Buybacks shouldn’t matter, but they do

In my first post every year, I emphasise that I won’t write about theories by Nobel Prize-winning economists, but instead focus on what happens in the real world. One case in point (and one of my favourite peeves) is the capital allocation theory of Modigliani and Miller.

Franco Modigliani and Merton Miller developed their theory in the 1950s, and Modigliani won the 1985 Nobel Prize in Economics for this and other theories he developed in his career. I’ll keep things short, but assuming that markets are frictionless, investors are neutral, and dividends and capital gains are taxed the same, the managers of a company should be indifferent between paying dividends and buying back their own shares. And indeed, if a company chooses to buy back shares instead of paying dividends, it should not increase the share price of the company because while the number of shares in circulation is reduced, the company also has less cash due to the buyback.

In the world of economic textbooks, these two forces exactly offset each other, but as Victor Haghani and James White from Elm Partners have neatly summarised, when it comes to Modigliani-Miller, the adage holds true that in theory, theory and practice are the same, in practice, they are not.

Gabaix and Koijen estimate empirically that share prices in the US rise $5 for every $1 in buyback. That seems large to me. Meanwhile, a good rule of thumb developed by Grinold and Kahn is that share prices react to buybacks as a function of the daily volume traded. The price impact is roughly the daily volatility of the stock multiplied by the square root of the share of daily liquidity bought back. Since most stocks have a daily volatility of around 1%, a buyback programme that buys back 1% of daily volume for 100 days will increase the share price by 1% x SQRT(0.01) = 0.1% per day for 100 days. That is a large 10.5% at the end of the 100-day buyback programme.

Haghani and White created their own estimate:

In a world with two assets – cash and equities – how can investors in aggregate sell some of their holdings back to companies, and at the same time, keep the fraction of their portfolio allocated to equities constant? The price of equities must go up!

“Consider an investor holding $50 in equities and $50 in cash who wants to maintain this 50/50 asset allocation. If he sells 3% of his equity holding, he’ll then have $51.50 in cash. He’ll need $51.50 in equity holdings to stay at a 50/50 allocation. His equity holdings will have to increase in value by 6% to be worth $53.00, so that after he sells $1.50 into the buyback, he’ll be left with the desired $51.50 of equities. So a 3% buyback will cause the whole market to rally by 6% if all investors want to keep a constant asset allocation of 50/50. The impact is greater if investors want to maintain a higher allocation to equities, and lower for lower desired allocations. If investors want to maintain 75% in equities, then buybacks of 3% of the market would require the market to go up 12%!”

Whatever the ratio between buyback volume and share price reaction in practice, it is never zero. Which is why buybacks make so much sense for a company that wants to lift a struggling share price.