Barron's/S&P Have It Wrong: Buybacks Can Be Good

Barron's reports this week, in an article entitled "Buybacks That Bite Back", that share repurchases don't produce lasting benefits for the companies that do so. The article is based on a study conducted by two analysts at Standard & Poor's. This would be interesting news if the study wasn't so flawed.

First off, the analysts look at a time frame of only 18 months, from the start of 2006 to mid 2007. Immediately you must ask yourself is this a long enough time frame to make any conclusions about "lasting benefits"? Not in my book. Second, how are they judging whether any benefits acrue? By looking at stock price performance of course. That assumes that stock price performance, over an 18 month time frame, is an accurate representation of business performance. I.e., the Efficient Market Theory lives!

Now I haven't actually read the study, because I don't want to pay for it. I am going by what Barron's says and by what the two analysts say in a video interview available on S&P's website. The crux of their argument is that companies that bought back the most stock relative to their market capitalization had the worst stock performance over this period. Yet it is astounding to me that they dare to conclude from this that buying back more stock was the causal factor of the underperformance. The term "spurious correlation" comes to mind.

In their defense, they do admit in the video that the worst performers were in sectors like homebuilders, while the best performers were in strong sectors like energy and tech. Earth shattering! Yet, they conclude that the degree of repurchase activity had a bearing on stock performance. A bit of a stretch perhaps.

They also note that in fact those companies that bought back a lot of stock did so at a premium, because at the end of the study their share prices were lower than when they started buying shares. Shockingly, they uncovered that even public companies don't have crystal balls. And all this time I thought that public companies bought back stock only when they knew it wasn't going to go down.

In further defense of the analysts, they do say in the video that one should view repurchase activity as part of the big picture, and as just one option for allocating capital. I agree. If they had listened to themselves they never would have published this study.

Here is the scoop. Buybacks are a good thing if the company in question has excess capital and is buying shares at a price that over time provides a better return on this capital than other options. If their stock subsequently goes down, they should buy more. It doesn't mean that the first purchase was wrong. I am sure that in most cases the companies who bought back a bunch of stock in 2006 did so because their stocks were getting hammered and thought it was a good use of capital. Two years from now it may prove that it was just that. Since most of these companies are in tough sectors, it's no surprise that their stocks have continued to go down, just as it is no indication that buybacks are a bad thing. Once business improves, their earnings will be spread over fewer shares. That's a good thing last time I checked.

Sure, companies that issue tons of stock options to executives and buy back shares to offset this dilution, regardless of price, are probably destroying value. It's pretty easy to look at a particular company and see whether its buybacks create or destroy value. Do they buy shares opportunistically when the stock is reasonably undervalued based on information available at the time? Has their fully-diluted share count declined significantly over time? Was it a good capital allocation decision versus other options at the time?

Studies like this that make such sweeping generalizations based on limited data, and which are then regurgitated by the press do no service to investors. Isn't it just possible here that the companies that bought back the most stock over this period did so BECAUSE their stocks were performing poorly?

Holy chicken and egg, Batman!