Written by Andrew Sather
When a company creates profits, it has a decision to make. It’s all between how much of the earnings they should pay back to shareholders as a dividend, and how much they should reinvest in the business. Keeping the earnings is known as retained earnings.
Many investors can be led astray by the deceitfulness of retained earnings because on the surface it sounds like a great idea.
Warren Buffett is a proponent of retained earnings, after all, his company Berkshire Hathaway retains all of its earnings. His argument is that the company can compound the retained earnings at a much higher rate than the investor can if it was instead paid out as a dividend.
In fact, the better a company is at this– evidenced by high efficiency ratios such as return on assets, return on equity, etc– the stronger the argument for retained earnings tends to become.
Many of the “growth stocks” of today, those that are popularized by the media that usually have high rates of earnings growth, don’t pay out a dividend and use the excuse of retained earnings quite frequently. It’s almost expected by growth investors that their companies won’t pay a dividend.
The logic is this. If a company can utilize earnings and return 20% on that money, meaning create 20% more profits for the company, then that’s a superior compounding rate than an investor can expect to get in the market. And it’s true, investors can’t expect 20% returns in the market most of the time.
So then investors should be in favor of retained earnings most of the time, right?
Not so fast.
There’s several reasons why I disagree. It’s not that I don’t agree with the base level logic. I do, I get it. But you have to look deeper than that, and really examine the implications. Look at all possible outcomes of when a company does this.
Let’s not just blindly follow successful investors like Buffett. After all, his company Berkshire implements this standard for their own company– but look at many of Buffett’s past and present investments. They tend to pay great dividends or have high levels of dividend growth, indicating a stock that retains less of its earnings.
Valuations Matter, Alot
The first and most easiest point to understand is the valuation effect. I mentioned earlier that many growth stocks tend to be the biggest perpetrator of retaining earnings. They also tend to carry high valuations.
For beginners, a high valuation means a stock is expensive. And the higher a valuation you pay for a stock, the less of a positive effect retains earnings has on your results.
Bear with me, there’s some math involved.
Say you bought at stock at a P/E of 20. Let’s make the bold assumption that all stocks have an average P/E ratio over the years, so most stocks will tend to trade in a range close to that average (let’s use the median to filter out stocks that got extremely expensive and would push the average too high).
The median P/E historically has been 15.
If you think your stock doesn’t apply to the average or median P/E because you have a special snowflake, good luck to you. Thinking you can always find the exception to the rule is a terrible way to do investing, or even live life.
Yes, the P/E ratios move over time. But if you think again this won’t apply to you because you’re going to sell when the P/E is at a high range– essentially timing the market– start again at square 1. It’s been proven all over the place that you can’t time the market effectively.
Now, we don’t know where a price will go on any given day. It’s our wild variable, so let’s hold it constant.
Say you bought a stock at a P/E of 20. Hold the Price (numerator) constant. For that stock to trade at the historical median P/E of 15, a company would have to increase earnings by 33%.
[The math: multiply the P/E by 3/4 to change 20 to 15. We’re not changing P, so you have to multiply E by 4/3, or a growth of 33%]
Say someone else bought a stock at a P/E of 10. For that stock to trade at a P/E of 15, the company could lose 33% of their earnings for the same result as above.
[The math: multiply P/E by 3/2 to change 10 to 15. Multiply the E by 2/3 when P doesn’t change]
Yes I just made a strong case for the validity of the P/E ratio. But this applies heavily to retained earnings as well.
You can see that increasing the earnings by a certain amount will change the returns for an investor. If we are going to argue that a company will keep earnings to grow future earnings, you can see this has different implications for the investor depending on his P/E entry point.
The higher the P/E of a company, the more an investor depends on high earnings growth to get a good ROI.
Of course the investor could also achieve the same goal with an increasing price, but remember you don’t want to depend on that. Now you’re playing the timing game and the greater fool theory– that someone will buy at a more expensive price point than you– and it’s just not sustainable or realistic.
Notice that the lower the P/E, the less the investor leans on those earnings. The company could actually shrink their earnings and the investor would still come out even or maybe even ahead. You can see in this case that paying those earnings out in a dividend would be better for the investor than retaining them and plowing them into a shrinking business.
I really simplified this. But it’s just scratching the surface.
My biggest conclusions are coming up below.
The Return on Equity Argument
Another simple logic argument made in favor of retained earnings is this. Well if a company has proven to have a high return on equity, they should reinvest more capital into the business.
What’s equity? It’s the number of assets above the liabilities number. So say we pump 100% of earnings into the business. We’re essentially increasing equity by this much. Using earnings to buy assets. So if a company has an ROE of 20% (they turn 20% of their equity into earnings), we’re essentially increasing earnings by 20%.
Average stock market returns are 7-11% depending on who you ask and what variables you include. Surely you won’t find returns in a different stock of 20% on average. So the 20% increase in earnings should be more reliable.
And that argument is right. Take the same valuation-type math analysis we did earlier. We hold Price constant because it can’t be predicted. So if you add 20% to the Earnings (denominator) of the P/E ratio… then to achieve the same P/E ratio you’d have to increase the price side by 20%.
Which equals a 20% ROI. Nice, right?
Here’s where it gets tricky. The valuation paid again effects how much the ROE affects our investment returns. Remember the first point I made. The change in earnings affects the higher valuation investors than the lower ones.
So again… the extent that higher ROE effects our returns will depend on valuation. The higher ROE results in higher earnings growth, which we already proved, affects investment returns differently.
But We’re Buying at Low Valuations…
Here’s where value investors take their special snowflake card out. They will argue that the above doesn’t apply because they buy at low valuation. Then they point to Buffett.
I mean surely they’re already maximizing valuation advantage so any extra earnings is gravy right? And an increase in earnings should only exponentially increase returns, right?
In a sense, yes.
You’re doing the right things by buying at a low enough valuation to get a margin of safety. And hopefully your company is increasing earnings rather than decreasing, and yes you can always sell a stock doing this poorly and find a company doing this better than the rest.
That is all true. But remember that you get diminishing returns from all this as your margin of safety grows. And value investors are all trying to maximize margin of safety.
The question might now become finding the balance point where more earnings make sense and where more dividends make sense, and applying that to each opportunity. Sure we can do that, but consider this.
We preformed all these thought experiments in a vacuum. Assuming a perfect world and reversion to the mean. We all know that the ideal never happens so we have to account for that as well.
The fact that the market’s price constantly changes as does business results means that we can’t come up with a reliable method at predicting where retained earnings makes more sense and where dividends make more sense. It becomes a unique situation. Unique for each company and each individual.
I’m looking for reliable, consistent and simple strategies for my investments. I truly don’t believe you can come up with a model to determine where retaining earnings makes sense. Clearly ROE or any other efficiency metric won’t work.
The Reliable Solution: Dividends
Here’s something we can concretely measure. I’m always striving to buy companies with increasing dividends. But even in the pessimistic cases where a company isn’t increasing dividends– you’re still receiving an ROI regardless of what the market does.
What about where a company ceases to pay dividends? If you’re consistently buying stocks where this happens you have deep seated problems with your approach. Fix those before worrying about this.
This point is so key. When I own a dividend paying stock, I’m betting on averages. By using low valuations, I’m betting that the stock will eventually trade in a range close to median valuations. I’m betting that the company will pay dividends, and that over time this will lead to serious compounding.
I see “maximizing earnings” by reinvesting them in a business as a very short term strategy. Over the long term, dividends have more to do with return than share price (this has been proven historically).
And there’s no guarantee that a company’s share price increases because of increased earnings. A company could just as easily shrink earnings and the price could still go up.
Here’s my main point.
Earnings will fluctuate. Increasing these earnings help but are no guarantee. That means by maximizing retain earnings, you are doing it at the expense of the dividend to the shareholder. That’s a serious opportunity cost.
See– we’re trading one probability for another. Do you believe that earnings leading to share price appreciation is more reliable than dividends paid.
And what about the bear market? Price appreciation will all but vanish at these times. Earnings almost inevitably contract as well. Well what did that bump in earnings growth do for you during the bull?
It’s wasted once a bear hits. Oh, so you’re going to just sell before the bear hits to realize these gains? You’re treading on dangerous waters there buddy. Timing the market is a fools game.
But a value investor is smarter than that, right? They’re just going to sell at fair intrinsic value, right? I discussed that last week and why selling based on intrinsic value is also a fools game.
Contrast this to a dividend strategy. An investor holding for the long term welcomes a bear market, because they are still able to collect those dividends. And those compound very nicely over time, in fact they compound more during a bear because you are able to reinvest in the company and buy more shares.
When it comes to money, it’s much better for the investor for them to be able to reinvest their capital rather than the company. Of course it’s better for the company to be able to reinvest those earnings because it helps their business results. Remember, business results doesn’t equal stock market results.
I’ve talked about dividend compounding again and again and again. And why it’s one of the core tenets of my approach. Let me close by making the simplest argument of all.
The Goal of Investing is Income
This is seriously the most simple reality of investing, and it gets lost in the stock market complexity. Investors start by wanting to live an independent lifestyle, where they don’t need income from a job.
Well you still need income, right? So what you’re really striving for is an income stream to replace the income stream from your work. Business owners know this and most investors intuitively know this.
But investors get trapped down the rabbit hole of achieving that “just enough” amount of capital to sustain their lifestyle with investment income, and then laser focusing on that goal at all costs. Forgetting that wealth is made through compounding interest, and that interest is made from income.
Ok, maybe you are a bright stock picker who is going to compound his wealth through selling stocks at a gain and then buying more winners.
But understand that this path, if a primary foundation of your strategy, is essentially a market timing approach. Don’t delude yourself into thinking it’s not.
Investors buy bonds for the income, fully understanding that the price won’t appreciate and assuming some risk for doing this. Smart real estate investors focus on the cash flow of their investments, the amount of income received from their rents, rather than focusing on buying and flipping.
Ever notice that the successful flippers only come out to the public eye during a strong boom, and then seemingly disappear when things crash? That’s because their strategy eventually fails. It’s short term.
With dividend investing and reinvesting…
You’re consistently increasing the number of shares you own. This in turn increases the income you are receiving. Simple compound interest.
Contrast that with the long term investor receiving no dividends. Over a long enough period of time, they’ve still received no income. What kind of an investment strategy is that?
Yes I know there’s always exceptions to the rule and cases where I’d be wrong. Some companies do compound earnings better than others and create higher ROIs than the average. But I’m banking my strategy on “majority of the time” cases. That, in my humble opinion, is more reliable.
The author doesn’t hold any securities that may be listed.