“Ideas become part of who we are. People get invested in their ideas, especially if they get invested publicly and identify with their ideas. So there are many forces against changing your mind. Flip-flopping is a bad word to people. It shouldn’t be. Within sciences, people who give up on an idea and change their mind get good points. It’s a rare quality of a good scientist, but it’s an esteemed one.”
If you meet the Buddha on the road, kill him – this old Buddhist saying implies that we need to be able to destroy our most cherished beliefs. We can grow only if we are able to reassess our belief system. To do this we need to detach ourselves from our beliefs and examine them, and if we are wrong then we must have the mental strength to admit that we are wrong, learn, and move on.
Softbank has been a very successful investment for our firm. I thought I’d own this stock for decades. It would be hard to find someone who had written as many positive articles on Softbank and its founder Masayoshi Son as I have. However, a few weeks ago we sold our Softbank stock. I met the Buddha and killed him, at least for now.
To understand why we sold it, you need to understand why we bought it. We bought Softbank when it was an unknown company to US investors. At first, our clients thought Softbank was a bank. At the time we were buying a proverbial dollar for 40 cents.
The company was misunderstood. It owned 80% of Sprint. Sprint’s debt was consolidated on Softbank’s balance sheet, and thus it made Softbank look overleveraged. That was the wrong way to look at it. Since Sprint was an equity investment, if it went bankrupt Softbank would not be responsible for Sprint’s debt. It was hard to tell at the time whether Sprint was going to survive, but Softbank’s stock was so cheap that even if Sprint went bankrupt, we’d still make money. Most of Softbank’s assets (with the exception of Japan Telecom) were public, and thus it was easy to value them. (They included Alibaba and Yahoo! Japan).
And then there was also Masayoshi Son, Softbank’s charismatic founder, whom I described as a guy who embodied the investment prowess of Warren Buffett, Richard Branson’s daring and grit in competing against Goliaths, and Steve Jobs’ vision of the future. (You can read my thoughts about him here). Mr. Son has built Softbank from nothing into a company with a $100 billion market cap.
Thus Softbank was a misunderstood, significantly undervalued company run by a phenomenal founder who owned 20% of the company – that is why we owned it.
Enter Vision Fund 1, Mr. Son’s $100 billion bet on the Singularity – a version of the future where computer intelligence surpasses human intelligence. The sceptic inside of me had some reservations about the sheer size of the fund, but Mr. Son had an incredible investment track record, and the gap between the value of Softbank’s assets and where the stock was trading had widened. (Alibaba’s stock and its earnings went up, and Japan Telecom went public at a higher valuation than we expected).
In our models, even if we took down the value of Softbank’s equity investment in the Vision Fund from $25 billion to zero, the stock was still significantly undervalued.
When Softbank bought WeWork, here is what I wrote about it:
The truth is that my firm is not as big a fan of WeWork as is Softbank’s CEO and Chairman, Masayoshi Son. WeWork is to us a real estate company that borrows long-term by buying office real estate or entering into office leases, and lends short-term by breaking large office spaces into small offices and leasing them out fully furnished for periods ranging from days to months.
My firm spent a lot of time analyzing WeWork’s competitor IWG (which used to be known as Regus) a few years ago. IWG actually pioneered this business model. We came to the conclusion that the only number which matters for this business is utilization (percent of real estate leased). If WeWork is able to lease out 90% of its space, it will be a highly profitable enterprise. If the utilization drops to, let’s say, 60%, then it will start losing money and implode. This is a classic high-fixed-cost business with variable and cyclical revenues.
This quote by F. Scott Fitzgerald has been very useful to me lately: “The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function.”
And hold opposing ideas I did. I wrote an article (read it here) describing the DotCom 2.0 of private equity markets, which Softbank’s $100 billion Vision Fund helped to inflame– that’s one opposing idea.
But then on the other side, the disparity between the value of Softbank’s real assets (Alibaba, Japan Telecom, Yahoo! Japan) net of debt exceeded the market capitalization of Softbank by $100 billion. Had the DotCom 2.0 bubble imploded, this gap would have afforded us a possible $25 billion write-down of Softbank’s equity stake in Vision Fund 1, which owned the bulk of WeWork.
What changed to cause us to sell?
This brings us to the true point of this article: How do we make sell decisions?
We sell for three reasons: First, a company reaches our fair value level. Second, we find something more attractive with a better risk-adjusted return. Or third, things change.
Let’s look at our sell decision through this lens. Throughout our ownership, despite its appreciation, Softbank stock always traded at a discount to its fair value. Even when we sold it, it was still trading at a significant discount to the fair value of its assets, even if Softbank’s share in Vision Fund 1 was worth zero.
We have plenty of cash in our portfolios, so we did not sell it to buy something else.
However, in light of recent developments we started to question the future value of Softbank’s assets and, just as importantly, the future levels of its debt.
A few months ago, Masayoshi Son announced a new, even larger fund – Vision Fund 2. I wrote about my concern with its funding (read it here):
Softbank is contributing $38 billion to the second fund – a large sum of money even for Softbank. Though Softbank’s large discount to fair value could afford seeing the first fund do poorly, we won’t have that luxury if both funds collapse…
We’ll be looking very carefully to see where Softbank gets this money and how the deal is structured. I suspect that Softbank will be selling its stake in T-Mobile–Sprint once the deal is done. Mr. Son has mentioned in the past that companies in this portfolio that have matured will be the source of capital that will flow into the ones that have growth ahead of them. This is one of the reasons why Softbank took Japan Telecom public.
I hoped that the WeWork debacle would cool down Mr. Son’s ambitions for Vision Fund 2. It looks like I was wrong. My concern now is that any shortfall of outside capital for the fund will be filled by Softbank. Mr. Son’s ambition for Vision Fund 2 appears to be larger than for Vision Fund 1. The sale of Sprint would not be enough to fund Vision Fund 2. Softbank would have to borrow tens of billions of dollars, despite owning only one cash-generating asset, Japan Telecom.
Japan Telecom has its own problem – in a year it will have a new competitor that may pressure its profitability – so we reduced our earnings assumptions and thus our valuation of that business.
The final straw that broke our back on Softbank was when the company started throwing good money after bad money to bail out WeWork and save face. (We sold it as soon as we learned that it would try the bailout). I really don’t know if WeWork can be saved; but if anyone can do it, it will be Mr. Son, who did two very difficult turnarounds in Japan and stopped Sprint from bleeding money. However, I doubt WeWork is worth $8 billion. IWG, though it doesn’t have beer fountains and ping pong tables in its offices, has revenues similar to WeWork’s, is very profitable, and has a market capitalization of $4 billion. If WeWork is really worth $8 billion, I’d suggest that IWG management install beer fountains and ping pong tables and forgo profitability. They could instantly double shareholder value.
Finally, part of our “Softbank’s true indebtedness is misunderstood” thesis was based on Softbank’s selling Sprint and improving the optics of its balance sheet, thus narrowing its discount to fair value. (Sprint’s debt would become T-Mobile’s debt.) This thesis is likely to have to be thrown out of the window after Softbank borrows tens of billions to fund Vision Fund 2 and to fill the endless hole of WeWork. If Softbank’s debt skyrockets, a 50% discount to fair value will become a permanent feature of Softbank stock. This is why we sold Softbank.
This is what I think today. If things change, I’ll change my mind.