A portfolio report says it harvested $20,000 in losses. That is a transaction result, not a $20,000 improvement in your finances. The next questions matter more: How much can you use? At what rate? And what happens to the replacement investment?
Start with a deliberately simple example. A taxable investment bought for $100,000 falls to $80,000. You sell it and invest the proceeds in a suitable replacement that is not substantially identical. Assume the entire $20,000 loss offsets capital gains that would otherwise be taxed at 20%. The current tax reduction is $4,000. These are invented assumptions, not a client result or a quoted tax benefit.[S1][S4]
The tax bill may have changed dates, not disappeared.
The replacement starts with an $80,000 purchase basis. Suppose it later sells for $120,000. Its gain is $40,000. In a simplified comparison in which the original investment also reaches $120,000, keeping the original $100,000 basis would have produced a $20,000 gain. At a 20% tax rate, harvesting produces $4,000 more tax at that later sale. That matches the initial $4,000 reduction in nominal dollars, before costs. The identical future values isolate the tax arithmetic; real replacement holdings will not necessarily perform the same way.[S4]
Timing still matters. Paying a dollar later is different from paying it today. If the extra $4,000 tax occurs ten years later, a 5% annual discount assumption puts its present value at about $2,456. Subtract that amount and an invented $500 of additional costs, measured in today's dollars, from the initial $4,000 reduction. The tax-only present-value difference is about $1,044. The discount rate is a valuation assumption, not an expected investment return.
Change the future tax rate, the sale date or the costs, and the answer changes. It can become negative. A full decision also requires a comparison of the investments themselves, not just their tax bills.
A usable loss and a carried-forward loss are different.
The example assumes the whole loss is usable immediately against capital gains. Actual capital-loss netting comes first. For individuals with an excess net capital loss, the annual deduction against other income is generally limited to $3,000, or $1,500 when married filing separately; unused losses can carry forward. A large carryforward is not the same thing as an equally large current tax reduction. The interactive example does not model carryforwards.[S1]
Daily is a monitoring schedule, not a trading target.
The standard should be straightforward: check for opportunities without making the number of trades a measure of success. A useful tax-aware review asks whether a loss is usable, whether the replacement fits, whether the household can avoid a wash sale, and whether the economics justify the costs. Sometimes the most defensible decision is to leave the portfolio alone.[P2][S2]
Direct ownership of individual stocks creates more control over which holdings to sell than ownership of a single fund share. That flexibility comes with work: portfolio restrictions and replacement choices can create benchmark deviations, and additional fees can offset benefits. Direct indexing is an implementation choice, not a guarantee of a better result.[S5]
Academic work by Chaudhuri, Burnham and Lo illustrates why assumptions deserve attention: their historical model reported a lower tax-alpha estimate after adding wash-sale constraints. We reviewed the published abstract, not an independent replication. Its estimate is not a forecast for your account, and no research return is used in our illustration.[S8]
The better question: After the full tax cycle and the implementation costs, did this decision improve the plan?