What happens after an S corporation files a California refund claim?
Edvin Givargis Published 5 minute read
The short answer
Three things, and the well-run claim plans for all of them on day one. The claim gets examined: the Franchise Tax Board does not treat amended returns requesting money back as ministerial, refund claims of any size draw review and sizable ones draw full audits, with a power of attorney, scheduling letters, and document requests that read like any other examination except that the taxpayer is the moving party and carries the burden. The claim cascades: an S corporation's amended California return does not end at the entity's own 1.5 percent tax, because the re-computation that produces the entity refund also rewrites the shareholders' K-1s, and the shareholder-level money, usually the larger money, moves only if the shareholders file their own claims. And the clocks run separately: the entity's limitations period and each shareholder's limitations period expire on their own schedules, four years from the later of the original due date or the timely filed return, or one year from the overpayment, whichever is later, so a claim strategy that perfects the entity's position while a nonresident shareholder's April deadline lapses has thrown away the point of the exercise. The classic modern fact pattern makes the mechanics concrete: a California software company that sourced its licensing receipts to California for years discovers that market-based sourcing assigns those receipts to where its customers are, amends the open years, and watches its sales factor, its entity tax, and its nonresident shareholder's California-source income all fall together, at which point the file's success depends on sequencing, documentation, and stamina for the audit that follows.
The substance: re-sourcing receipts, and what it does to each shareholder
For taxable years under California's market-based regime, receipts from sales other than tangible personal property are assigned to California only to the extent the customer receives the benefit, or uses the intangible or licensed property, in California. For a software business selling nationally, the difference between the old habit, sourcing everything to the home state where the code was written, and the market rule can be most of the sales factor, and the amended returns implement the correction: receipts re-assigned by customer location, supported by the customer data available, with billing address serving as the practical proxy where actual use location cannot be established, a proxy the regulations tolerate as reasonable approximation but an auditor will test, which is why the customer-location workpapers are the heart of the file. The shareholder cascade then splits cleanly along residence lines. A nonresident shareholder is taxed by California only on California-source income, and the pass-through's apportionment is what sources it, so the amended K-1 that carries a collapsed California-source number is the nonresident's entire case; the shareholder files a personal refund claim on the strength of it, and the dollar movement at a top-bracket individual rate routinely dwarfs the entity's own 1.5 percent refund. A resident shareholder presents the asymmetry that surprises people: California taxes its residents on all income regardless of source, so re-sourcing the entity's receipts changes nothing about the resident's California base, no amended resident K-1 is needed for California purposes, and the resident's relief valve, if other states now tax more of the same income, is the other state tax credit rather than the sourcing itself. Stating that asymmetry early prevents the pointless amendment of returns that cannot change and focuses the deadline management where it belongs, on the nonresidents.
The procedure: deadlines first, then the audit the claim invites
The limitations analysis belongs at the top of the engagement memo, in a table, entity and each shareholder, because the periods interact but do not merge. Each taxpayer's claim must be filed within its own window, the shareholder claims cannot be perfected until the entity's amended K-1s exist, and the schedule therefore runs backward from the earliest shareholder deadline, not forward from when the entity work happens to finish. Where a deadline is close, a protective claim, filed on the available numbers and perfected later, preserves the year; discovering the shareholder deadline after polishing the entity file is the unforced error this area exists to prevent. Then comes the examination. A refund claim audit inverts the usual posture: the taxpayer asserted the facts, so the taxpayer proves them, and for a sourcing claim that means producing the customer-location analysis, the revenue detail behind it, and the tie-out from the workpapers to the amended returns, years after the returns were filed, often after the people who built them have moved on. The file should be assembled for that future on the day the claim is filed: the sourcing workpapers, the data extracts, the methodology memo explaining the proxy used and why it is reasonable, and the deliverable set, archived where the eventual responder can find them, because reconstructing a claim's support from scattered drives while the auditor waits is how strong claims develop weak records. Expect the audit to test both direction and degree, whether market sourcing applies and whether the customer-location data supports the split claimed, and expect the state to look for offsets in the same years, since a claim opens its year to adjustment against the refund even where the state could no longer assess affirmatively. Interest runs on what is ultimately refunded, at the state's credit rate, which rewards patience but never substitutes for it.
Practice notes
The engagement checklist: limitations table for entity and every shareholder before substantive work begins, with protective claims for any window inside the working horizon; residency sort of the shareholder roster, nonresidents drive deadlines and amendments, residents get the other-state-credit conversation instead; customer-location workpapers built to audit standard at filing, with the methodology memo attached to the file rather than implied by it; and the archive discipline, deliverables, data, and tie-outs stored against the examination that should be assumed. Under audit, the burden posture shapes the tone: produce affirmatively, organize the support around the claim's own logic, and treat the state's offset review as expected rather than aggressive. And counsel the prospective fix alongside the retrospective one, because a company that needed to amend three years of sourcing needs its current-year apportionment built on the same customer data going forward, or the next engagement is this one again with fewer open years to save.
This article states the law as of September 16, 2026
Statutes, rates, thresholds, and agency practice change, and a different set of facts can change the answer. Before acting on anything discussed above, contact G&G State Tax Group to confirm what has changed since this was written and how the rules apply to a specific situation.
G&G State Tax Group, LLC is a state and local tax advisory firm. The firm provides state and local tax consulting and representation in state and local tax controversies. G&G does not prepare or file tax returns, perform attest services, or provide bookkeeping, and is not a CPA firm.