Six assignments, and what actually turned on them

Four separate reports would have used four effective dates, four sets of assumptions, and four definitions of value — and the estate needed all of it as of one date of death, two years earlier. Reconciling them would have fallen to the personal representative, who is not an appraiser.

One engagement, one effective date, one set of assumptions, covering all four asset classes. The retrospective date meant reconstructing four different markets as they stood two years prior — a research problem, not merely an older one. The machine tools required a physical inventory; the firearms required individual identification and condition grading; the business interest required five years of financials and the operating agreement.

The estate filed with internally consistent values and a single appraiser available to answer questions. Where an estate is considering the six-month alternate valuation date, that consistency matters even more — the election is all-or-nothing across the entire gross estate, so the whole schedule has to be built on the same footing.

Arizona has no fixed valuation date for community property. It is neither automatically the date of service nor the date of trial — the superior court has wide discretion and must choose a date that produces an equitable result. Valuing to one date would have handed the other side an easy line of attack.

The date question was raised with counsel before any work began, in writing. Because it was genuinely contested, the engagement developed the conclusion to both candidate dates in a single assignment, with the difference between them explained and attributed to identifiable causes rather than left as a gap.

Counsel could argue the legal question of date selection without re-hiring an expert, and the report could not be attacked for having quietly picked a side of a disputed issue. Two dates in one engagement also cost materially less than two engagements.

The thresholds are not tested per gift. The regulations require the donor to aggregate the deduction claimed for all similar items of property contributed during the tax year — and that aggregation crosses donee organizations. Grouped, the gifts exceeded $5,000, which meant Form 8283 Section B and a qualified appraisal, and Section B had to be completed for each institution.

A single qualified appraisal covered the group, providing the required information for each item rather than a group total, and signed within the 60-day window the regulations allow. The report carried the education and experience statement and the appraiser declaration the regulations require, which generic valuations routinely omit.

The deduction was substantiated in a form that meets the regulatory definition. Had it been filed without the appraisal, the deduction was exposed to disallowance on substantiation grounds alone — entirely separately from whether the values were right.

Two errors, in opposite directions, both material. The denominator included land — which the rule excludes, since only the market value of the structure counts — making the ratio look far smaller than it was. And the numerator omitted contractor overhead and profit, sales tax on materials, and the market value of materials the owner intended to supply and labour the owner intended to donate. All three must be counted.

The structure was valued on the basis the floodplain administrator would accept, with the valuation date stated as before the start of construction and the basis stated explicitly — the part a general market appraisal usually gets wrong. The cost side was rebuilt against FEMA's own inclusion list.

Corrected, the project sat just over the threshold. The owner learned that before the permit rather than mid-build, and could choose between reducing scope and designing for full compliance. The determination itself belongs to the floodplain administrator — the appraisal supplied a credible input, which is what the burden on the applicant actually requires.

The number was not obviously wrong, which is what made it dangerous. The defects were structural: the intended use was generic, the value definition was never stated, the scope of work did not disclose that no unit had been inspected, and the comparables were unsourced.

A rebuttal analysis addressing the report on its own terms, section by section, plus an independent valuation developed to the correct definition of value with sourced market evidence and a disclosed scope of work. Not an attack on the conclusion — an examination of whether the conclusion was supported.

The cross-examination wrote itself from the certification and the scope of work. The lesson generalises: reports usually fail on the parts nobody reads, not on the number everybody argues about.

Actual cash value is measured as of the date of loss, and under the replacement-cost-less-depreciation method it does include a deduction for depreciation. But the Arizona Supreme Court held in 2022 that an insurer may not depreciate labour costs. On a contents schedule of any size, that single distinction moves real money.

The schedule was valued to the date of loss, with the depreciation applied to materials and the labour component identified separately — so the two could be seen rather than blended into a single number nobody could interrogate. The policy's own loss settlement provision governed the basis, and the report cited it.

The dispute became arithmetic instead of argument. Where a policy pays actual cash value first and releases the depreciation holdback on completion of repair or replacement, having the components separated also tells the insured what they are entitled to recover and when.

The report on file concluded to fair market value in continued use. The bank was underwriting against orderly liquidation value, because that is what it would face if it ever had to act. Neither party had misread the report; the seller had simply commissioned the premise that answered his question, and the bank needed the one that answered theirs. Nobody had noticed, because the word ‘value’ appears in both.

A new engagement was scoped to produce both premises in one report, with the definition of each quoted in full and attributed, and a plain statement of why they differ. The equipment schedule itself barely changed — the assets were the same assets. What changed was that the report said what it was measuring, twice, and showed the gap rather than leaving each reader to assume it away.

The premise, not the equipment. An identical machine list, inspected the same way on the same day, supports materially different conclusions depending on who is assumed to be buying and how long they have. A report that states only one premise is not wrong, but it can only answer one of the questions in the room.

The tooling package had never been itemised. It included workholding, fixtures, tool holders and a probe, and the owner's understanding was that the fixtures were his personally, having been built in-house over years. The buyer's understanding was that a walk-through of a running shop showed him what he was buying. The lender's filing described the machine.

The schedule was rebuilt to list the tooling item by item, and the machine was valued twice: once including the listed package, once excluding it. Nothing was decided about ownership — that was a question for the parties and their counsel, and the report said so. What the report did was make the size of the disagreement visible before it became a dispute about money rather than about a list.

An unlisted inclusion is functionally an exclusion, because after completion nobody can prove what was in it. ‘With tooling’ is not a description of an asset. It is a description of a bundle whose contents nobody wrote down, and it produces this argument reliably.

They were not equal. One unit sat on a platform the manufacturer still supported, with a service agreement written so that it survived a transfer of the equipment. The other was on a platform whose support had ended, with a service arrangement that terminated on any change of ownership. Nothing about the physical inspection distinguished them, and nothing in the depreciation schedule did either.

Platform support status and contract transferability were established in writing before either unit was valued, rather than assumed from the model designation. The report valued them separately, stated the basis for the difference, and disclosed the assumption it made where written confirmation could not be obtained.

Serviceability, not the machine. Model year is a proxy for the things that actually drive medical equipment value — parts, qualified service, and whether coverage follows the asset — and it is a poor one. Two units can be identical on paper and materially different as assets.

The low reading was accurate and meaningless. The meter had been replaced after a failure and had restarted from zero; the machine's actual service life was in line with the rest of the fleet, and its undercarriage showed it. Nothing in the schedule was false. The schedule simply recorded what the meter said.

The meter replacement was identified and disclosed rather than averaged away, and the machine's condition was assessed on the wear components directly, with the undercarriage treated as its own analysis. The report stated plainly which reading it had relied on for each unit and where a reading had been disregarded and why.

The disclosure. A hour meter measures elapsed running time, not work done, and it can be replaced. A report that leans on hours without saying so, and without saying what it did when the hours were not credible, hands a lender a number it cannot test.

The agreement did not say which value. It named no standard, said nothing about whether discounts for lack of control or marketability applied, and fixed no effective date beyond a reference to the appraisal itself. Each side had by then commissioned a report, and the two reports had made different assumptions on all three points. The gap between them was attributable almost entirely to those assumptions rather than to any disagreement about the business.

The construction of the agreement was identified as a question for counsel rather than for an appraiser, and said so in writing at the outset. The engagement then developed the interest on each competing reading, clearly labelled, so that the decision-makers could see precisely what turned on the construction rather than arguing about a single number whose assumptions were buried.

The drafting. A buy-sell agreement that fixes a mechanism but not a standard of value has deferred the hardest question to the moment when the parties are least able to agree on it. The appraiser cannot resolve it — doing so would be construing the contract.

Most of the client base had arrived through one practitioner personally, and much of the work was not delegated. The practice had never previously lost a principal, so there was no internal evidence of what happened to clients when one left. The two reports in the matter differed less on the total than on the split between goodwill attaching to the business and goodwill attaching to the individual.

The allocation was built from evidence rather than from a proportion that seemed reasonable: how clients originated, what proportion of work was delegable, what the marketing was actually built around, and what practices of that type demonstrably did when they changed hands. Where evidence was thin the report said so. The legal treatment of the personal component was expressly left to counsel and the court, on the footing that Arizona has not settled it.

The evidence behind the split, and the candour about what remains unsettled. An allocation asserted as a percentage is the easiest thing in a valuation report to attack, because there is nothing underneath it to test.