Who Owns the Song?

In November 2024, Blackstone issued $1.47 billion in asset-backed securities. The collateral was not real estate or corporate debt. It was the songs of the Red Hot Chili Peppers, Neil Young, Shakira, Leonard Cohen, Fleetwood Mac, Beyonce, and dozens of other artists whose recordings and publishing rights Blackstone had acquired through its Hipgnosis Songs Fund purchase earlier that year. The bonds were rated. The cash flows were modeled. The music was the mortgage.

This is what the music catalog market became in the years between 2018 and 2026: a structured finance product, dressed in the language of art and culture, underwritten by the same institutional logic that prices any long-duration income stream against a risk-free benchmark. Between 2020 and 2025, an estimated $26 billion flowed into music catalog acquisitions globally, from private equity firms, sovereign wealth funds, major publishers, and specialist vehicles built for the specific purpose of buying other people's songs.

The investment thesis was always straightforward. A catalog of hit songs generates royalty income from streaming, sync licensing, performance rights, and mechanical reproduction. That income is recurring, largely recession-resistant, and grows as streaming penetration expands globally. The question was only what multiple of annual earnings a rational buyer would pay for that stream. At the peak of the market in 2021, with interest rates near zero, the answer was 22 to 30 times net publisher share. The $500 million Bruce Springsteen deal, the Bob Dylan catalog sale to Universal for a reported $300 to $400 million, the Hipgnosis acquisitions of Neil Young and Shakira — all priced in an environment where cheap capital made long-duration cash flows look irresistible.

Then interest rates moved. The same DCF math that justified 25x multiples at a 2 percent discount rate produced a very different answer at 4 to 5 percent. Hipgnosis Songs Fund, the publicly traded vehicle that had built its portfolio at peak multiples, saw its share price collapse as net asset value was written down. Blackstone acquired the fund in July 2024 for $1.58 billion, took it private, rebranded it as Recognition Music Group, and proceeded to sell a portion of the assets to Sony Music Publishing while securitizing the remainder. The era of public-market scrutiny on every music transaction was over.

What remains in 2026 is a market that is selectively hot at the top and considerably more disciplined below. Britney Spears sold her catalog to Primary Wave for a reported $200 million. Concord acquired 8,500 songs from Spirit Music Group for approximately $360 million. Warner Music and Bain Capital announced a joint vehicle with up to $1.2 billion earmarked for catalog acquisitions. Blue-chip legacy catalogs — the Springsteen and Dylan tier — still command 18 to 24 times net publisher share. Mid-tier and independent catalogs trade at 8 to 14 times, with a 5 to 15 percent haircut applied to any catalog exposed to AI voice cloning risk.

That last factor is the newest and least understood variable in the market. Several majors and publishers signed AI training licenses with model developers in 2024 and 2025, the terms of which remain largely confidential. Buyers now price in the option that a catalog can generate additional revenue from AI licensing — but they also price in the risk that AI-generated music eventually compresses the streaming market share of human-recorded catalogs. The artists who signed away their masters a decade ago have no say in any of this. The catalog buyers do.

The songwriter, whose work generates the royalty stream that makes the bond possible, typically earns a percentage of that stream determined by a contract signed years or decades earlier, in a negotiation where they almost certainly had less information and less leverage than the institution now holding their rights. The bond is real. The question of who it was built for has a clear answer, and it is not the person who wrote the song.