Who Gets the Cost Segregation Deduction in a Partnership?

TL;DR: Nobody at the entity level. Who gets the cost segregation deduction in a partnership is answered by the operating agreement, which splits the depreciation and sends each piece out on a K-1. Whether a partner can actually use their piece is a separate question, answered by their basis and by two more limits stacked on top of it.

Key Takeaways:

• The split follows the operating agreement, not the ownership percentages. It only has to carry substantial economic effect under the capital account rules.
• A partner's loss stops at their outside basis. At-risk and passive rules cut it further. What gets stopped carries forward instead of disappearing.
• With a section 754 election in place, a buying partner gets their own basis adjustment, and that adjustment can qualify for bonus depreciation on its own terms.

Who gets the cost segregation deduction in a partnership?

The partners do, in whatever proportions the operating agreement says.

An engineering-based study does the same work on a partnership-held building that it does on one you own outright. Components get reclassified into 5, 7, and 15-year lives. The deduction comes out the same size. Then it hits a partnership return, which is not a taxpayer. The LLC computes the deduction, allocates it, and reports each partner's share on a Schedule K-1. From there it is your problem, and it behaves differently depending on who you are.

That is the whole structural difference, and it is the source of every complication below.

Does the split have to match everyone's ownership percentage?

No. And that is the point of putting the building in a partnership in the first place.

Section 704(b) lets a partnership hand out a single item, depreciation included, in a way that does not follow the ownership percentages. The allocation has to carry what the regulations call substantial economic effect. Deals use this constantly. The sponsor takes 20 percent of the cash and 5 percent of the depreciation. An investor who came in for the tax benefit takes an outsized share of the losses and gives up the corresponding upside later.

The test has teeth. Treas. Reg. section 1.704-1(b)(2)(ii)(b) says an allocation has economic effect "if and only if" three things hold for the life of the partnership. Capital accounts get maintained under the rules in (b)(2)(iv). Money paid out at liquidation follows the positive capital account balances. And a partner whose account is negative at liquidation has to put the money back. That last one is the deficit restoration obligation. Most operating agreements skip it, because nobody wants to sign up to write a check on the way out.

The regulation builds in a workaround. An agreement with no deficit restoration obligation can still pass, under the alternate test at (b)(2)(ii)(d), if it carries a qualified income offset. That clause routes income back to a partner who unexpectedly goes negative, until the hole is filled.

None of this is boilerplate. Say your agreement splits depreciation one way and keeps capital accounts another way. The split gets thrown out, and the deduction goes back to the partners according to their real interests in the partnership. So a special allocation of cost segregation depreciation is worth exactly as much as the drafting behind it. Have the person who wrote the agreement look at it before the study lands, not after.

Why does a partner get a big loss on the K-1 and still not use it?

Because three separate walls sit between the K-1 and the return, and each one has to be cleared in order.

First is basis. Section 704(d)(1) allows a partner's distributive share of loss "only to the extent of the adjusted basis of such partner's interest in the partnership at the end of the partnership year in which such loss occurred." Outside basis is roughly what you put in, plus your share of income and liabilities, minus distributions and prior losses. Cost segregation is very good at producing a first-year loss bigger than a partner's basis. The excess is not lost. Section 704(d)(2) carries it forward until basis shows up to absorb it.

Second is at-risk. Section 465 asks a narrower question than basis does: how much of this are you actually on the hook for. Nonrecourse financing and guarantees change the answer.

Third is passive. Section 469 asks whether the activity is passive to you. A limited partner writing checks is almost always passive, and a passive loss offsets passive income and nothing else. A partner who meets the real estate professional test, and materially participates, sits somewhere else entirely. Two people can hold the same percentage of the same LLC, receive identical K-1s, and get completely different results, because the passive rules are about them, not about the building.

This is where partnership cost segregation goes wrong most often, and it goes wrong before the study is ordered. Somebody models a deduction, divides it by ownership percentage, and shows an investor a number that partner has no capacity to use for years.

What happens when a new partner buys into a building that already had a study?

They inherit the schedule, unless the partnership makes an election.

The default is unpleasant for a buyer. The partnership's basis in the property does not change when an interest changes hands. The building keeps depreciating on the old schedule. The new partner picks up whatever share of what is left the allocation gives them, even though they just paid today's price to get in.

Section 754 fixes the mismatch. Once made, the election "shall apply with respect to all distributions of property by the partnership and to all transfers of interests in the partnership during the taxable year with respect to which such election was filed and all subsequent taxable years." That last part is the part people skim. It is not per-transaction. It is on until it is revoked, and revocation is available only under the regulations. Everything that happens afterward comes under it, which includes the transfers where the adjustment goes the wrong direction.

With the election in place, section 743(b) gives the transferee a basis adjustment on a sale, an exchange, or the death of a partner. The adjustment belongs to that partner alone and does not touch anyone else's numbers. Section 743(d)(1) can force the adjustment with no election at all, when the partnership has a substantial built-in loss. That means one of two things. The partnership's adjusted basis in its property runs more than $250,000 above the value. Or the new partner would take more than $250,000 of loss if the partnership sold everything at market value that day.

The study is what makes the adjustment worth having. Section 755(a) allocates the adjustment "in a manner which has the effect of reducing the difference between the fair market value and the adjusted basis of partnership properties." A completed cost segregation study is the record of where that difference actually sits. Without one, the step-up drifts toward 39-year and 27.5-year property, and a real number turns into a slow one.

Does the new partner's step-up qualify for bonus depreciation?

Often yes, and this is where the detail matters, so here is the pro-level version.

Treas. Reg. section 1.168(k)-2(b)(3)(iv)(D)(1) treats a section 743(b) increase as qualifying used property. Two conditions. Neither the buying partner nor any predecessor held a depreciable interest in the piece of property the adjustment lands on under section 755. And the transfer meets the used property acquisition rules. Relatedness gets tested by comparing the seller and the buyer of the partnership interest, under (D)(2).

Three fact patterns in the regulation's own examples defeat it. A buyer related to the selling partner fails the acquisition requirements. A transferee taking the interest at death under section 1014 fails, because section 1.179-4(c)(1)(iv) is not satisfied, though the step-up at death is doing plenty of other work in that situation. A partner who already held a depreciable interest in that property fails as to the portion they held. Timing has a trap of its own. Buy a partnership interest and dispose of it inside the same tax year, and bonus on the adjustment from that first purchase is gone.

Two neighboring rules go the other way and are worth knowing before somebody quotes a number in a model. Remedial allocations under section 704(c) do not qualify, under (b)(3)(iv)(A). Section 734(b) increases, the kind generated by distributions rather than transfers, do not qualify either, under (b)(3)(iv)(C). A 743(b) adjustment and a 734(b) adjustment are both basis increases created by the same election, and only one of them gets bonus treatment.

One more piece of plumbing. Under section 1.168(k)-2(f)(1)(ii)(G), each partner's 743(b) adjustment, for each class of property, is treated as its own class of property for purposes of the election out of bonus. So a partnership can elect out for one partner's adjustment without touching anyone else's, which is useful when one partner wants the deduction and another wants the income.

Bonus is currently 100 percent and no longer phasing down for property acquired after January 19, 2025, under the law signed in July 2025. I wrote up what that changed here. Acquisition date drives it, not placed-in-service date, which surprises people every time.

Should the study happen before or after a partner buys in?

Before, in nearly every case I see.

A finished study gives the section 755 allocation something real to work with, so the incoming partner's step-up lands on the short-life components instead of drifting into the long ones. It also gives everyone at the table a number they can plan around while the deal terms are still moveable. Special allocations, the 754 election, and the timing of the buy-in are all easier to arrange before the paperwork is signed than after.

There is one situation where the order barely matters, and that is a partner who cannot use passive losses this year or next. For that partner the deduction is a carryforward either way, and the sequencing question is smaller than the participation question.

Frequently Asked Questions

Does the partnership itself save any tax from cost segregation?

No. A partnership computes the deduction but does not use it. Every dollar goes out on a K-1 and gets used, limited, or suspended at the partner level.

Can we give one partner most of the depreciation?

Yes, if the operating agreement supports it. Section 704(b) allows a special allocation that carries substantial economic effect. That takes three things: capital accounts maintained under the regulations, liquidation payouts that follow positive capital account balances, and either a deficit restoration obligation or a qualified income offset. An allocation the agreement does not support gets thrown out and redone.

Is the section 754 election a one-time thing?

No, and that is the part worth thinking about. The statute applies it to all distributions and all transfers for that year and all subsequent years. It works in both directions, including transfers where the adjustment reduces basis.

A partner inherited an interest. Does the step-up get bonus depreciation?

The section 743(b) adjustment on a transfer at death does not qualify for bonus. The adjustment itself is still available, and it still depreciates. It just does so on the normal schedule for the class of property it lands in.

Let's look at your situation

Most of the money lost on partnership cost segregation is lost in the operating agreement and the timing, not in the study. By the time the report is finished, the allocation is what it is.

I offer a no-cost analysis, delivered through CSSI. No obligation, no pitch, nothing to gather ahead of time. If a study will not pay for itself, I will tell you that.

Book a time: https://calendly.com/david-wiener/cs
Or call 770-224-8504, option two.