The Halo Effect

I blogged briefly about The Halo Effect the other day. Phil Rosenzweig gives a clear analysis of the delusions that undermine the credibility of a great deal of popular management writing. He debunks the In Search of Excellence fad in which people claim to unravel a simple winning formula for business success.

Personally I’ve always disliked In Search of Excellence and generally get irritable when people bang on about excellence in organisations. I’ve had a lot more time for Jim Collins’ Good to Great. which struck me as a much more solid effort that took effective pot shots at a lot of ego-based gimmickry in business. However, Rosenzweig raises a series of well-argued objections to the Collins methodology.

Overall, he identifes nine common flaws, some or all of which tend to colour efforts to explain the factors that cause businesses to succeed.

Here’s how Craig Howe summarises them in his Amazon review.

1. The Halo Effect – many performance drivers are simply attributions based on prior performance.

2. The Delusion of Correlation and Causality – Two things may be correlated but we may not know which is the base cause.

3. The Delusion of Single Explanations – Many explanations are highly correlated; the effect of each one is usually less than suggested.

4. The Delusion of Connecting the Winning Dots – It is difficult to isolate the reasons for success. There is no way of comparing them with less successful companies.

5. The Delusion of Rigorous Research – If the data are not good, it does not matter how sophisticated the research methods appear to be.

6. The Delusion of Lasting Success – Almost all high-performing companies regress over time.

7. The Delusion of Absolute Performance – Company performance is relative, not absolute.

8. The Delusion of the Wrong End of the Stick – Highly-focused companies are often successful; yet highly-focused companies are not all successful.

9. The Delusion of Organizational Physics – Despite our quest for certainty and order, company performance does not obey the laws of nature and science.

For example, we all see those surveys that show places that are great to work also give great performance. But because of the Halo effect, people tend to believe high performing companies do have great practices, so any research based on surveys of people’s judgements will be skewed by this. And of course, since nothing succeeds like success, it’s quite possible that great practices are actually funded by high performance, not the other way round. And I’ve only touched on the first two of the nine delusions. You’d need to read it to really appreciate the quality of thought.

Rosenzweig does identify two pieces of research that he believes are rigorous enough to mitigate these all-too-human cognitive biases. The first, done by MIT, investigated whether company performance was affected by the CEO’s personal management style. You’ll have to read the book for details of how this was done, but it wasn’t based on reading what newspapers said or asking managers to give style ratings. The result?

Bertrand and Schoar found that individual managers indeed have preferred personal styles when it comes to investment and financial policies, and that these preferences explain about 4 percent of the variance of company performance… That’s a statistically significant finding but it’s hardly a seductive story. You won’t excite many managers by saying: If you do these things, all else equal, you might improve company performance by about 4 per cent.. But rigorous science doesn’t always lead to a riveting story.

In the second example, folks from McKinsey and the LSE test the correlation between specific management practices and company performance, again avoiding the Halo effect. The result: a company that adopted best practices (let’s not get into what that means) across the boards tended to outperform a laggard by 10 percent.

Essentially, the thorough research has no sizzle. Rosenzweig talks about two worlds; one which speaks to practising managers and trades in stories, the other which aims for rigour but, er, doesn’t produce many gurus like Jim Collins who gets up to $150k for a speech.

Where do we go from here? Essentially, Rosenzweig argues that managers need to acknowledge that while stories are fine, in the real world they’re dealing with constant uncertainty. It’s a position very similar to that of Richard Farson, who I’ve blogged about before. In my work, I see a lot of folks who essentially have a circle of concern way outside their circle of influence, partly because of the kind of excellence-fetishism that management porn creates. It’s one of the reasons I tend to recoil in horror at the heavily diagrammed programmes for change in organisations.

I suppose the quality in Good to Great that I appreciated was its emphasis on a certain kind of humilty, although Rosenzweig rightly nails Jim Collins for trying to uncover the “physics of business”. It does seem to me that so many of the brittle “truths” of business are really highly subjective and we’d do well to realise that we really only get to manage small parts of a much bigger more complex system. And one of things we need to manage is all the cognitive biases that make us look for a great leader (or think that it’s us).

Anyway, it’s a great book, I’d encourage to take a closer look.

4 thoughts on “The Halo Effect

  1. Andy

    Sounds like a good read and one I’ll add to the list. Thanks for the extended post.

    I’ve just finished “Managing without Leadership” by Gabriele Lakomski. It grapples with similar sounding themes pointing out many of the holes in both cognitive theory and associated theories of leadership and organisational functioning. The tone is academic and also very readable – well worth a look.

  2. Johnnie Moore

    Andy: I just did the “look inside” on Amazon for that book and it does sound very interesting.

    The 4% figure that Rozenszweig cites is very thought-provoking: it flies in the face of received thinking (including my own) about the impact of leaders. It may well fit with the Lakomski’s contention (as I understand it) that “leadership” is really a figment of our imagination.

    I was going to talk about the fundammental attribution error in my post but decided not to complicate things. But that cognitive bias may help account for a human tendency to attribute “leadership” qualities where none exist.

  3. Jon Husband

    I like Richard Farson’s work. I think not too many people know of it.

    OTOH, I am not cazy about Good To Greatreat, for reasons I won’t bore you with here. Maybe the next time we have a drink, or in a Skyeversation.

    There’s an interesting book on leadership I read a long time ago that popped into my memory today … Artists, Craftsmen and Technocrats – The Dreams, Realities and Illusions of Leadership(Patricia Pitcher)


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.