When the Gate Closes: Sizing Illiquid Allocations for the Redemption Window That Fails

The quarterly redemption gate has become one of the most persuasive selling points in private-markets fundraising. Limited partners are told they hold a stake in illiquid assets with a scheduled exit: tender offers at net asset value, quarterly windows, reasonable notice periods. The framing suggests a middle path between a closed-end drawdown fund and a daily-dealing mutual fund—access when needed, patience when not.

The framing deserves more scrutiny than it typically receives in committee minutes. The gate is not a schedule. It is a conditional promise, and the condition is the problem: it binds only when the manager chooses to honor it, and the manager's incentive to honor it weakens precisely when redemptions cluster.

The Advertised Window Versus the Effective One

Consider the structure honestly. A non-traded REIT or interval fund may promise quarterly liquidity of, say, five percent of net assets. In calm markets, that promise is nearly always kept. In stressed markets, two things happen simultaneously: redemption requests surge beyond the cap, and the manager invokes suspension provisions or prorates fulfillments. The window that existed in the term sheet narrows to a sliver exactly when a committee might actually need it.

This is not hypothetical. Post-2016 non-traded REIT redemptions, the 2020 suspensions across several semi-liquid real estate vehicles, and the gating episodes in certain credit interval funds during the spring of 2020 all followed the same pattern. Liquidity was available in every quarter except the ones that mattered.

Committees that underwrote these allocations using the advertised window—the steady-state quarterly capacity—systematically overstated the liquidity they were buying. The correct input was never the average window. It was the worst window, the one that coincides with a drawdown elsewhere in the portfolio.

Modeling the Failure Modes

A disciplined approach treats the redemption gate as an option held by the sponsor, not the investor, and then models three distinct failure modes.

Suspension. The manager halts redemptions entirely, typically citing asset-level illiquidity or regulatory concerns. The investor's position becomes, for an indeterminate period, indistinguishable from a closed-end commitment. The committee should ask: if this gate closed today and stayed closed for four quarters, what obligations in our own portfolio would go unfunded?

Proration and queueing. Requests exceed capacity and are fulfilled pro rata, with unsatisfied requests often rolling forward. This creates a queue in which the investor's place depends on everyone else's behavior—a coordination dynamic that rewards early panic and punishes patience. Modeling a queue requires assumptions about fellow holders, an exercise almost no committee performs.

Forced secondary disposition. When an institution must raise cash and the gate will not open, the only exit is the secondary market, where discounts on semi-liquid positions during stress have historically been punishing. The spread between NAV and achievable secondary pricing is the true cost of the mirage, and it should be entered into the allocation decision as an explicit haircut.

Running these three scenarios against a proposed allocation produces a very different sizing answer than the fundraising deck does. An allocation that looks modest at five percent of a portfolio can become a genuine solvency question when the committee models a two-year suspension coinciding with a capital call from its closed-end commitments.

Correlation Is the Hidden Variable

The deeper problem is that gate failures are not idiosyncratic. The conditions that trigger suspensions—broad asset-price declines, credit dislocation, sponsor distress—tend to coincide with the conditions that stress the rest of an institutional portfolio. The liquidity offered by semi-liquid structures is therefore negatively correlated with the portfolio's need for it.

This inverts the usual diversification logic. An asset that provides liquidity in calm markets and withholds it in stressed ones is not a diversifier of liquidity; it is a concentrator of liquidity risk. Committees should treat gate capacity as a conditional variable correlated with equity-market drawdowns, not as a fixed schedule independent of market state.

Governance Implications

Three practices would materially improve committee treatment of these structures.

First, every approval memo for a semi-liquid allocation should state, in writing, the assumed worst-case liquidity—the suspension scenario—and the funding plan that covers it. If no funding plan exists, the allocation is implicitly sized to the advertised window, and the memo should say so.

Second, committees should monitor the redemption queue and gate status of vehicles they hold, not merely their NAV. A fund trading at a persistent NAV discount or reporting rising tender backlogs is transmitting information that quarterly statements obscure.

Third, the allocation review should ask what the sponsor's own incentives look like under stress: management fees continue whether the gate opens or not, which means the sponsor's downside is reputational while the investor's is financial. That asymmetry belongs in the diligence file.

Sizing to the Worst Window

None of this is an argument against semi-liquid vehicles. For patient capital with genuinely long horizons, the structures can offer attractive access to private assets at institutional scale. The argument is narrower and more uncomfortable: the liquidity is not what the marketing implies, and the sizing decision should be made against the gate that closes, not the gate that opens.

Institutions that internalize this will size these allocations more conservatively than their peers. That conservatism will look like a cost in calm markets. It is, in fact, the premium paid for owning the truth about when the door actually opens.