[nrao123]

This is what I wrote to a friend in an email on this post:

A few challenges with the trying to replicate the Berkshire model on the internet:

##Lower cost of Capital (Float): One of the unsaid rules of BK is that their cost of capital is cheaper than most operating companies because of thier insurance float (Ajit Jain, General RE, GEICO etc). ​So, they can take an existing business & assume absolutely no changes to their changes & still make better ROI because their capital structure is more efficient. It is a different matter that they take portfolio companies like BNSF & use float to expand their capital projections (this is like making a private investment vs a public investment. From the BK 2015 Letter on how capex is linked to float:

After a poor performance in 2014, our BNSF railroad dramatically improved its service to customers last year. To attain that result, we invested about $5.8 billion during the year in capital expenditures, a sum far and away the record for any American railroad and nearly three times our annual depreciation charge. It was money well spent. BNSF is the largest of our “Powerhouse Five,” a group that also includes Berkshire Hathaway Energy, Marmon, Lubrizol and IMC. Combined, these companies — our five most profitable non-insurance businesses — earned $13.1 billion in 2015, an increase of $650 million over 2014.*

##Bullet proof Revenue (Moat): One of the operating assumptions behind BK's acquisition is that the business does well regardless of management. Therefore, if the revenue assumptions hold (i.e. this is the moat part of the strategy)- then the float kicks in & does magic.

However, the challenge with using a BK model in tech i.e. acquiring companies & being hands off without PRODUCT ENERGY- the products atrophy & revenues goes. The reason that BK hates tech is because there is moats are hard to sustain & it is too competitive.

I am looking at Tiny's portfolio: http://www.tiny.website/ & I am hard pressed to think of any product that has moat. I would suspect that once the founder leaves in almost all of these cases, the products will atrophy.

[biggy31]

Without getting into all the details about what Berkshire is and what their methodologies are, the author merely wants to be nothing more than a "conglomerate" of tech/internet businesses, and was emphasizing his ability to close fast(er).

I think we can all agree thats about as similar as it gets to Berkshire Hathaway, and what the author is doing is not really that unique or similar.

I view the article as nothing more than a way to get press and hope to try and improve the author's deal flow in case anyone who reads the article is interested in selling their business, and if I had to guess the author got at least a few leads from it.

[erdevs]

This is an outstanding comment. Few people understand what Buffet and Munger mean when they discuss "moat" and even fewer understand the importance of cost of capital in BK's investment successes.

It may be possible to build a BK of the internet but it's not easy to imagine how.

[dave_sullivan]

Buffett has famously stayed out of tech companies because he (self admittedly) doesn't understand them.

A BK of the internet would buy ugly tech companies, turn them around, and sell them. Buffett is definitely not the only guy in that business, but he's probably been the most successful. An internet version of that would have to really really understand technology and the markets driving it, along with understanding the modern financial system and sources of capital (to finance buying the companies). It is kind of a weird combination.

However, I do think we're going to start seeing waves of consolidation in tech, and maybe some interesting turnarounds.

[tim333]

Buffett has bought some tech in recent times - IBM and Apple. More established stuff with cash flow.

[dave_sullivan]

Indeed. What do you think he meant when he talked about not wanting to invest in tech companies in the past? Or did he never say that?

[zeusk]

The moats in tech are mostly around IP and hardware, look at Apple, Google, Microsoft, Intel, Nvidia and AMD. The larger ones have a moat complemented by a warchest for acquisition and lawsuits.

[mooreds]

> The moats in tech are mostly around IP and hardware

And network effects. That was true with most of the OS providers since the OS was an independent component, and obviously matters for some of the modern bigcos (Facebook, Amazon to some extent [third party sellers], Ebay).

[hobofan]

> One of the operating assumptions behind BK's acquisition is that the business does well regardless of management

Is that the case? I am currently reading the collection of Warren Buffets' Annual Letters to the Shareholders (https://www.amazon.com/Berkshire-Hathaway-Letters-Shareholde...) and in every letter, he attributes a significant part of the success of their companies to good management.

[nrao123]

He likes to praise his management and his management is sometimes good. BUT - when he buys or invests in a business - he wants to only do it such that ROI will work even if the management is not good. Thats the difference between a tech small business where the founder / product guy has to be switched on and be excellent otherwise - your business goes haywire. Here is an example off Buffett and his thinking on management:

I knew nothing about the management of Moody's. The - I've also said many times in reports and elsewhere that when a management with reputation for brilliance gets hooked up with a business with a reputation for bad economics, it's the reputation of the business that remains intact.

If you've got a good enough business, if you have a monopoly newspaper, if you have a network television station - I'm talking of the past - you know, your idiot nephew could run it. And if you've got a really good business, it doesn't make any difference.

http://www.businessinsider.com/warren-buffett-good-business-...

[0898]

"I always used to tell Gates that a ham sandwich could run Coca-Cola" [Buffett]

[tim333]

He prefers business any idiot could run but buys other ones as well such as IBM and some airlines recently.

[Nomentatus]

From my reading this year: starting out, with his mentor, he invested in companies with poor management the could be improved. He dropped that practice long ago, however, in favor of buying good management along with a good business. Management parachuted in often does badly, since every business is different. He does look for businesses with an edge (such as a brand) that will persist, quite true; even when he doesn't understand exactly what that persistent edge in the market is (he or Munger gives the example of breakfast cereal, that way.)