Two weeks ago, when SB Energy’s SB Energy decided to delay its ambitious IPO for the construction of a massive data center, concerns were raised about the future of the project (“SB Energy Delays IPO Funding World’s Largest Data Center Amid Investor Revolt, Public Outcry”, Sep 22). Fast forward to today, and it seems that the challenges faced by SB Energy are not unique.
Just recently, Australian data center operator, Firmus, also encountered difficulties with its IPO, ultimately closing the books without a clear pricing structure in place. The initially proposed valuation of A$11 per share, which would have valued the company at A$43.7 billion ($30.4 billion), was slashed by 25% to A$8.25, with the company struggling to maintain it at A$8. This turn of events sheds light on the broader issues plaguing the AI supercycle, particularly when it comes to financing mega data center projects.
Firmus’ journey from a $5.5 billion private valuation to a $30 billion IPO in just six months raises eyebrows, especially considering that the company has only completed 4% of its planned capacity. The reliance on a new valuation metric, “EV+1/EBIT+2,” to justify the inflated IPO price further highlights the challenges faced by data center developers in securing funding for their projects.
Moreover, the decision to allocate half of the IPO to existing strategic and financial investors, including Nvidia, Coatue, Blackstone, and Jane Street, has raised concerns among potential investors. The circular financing arrangement, where IPO proceeds would fund the purchase of Nvidia GPUs for Firmus’ projects, adds another layer of complexity to the situation.
Ultimately, the crux of the issue lies in the mismatch between the company’s equity value and its staggering debt load. With the equity potentially worth less than the debt, the IPO’s fate hangs in the balance, highlighting the precarious position of AI infrastructure projects in the current market environment.
As Firmus prepares to lodge its prospectus and begin trading, the uncertainties surrounding the IPO serve as a cautionary tale for the broader AI industry. The recent setbacks faced by both SB Energy and Firmus underscore the challenges of financing mega data center projects in an increasingly volatile market. The future of the AI supercycle remains uncertain, with the fate of data center projects hanging in the balance.
