Principles: Shares Repurchases & Dividends

Criticisms on Culture of Share Repurchases

It is value accretive for a company to repurchase its shares under two conditions:

1. It has enough cash (including sensible borrowing capacity) beyond the near term needs of the business

2. Its shares are selling in the market at a price below a conservatively calculated intrinsic value

We add a caveat that shareholders should have been supplied with all the information necessary for them to estimate the intrinsic value. Otherwise, insiders could take advantage of their uninformed partners and buy out their interests at cheaper than true value.

If a company’s shares are selling well below intrinsic value, repurchases usually make sense. In the mid 1970s, it was very value accretive to do so, but very few businesses did so. These days, repurchases are in fashion, but they are generally done for the wrong reasons; either to pump up or support the share price.

Of course, the shareholder who chooses to sell benefits from having an extra buyer, regardless of their motives. However, the existing shareholder is hurt by repurchases above intrinsic value.

We could argue that it is natural for the CEO to be optimistic about their own business, but there is also evidence that companies are repurchasing shares to offset the shares issued from share-based compensation (SBC). These SBC options are granted at much lower prices when exercised.

Effectively from a capital allocation lens, management is “buying high” and “selling low”!

The decision of doing SBC and repurchases are not logically related, so they should be considered independent of each other. Just because shares have been issued when SBC options are exercised does not mean shares should be repurchased at above intrinsic value. Correspondingly, shares that are selling below intrinsic value should be repurchased regardless of SBC.

It is not that complicated when assessing the effectiveness of repurchases. When the rule of repurchasing below intrinsic value is followed, gains accrue towards existing shareholders. So then it is puzzling that repurchase announcements almost never refer to a price ceiling, above which it would not make economical sense to buy.

Why do most management seem to neglect the price that they are repurchasing at? Would they behave the same way if they were managing a private company with just a few partners? Clearly, by answering this question, we will come to the conclusion that it is rare to find management who treats their public shareholders as partners.

Blindly repurchasing shares can be value destructive, but this point is rarely communicated when Boards approve of repurchase programs. This tends to be drowned out by ever optimistic management.

Criticisms on Culture of Dividends

Shareholders receive dividends payout reports, but they are not told why the particular dividend policy is best for them. It is again rare to read about communications from management about the rationale for dividend policy.

This is an important exercise because we believe management and owners should think hard about the circumstances under which earnings should be retained or distributed.

First, we need to distinguish what we deem as “restricted earnings”. Not all earnings are created equal. Some businesses require earnings to be retained in order to maintain its economic position, for example, utilities companies continuously need to reinvest earnings into infrastructure. So for these types of businesses, even if the payout ratio is tiny, by continually distributing out earnings they will lose their long term competitive advantages.

In this discussion we focus on the more interesting “unrestricted earnings”, this is the portion that management can decide to retain or payout.

Generally, managements prefer to retain these earnings for obvious reasons such as growing the corporate empire and operating with financial comfort. However, we think that there should be one valid reason for retaining earnings: every dollar retained should at least create more than a dollar of market value. This will only happen if retained earnings produce higher incremental returns than what is available for shareholders.

Of course, this analysis should be backed by historical evidence or thoughtful analysis of the future. We are also aware that such analysis is difficult and subject to error because the reinvestment rate is not a pre-determined number, but rather it fluctuates.

Owners must make estimations on the average reinvestment rate over the near future. However, after an informed analysis is made, the decision to retain or payout dividends is easy. If the expected reinvestment rate is high, management should retain earnings, otherwise they should payout dividends.

It is interesting to note that similar to the decision of share repurchases when running a private company with a few partners, managements usually have no problem thinking rationally when asking whether subsidiaries should distribute earnings to the parent.

The problem comes when the payout decision is made to public shareholders; management have trouble putting themselves in the shoes of their shareholders.

To further compound this problem. As an outside shareholder, it is difficult to know if management is misallocating the retained earnings. This is especially true when the company has an outstanding core business that generates big earnings which can hide the failures of capital allocation elsewhere. In such cases, shareholders would be better off if retained earnings were used only for the core business and the remainder paid out as dividends.

Managements of high returns businesses who consistently deploy retained earnings into sub-par ventures with low returns should be accountable for these allocation decisions, regardless of the high profitability of the overall enterprise.

This is not suggesting that dividend policies should vary according to the fluctuating reinvestment opportunities. We understand that it’s preferable that dividends are consistent and predictable. Therefore, payouts should reflect long term expectations of both returns and earnings, since the long run expectation changes infrequently, the dividend policy should logically follow suit.

This conclusion does not negate our initial claim that management should communicate with shareholders the rationale of dividend policy and provide commentaries on how retained earnings have been deployed (or at least provide information in reports for shareholders to arrive at the verdict).