Softbank was Simply Hyper Fuel on WeWork's Already Raging Inferno

In Reality just a Real Estate Company, Albeit a very Sexy one

WeWork was hemorrhaging money long before SoftBank showed up, burning through $500-$750 million of cash for every $1 billion in revenue it generated.
Those expenses have fairly evenly split between those specific to operating the buildings and those to build and support the company, such as sales, marketing and administration.
The big carrot the company has been dangling is the promise of dramatically improving building operating margins (ex all support spending) from 15%-30% upon achieving critical mass in building maturity, as it defines buildings operational for 24 months.
However, as of the end of June, memberships accounted 87% of available workstations. The company sited London as an example of a mature market at 93%. An additional few thousand memberships would barely appear as a drop in the roughly $2.5 billion cash burn bucket looming for the year.
“You Name It” as a Service is valid economically only if it can scale and leverage. With high fixed costs, virtually no barriers to entry limiting WeWork’s ability to significantly raise rates, it seems highly unlikely that office space will produce that effect.
Softbank will not be interested in running a real estate leasing company; its best hope is to stabilize WeWork and sell it to a competing real estate developer for pennies on the dollar, if that.

SaaS: Space as a Service

WeWork has created dynamic office environments that it conceptualizes as “Space as a Service,” which has indeed proven highly popular with clients worldwide, growing to serve 527,000 individual memberships by the end of June, 2019. But it has done so by offering very attractive price points and amenities, the build-out and support of which far exceed the operating profit the facilities themselves generate. For unlike the cloud based S(oftware) as a Service that WeWork is attempting to emulate, new members don’t leverage over existing cloud software platforms: the real estate build out must continue to perpetuate growth.

One of the first points WeWork presents in its recent S-1 is that can sell its individual workspaces at some 60%-65% lower than what it would cost a company to build out a comparable space on its own,, as represented in the figure below.

WeWork’s Pitch: Comparing the Cost Benefit to a Landlord of a WeWork rental versus a traditional lease.(!)

While this scenario is terrific news in marketing to potential clients, it is in fact not useful information to potential investors, as it doesn’t take into consideration WeWork’s cost in the creation of that space.

The left three columns in the chart below replicate WeWork’s from its S-1. The two right columns are its averages of its estimated 2019 revenue and costs per category (whereas the left three columns are based off of ten specific markets only).

WeWork is currently generating a slight –on average $500 per member—annual operating profit when considering only costs directly associated with the operation of the space. However, the company’s overhead and capital expenditures dramatically overshadow that profit, currently by over $8,500 per average member annually.

The underlying flaw in the WeWork Model: It’s buildout costs exceed average annual revenue per average workspace by $8,500.(2)

Memberships are at 87% of Workspace Capacity. How Much More Leverage Could There Be?

A significant number of variables are involved in determining the scale the company would need to reach to absorb that overhead, even presuming some surely to be expected headcount reduction. One that the company pushes aggressively is the dramatic ramp to full occupancy as locations enter their third year of operations, which the company considers the age of maturity.

However, data provided by the company suggests an existing capacity of almost 90%, based upon single memberships as a percentage of singe workspaces. Further, WeWork cites London as one of its highest occupancy markets, at 93%. Were WeWork to have produced companywide 93% occupancy for the period ended June 30, 2019, at similar operating profit per member, the incremental income would have been less than $2 million – not much of a dent in the roughly $2.5 billion the company is on target to burn this year.

“Obligations under operating and finance leases signed as of June 30, 2019 were $47.2 billion.”

WeWork is, perhaps rightly, proud if its $4.0 billion of committed revenue backlog, giving it featured billing in a bold yellow bar chart on page 4 of the S1. However investors must be diligent in their reading to find WeWork’s committed lease obligations, which are mentioned, in small print with no braggadocios yellow bars illustrating their growth, on page 117.

As of June 30, 2019, $17.9 billion were recorded on the balance sheet. One can only wonder, whilst jubilant WeWorkers were rushing about the planet committing themselves to an additional $30 billion in leases, if anyone at SoftBank, or anywhere else for that matter, ever bothered to attempt quick back-of-the-envelope calculation on how many tens of billions of dollars they were actually committing to spend.

One can only imagine that those who survive the draconian headcount reductions that are certainly eminent will have the same cavalier enthusiasm for the return visits asking ever so politely to be excused from what is likely to be an extensive number of these obligations.

Notes:

Comparison data represent annualized costs based on an average of ten select city centers in the US, Europe, South America and Asia that are representative of our key markets. “Standard lease” includes the costs associated with food and beverage, events, utilities, insurance, property taxes, facilities management and base rent. “Build-out” includes the costs associated with construction, procurement and design services offset by a tenant improvement allowance amortized over an illustrative five-year lease.
2. Represents annualized average membership and service revenue per WeWork membership for the six months ended June 30, 2019 across the same 10 select city centers as above.]
Calculations based upon entire company as opposed to the 10 select markets of the WeWork chart, so not quite apples to apples, but in fact likely more favorable to WeWork as their data focuses on more expensive markets. Gross capex used as WeWork net landlord buildout contributions in lease costs so would be double counting.