Hacker News Reader

Hacker News

Comment by baxtr | Hacker News Reader

Parent Full thread

baxtr

Do you mind sharing a link of your linking for (a)? Thank you

View on HN

Buffett routinely mentions it in his annual letters:

https://www.berkshirehathaway.com/letters/letters.html

A good place to start is by reading his thoughts on See's Candy as of 2007 (~four decades after he bought it):

https://berkshirehathaway.com/letters/2007ltr.pdf (search for "See's Candy")

I'll summarise the thinking for you. If you got $1m in profit that sounds great. But if you had to invest $1bn to get it, that sounds less good, because you could have made more by putting the money in the bank at a much lower risk.

Profit only makes sense when considered against the amount of capital required.

Isnt thay accounted for in the revenue - profit equation?

The cost of capital is expressed in their balance sheet as expenses or depreciation. Pay back loans, investors, etc are all considered when calculating profit.

In theory but that's where the accounting games come in: how you choose to capitalize the expenses over what time horizon and how you recognize the returns are extremely relevant. While you might point out that there are commonly accepted accounting principles, you'll also note that people use all kinds of different approaches for different types of businesses were where they argue the commonly accepted model is not quite the right fit the shape of the business

The amount of capital tied up in fixed assets is only one component of the capital required. Many other short-term and long-term assets can be components. Every business is different.

For example, a consulting firm may have almost no fixed assets, but let's say its customers are mostly large corps that take 90-120 days to pay invoices. When the firm gets hired for a new project it must cover its expenses for 90-120 days until it gets paid. As a going concern, the firm requires capital equal to 90-120 days of revenues to finance its accounts receivable. If the firm's revenues grow from, say, $100M/month to $120M/month, all else remaining the same, the firm will require an additional $60M to $80M in capital = ($120M/month - $100M/month) / 30 days/month * 90 to 120 days to collect.

Thanks for that.

So in other words: something like amazon is not a good business for Buffet since they have to finance the inventory for a couple of month at least?

Change in inventories is only one component of capital required. There are many other components. There are great businesses that carry lots of inventory. Every business is different.

Amazon, in particular, is a large entity incorporating numerous businesses (AWS, Prime, Alexa, Merchant Services, proprietary brands, etc.) that are different from each other. Each should be analyzed on its own, because their dynamics are different. For example, Merchant Services is a platform for third-party sellers. They, not Amazon, are the ones who pay upfront for the inventory. They pay Amazon to store and manage that inventory. When a product in that inventory sells, Amazon gets to collect the money upfront from consumers. The sellers get paid sometime later.

Here's a decent primer on estimating return on invested capital: https://www.morganstanley.com/im/publication/insights/articl...

Yeah sorry, I should have said online retailer. I had amazons original business in my mind.

Of course you’re right that it’s a complex business nowadays.

The portion of invested capital tied up in inventory is not sufficient to judge a retailer. What we want to know/estimate is the retailer's return on invested capital (ROIC). A retailer with significant competitive advantages that can generate above-average ROIC for many years to come is a good business. If, on top of that, that retailer can be acquired at a sensible valuation in relation to such future ROICs, it would also be a good acquisition.

https://www.berkshirehathaway.com/letters/letters.html

Reply on news.ycombinator.com