Dividends vs. Buybacks

One final thought from the Jeremy Siegel article I referenced in my previous post - Siegel equates dividends and stock buybacks.  This is important, because in his quasi-mea culpa, The Future for Investors in which he clarified his view from Stocks for the Long Run, he observed that - geez, much of the return of stocks is actually based on dividends.

Dividends, once upon a time used to offer stock investors higher present yields than bonds, because investors saw dividends as far riskier, since unlike bond covenants, there was no obligation of the company to declare one, whereas bonds had fixed payments according to the covenants in the issues.  The low multiples required to offer 4-7% yields  - normal sums for much of the period Siegel studies, is a big reason why returns have been so strong since 1926, the starting point for the S&P and the oft-cited Ibbotson stock return analysis began.

Dividends today generally still offer lower yields than bonds (though with ultra-low bond yields, dividends are now becoming attractive for income investors).  Nevertheless, few firms yield above 4%, leaving the investor wondering if in fact historical returns are even possible in this market.  The lower yield should cause Siegel to argue for lower earnings expectation go forward.  Instead, Siegel argues that while dividends are lower, companies are instead repurchasing stock, which is also a cash distribution to shareholders and one which drives higher the (tax deferrable) gains in stock prices.  Or does it?

I would argue that the apparent equivalence of dividends and buybacks is a case of sensible accounting misleading investors.  The apparent equivalence is in the accounting treatment of a buyback and a dividend in terms of the effect on the equity account of a company.  On the surface, both moves are a form of cash distribution from the enterprise. This overlooks the fact that a dividend does not direct the use of the shareholder's money, whereas a buyback most certainly does.  Instead of seeing these as alternative strategies for reducing the capital account, dividends should be seen as such, but buybacks should better be seen as investments, because buybacks only make sense if the asset being purchased (the company's own stock) is cheap.

Quick accounting review - when a company declares a dividend, there are the usual credits and debits - the retained earnings, a liability (credit) account, is debited and the dividends payable account - also a liability and credit account is credited.  Debiting a credit account reduces the account, and crediting a credit account increases it, so the net liabilities of the company remain the same, only retained earnings declines and dividends payable increases.  When the dividend is paid, the div payable account is debited and cash, an asset account - a debit account - is credited, reducing both the payable and the cash.

When a buyback is conducted, the capital account - that is, shareholders equity - will aslo be reduced, but in quite a different way.  Retained earnings is not affected (unless there is a subsequent retirement of the shares).  Cash, a debit account is credited, and the offsetting debit comes in the form of a debit to the treasury shares account, a liability and credit account.  The debit makes the treasury shares a negative value on the balance sheet (which you regularly see if you look at a company with large amounts of treasury shares, they are always carried "at cost" as a negative value which reduces equity).

It is no accident that like other investments under the cost method, the value on the books is held at cost.  This is because in actual practice, the company is making an investment in its own income stream.  Buying out other equity partners only makes sense if the value you would pay is less than the future cash flows of the shares that have been repurchased, properly discounted.

Companies should buy their own shares, if and only if, other investment opportunities (in physical plant, acquisitions, etc) are unattractive and the shares are trading at a big discount to intrinsic value.  Buffett, believing the intrinsic value of BH to be much higher than book value, has announced an open buyback of shares when they fall close to book value.  In this way, he is honoring this principle.  If, however, shares are near intrinsic value, a firm should not repurchase, as this will only lead to more selling from investors who believe the cash from the future flows, properly discounted, will not be worth more than the money that had to be laid out to acquire the shares.  If the shares are genuinely overvalued, such as tech stocks in the 1990s, or bank stocks circa 2006, repurchasing does not return cash to shareholders - it destroys value by overpaying for a business (which just happens to be the one the shareholders own).

Ask yourself, if you held a stake in a partnership and one of the other partners offered you a chance to buy part of his stake but you thought the price was outragous, would you feel better if instead of paying from your bank account, you instead used the business' profits to buy him out?   Wouldn't you rather have your share of the profits distributed to you to invest in something else with a more attractive price tag? 

Yet overpaying is effectively what you are doing when you let management overpay for buybacks.  (That they are often the partners whose stake is being acquired should give you pause to question whether your interests as an OPMI and theirs are the same).

This is why one of the first things I look at is the equity accounts - how much stock are people repurchasing and how much are they awarding to themselves.  It is the single best indicator of management's attitude toward shareholders.  One reason I decided to buy Intel stock (and Microsoft for that matter) is that they were significant net repurchasers of stock which I believe was trading at a significant discount to intrinsic value.  Both stocks have increased in price and so I would expect repurchases to be tempered and instead for dividends to be increased at a faster rate.

You could argue that CL, which continues aggressive repurchases with the stock at all time highs, may be making a big mistake.  I concede that this is a possibility, although my own valuation of the stock puts it in low triple digits on the strength of overseas growth, where CL is gaining an increasing share of a market that is growing rapidly along with incomes.

Why the efforts to claim they are equivalent?  At one time, companies argued that it allowed investors to take advantage of tax benefits associated with the deferral of capital gains (and what were lower tax rates on any gains).  Thus it was an attempt to offer shareholders tax efficiency.  Unfortunately, for most small investors, the sums are held in accounts with special tax treatment, such as IRAs or 401(k)s, so the tax advantages are mostly lost on such investors.  In any case this was mostly a cover for a less shareholder friendly strategy.  Repurchasing stock reduces both the equity account (which, all else equal increases ROE), and also has the effect of reducing shares outstanding boosting EPS, changes in which are heavily reported.  it allows management to massage KPIs (and often to produce significant compensation awards for themselves).

For Siegel, a man who supposedly studies valuation, not to focus on this sort of difference is pathetic.  He should be out educating investors about the significant difference between these two concepts, rather than assuming that one form of capital reduction is as good as another.

I highly recommend that if you are really focused on intrinsic value, the biggest single criterion for the evaluation of management is capital allocation, and one of the best metrics available to you is their use of the capital account, particularly as it relates to distributions from the firm.  If they buy back stock in periods of high prices and low, you have to wonder what the real motivation is.