[EconTalk] Rory Sutherland on Alchemy

Rory Sutherland on Alchemy

11/11/2019 by EconTalk: Russ Roberts

Web player: https://podplayer.net/?id=86605259
Episode: http://files.libertyfund.org/econtalk/y2019/Sutherlandalchemy.mp3

Author and Advertising Executive Rory Sutherland of Ogilvy talks about his book Alchemy with EconTalk host Russ Roberts. Sutherland makes the case for the magic (yes, magic!) of advertising and branding in helping markets work well. This is a wide-ranging conversation on consumer choice, public policy, travel, real estate, and corporate decision-making using insights from behavioral economics and decades of experience in the world of advertising.

Listen Date: 2020-01-30


  • Rory Sutherland was awesome fun to listen to; with that bombastic sense of speaking coupled with the plummy English accent, and the ability to bring in examples at every stage. The downside of that is that the overall charm and persuasiveness of the speaker means that it’s that much more difficult to think about the truth or not of the arguments. I’m still not sure if I’ve done it properly.
  • The point when Rory Sutherland mentions Luca Dellanna and The Control Heuristic brings back memories of the Econ Talk episode which talks about being addicted to knowing, and I’m surprised Russ Roberts didn’t bring that up.
  • There was a bunch of talking about ergodicity and non-ergodicity without ever explaining what it was, and I’ve forgotten. Eventually, from context, I worked out that non-ergodicity probably refers to events being dependent on each other, and not the ‘every event is independent and uncorrelated.’ And Wikipedia says: “In probability theory, an ergodic dynamical system is one that, broadly speaking, has the same behavior averaged over time as averaged over the space of all the system’s states in its phase space. In physics the term implies that a system satisfies the ergodic hypothesis of thermodynamics.” Eurgh.
  • Okay, going further down the page, there’s a bit about Markov chains, which is: “In a Markov chain with a finite state space, a state i {\displaystyle i} is said to be ergodic if it is aperiodic and positive-recurrent (a state is recurrent if there is a nonzero probability of exiting the state, and the probability of an eventual return to it is 1; if the former condition is not true, then the state is “absorbing”). If all states in an irreducible Markov chain are ergodic, then the chain is said to be ergodic.

    Markov’s theorem: a Markov chain is ergodic if there is a positive probability to pass from any state to any other state in one step.” That seems slightly more comprehensible.

  • I thought this argument was insightful: “If you ask 10 people to hire one person for your company, you get an extremely different result than if you ask one person to hire 10 people. In both cases you’re going to get 10 new employees, but they’re going to be a very different mix. Explain that.” / “So it seems to me that when we make a single hiring, our instinct for variance reduction is very, very high, or blame avoidance, you might say, in a corporate setting. And so we’re going to make a very conservative hire. If you have 10 people, you’re going to go much wider and you’re going to look for complementarity rather than conformity.”
  • Just because it’s insightful, I don’t think it’s necessarily prescriptive on one side or the other, you need to be aware of whether you want the safe option or diverse options.
  • The one main takeaway from this episode, shorn of all the examples, doubt over whether it’s flamboyance or solid, the actual validity of the examples, was this line: “And it struck me as a really interesting point, which is just that we tend to be blind to the affect that choice architecture has over the choices we make.”
  • While this was playing, and the conversation was happening so rapidly that I didn’t really have the time to make connections through – I was thinking about my own decision making and choices in relation to choice architecture; and was surprised that the topic currently in the zeitgeist – the attention economy – didn’t make it to the discussion. Sure, the choice architecture can force you down certain paths. If you have attention to spare, or the ‘cognitive surplus’, you could, I guess, go chasing down rabbit holes of different options and approaches rather than let somebody else’s algorithm do the work. But to even know that there are different options requires having spent a certain amount of time obsessing about either the problem or the solution, just for the sake of the solution.
  • And in that context, we get to the previous episodes on Waldenponding vs Stillness; and if you live in a choice architecture which is not very flexible; choosing waldenponding or stillness helps you devote attention and mental energy to exploring beyond it.
  • The discussion about political correctness at comedy clubs is something which Tea With Alice used to explore in detail. Alarmingly repetitive detail at times; but even so, the way it used to go indepth over there was much more interesting than this off the cuff mention.
  • The point about choosing brands; or not switching out because you’re terrified of even small chances of bad experiences, and how that ties up with sunk cost fallacies – okay, sure, fine. But to bring in my experience – I’m now in my thirties, and wealthy. I live in abundance and not scarcity; and I’m willing to take the risk on weird brands or experiences because I don’t necessarily have that much to lose. I think that for people like that – and perhaps societies like that – it’s worth trying to overcome sunk cost bias or brand name bias.
  • I need to learn more about TRIZ.
  • I full-mouth laughed when I heard this story: “The worst case of which was of course just after we’d had young children, my wife sends me out to buy a wide slice toaster and I come back with a bread slicer arguing that what we need isn’t a wider toaster, it’s narrower bread. Now, for all sorts of reasons, this was a totally ridiculous thing to do. I mean, having two very young children around with a sort of effectively a circular saw device sitting in the kitchen was never going to be the best idea to begin with, but it does force you to do that a lot.”
  • The idea that you should redesign the train map every few years was interesting, but I’m not sure he’s considered the downsides of it.
  • My beloved seat61.com got a shout-out. And again, coming back to the earlier point about having to be slightly willing to go down rabbit holes for the sake of rabbit holes; you wouldn’t need the easy presentation of seat61 if you were an Indian kid who had grown up reading Trains at a Glance for fun. (We didn’t have all that much else to do.) In that case, you can figure out alternate connections yourself rather than relying on Google Maps’s suggestion. But you still wouldn’t know about other arbitrages, like where you could have dinner while waiting, etc etc.
  • That, I suppose, is what changes something from quantitative optimisation to narrative optimisation.
  • Narrative optimisation. I should think about that and write about that more.
  • “I hate commodities” cut a little bit too close considering what I do is make and sell a commodity.
  • I thought this was extremely insightful: that the same event (rising house prices) can lead to different responses in different consumers (sit tight and wait for more appreciation, or sell , take the gains, and move to the suburbs). And that this makes a mockery of price curves, which means that simple economics is doomed.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s