Having failed at acquisition, Microsoft is coming back at Yahoo another way:

According to people familiar with the matter, Microsoft has proposed to Yahoo a deal related to advertisements that run next to Internet search results, a large business that is dominated by Google Inc. The move by Microsoft appears to be an attempt to stop Google from entering a search-related deal with Yahoo that's now under discussion and could be announced in coming days.

In a statement, Microsoft said only that it has raised with Yahoo the possibility of a "transaction" that isn't an acquisition of all of Yahoo, and declined to be more specific. However, Microsoft also said in the statement that it "reserves the right to reconsider" the possibility of a bid for the company, depending on developments or talks with Yahoo or its shareholders or other parties.

The language in the statement, while vague, appears to represent a notable shift in Microsoft's stance. In recent days, Microsoft had told Yahoo representatives that it no longer intended to pursue a takeover of the company, according to people familiar with the matter.

In a statement, Yahoo said that it "has confirmed with Microsoft that it is not interested in pursuing an acquisition of all of Yahoo at this time." But Yahoo added that it remains "open to pursuing any transaction which is in the best interest of our stockholders." The company said its board will review its alternatives "including any Microsoft proposal."



Most finance literature suggests that companies do too many mergers. Many people--including corporate managers--think of a merger as a way for a company to acquire a valuable asset. This only works, however, if the other company does not know that it has a valuable asset on its hands, and its shareholders will therefore sell to you on the cheap. This does sometimes happen, of course, but more often not. That's why it's the stock of the target company, rather than the acquirer, that rises on the news of a buyout. Stupid mergers have destroyed an enormous amount of buyer's shareholder value.

Mergers should really only be undertaken in a few situations: where there is an undervalued asset the company doesn't know about (rare), when there are economies of scale to joint operations (pretty rare, and likely to be eaten up by the costs of combining operations), or when there are significant barriers to doing a deal. The most prominent case of that last is something called co-specialized assets: when doing a deal would require the supplier to invest heavily in specialized equipment to produce for the buyer.

Once you have, say, totally retooled your plant to produce widgets for Acme, Inc., you're in a kind of vulnerable position. Another changeover would be expensive, so they have a great deal of negotiating power. These kinds of problems can sometimes be resolved by contracts, but sometimes they can only be resolved by moving the firms under one roof. Proprietary information is a special case of this, where the owner of the information is vulnerable.

But managers prefer mergers to side deals, since it gives them an empire. It's only when mergers are frustrated, as now, that they start exploring the deals that should have been a first, rather than a last resort.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.