EconResponse: Esther Dyson on the Attention Economy

This week's EconTalk is Esther Dyson on the Attention Economy and the Quantification of Everything. She is approaching two topics, one the on and offline market for attention, and then goes into quantification of things, by which she means keeping metrics on different aspects of like like community etc. I intend to start writing about EconTalk every week as it is probably my favorite thing to listen to. Unfortunately, this was not my favorite episode, but it did spur me to write about some of the things I didn't agree with, so hey, there's that!

In the attention segment, there are a few main insights. Attention is something people both intrinsically desire and intrinsically enjoy giving. Attention is also a service (though she doesn't use the word for some reason); it can not be stored or transferred by arbitrage as far as we can tell. At the tail end of this segment she and Russ discuss briefly the idea that productivity increases seem in part to allow people to do more non-measurable, non-monetized things like walk in the woods, pay attention to each other, etc. which make the measured economy look worse but make us better off. This provides a bit of a segue into the next section on measurement. I say a bit, because while Ms Dyson seems to like the idea that we are moving towards a barter economy (her words) that could not be measured easily, and seems to tacitly realize that many things we care about are neither monetizable or measurable, it isn't clear to me then why she thinks that quantification is so valuable.

She does have some good points about quantification, that as a former supply chain guy I can absolutely appreciate. Actually keeping track of who does what in written form can be a big help, and she makes a good point that we should probably do a bit more of that in our lives. My own example would be keeping track of which suppliers are late on deliveries and by how much. Our brains don't keep good track as we tend to only remember the times such issues became problems, like an order arriving two days late causing us to stop production. Often, when we keep track of things we find that some are consistently late by a day, and some are randomly late by a larger margin. Those all affect how you order and keep inventory, as the former just requires ordering a day earlier than normal, while the latter requires keeping more stock on hand. Not an exciting example, but one that costs companies millions when they get it wrong. 
One issue with her idea, besides the facts that many things that really matter are very difficult or impossible to measure, and we also don't know what we should be doing with those measurements to affect positive changes, or even what normal or good is... and many more... is the other side of the coin: we often get what we measure. In other words, once you start keeping track of something like steps walked, pounds recycled or whatever, with the understanding that more/less is better, people unconsciously start to adjust behaviors to seek making those metrics larger or smaller, even if doing so is worse over all. This is a frequent problem in business, where maybe you measure days to order fulfillment from order date. Well, there are two ways to get that: low inventory and efficient production processes, or just lots of inventory. The former is great and what you are shooting for, but the latter is expensive for the company yet easier to achieve for the workers and managers. Guess which one happens when managers are implicitly graded on the matter. See also my post about whether incentives imply the ability to pursue them; just as incentives tell our brains that we should be able to affect the outcome, metrics tell our brain what to focus on. How many economists talk about over all happiness, or even wealth, instead of GDP, just because GDP can be more easily measured?

Effectively, a good take away would be to pay more attention to what we do in general, and if something really matters, keep written records of what happens that we care about. Our brains are not naturally attuned to give us accurate measures of these things, and can be tricked in various ways, and so keeping a journal or record can really benefit us. Sites like Beeminder can really help us when it comes to such things. However, I think a broader understanding of the topic would point out that many things we care about can not be measured, and even if we can, we don't always know what to think or do about those numbers. Russ brings up this issue, but Esther does not seem comfortable discussing it. I find this to be the case with many engineers and economists I have spoken to, as well as many business managers: a very strong faith that situations can be reduced to numbers, and once we do that, we can then tweak the system to get the numbers we want. 

Russ asks an interesting question as to whether introspection happens less frequently in the iPhone age, and Esther replies (almost in my own voice) that she wonders how much happened before. I personally suspect that if we took away everyone's little toys we distract ourselves with we wouldn't end up with philosophers, but just really bored people. I often think that sort of belief implies that if you took away all my books and things I would become a great athlete (or at least a runner) which is not likely at all. I think we each have things we like to do, and seek to do such things even when we can't obviously do them, and taking away our ability to do them simply frustrates us, and doesn't drive us in different directions. I would argue people don't engage in introspection as much as we should, but I don't think anything in particular is to blame.

Another point Esther makes along those lines is that "Abraham Lincoln used to give 3 hour speeches, and who would sit through that today?" The important point, besides 'what politician gives good 3 our speeches today?' is "How do you know people sat through all of that?" Russ points out, correctly I think, that people habitually sit through long movies, perhaps with an intermission, and other things. Esther laments a bit that people are more interested in short videos than books or longer things. However, I think that she misses the larger point that there are few things good enough to spend lots of time on, but many that are good enough for a few minutes. The internet has not caused us to only like short things, but rather has allowed people with only two minutes of good story to show us that story instead of only those who have two hours of things to say. I would much rather watch three hours of random cat videos on YouTube than have to sit through Avatar again. Now I have the choice, whereas in the late 80's if I wanted to watch a movie, I was limited to the things the movie industry wanted to make.

She also talks about investing a bit, but I don't really have a lot to say about it. Arguably, she doesn't either; she says a lot of things that are general media boiler plate things I have heard dozens of times yet contain no insights (or even accuracy.)
Then space... but not why people would want to go there other than "Hey! Space!"


All in all, a relatively disappointing episode of EconTalk. Esther is not a dynamic speaker, which is forgivable, but her points and ideas are somewhat fuzzy and poorly thought out and communicated, which is less so. Perhaps the latter is due to the former, as speaking extemporaneously is never easy, even when not in front of people. However, I think many of her ideas are somewhat facile, and contradict each other on deeper reflection. She implies that we should, in the normative, it makes a better person sense, spend more time doing long things like reading or listening to three hour political speeches with little regard to what people actually like, or whether spending that much time on something long is actually ideal. Is reading a long, bad book actually better than reading 30 well written blog posts? Personally, I find much more benefit in terms of thinking from reading a handful of really insightful posts than a long, cumbersome journal article. As Tyler Cowen put it, most articles are just blog posts fluffed out to 15 pages and most books are just articles with more filler. I absolutely agree; length does not imply quality, just length. I don't think this podcast would have benefited from being longer, but maybe I just don't have the attention span.