14
EARNINGS GROWTH ISN'T ALWAYS A GOOD THING
I now want to expand on the discussion introduced in the previous chapter regarding the desirability of earnings growth.
Earnings per share (EPS) is calculated by dividing a company's net profit after tax by the number of ordinary shares on issue. Investors often include a requirement for positive EPS growth in their stock screen. Chairpersons and CEOs love writing in the annual report that they've delivered EPS growth for the most recent financial year or for the past several years. They present this information in the form of bar graphs showing progressively higher bars for each successive year. Reported growth gives shareholders a warm, tingly feeling because their shares are earning more every year. The reality is sometimes it's all a bit of a con.
For starters, information regarding how much a company has earned, expressed either as total profit or as EPS, isn't telling us the whole story. Earnings need to be related to the amount of capital being used to achieve that profit. I really get frustrated when I see headlines like ‘Customers of XYZ Bank Outraged after Record $1.2 Billion Profit'.
All the tongues start clicking and the fingers start pointing. ‘The banks are making too much money', I hear people say. I say if XYZ Bank has shareholders' equity of $8 billion then the profit figure is totally reasonable. But if it's returning $1.2 billion on shareholders' equity of $16 billion I'd be telling shareholders to find another place to invest their money, because the bank's ‘record profit' is too low. Profits should always be related to a capital base when judging their appropriateness.
The problem is, buying stocks with growing EPS seems intuitively to be a sensible thing to do. But sometimes it can be the wrong thing to do, just as it can be wrong to select a stock solely because it has a low PE. For example, earnings will grow every year in a neglected bank account even if the bank is paying a meagre 1 per cent interest per year, but that doesn't mean you'd rejoice in the result.
Let's explore when EPS growth is a good thing and when it isn't. Table 13.1 (reproduced in table 14.1) will assist in the discussion. Remember we have assumed a required rate of return of 10 per cent.
Look at the top half of the table — that is, the top nine rows where the ROE is 9 per cent or less. Note two things. Firstly, as the ROE increases (moving down the table in any column) so too does the PE ratio. Secondly, as you move your eye from left to right within any row on the top half of the table, the PE falls as earnings growth increases. For many people this would seem to be counterintuitive. The price being placed on the same current earnings is becoming progressively lower as the rate of earnings growth increases. It's demonstrating that earnings growth can be a bad thing.
Let's take another tack in this discussion. All other things being equal, how does a company grow earnings? It does so in lots of ways. A gold-mining company could uncover a rich new seam of gold, or an oil and gas company could find a huge new oil reserve. A biotech company could discover a cure for the common cold, or a computer hardware company could revolutionise the microchip.
Table 14.1: fair PE for different return on equity and earnings growth
It all sounds very exciting. But for most mature companies, earnings growth is typically delivered in a more pedestrian manner. These companies simply retain a proportion of each year's profit and reinvest it in more of the same business activities they've been undertaking for years.
Following this logic, the reason investors are prepared to invest a dollar is because they hope to see more than a dollar delivered back in the future. They should feel the same about the dollars the company is retaining and reinvesting in the business. After all, as shareholders they own the business, so the money being reinvested by management is theirs. Integral to this process is what management actually does with the money it retains; it's one of the factors that determines the rate of future earnings growth. For example, assuming all else remains constant, a company that is currently achieving an ROE of 5 per cent, and will continue to do so, needs to retain and reinvest every dollar of earnings to achieve a target of 5 per cent growth in earnings per share. Nothing will be left over to pay a dividend. Using a purely mathematical argument, if it were to pay out half the profit as a dividend it could achieve only a 2.5 per cent growth in earnings per share the following year. We can express the earnings growth rate with a simple equation:
But think about it. If your aim as an investor is to achieve a minimum return of 10 per cent and management can achieve only 5 per cent on funds it invests on your behalf, you'd prefer they didn't. You'd prefer all profits be paid out as a dividend to provide you with the chance of obtaining a 10 per cent return in some other form of investment. This is why, in table 14.1, progressively lower PEs are awarded to companies delivering progressively higher earnings growth but at returns on equity lower than you demand (in this case 10 per cent). They are effectively destroying your wealth.
Refer back to the table and look at the row in the middle with all the 10s. This is where the ROE is 10 per cent, the rate you've chosen as your minimum required rate of return. At 10 per cent you don't care how much growth is being achieved — any level of earnings growth being generated from the reinvestment of retained earnings is just meeting your target. Therefore the price you place on these earnings (the ‘P') keeps step with the increase in the earnings (the ‘E'). So the PE ratio doesn't change with increased rates of earnings growth.
Finally, look at the bottom half of the table. Here the ROE is greater than your required rate of return. For any ROE, as earnings growth increases so too does the PE. But remember that in our perfect world every PE on the table represents fair value. As an investor you are indifferent between them if you have to pay that price to own them.
WHY MANAGEMENTS RETAIN PROFITS DESPITE DELIVERING A POOR ROE
You'd think boards and senior managers would understand the above financial logic even better than shareholders do — and probably they do. Yet the managements of many companies, with histories of delivering low ROE, still choose to retain significant proportions of annual profits. Why?
Consider a board meeting of a hypothetical company. It's bristling with managerial self-interest. One director kicks off the meeting by suggesting that since they're devoid of profitable new investment opportunities it would be best to pay out a larger dividend to shareholders this year. The others think that's a stupid idea. The directors have been granted copious volumes of options as part of their remuneration package, and their options become more valuable as the company's share price increases. By retaining earnings they have a better chance of the share price increasing in absolute terms.
They also know that investors like to see growth in earnings per share, something they also have a better chance of delivering if they retain earnings. After all, growing earnings per share looks good in the glossy graphs that adorn the first few pages of each year's annual report. If they pay out too much of the profit it lessens the chance that these ‘desirable outcomes' will be achieved. Then the chairman settles the discussion: ‘Hey team, I have to stand up in front of the shareholders at the AGM next year and tell them earnings are still growing. So I need as much of that money as we can hang on to. Reinvestment at any rate will help boost next year's earnings.' Then he turns to the director who first suggested that there were no suitable high-return investment opportunities and says: ‘It seems you're not aware that the company's acquisitions team is looking at a couple of takeover targets at the moment. The returns they're offering aren't great but the CEO and I want this company to get bigger. So let's hang on to the cash. We might need it down the track if the acquisitions team come up with something.'
None of the above discussion is in the shareholders' best interests, but unfortunately dividend policy is not always based on maximising shareholder wealth. The concept of companies having a better chance of delivering earnings growth by reinvesting retained earnings has long been appreciated. This from more than one hundred years ago:
Rich men whose income is in excess of their wants, can afford to forgo something in the way of yearly return for the sake of a strong prospect of appreciation in value. Such men naturally buy bank and trust-company stocks, whose general characteristic is a small return upon the money invested, but a strong likelihood of appreciation in value. This is owing to the general practice of well-regulated banks to distribute only about half their earnings in dividends and to credit the rest to surplus, thus insuring a steady rise in the book value of the stock.58
Considered purely in mathematical terms it makes sense that the reinvestment of retained profits results in a larger book value and enhanced earnings down the track. And this belief forms the bedrock of most valuation formulae, as I'll discuss later. But that's mathematics. The question remains: does this actually play out in real life? Surprisingly, there are studies that have shown that higher profit retention doesn't automatically lead to larger profits down the track. Let's take a look.
A study undertaken by Robert Arnott and Clifford Asness demonstrated a positive relationship between higher payout ratios and higher subsequent 10-year real earnings.59 A later study by Ping Zhou and William Ruland came to a similar conclusion.60
These findings defy intuition. What's going on?
The researchers argue that dividend policy doesn't comply with our neat mathematical rules. In the real world, dividend policy is used as an instrument of wealth distribution (unfortunately not always to shareholders!); it is not always used as an instrument of wealth creation, as it should be. This subtlety escapes most investors. The reality is that dividend policy is influenced by many conflicting factors. I introduced a couple of these in the hypothetical board discussion above, but let's look at a couple more here.
Firstly, managements prefer not to reduce the annual dividend in any year of reduced profit. Their fear is that a lower dividend will send out a negative signal to the market about the company's future prospects. To keep the dividend steady they need to increase the payout ratio when the profit is down. This typically occurs at low points in the economic cycle, times when it might be better for management to retain profits for reinvestment. Even so, as the economy recovers, typically so do company profits. This presents a scenario where profit growth follows a high payout period, not low.
Secondly, managements can at times be accused of ‘empire building' at the expense of the shareholders' best interests. Dividend policy is dictated by their focus on retaining earnings to accumulate a war chest of capital so they can acquire target companies. Ben Graham referred to this as ‘personal aggrandisement'. Acquisition activity often reaches its peak when stock markets and assets are overpriced, which means management is retaining and reinvesting profits at the high point of the cycle, a time when it might be better to distribute them. Acquisitions are made at inflated prices, destroying shareholder wealth, so retained profits aren't converted into higher future earnings. Instead higher retention rates are followed by periods of lower profit.
As discussed, dividend policy can be further distorted when options are used as a form of management compensation. If management hold call options and/or their compensation is linked to an increase in the company's share price or earnings per share, there's a strong incentive for them to retain profits rather than pay them out as dividends. This might run counter to the best interests of other shareholders, particularly when the retained profits are invested at low rates of return.
Another factor influencing dividend policy is tax. In Australia the impost of double taxation on company earnings was removed in 1987 by the Hawke–Keating government. Since then, tax already paid by the company is added back to the dividends received by the shareholder and this ‘grossed up' amount is taxed at the shareholder's marginal personal tax rate. This system of dividend imputation is fair but not enjoyed universally. Double taxation is one of the reasons lower dividend payout rates are seen in a number of countries around the world. This driver of profit retention is independent of future profit generation opportunities.
Chapter summary
- If profit is to have any real meaning, it should be related to the amount of capital employed to generate it.
- Earnings growth is undesirable when it's being delivered by the investment of capital at rates of return lower than an investor's required rate of return. Earnings growth is desirable when it's delivered by the investment of capital at a rate of return equal to or higher than an investor's required rate of return.
- A simplistic, formula-based determination of earnings growth is earnings growth = ROE × earnings retention rate. However, in real life the retention and reinvestment of earnings doesn't always deliver an increase in future earnings.
- Dividend policy is typically used as an instrument of wealth distribution rather than of wealth creation.