september 16, 2021 research

Please see slide 26 for important disclosures.

Sam Doctorac Chief Strategy Officer | Head of [email protected] @CryptoSamDoctor | @Sam_Doc

Mark Chin | [email protected]

A key factor determining investmentreturns from Bitcoin mining is an estimateof future Hashrate and difficulty. Whetherdirectly investing in Bitcoin mining assets orin mining stocks, this estimate is critical. Themining investment’s share of future Hashratedetermines future Bitcoin earnings. Our“Bitcoin Reserve” metrics estimate the BTCexpected to be mined by 1PH/s (5BTC) and1MW (171BTC), operating through 2030.

We assess Hashrate could reach 198EH/s byyear end 2021, versus our prior 145 EH/sestimate. This is a significant upwardrevision, as Hashrate lost after China’s miningban has recovered faster than expected.

Long‐term Hashrate is price dependent –profitability is a key determinant of bothminer balance sheet health and confidence,and investor funding. In this context, weshow that the $353 / PH/s in daily revenueand $385 / MWh using S19 rigs are both inthe 95th percentile of their past two‐yearhistory. Mean reversion over time would bea reasonable assumption that needs to befactored into investment models.

Transaction (Tx) fees have proved lowerthan anticipated recently.We suspect a shiftof trading from Asia to the West, coupledwith a reduction in overall trading volumes,may be a factor, reducing on‐chaintransactions. Over time, increased Layer 2deployment would also lower Layer 1congestion and thus fees. We are loweringour long‐term estimates, with Tx fees not

expected to exceed block rewards until 2028.This estimate also factors into our revenueframework in this report.

We calculate theoretical Bitcoin pricesneeded to maintain revenue per PH/s atrecent levels given our estimate of futureHashrate. At an estimated 198EH/s Hashrate,BTC would need to be $62k to maintaincurrent profitability, but at $685k if Hashrateexceeded 1500 EH/s. Conversely, Hashratecould come in under expectations if price isnot supportive – Hashrate is a function ofprice, and we refrain from predicting price,focusing instead on the price that achievesbenchmark revenue per PH/s at a givenHashrate.

Current rig pricing of $80‐100 per TH/s couldresult in 550‐900 day payback periods ifrevenue per PH/s reverted to the $154median / $189 mean. This could pose a riskto the miner ecosystem. We had previouslyanticipated an inventory glut in the marketowing to order cancellations by Chineseminers following the ban. We were wrong. Inour view, the global semiconductor shortageallowed rig manufacturers to push out someof their anticipated wafer starts, consideringfoundries like TSMC / Samsung have noproblem filling such freed‐up capacity rightnow. This helped support high rig prices.

We are also updating our estimated Bitcoinnetwork power cost curve, which we assessshifted up with the loss of low‐cost Chinesepower. However, it should trend back from a

$40/MWh median now, to $30/MWh overthe next couple of years. This estimate isbased on individual miner conversations andis approximate, although the curve’s trendand contour are clearer.

We model the MWh needed to produce 1BTC based on various theoretical powerefficiencies (W / TH/s) and the cost toproduce 1 BTC, including labor, as well as theincremental cost per BTC for each 1c /kWh($10/ MWh) increase in the cost of power.This retrenchment model helps identifywhen a miner would withdraw capacitybased on network Hashrate and equipmentefficiency, compared with the prevailingBitcoin price.

Risks to our Hashrate estimates: upside riskwould be driven by easing infrastructureconstraints, coupled with a surging price.

Downside risks appear to have substantiallyplayed out, driven by China’s crackdown.

Takeaways• We expect 198H/s Hashrate by year end,

with risks skewed to the upside• Median power cost is now estimated at

$40/MWh• While we do not forecast Bitcoin price, we

dive into the sensitivity of profitabilityand payback periods to revenue per PH/s,in turn driven by Transaction fees, BTCprice and network Hashrate

• Finally, our “Bitcoin Reserve” stands at5BTC / PH/s and 171BTC/MW through2030

Key Takeaways

• We expect 198H/s Hashrateby year end, with risksskewed to the upside

• Median power cost is nowestimated at $40/MWh

• While we do not forecastBitcoin price, we dive intothe sensitivity ofprofitability and paybackperiods to revenue per PH/s,in turn driven byTransaction fees, BTC priceand network Hashrate

• Finally, our “BitcoinReserve” stands at 5TC /PH/s and 171BTC/MWthrough 2030

• Contact our sales team at [email protected] with questions or to subscribe to our paid research

Research2021 Year End Bitcoin Hashpower Estimates:Raising to 198 EH/s; Updating Cost Curve, Mining Costs and Mining Reserve Estimates

September 16, 2021

BitO

oda Re

search

2 © 2021 BitOoda Holdings Inc. Sep 16, 2021

Hashrate and Mining Economics

• Miners receive rewards as compensation for validating transactions in two forms:(1) each block consists of a number of transactions awaiting validation for a fee,paid by the transactor, and (2) the block includes a reward – currently 6.25BTC perblock until mid‐2024 – of freshly minted Bitcoin that the protocol generates as anincentive for mining

• Miners contribute their computing power into a mining “pool” to try and solve abrute force puzzle. The first pool to solve the puzzle receives the reward, and themining pool distributes the total daily rewards (after pool fees) to the differentparticipants in the pool, based on their share of contributed computing power.This pool system helps stabilize cash flows and ensures that even small miners geta fair share of rewards, with less of an impact of luck

• The number of blocks mined each day is targeted at roughly 144, with thecomplexity of the puzzle changing roughly every 14 days, rising or falling asneeded to maintain the 144‐block production rate

• In addition, the number of BTC rewards in each block halve every 4 years, thenotion being that block rewards are a temporary incentive until time and usagegrow transaction fees to the point the network becomes self sufficient

• Thus, the number of Bitcoin minted each day remains roughly the same, halvingevery 4 years

• If network computing power or “Hashrate” increased over time, the share of thetotal daily rewards earned by one unit of mining “Hashpower” would fall

• This paper develops a methodology for estimating Hashrate over the lifetime ofa mining investment, given that Hashrate is a critical driver of the viability of amining project

Miner

Mining Pool

Blockchain

Computing Power

Miner

Revenue Considerations• Transaction Fees influence total daily BTC mined

• Network growth drives each miner’s share of BTCmined

• Revenue = Bitcoin Price x BTC Volume Mined

• Thus, we need to develop reasonable forecasts aroundeach of these three key variables

Expense Considerations• The biggest operating expense is power

• Labor and maintenance costs are relatively modest

• Capex is significantly influenced by rig price andavailability, which are driven by Bitcoin price andsemiconductor shortages

• Decisions such as immersion vs. air cooled and buildingvs. containerized, as well as existing facilityinfrastructure, influence timing, investment andpayback periods

Hashrate estimates are the most important revenue driver for a mining project, more than price

BitO

oda Re

search

© 2021 BitOoda Holdings Inc. Sep 16, 20213

BitOoda 2021 Hash Estimates~198 EH/s by Year End

Figure: Historical and Estimated Target Hashrate 2017‐2030Historical as of 9/7/21; Estimates as of current, 7/2/21, 5/22/21, and 7/15/20

• Our projections call for a target Hashrate of ~198EH/s by year end (YE) 2021• On July 2nd, we had expected Hashrate to be roughly 145EH/s• Hashrate recovery following the China ban (loss of ~100 EH/s) has exceeded

our expectations, with the target Hashrate currently at 131.7 EH/s• We continue to expect power infrastructure to be the gating factor in mining

expansion

Source: BitOoda estimates, CoinMetrics

Log Scale

BitOoda Hashrate EstimatesBitOoda Hashrate Estimates

1,000 PH/s

2,000 PH/s

4,000 PH/s

8,000 PH/s

16,000 PH/s

32,000 PH/s

64,000 PH/s

128,000 PH/s

256,000 PH/s

512,000 PH/s

1,024,000 PH/s

2,048,000 PH/s

Sep-16 Sep-18 Sep-20 Sep-22 Sep-24 Sep-26 Sep-28 Sep-30

Historical5/22/21 Est7/15/20 Est7/2/21 EstCurrent

144,915 PH/s

138,680 PH/s

197,848 PH/s216,497 PH/s

254,653 PH/s

289,729 PH/s

333,665 PH/s

80,000 PH/s

130,000 PH/s

180,000 PH/s

230,000 PH/s

280,000 PH/s

330,000 PH/s

Jul-20 Jan-21 Jul-21 Jan-22 Jul-22

5/22/21 Est 7/15/20 Est7/2/21 Est Current

BitO

oda Re

search


USD per PH/sDaily Revenue in USD

Figure: Bitcoin mining revenue in USD per PH/s over the past two years

• Earnings per PH/s per day are currently $353, in the 95th percentile compared to the past 2 years

• Over the same period, the mean / median daily earnings have been $189 and $154 respectively

Source: BitOoda, CoinMetrics, Bloomberg

BitO

oda Re

search


USD per MWhDaily Revenue Using Latest Gen Rigs

Figure: Bitcoin mining revenue in USD per MWh using then‐current gen rigs over the past two years

• Miners using latest‐generation S19 class mining rigs are currently earning $385 per MWh, in the 95th percentile compared to the past 2 years of history

• Over the same period, the mean / median daily earnings have been $140 and $99 respectively


BitO

oda Re

search


Mining Revenue AnalysisBTC Earnings / Block Other Driver With Hashrate

• Mining revenue in BTC terms is a function of network payout and network size• Over time, transactions fees should increase to help offset declining block

rewards. The process could be aided by second‐order services, such as the Lightning Network, that consolidate lower‐value transactions, increasing system throughput while keeping individual transaction fees low

• Our network analysis helps estimate long‐term BTC‐denominated revenue, factoring in block rewards and Tx fees per PH/s and per MW

• We expect Transaction (Tx) Fees to grow and helpoffset declining block rewards, crossing over to exceedthe block reward post the 2028 halving

• At the same time, we expect the network Hashrate tocontinue to grow, which will drive a monotonic declinein BTC earned per PH/s

• Tx Fees / block are a function of multiple competingfactors:

• Rising BTC price would inhibit BTC‐denominated fee growth

• High fees inhibit smaller transactions

• Layer 2 projects, such as the LightningNetwork, could consolidate smallertransactions, creating the ability to pay higherfees on the primary Bitcoin blockchain

• However, if enough transactions migrate to beconsolidated on Layer 2 solutions, networkcongestion on the primary blockchain mayease, lowering fees – particularly in BTC terms

Source: BitOoda estimates, Bloomberg, CoinMetrics

Year Hashrate BTC Reward / Block Tx Fee / Block Total Bitcoin per Day

2013 8 PH/s 25.00 BTC 0.08 BTC 3612 BTC2014 291 PH/s 25.00 BTC 0.09 BTC 3613 BTC2015 743 PH/s 25.00 BTC 0.21 BTC 3630 BTC2016 2,273 PH/s 12.50 BTC 0.64 BTC 1893 BTC2017 13.8 EH/s 12.50 BTC 5.92 BTC 2653 BTC2018 40.2 EH/s 12.50 BTC 0.13 BTC 1819 BTC2019 92.7 EH/s 12.50 BTC 0.15 BTC 1822 BTC2020 133.1 EH/s 6.25 BTC 0.77 BTC 1010 BTC2021 197.8 EH/s 6.25 BTC 0.91 BTC 1032 BTC2022 334 EH/s 6.25 BTC 1.11 BTC 1059 BTC2023 537 EH/s 6.25 BTC 1.32 BTC 1090 BTC2024 623 EH/s 3.13 BTC 1.56 BTC 675 BTC2025 791 EH/s 3.13 BTC 1.82 BTC 713 BTC2026 976 EH/s 3.13 BTC 2.13 BTC 757 BTC2027 1,156 EH/s 3.13 BTC 2.47 BTC 806 BTC2028 1,326 EH/s 1.56 BTC 2.81 BTC 630 BTC2029 1,438 EH/s 1.56 BTC 3.18 BTC 682 BTC2030 1,541 EH/s 1.56 BTC 3.52 BTC 731 BTC

BitO

oda Re

search


Historical USD / PH/s:Implied Future BTC Prices

Figure: Implied price range of Bitcoin to match the current range of daily earnings per PH/s, at the higher future network Hashrate

• Over the past two years, revenue per PH/s has ranged from $71 to $415 per day, as seen in the box plot below left

• Based on our estimated forward network Hashrate and daily BTC mined (including Tx fees), we estimate the price Bitcoin would have to be in order to match the historical economics of mining per PH/s

• On our estimated 198 EH/s by YE 2021, BTC would need to be at $62,324, and at $685k by YE 2030, to earn the same $353 / day a miner earns per PH/s per day today, and ~$36k and ~$400k respectively, to match mean earnings of $189 / PH/s per day


$36,244

$398,419

$62,324

$685,112

$10,000

$20,000

$40,000

$80,000

$160,000

$320,000

$640,000

$1,280,000

2021:198EH/s

2022:334EH/s

2023:537EH/s

2024:623EH/s

2025:791EH/s

2026:976EH/s

2027:1156EH/s

2028:1326EH/s

2029:1438EH/s

2030:1541EH/s

Log S

cale

Implied Future BTC Price

Mean Earnings Current Earnings

Daily Revenue per PH/sAs of 9/15/2021

Price Implied by Historical Daily PH/s Earnings

BitO

oda Re

search



Daily USD Rev / PH/s, Current Range

Minimum25th

Percentile Median Mean75th

Percentile Current Max

Year End Hashrate Daily BTC BTC / PH/s / Day $71 $107 $154 $189 $246 $325 $415

2021 198 EH/s 1031.71 0.0052 $13,615 $20,519 $29,532 $36,244 $47,175 $62,324 $79,583

2022 334 EH/s 1059.32 0.0032 $22,364 $33,703 $48,507 $59,531 $77,485 $102,368 $130,717

2023 537 EH/s 1090.40 0.0020 $34,987 $52,727 $75,887 $93,134 $121,222 $160,151 $204,501

2024 623 EH/s 674.84 0.0011 $65,539 $98,771 $142,156 $174,464 $227,080 $300,004 $383,082

2025 791 EH/s 712.67 0.0009 $78,820 $118,785 $170,961 $209,816 $273,094 $360,795 $460,708

2026 976 EH/s 756.88 0.0008 $91,585 $138,022 $198,649 $243,797 $317,323 $419,227 $535,321

2027 1,156 EH/s 805.68 0.0007 $101,875 $153,530 $220,969 $271,189 $352,976 $466,331 $595,468

2028 1,326 EH/s 629.92 0.0005 $149,460 $225,242 $324,180 $397,858 $517,846 $684,147 $873,603

2029 1,438 EH/s 682.33 0.0005 $149,643 $225,518 $324,578 $398,345 $518,481 $684,985 $874,674

2030 1,541 EH/s 731.19 0.0005 $149,671 $225,560 $324,637 $398,419 $518,577 $685,112 $874,835

Historical USD / PH/s:Implied Future BTC Prices ‐ Table

• $71‐415 daily revenue / PH/s over the past two years• As Hashrate increases, BTC flow per PH/s falls• Price of Bitcoin needs to increase to offset the lower volume, if daily

revenue per PH/s is to be comparable with recent history

Figure: Implied price range of Bitcoin to match the current range of daily earnings per PH/s, at the higher future network Hashrate

What would BTC price have to be in order to sustain earnings at levels comparable to recent history at future Hashrates

What would BTC price have to be in order to sustain earnings at levels comparable to recent history at future Hashrates

BitO

oda Re

search



Investment Payback:Days to Recover Capex

• We examine the days needed to recover capex, after power and labor costs

• Investment recovery periods could really stretch out as equipment cost rises

• This is especially a concern if rigs are procured at current prices and revenue per day reverted to the mean / median levels of recent history

• 500‐900 day paybacks, excluding infrastructure / container / ancillary capex, could prove prohibitive

Figure: Days to recover rig investment as a function of revenue per PH/s per day, assuming $40/MWh median power cost, PUE of 1.12, and 34W / TH/s efficiency and $7/MWh labor cost

Daily USD Rev / PH/s, Current RangeMinimum25th Percentile Median Mean 75th Percentile Current Max

Capital Cost / TH/s $71 $107 $154 $189 $246 $325 $415$10 364 Days 158 Days 91 Days 69 Days 49 Days 36 Days 27 Days$20 728 315 181 137 99 71 54$30 1092 473 272 206 148 107 81$40 1455 630 362 275 198 142 108$50 1819 788 453 344 247 178 135$60 2183 945 543 412 296 213 162$70 2547 1103 634 481 346 249 188$80 2911 1260 724 550 395 284 215$90 3275 1418 815 619 444 320 242

$100 3638 1575 905 687 494 355 269$110 4002 1733 996 756 543 391 296$120 4366 1890 1086 825 593 426 323$130 4730 2048 1177 894 642 462 350$140 5094 2205 1267 962 691 497 377$150 5458 2363 1358 1031 741 533 404

Capex costs of $80‐100 per TH/s could result in

prohibitive payback periods if the revenue mean reverts

Capex costs of $80‐100 per TH/s could result in

prohibitive payback periods if the revenue mean reverts

BitO

oda Re

search


Power Cost CurveBitOoda Estimates $40/MWh Median Cost

Figure: Power cost curve: Mapping estimated power cost vs. share of network capacityAs of 9/16/2021

• In July 2020, we published a survey‐based cost curve estimate for Bitcoin mining power costs

• Much of the low‐cost power was in China and is now shut down, raising the average network power cost by our updated estimates

• However, we expect the curve to shift down and to the right over the next 18‐24 months as new capacity seeks the lowest cost

• We estimate the current median power cost at $40 / MWh, dropping to $30 / MWh over the next 18‐24 months


50%$40

50%$30

$0

$10

$20

$30

$40

$50

$60

$70

$80

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%

Powe

r Cos

t per

MW

h

Percentile of Network

Bitcoin Power Cost Curve

9/15/2021

18-24 months

BitO

oda Re

search


MWh per BTC MinedAcross Rig Efficiencies

• As Hashrate increases over time, we anticipate it to take an increasing amount of power to produce 1 BTC

• We assess it will take over 500MWh by year end to produce 1 BTC using older S9 class rigs that consume 100W / TH/s

• Later in the decade, it will take over 5000 MWh on S9s, although that is a theoretical notion – the machines would have been turned off long before then


0 MWh

1000 MWh

2000 MWh

3000 MWh

4000 MWh

5000 MWh

6000 MWh

198 EH/s 334 EH/s 537 EH/s 623 EH/s 791 EH/s 976 EH/s 1,156 EH/s 1,326 EH/s 1,438 EH/s 1,541 EH/s

2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

100 W

80 W

50 W

40 W

35 W

30 W

25 W

20 W

15 W

10 W

5 W

Figure: Power consumption per Bitcoin mined, by theoretical power efficiency, in MWh, based on year‐end Hashrate estimatesPUE (Power Usage Effectiveness) assumed at 1.12

MWh needed to produce 1 BTC, at different power

efficiencies

MWh needed to produce 1 BTC, at different power

efficiencies

BitO

oda Re

search


MWh per BTC MinedUsing Latest Gen Rigs

Figure: Power consumption per Bitcoin mined, by theoretical power efficiency, in MWhPUE (Power Usage Effectiveness) assumed at 1.12

• The price of Bitcoin affects the viability of operating older‐gen rigs• Theoretical efficiencies show the relationship of power consumption

to network Hashrate


Power consumption to produce 1 Bitcoin, based on device power efficiency (Watts / TH/s)

Year End 133 EH/s 100 W 80 W 50 W 40 W 35 W 30 W 25 W 20 W 15 W 10 W 5 W

2021 198 EH/s 515 MWh 412 MWh 258 MWh 206 MWh 180 MWh 155 MWh 129 MWh 103 MWh 77 MWh 52 MWh 26 MWh

2022 334 EH/s 847 677 423 339 296 254 212 169 127 85 42

2023 537 EH/s 1325 1060 662 530 464 397 331 265 199 132 66

2024 623 EH/s 2481 1985 1241 993 868 744 620 496 372 248 124

2025 791 EH/s 2984 2387 1492 1194 1044 895 746 597 448 298 149

2026 976 EH/s 3467 2774 1734 1387 1214 1040 867 693 520 347 173

2027 1,156 EH/s 3857 3086 1928 1543 1350 1157 964 771 579 386 193

2028 1,326 EH/s 5658 4527 2829 2263 1980 1698 1415 1132 849 566 283

2029 1,438 EH/s 5665 4532 2833 2266 1983 1700 1416 1133 850 567 283

2030 1,541 EH/s 5666 4533 2833 2267 1983 1700 1417 1133 850 567 283

BitO

oda Re

search


Cost to Mine 1 BTCUsing 100W / TH/s Rigs

• By YE 2021, it should take $3,600 in labor and $5,155 in power cost per 1c/kWh price to produce 1 BTC

• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 100W machines would be $3608 + 5155 x 2 = $13,918

• By YE 2024, the same miner would spend $66,995 on this older equipment, assuming Bitcoin price is higher than that (i.e., $66,995)

Source: BitOoda, CoinMetrics, BloombergFigure: Cost to mine 1 Bitcoin using a 100W/TH/s rig, where each grey segment represents1 c / kWh; for example, the top of the stack shows labor cost plus 10 x cost per 1 c / kWh

Year EndHashrateMWh / BTC

$3,608 $5,927 $9,272 $17,369 $20,888 $24,271 $26,998 $39,609 $39,657 $39,665 $5,155 $8,467 $13,246

$24,813 $29,841

$34,673 $38,569

$56,584 $56,654 $56,664

$0

$100,000

$200,000

$300,000

$400,000

$500,000

$600,000

$700,000

2021: 198EH/s515 MWh

2022: 334EH/s847 MWh

2023: 537EH/s

1325 MWh

2024: 623EH/s

2481 MWh

2025: 791EH/s

2984 MWh

2026: 976EH/s

3467 MWh

2027: 1156EH/s3857 MWh

2028: 1326EH/s5658 MWh

2029: 1438EH/s5665 MWh

2030: 1541EH/s5666 MWh

Cost to mine 1 BTC with 100W / TH/s Machines

Labor Cost

Incremental Cost per 1 c / kWh

BitO

oda Re

search




• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 80W machines would be $2,887 + 4,124 x 2 = $11,135


Source: BitOoda, CoinMetrics, BloombergFigure: Cost to mine 1 Bitcoin using a 80 W/TH/s rig, where each grey segment represents 1 c / kWh; for example, the top of the stack shows labor cost plus 10 x cost per 1 c / kWh


$2,887 $4,741 $7,418 $13,895 $16,711 $19,417 $21,599 $31,687 $31,726 $31,732 $4,124 $6,773 $10,597

$19,850 $23,872

$27,739 $30,855

$45,267 $45,323 $45,331

$0

$100,000

$200,000

$300,000

$400,000

$500,000

$600,000

2021: 198EH/s412 MWh

2022: 334EH/s677 MWh

2023: 537EH/s

1060 MWh

2024: 623EH/s

1985 MWh

2025: 791EH/s

2387 MWh

2026: 976EH/s

2774 MWh

2027: 1156EH/s3086 MWh

2028: 1326EH/s4527 MWh

2029: 1438EH/s4532 MWh

2030: 1541EH/s4533 MWh


Labor Cost


BitO

oda Re

search





• By YE 2024, the same miner would spend $33,496 on this older equipment, assuming Bitcoin price is higher than that (i.e.. $33,496)



$1,804 $2,963 $4,636 $8,684 $10,444 $12,136 $13,499 $19,804 $19,829 $19,832 $2,577 $4,233 $6,623

$12,406 $14,920

$17,337 $19,285

$28,292 $28,327 $28,332

$0

$50,000

$100,000

$150,000

$200,000

$250,000

$300,000

$350,000

2021: 198EH/s258 MWh

2022: 334EH/s423 MWh

2023: 537EH/s662 MWh

2024: 623EH/s

1241 MWh

2025: 791EH/s

1492 MWh

2026: 976EH/s

1734 MWh

2027: 1156EH/s1928 MWh

2028: 1326EH/s2829 MWh

2029: 1438EH/s2833 MWh

2030: 1541EH/s2833 MWh


Labor Cost


BitO

oda Re

search




• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 40W machines would be $1,443+ 2,062 x 2 = $5,567




$1,443 $2,371 $3,709 $6,948 $8,355 $9,709 $10,799 $15,844 $15,863 $15,866 $2,062 $3,387 $5,298

$9,925 $11,936

$13,869 $15,428

$22,634 $22,661 $22,666

$0

$50,000

$100,000

$150,000

$200,000

$250,000

$300,000

2021: 198EH/s206 MWh

2022: 334EH/s339 MWh

2023: 537EH/s530 MWh

2024: 623EH/s993 MWh

2025: 791EH/s

1194 MWh

2026: 976EH/s

1387 MWh

2027: 1156EH/s1543 MWh

2028: 1326EH/s2263 MWh

2029: 1438EH/s2266 MWh

2030: 1541EH/s2267 MWh


Labor Cost


BitO

oda Re

search




• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 35W machines would be $1,263+ 1,804 x 2 = $4,871




$1,263 $2,074 $3,245 $6,079 $7,311 $8,495 $9,449 $13,863 $13,880 $13,883 $1,804 $2,963 $4,636

$8,684 $10,444

$12,136 $13,499

$19,804 $19,829 $19,832

$0

$50,000

$100,000

$150,000

$200,000

$250,000

2021: 198EH/s180 MWh

2022: 334EH/s296 MWh

2023: 537EH/s464 MWh

2024: 623EH/s868 MWh

2025: 791EH/s

1044 MWh

2026: 976EH/s

1214 MWh

2027: 1156EH/s1350 MWh

2028: 1326EH/s1980 MWh

2029: 1438EH/s1983 MWh

2030: 1541EH/s1983 MWh


Labor Cost


BitO

oda Re

search








$1,082 $1,778 $2,782 $5,211 $6,267 $7,281 $8,100 $11,883 $11,897 $11,899 $1,546 $2,540 $3,974

$7,444 $8,952

$10,402 $11,571

$16,975 $16,996 $16,999

$0

$20,000

$40,000

$60,000

$80,000

$100,000

$120,000

$140,000

$160,000

$180,000

$200,000

2021: 198EH/s155 MWh

2022: 334EH/s254 MWh

2023: 537EH/s397 MWh

2024: 623EH/s744 MWh

2025: 791EH/s895 MWh

2026: 976EH/s

1040 MWh

2027: 1156EH/s1157 MWh

2028: 1326EH/s1698 MWh

2029: 1438EH/s1700 MWh

2030: 1541EH/s1700 MWh


Labor Cost


BitO

oda Re

search



• By YE 2021, it should take $902 in labor and $1,289 in power cost per 1c/kWh price to produce 1 BTC

• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 25W machines would be $902 + 1,289 x 2 = $3,480

• By YE 2024 end, the same miner would spend $16,748 on this older equipment, assuming Bitcoin price is higher than that (i.e., $16,748)



$902 $1,482 $2,318 $4,342 $5,222 $6,068 $6,750 $9,902 $9,914 $9,916 $1,289 $2,117 $3,311

$6,203 $7,460

$8,668 $9,642

$14,146 $14,163 $14,166

$0

$20,000

$40,000

$60,000

$80,000

$100,000

$120,000

$140,000

$160,000

2021: 198EH/s129 MWh

2022: 334EH/s212 MWh

2023: 537EH/s331 MWh

2024: 623EH/s620 MWh

2025: 791EH/s746 MWh

2026: 976EH/s867 MWh

2027: 1156EH/s964 MWh

2028: 1326EH/s1415 MWh

2029: 1438EH/s1416 MWh

2030: 1541EH/s1417 MWh


Labor Cost


Figure: Cost to mine 1 Bitcoin using a 25 W/TH/s rig, where each grey segment represents 1 c / kWh; for example, the top of the stack shows labor cost plus 10 x cost per 1 c / kWh

BitO

oda Re

search



• By YE 2021, it should take $722 in labor and $1,031 in power cost per 1c/kWh price to produce 1 BTC

• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 20W machines would be $722 + 1,031 x 2 = $2,784




$722 $1,185 $1,854 $3,474 $4,178 $4,854 $5,400 $7,922 $7,931 $7,933 $1,031 $1,693 $2,649

$4,963 $5,968

$6,935 $7,714

$11,317 $11,331 $11,333

$0

$20,000

$40,000

$60,000

$80,000

$100,000

$120,000

$140,000

2021: 198EH/s103 MWh

2022: 334EH/s169 MWh

2023: 537EH/s265 MWh

2024: 623EH/s496 MWh

2025: 791EH/s597 MWh

2026: 976EH/s693 MWh

2027: 1156EH/s771 MWh

2028: 1326EH/s1132 MWh

2029: 1438EH/s1133 MWh

2030: 1541EH/s1133 MWh


Labor Cost



BitO

oda Re

search



• By YE 2021, it should take $541 in labor and $773 in power cost per 1c/kWh price to produce 1 BTC


• By YE 2024 end, the same miner would spend $10,049 on this older equipment, assuming Bitcoin price is higher than that (i.e., $10,049)



$541 $889 $1,391 $2,605 $3,133 $3,641 $4,050 $5,941 $5,949 $5,950 $773 $1,270 $1,987

$3,722 $4,476

$5,201 $5,785

$8,488 $8,498 $8,500

$0

$10,000

$20,000

$30,000

$40,000

$50,000

$60,000

$70,000

$80,000

$90,000

$100,000

2021: 198EH/s77 MWh

2022: 334EH/s127 MWh

2023: 537EH/s199 MWh

2024: 623EH/s372 MWh

2025: 791EH/s448 MWh

2026: 976EH/s520 MWh

2027: 1156EH/s579 MWh

2028: 1326EH/s849 MWh

2029: 1438EH/s850 MWh

2030: 1541EH/s850 MWh


Labor Cost



BitO

oda Re

search








$361 $593 $927 $1,737 $2,089 $2,427 $2,700 $3,961 $3,966 $3,966 $515 $847 $1,325

$2,481 $2,984

$3,467 $3,857

$5,658 $5,665 $5,666

$0

$10,000

$20,000

$30,000

$40,000

$50,000

$60,000

$70,000

2021: 198EH/s52 MWh

2022: 334EH/s85 MWh

2023: 537EH/s132 MWh

2024: 623EH/s248 MWh

2025: 791EH/s298 MWh

2026: 976EH/s347 MWh

2027: 1156EH/s386 MWh

2028: 1326EH/s566 MWh

2029: 1438EH/s567 MWh

2030: 1541EH/s567 MWh


Labor Cost



BitO

oda Re

search




• Thus, if a miner’s power cost was 2 c/kWh, the cost to produce 1 BTC on 5W machines would be $180 + 258 x 2 = $696




$180 $296 $464 $868 $1,044 $1,214 $1,350 $1,980 $1,983 $1,983 $258 $423 $662

$1,241 $1,492

$1,734 $1,928

$2,829 $2,833 $2,833

$0

$5,000

$10,000

$15,000

$20,000

$25,000

$30,000

$35,000

2021: 198EH/s26 MWh

2022: 334EH/s42 MWh

2023: 537EH/s66 MWh

2024: 623EH/s124 MWh

2025: 791EH/s149 MWh

2026: 976EH/s173 MWh

2027: 1156EH/s193 MWh

2028: 1326EH/s283 MWh

2029: 1438EH/s283 MWh

2030: 1541EH/s283 MWh


Labor Cost



BitO

oda Re

search


12%

30%

18%

10%

7%

6%5%

4%3%

4%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Q4 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

Annual share of BTC Mined / PH/sBlock Reward Tx Fees

Reserve of ~5 BTC / PH through 2030Front End Loaded

Figure: Projected BTC earned per PH/s annually through 2030

• A PH/s of capacity running continuously will earn 5.013 BTC by the end of 2030

• 12% of that “mining reserve,” as we call it, will be mined in the remainder of 2021 and another 30% in 2022

• This underscores our thesis that time to market is a key factor for any project


Over half the BTC mined over next decade by 1PH/s to be mined in

next 24 months

Over half the BTC mined over next decade by 1PH/s to be mined in

next 24 months

Annual BTC Mined per PH/s

Year Block Rewards Tx Fees TotalPercent Of

Decade Total

Q4 2021 0.521 BTC 0.074 BTC 0.595 BTC 12%2022 1.290 BTC 0.207 BTC 1.497 BTC 30%2023 0.772 BTC 0.149 BTC 0.921 BTC 18%2024 0.385 BTC 0.130 BTC 0.515 BTC 10%2025 0.232 BTC 0.126 BTC 0.358 BTC 7%2026 0.187 BTC 0.118 BTC 0.305 BTC 6%2027 0.154 BTC 0.113 BTC 0.267 BTC 5%2028 0.090 BTC 0.111 BTC 0.201 BTC 4%2029 0.059 BTC 0.114 BTC 0.173 BTC 3%2030 0.057 BTC 0.122 BTC 0.180 BTC 4%

Total 3.747 BTC 1.266 BTC 5.013 BTC 100%

BitO

oda Re

search


~171 BTC / MW ReserveThrough 2030

• Our equipment upgrade estimates suggest we could ultimately get 51 PH/s per MW of capacity, compared with 26 PH/s with current S19‐class rigs, including a 12% PUE

• This leads us to a ~171 BTC “mining reserve” per MW capacity through 2030• Two key assumptions that may not hold true, though, are continuous operation

and immediate upgrades to the latest technology• As a practical matter, the BTC reserve could be meaningfully lower if we skipped

a generation while upgrading, and/or if there is intermittency of power supply


Rig upgrades drive growth in PH/s per MW, slightly

flattening the payout curve

Rig upgrades drive growth in PH/s per MW, slightly

flattening the payout curve

9 PH/s22 PH/s 26 PH/s

35 PH/s 40 PH/s 45 PH/s51 PH/s

0 PH/s

20 PH/s

40 PH/s

60 PH/s

S9Class

S17Class

S19Class

5nmClass

3nmClass

2nmClass

1nmClass

PH / MW

Figure: Projected BTC earned per MW annually through 2030

Annual BTC Mined per MW

Year Block Rewards Tx Fees TotalPercent Of

Decade Total

Q4 2021 13.502 BTC 1.929 BTC 15.431 BTC 9%2022 33.442 BTC 5.371 BTC 38.813 BTC 23%2023 26.750 BTC 5.171 BTC 31.921 BTC 19%2024 15.209 BTC 5.151 BTC 20.360 BTC 12%2025 9.187 BTC 4.966 BTC 14.153 BTC 8%2026 7.377 BTC 4.661 BTC 12.038 BTC 7%2027 6.920 BTC 5.103 BTC 12.023 BTC 7%2028 4.045 BTC 5.019 BTC 9.065 BTC 5%2029 2.668 BTC 5.128 BTC 7.797 BTC 5%2030 2.933 BTC 6.296 BTC 9.229 BTC 5%

Total 122.034 BTC 48.795 BTC 170.829 BTC 100% 9%

23%

19%

12%

8%

7%

7%5%

5%5%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Q42021

2022 2023 2024 2025 2026 2027 2028 2029 2030

Annual share of BTC Mined / MWBlock Reward Tx Fees

© 2021 BitOoda Holdings Inc.

BitO

oda Re

search

Sep 16, 202126

Disclosures

Purpose

This research is only for the clients of BitOoda.This research is not intended to constitute anoffer, solicitation, or invitation for any securitiesand may not be distributed into jurisdictionswhere it is unlawful to do so. For additionaldisclosures and information, please contact aBitOoda representative at [email protected].

Analyst Certification

Sam Doctor, the research analyst denoted by an“AC” on the cover of this report, hereby certifiesthat all of the views expressed in this reportaccurately reflect his personal views, which havenot been influenced by considerations of thefirm’s business or client relationships.

Conflicts of Interest

This research contains the views, opinions, andrecommendations of BitOoda. This report isintended for research and educational purposesonly. We are not compensated in any way basedupon any specific view or recommendation.

General Disclosures

Any information (“Information”) provided byBitOoda Holdings, LLC, BitOoda Digital, LLC,BitOoda Technologies, LLC or OodaCommodities, LLC and its affiliated or relatedcompanies (collectively, “BitOoda”), either in thispublication or document, in any othercommunication, or on orthrough http://www.bitooda.io/, including anyinformation regarding proposed transactions ortrading strategies, is for informational purposesonly and is provided without charge. BitOoda isnot and does not act as a fiduciary or adviser, orin any similar capacity, in providing theInformation, and the Information may not berelied upon as investment, financial, legal, tax,regulatory, or any other type of advice. TheInformation is being distributed as part ofBitOoda’s sales and marketing efforts as anintroducing broker and is incidental to itsbusiness as such. BitOoda seeks to earnexecution fees when its clients executetransactions using its brokerageservices. BitOoda makes no representations orwarranties (express or implied) regarding, norshall it have any responsibility or liability for theaccuracy, adequacy, timeliness or completenessof, the Information, and no representation ismade or is to be implied that the Informationwill remain unchanged. BitOoda undertakes noduty to amend, correct, update, or otherwisesupplement the Information.

The Information has not been prepared ortailored to address, and may not be suitable orappropriate for the particular financial needs,circumstances or requirements of any person,and it should not be the basis for making any

investment or transaction decision. TheInformation is not a recommendation to engagein any transaction. The digital asset industry issubject to a range of inherent risks, including butnot limited to: price volatility, limited liquidity,limited and incomplete information regardingcertain instruments, products, or digital assets,and a still emerging and evolving regulatoryenvironment. The past performance of anyinstruments, strategies, products or digital assetsaddressed in the Information is not a guide tofuture performance, nor is it a reliable indicatorof future results or performance. Commoditytrading involves substantial risk of loss.

Ooda Commodities, LLC is a member of NFA andis subject to NFA’s regulatory oversight andexaminations. However, you should be awarethat NFA does not have regulatory oversightauthority over underlying or spot virtualcurrency products or transactions or virtualcurrency exchanges, custodians or markets.

BitOoda Technologies, LLC is a member of FINRA.

“BitOoda”, “BitOoda Difficulty”, “BitOoda Hash”,“BitOoda Compute”, and the BitOoda logo aretrademarks of BitOoda Holdings, Inc.

Copyright 2021 BitOoda Holdings, Inc. All rightsreserved. No part of this material may bereprinted, redistributed, or sold without priorwritten consent of BitOoda.

september 16, 2021 research

Documents