An accrued preferred return is not deferred. It's a tax bill that shows up before the cash does.
Most sponsors don't believe that, and the reason they don't is reasonable. The pref didn't get paid. No money left the account. So the working assumption is that nothing hits the LP's K-1 until the day a distribution actually clears. That logic is clean, it's intuitive, and if your operating agreement uses targeted allocations, it's wrong.
This is one of those gaps where a smart operator believes the obvious thing right up until the mechanics prove otherwise. So let's walk the mechanics.
What targeted allocations actually do
Most modern syndication and fund agreements abandoned the old layered allocation approach years ago. The replacement is targeted allocations, and the name is a fair description of the job.
Plain English: at the end of each year, the agreement looks at the distribution waterfall, asks “if we liquidated the partnership today at book value, who would walk away with what,” and then allocates income or loss so that each partner's capital account lands on that number. The allocations chase the waterfall. The waterfall sets the target, and the income allocation is whatever it takes to hit it.
That single design choice is where the trouble starts, because the preferred return lives inside that waterfall as a priority claim. The LP is owed their capital back plus a pref before the sponsor sees promote. Whether that pref gets paid in cash or simply accrues on the books, it still sits in the waterfall every year, and it still raises what the LP would collect in that hypothetical liquidation.
So the capital account target moves. And when the target moves, the allocation has to move with it. The pref doesn't wait politely for a check to be cut. It compounds the LP's claim in the background, year after year, paid or not.
The pref doesn't wait politely for a check to be cut. It compounds the LP's claim in the background, year after year, paid or not.
Phase one: the early years feel fine
Here's why this stays invisible for so long.
In the first couple of years, a well-structured deal is throwing off losses. You run a cost segregation study, you accelerate the depreciation, you layer bonus on top, and the deal generates paper losses that dwarf the cash flow. Those losses get allocated to the LPs, because the LPs are the ones who funded the deal and would eat the loss in a downside. That part works exactly the way everyone expects.
And because there's no net income to allocate yet, there's no phantom income yet either. The accrued pref is quietly raising the LP's liquidation claim, but with losses flowing the other direction, nothing surfaces on the K-1 that anyone questions.
This is the trap setting itself. Not because anyone did anything wrong, but because the early years reinforce the belief that an unpaid pref is a non-event. The reader sees losses, sees no income, and concludes the pref is dormant. It isn't dormant. It's accruing.
Phase two: the income and exit years send the bill
Then the deal turns. Lease-up finishes, the property stabilizes, and now there's actual profit to allocate. This is where the accrued pref stops being theoretical.
The targeted allocation engine looks at the waterfall and sees that the LP's claim has been climbing the entire hold, fed by a pref that was never paid in cash. To get that LP's capital account back to its target, the partnership now has to allocate income to them. Income equal to the accrued pref they were owed but never received.
That's the phantom income. The LP picks up taxable income on dollars that never hit their bank account. They get a K-1 with a number on it, they owe tax on that number, and when they go looking for the matching distribution, it isn't there. It's still sitting in the waterfall as an unpaid priority.
The exit makes it sharper. When the property sells, the gain gets allocated back toward the partners who absorbed those early losses, because the capital accounts have to true up. The unpaid pref pulls even more of that gain toward the LP, since their claim is still elevated by every year of accrued-but-unpaid return. And a meaningful slice of that gain isn't even capital gain. The depreciation you accelerated through cost seg comes back as Section 1245 recapture, taxed at ordinary rates, allocated to the same LPs who enjoyed the losses on the front end.
So the LP gets it from both directions. Phantom income on the accrued pref during the income years, then a heavier and partially ordinary gain allocation at sale. All of it driven by a return they were promised and have not yet been paid.
Why none of this is a flaw
It would be easy to read this as a defect in targeted allocations or a reason to fear an accrued pref. It's neither.
Targeted allocations exist because they keep the capital accounts honest. They make sure the tax follows the economics, which is exactly what you want when the IRS is the one grading the substantial economic effect of your allocations under Treas. Reg. §1.704-1(b). An accrued pref that compounds an LP's claim is doing precisely what it's supposed to do. The structure is working.
The failure isn't in the structure. It's in the modeling. The sponsor who treats an unpaid pref as a tax non-event isn't reading the agreement they signed. They're modeling the cash and ignoring the capital account.
And that's the recurring pattern in this business. The complexity isn't there to trap you. It's there because the economics are genuinely intricate, and the people who take the time to understand the intricacy are the ones who don't get surprised by a K-1. The game is winnable. It just rewards the operators who model the full lifecycle instead of stopping at the bonus depreciation line.
The takeaway
So the question to carry into your next deal is simple. Are you modeling what your LPs collect, or what their capital accounts are required to reflect? Because under targeted allocations, those are two different numbers, and only one of them drives the tax.
Are you modeling what your LPs collect, or what their capital accounts are required to reflect? Under targeted allocations, those are two different numbers, and only one of them drives the tax.
An accrued preferred return is never neutral. Accrued does not mean deferred. It means the tax arrives before the cash, lands on the LPs who funded the deal, and gets heavier at exit. Model the lag before you sign the term sheet, not after the K-1 goes out.