Buybacks are a Capital Structure Decision not an Investment Decision

As the stock market raced ahead of the real economy in recent years, the increasing volume of buybacks was criticized by those that felt that firms were channelling funds away from productive investment into seemingly wasteful financial engineering. Now that they have passed a peak, the decrease in buybacks is being interpreted as a signal that management believes its stock is overpriced, and therefore not worth buying. Buybacks are not deserving of this double-edged rebuke, nor the confusion that surrounds what they represent.

Buybacks are first and foremost a capital structure decision and they are much more similar to dividends than is often suggested. Furthermore, the supposed trade-off between investing to grow and investing in one’s own stock confuses the sequence of decisions in the capital allocation process and needlessly adds complication.

Buybacks and dividends both take cash from the company and give it to shareholders. Dividends treat all shareholders equally; buybacks on the other hand change the composition of the shareholder base and in doing so increase leverage only for continuing shareholders. Buybacks are better than dividends for continuing shareholders if the market value of the firm subsequently increases, which is more likely to happen if the market value is below intrinsic value. But buying back stock when the market value is below intrinsic value, though often lauded, is not necessarily a good capital allocation decision. The firm may have capital investments that would have created even more value for shareholders.

A good capital allocation process looks first at opportunities to invest in projects that are sufficiently likely to earn a return in excess of the firm’s cost of capital. The correct cost of capital to use is market implied, not the theoretical (and, almost always, wrong) one that comes from a spreadsheet in Canary Wharf or Wall Street.

Next, the process turns to how these capital investments should be funded. If funded entirely from profits (ie equity), then leverage will fall. This is good only if leverage was too high to start with. If funded with a mix of debt and retained profits, firm leverage can be held constant or adjusted to a closer-to-optimal level. Only if there is excess cash flow following this, should the firm look to return these to shareholders.

It can use dividends or buybacks to do so. Dividends are more likely to be used if there is a long term decision to return a percentage of profits. Buybacks are seen as more flexible. This is convention and need not be the case. Of course, shareholders can decide to create a synthetic dividend during a buyback or use proceeds from a dividend to buy additional stock.

Separately, the idea that buybacks signal a cheap stock is flawed. If a company’s stock is trading at a relatively low multiple, it may be that the market’s estimate of the cost of equity is high, rightly or wrongly. Or it could be that growth prospects are low. Mixing these up and using the outcome to justify buybacks is bad for shareholder value creation. A high multiple may be a sign of growth opportunities with high returns so you are less likely to see a lot of cash left over for buybacks/dividends. But the stock may still be cheap.

As stock prices go up the hurdle rate for internal projects comes down and firms channel more into investment projects, leaving less to return to shareholders. If the hurdle rate subsequently goes up (ie share prices fall) it doesn’t make the previous decision to invest a bad one, but may mean it is harder to justify investing additional cash into the same project. All projects should be NPV positive at the time of funding, decisions to increase or decrease the size of the project at a late date should similarly be NPV positive.

So, capital returns to shareholders should naturally be low when valuations are high and vice versa. The reduction in capital return that has been noted after the rise in share prices is what one should expect, all else equal. The fact that a peak in buybacks didn’t happen earlier as the market advanced, probably reflects executives being too conservative in estimating IRRs on projects and being paid for non-productive activities such as reducing the number of shares outstanding.

Buybacks are not an investment in one’s own shares; they are a means, just like dividends, of reducing the amount of shareholder equity and hence changing the capital structure of a company. Constructive discussions about the role of buybacks should take this as a starting point and, with some hope, buybacks will shake off their unjustified reputation as a dark force in the capital markets.

What Investors Don’t Get About Buybacks

Shareholders trust the management teams of companies to make sensible use of the profits generated. Investing those profits wisely can greatly increase future profitability and, hence, shareholder return. Overpaying on an expansion project, for example, will do the opposite. Recently there has been increased focus on companies using their cash stacks to buy their own shares rather than looking for good ways to invest to grow. The question I am going to address is (and it’s an important one) is Does it matter at what price a company buys its own shares?

The intuitive answer, and one that many professional investors and commentators would give, is “of course, buying at a lower price is always better”. The reasoning is that companies that buy their stock before the price goes up get more bang for their buck than those that buy before it goes down. As a result, investors scold companies that buy at or near peak levels in their share prices (this article from Fortune is one of a very large number I found online).

But it actually does not matter. In fact, I can reverse the intuitive answer with two simple arguments:

(1) The shares the company buys get cancelled. The company is not investing in its own shares. Shareholders have no claim on the shares that are bought back and so do not gain or lose if this is done at a high or low price.

(2) When a company buys back its shares, it is buying them from shareholders. Shareholders always prefer to sell shares at a high price and therefore should thank any company that buys from them at such a lofty level.

Argument (1) is pretty solid but I think it’s worth digging into (2) in more detail. Shareholders as a group should prefer the “buy high” approach, but at the individual level those that sell to the company at the peak are better off (following a subsequent fall in price) than those who choose to hang on. That the latter group did not sell at the peak is not the fault of the company or in any way linked to a buyback or lack thereof.

Here’s another way to think about it: if companies had perfect foresight on their share price would they still buy back shares today if they knew the price would be lower in a week?

This is not just a theoretical exercise since companies do have inside information that can give them a sense of the short term trajectory of their share prices.

The answer, again, is that it does not matter. Neither the shareholders nor the company is better off if the buyback is delayed. The company can buy more with a fixed pot of cash if it waits, but the total value of equity following the buyback will be the same regardless of the timing.

However – and here’s where the intuitive answer starts to look (somewhat) good again – the share price will actually be slightly higher in the case when the company delays the buyback (let me know if you want proof of this). Any shareholder who would not sell regardless of when the buyback takes place is better off if the company waits a week. But, shareholders on average will not gain from this “buy low” strategy and, if the company does not buy at today’s higher price, shareholders miss out on the option of cashing in at a better price.

The discussion should not focus on whether timing buybacks in a certain way creates shareholder value (it does not), but how the decisions of individual shareholders to sell or not seem better or worse as a result of the timing. The actions that a subset of the shareholder base might take should not determine the way in which companies divvy out the spoils, so shareholders would be better off redirecting their attention to monitoring the use of the cash that actually stays in the company.