The 90 percent asset test looks mechanical until a fund misses it by timing a property sale wrong

The working capital safe harbor being QOZB-only is the thing that bites recycling funds hardest, and almost nobody reads that distinction in the PPM.

The structural point here is correct and the layering distinction matters. The working capital safe harbor in the QOZB regulations gives the business entity up to 31 months to deploy cash into a qualified project, provided a written plan exists and the schedule is followed. That protection does not travel up to the QOF itself. Cash at the fund level on either testing date is exposed, full stop, and a manager who confuses the two has a documentation problem at minimum.

The assumption doing the most work in most recycling pitches is that the gap between sale closing and new acquisition closing will be short and predictable. In practice, OZ deal timelines compress on the buy side because the seller knows the buyer has a deployment deadline. That asymmetry costs basis points before construction ever starts.

The risk you did not name is the penalty's cumulative character. The monthly charge accrues on the shortfall amount at the federal underpayment rate plus five percentage points, and in a fund that sits on a large cash balance across two testing dates, that is not a rounding error. The penalty does not threaten the investor's gain deferral, but it does come out of the fund's economics, which means it comes out of the investor indirectly.

On reasonable cause: the IRS has discretion to waive if the manager demonstrates the failure was outside their control and they acted in good faith, but this is a facts-and-circumstances argument, and the burden is on the fund. Whether a particular fund's documentation would support that argument requires a tax professional to assess, not a forum.

Two things worth pulling from the operating agreement before committing capital: whether the agreement specifies where redeployment proceeds sit at each testing date, and whether the manager has discretion to restructure the entity holding the cash to manage the testing date exposure. Does the subscription deck you reviewed describe a recycling strategy, and did the sponsor address the QOF-level testing date in the disclosure?

What's the fund's realistic redeployment window in Phoenix versus Memphis, because those acquisition timelines are not the same.

What did the PPM say the average days-to-redeploy was on their last deal?

The penalty math is worth running because it changes how you read a recycling strategy on paper. The federal underpayment rate has moved around, but even at six or seven percent plus the five-point add-on, a fund holding ten million in sale proceeds through a testing date shortfall is looking at something in the range of nine to eleven thousand dollars per month on the full shortfall amount, and that compounds across months if redeployment drags. A two-asset fund with one property sold is not holding a marginal shortfall; it is potentially holding zero qualifying assets at the QOF level until the new acquisition closes, which means the penalty base is the entire deployed capital, not a rounding error.

The reasonable cause argument the original post flags is real but genuinely narrow. The IRS has accepted it where a fund documented a specific redeployment plan, had a signed purchase agreement in place, and the delay came from an external cause outside the manager's control. "We were still sourcing the next deal" does not clear that bar, and most operating agreements I have seen do not actually commit the manager to a cure timeline, they just describe the penalty as a fund-level cost that reduces distributions. That allocation of risk to investors without a corresponding obligation on the manager to avoid the shortfall is the thing I would press on before committing capital. What does the subscription agreement say about who bears the penalty economically, and is there a cap on how many testing dates the manager can miss before it constitutes a breach?

Does the operating agreement even define "testing date" or just defer to IRC 1400Z-2?

The misconception worth correcting is that the penalty itself threatens an investor's deferral election, which the original post already notes it does not, but managers sometimes frame the penalty risk as existential to deferral when the actual exposure is to the fund's net returns and the manager's reputation. What often gets missed is that some funds address the testing date gap not through entity structure but through the timing of when capital calls go out, specifically by delaying the call until the replacement acquisition is close enough to closing that the redeployment window is days rather than weeks, which keeps the QOF-level cash exposure narrow. That approach has its own problem: investors who committed capital are sitting on an unfunded commitment that is not yet generating qualified gains treatment, and the 180-day reinvestment clock from the investor's triggering event keeps running regardless of when the fund calls the money. The subscription documents should specify who bears the penalty economically if the test is missed, because an operating agreement that simply says the fund will use commercially reasonable efforts to comply is not the same as one that caps manager fees or clawbacks carry in a penalty year.

What does the fund's distribution waterfall say about a year in which a penalty reduces cash available for distribution, specifically whether the preferred return calculation treats the penalty as a fund expense that hits the limited partners first?

The thing I have never seen work is a manager who describes the recycling strategy in the deck but runs a single-entity structure at the QOF level and claims the working capital safe harbor will cover the gap. It does not cover it at that level, the cash is naked on the testing date, and the reasonable cause argument they have ready is usually a letter from counsel that was written six months after the fact and amounts to "we meant to redeploy faster." The IRS does not find that compelling and the penalty accrues monthly while that letter is being drafted.

What I have seen instead is funds that solved this by never letting the QOF hold sale proceeds at all. The recycle goes through a QOZB that was structured with a 31-month working capital plan documented before the first dollar moved, the QOF interest in that QOZB stays on the books the entire time, and the QOF never technically holds cash because the business entity is holding it. That requires the acquisition timeline to be modeled into the QOZB agreement before the sale of the first asset closes, not after. I have looked at three Pacific Northwest-focused OZ funds in the last eighteen months and only one had that structure in the operating agreement rather than in a slide deck footnote. My actual question to you is whether the fund you are looking at can show you the QOZB-level working capital plan in writing with dates, not just tell you the safe harbor applies.

Two-asset fund selling one property is basically a coin flip on whether redeployment clears the testing date, and I don't know why more subscription docs don't just say that plainly instead of burying it in risk factors.

What I want to know before committing to any recycling-strategy fund is how many prior deals they've actually recycled and whether any of them crossed a testing date while the proceeds sat at the QOF level, not the QOZB, the fund itself.

@op you mention "reasonable cause argument" like managers have one ready, but has anyone actually seen that argued successfully with the IRS on a missed 90 percent test, or is that mostly a theoretical out that sounds better in the PPM than it performs in practice?

Recycling into a 1031 simultaneously buys redeployment time and resets the basis clock, which changes which structure wins.

Recycling strategy funds should price the penalty drag into their IRR model at subscription, not treat it as a contingency.

The risk that compounds this is a fund using a subscription-based capital call structure, where investor closings happen on a rolling basis. Each new closing can reset which date counts as the "first six-month period" for that tranche, so the manager may be managing multiple overlapping testing windows simultaneously, and a redeployment delay that is harmless for one cohort of investors lands directly on a testing date for another. Ask the sponsor how the fund tracks testing dates across multiple closing cohorts, because if the answer is vague, the exposure is probably larger than any single sale-and-redeploy gap.