Mind the GAAP

Den Howlett Profile picture for user gonzodaddy August 15, 2019
Can firms justify the use of non-GAAP measures? Maybe. But most of the time they serve to obfuscate or are there simply to help juice exec comp.


Last weekend I got into something of a Twitter fight over the use of GAAP and non-GAAP measures when reporting results. Colleagues couldn't understand why I am so agin' the use of non-GAAP numbers in reported results. Many are of the view that some degree of non-GAAP reporting is essential in order to make financial results more readily understood. At the risk of being classed as 'out of touch,' it's worth highlighting a few cases and explaining why I think non-GAAP is generally (but not always) a bad idea. 

First up, I'm not defending GAAP as such. I agree there are plenty of ways in which you can poke holes in the way it pushes firms into accounting methods that at times defy even the smartest of people. But...GAAP applies to all publicly quoted companies and therefore acts as a standard against which to assess financial performance. GAAP is rules-based and that serves to reinforce the standards nature of its application. 

The problem as I see it is that rules are almost always going to be broken by those who want to juice the accounts or show better than actual outcomes. Here is one astonishing example:


Does that make any sense to you? More to the point, since when does ANY company get to decide how it creates rules of its own making without misleading stakeholders? This is an extreme case and has yet to be tested in the court of financial analyst opinion. My hope is that advisors steer WeWork away from this dangerous path. 

Non-GAAP nonsense

Most I speak with see non-GAAP as a way for executives to ensure that the measures by which their bonuses are assessed remain intact. I can see how that makes sense for execs but then you end up in ridiculous situations.

Uber recently reported an eye-watering $5.4 billion GAAP loss for its Q2 2019 reporting period which magically fell to a more palatable $656 million after adjustments. Included in those adjustments was a whopping $3.9 billion in stock-based compensation. To put this into perspective, on a GAAP basis, Uber has piled up losses in excess of $14 billion. You can legitimately argue that one-time expenses associated with events like an IPO or acquisition should be taken into consideration when endeavoring to assess trading trends but the fact remains those costs have to be met somewhere along the line.

Here's another case in point, illustrated by MarketWatch's Francine McKenna in Salesforce.com acquisition of related company mixes complex transactions with a big writeoff in which she says: 

Salesforce.com Inc.’s fraught acquisition of Salesforce.org will force an estimated $200 million profit hit, but the transaction — which also involved another deal for a company Salesforce Ventures invested in — could produce “ghost revenue” the company can use to meet executive bonus targets.

The mechanics of how this might play out are beyond the scope of this story but McKenna's point, supported by academic opinion on the use of accounting maneuvers is that:

Companies complain the revenue write-downs affect comparability and result in numbers that are not representative of what they would have produced if GAAP rules didn’t make them eliminate the revenue forever. Since they will never be recognized under GAAP, companies instead create a “cookie jar” of revenue that can be recognized using non-GAAP metrics any time the company needs a boost to meet earnings estimates or to hit executive compensation targets.

It's hard to disagree but more to the point, it is hard to spot without undertaking a significant amount of detective work or, alternatively asking the questions outright on earnings calls. Unfortunately, analysts rarely, if ever do that. 

Salesforce isn't alone by any stretch of the imagination. According to McKenna:

Companies such as Broadcom Inc. AVGO, -1.39%  , Norwegian Cruise Line Holdings Ltd. NCLH, +0.67%  , and Symantec Corp. SYMC, +0.00%   have reportedly been adding them back to GAAP revenue numbers on an adjusted basis. 

The technique raised questions after a former Symantec employee complained about the practice internally and prompted an audit committee investigation in May 2018 into “the company’s public disclosures including commentary on historical financial results, its reporting of certain non-GAAP measures including those that could impact executive compensation programs, certain forward-looking statements, stock trading plans and retaliation.”

GAAP exceptions?

Should there be exceptions? Maybe. One with which I have little quarrel relates to the use of 'constant currency' measures. This is a technique SAP routinely uses. When you are trading across multiple currencies there are going to be times when you get the benefit of tailwinds while at others you suffer from headwinds. This is especially true for firms that have large scale US operations but which report in either Euros or GBP.

'Normalizing' numbers to reflect currency impacts makes sense when endeavoring to understand sales and expense momentum but even with that caveat in mind, I'd want to know what hedging arrangements were made and their impact. Why? Currency volatility can play havoc with cashflow. Or it can be an unexpected windfall. Either way, how well firms hedge this factor is a good indicator of fiscal prudence. 

One measure I despise when looking at GAAP/non-GAAP is stock-based compensation. I don't care whether analysts discount it as non-cash and therefore of no consequence, the fact is that SBC is a cost to the business. Excluding it makes no sense to me because the only way the firm can compensate is by juicing the stock price. In other words, it ends up that investors foot the bill rather than the firm. 

My take

The use (and abuse) of non-GAAP measures is an ongoing discussion. Customers may not care but in the long run, are they being as well served as investors when firms use accounting tricks to encourage ever upward motion of the firm's stock price while execs pocket compensation tied to stock performance? I am a firm believer in outsized gains for outsized performance. But not at the expense of accounting tricks that are often opaque, and, at times, nonsensical. Feel free to correct me but you're going to need rock solid reasoning before I accept arguments in favor. 

It is with that in mind that I chose the image of a clown to illustrate this story. 


A grey colored placeholder image