Is TSLA Berkshire Hathaway's new mystery investment?

Yesterday, I pointed out in this blog post that, based on data I gathered from 13F filings, a large number of TSLA shares seem to have gone missing during Q3, to the tune of 100M shares.

I turns out I made a small mistake, so the number is actually a little over 50M shares, but nonetheless this is a large number of shares that went missing, which can't simply be explained away by retail buying, delta hedging, and smaller institutional investors increasing their stakes.

Twitter user @s17_scott and TMC user Thekiwi pointed out to me that Berkshire Hathaway omitted one company from its 13F filing, and instead disclosed this newly acquired stake separately to the SEC in a confidential filing, because it was worried that disclosing the position to the public would move the market too much.

According to this BI article, there is a $25B difference between the value of the portfolio Berkshire Hathaway declared in its 13F and its 10-Q ($220B vs $245B). In actuality, they declared $245.5B in their 10-Q and $229B in their 13F, a difference of $16.5B.

The BI article also mentions a $6B investment into five big Japanese trading companies made over a 12 month period, which was made public in early August, and which supposedly explains part of the disparity between the 13F and 10-Q.

This seems to be true, because there is also a disparity of $5B between Berkshire's 2nd quarter 13F and 10-Q. In other recent quarters, the differences were:

• Q2'20: $202.5B vs $207.5B = $5B
• Q1'20: $175.5B vs $180.5B = $5B
• Q4'19: $242B vs $248B = $6B

So the current $16.5B difference is definitely not normal.

The BI article also mentions that this has happened once before in Q2'15, when Berkshire reported $110.5B in equity holdings in its 10-Q vs $107B in its 13F, a difference of $3.5B, only to file an amendment 3 weeks later, containing a previously undisclosed investment worth $2.5B.

Assuming that $5-6B of the current $16.5B disparity is due to the investment in the Japanese trading giants, what can the other $11B be attributed to? Could Berkshire's currently confidential new investment be TSLA?

Another important piece of information is the size of this position, and the implications it has on 13D/13G filings. Companies have to disclose an acquisition of a 5%+ stake in a company within 10 days to the SEC. So unless Berkshire was allowed to keep this confidential as well, this means the $11B must be less than 5% of the company in which it was invested. That means the company's market cap must be >$220B, which really limits the options. There are only 25 such companies listed on US exchanges.

Furthermore, Berkshire already owns stock in nine of them, and is one of them:

1. Apple
2. Amazon
3. Berkshire Hathaway
4. Visa
5. Johnson & Johnson
6. JPMorgan Chase
7. Procter & Gamble
8. Mastercard
9. Bank of America
10. Coca Cola

There are only fifteen other $220B+ companies listed on the NYSE and NASDAQ.

None of this proves anything, but when combined with the ~50M TSLA shares that went missing in Q3, there are some strong signs showing Berkshire's mystery new investment could be TSLA.

If it is indeed TSLA, this $11B would amount to ~25M shares at the stock price at the end of Q3 ($429), and would go a long way towards explaining the ~50M missing TSLA shares. Fifty million shares can't be explained away by retails, small institutions, and delta hedging, but 25M can, especially with 25M shares worth of buying pressure pushing up the stock price as much as it did in Q3, because a rising stock price requires a lot of shares to be bought for delta hedging purposes, as I've explained in my blog previously:

The Mechanisms That Fueled TSLA's Meteoric Q1'20 Rise

So did Berkshire acquire a large stake in TSLA? There's no proof, but it seems very plausible. We should find out for sure what Berkshire's mystery new acquisition is over the next few weeks or months.