Wednesday, February 1, 2017

How to allocate capital

The biggest fallacy out there is that each dollar reinvested by companies will automatically translate into more profits.

Unfortunately, real life does not work this way. There are issues with that. There is law of diminishing returns. There is the competitive nature of business and various physical constraints.

Most dividend investors intuitively understand this. For example, a company like Target (TGT) is a better value at $62 than at $85. Therefore, it makes sense to invest dollars into Target at a lower valuation, since that provides better margin of safety and higher future dividend income potential. Buying a stock at 12.60 times earnings and 3.80% yield is better than buying at 18 times earnings and a 2.80% yield. Of course, as the company has not really grown its store base for several years, we may be seeing the limits to its growth. This is why paying a dividend may be the optimal capital allocation decision.

Most great companies that we talk about on this site have been able to grow dividends per share for decades. These companies have accomplished that because they are of high quality, have branded products, and are able to generate high returns on invested capital. These companies generate a lot of excess capital, that they do not know what to do with. This is how companies like Coca-Cola (KO), Altria (MO) and McDonald's (MCD) have delivered substantial shareholder wealth over the past 50 years. Having too much cash is a good problem to have of course. It allows you to grow your business, and still shower your shareholders with cash. I will discuss a few examples below, which will provide more detail on my main thesis above.

The first example I have is Apple (AAPL). Apple (AAPL) developed the blockbuster iPod and iPhone products when its R&D expenses were much lower than today. The company spent somewhere between $400 - $500 million per year on R&D from the time Steve Jobs returned to 2005. Then it raised its R&D expenses to $700 million for 2006 and 2007, which is when the iPhone was put in motion. The iPhone and iPod have not cost a lot of money, relative to the hundreds of billions in cash flow they have generated for shareholders. The company is now spending $10 billion per year on R&D, but it has not really been able to create groundbreaking products that would deliver the commercial success that the iPhone, iPod have had. ( perhaps the iPad in 2010 was the last great product that the company "invented"). The rest was probably spent determining how to get new versions of existing products, which is fine, but may not generate great returns for shareholders on a go forward basis. But I may argue that the return on investment has not been as good today at $10 billion per year spent on pumping out new phone versions every year, versus the $500 million/year spent on creating new innovative products. Check my analysis of Apple for more information.

In terms of the law of diminishing returns, let's look at REITs. If all REITs had a lot of cheap money and they wanted to grow, they can all get the cheap money today (depending on their credit rating). However, when allocating the money, they may end up bidding against each other for projects. If you have ever bid on anything, you know things could get out of hand and you might end up overpaying in a bidding frenzy. Paying too much for an asset, automatically decreases future investment returns. We can easily calculate that if a triple net lease costs $500,000 and generates $40,000/year, you can earn an annual return of 8%. If bidding for the property increases its price to $800,000, you would still generate a decent return of 5%/year.  However, this is a much lower return on a forward basis. If the company's cost of capital is higher than 5%, it may not take on this project. If a company is not be able to allocate new funds at an attractive rate of invested capital,  might be better off distributing that cash back to shareholders. I hope it is not a surprise for you that if a company is unable to find attractive projects that offer rates of return above its cost of capital, it should return that cash to shareholders. Paying a dividend imposes discipline on company's management to focus on a few high earning projects, rather than pursue any project, without taking into effect minimum profitability requirements.

In another example, if you are Coca-Cola (KO), and you sell 3% of all liquids consumed in a day by people around the world, you have limits on growth too. Getting your product everywhere, convincing everyone they want Coke, and not PepsiCo (PEP) or Beer, or free alternatives such as tap water, is placing a natural limit on how much growth you can get each year. If you spend all funds on growth or new products, you could end up losing money. This could happen when you reach point where each additional dollar spent generates less than $1 in profit. Developing new products is costly and risky, and could deliver losses no matter how much effort is put into the project. Check the New Coke fiasco from 1985 for reference. Even if a company finds a way to diversify into a similar field or a new one, it may still not do very well with it. There are risks with finding new opportunities, and you have to avoid overpaying for them. Even if you manage to buy them at the right price, you may still have issues with integrating different systems and cultures. This could be a challenge even if you have a merger of two equals.

Another issue with capital allocation is with share buybacks. A $1 dividend represents a $1 for each shareholder. If a company spent on the money on share buybacks, this is like a mandatory DRIP plan for remaining shareholders. The shareholders who sell get cash for their stock, while the others still hold shares (which now represent a slightly higher level of ownership in the enterprise). The hope with share buybacks is that the earnings per share could go up as a result of the share repurchase (assuming that the buybacks are not done to offset dilution caused by exercising employee options). This affects management bonuses, which is why we have seen a push for more share buybacks over dividends. But, what if management ends up overpaying for shares? If a company is worth 20 times earnings, but management ends up paying 30 times earnings, it would have wasted shareholder capital. This usually happens when companies are unable to grow organically, which is why they resort to financial engineering tactics such as buybacks to mask decelerating earnings growth. Usually you may get the double whammy of paying a high multiple for shares, and then seeing those multiple contract (since earnings expectations are reduced now) Earnings per share will grow quicker if management buys back shares at a discount. But the money would have been squandered because management is paying $1.50 for $1 worth of stuff they are buying. I highly doubt that a large portion of shareholders sit down and ever think about this stuff. Instead we hear polarizing opinions stating that "dividends" are bad, but "share buybacks" are good. I say it depends on the situation, but dividends are usually better than share buybacks.

I have seen situations where companies paid 20 times earnings per share, repurchased shares, and now the multiple is 10. This happens because the marketplace assigns varying P/E ratios to companies depending on its estimates for profitability, growth, interest rates, etc. Is that a smart deal? I have also observed companies such as General Electric (GE) buy back shares at high prices during the 2000s, only to sell them at a lower price in 2008, during the Global Financial Crisis. For buybacks to be successful, management has to get more value than the amount they are paying. Otherwise, they are squandering your resources.

The unfortunate reality is that management and shareholders have two competing objectives. Shareholders risk their capital in order to make as much money as possible. Management on the other hand wants to make as much as possible for themselves. So a share buyback over a dividend benefits earnings per share regardless of whether it it turns out to be a good allocation of capital or not. This benefits management stock options becoming more valuable, rather than improving the lives of shareholders. This short-term massaging of earnings per share really benefits management, more than shareholders. Depending on how compensation is set up for management, we may have an incentive to offer discounts to customers right around the end of a quarter or an year, just so that the company can hit its sales numbers. This short-term game only benefits management, and not shareholders. This is one reason why I really like getting a cash dividend paid to me, which serves as a sort of proof that earnings are real and not manufactured by a creative accounting department.

I haven't even started discussing the fact that when a company finds itself in a lucky position to earn a lot of money due to it being in the right cycle at the right time, it may end up hiring too much people, and become a slow moving bureaucracy. Having too much cash on hand can cause people to defend the status quo, and their position, rather than do what's best for long-term shareholders.

Today, we discussed a few examples that illustrate why it makes little sense for management to retain and reinvest all profits. It is a fallacy to assume that every dollar retained or reinvested by a company's management will automatically result in growth in intrinsic value for shareholders. In most situations, shareholders would be better off imposing some discipline on management, and receiving a dividend from excess cashflows.

Full Disclosure: Long KO, PEP, GE, MO, MCD

Relevant Articles:

Dividends versus Share Buybacks
Are Dividend Investors Benefiting from Stock Buybacks?
Special Dividends Unlock Hidden Value in Stocks
Does Paying a Dividend Reduce a Company’s Value?
Altria Delivers Dependable Dividend Growth and High Total Returns

Popular Posts