Data Centers

Sector thesis

Data centers are the physical buildings and infrastructure that store, process, and move digital information—basically the invisible backbone of the internet, cloud computing, and AI. Right now, they're experiencing explosive demand because AI models require enormous amounts of computing power, and every major tech company is racing to build or lease capacity to train and run these systems. This isn't a temporary spike; it's a structural shift. As AI workloads grow and companies move away from owning their own servers toward renting cloud capacity, data center operators are in a position to capture that spending for decades. The sector breaks into three main pieces: hyperscale operators (companies like Amazon, Google, and Microsoft that build massive facilities for their own use and rent excess capacity), independent data center landlords (firms that own buildings and lease space to multiple customers), and infrastructure providers (companies supplying power, cooling, and networking equipment). Each has different economics and risk profiles. The biggest risks are real. Data centers consume enormous amounts of electricity, and power costs are rising—some facilities may struggle to find reliable, affordable power. There's also execution risk: building new capacity takes years, and if demand softens or competition intensifies, operators could be left with expensive empty buildings. Regulatory uncertainty around energy use and environmental impact is growing too. Finally, the sector is capital-intensive, meaning companies borrow heavily, so rising interest rates directly hurt profitability. For a retail portfolio, data center exposure works as a long-term infrastructure play, similar to owning utilities or pipelines. Watch quarterly earnings reports for metrics like occupancy rates (what percentage of space is rented) and power pricing trends. This sector rewards patient investors but punishes those chasing short-term momentum.