Daily Archives: January 21, 2010

Strongly held assumptions resist data!

Planet Money yet again provided a great example of how preconceived notions resist contrary information.  First a bit on context, although the details of the episode aren’t totally relevant to the point I’m making here.  The episode focuses on the idea of taxing ‘Cadillac’ health care plans and they get an economist from MIT to argue for the proposition saying that the evidence is that such a plan does not negatively impact upon public health, controls cost and will likely lead in increased wages.

The sound engineer objects strenuously.

Madcap hilarity ensues as both try to convince each other by using criteria the other rejects.

The engineer relies on personal, anecdotal evidence (“I’ve seen this…I know someone that…”) while the economist relies on data (‘We’ve got X number of studies that say Y’).  There’s no convincing going on here and it reminds me (whoa…name dropping alert) of Kierkegaard’s  Either/Or (eh…it’s not that impressive, I only know about it because I heard a summary of the work via the Teaching Company) where he says everyone has to make a decision between the religious and aesthetic life and there’s no criteria upon which to base the decision which doesn’t presuppose the answer.

This is the same situation.  The economist could have cited studies until he was blue in the face and the engineer would still say:  “Yeah, but I don’t know anyone like that.”

I suspect this happens a lot more than we realize, even in the intelligence field and the problem lies in the fact that the protagonists don’t set the ground rules and aren’t clear what they’re talking about.  They both value different aspects of the issue their studying and what metrics to use to define success.  Of course I suspect any attempt to reach an agreement on what to talk about would result in it’s own dispute but at least these guys would know what they were arguing about rather than having to have a proxy fight about something else.

The other interesting point is one that was discussed by Richards Heuer when he says:

  • Mind-sets tend to be quick to form but resistant to change.
  • New information is assimilated to existing images.

Good examples of both of those here.


Economic heresy

Planet Money had a couple of podcasts about the economy of Denmark recently and presented some interesting facts:

Denmark has one of the lowest poverty rates in the world and lowest income disparities. At one point the jobs picture was so good unemployment dropped below 2 percent, something some economists would tell you should be impossible. Despite having the highest taxes in the world, economist Ove Pedersen says Denmark’s economy has grown steadily. Also, Danes are the happiest people in the world, at least according to one analysis.

‘Yeah’ I hear you say, ‘But they’re backs are breaking under the strain of living under this tax crazy, soul-sucking socialist hell.’  Well, what do the young people (the ones expected to pay the most while getting the least out of this system) think about this?  According to one student who was asked about how she felt paying such high taxes:

“I think it is terrific. . . . I get a little bit angry because constantly in Denmark there’s this talk that we have to lower the taxes, lower the taxes, lower the taxes. And I can only say I’m very young, I am only 25 years old, and already the system has provided me with a great education and help whenever I need it. I have been able to go the library whenever I needed it. I have not been to the hospital many times in my life, but when I have been it has not been a problem. I mean, I think we are so privileged that it is so wrong to attack this system.”

These podcasts are the basis of a really interesting post (and the point of this post) on the Baseline Scenario that talks about out (American) assumptions about capitalism and what economic system is the most efficient and productive.  He discusses this and expands upon two arguments:

  1. First, it’s not obviously true that the more free your markets, the faster you grow.
  2. …it stretches plausibility to argue that more free markets and smaller government always lead to more growth…

What I really like about all these posts is how they present a whole different view of reality from what we’re used to AND a demonstration of how we can internalize assumptions without even being aware of it.

From Baseline Scenario again (this time following up from the student’s comment):

Obviously not very many Americans feel this way; since the Reagan Revolution if not the 1970s, there has been an increasingly widespread belief that government spending is wasteful, and therefore people want to hold onto their money. But there’s nothing irrational or bizarre about thinking that high taxes and high benefits are good, and you don’t have to agree with her to see that.

So I think it’s telling that on an issue as basic as this–is small government necessary for economic growth–they have internalized the economic orthodoxy so completely that Denmark becomes “Bizarroland.”

I like Planet Money but they do clearly have an intellectual framework within which they base their reporting.  When they bring in people with contrary opinions it’s clear that they are introducing ideas that are contrary to (their) conventional wisdom and (at least to me) comes across as therefore less than credible.  (I really could develop this more as I can’t point to specifics here but rather just a feeling I get as I listen to their podcasts.  I haven’t really dug into this question enough to pinpoint what it is about their reporting that gives me that impression.)

Intelligence agencies struggle with the very same issues except they have an added layer of difficulty in that:

  1. fewer people get to see the assumptions which drive the analysis because of security
  2. most of the producers and consumers of intelligence share a worldview, if not an organizational culture, which predisposes them to not just view information through a particular lens but also in terms of what information is deemed worthy of viewing.

So, how to overcome those biases?  Well, there are the usual platitudes of ‘include more people’, ‘search for more sources’, etc. but it’s hard (impossible?) to systematize those things.  Likewise, it’s just not practical to subject every piece of analysis to an academic style peer-review.  I suspect there is no perfect answer but I’d recommend a couple of easy steps

  1. reducing products to the lowest classification possible (nothing new here)
  2. Actively develop ‘push’ and ‘pull’ distribution systems (beyond individual products to the shop/agency level).  This needs to be more than the occasional email or period homepage update.  If you aren’t already the ‘go to’ agency (any regardless of what you think, you probably aren’t) ‘marketing’ of your products needs to be more than an afterthought.  Your brilliant product isn’t worth jack if nobody knows about it.
  3. An open forum for continuing conversations on the product.  Paper/pdf/powerpoint/etc. products need to be heading in the direction of the dinosaur or, at least, should be very rare indeed.  For products to maintain their usefulness, they need to not just have the ability to be continuously updated but debated within the community (and maybe even outside when circumstances allow).