the surface footprint for one of these look like, either when it's in operation or after operation? You know, should I be imagining some giant industrial facility or should I be imagining a 10 by 10 square of concrete with a tower on top of it?

Austin: The old school is very low footprint. Even when the drilling rig is out there, you're talking about maybe a little more than an acre – that is what you're working with on the surface location. It actually gets way smaller once you move the drilling rig off – you're down to the size of a small yard. 

On the surface most often (in the U.S. lower 48), you have the pump jack, or it's just a little wellhead, then you have a few tanks that are usually right next to the road, so they're out of the way. That's all together, like the size of a yard, essentially. 

It's a little bit bigger with the horizontal wells because you have a much bigger location needs, you have more equipment that you’re trying to get on there for the bigger wells, so maybe it's a few acres, and it stays a little larger. 

Each individual one is small, and it used to be you'd only have like one or two wells per section, but now with the horizontal wells, sometimes you have like 20 wells per section and they're generating just an enormous amount of fluid, so as an area, it can become somewhat industrial. Along every road, you have this well site with a pad with 20 wells on it, then you have to have your saltwater disposal facilities and all this kind of stuff; it can become somewhat industrial in the most intensively drilled areas.

Patrick: We've been talking about acreage, which is a natural unit of measurement for a relatively small section of the world – Americans who've been involved in real estate, farming, and oil and gas development. For folks who aren't familiar with how large an acre is, it's approximately 200 feet by 200 feet, or (plus or minus) 65 meters by 65 meters – these are not gigantic bits of area that we're talking about. You've been in buildings that are larger than an acre.

Learning curves in fracking and geothermal energy extraction

Patrick: So another concept that I learned from you is the concept of learning curves, which turn out to be generalizable in a variety of energy fields – we had a learning curve associated with solar productivity, for example, where the photovoltaic cells got progressively cheaper to produce, and progressively more effective per dollar invested over the course of 20 years of technological development on them.

[Patrick notes: Discussed in substantially more detail with Casey Handmer.]

Patrick: You've explained how some of the learning curve works in fracking to me; what are some concrete examples of the kinds of engineering where we just did not know to do that back in the day, and then we tried some things and now are more effective for it?

Austin: The most durable one by far is with the drill bits. Our drill bits, we used to have these “tricone bits” – roller bits where they have these little buttons on them, maybe some diamond on there to try to make them a little bit more durable, but you could only drill with them for 50 hours. They were slow, less than 50 feet an hour; a lot of times more like 20 or 30. 

We switched to these PDC bits, which are more like a solid one-piece thing. They have lots of these little diamond cutters on them – one drill bit may have 100 diamond cutters or 50, depending on the design. These bits are so much more durable, and they've gotten so much better basically all because of the cutters.

Diamond drill bits are not quite forever

Austin: Diamond is very hard, so the abrasive formations can't wear it away – but of course it's very vulnerable to cracking and impact damage. Over the years we've gotten much better at making the diamond cutters where they are extremely hard, but less sensitive to impact damage.

It allows you to be way more aggressive drilling – you put more weight on it and go. Now it's normal to be drilling at hundreds of feet an hour, just flying. It's actually become a situation where the on-bottom drilling has sped up so much that you spend most of your time running casing and doing other nonproductive activities.

Patrick: I'm curious, do we use natural or artificial diamonds on the drill bit?

Austin: Last year I actually got to go to a factory where they make these – it's one of the craziest things I've ever seen. It's artificial. They take the diamond powder and mix it with a catalyst, cobalt usually, then they have these giant presses that press it together up to a million pounds – and they also have to heat it, so while they're doing this, they also heat it up to, like, 1000°C or something. It's just this absurd machine, absurd engineering technology. They actually have them in these giant vaults because sometimes they break and pieces fly everywhere. 

Most of those factories are in Utah by Brigham Young, because one of the early pioneers was there and it led to this whole cluster.

Patrick: It is one of these wild fun facts about infrastructure:  It is so useful to maintain the uptime of a drilling operation that we have factories which presumably cost hundreds of millions of dollars to create the factory directly, plus hundreds of millions to a billion dollars in R&D to discover exactly the process that successfully melts the cobalt and the artificial diamonds – to then put presumably many, many diamond rings worth of diamonds down a mucky, oily, disgusting hole to maximize the efficiency of this operation. 

(But hopefully, you get most of them back. You don't want to leave substantial portions of your equipment down the hole, except for the parts that are designed to be down the hole.)

Austin: Yeah, and I think their R&D is all destructive; they spend like five or ten percent of what they make just going through testing to incrementally improve it each time. 

That's the biggest learning curve, and it looks like a learning curve where it's this incremental improvement: the more footage you drill, the more data you have, the more cutters you buy, the more testing they do. Then it gets harder and harder to keep improving them.

Where fracking goes from here

Patrick: Should we expect further incremental improvements? We know what the last 20 years of history look like, but in the next 20 years in the future, should we expect to get more oil and gas resources from what we currently know is available? Or are we approaching some physical asymptote yet?

Austin: I think it’s really a question of, “where can we go?” We have the same problem: fracking really hasn't made it very many other places. I think maybe Argentina of all places is maybe the only place you could say that has anything that looks like a U.S. fracking industry, but I think they still are importing oil and gas, which is kind of embarrassing considering the enormous resource they have down there. 

So I think there's still tons more gas to find. Fracking works much better for gas because the molecule is small, and it fits through all the little cracks – as opposed to oil, which is kind of sticky. 

I think the future for oil probably looks much more like Canada, like oil sands. Another technology that has improved alongside fracking that doesn't get near as much hype is steam-assisted gravity drainage. That's what they do in Canada; that's what we should do in Venezuela, if they weren't so crazy. There's enough oil in those fields in Venezuela and Canada to produce at our current rate for hundreds of years or something, so it's not really a question of resources.

If we got much, much better at steam-assisted gravity drainage, which has been happening over the years, there are probably a lot of resources in the U.S. too – that are much worse than Canada or Venezuela, but you can actually drill, build pipelines here and all that kind of stuff that would allow us to take advantage of it.

I think for oil, the future is probably heavier – and gas, we still have lots of gas reserves. The question is, are any other countries going to actually pick up this technology?

Patrick: One thing I've been learning as I've been on my geothermal bender is that the cross-application of fracking surface engineering technology to extracting geothermal heat resources turns out to be much higher than my naive expectation of that.

That might be a discussion for some other day, but do you just want to give people the quick rundown of, “oh, that might not be a totally pie-in-the-sky idea anymore”?

Austin: Yeah. It gets back to the bits: generally in geothermal, you want to go hot. Where we drill oil and gas, we drill in sedimentary reservoirs; in geothermal, you're dealing with more like the granite kind of stuff, which is extremely hard, extremely abrasive. This is the big drilling challenge for bits.

Ten years ago, you just wouldn't even be able to drill it with the PDC technology we had. People were looking at doing all sorts of other crazy stuff like air hammers or laser plasma drilling – but because there are some oil and gas formations that are that hard and that abrasive, it has pushed the technology to where now you can drill granite.

And I mean, literally ten years ago, you would drill like 50 feet and your bit would be dead, and you could only drill 50 feet at 10 feet an hour. Now you can drill at hundreds of feet an hour in granite, and your bit can last for several thousand feet, what the geothermal people have done – but as is always the case, geothermal is much behind oil and gas.

So in these very hard, very abrasive oil and gas formations, people are drilling 3 mile laterals in this stuff, just flying, and only using like, you know, 2 or 3 bottom hole assemblies. 

Another key unlock: we can make the diamond bits super hard, but there's only so much we can do on their fracture toughness, like impact damage. but we have other tools now, like downhole stabilizers, that limit the impact damage. 

When you're thinking about this, you're controlling the drilling stream, you know, a bunch of 30-foot sections of pipe screwed together, miles and miles long, so there’s an accordion-type effect here, where if you adjust a little bit at the surface, it can be much more dramatic at the bottom – and there's only so much control you have. It's almost impossible controlling at the top to keep your bit from occasionally just ramming against the rock with dropping a bunch of weight on it. If you have these stabilizers downhole that prevent the impact damage, that's how you're getting these runs where you might be getting 5,000-7,000 feet or a whole mile out of one run for one BHA, for not much more cost than what we have today for a BHA. When you start looking at that, it starts looking pretty good for what you can do in granite because it's the same type.

Patrick:  So I suppose the upshot for this is, back in the day we could only tap oil resources that were either literally bubbling up to the surface of the earth where we could see it or a couple of hundred feet down. Geothermal energy that people are most familiar with is in places like Iceland, which are blessed by nature, where superheated steam literally geysers to the surface of the earth. You just have to bottle that energy for human use.

But due to crossover technology from the oil and gas industry and fracking and similar, we can go much deeper to find the heat where the heat actually is, and then bring it to places where it might be useful for humanity. We should expect that technology to get better over the course of the next couple of decades, as we improve our engineering understanding of drill bits and, you know, not dropping the multi-ton device at the bottom of a very deep hole and having it shatter on us (an important thing to do).

Austin, thanks so much for taking the time today to have a conversation. For people who, like me, might not have come from this industry but love learning about it, where can they follow you on the Internet?

Austin: I'm on Twitter at @Vernon3Austin. And then I also, the main place would be my blog. (I haven't written a whole lot about oil and gas on there, but I have lots of geothermal stuff that covers very similar concepts.)

Patrick: We'll have to do an episode on geothermal one day – some other time. 

Austin, thank you very much for coming on the podcast today. Thanks very much everybody, I'm Patrick McKenzie, patio11 on Twitter as always, and we'll see you next week.

Austin: All right. Thank you.