Corporate cash and capital preservation: Using tax-loss harvesting to offset realized gains
Sponsored content. This article was supplied by a third party and Business Money has been paid to publish it.
The finance director at a mid-sized manufacturing firm opened the November portfolio review and found something awkward. The treasury book had booked a $1.4 million gain on a bond position sold in June, and the tax provision had quietly grown teeth. Nothing was wrong. Everything was profitable. That was exactly the problem.
Corporate cash does not sit still anymore. Operating surpluses get parked in short-duration funds, laddered notes, and equity sleeves run by an outside adviser. Those positions get traded. Trades create realized gains, and realized gains create a tax bill that lands in the same fiscal year whether or not anyone planned for it.
Harvesting losses is how disciplined treasury teams bring that bill down without changing what they actually own. The idea is old and the mechanics are unglamorous. Sell the positions sitting underwater, book the loss, apply it against the gain, and put the money back to work in something close enough to keep the allocation intact. Done properly, it is not a trick. It is housekeeping.
Why realized gains sneak up on treasury teams
Most finance teams track unrealized performance. Marks move, statements arrive, nobody panics. Realization is a different event entirely. The moment a manager sells, the gain stops being a number on a screen and becomes a line on the return.
That gap is where the surprises live. An outside manager rebalances in March. A bond gets called in July. A concentrated position gets trimmed because a risk limit says so. None of those decisions passes through the tax desk, and by the fourth quarter the company is staring at a number nobody forecast.
The remedy is not to trade less. It is to look at the whole book at once. Gains in one sleeve and losses in another belong to the same taxpayer, and the tax code treats them that way: capital losses offset capital gains before anything reaches the return, and only the net figure matters.
Building the harvest around the fiscal calendar
Timing is the whole discipline. A loss counts for the year it is realized, so December 31 is a hard wall rather than a suggestion.
Good teams start in October. Pull a lot-level report from every custodian, sort by unrealized loss, and flag anything below cost with enough size to matter. Then match those candidates against the year’s actual realized gains instead of some vague ambition to save tax. The exercise takes an afternoon, and it converts an unpleasant surprise into a manageable number.
Then decide what you actually want to keep. Some underwater positions are still doing their job in the allocation, and selling them purely for tax reasons swaps a real risk decision for a paper one. That is a bad trade. Harvest where the thesis has already broken. Those losses cost you nothing to take.
Staying clear of wash-sale and superficial-loss rules
This is where good plans fall apart. Sell a security at a loss, buy it back within thirty days, and the deduction vanishes. The United States calls that a wash sale, and Canada calls the same thing a superficial loss. The window runs thirty days before and thirty days after the sale, so a routine reinvestment on either side can quietly kill the benefit.
The rule reaches further than most people expect. It follows purchases made by a spouse, by a corporation under common control, and inside retirement accounts. For a company running several managed sleeves at different custodians, that matters enormously, because one manager can buy back exactly what another just sold and nobody notices until filing season.
The way through is exposure, not identity. Sell the fund and buy a different fund tracking a different index in the same asset class. Sell one issuer’s investment-grade paper and buy another’s at similar duration. You keep the risk profile and you keep the loss. Write the substitution list before December so nobody is improvising under pressure.
Turning a harvested loss into capital preservation
A harvested loss is cash the company keeps. That is the entire point, and it deserves saying plainly, because tax work gets filed under compliance when it is really a source of return.
Excess losses are not wasted either. When losses run past the year’s gains, they carry forward and sit on the books as a shield against future realizations. A rough year becomes an asset for the year after, which is exactly the asymmetry a treasury function should want.
The basis reset is the other half of the story. Repurchasing at a lower cost basis means a larger gain later, so harvesting defers tax rather than erasing it. That is still a win when the deferral is long and the saved cash earns a decent rate in the meantime, but model it honestly. Anyone selling this as free money is selling something else.
Making it a process instead of a year-end scramble
Write the policy down and name who owns it. Give outside managers a standing instruction that loss harvesting is in scope, and coordinate across accounts so nobody triggers a wash on someone else’s sale. Keep lot-level records, because a disallowed loss gets added to the basis of the replacement shares and somebody has to track that number for years.
Repeatable tax-loss harvesting strategies turn all of this into a checklist instead of a November emergency. The same discipline applies to the rest of the capital structure: share repurchases carry their own consequences, and it is worth understanding the tax implications of treasury stock transactions before two programs collide in the same quarter.
The finance director from that November review did the boring version of all this. The team pulled lot-level data across both custodians, found roughly $900,000 in unrealized losses, sold the positions where the original thesis had already failed, and reinvested in near-equivalents the same afternoon. The allocation barely moved. The tax provision moved a lot.
None of that required a clever structure or an aggressive reading of the code. It required a calendar, a lot-level spreadsheet, and somebody willing to look at the whole book instead of one account at a time.
Capital preservation usually gets described as a defensive posture, and mostly it is. But defense has an offensive edge, and this is where it shows up. The gains are already booked. What is left to decide is how much of them you hand over, and how much stays working for the company.