Data center spending has become such a force that it’s no longer simply a tech story—it’s headline news. But how protected are investors from dotcom-style setbacks?
Data centers: Boom, bust, or somewhere in between?
We’re approaching a historic inflection point: The U.S. may soon be spending more money building data centers than offices (Exhibit 1).
In 2025, power consumption by U.S. data centers rose 22% from the previous year, exceeding the entire generating capacity of New York and New Jersey combined.1 Data center spending has become such a force that it’s no longer simply a tech story—it’s headline news.
How did we get here?
The dawn of the data center
The first “data center” came into existence in December 1945, when the U.S. military turned on what was arguably the world’s first programmable computer: the Electronic Numerical Integrator and Computer (ENIAC). It weighed 27 tons, contained 18,000 vacuum tubes, and occupied 1,800 square feet of floor space in its own dedicated building. It could perform a whopping 500 floating-point operations per second, which enabled the first fully automated Monte Carlo calculations.2
We’ve come a long way since then. The NVIDIA Blackwell Ultra chip is 30 trillion times as powerful as all of ENIAC—and a heck of a lot more portable.
As computers became smaller and more powerful, two innovations fueled the rise of the data center as we know it today: local area networks (LANs) and the internet.
During the 1980s, LANs allowed businesses to dedicate certain computers to specific tasks: data processing, print processing, and storage. This led to the birth of servers, which combine higher processing power and reliability to manage requests from multiple client computers at the same time. Companies would often keep their servers in specially cooled rooms on their premises (giving rise to the term “on-prem”), and IT staff began referring to this area as the “data center.”
Voya underwrote its first infrastructure transaction in 1990 and, since then, the energy & infrastructure team has remained a core pillar of our investment grade private credit offering. We manage a circa $13 billion portfolio that focuses on single- and multi-asset secured financings through a traditional project finance structure, plus investments in corporations whose business models have infrastructure characteristics. Since Voya began classifying deals into infrastructure, corporate, and ABF in 2002, the infrastructure team has maintained a zero credit loss rate—successfully preserving principal through commodity cycles, financial crashes, geopolitical shocks, and the pandemic, while also generating competitive returns. The team has deep sector experience across the energy and infrastructure landscape, developed over multiple decades of transaction underwriting. It is our pleasure to distill that expertise into this series of educational insights for our current and prospective clients. While digital infrastructure has been the topic du jour for the past several years, we intentionally wanted to see how the asset class evolved before penning a letter addressing it. As you will notice from our portfolio, we have been cautious about making investments in digital infrastructure. While we believe that AI will be a transformational technology for mankind, we also believe that the market doesn’t fully appreciate some of the risks that are embedded in individual financings. In this letter, we will provide you with a history of the asset class, the types of transactions we see in the market, and our roadmap for investment in the space. |
A note about risk
The principal risks are generally those attributable to bond investing. Holdings are subject to market, issuer, credit, prepayment, extension, and other risks, and their values may fluctuate. Market risk is the risk that securities may decline in value due to factors affecting the securities markets or particular industries. Issuer risk is the risk that the value of a security may decline for reasons specific to the issuer, such as changes in its financial condition.