Links for 21 February 2012

First, a wonderful post by John Cook, on how to deal with singularities when integrating functions.

Second, a post in which Chris Dillow asks whether Management matters. He describes a study of medium-sized manufacturers which seems to say it does. He doesn't agree, because his idea of what "management" means is different.

The practices it identifies are largely about whether there are good feedback mechanisms in place. Is the production process sufficiently well monitored that errors can be eliminated and efficiencies identified? Is there good performance appraisal of employees? Are goals clear and sensible? And so on.

What we have here, then, is not a story about CEO’s “strategic vision”, or about the power of great individuals, but about day-to-day administrative structures. Common sense says these must matter, and must be basic good practice for any firm.

I think he is right, but there could be even more to this. Few things are so disruptive of good administration as management which wants to be doing things all the time, trampling their way through the organization like a herd of elephants cavorting in a field, shaking things up in reorganization after reorganization. Thus, the fact that these organizations have good practices may mean simply that they are lead by sensible people who know to put good systems in place and generally stay out of the way. In any case, I am entirely in favor of calling the things which we do "administration" rather than "management", as Joel Spolsky puts it in this recent essay.

The “management team” isn’t the “decision making” team. It’s a support function. You may want to call them administration instead of management, which will keep them from getting too big for their britches.

Of course, maybe it is simply selection bias. There are times when management do have to shake things up if the firm is to have a chance of returning to good health, and maybe these firms have been doing well for other reasons altogether, and their systems have had a chance to mature. We see that firms which have good systems seem to be doing well, and we assume that the one caused the other.

Um, but there is this awkward point

However, Bloom and colleagues find that less than one fifth of the variance in total factor productivity can be explained by these management practices. I’m surprised by how low this is.

Indeed.

Third, another old post by Dillow, in which he asks whether politics should be based on morality.

My answer would be that it will be whether we want it to be or not, since morality seems to be a basic drive in human beings. We cannot help it. This is not always a bad thing but it can be, especially in a democracy where the median voter will not be unduly discomfited if he gives in to his taste for revenge, moral outrage, and spasms of maudlin sentimentality. I find Jonathan Haidt's work quite persuasive.

How much of moral thinking is innate? Haidt sees morality as a "social construction" that varies by time and place. We all live in a "web of shared meanings and values" that become our moral matrix, he writes, and these matrices form what Haidt, quoting the science-fiction writer William Gibson, likens to "a consensual hallucination." But all humans graft their moralities on psychological systems that evolved to serve various needs, like caring for families and punishing cheaters. Building on ideas from the anthropologist Richard Shweder, Haidt and his colleagues synthesize anthropology, evolutionary theory, and psychology to propose six innate moral foundations: care/harm, fairness/cheating, liberty/oppression, loyalty/betrayal, authority/subversion, and sanctity/degradation.

Chris Dillow feels that we should try to keep morality out of politics, and offers excellent reasons to do so. I believe that what he wants is desirable, but impossible. As he points out in his post, we should be looking at structures and asking why things turn out the way they do. I fear that his post is thought provoking but fundamentally similar to the moralistic thinking he deplores.