Google Tries Selective Intervention?
| Peter Klein |
Can a large firm do everything a collection of small firms can do, and more? If not, how do we understand the limits to organization? Arrow focused on the information structure inside firms. I favor Mises’s economic calculation argument. Williamson’s preferred explanation for the limits to the firm is the impossibility of selective intervention — the idea that higher-level managers cannot credibly commit to leave lower-level managers alone, except when such selective intervention would generate joint gains. Williamson’s argument is not, however, universally embraced (or even understood the same way — see the comments to Nicolai’s post).
Google apparently sees things Williamson’s way and has formulated an explicit policy on “autonomous units” designed to address the problem. Such units ”have the freedom to run like independent startups with almost no approvals needed from HQ, ” reports TechCrunch. “For these divisions, Google is essentially a holding company that provides back end services like legal, providing office space and organizing travel, but everything else is up to the pseudo-startup.” Can it work? Insiders are doubtful. The TechCrunch reporter even frames Williamson’s thesis in this folksy way:
There’s a lie that companies and entrepreneurs tell themselves in order to commit to an acquisition.
Oh, we’re not going to change anything! We’re just going to give you more resources to do what you’ve been doing even better!
Yeah! They bought us for a reason, why would they ruin things?
It usually works for a little while, but big company bureaucracy– whether it’s HR, politics or just endless meetings– almost always creeps in. It’s a law of nature: Big companies just need certain processes to run and entrepreneurs hate those processes because they stifle nimble innovation.