Cipher Mining: The Rerating Story Isn't Over Yet

Summary Cipher Mining remains a top Bitcoin miner, with strong management, staged buildouts, and a focus on equity over short-term debt. CIFR's operational metrics have surged: hashrate, fleet efficiency, energized capacity, and liquidity have all seen significant growth in the past ￰0￱ recent 10-year AI hosting deal with Fluidstack, backed by Google, secures $3B in revenue and justifies CIFR's premium ￰1￱ GAAP net losses and crypto volatility risks, I maintain a Buy rating, seeing further upside from pipeline expansion and demand for AI compute. I've had a high conviction in Cipher Mining (NASDAQ: CIFR ), and in my view it has been one of the best-run Bitcoin ( BTC-USD ) mining companies.

I initiated coverage on CIFR (in an article titled Cipher Mining’s Strategic Planning Will Pay Off ) in December 2023 at $3.3 per share, with a Buy ￰2￱ is up nearly 500% since initiating ￰3￱ subsequent coverage in July last year ( Cipher Mining Remains An Attractive Bitcoin Miner ) remained a ￰4￱ last time I revisited CIFR a year ago, in October 2024, I maintained a Buy ￰5￱ conviction has been ￰6￱ is not a miner I need to check up on every other ￰7￱ me, it has been a long-term play. Author's CIFR rating history (Seeking Alpha) My conviction has been high in management’s decision-making and their strategic ￰8￱ most other miners have mostly chased rapid capacity growth funded by short-term debt, CIFR has prioritized staged ￰9￱ company has shown a build-out financed by equity and structured obligations rather than immediate short-term bank ￰10￱ - Up A Lot, Not Just In Stock Price CIFR isn't just up in stock price, but the operational metrics are up a lot from a year ￰11￱ buildouts this year have been more aggressive.

Cipher's hashrate has grown from ~7 EH/s in mid-2024 to ~23.6 EH/s as of September ￰12￱ rigs increased from roughly 45,000 to 114,000 ￰13￱ efficiency has improved to around 16.8 J/Th (as of last month’s operational update ) from over 25 J/Th last year. Cipher's current fleet efficiency is one of the lowest (best) among publicly traded ￰14￱ for readers who may not follow mining metrics: a lower J/Th figure for fleet efficiency means improved operating margin per bitcoin ￰15￱ and energization timeline (Cipher Mining website) The energy base expanded as new MW came online across sites, and more power deals were locked in ahead of the HPC ￰16￱ now controls ~777MW of energized capacity across six active ￰17￱ runs 40 ￰18￱ and Chief sites run 40 MW each (which is 80 MW combined).

Barber Lake runs 300 MW energized under Phase ￰19￱ runs 207 MW, and Black Pearl runs 150 MW under Phase ￰20￱ II expansions at Barber Lake and Black Pearl will total 650 MW when they become ￰21￱ since the start of this year shows the same ￰22￱ has risen from about $5.6 million at the end of 2024 to $62.7 million by ￰23￱ holdings have grown to 1500 BTC, worth around $160 million at current Bitcoin spot ￰24￱ on equipment reached $183 million as of Q2, which shows heavy investment in new ￰25￱ injections have come through the $50 million PIPE from SoftBank in Q1 and $172 million in convertible notes in Q2, then the more recent $1.1 billion 0% convertible senior notes offering , which was initially $800 million but was upsized due to strong institutional ￰26￱ balance sheet has scaled with the fleet and operational metrics, the cash position is healthy, and the company remains funded for the next leg of buildouts and HPC ￰27￱ good thing is that debt maturities are long-dated; the zero-coupon convertibles mature in October 2031, and redemption is available starting in October 2028 if conversion conditions (CIFR trades at least 130% above the $6.45 conversion price) are ￰28￱ Has Rerated, Now Faces Derating Risks?

Cipher's phased buildout has made the company scale up swiftly from what would previously be considered an underdog a few years ago to being among the ranks of large public miners ￰29￱ context, in 2022 Cipher had a mere operating hashrate of 2.8 EH/s mining capacity as of ￰30￱ of 2023-end, Cipher was still considered an emerging miner with 7.4 EH/s after the Odessa data center was completed. Today, the 777 MW energized capacity and 23.6 EH/s self-mining hashrate rival older ￰31￱ miners that pivoted to HPC and AI hosting saw valuation rerates, especially those with long-term hosting ￰32￱ ( WULF ) is one of the clearest examples. TeraWulf’s Fluidstack deals and Google support drove a ￰33￱ 8’s ( HUT ) pivot and funding actions also attracted fresh investor interest.

Cipher’s latest 10-year AI hosting agreement with Fluidstack is undoubtedly a major HPC ￰34￱ the delivery of 168 MW of critical IT load and up to 244 MW of gross capacity at Barber Lake expected by September next year, Cipher secured roughly $3 billion in contracted revenue over the contract term, with Google backstopping about $1.4 billion of Fluidstack’s obligations and taking a 5.4% equity stake in ￰35￱ has been rerated higher since the news was released late last ￰36￱ week, Nvidia ( NVDA ) and BlackRock ( BLK ) announced a $40 billion acquisition of Aligned Data Centers , valuing power capacity at about $8 million per ￰37￱ $8 million per MW valuation was about 160% higher than the valuation for listed miners whose power capacity was mostly being valued at $3 million per ￰38￱ $8 million/MW figure from the acquisition serves as a good benchmark for how institutional investors currently price large-scale HPC and AI data infrastructure power ￰39￱ compared with Cipher, there is some premium on Cipher’s energized capacity valuation.

Cipher’s market cap is roughly $7.85 billion and its enterprise value around $8 ￰40￱ ~777 MW of energized capacity currently, that gives Cipher an implied valuation of roughly $10.3 million per MW of energized power, about 29% above the Aligned Data Centers benchmark (but that is if we value the company based on its energized power alone). And as we know, energized power isn't Cipher's only asset; hence, CIFR has the potential to even trade higher, and I think CIFR doesn't face derating risks ￰41￱ we consider the power capacity to be energized in the near term according to Cipher's roadmap, I think the premium is ￰42￱ Stingray site has 100 MW capacity to be energized next ￰43￱ energization timeline for the other sites is at least two years away; some sites, such as Barber Lake Phase II, are not expected to come online until ￰44￱ timeline for the Reveille is Q2 2027, and the 500 MW Milsing site has a timeline of 2028 (it’s not yet clear whether it will be energized in phases or the whole 500 MW will be energized at once).

McLennan site also has an expected power capacity of 500 MW; the timeline for energization is 2028 ￰45￱ site at Mikeksa is also a 500 MW capacity sitting on 100 acres with a 2028 energization ￰46￱ Barber Lake Phase II energization is expected in 2029. Cipher’s Q3 results are to be announced next week, and I believe the results will show strong sequential top line improvement, as Bitcoin production was stronger in Q3 compared to ￰47￱ mined 444 BTC in Q2, and over 700 BTC were mined in Q3 while Bitcoin maintained an average price well above $100,000. GAAP net loss could persist, but I believe the Street already expects this (consensus estimates -$0.08 EPS), and any negative market reaction should be ￰48￱ One of the main risks to Cipher's valuation is the fact that Cipher is still reporting net ￰49￱ on an adjusted basis it shows improvement, but GAAP net losses have persisted for several consecutive quarters.

Q2 saw adjusted earnings of $30 million, or $0.08 per diluted share, but still posted a GAAP net loss of $46 ￰50￱ is also not yet immune to the volatile crypto market cycle; the 1,500 BTC on the balance sheet means exposure to Bitcoin price swings remains ￰51￱ CIFR’s impressive price surge from when I first covered it about two years ago could be a profit-taking opportunity for investors who were early ￰52￱ I personally believe that there is still more room for CIFR to rerate higher because of the sites and pipeline capacity under Cipher's ￰53￱ is also potential to ink more HPC deals or extend existing ones, just like WULF saw a $9.5 billion extension of its Fluidstack deal this week.

I wouldn’t rule out more momentum for CIFR in the near term, especially if Bitcoin regains momentum or if Cipher announces a subsequent HPC deal, which isn't also out of place to anticipate. I also think the institutional valuation for HPC infrastructure per MW might not be near the ceiling yet, though the $8 million per MW valuation in the Aligned Data Centers acquisition is already 160% over the range of miners valued around $3 million per MW. I believe the growing institutional demand for AI compute could continue to drive valuations for energized or near-ready MW capacity ￰54￱ current levels, I still believe CIFR is reasonably priced if we consider the total pipeline capacity.

I'm still leaning towards a Buy here despite the impressive run CIFR has had.