The Waterfall Strategy

Home / Articles

Is an Installment Sale Worth It? 3 Times It Won't Help Losses

By Hans Goldstein · Updated 2026-09-27

An installment sale is worth it for passive losses only when there are losses that cannot be used today and enough gain to meet them. If your losses already offset your income, if your deals pay income instead of losses, or if the exchange you are doing creates no new losses, the waterfall has nothing to do. Spreading the gain may still beat a cash sale by a little, but that is a bracket tool, and it comes with years of locked-up money.

The book devotes three full cases to where it fails. These are illustrative composites run on the book's engine, not real clients. The numbers show the pattern; yours will differ.

Pattern 1: The pro who runs his own rentals

Case 11 (illustrative). A 56-year-old owner of eleven rentals runs every one himself and logs 2,000 hours a year. He is a real estate professional who materially participates: the real thing. His rentals throw off about $80,000 a year of losses, and they already offset his other $500,000 of income every year. Nothing sits on Form 8582.

He sells a building for $2 million with a $1.4 million gain, in California. The engine ran $1 million over five years, level.

Plan (Case 11, engine output) Year-1 tax on the gain 10-year tax on the gain Ahead after 10 years
Sell for cash $436k $436k (baseline)
Installment sale, five years $248k $364k $107k
Same structure without his rental losses $80k

The structure beats cash by spreading $1.4 million across five tax years. But no year goes to zero, and the roughly $26k gap is mostly a federal bracket effect plus California (which ignores professional status and treats all rentals as passive). You can't unlock a door that's already open. A 1031 into another building he runs himself came out ahead of the note in the engine.

Two rules make this case: a professional's rental gain from a materially participated activity is nonpassive (Temp. Reg. §1.469-2T(c)(2)(i)(A)), so it cannot absorb anyone's passive losses; and when he meets the real estate professional safe harbor (more than 500 hours in the rental activity), his rent and sale gain are outside the 3.8% net investment income tax (Reg. §1.1411-4(g)(7)), so there is no 3.8% to save.

How to spot this case in two minutes:

Pattern 2: The investor whose deals pay out

Case 12 (illustrative). A dermatologist earning $520,000 invested for years as a limited partner. The deals are mature and pay about $90,000 a year of passive income. Every paper loss is already used against that income. No carryforward. She also owns a remodeled rental worth $2.9 million with $2.25 million of basis, a cost segregation study, and a $300,000 loan.

Three strikes:

Plan (Case 12, engine output) Year-1 tax on the gain 10-year tax on the gain Ahead after 10 years
Sell for cash $291k $291k (baseline)
Installment sale, five years, $1.2M $216k $277k $44k
Same structure without her LP income $50k

$13k of tax saved over ten years, for $1.2 million locked up for five. The version counting her deals came out behind the plain version: income stacked on each payment pushed it into a higher bracket. She took the cash.

How to spot this case:

Pattern 3: The all-cash trade-down

Case 13 (illustrative). A California couple, both 66, sells a $3 million building with no loan and exchanges into a $1.5 million replacement bought with cash. About $1.38 million of equity is left over and becomes boot. They take it over ten years instead of as cash, counting on the new building's depreciation to shelter it.

It does not. In an exchange, carried-over basis keeps the old building's depreciation schedule. Only excess basis, added with new debt or new cash, is treated as newly placed in service and, for used property, only that excess can take bonus depreciation (Reg. §§1.168(i)-6(d)(1), 1.168(k)-2(g)(5)(iii)). They added neither. No new debt, no new basis, no new losses.

Plan (Case 13, engine output) Year-1 tax on the gain 10-year tax on the gain Net position after 10 years
1031 + cash boot $504k $728k $3.04M
1031 + boot over ten years $65k $654k $3.25M

Spread boot still came out $208k ahead, the same with or without the "waterfall" losses, because there were none. Every payment pays full freight, federal and California. It can still be worth doing if they will not need the money for ten years; it just is not a waterfall. A Delaware Statutory Trust interest for the leftover, or a brand-new replacement whose short-life parts can take bonus, are the alternatives to model.

How to spot this case:

Ask for the replacement's depreciation schedule before closing. One page tells you whether there is a waterfall or just a spread.

The spectrum: when it is dumb

The book ran one $2 million sale about a hundred ways (round numbers, engine output). Measured as "ahead after 10 years" against cash, households with no losses and $500k of income came out about $95k ahead, a high-basis seller about $79k, and an LP investor whose deals pay income about $68k. The book's tier for under $100k: not worth locking up money.

Skip it, or spread very little, if:

Compare the math for your own sale in installment sale vs. lump sum.

Everything that can break it (short list)

Bottom line

The waterfall needs two things: gain on a schedule and losses that cannot otherwise be used. The active professional has no stuck losses, the investor with paying deals has no reservoir, and the all-cash trade-down creates no new losses. In each, spreading may still edge out cash, but only through brackets and time, in exchange for years of locked money. Know which one you are paying for. When it does fit, the waterfall explainer shows how, and the free book gives all thirteen cases, including these three.

Questions to ask your CPA

Get the full playbook. The Waterfall Strategy, the 20-minute version and the one-page Cliff Notes, free.

Send me the books Try the calculator

Educational only, not tax, legal or investment advice. Examples are illustrative. Have your CPA or tax attorney review your facts before you act. Hans Goldstein is a licensed insurance agent (CA Insurance License #4273294) and is not a CPA or attorney. He is paid a commission only if a structured installment sale is funded; seller financing pays him nothing. Disclosures.