12 warning signs of unreliable forecasts from Tarbell and Trugman

“It’s good to be talking about business forecasts with a lot of CICBV members in the room,” began Gary Trugman at his keynote session on working with financial projections in Miami at the ASA/CICBV Annual Business Valuation Conference.  “You’re all already familiar with hockey sticks.”

Trugman and his co-presenter Jeff Tarbell note that USPAP doesn’t address forecasts directly in Sections 9 or 10.   SSVS-1 refers to projections in sections on collecting data, and in DCF analyses.  But, there’s nothing about the degree to which appraisers need to audit forecasts—which is part of the reason that a standard limiting condition in many valuation reports is something like the language below.

We do not provide assurance on the achievability of the results forecasted by [ABC Company] because events and circumstances frequently do not occur as expected; differences between actual and expected results may be material; and achievement of the forecasted results is dependent on actions, plans, and assumptions of management.

Correctly or incorrectly, some appraisers may try to account for forecasts they don’t trust via the company-specific risk factor.  “The courts are catching up with anything that looks like this practice,” Tarbell said.

How good are forecasts?  First, “30-40% of the companies we work with don’t have a meaningful forecast,” said Trugman.  “But, even when you can obtain one, you often face unreliable assumptions, and they may be unwilling to make changes you suggest.”  So, when do you just do your own forecast?

Particularly with public companies “you’re going to have a hard time justifying numbers you made up as opposed to numbers management made up,” says Tarbell.  “You could be asking management to revise numbers that may have already been presented to analysts or others.   You’re opening a lot of doors no one wants to open.”  But, if you do your own, the best hope is “to get management to sign off on what you’ve done.   Maybe they don’t have a balance sheet forecast so we fill in the blanks, send it to management, and ask them to agree to it with all the disclaimers that you can’t predict the future.”

Trugman and Tarbell feel that if you can’t obtain a forecast, or adjust a weak forecast, or create your own, your option is to reject the income approach.   And, they warn all appraisers to recognize two rules about nearly all forecasts:

1.  Management tends to over-estimate projected cash flows:

– The distribution of future cash flows is not likely symmetric.

– Downside often exceeds upside due to capacity constraints, market size, competition, etc.

2.  The appropriate projected cash flow for discounting is the statistically expected value, meaning a probability-weighted expectation of future results.

– This is not the particular outcome with the highest probability of occurrence.

Pratt and Grabowski’s Cost of Capital 4^th Edition is due out very soon.  Tarbell reports that these two points are freshly emphasized in the update.  And, Tarbell and Trugman recognize that there are a dozen Indications of possible unreliability:

1. Forecast results are notably different than past results.—“It’s OK to be wrong in a forecast; all the public companies are.  But by looking at past forecasts you can see patterns of unreliability,” says Trugman.
2. Forecast was prepared by a party with an interest in the valuation outcome.
3. Resulting value is not consistent with the values from other methods used.
4. Forecast was prepared by CEO/CFO without input from business unit heads. “If the CEO hasn’t spoken to sales and marketing, you may see very different results,” says Tarbell.
5. Forecast is inconsistent with analyst expectations for public comps. “Are growth rates and margins consistent with what analysts are projecting for public companies in your industry,” asks Tarbell.  “There better be a good explanation if a forecast is different that other companies who are all competing for the same market share.”
6. Forecast income statement without balance sheet and statement of cash flows. “It may not be very safe when you’re missing such critical inputs to the DCF method such as working capital, CAPEX/depreciation, or financing needs,” warns Trugman.  He sees many clients who project faster than historical growth, but in fact when balance sheet forecasts are prepared, it turns out that you outstrip cash resources very quickly and there’s no way to support this growth.  “The company may have exceeded its borrowing capacity early in the projection cycle,” Trugman explains in the most typical case.
7. Forecast assumes capital spending at levels that are not financeable.  “We often see a forecast that assumes a doubling of some factor in the middle of the cycle,” says Tarbell.   This is a clear warning sign.  “A leveraged analysis may be more appropriate where the subject company has significant capital needs over the course of the forecast,” Trugman agrees.
8. Forecast ends on the peak or trough of a business cycle.  “What do we do with forecasts now, for instance,” asks Trugman.  “The answer is we’re looking at longer business cycles, even beyond the standard five year projections.   How did the business do during the last downturn?”
9. Forecast not accompanied by a detailed schedule of assumptions.  Tarbell says “even if there aren’t detailed notes, can management explain the significant assumptions and particularly any of those that are inconsistent with the past.   I don’t like those.”
10. Forecast not achievable without additional financing or acquisition.
11. Forecast hinges on one or two extraordinary assumptions.  “If good results are tied to the outcome of one key assumption, you’ll need to examine that assumption very carefully, and be very suspicious of the projections,” Trugman warned.
12. Too much of the indicated value is coming from the terminal value.

Trugman and Tarbell’s indicators of forecast reliability appear in the table below

Determining Forecast Reliability

• Revenues

– Are growth rates consistent with history?

– What are new revenue streams based on?

– What is the timing of new revenue streams?

– Are changes in revenues consistent with industry information?

• Expenses

– Forecast should be based on normalized operations

– Fixed vs. variable cost analysis

– What do variable costs vary against? Revenues? Payroll? Square footage?

– Is this consistent with history?

– What is basis for research and development costs?