Developing Pricing Multiples Using the Direct Market Data Method

Uncertain How to Assess the Legitimacy of Price Multiples? Here Are Some Tips!

Rand Curtiss explains how to used reasoned judgment, statistical techniques, and comparative analysis to defensibly develop and support pricing multiples in an appraisal.

Many of us have participated in business conference sessions concerning the Direct Market Data Method (DMDM), which uses private guideline company transaction data from the IBA Market Database, BIZCOMPS, Pratt’s Stats, Done Deals, and other sources such as Securities and Exchange Commission (SEC) filings.  A subject that did not receive sufficient attention was how to determine pricing multiples for a subject company with such data.

“Assume that we have done a comprehensive search for transaction data, eliminated outliers,  made appropriate screening adjustments (such as for size) and compiled a valid and sufficiently  large sample(between 5 and 20 depending on the degree of precision required) for statistically confident measures of central tendency—the mean,  median, and standard deviation—and more for advanced measures such as percentiles).”

IBA’s valuation courses and Ray Miles’ DMDM monographs emphasize the importance of developing and substantiating a price multiple.  Nonetheless, it appears from review of numerous demonstration reports and consultations with IBA members that this point may not have been clearly communicated and/or there is uncertainty as to how to accomplish this.

The DMDM pricing multiple is, like the specific company equity risk premium in the Build-up  Method for developing a discount rate, a moving part that requires both appraisal skill and  judgment to develop.  IBA and other professional society standards require that our findings be replicable, supportable, and appropriate.  We cannot simply proclaim a multiple without proof.   We must develop and present a basis.  The phrase “in my professional opinion,” proffered with no evidence, is insufficient.  The blind application of average or median multiples without basis is likewise unacceptable.

So how can we develop and defend our multiples?  First, some general suggestions: 

1. It is helpful to first establish bounds for the range of possible multiples by inspecting the transaction data.
    
2. Then we select a likely multiple within the range.
    
3. Although the quantitative techniques discussed below suggest a formulaic approach, Revenue Ruling 59-60, Section 3, Paragraph 1 instructs us to exercise “common sense, informed judgment, and reasonableness.”  Case facts and circumstances usually demand this.  As an example, if we are valuing a single-location retail establishment, we might think that its dependence on its owner-manager merits a below average multiple.  However, it is probable that most of the sample transactions involved key owner/managers, so this risk factor was already reflected in those multiples.  On the other hand, the same business, with a favorable (cheap, long-term, and ironclad) lease, would probably be superior relative to the norm.  This would justify an above average multiple.
    
4. The benefit of quantitative techniques is not so much their (potentially misleading) precision, but their logical and comprehensive inclusion of relevant factors.  This builds confidence in our conclusions and explanations.
    
5. The price multiple reflects the relative desirability of the subject compared to the sample.  If it is more desirable, its multiple should be above average, and the converse is true.
    
6. Relative desirability is the Market Approach analogue to risk and return in the Income Approach. 

Now to the essence of our discussion. I have used three different methodologies to develop and support price multiples: 

1. Reasoned judgment: in some engagements, the business is so atypical that I have been unable to locate comparative industry financial data.  In such cases, I so state, and do the best I can with what I have.  I consult business brokers and colleagues to reinforce my judgment.
    
2. Comparative analysis: when such data are available, a thorough workup often leads to a clear conclusion as to the subject’s relative desirability based on its financial characteristics (profit margins, turnover ratios, liquidity, leverage, and rates of return).  I use this method in the large majority of my assignments. (Real estate appraisers use what they call “quality adjustment grids” to accomplish this.  Perhaps someday we business appraisers will develop one.)  I believe that this is the closest approximation to replicating the behavior of typical, reasonably informed, willing buyers and sellers among the three methodologies.  The Excel-based financial analysis template I use for this purpose is free for the asking at rc@businessval.com.
    
3. Statistical techniques: in cases where I have a very large sample and descriptive information, such as that found in Pratt’s Stats, I have used linear regression analysis to develop pricing models.  (I also used comparative analysis).  I am comfortable with this, but not with more advanced techniques in which I am not qualified.  I believe that this overshoots what typical willing buyers and sellers consider.  On the other hand, when these results are strong, they provide compelling evidence, particularly when charted or graphed.

To summarize, regardless of the methodology chosen, our objective is to use our sample data to explicitly, logically, and defensibly support our pricing multiples.

Rand M. Curtiss, MCBA, FIBA, ASA, CBA is the President of Loveman-Curtiss, a business appraisal firm located in Shaker Heights, Ohio. This article first appeared in the First Quarter, 2011, issue of Business Appraisal Practice (BAP).