The S-Value Premium

Benefit or Boondoggle?

Is the S corporation premium defensible? In this article, the author shares her views on this matter by answering the following questions: Should there be a premium applied to the S corporation whose value has been determined relative to the publicly traded C data by which it has been valued?” Are the assumptions we use to compare S and C attributes reasonable? Do they make sense? If not, what might we use instead?

As a profession, business appraisers continue to chew on the issues surrounding the value advantages derived from the tax attributes of pass-through entities (PTEs)— in particular S corporations vs. those of C corporations.  Thoughtful research and analyses have been performed and published.  Models have been proposed and widely used.  Debates have been held in court and out.  However, all this activity has resulted in a thicket of theory and practice that is challenging to maneuver with clarity and common sense.

After some years of trying different paths through the thicket, this author has identified a number of anomalies in current practice that make the resulting PTE premium questionable.  They can be captured in the following questions:

1. “Should there be a premium applied to the S corporation whose value has been determined relative to the publicly traded C data by which it has been valued?”[1]
2. Are the assumptions we use to compare S and C attributes reasonable? Do they make sense?
3. If not, what might we use instead?

 

To answer these questions, we will explore each baseline assumption in our S-valuation model.  For simplicity, we will abbreviate certain terms, as follows:

• S refers to S-corporations and other PTEs —all privately held.
• C refers to publicly-held C-corporations.
• S-premium refers to the value premium associated with S-corporation (and other PTE) tax attributes.
• Investor, shareholder, and owner refer to the parties that hold ownership interests in publicly and privately-held We use them interchangeably.

Assumption 1: It is possible to create a proxy C that has the tax attributes of a publicly-held company (in order to match the rates of return data we use from the public markets) but is “otherwise identical” to the subject S.

But, is this realistic?  We can make certain adjustments to the S benefit stream and capital structure to make it more like a C.  But, as we know, foundational dissimilarities that cannot be adjusted away still remain between a C and an S.  We suggest, if we apply the term “otherwise identical” to the proxy C, we may be setting up a stringency of comparison that is neither realistic nor useful in exploring the need for an S-premium.  The term “proxy” is sufficiently precise.

Assumptions 2 and 3: The S-premium is based on the single-level tax paid by S shareholders vis-à-vis the double level of taxes that would be paid by shareholders on an “otherwise identical” C.

The rates of return we use from the public markets embed shareholder expectations for all entity and shareholder-level taxes.  This automatically makes them a mismatch to the S benefit streams being valued, from which no payment of taxes will be made at the entity level and only income tax will be paid at the shareholder level.

How do we realistically address the “double taxation” and the “mismatched rates of return” issues?

First, as unlike as they are in most respects, C and S investors have two fundamental goals in common: (1) solid returns (current income and/or longer-term capital appreciation); and (2) the avoidance or minimization of taxes.  This allows us to reasonably compare C and S investors and their tax attributes when addressing the S-premium debate.

Second, both the dividend and capital gains tax effects that are supposed to play a major role in obtaining an “otherwise identical” C benefit stream (or adjusting the market-based rate of return) and in quantifying the S-premium, are questionable.

Plentiful market evidence concludes that not only are dividends of mixed interest to both C and S investors, but also they are not always distributed; are distributed at varying levels (but always modest for Cs); and that dividend taxes paid by C shareholders may be close to zero due to the presence of untaxed and tax-favored institutional investors.  Furthermore, virtually 100% of the market data we use for most S valuations come from non-dividend-paying microcap portfolios, not the dividend-paying S&P 500.

Other than the S&P 500, actual C payment of dividends (and subsequent payment of dividend tax by investors) is uncertain at best.  Thus, we suggest that the proxy C flows and market rates of return we use do not actually reflect the payment of dividends or dividend taxes.  And, it is not meaningful to develop a proxy dividend tax on S distributions.

What about capital gains?  At the surface, C and S investors take capital gains and pay taxes on them quite differently.  As a single example, holding periods for C stock can range from a day to an average of 3.8 years (calculated over the period from 1926 to 2014), changing hands many times over a decade.  However, S stock can be held for decades by the same investors.  But, digging deeper, we see important similarities: both C and S investors can select holding periods at will; many intersecting factors affect their selling choices; and both can delay or mitigate capital gains taxes indefinitely by various investing strategies.

Given this, we suggest that market capital gains and related taxes are not reasonably assured or quantifiable.  Thus, quantifying an S capital gains tax is only appropriate when it is known or knowable that the S will be sold in a stock sale within the five-year forecast period.  Under any other scenario, an S-premium for capital gains tax is purely speculative and not defensible.

As for adjusting the market rate of return downward to remove the effects of embedded dividend and capital gains taxes, we could use the newly-minted Fannon-Sellers method.  However, their market data-based sample calculation demonstrates a 1.3% increase in the total rate of return for both taxes together.  The academic research they cite indicates that even this small adjustment may be too large.  Realistically, a market rate of return adjustment for dividend and/or capital gains tax effects is not material enough to be necessary.

But, there is another far more important flaw in our current calculation process.

Currently, when we develop the benefit stream to be valued, dividend and capital gains taxes are not deducted from proxy C entity-level cash flow, because they are not paid directly by the entity.  They are paid directly by the shareholders.  Yet, we suggest, if they will never be paid by the entity, they cannot be considered “savings” for the entity and rolled into an add-on S-premium.

We suggest the S-premium creates excess value that cannot be supported by the entity’s actual net cash flow.  Based on its historical and forecasted fundamentals, the entity will not generate the net cash needed to make this excess value available to the shareholders.  The result of our current calculations is overestimation of S value.

So—how do we solve this dilemma: C income tax is paid at the entity level; S income tax is paid at the shareholder level; yet, we normally value an entity-level benefit stream?

Whether intentional or serendipitous, the Delaware Chancery Court’s ruling in Delaware MRI and Barr’s Modified Delaware MRI Model capture and resolve it cleanly.

But, we encourage the reader to step outside our current rigid view of who pays corporate income taxes.  We suggest all taxes on corporate pre-tax income are paid by the shareholders.  Some are paid directly (S individual income taxes).  Some are paid indirectly (C corporate income taxes) through the company as the shareholders’ agent.  Both direct and indirect income taxes have the same base and result.  They are paid on the entire pre-tax income of the firm and reduce the net cash flow available to shareholders for their discretionary use, now and in the future.

If the investor/shareholder is the beneficiary of entity value and also the actual payer of entity income taxes, can we not just drop the entire “entity-level/shareholder-level” tax debate?  We could develop a realistic effective tax rate for the proxy C and a blended, realistic effective tax rate for the S shareholders, compare these rates, and address the differential using a methodology of choice.

While this article is too brief to explore this further, it is worth considering.  It certainly would provide simplicity, clarity, and realism to our S valuation model.

Assumption 4: Since the complexities of developing realistic effective tax rates for both business entities and individuals are great, it is most conservative and reasonable to use statutory tax rates.

There is no question that estimating precisely accurate tax rates for both corporations and individuals is difficult or impossible, and statutory rates seem like a simple solution.  However, in the real world, individual and corporate taxpayers rarely pay taxes at statutory rates.  As a result, our models skew value, often substantially.

How can we resolve this issue?

Eric Barr suggests we focus on developing reasonably accurate individual income tax rates for application to S pre-tax income.  His “with and without” methodology and simple income tax preparation software will achieve this nicely.  We can, then, use an after-S-income-tax benefit stream to backsolve for a comparable proxy C effective tax rate and use it in the valuation.

For a less precise but more direct approach, we can use the average tax rates paid by individuals in various AGI (Adjusted Gross Income) ranges, as provided by a number of respectable online sources.  In addition, Damodaran’s website provides an extensive database of average corporate income tax rates for industries and the market as a whole.

Either of these, or a combination of them, is more defensible than using statutory rates.

In conclusion, when the messy facts of real-world C investing are taken into account, most of the C assumptions and attributes we compare and use in S valuation—and S-premium calculation—are not sensible or necessary and should be dropped.  The useful ones that remain can be approached more effectively, using simplified models and realistic tax rates to generate more defensible results.

The S value premium a benefit?  A boondoggle?  Given the facts, you decide for yourself.

[1] Ibid. p. 584.

Sarah von Helfenstein is Founder and CEO of Value Analytics & Design, LLC. She has created, launched, managed, and consulted to new initiatives inside/for established enterprises and on a stand-alone basis. For the past twenty years, she has served the valuation profession as financial economist and ethnographer, educator, champion, and change agent.  Ms. von Helfenstein can be reached at: (617) 401-1122 or by e-mail at svonhelf@valueanalyticsanddesign.com.