Balance of payments

The PE ratio comes up short for many analysts as they seek to determine returns, writes Jake van der Kamp

PUBLISHED : Monday, 30 April, 2012, 12:00am
UPDATED : Monday, 30 April, 2012, 12:00am


It used to be that the investment measure known as the price/earnings ratio, normally referred to only as the 'PE', was the guiding benchmark of investment valuation. You bought stocks that had low PE ratios and shunned those with high PE ratios, except of course if those high PE stocks also showed stellar earnings growth.

Except, except, except. This is exactly what happened to the PE ratio. There are too many exceptions. It does not make a good benchmark all on its own. It is why investment analysts look for other definitions of earnings or returns on investment. The PE ratio gives you only a broad-brush picture, and even then can be distorted.

Start with the difficulty that when considering whether to buy a stock you cannot use the overall net profit of a company for the PE ratio. You are not buying an entire company. You are buying shares in a company. You therefore need to calculate what the earnings per share of the company are. This requires that you have a figure for share capital so that you can divide the profit by the number of shares in issue.

It seems straightforward, but it isn't. We shall do this simply. Take, for example, Company A, which in Year 1 has 100 shares in issue and makes a net profit of $100. We shall thus make its earnings per share $1.

Now assume that on Day 1 of Year 2 Company A buys Company B and pays for it by issuing 100 new shares. We now have 200 shares in issue but, because Company B is just as good at its business as Company A, it also makes a net profit of $100 in the year. At the end of the year the enlarged Company A has $200 profit for 200 shares in issue. The earnings per share therefore remain $1.

But now let's take the case where Company A buys Company B on the last day of Year 2 and pays for it in the same way by issuing 100 new shares. We now have something different. Company B's earnings for 364 of the 365 days in that year are attributable to its original owners. Company A does not get that money.

At the end of the year, however, right on the last day of that year, Company A suddenly has 200 shares in issue. We now have only $100 in the year's profit (Company A's own) divided by 200 shares and the earnings per share is only 50 cents, where the previous year it had been $1. This appears to represent a serious diminution in profitability. Yet Company A is in no way weaker or more poorly performing than it was before.

The anomaly can only be addressed by taking not the issued share capital at the end of the year but the weighted average number of shares in issue during the year.

Thus, to draw yet another scenario, if Company A buys Company B exactly midway through the year, then we give Company A $50 of Company B's $100 profit for the year and, for the purpose of calculating earnings per share, we take only 50 of the 100 new shares that Company A issued.

We now have $150 profit for a weighted average of 150 shares in issue and we are back where we want to be at an earnings per share figure of $1.

This exercise has to be performed for a range of company activities. Another prominent one, for instance, is weighted average adjustments done for shares issued under share option offers made to key members of a company's staff.

The crucial point about these exercises is that few analysts making projections on future profits ever get the weighted average share capital numbers right. On what date did the company issue the new shares? Errors easily creep into such guesswork and have a direct impact on earnings per share calculations.

In my own experience as an investment analyst, I have seen colleagues frequently get even historical earnings per share figures wrong, although they are stated by the companies in their accounts.

And then we have the question of what is a profit in the first place. It may seem an easy question to answer. You buy something for $100, you sell it for $120, and you have a $20 profit. Ask an accountant, however, whether it is really so simple. He will tell you that this description is a cash flow statement, not a profit, and even then too simple. Did you catch the MTR to a different part of town to sell your 'something'? Deduct that MTR fare from your supposed profit then. It was an additional cost.

There are many more questions to ask when calculating a profit, and most of them are much more involved. There is, for instance, the matter of depreciation and amortisation. If the 'something' was bulky and had to be taken to the buyer by a delivery van, you will have to make a deduction for the purchase cost of the van.

Let's say you assume that the delivery van will last five years before it breaks irretrievably down. You must now deduct a fifth of the van's cost every year from your profits. You must also deduct the van's operating costs every year.

But can you be sure that these vans will last five years? What if you have hired bad drivers who like to pretend they are Formula 1 stars and drive these vans into wrecks in just two years? Can you still amortise them over five years? How long should the amortisation period then be? The answer to this question can make a significant difference to earnings.

It is for such reasons that auditors never say of a set of company accounts that they give you an absolutely accurate record of the company's doings. The most they will say above their signatures is that the accounts represent 'a fair view'. It is the most they can do. There is invariably an element of judgment involved.

All of this assumes that the auditors have done their work faithfully. It is only rarely they have not. False audits can ruin an auditor's name, and most of the times that you see a set of accounts issued by a company already listed, you can accept what you are told.

But it is not always so with initial public offerings. The existing owners of these companies set a great deal of store by getting as high a price as they can, and they are not always above finding less reputable accountants to help them with prospectuses that stretch things past 'a fair view'.

A notable recent example was a mainland company that declared its forest operations had been profitable over the previous three years when, in fact, no transactions had taken place.

The accountants simply took an inflated value of the forest assets and produced profit figures that would have resulted if timber had been sold at such values. It was done because the company needed a three-year profit history to list itself on the market.

Hong Kong's listing committee actually accepted this subterfuge because it carried an accountant's signature.

Measures of profitability in such cases become so wholly unreliable that other means of evaluating a company's earnings prospects must be found. Among them are such things as enterprise value, the originators of which attempted to copyright their calculations, and EBITDA, earnings before interest, tax, depreciation and amortisation.

Greater attention has also come to be placed on price-to-cash flow calculations and even dividend discount models, an older form of valuation current when people set greater store by dividends and dividends were steadier.

Of course there also are measures that do not take earnings into account, such as price to book, which measures the share price to the balance sheet value per share of the company's assets. People also look at measures that do not compare earnings directly to share prices. An example is return on equity, a percentage calculation of earnings per share to shareholders book equity per share.

But earnings measures are still vitally important, as they are the ones that tell you how much you get back from your investment. What most professional investors look at can be summarised as ROIC - WACC. This stands for return on invested capital less the weighted average cost of capital.

First you calculate a percentage figure for what your money costs you, including not only an interest rate on borrowed money but for your own capital as well. You don't treat your own money as having no value, after all. Put it all together and you have WACC - weighted average cost of capital.

You then ask yourself what your preferred definition of the profits generated by your investment are as a percentage of all the money, borrowed and own capital, that you put into them. That's ROIC - return on invested capital.

If ROIC is higher than WACC, then your investment is generating new wealth for you. If it is lower it is destroying your wealth. With variations and many quibbles, this is a calculation almost all professional investors use.

The trouble is, of course, that it is also known to the accountants and investment relations officers of any company that values its existence only by a rising share price. They then do their best to state their profits in such ways that most people's figure for ROIC is pushed up. Gradually the value of the ROIC - WACC calculation is eroded and professional investors must sharpen new tools.

But this only brings us back to our first question. What happened to the good old price/earnings ratio is that no definition of earnings has ever been definite and no investment measure that relies on earnings can count on being the same today that it was yesterday.

What it boils down to is that you have to employ a little thought when making investments, that you must rely heavily on your own brains when staking your own money. Call this the cardinal principle of investment and you are already better off than any investment ratio will make you.