Joint Ventures and Private Credit

The newest structures solve one problem. The largest technology companies now want data center capacity on a scale that strains even their own balance sheets, and they would prefer not to carry all of it as owned property and debt. The answer has been to split ownership from use: let outside investors own the building, let the technology company lease it, and design the contracts so that each side holds the risk it prefers.

The clearest recent example is Hyperion. On October 21, 2025, Meta announced a joint venture with funds managed by Blue Owl Capital to develop and own the Hyperion data center campus in Richland Parish, Louisiana. Under the announced terms, the Blue Owl funds hold eighty percent of the joint venture and Meta holds twenty percent. Meta leases the capacity under agreements with a four-year initial term and options to extend. To give investors comfort beyond that short initial term, Meta provided the joint venture with a residual value guarantee covering the first sixteen years.

Each piece of that structure does a job. The joint venture, or JV, separates ownership of the asset from the company that uses it. The eighty-twenty split puts most of the equity, and therefore most of the capital requirement, with the outside investor. The short initial lease gives Meta flexibility: if its needs change, it is not locked into a long commitment. And the residual value guarantee is the bridge. It promises that if Meta does not renew and the property is worth less than an agreed amount, Meta will cover some or all of the shortfall. The investors get downside protection close to that of a long lease, while Meta keeps the option to leave.

The debt came from a different place. Press reporting on the deal described the joint venture's debt as bonds sold to large institutional investors through a special purpose vehicle, rather than a traditional bank loan, and Meta discussed the arrangement in its quarterly filing with the Securities and Exchange Commission. That is the private credit channel at work. Private credit refers to lending by non-bank institutions such as insurance companies, pension funds and dedicated credit funds. These investors want long-dated, fixed-income assets backed by strong counterparties, and a data center tied to a company of Meta's credit quality fits the brief.

Why do this rather than borrow directly? Three reasons are commonly given, and they can be stated without guessing at any company's private motives. First, capacity. Spreading the capital requirement across outside investors lets a company build more at once. Second, flexibility. A lease with extension options is easier to adjust than owned property. Third, balance-sheet presentation. Depending on how the accounting rules treat the arrangement, a joint venture of this kind may not be consolidated onto the company's own balance sheet, which affects reported debt levels. Analysts debated exactly that point after the Hyperion announcement, which is a sign that the structure's benefits and its opacity are two sides of one design.

The joint venture form is not new to the sector. The Equinix xScale ventures with GIC, described in the previous module, apply the same principle: an operator contributes expertise and a minority stake, and a financial partner contributes most of the equity. What is new is the size of the individual projects and the direct involvement of the end user as both tenant and co-owner.

Now the risk transfer, stated plainly. In a conventional lease, the landlord bears the risk that the tenant leaves when the lease ends. In the Hyperion structure, much of that risk is returned to the tenant through the residual value guarantee, but only up to the guaranteed amount and only for the guarantee's term. Investors bear construction risk alongside the developer, credit risk on the tenant, and whatever residual risk lies beyond the guarantee. Faraday's read is that these structures are best analyzed as corporate credit wearing real estate clothing: the building matters, but the guarantee from a creditworthy tenant is what the bondholders are really underwriting.

There is a second-order effect worth watching. As more capacity is financed through joint ventures, private credit and securitizations, a growing share of data center exposure sits outside the public REITs and outside bank balance sheets, in the portfolios of insurers and credit funds. That diversifies where the risk lives. It also makes the total exposure harder to see, because private placements disclose less than listed securities.

Read any new financing structure through one question: who pays if the tenant walks away? In a long lease, the landlord's investors do. In a guaranteed joint venture, the tenant does, up to a limit. In an asset-backed security, the bondholders do, after the equity is gone. The label on the deal matters less than the answer.