Hi,

Sadly, it’s time to hang up my skis for the year. This season had its moments. Untracked 18” of powder on March 3rd? March 30th at Taos? In a last gasp of winter it snowed a few inches on April 29th (yesterday as I write this). But I’m excited for summer. I love being outside in the hot Colorado sun.

I’m grateful for the continued amazing work I’ve gotten to do so far this year. I’m grateful that I still love playing with models and helping people make good decisions. I hope I get to keep doing this for a long time to come.

I was inspired to write about Uber again when I read about their IPO and clicked through to their SEC filing, the S-1. The absurdity jumped out quickly.

Hope you are well!

Have a big model that nobody really understands, yet is critical to your work?

Trying to raise debt or equity capital?

Need a model to get a corporate approval to move forward with your project?

Need to add capabilities to your current model?

I have numerous sample models and am happy to help you understand your options for improving your analytic tools.

Curious? Contact me for a demo.

Unpacking Uber’s Spin Machine

Uber is preparing for an IPO and the spin machine is redlining. Consider two quick examples, and then a third, in depth.

Quick Example #1: Now they’re comparing themselves to Amazon.
Amazon, famously, lost money for a long time. It was unclear if they were ever going to be profitable. But it’s quite the pivot for Uber to say, “Cars are to us what books were to Amazon.” Cars aren’t the real business? Then finish this sentence: As AWS is to Amazon, ______ is to Uber. Not easy, right? It’s a deus ex machina solution to profitability: hey guys, I can’t tell you how, but something is going to come along and make us gobs of money. I promise. (Except for all the risk factors we identified.) What could go wrong?

Quick Example #2: Drivers are Customers. Customers are End Users.
We will go into accounting games in more depth on #3, but this trick makes revenue and gross profit larger. Apparently Uber says “Because end-users access our platform for free and we have no performance obligation to end-users, end-users are not our customers.” (Uber S-1 p.138) This article does a good job laying out the implications of this policy. Also, War is Peace and Freedom is Slavery. Jamie Canizales summed this up: “This is just pure sophistry. Riders pay for the service, Uber provides insurance and customer service to riders.” By the way, Uber has long faced criticism that drivers should be employees, not independent contractors. Calling them customers instead is great jujitsu.

Example #3: Uber has their own definition of profit.
It seems that Uber has abandoned GAAP accounting. My general sense is that these things don’t end well--remember Groupon? Jamie Canizales again: “Once you leave the GAAP playing field, no way to identify the players, the score or even the rules.”

First, below, I’ll unpack the paragraph which describes their new accounting measure. Then we’ll see the impact of this on Uber’s historical performance reporting.

     Adjusted EBITDA. We define Adjusted EBITDA as net income (loss), excluding (i) income (loss) from discontinued operations, net of income taxes, (ii) net income (loss) attributable to redeemable non-controlling interest, net of tax (iii) benefit from (provision for) income taxes, (iv) income (loss) from equity method investment, net of tax, (v) interest expense, (vi) other income (expense), net, (vii) depreciation and amortization, (viii) stock-based compensation expense, (ix) legal, tax, and regulatory reserves and settlements, (x) asset impairment/loss on sale of assets, (xi) acquisition and financing related expenses, and (xii) restructuring charges.
--Uber S-1, page 22

First, it’s funny that they start with EBITDA to come up with their measure, “Adjusted EBITDA.” EBITDA was at one time a rogue measure. For example, cellular phone companies looked bad, because Operating Income included depreciation from all that network capex and amortization of license costs and goodwill. EBITDA was a clever way to get people to focus on the cash profits, and it does make sense for that. Still, telecom networks seem to always need new capital spending, and that spending is in cash! But Uber’s Adjusted EBITDA is really a “earnings before the bad stuff” measure.

So, let’s break this down and think about the worst-case scenario. Say you’re under pressure to show profits in your company. Sadly, you’re not profitable. What if you could exclude some expenses from your definition of profits? Of course you’ll remind people that you know what profit means and you reported the true losses, but let’s talk about “profits” instead, or as Uber calls it, “Adjusted EBITDA.”

So how does Uber redefine EBITDA? First, note that they work up from net income to include some “duh” items because otherwise the inclusions look so egregious.

Net income (loss) excluding
(i) income (loss) from discontinued operations, net of income taxes,

          Well, is that a loophole big enough to drive a truck through? Say I declare some project to be an independent operation, and when it fails I just say all that expense doesn’t count against profits. Brilliant. They’ve already said this doesn’t follow GAAP, so they get to define “discontinued operations” however they like. The “net of income taxes” clause is funny because I bet these were losses from these discontinued ops--which would have zero tax. But it’s awesome because the eye reads that “taxes” aren’t part of EBITDA, which makes sense when you scan it. Apparently IBM did this for years as they downsized.

          (ii) net income (loss) attributable to redeemable non-controlling interest, net of tax Um, is this a second way to remove expenses? I could set up a company where Uber owns 49% and has an option to gain control inexpensively. Expenses would disappear. If it’s a success the 51% owners get bought out perhaps and make money on the side?

(iii) benefit from (provision for) income taxes,

Fair enough.

(iv) income (loss) from equity method investment, net of tax,

I’m told this one is ok.

          (v) interest expense, (vi) other income (expense), net, (vii) depreciation and amortization,
Yes, classic EBITDA. Go on.

          (viii) stock-based compensation expense,
Is this typical? I get that it’s non-cash, but it is an operating expense. Could it be another small part of a stock-price Ponzi variant? If you don’t want to dilute shareholders you have to buy back shares, which takes actual cash.

(ix) legal, tax, and regulatory reserves and settlements,

          This feels funny, too. The legal and regulatory reserves feels like a way for Uber to take its macho “that’s just a cost of doing business” attitude towards potential legal settlements and regulatory fines and have it not count against “profits.”

(x) asset impairment/loss on sale of assets,
Cute. Loss on sale of assets is, of course, appropriate. But “asset impairment” sounds like a term of art. What if I spent money on something I’m choosing to capitalize? (Remember, this is non-GAAP, so it could be anything.) I’m building an asset! But then the project fails, as many big-company projects do. So now the asset is worth zero. You’d have to expense the full loss. But if we put “asset impairment” expenses after “profits” then I’d have a fifth way to bury expenses.

(xi) acquisition and financing related expenses,
Another cute one. Financing-related expenses are part of interest and excluded as being the “I” in “EBITDA.” But “acquisition expense” seems like a legitimate part of corporate development expenses.

My friend Craig Caukin, an auto lending expert, said, "Uber started an ill-founded/ill-fated leasing company to lease cars to drivers on three year leases. What they didn’t realize is that most drivers don’t last nearly that long and they got killed." When you're in search of a profitable business model you try lots of ideas. Craig wondered, "Acquisition/sale/financing loss? Who knows where this might show up?"

and (xii) restructuring charges.

Oh, stop, you’re killing me. So it I shuffle the deck chairs I can charge expenses to it and, magically, “profits” go up!

So what’s the impact of this?

The good news for Uber? It’s a quick way to gain 11% on your EBITDA margin. The bad news: Adjusted EBITDA is still -16%. And that’s their best case scenario for 2018. But the trend line isn’t bad either way. Still, it’s a long way from huge losses to huge profits. And just because you’re losing less money doesn’t necessarily mean you’ve cracked the code.

Maybe Uber will make it. Companies move from losses to profits all the time. It’s just a matter of what the coherent story and assumptions are that get you there. Oh, what I’d give to see their projections and the assumptions behind those! But I’m wary when companies start making up their own accounting measures. Stay tuned.