A valuation is often viewed as a number crunching exercise with readily available inputs and assumptions available, but it typically involves many subjective assessments, choices and assumptions that are prone to bias in a valuation. That is often driven by the underlying purpose for the valuation and if not managed properly, can give a result that may be limited in its usefulness.

Common Sources of Bias

Forecasting: a forecast or projection of future cash flows from a business is a key input to a valuation model based on future cash flows. In preparing as a forecast, there are many sources of potential bias. For example, there can be too much reliance on personal experience, intuition instead of independent information and data in estimating revenue growth rates and profitability metrics. Even if objective information is utilized, confirmation bias can result in more weight in the analysis being given to information that confirms existing optimistic beliefs that may be optimistic or pessimistic. When estimating the profitability of a business, the historical performance is often given significant weight, but those historical results are often subject to adjustments intended to normalize the results which can be selectively included or excluded.

Valuation Inputs: beyond the forecast assumptions, there are various other inputs assumed in a valuation model related to working capital and capital expenditure requirements, identification of redundant assets, discount rates and terminal value adjustments. For example, discount rates should reflect the risk of achieving the future cash flows forecasted but there are several choices among alternatives in building the discount rates that are subject to bias. Any one of these inputs, if misapplied or selected without any objective basis, can result in significant variations in valuation conclusions.

Valuation Multiples: market participants may rely on a relative valuation or market approach as the primary valuation approach or as a secondary approach. Obtaining relevant data from truly comparable companies that are publicly traded can be difficult and there may be a temptation to use companies that are not comparable due to size, product mix, end markets, etc. In selecting valuation multiples from open market transactions, transactions may be selected that are too old or not relevant for many of the same reasons related to publicly traded companies. Also, certain valuation multiples from comparable companies can included or omitted to achieve the objectives of the valuation and minimize those that conflict with the objectives.

Application of Discounts or Premiums: the use of discounts such as illiquidity and minority discounts or a premium for control are more typical in private company valuations for shareholder disputes, income taxes and other disputes. Since there is limited objective information on discounts and premiums, a valuation conclusion can be subjectively decreased or increased by using subjective adjustments for various situations.

Responses to Bias

Corroborate Forecast Inputs: when estimating growth rates in revenues, factors like industry growth rates and market share should be considered. Independent information that conflicts should not be dismissed but rather used to stress test the forecasts. For example, a business may be expected to grow faster than industry average during the short to medium term but over the long-term, businesses tend to revert to the average in the longer term. To address bias in normalized financial results or where there is a limited history of operations, to the extent possible, the historical profit margins as a percentage of revenues should be corroborated with independent industry evidence for reasonability.

Corroborate Valuation Inputs: to the extent possible, other valuation inputs that have a material impact on the valuation should be based on objective verifiable information. This includes historical information related to inputs such as working capital and capital expenditure requirements and market-based information related to calculation and selection of discount rates. While historical data specific to the company is usually strong evidence for inputs such as working capital and capital expenditures, industry data should also be utilized where there is limited historical evidence or data available in an early stage business.

Cross Check the Results: where possible, a secondary valuation approach should be used to ensure the valuation conclusions from the primary valuation (typically a cash flow based method) approach are reasonable and consistent lending further support to the inputs and assumptions used in the primary valuation approach. This typically involves comparing the valuation multiples of a businesses with those of other comparable companies or to prior transactions in the shares of the subject company and if properly carried out, such an analysis can help stress-test the primary valuation method.

Valuation Range: any value that is obtained for a business is first and foremost an estimate and as accordingly should be quantified as a range of estimates to accommodate the inherent margin for error. This can be based on application of multiple scenarios and or presentation of a best-case (high) and worst-case (low) estimates of value. The output that is presented should reflect the estimates of value and the inherent uncertainty of those values.

Conclusion

Bias in valuations cannot be eliminated as there will always be inherent estimation uncertainty from the forward-looking nature and many assumptions used. However, building better valuation models that effectively use available objective and independent information is an effective way of addressing the bias and the uncertainty arising from macroeconomic, industry and company specific conditions.

Contributed by Michael Frost, CPA, CA, CBV, from Mowbrey Gil. This piece was produced as a part of the quarterly Canadian Overview, a newsletter produced by the Canadian member firms of Moore North America.