Jan. 21, 2023

Bitcoin Park Open House: Bitcoin Mining - What to Expect in 2023

The player is loading ...

Search Bitcoin Park in your favorite podcast app to listen to other discussions hosted at the Park!

This conversation was recorded at Bitcoin Park’s January 2023 Open House, a monthly meetup event at Bitcoin Park in Nashville, TN featuring panel discussions on various Bitcoin-focused topics.

January 2023’s Open House topic: Mining to the Moon

Recorded: January 11, 2022

Bitcoin Mining: What to Expect in 2023

with:

Bitcoin Park is a community supported campus in Nashville, TN focused on grassroots bitcoin adoption and a home for bitcoiners to work, learn, collaborate, and build.

Our aim is to produce and publish conversations from future events, so please like and subscribe on your favorite podcasting app to stay up to date.

To learn more, please visit and join: https://bitcoinpark.co and/or join our Discord where we are most active: https://discord.gg/YZ8x7mfFKj

Follow us on Twitter:

Bitcoin Park: https://twitter.com/bitcoinpark_

Matt: https://twitter.com/ODELL

Rod @bitkite: https://twitter.com/bitkite

Transcript
Unknown:

This panel is gonna be a banger. We got Jonathan Kirkwood from 10/31, Cassie Clifton from Galaxy, Harry Sudhak

from, Grid, and then Matt's gonna moderate this. So give these guys a round of applause, please.

Unknown:

We're gonna do 30 minutes. We're gonna have fun with it. I'm very excited about this panel. I think we got a great group of people up here.

So 2022.

Well, first of all, I don't know if anyone looked at the price, but we're over 18 k, so we're rich again. So we're gonna act a little bit.

Unknown:

Oh, is that for real? 2022.

Unknown:

Is that for real? That's what I said.

Unknown:

Tell me. Because over the price first verified.

Unknown:

2022

was absolutely brutal for miners,

and I think

it's hard to talk about what's next for mining in 2023

unless we first start with what went wrong in 2022.

And I know Marty said he didn't wanna pick on CoreSite, but we should just pick on CoreSite.

They were absolutely massive. They were crushing it. Right? They were the darling.

Everyone wanted to be them, or a lot of people wanted to be them.

And

now they're on the midst of bankruptcy.

They filed. Yeah. Why did that happen,

and can that be avoided, or is that just something that happens every cycle no matter what? And we will start with Harry

because he's very excited.

Unknown:

Never not excited.

I think in order to talk about kinda what went wrong in 2022,

we need to talk about what happened in 2022.

And in the context of mining, you know, to me, that really means sort of 3

key things.

The first is that global hash rate exploded,

and it exploded on top of the minor China ban. So minor China said,

get the ASICs out of here. We're shutting down this sector. This is not part of our economy anymore. We saw global hash rate get cut in half.

And and all of us with hash rate active were saying, this is incredible. Our margins are are expanding. I love every minute of this. It's gonna take forever to recover to previous levels even though price was still ripping.

And we were wrong.

So instead of it taking a long time, it took a very short period of time, all of that hash rate got either moved to the US or existing operations

that were under development in the US

hit their expansion numbers, and global hash rate ripped past the previous,

China included all time high

and continued to go further.

And that happened

when Bitcoin price was on its way up at the beginning at the beginning or end of 21 into beginning of 22, and it just continued no matter how far price bled.

So the competition for those 900 Bitcoin a day

was unbelievably

intense.

The second thing that happened was

we saw Bitmain and Microbt

deliver

machine volumes at higher efficiencies

at massive scale.

So the way that that hash rate climbed was that everybody who put in those preorders in 2021 for 100 of 1,000,

tens of exahash of units

got delivered. The machines were high quality in the s 19 j pro series, in the m 30s plus and s plus plus series. All those

machines hit their numbers. There was no s 17, you know, hardware failure debacle.

So the quality of the hardware that got delivered was really high.

And then what's the third thing that happened is that the financing structures that the private and public companies

entered into to buy all of those machines

blew up in everybody's face.

So

you've got compressed margins from Hash rate rising.

You've got machines all getting delivered,

en masse. And then you've got all the financing structures that paid for those machines

going totally south,

and punishing all of the miners who now can't service that debt, who whose collateral is now worth far less because Bitcoin priced down, miner priced down.

And so we get to the end of this year,

and everybody

is sort of on the battlefield either bleeding or dead.

Unknown:

Now we're now. Do you wanna add about, you know, Core Scientific

what, Core Scientific started, like their business plan and the fact that they pivoted

to and, initially, they were providing hosting,

and then they pivoted. They took the cheap, liquidity that was in the market

and started mining on their own, which they failed miserably at.

Unknown:

Yeah. So so why did why did Core struggle?

Number 1 is that, like, they, you know, they were incredibly large and successful and continue to be. Right? Just because they've filed for bankruptcy doesn't mean those assets disappeared. Right? An airline files for bankruptcy. The plane still sits on the tarmac.

And so, you know, so those those facilities and those assets still exist. So it's important to kind of acknowledge that along the way. But, like, why did they struggle was because

they had a lot of debt that they had to service.

Their margins got significantly compressed

because we talked about hash rate going up. But also, you know, the nature of their powers appears you know, I don't have any information about it, but what it what it seems like in all the court documents is that they had variable rate power. And,

you know, apparently, there was a war in Eastern Europe that tripled the price of natural gas for all of us.

And so you're seeing margin compression on the revenue side on the exahash, but you're also seeing your cost of goods sold triple or quadruple because the price of gas went from a dollar in MCS to $5.

Unknown:

I think the other thing that has been touched on more indirectly and just, like, more directly hit on this is that the assumption that everyone was operating on is that number continues to go up. So people took out these massive amounts of financing at basically the all time peak to buy machines

at the all time high. So now they're literally purchasing at the very top, and they're stuck holding these bags, and they can't afford it. And they're also assuming, like, they had they had these hash rate assumptions that Harry had mentioned, and they're assuming that price is at least gonna stay where it is or continue to go up. Like, no one was expecting the levels that we're at today. And so they weren't building risk tolerant models to, like, expect for this type of downturn. And so now they're completely underwater on on the liabilities that they have outstanding,

and the power costs that are rising. So they're just being squeezed on all sides. It's a macro backdrop. It's energy prices rising,

and it's Bitcoin price. So all of these things kind of just blending together to this perfect storm. And and

Unknown:

I just wanna give one number, which I think is really relevant. Because, like, we see Bitcoin at today. It's 18,000,

and, like, that doesn't feel low,

because it isn't. Historically, this is a great price for Bitcoin.

But to understand sort of the mining economics,

we are at all time lows. We're low like, we're we're making less money on a on a terahash day basis

right now than we would have when Bitcoin went to 32100 or when, you know, Bitcoin dropped to a1000,

during a previous epoch. So to understand sort of the nature of, like, where you see Bitcoin price on the chart versus where you see mining economics on the chart, we are at all time, all time, all time lows even though Bitcoin is, like, still reasonably high. Yeah. And when we mentioned cheap liquidity, these miners, large scale miners, were taking out loans at 15 to 18%

Unknown:

interest rates when, you know, Bitcoin was at 40 to 60,000.

But

at this rate, they are, you know, unable to service those loans. And so that's when she's talking about the liabilities or, underwater.

Unknown:

Is the

is the bloodbath over yet in in mining or In mining. Is the global bloodbath over yet? Like

It's it's tricky. Right? So, like, I think that energy prices, like, I think the I think the top is in.

I think we're gonna get cheaper power from here.

And so that's really good,

if I'm right.

I think that, you know, is the bottom in for Bitcoin? No idea.

Is the top in for Hash rate? I think that we're we're getting pretty close to the marginal levels. I guess to reframe my question,

Unknown:

obviously, every miner's in a different situation.

This cycle

all the cycles in Bitcoin, if you if you stay around long enough, they they tend to blur together. There's a lot of similarities, but there's also major differences that have this time is different. Right? There's major differences every cycle. This cycle, the major difference I saw

was

massive ASIC collateralized loans,

usually by public traded companies or at least very large companies,

that were able to access this cheap external capital, right, finance it,

and and really expand their operations significantly.

Is the bloodbath

over and I know

I'm saying this to 2 people

that work with large mining operations.

Is the bloodbath over for those large

mining firms, or is is that still a story that's being told? Like, when we look back, like, are we mid chapter in the history book, or is it towards the end? I don't think we're at the end. I think we're towards the end.

Unknown:

I think that like 2023

is really gonna be a year of survival. Like, no one really has an idea. Like, I don't have a magic ball to tell you what the price is gonna be. And unfortunately, like, we're all just kind of like victims of the price until it goes up. And so at the end of the day, like, it's going to be a struggle to survive. And people there are a lot of miners that will need an inflow an influx of cash in order to survive.

And capital markets are effectively closed unless you have friends and family and really really great relationships in that arena.

So it's gonna be really tough for a lot of people out there. There are a lot of people that are like struggling to keep their heads above water and are are doing it really gracefully, honestly.

And I'm impressed with some of them. But at the same time, like, I don't think it's quite over, but I think most of the shakeout is here. I will say I think there will be additional supply of a 6 coming onto the market.

Not just from like the Celsius of the world, but it's like, they're still manufacturing a shit ton of these machines, like over a 100,000

manufactured a month. And so it's just like, what does that do to the market? So we'll continue to see, like, all all time low ASIC pricing, which will be good for growth in a lot of companies that are well positioned.

Unknown:

That makes a lot of sense to me.

Do we think

this idea

if you ask most people in

2021

early 2022,

they would say the I think they would say the majority of Hash rate will be dominated by large

firms, large mining firms, that that Hash rate will centralize.

The firms that have access to cheap capital will be able to dominate.

Most people did not think

that the Fed would raise so hard so quickly. Most people did not think capital markets would

seize up so quickly.

I I mean, I remember Marty in January,

January 2022. He said he made the cardinal sin on the podcast. He's he's spoken absolutes, and he said there is no way the Fed will raise rates.

And I was caught off guard too, but I just didn't use the absolute.

Is that still the is that still well, like, what do you guys think? Do you guys think that that the majority of hash will be dominated by large firms, or do we do we expect kind of a distribution of hash out of,

you know what what did Rob say? Chicken Shack Chicken Shack operations. Right? I like that term.

Unknown:

Wanna give everyone else a chance to answer. I mean, my high level thesis is yes. That's largely because, like, you want the lowest cost power. In order to have low cost power, you need money. In order to have money, you probably need to have some sort of, like, large asset situation on your balance sheet. So I just don't like, you're not gonna get a PPA at a low price,

or you're not gonna be able to, like, you know, colocate with the solar miner unless you are deemed as creditworthy,

and have, like, a large operation. I'll take the other side.

Unknown:

I think over the next,

kinda short term future of, let's say,

2 to 4 years that there's gonna be a a larger push

for going,

after

pro

regulatory

locales where you've had declining

population or deindustrialization,

and you can put a, you know, 5 to 25 megawatt site

instead of a, like a gigawatt site in in some locations. And, you know, what you're doing is you're providing a benefit to the community. And that's why it's pro regulatory because they need someone to come in and buy the power so that the average citizen

isn't,

footing the bill. So their price, their their cost per kilowatt isn't going up for their electricity bill. And I think

over, let's say, this decade, the the overall cost for power,

I think will trend up. I think we're probably trending down

right now, but,

I believe we're probably in cheap peak energy

right now, at least until we start,

doing more nuclear.

But for, like I said, 2,

2 to 4 years, we will be,

probably focusing more on 5 to 25 megawatt

and the locations that are needed for humanity to, to continue to thrive. And then after that, I think

that the sats from mining will be a byproduct

or will, you know, be just one of the revenue streams

from the heat that's generated from the mining and that, you know, as efficiencies are

driven forward,

you know, it's gonna be who can, acquire as much use out of that kilowatt

that is being purchased from either either the grid or being manufactured from, you know, a well site. You know, how much can how how much use in utility for humanity can be

obtained. And so is that heat that is being, you know, pumped to,

the the small towns or the the cities, is it you know, is that heat then being

recycled to mine more, or is it being used for some other,

use case for humanity?

Unknown:

I'm gonna have it both ways,

and say just that that

the era of sort of unconstrained

ability to stand up mining operations is over. The need for structural advantages,

has now begun.

So if you you know, what what you used to be able to do is,

you know,

buy some piece of hardware and plug it in your garage, and you're probably

not super profitable, but you're reasonably cash flow positive, and you can kinda justify it. And, you know, as long as your Excel model straight lines Bitcoin price and straight lines hash rate, like, it'll be fine. You'll pay back in, like, 18, 20 months.

And then that doesn't happen. So, you know, the the sort of the low barrier to entry

mining opportunities,

like, those are gonna be where the struggle,

will continue.

But if you're able to have differentiated access to power, whether it's through large scale creditworthiness

or positive externalities

in a community,

and there's also a bunch of other ways that you can unlock these types of of true structural cost advantages

on how you power your ASICs. Like, that's what's required to move forward at this point.

And so, you know, we're seeing a lot of people do interesting stuff. And whether that's, you know, some of the the, you know,

off grid hydro in other countries, like, that's exciting and interesting. And I don't know whether or not that's gonna work at scale, but I do know that they're generating

kilowatt hours in a differentiated

way, and that's now become table stakes.

Unknown:

You have something to add, Cassie?

Unknown:

No. I was just gonna make a comment. It's fine.

Unknown:

That's the point.

Unknown:

I saw the mic raise, so I was being I don't remember. It's fine.

So where do where do we think the greatest opportunities lie then for miners right now?

Unknown:

I mean, I think, like, the the screamingly obvious one is that the cost to the cost per tera hash is is way, way, way, way lower. Right? So you're you know, imagine, you know, imagine that that you get to buy Bitcoin,

you know, at these at these much lower levels. Like, it's just easier to be in a strong financial position

when you're capitalizing your site at 5 to $15 a terahash,

not 50 to a 150, which is basically what happened over the last, you know, the previous 18 months.

Unknown:

Yeah. The the kind of the question that I have and, you know, as as an investor in in companies that are building out the these infrastructure

is,

yes, you can

enter at these lower

price points today than what you were 2 years ago. But, you know, if you're wanting to make sure that you're building a sustainable business that's going to last not just 2 years, but, you know, maybe 1 to 2, epochs,

you know, is there gonna be enough

sats from the mining that is going to be able to subsidize

or to facilitate

the the site. I

I I really think that there needs to be

an a really leaning end into what you determined the externalities

of mining and all of the benefits that it will provide to the to the local population. Otherwise,

I don't think that in 6 years that there's gonna be enough mining reward or subsidy and reward to,

to

to economically

sustain

a, you know, a 100 megawatt, 500 megawatt site. Why so bearish?

Unknown:

So you're basically speaking towards, like, these massive brand new warehouse sites that are being built all over the place. Correct.

Unknown:

Even, like, even some of, like, our internal models,

super low cost power, like, locked in at a fixed rate, like, you're looking at potentially having to curtail, like, 70%

of the time just depending on, like, where Hash rate is. And and that's sorry. Curtail 30% of the time. So you're looking at, like, 70 70% uptime.

So,

yeah, you have to be a lot more strategic. You need to have, like, other ways that you're diversifying your revenue. And, like, at the end of the day, what you really need to do is to understand the market that you operate in. And you need to under like the power market that you're operating in. And so you need to get a lot smarter on that front to be able to survive in the future. I'm kinda curious. I'm just gonna switch it up a little bit.

Unknown:

I'm curious what you guys think.

Basic collateralized

loans,

do we are they gonna be massive next cycle as well?

Or is there gonna be changes to how they're structured?

Is was this was this a one cycle phenomenon?

Unknown:

Yeah.

So this is, like,

for context and, like, full disclosure,

the primary, like, role that I'm in is effectively, like, managing my fight for minors. So, like, people come to me and ask for loans, and I have these initial conversations and like walk them through diligence and and get a feel for whether or not they qualify.

Brandon and I in the audience here, Allstate Galaxy, we've been working on a project recently that's effectively, like, revamping

ASIC backed loans as we know them.

The traditional model that we've seen doesn't work. There's a lot of reasons it doesn't work. The primary reason being that, like, we built

a fixed interest rate, like, short term loan effectively, like, over 12 to 18 months

and offer that in the most volatile asset class in the world.

You know, there there's no pick interest that we're offering. There's no cash flow sweeps. There's no real way to derisk. And so what ended up happening at Galaxy, at least, is we ended up turning down 97% of the people that came knocking on our door asking for money. Our loans have performed,

you know, it it certainly hasn't been like a a pleasant experience for us,

and the diligence is

insane. Like there's a massive diligence undertaking. But, yeah, I would say they will look very different.

We'll be thinking about things like not just converts, but potentially forced converts if we're looking at doing them.

We'll be looking at things like cash flow sweeps, like pick interest.

We'll be looking at like min and max paybacks periods versus just flat rate term loans because

if Bitcoin price, you know, goes up, like, we wanna derisk. We wanna cash flow sweep and, like, not just sit there collecting, like, our meager little income while they're making like, raking in money and then doing who knows what with it. It's gonna have debt, you know, there's gonna be, a ton of different, like, covenants on leverage and debt to equity.

So there's there's gonna be a lot of things that we're gonna use to help derisk these loans in the future. I'm gonna do what every miner does when things go badly, blame the manufacturers.

Unknown:

And and I'm and I mean

and and I mean that seriously. Like, the reason that the ASIC backed loans

I'll keep laughing.

The reason that these loans went so upside down is because the manufacturers felt comfortable pricing

the machines

in future Bitcoin revenue

terms.

And that's crazy. Right? So, like, the the cost to produce the hardware is what it is. But the price that they were selling the hardware at and that the market was tolerating at the time, to be clear, they weren't, you know, they weren't forcing miners to do things they didn't wanna do.

The market priced it.

They were basically saying,

here's what the next 200 to 400 days of revenue look like. This is the cost of the machine.

Unknown:

But isn't that what always happens? That's always been the case.

Unknown:

Thus far. But this if I were to if I were to be the the ASIC backed loans czar and waving my magic czar wand on top of the the lending markets, I would say that I'm willing to lend into a fixed price environment for the hardware, whether Bitcoin's trading at 10,000

to 150. It at the market price or whatever. You can only collateralize the machine at sort of the minimum

the minimum clearing price,

Unknown:

Equitation values is exactly that has to do more with the I mean, let's just I I think we can probably agree that the last 3 years is probably an exaggeration

of liquidity

from, let's say, the fed,

the the entire stimulus is that that have occurred

and kind of the the easygoing

mentality of the the last, like,

12 year bull market

of just equities and everything. So

Unknown:

50 year bull market.

Unknown:

I mean, I think, you know, we

let's so let's set that aside, but also the the question of, you know, will you have ACEC backed loans? I mean,

it's gonna be a commodity.

And all commodities,

like the production of commodities, they get loans. You get to to produce oil, gold,

agricultural goods. So, yes, there will be continued

lending

to,

to for for ACEC backed loans, but

there's not gonna be the same or I would not think there's gonna be the same exaggeration in the amount of liquidity

that's being pumped into the system

because, I mean, it's still

a a buyer and a seller meeting,

Unknown:

and making a price. I'm just thinking about how to price the collateral. Right? So if I'm like, at the end of the day, like, why do I, like, why do I think about an ASIC backed loan as a financial instrument? It's because I post a certain amount of collateral in exchange for cash to purchase,

you know, something,

or run the business. And so, like, how do I value that collateral is no longer gonna be, you know, whatever, 60% of the current ASIC price, which was basically what was done

historically. It's gonna be, this is sort of my lower bound on the risk model, and the risk model's gonna have a much lower lower bound. And that's the amount that I'm willing to finance against this piece of collateral. The same way that if you go to Unchained and say, I want a Bitcoin backed loan. Right? There's this concept of LTV, which is if I put if I give you one Bitcoin, I'm gonna get some percentage less than that current day's value in cash back for it. I I just think that we're gonna see the same mechanics. We're just gonna see them in sort of a more conservative

iteration.

Unknown:

Yeah. I agree with that. And I think, like, to kinda prove this point and to offer, like, some more concrete numbers here, like, we only really look at liquidation value when we're thinking about how to price the underlying asset that we're underwriting and that we're offering a loan against. And so, like, for example, if I say, today, I'm looking at new machines on the market and they're priced at $10 a terahash.

I think, you know, and and we're using, like, reasonable data that we have knowing, like, okay, let's say 200,000

Celsius machines are coming onto the market pending few weeks. Right? So I think that an actual liquidation value for a theoretical, I don't know, 200,006 we might be underwriting is actually $5 a t. That's where I feel 100%

confidence that I can get all of my money back. And so that's the liquidation value that we're offering. That's separate from the LTV. So down from there, let's say I might offer a 50% LTV. So now I'm giving you 2.50 on your $10 terahash machine that you paid that that that you paid for or that it's maybe you didn't pay for it, maybe that's the value that it is today. So that's how we're thinking about that. Additionally,

we'll probably also have margin provisions in there, which is really unfortunate. But at the end of the day, like, if you don't have additional ASICs to post, we're gonna be repoing your machines.

Unknown:

And that's the conservative outcome that happens when you blow out 100 of 1,000,000 of dollars, if not 1,000,000,000 of dollars of cash across

Unknown:

10 to 15 different companies that are upside down. Yeah. I mean, I'm just gonna go with next cycle, people are gonna do crazy insane shit and just lose their shits.

Unknown:

We we we are lowercase right. You are

Unknown:

right. There's gonna be, like, nothing conservative about next Crazier.

Yeah. It always gets crazier. You always can bet on crazier.

But before we hit q and a, I'm curious.

Whole mining, bullish or bearish? Why, Harry?

Unknown:

Bullish on it happening, bearish on unit economics.

Unknown:

Okay.

Cassie.

Unknown:

I live in Puerto Rico, so I have island energy prices.

Unknown:

So I'm a little bit bearish on home mining, but, I am, bullish on,

home mining and, locales where they're gonna be able to harvest their heat. I mean, we are we're we're seeing,

the latest with crypto cloaks.

You can get a black box from Steve Barber and then pump in

that heat,

from distributed hat. Oh, you're showing from the CryptoCloaks.

Unknown:

That's a 3 d printed shroud they have there.

So you can pump it into your heating system.

Unknown:

Correct. I think these externalities are gonna be what is you know, really exciting for individuals

to be able to go after to lower their all in cost for heating their homes, living their lives.

Unknown:

I like that answer.

I like Jonathan's answer the most.

What percent yeah. Before we get to q and a, what in how many years are we doing? 50. It's all done. 50 year in 50 years sorry. In in

What's all done? Okay. We won't do 50, and we won't in 30 years,

what percentage

of world energy production will be used by Bitcoin mining? We'll start with Harry.

Unknown:

14 a half percent.

Unknown:

14 a half percent, Cassie.

Unknown:

Take my dart and throw it at a board.

I think

I think less than 10. Less than 10. That also assumes, like, continued population growth and, like, we now electrify, like, you know, the other 40% of the world, 60% of the world. Fair. Jonathan.

Unknown:

I think, long term, it's gonna be less than 5, probably less than, you know, could be down to less than 1 because

we're gonna need a tremendous amount

of coordination to be an interplanetary

species

and fully anticipate that this century. You can always count on Jonathan. Answer.

Unknown:

Jonathan was just hiding his bearishness behind, like, bullish on humanity.

Thank you, guys. Let's thank our panelists.

Unknown:

Thank your moderator.

Thank you,