Does the S-Corp Premium Really Exist?

The underlying argument is whether an entity that does not pay taxes is inherently more valuable due solely to this tax status. All things being equal, if the owner of a pass-through entity receives cash flow that the owner of a C-Corporation would not, it stands to reason that more value should be attributed to the pass-through entity. Conversely, should additional value be awarded to an entity that made a tax-based election that many other companies could easily make?

Furthermore, in the valuation universe where the fair market value standard relies on the assumption of a hypothetical buyer, can it be assumed the buyer would be able to maintain the pass-through status? In some cases, studies have shown that owners of pass-through entities have been taxed more heavily than standard C-Corporations based on effective rates (“Entity Choice and Effective Tax Rates,” Quantria Strategies, LLC). There have also been studies such as the Denis & Sarin S-Corporation study and Van Vleet S-Corporation Economic Adjustment Model study that argue S-Corporations indeed have more value and they attempt to quantify the value through various formulas considering the tax implications of the entity types.

I am not purporting to know the answer to this question, but given the confusion, I have performed an independent analysis of transactional support for the S-Corporation premium in order to see if the trends in completed transactions coincide with any of these theories. In some ways, this analysis is similar to approaches used in other studies since 2002 as it relies on actual transactional data. However, I believe that the eliminations of transactions made in this analysis minimize any distortion of results by removing some data points that are inconsistently reported.

My hypothesis states if pass-through entities are more valuable than standard C-Corporations, than it stands to reason that this would be evidenced in completed transactions. Essentially, the sales of pass-through entities should produce multiples higher than those of comparable C-Corporations. For this analysis, I relied upon transactional data contained with the Business Valuation Resources, LLC Pratt’s Stats database. Although the breadth of data necessary to conclusively determine the existence of an S-Corporation Premium is not available, this analysis should provide practitioners with a reason to consider if a premium is justifiable rather than hastily applying it simply due to the corporate tax structure.

Within my analysis, I considered all of the results with the perspective of “fair market value.” As a result, this assumes a hypothetical buyer and seller. Within the data, there will be sales of S Corporations where their “S” status was voided due to the structure of the purchaser. Although there are many other layers of adjustments that could be made, I largely ignored the buyer type adjustment as data was not available within Pratt’s Stats beyond public or private buyer status and due to practitioners generally ignoring this factor.

For my analysis, I relied on four different approaches, however, the master data set followed these procedures:

1. All of the transactions that occurred in the United States were exported into an Excel workbook. This included data from 1995-2015 and was run on December 1, 2015.
2. Asset Sales were removed; leaving only Stock Sale transactions.
3. Data was sorted by Company Type and grouped into the categories of C-Corporations (C-Corporations and PC) and Pass-through Entities (Limited Company, Limited Partnership, LLC, LLP, LTD, Partnership, and S-Corporation)

Once the master data set was compiled, four different approaches were performed. Within each approach, the median and average metrics for MVIC/Sales, MVIC/Gross Profit, MVIC/EBITDA, and MVIC/EBIT were compared between the C-Corporation and Pass-through Entity categories. The first and third approaches consider the data from a high level broken down only by industry code, however, the third approach removes variables related to earn-outs, non-competes and employment agreements that may skew the data. The second and fourth approaches consider the element of time and is designed to test if any economic or other stimuli influenced the value of the S-Corporations. The fourth approach excluded additional variables similar to the second approach.

 

Approach 1

Assumptions and Analysis

The transaction detail was sorted by SIC code and broken down into groupings by the 1,000’s (0-0999, 1000-1999, etc.). This sorting was performed to minimize the impact of a few highly volatile industries skewing the comparison. Although in later approaches certain exclusions were made, all data remaining after the initial procedures was included in this analysis. An average and median for each grouping was developed for both the C-Corporations and pass-through entities and then compared to see which entity type produced a higher multiple. The comparison resulted in three possible outcomes: a higher C-Corporation multiple, a higher S-Corporation multiple, or a similar multiple. A multiple was deemed similar if the variance between the two multiples were within 5% of the C-Corporation multiple.

Approach Results

Based on this approach, it appears that in 85% of the comparisons, the C-Corporation entities produced a higher multiple than the pass-through entities. To verify that these results were not skewed by the recession or other time-related factors, Approach 2 was performed to analyze the data on a year-by-year basis.

 

Approach 2

Assumptions and Analysis

Similar to Approach 1, the transaction detail was sorted by SIC code and broken down into groupings by the 1,000’s (0-0999, 1000-1999, etc.). This data was further sorted by transaction year with groupings from January 1 through December 31 of each year (1995-2015). All data remaining after the initial procedures was included in this analysis. An average and median for each grouping was developed for both the C-Corporations and pass-through entities and then compared to see which entity type produced a higher multiple. The comparison resulted in three possible outcomes: a higher C-Corporation multiple, a higher S-Corporation multiple, or a similar multiple. A multiple was deemed similar if the variance between the two multiples were within 5% of the C-Corporation multiple.

Approach Results

The year-by-year analysis results proved interesting as the percentage of C-Corporation transactions with higher multiples dropped from 85% to approximately 62%. After a more thorough review, it appears that this is the result of smaller subsets of transactions being compared rather than any time-related factors. Between 1995-2005 and 2005-2015, the first period had 7 more transactions with higher S-Corporation multiples, however, the 2015 information available wasn’t a full year. Overall, this approach similarly resulted in C-Corporations having higher multiples.

 

Approach 3

Assumptions and Analysis

Transactions including earn-outs, non-competes, and employment agreements were removed. The remaining detail was sorted by SIC code and broken down into groupings by the 1,000’s (0-0999, 1000-1999, etc.). These transactions were excluded as an exact depiction of the impact these items had on the overall deal value which were not consistently reported. In an effort to minimize data that isn’t comparable, these transactions were excluded for this approach. An average and median for each grouping was developed for both the C-Corporations and pass-through entities and then compared to see which entity type produced a higher multiple. The comparison resulted in three possible outcomes: a higher C-Corporation multiple, a higher S-Corporation multiple, or a similar multiple. A multiple was deemed similar if the variance between the two multiples were within 5% of the C-Corporation multiple.

Approach Results

In this approach, the C-Corporation entities still produced a higher multiple than the pass-through entities as they represented the higher multiple approximately 78% of the time. To verify that these results were not skewed by the recession or other time-related factors, Approach 4 was performed to analyze the data on a year-by-year basis.

 

Approach 4

Assumptions and Analysis

Similar to Approach 3, transactions including earn-outs, non-competes, and employment agreements were removed. The remaining detail was sorted by SIC code and broken down into groupings by the 1,000’s (0-0999, 1000-1999, etc.). This data was further sorted by transaction year with groupings from January 1 through December 31 of each year (1995-2015). An average and median for each grouping was developed for both the C-Corporations and pass-through entities and then compared to see which entity type produced a higher multiple. The comparison resulted in three possible outcomes: a higher C-Corporation multiple, a higher S-Corporation multiple, or a similar multiple. A multiple was deemed similar if the variance between the two multiples were within 5% of the C-Corporation multiple.

 

Approach Results

Similar to the change between Approaches 1 and 2, the year-by-year analysis resulted in the percentage of C-Corporation transactions with higher multiples dropping from 78% to approximately 59%. Part of this decrease is most likely the result of smaller subsets of transactions being compared rather than any time-related factors, however between 1995-2005 and 2005-2015, the second period had 21 more transactions with higher S-Corporation multiples. This is contrary to what occurred in Analysis 2. As discussed previously, the 2015 information available wasn’t for a full year, which may increase this differential. It is not clear why this change may have occurred over these time periods. Additionally, given the eliminations inherent in this approach (removing transactions with earn-outs, non-competes, and employment agreements reported) it is possible that it caused the multiples to be more balanced for the C-Corporations and pass-through entities. In each case the C-Corporations still have higher multiples overall.

 

Conclusion

I am not suggesting this is a flawless approach or that this data couldn’t be further analyzed to provide additional conclusions. In fact, I realize that not every transaction is perfectly comparable, there are unidentifiable niche industries where higher multiples could be skewing the data, and some transactions may have been misclassified when initially reported. However, as a whole, I believe it does provide some insight into the question at hand. Fundamentally, the data in which many valuation professionals rely to develop indications of value, also indicates that S-Corporations do not generate higher multiples than their C-Corporation counterparts. Further analysis is necessary, however, this should cause practitioners to consider all relevant factors prior to applying an S-Corporation Premium.

This initial analysis runs counter to the valuation theory that many professionals believe and therefore the impact of a company’s tax status on value is an issue that deserves further investigation and eventually a conclusion from the valuation community. As the valuation of a company already requires substantial subjectivity and professional judgment, it is unreasonable for an issue with a potentially quantifiable solution, such as the impact of tax status on hypothetical earnings, to add another layer of subjectivity and discord within the profession.

—————

 

Robert Carter, MS, CPA, CVA, CFE, CEPA is the Senior Manager of the Litigation, Forensics & Valuation Services Department for Hertzbach & Company, P.A. He specializes in performing business valuations, forensic accounting and fraud investigations, and litigation support services. He is an adjunct instructor for Stevenson University’s Graduate Forensic Studies Program and has taught courses for the National Business Institute and Maryland Chapter of the National Association of Certified Valuators and Analysts (NACVA). He currently serves as the President of the Maryland Chapter of the NACVA.