Non-Voting Shares are in Vogue: Do (Lousy) Accounting Rules Play a Part?
I’m starting a company and I want you to invest. I will put up $100 of my money for $10 million of yours. I will own 100 shares of voting stock, and you will own 10 million shares of non-voting stock. My voting stock entitles me to make all the decisions.
You’ll pass on my offer? I thought so. But according to a recent WSJ article the kind of arrangement I have proposed is not unprecedented. It’s also, more or less, the deal that Snapchat offered through its IPO.* Other tech companies have recently done almost the same thing, or are making similar plans for when they go public.
Personally, I stay away from these sorts of investments. They offend me. But, I am certainly not writing this post to discuss their merits or the rationality of those who invest in them. I choose to write about the accounting issues only. The recent attention by the press lends an immediacy to a feature of the system of Shareholder-Oriented Financial Accounting (S-OFA) that I am developing.
Keeping Up Balance Sheet Appearances
As I described in a post way back in 2007, the balance sheet classification scheme for the claims on assets can be a much bigger deal than the classifications of the assets themselves. Under U.S. GAAP (or IFRS), calling an issued derivative a liability, for example, could mean (horror of horrors) measurement at fair value on each balance sheet date. And even without fair value measurement, the effect on debt/equity ratios could put any company on the wrong side of a debt covenant. Banks are particularly sensitive to this because their profits are particularly reliant on flirting with the edges of the minimum capitalization ratios required by their prudential regulators. Consequently, it might come as no surprise to you to learn that U.S. GAAP for this topic boils down to an instructional guide used by financial engineers for avoiding liability classification of the complex contracts they are hired to cook up.
The foregoing is important background to the non-voting stock problem I started out with, but in point of fact, classification of common shares under GAAP – voting or non-voting – is straightforward. The end of their accounting story is that all common shares, regardless of relative voting rights, are classified as owners’ equity (since they don’t possess any of the characteristics of a “liability”).
But that’s not the end of the economic story. What if the Snapchat founders vote to acquire another company – maybe at an inflated price and using common shares as consideration – that just so happens to be owned by a related party? Or to give themselves a bonus, or options to purchase more shares? Or to sell the company while reaping a huge ‘golden parachute’?
My point is that some classes of equity may be more equal than others, even though the accounting makes no such distinction. Recent trends are telling me that the games won’t stop unless everything about liabilities and equity are changed completely. Here’s how to do that:
First — Claims on assets are classified as either: (1) the residual interest; or (2) non-residual interests. (Note that these terms replace “owners’ equity” and “liabilities,” respectively. More on this, later.)
Second — The residual interest is the lowest-priority present claim on assets. In the case of a tie, such as between identical instruments other than their voting power, the class of securities with the least voting power per share is deemed the residual interest.
Third — A change in non-residual interests is offset by a change to the residual interest to the extent it is not offset by a change in assets. Similarly, a change in assets must be offset by a change in the residual interest to the extent it is not offset by a change in non-residual interests.
Fourth — The change to the residual interest is measured as the change in assets, less the change in non-residual interests.
That’s pretty much all there is to it. Except to explain, as promised, the jettisoning of “owners’ equity” and “liabilities.” Both terms harken back to simpler times, and “liability” in particular has been difficult to define. It feels like at least 15 years ago since the FASB announced its intention to revise the definition of “liability” in its conceptual framework due to well-recognized flaws. It has made a lot of changes to the conceptual framework since then, but as to the definition of a liability, nothing doing. As to owner’s equity, the example of common shares with different voting powers already says enough about the problem with the current definition.
S-OFA will not require a definition for liabilities, or whatever may take its place. This is because true accounting logic requires recognized claims to be the negative of assets. Therefore, a non-residual interest is not eligible for recognition unless it would be recognized as an asset, under S-OFA, of some other entity.
I can promise you that what I am proposing will be an integral part of S-OFA. That’s because I am the only member of the S-OFA rulemaking board. But there is also nothing to prevent a majority of the FASB from greatly simplifying U.S. GAAP for the better by adopting the same. I mention this in part because it came darn close in 2007 when it published a preliminary view that owners’ equity should be limited to what they termed “basic ownership interests.” It was an idea that happened to be floated during the convergence era; and it was dropped like a hot potato when the European banks threatened to disavow IFRS should such a horrible thing come into being.
Whither S-OFA or U.S. GAAP, my proposal will take the fun and games out of financial engineering for an illusory accounting result; and the moguls like Snapchat’s founders will having something else to chew on as they plot to undermine shareholder democracy. For example:
- Dividends on non-residual classes of stock would be expensed.
- Stock options granted to employees would be classified as liabilities under U.S. GAAP instead of owners’ equity. Because they are derivatives, they would be measured at fair value through net income. In effect, we would be shifting from grant-date valuation to exercise-date valuation.
- Non-controlling interests would be re-classified from owners’ equity to liabilities with dividends expensed.
Perhaps it is wishful thinking on my part, but with convergence in the rearview mirror, the increasing prevalence of multiple classes of common stock could be the bridge too far for the FASB; and it will revert to something resembling its 2007 preliminary views document.
Yes, I know, I must be kidding myself. The more I blog and learn, the stronger I come to feel that the best, if not the only, path to real improvement in financial accounting starts out with a blank slate. Let’s compare the hundred-or-so pages that will be S-OFA with the thousands of pages of U.S. GAAP. Whatever the outcome, my work will be done.
—————-
*Thanks to Bob Jensen for bringing the article to my attention through the AECM.