1/ Get a cup of coffee. In this thread, I'll walk you through the "magic of retained earnings". This is the basic theory behind why stocks grow exponentially over long periods of time. As investors, we'd do well to understand this theory -- and the assumptions it's based on.
2/ Warren Buffett's 2019 letter to Berkshire shareholders has a section titled "The Power of Retained Earnings". In this section, Buffett describes how businesses can deliver enormous benefits to their owners by *retaining* and *compounding* a portion of their earnings:
3/ Let's break down this key insight from Buffett's letter. Imagine we have a business that earns $100M per year. Let's say this $100M neither grows nor shrinks over time.
4/ Suppose the business has 10M shares outstanding. Then, *per share earnings* work out to $100M/(10M shares) = $10/share. Each year, let's say the business distributes its entire $100M of earnings to its owners. This is done via a $10/share dividend.
5/ Now, suppose we want a 12% per year return from buying and holding this business. How much can we pay for each share? That's easy. As each share gives us $10 per year, we can pay ~$83.33 per share. Because 12% of ~$83.33 = $10, we'll get our desired 12% return.
6/ Now, suppose 5 years down the road, someone else wants to buy our shares from us. Let's call this guy Peter. Suppose Peter also wants a 12% return from owning the shares. How much will Peter be willing to pay us per share?
7/ Well, Peter's calculation is *exactly* the same as ours. Once Peter buys our shares from us, he will get the same $10 per share per year dividend. So, rationally, Peter should pay us the *same* ~$83.33 per share -- 5 years from now. Only then, Peter will get *his* 12%.
8/ Thus: A "rational" price to pay for the shares TODAY is ~$83.33 per share. And a "rational" price for them 5 years from now is ALSO ~$83.33 per share. That is, we can't really expect the shares to rise in value over time.
9/ Of course, the market can do irrational things in the next 5 years. So, our shares may rise or fall. They will most likely NOT remain flat at ~$83.33 per share. They will most likely fluctuate.
10/ But the point is: the "intrinsic value" of our shares, discounted at 12% per year, will remain at ~$83.33 per share *forever*. Over the long run, IF we believe that market prices will roughly mirror intrinsic value, we cannot rationally expect our stock's price to rise.
11/ That is, there's simply NO good reason why we'd expect someone to pay more for our stock 5 years from now -- barring "macro" factors that decrease our 12% discount rate (eg, falling interest rates, equity risk premiums, etc.).
12/ Over the long run, we can reasonably expect *exponential* growth in stock prices ONLY from businesses that compound their *intrinsic value* exponentially. That means *fundamentals* like earnings, cash flows, and dividends MUST grow exponentially over time.
13/ So, let's change our assumptions about our business. Let's say its dividends will *start* at $10/share in Year 1. But from there on, dividends will *grow* at 10% per year. Now, what happens to our business's "intrinsic value" over time -- at the same 12% discount rate?
14/ It turns out that intrinsic value now grows at 10% per year -- in line with dividend growth. Initially, intrinsic value is $500/share. In 5 years time, it grows to ~$805.25/share. Calculations:
15/ So, for a business that delivers exponentially growing per share dividends over time, it's *rational* to expect the stock price to ALSO grow exponentially at the *same* clip. As time passes, businesses that return MORE cash to shareholders become MORE valuable.
16/ Of course, businesses cannot keep growing their *dividends* forever -- unless they also keep growing their *earnings* at least as fast. So, over the long term, exponentially growing earnings are necessary to sustain an exponentially growing stock price.
17/ But here's the thing: At most businesses, *earnings* growth isn't free. It takes *capital*. For example, if Nike wants to grow earnings by selling more shoes, it has to build more factories, keep more inventory, spend more on marketing, etc. All this requires capital.
18/ Similarly, if Starbucks wants to grow earnings by selling more coffee, it has to open new stores, buy more espresso machines and coffee beans, employ more baristas and invest resources into training them, etc. All this again requires capital.
19/ So, where do businesses get capital to fund earnings growth? One option is to issue debt. Another is to increase "float" -- payables to suppliers, pre-paid revenue from customers, deferred taxes to the government, etc.
20/ But the most common way for businesses to fund their growth is to *retain* a portion of their earnings. That is, give out only a *part* of earnings as dividends to owners. *Re-invest* the rest into new projects, acquisitions, etc., to drive earnings growth. Like so:
21/ Businesses that are able to successfully execute this loop year after year -- earning good returns on *both* "legacy" capital and "new" capital added via retained earnings each year -- tend to deliver exponentially growing stock prices over the long run.
22/ Here's a simple math model for such a business. Each year, the business starts with some capital -- on which it earns R%. D% of these earnings are dividended out. The rest, (100 - D)%, are retained and added to the capital pool for next year. And this repeats forever.
23/ As the math above shows, under these assumptions, intrinsic value grows at the same rate as earnings and dividends. Thus, the compounding of retained earnings is the engine that drives long term exponential stock price growth in wonderful businesses.
24/ But of course, just because a business retains earnings does not mean intrinsic value will compound. For that to happen, these retained earnings have to be *deployed* intelligently and earn a meaningful *return*.
25/ Failed acquisitions, investing in low return projects, share buybacks at fancy prices, etc., are common ways businesses *destroy* the value of retained earnings. In such cases, owners would be better off if the earnings were just dividended out to them, and NOT retained.
26/ Buffett's "dollar test" is a good way to think about whether retained earnings are adding value at a business or not. Every dollar retained by the business should create more than $1 worth of future intrinsic value, when adjusted for inflation and the time value of money.
27/ For example, suppose a business we own retains $1 instead of giving it to us as a dividend. The business uses this $1 to buy an asset that will last 10 years. And the asset will earn $0.20 per year for those 10 years. And these $0.20 will be given to us as dividends.
28/ Is this business using its retained earnings to add value for us or not? Well, the business *could* just give us $1 now. Instead, it's going to give us $0.20 per year for the next 10 years. So, we forego $1 now to get $2 over 10 years. That's a ~15% annualized IRR.
29/ IF we're not aware of any opportunity to earn ~15% on our own, we're better off just letting the business retain our $1 and re-invest it on our behalf. But IF we know of an opportunity that can get us, say, 20%, we'd be better off if the business just gives us the $1 now.
30/ In addition to this kind of "opportunity cost", we should also consider the impact of inflation and taxes when weighing such "dividend vs retained earnings" trade-offs. But this thread is already too long! So, we'll cover these topics another day.
31/ To learn more about the power of compounding retained earnings, plus other related concepts, please join me tomorrow (Sun, Jan 23) at 1pm ET via the Callin app (@getcallin) for a new episode of Money Concepts. Link:
callin.com/link/umjHPSxogS
32/ If you're still with me, thank you very much! Please stay safe. Enjoy your weekend! /End
