When Buffett bought Coke, it was argued that he had overpaid for the acquisition. Coke sported a yield of 4% and seemed “overvalued” given that the Treasuries were yielding 8.67% at that time (T-Bill Rates).
In hindsight, this was among one of his best investments. Adjusting for splits and dividends, Buffett paid $3.75 per share in 1988. At the current market cap of $162 billion, the stake is worth $10 billion. This is a compounded 10% annual rate of return (ignoring dividends). This itself is nothing to scoff at, but if we consider the dividends, the situation changes drastically.
Coke paid a dividend of $1.88 in 2011. Given that Buffett paid $3.75 for a share, this is a 50% yield on cost! This is one of the best examples of the following Mungerism:
“Investing is where you find a few great companies and then sit on your ass."* * *
-- Charlie Munger
The aim of dividend growth investing is to do something similar to what Buffett did in 1988. You find companies that are not necessarily undervalued but have a history of rewarding shareholders with dividends which increase faster than the rate of inflation. If the company can do this for many years than the yield on cost will increase with each year. If you are lucky, it might even beat 50% yield on cost.
In theory, this is a very clean and simple idea. It follows an easy-to-understand-and-implement framework. You don’t have to worry about the market going up or down. You don’t have to worry about earning surprises or short-term pains. As long as the company is increasing dividends, you hold the stock. You can also use the dips to buy more and reduce your cost per share.
The only problem is - how do you pick a stock which keeps increasing its dividend? We don’t know the future. How is one supposed to find a Coco-Colaesque pick?
A dividend growth investor approaches this problem in the following way. He looks at the history of the company. If the company has been able to increase dividend in the past, he conjectures that it will be able to do so in the future. To increase his safety, he might also verify that the company has low payout ratio, a history of positive earnings and little or no debt. Some dividend investors will also look for good return on invested capital and they may also run a discounted cash flow model to verify that the company is undervalued.
* * *
In the discussion below, we are assuming that the company is not making any share repurchases. The extra cash it gets from the business, it either gives to the shareholders as dividend or puts them back in the business. Share repurchases have this nasty habit of being cumulative, and over the years they add up (well, nasty for the point I am trying to make). For example: If the Coke bought back all its shares except one then even if the dividend yield is 3% you will receive an astounding $4.86 billion in dividends. Also, your shares will be worth $162 billion, which is a large capital gain. The yield on cost will similarly be phenomenal.
Let us suppose that you buy Coco-Cola stock today. Coke has a market cap of $162 billion and sports a dividend yield of 2.8%. Fast forward 30 years and let us suppose that you want a yield on cost of 25%. What should be the market cap of Coke after 30 years so that a similar yield, i.e. of 3%, will result in a yield on cost of 25%?
The answer is $1,35 billion or $1.3 trillion! The market cap of Coke needs to be $1.3 trillion in 2042 for you to have yield on cost of 25% even when the company yields 3% (in 2042).
A basic question that we need to answer is this: How does the market cap of a company increase? If we are thinking about compounding the market cap over the long term, then an incontrovertible fact is increase in sales. There are only a few ways in which a company can increase its sales
- The company can increase the inflation adjusted price of the product. The cigarette industry has been doing this for a long time.
- The company can sell more products by convincing its customers to consume more of it. The per-person consumption of Coco-Cola in the U.S. has been going up more or less steadily for a long time. It can also sell more products by capturing market share from its competitors.
- The company can sell new products to its customer base. It can do so by acquiring new companies and brands or by introducing new products by itself.
- The company can start selling its products in a new market. If the business of the company can be scaled (example: Coco-Cola, McDonald’s) then it is possible to do so easily and in a capital-light way.
* * *
A large part of the world we live in is either developing or undeveloped. A significant part of the world population lives in these areas (China, India, most of Africa, most of South America and most of Asia). There are major opportunities for companies with proven scalable models like McDonald's (MCD), Coco-Cola, Pepsi (PEP), etc., to break into these areas to increase their respective sales. These companies will probably increase their sales at a respectable (inflation-adjusted) rate and reward their shareholders with increasing dividends.
There are a few favorite dividend growth investments which I am doubtful about. For these companies, the path to increased sales lacks the clarity one should demand from a dividend growth investment. Let us run down a few stocks on my list. Even if I am wrong about a particular company, I hope to convince you to look at your positions again under this light.
- General Dynamics (GD) or any defense company. General Dynamics can increase its sales by acquiring other defense companies, or capturing market share from existing ones. It can also increase sales if there is increased defense spending around the world. Acquisition or capturing market share is not going to increase the sales of the defense companies on the whole but increased defense spending will. In 20 years, do you see countries spending more money on defense or less? Do you want to take a direct bet on your opinion? Of course, you may believe that GD will be able to increase sales by breaking into new markets, but this is not as easy for GD as it is for Coke.
- Norfolk Southern Corp (NSC) or any railroad stock. If you run down the list of ways in which NSC can increase its sales, 1) and 3) are not something we can bet on. For a railroad company, increasing price is also difficult. They offer a commodity-like product and their is no brand awareness. So, 1) is more or less out too. The only thing left for them to do is to acquire a new set of railway lines and serve new customers. This is a very capital intensive task. One needs significant investment to put down new railway lines. Acquiring a set of railway lines from a different company is also not going to be cheap.
- Tech stocks like Intel (INTC). Intel might be a good value at the moment, but it is not a great pick for a dividend growth investor. The technology landscape is quite fickle. Let us run down our list quickly. 1) is difficult. Generally, when a tech company tries 2) they end up destroying sales of their previous products. A very good example is Apple (AAPL). The iPhone cannibalized the market of the iPod. I am willing to bet that a dividend growth investor will not be able to answer why either 3) or 4) would be a possible path for Intel to increase its sales. Intel has a lot of cash and it can make strategic acquisition to open new markets for itself, but this path is fraught with perils. Tech acquisitions have a knack for ending up badly. The complexity is on a different level than, say, Coco-Cola.