Written by Chris Walton, JDMost valuation disputes end with the court choosing one expert over the other. Sometimes the court builds its own model from the experts’ competing inputs. In Gillum v. Gillum (N.C. Ct. App. June 18, 2024), the court did something worse for both sides: it rejected both experts entirely and valued the business from a number it found in the company’s loan documents.
The case involved the valuation of a closely held construction company in a North Carolina divorce. One expert was rejected for inadequate normalization of the financial statements. The other was rejected for inconsistencies in the income approach. The trial court looked at the record, found a lender’s valuation in the company’s loan files, and adopted it. The appellate court affirmed, holding that courts are not required to choose the “least flawed” expert — when both valuations lack credibility, the court may look to any competent evidence in the record.
For family law attorneys and valuation professionals, Gillum is a warning about what happens when you lose control of the valuation outcome. A lender’s valuation — prepared for underwriting purposes, with different assumptions, a different standard of value, and a different purpose than a marital dissolution — became the basis for dividing the business. Neither side briefed it. Neither side’s expert tested it. And neither side could challenge the assumptions behind it, because it wasn’t their evidence.
Normalization is the foundation of any income-based valuation of a closely held business. The financial statements of a private company are rarely a clean reflection of the business’s economic earnings. Owner compensation may be above or below market rate. The company may run personal expenses through the business. Related-party transactions may be at non-arm’s-length pricing. Depreciation may be accelerated for tax purposes. The normalization process adjusts for all of these so that the income stream the appraiser values reflects what the business actually earns on an economic basis.
The first expert in Gillum failed to adequately normalize the construction company’s financials. For a construction business, normalization is particularly important and particularly challenging. Revenue can swing dramatically year to year depending on the contract backlog. Owner compensation in construction often includes vehicle allowances, equipment use, insurance, and other benefits that don’t show up on the W-2. Project-based revenue recognition can create timing distortions that make any single year’s income unrepresentative. If the expert didn’t adjust for these factors, the income stream being valued didn’t reflect the business’s true economic earnings — and the court recognized it.
For practitioners, the lesson is specific: in a closely held construction company, normalization isn’t a rote exercise. It requires understanding how the owner uses the business, how revenue recognition timing affects the income statement, and what the owner’s total compensation package looks like when all benefits are included. An expert who applies a generic normalization template to a construction company’s financials will miss the adjustments that matter most.
The second expert was rejected for a different reason: internal inconsistencies in the income approach. The court didn’t find the methodology wrong in concept — it found the execution unreliable. The specific inconsistencies aren’t detailed in the appellate decision, but the pattern is familiar to anyone who has reviewed valuation reports in litigation: assumptions in one part of the model that contradict assumptions in another part.
The most common income-approach inconsistencies in closely held business valuations include projection assumptions that don’t align with historical performance (projecting growth when the business has been flat, or vice versa) without adequate explanation, a discount rate that doesn’t reflect the risk embedded in the projections (optimistic projections paired with a low discount rate double-counts the upside), terminal value assumptions that imply a long-term growth rate inconsistent with the projection period, and normalization adjustments in the income statement that aren’t reflected in the balance sheet or capital expenditure assumptions.
Any one of these might survive cross-examination if the expert can explain the reasoning. But when the court identifies multiple inconsistencies, the cumulative effect destroys the model’s credibility. The court’s conclusion isn’t that the income approach is wrong — it’s that this expert’s application of it is unreliable. And once the court reaches that conclusion, the expert’s entire opinion is at risk.
With both experts rejected, the trial court turned to the record and found a valuation in the company’s loan documents. Lenders routinely require business valuations or financial analyses as part of the underwriting process for commercial loans, particularly for asset-heavy businesses like construction companies. These valuations are prepared for credit purposes — to determine whether the business can service the debt — and may use a different standard of value, different assumptions, and different methodologies than a marital dissolution valuation would.
The trial court adopted this lender valuation as the basis for the equitable distribution. The appellate court affirmed, holding that the trial court was not required to choose the “least flawed” expert when both lacked credibility. The court had discretion to look to other competent evidence in the record — and the loan documents qualified.
This is the holding that should concern both sides. A lender’s valuation isn’t designed for equitable distribution. It may focus on liquidation value rather than going-concern value (lenders want to know what they’d recover in default, not what the business is worth as a going concern). It may not adjust for personal goodwill, minority discounts, or other factors that matter in a divorce context. It may be stale — prepared at the time of the loan, not at the valuation date the court is using. And because neither side presented it as their evidence, neither side had the opportunity to cross-examine the assumptions behind it.
The practical implication: if your expert’s report has credibility problems, the fallback isn’t a less-perfect version of your own analysis. The fallback is whatever the court finds in the record. And that might be a number that neither side would have chosen.
Gillum represents a fourth judicial response to competing expert testimony that both sides find wanting:
Choose the more credible expert. This is the most common outcome. In Koch v. Koch, the Minnesota court found one expert’s income/market approach more credible than the other’s asset approach and adopted values close to the preferred expert’s conclusions. The court picks a side.
Build a hybrid from both experts’ inputs. In Ramcell v. Alltel, the Delaware Chancery Court blended the two experts’ models — weighting projections 70/30, averaging discount rates, and adjusting the terminal growth rate — rather than choosing one. The court constructs a third model from the pieces it finds most defensible.
Reject both and build from scratch. In Blue Blade v. Norcraft, Vice Chancellor Slights rejected both experts’ DCFs and built his own, cherry-picking inputs from each side. The court creates an independent valuation.
Reject both and look to non-expert evidence. That’s Gillum. The court doesn’t blend the models or build its own. It steps outside the expert testimony entirely and adopts a value from the documentary record. This is the outcome the parties have the least control over, and it’s the outcome that arises when both experts have lost credibility.
The common thread: courts will not be held hostage by flawed expert testimony. They have discretion to find value from whatever competent evidence is available. The expert’s report is the preferred source — but only if it’s credible.
If you’re retaining a valuation expert for a closely held business in a divorce — particularly a construction company or other business with cyclical revenue, heavy owner involvement, and complex financials — Gillum provides a pre-trial preparation checklist:
Is the normalization thorough and documented? Every adjustment should be identified, quantified, and explained. Owner compensation, personal expenses, related-party transactions, non-recurring items, and revenue recognition timing adjustments should all be addressed. If opposing counsel can point to an obvious normalization gap, the expert’s entire income analysis is vulnerable.
Is the income approach internally consistent? Projections should align with historical performance or explain the divergence. The discount rate should reflect the risk in the projections, not an independent estimate disconnected from the forecast. Terminal value growth should be consistent with the projection period’s trajectory. Walk through the model and ask: does every assumption tell the same story?
Has the expert addressed the specific challenges of the industry? Construction companies have backlog-driven revenue, project-based margins, equipment-heavy balance sheets, and workforce dependencies. An expert who values a construction company the same way they’d value a SaaS business hasn’t engaged with the industry’s specific valuation challenges.
What’s in the record if the expert is rejected? Before trial, review the documentary record for alternative valuation evidence: loan documents, insurance policies, buy-sell agreements, tax returns with balance sheets, prior appraisals. If your expert is rejected, these become the court’s source material. Know what they say and whether they help or hurt your position.
Not every closely held business valuation in a divorce requires both sides to retain competing experts. If the business is small, the marital estate is modest, and the cost of two full valuation engagements is disproportionate to the value in dispute, the parties may be better served by a single joint expert whose work both sides can cross-examine. A joint expert eliminates the “dueling advocacy” dynamic that courts increasingly criticize — and that produced the mutual rejection in Gillum.
Alternatively, if the business has a recent arm’s-length appraisal — for a loan, a buy-sell trigger, or insurance purposes — that appraisal may provide a credible starting point that both sides can adjust rather than replace. The irony of Gillum is that the loan-document valuation the court adopted might have been a more efficient starting point than the two experts whose work it replaced.
Gillum v. Gillum establishes a principle that should change how family law attorneys think about expert preparation: the court is not required to choose the “least flawed” expert. When both valuations lack credibility, the court may reject both and adopt alternative evidence from the record — including a lender’s valuation that was never intended for equitable distribution, was never briefed by either side, and may use assumptions that neither party would endorse. The way to prevent that outcome is to prevent the predicate: make sure your expert’s normalization is thorough, the income approach is internally consistent, and the industry-specific challenges are addressed. The expert who can survive cross-examination controls the valuation. The expert who can’t hands the outcome to the loan file.
If you’re preparing a valuation of a closely held business for a divorce proceeding and want a second set of eyes on the normalization and income-approach consistency before it goes to trial, happy to review it. The credibility issues that sink experts in cases like Gillum are easier to fix in draft than to explain on redirect.
Schedule a free consultation meeting to discuss your valuation needs.
