The Pension Fund's Dilemma: Why the Largest Pools of Capital Are Moving Toward Tokenized Assets

The California Public Employees' Retirement System — CalPERS — manages roughly $500 billion in assets on behalf of 2 million members. Its actuarial target return is 6.8 percent annually. In the current environment, hitting that number requires the fund to take on meaningful exposure to private equity, private credit, real estate, and infrastructure — asset classes that are illiquid, expensive to administer, and operationally complex. CalPERS is not unusual. It is a representative example of a structural problem that affects every major pension fund on earth.

The problem, stated plainly, is this: the portfolios of defined-benefit pension funds were designed for a world where investment-grade bonds yielded 6 to 7 percent. That world ended with the post-2008 decade of near-zero rates. Even with rates higher than they were between 2010 and 2021, the yield available from liquid, high-quality fixed income remains below the return targets that pension fund actuaries need. The gap has been filled, over the past fifteen years, by a dramatic increase in private market allocations — the so-called illiquidity premium. The question that tokenization poses to pension funds is whether the infrastructure being built to put private assets on-chain will eventually give them access to that premium at lower cost, with better liquidity, and with less operational friction than the current system provides.

That question does not have a clean answer yet. But the direction of travel is becoming clearer, and several of the world's largest pension funds are no longer watching from a distance.

The Illiquidity Premium and Its Hidden Costs

Private markets — private equity, private credit, infrastructure, and real estate — have delivered excess returns over public market equivalents for most of the past three decades. The excess return is commonly attributed to the illiquidity premium: investors who accept that their capital will be locked up for seven to ten years are compensated with returns that liquid markets, where capital can exit at any moment, do not offer. The mechanism is real. The question is how much of the premium is genuinely compensation for illiquidity, and how much is fee extraction by the intermediaries who structure and administer private fund vehicles.

The cost structure of a traditional private equity or private credit fund gives a sense of the problem. A typical fund charges a 2 percent annual management fee on committed capital plus 20 percent of profits above a hurdle rate. The management fee alone, on a $10 billion fund, is $200 million per year — before a single investment has generated a return. Administrative costs — fund accounting, legal, compliance, audit, transfer agency, and reporting — add further drag. The limited partners, including pension funds, bear these costs in exchange for access to deals and managers they cannot replicate in-house.

The result is that a meaningful portion of the illiquidity premium is transferred to the general partner before it reaches the pension fund beneficiaries. Studies of private equity net returns have consistently found that the average fund, after fees, performs roughly in line with public market equivalents over long time horizons — which means the illiquidity premium is largely consumed by fees in the median case. The funds that outperform are genuine, but access to top-decile managers is rationed and relationship-dependent in ways that disadvantage smaller allocators.

What Pension Funds Actually Need From Tokenization

Understanding the pension fund use case requires being specific about what large allocators find operationally painful about private markets in their current form.

The first pain point is capital call mechanics. Private equity funds do not take investor capital upfront. They issue capital calls over several years as investments are made, requiring pension fund treasury teams to hold uninvested capital in liquid instruments while remaining ready to fund calls on short notice. The administrative coordination required — tracking capital call notices across dozens of funds, moving cash to meet deadlines, reconciling capital accounts — is labour-intensive and prone to error. Smart contract automation of capital calls and distributions is one of the clearest near-term applications of tokenization for this segment.

The second pain point is reporting and transparency. Pension funds receive quarterly reports from private fund managers with a lag of 60 to 90 days after quarter-end. During that window, the fund's exposure to its private market holdings is effectively opaque: the last known value is stale, and no real-time information is available. For a pension fund trying to manage overall portfolio risk, currency exposure, or liquidity needs, this opacity is a genuine constraint. Tokenized fund interests on shared ledgers could, in principle, provide near-real-time transparency into portfolio composition, valuation marks, and cash flow events.

The third pain point is secondary market access. When pension funds need liquidity from private market positions — either because of benefit payment obligations or asset-liability management requirements — they must access secondary markets that are illiquid, relationship-dependent, and typically require discounts of 10 to 20 percent to NAV. The secondary market for private fund interests has grown significantly in recent years, but the mechanics are slow: a secondary sale can take three to six months from initiation to close, involving legal transfer of limited partnership interests, general partner consent rights, and bilateral negotiation. Tokenized LP interests with standardised transfer protocols and automated consent processes could reduce that timeline substantially.

Who Is Actually Moving

The gap between expressed interest and actual allocation in tokenized assets remains wide at most pension funds. But several large allocators have moved beyond exploration.

Canada's Ontario Teachers' Pension Plan has been one of the more active institutional investors in the tokenization space. The fund participated in a tokenized bond issuance by the European Investment Bank and has made direct investments in digital asset infrastructure companies. Ontario Teachers' investment horizon is long and its risk appetite is above average for a pension fund, but its engagement reflects a genuine belief that the plumbing being built now will matter for asset management over the next decade.

ABP, the Dutch civil service pension fund with approximately €500 billion in assets, has been exploring tokenized real estate as a way to reduce the administrative friction of its property portfolio. Real estate is one of the most administratively complex asset classes in a pension fund portfolio: each property involves legal title management, rent collection, maintenance contracts, insurance, and valuation — all of which require specialist administrators. The proposition of tokenizing property ownership is not primarily about creating a liquid secondary market; it is about replacing a fragmented administrative stack with a shared ledger that all parties to a transaction can read from a single source.

The Abu Dhabi Investment Authority, one of the world's largest sovereign wealth funds, has been actively exploring tokenized private credit as a complement to its existing direct lending exposure. At very large scale, the friction costs of private market administration are not a minor inconvenience but a meaningful drag on net return. A fund of ADIA's size employing dozens of people to manage capital call logistics, reporting reconciliation, and secondary transactions has a strong economic incentive to reduce that headcount through automation.

The Infrastructure Gap

The barrier to pension fund adoption of tokenized assets is not primarily appetite. It is infrastructure. Pension funds operate under fiduciary obligations that require them to invest in instruments with clear legal standing, auditable custody arrangements, and established valuation methodologies. Most tokenized asset products that exist today do not yet fully satisfy those requirements.

The custody question is the most immediate. Pension fund assets must be held by qualified custodians that meet regulatory standards in the fund's home jurisdiction. The market for qualified custodians of tokenized assets is still developing: BNY Mellon has launched digital asset custody services, as has State Street and Fidelity Digital Assets, but the coverage of asset types, jurisdictions, and blockchain networks is incomplete. A pension fund that wants to hold tokenized private credit interests issued on a permissioned blockchain has a smaller set of viable custodians than one holding listed equities.

The valuation question is equally important. Pension funds are required to value their holdings at fair market value and report those values to members, regulators, and auditors. For listed assets, this is straightforward. For illiquid tokenized assets without active secondary markets, establishing a fair value requires methodologies that are not yet standardised. The same asset — a tokenized senior secured loan to a mid-market company — might be valued differently by different fund administrators using different discount rate assumptions, creating inconsistency that regulators and auditors find uncomfortable.

The legal standing issue is jurisdiction-specific and remains unresolved in most markets. In the United Kingdom, the Law Commission's 2023 report on digital assets established a legal framework recognising certain digital assets as a third category of personal property. In the United States, the Uniform Law Commission has been working on a framework for digital asset treatment under commercial law, but state-level adoption is uneven. In the European Union, the MiCA framework addresses certain token categories but does not comprehensively cover tokenized fund interests. Pension fund legal teams in most jurisdictions are still navigating material uncertainty about whether tokenized LP interests would withstand challenge in an insolvency scenario.

The Timeline That Makes Sense

The honest framing for tokenization and pension funds is not that a revolution is imminent. It is that the foundation for a structural shift is being built, and that pension funds that understand the direction of travel are better positioned to participate in it efficiently when the infrastructure matures.

The near-term applications — capital call automation, real-time reporting on tokenized fund interests, secondary market facilitation for LP stakes — are achievable within existing regulatory frameworks and do not require pension funds to hold tokenized assets directly. They require the fund managers who issue private market vehicles to tokenize their fund structures, which is a decision being made by BlackRock, Apollo, KKR, and Ares, among others, on timelines measured in years rather than decades.

The medium-term application — direct pension fund allocation to tokenized private credit, real estate, and infrastructure as distinct asset classes, with standardised custody, valuation, and legal treatment — requires regulatory clarity that does not yet exist in most jurisdictions. The timeline for that clarity is plausibly five to eight years in the United States and Europe; faster in Singapore and the UAE, which have shown greater regulatory agility on digital asset frameworks.

The long-term possibility — a world where pension fund private market allocations are held as tokenized interests on shared ledgers, with automated capital call and distribution management, real-time transparency, and liquid secondary markets that price at narrow discounts to NAV — is not speculative as a direction. It is speculative only as to timing. Every major private market manager is building toward it. Every large custodian is building the infrastructure to support it. The regulatory frameworks are moving, unevenly but consistently, in the direction of accommodation.

The pension funds that will benefit most are the ones that are learning the architecture now — not because they need to make large allocations today, but because the transition, when it arrives, will favour investors who understand the underlying technology well enough to evaluate products, negotiate terms, and manage risk without relying entirely on advisers who are also learning as they go. That is a modest ask of a $500 billion institution. It is also, for most pension funds, a larger ask than it sounds.