Transcription
Um, very first simple point to make, which is just worth making before I say anything else, which is that psychology is technology. Although we think of them as completely separate things, all technological advance comes from the understanding of the properties of something. Whether it be silicon or stainless steel or nitrogen or whatever, it comes from understanding a property. As you get better at understanding these properties, technology gets better. And the properties of human behavior and the human brain should not be excluded from that quest. That's all I'm saying.
I'm just the co-founder of Ogilvy Change, which is part of Ogilvy, which, as I describe it, aims to solve the problems that ad agencies have never been asked to solve. Historically, the ad agency only really got asked, it had an interesting mixture of talents within the building, but it was only asked to solve problems which involved spending money on mass media. And my example, the problem of that actually is, it actually, it never really got to the deepest understanding of human psychology because mass media tends to be a fairly blunt instrument intervention. My view is that once digital technology came along and it was possible to actually intervene in far more subtle, kind of microsurgery ways, our understanding of the subtleties of human behavior had to increase commensurately.
And, um, I, what I tend to do, I always have a huge complaint. I've got nothing against economists, okay? I've got nothing against the forces of naive rationality, except for the fact that they always get to problems first. So if you think about it, it's much easier to be fired for being irrational than it is for to be fired for being unimaginative, for example. There's a huge rationality bias that pervades everything we do. And the problem I always have is that when you, when you go to engineers, you will always get engineering problems, engineering solutions. When you go to mathematicians and economists, you will always get solutions that involve incentives, direct financial incentives. Don't mind that, those kind of incentives are fine, sometimes they work. The thing I do begrudge is that those people are effectively handed the problem on a plate. They redefine the problem in terms of their own imagined solution, and no one else gets a look in. Sometimes it would help if a greater variety of, uh, solution approaches were just brought to bear in parallel.
I have a particular rant at the moment. You probably know about the problem with the overuse of A&E in the UK. Um, my, my advertising solution, which may be mad, is to go to the BBC and ITV and the major media and say, "Can you please not call it A&E?" It used to be called Accident and Emergency, and that gave, gave you a clue as to the circumstances under which you should use that part of the hospital. Um, it might even be good if you took the advertising idea further to actually have blood and screaming coming from A&E to discourage people from going there, except in the most desperate circumstances. But actually, what you call things affects how people behave. Accident of Emergency, that's a place I really shouldn't go to unless it's pretty desperate. A&E, it's like your best mate, isn't it? Okay? It's even an entertaining American TV channel, just to add further confusion to the whole thing. Um, brilliant advertising intervention.
Beautiful, beautiful idea. At Harvard, the phrase "designated driver" was created by a team of behavioral scientists, and they encouraged soap operas, sitcoms, and so forth to inject the phrase into their plot lines because if you create a name for something, we automatically assume it's a norm. And suddenly, because there's a word for designated driver, therefore this must be a normal and common behavior. It's extraordinarily potent. And this kind of thing leads me to become increasingly fascinated in what I occasionally call "mono ideas," which are minimalist, oblique, um, non-obvious interventions. Human behavior is a, individual human behavior, never mind mass human behavior, is a complex system. In complex systems, there isn't necessarily, um, a huge connection between the scale of the intervention and the scale of the effect. Strangely, the human brain understands this in some domains. We understand baby bio. We understand that when we're dealing with plants, adding a very small amount of some trace element can have huge effects on the health of the plant. If you talk about economic problems or business problems, for whatever reason, the mental model our brain leaps to is that of kind of Newtonian mechanics, where, you know, the scale of the effect is proportionate to the scale of the intervention. I think that's a fundamental mistake. So what I'm saying is that if we can get better at understanding the properties of human decision-making, we will get better at changing human behavior for the better.
Now, what Mark spotted just before me, and I will never come up with anything as good as that in my working life, is that you had a fundamental problem, which is that when it came to new syringe, old syringe, the bad decision was easier to make than the good decision. It was, human beings are naturally lazy. It was easier to do the wrong thing than it was to do the right thing. It's also cheaper to do the wrong thing than it was to do the right thing. He changed the choice architect so it became impossible to do the wrong thing. And changing the choice architecture of things, changing the way in which we present choice to people, or in which we make choice available to people, strikes me as an absolutely fascinating area of discovery.
Now, I'm sort of, I'm not a libertarian, but one of the quibbles I also have with libertarianism is it's all very well saying that people must be free to choose, but if the choices they think of are massively limited, or the choices they're offered are hugely limited, then that freedom of choice isn't particularly valuable to us as human beings. It becomes kind of placebo choice. I don't know if you know what placebo choice is. Basically, that thing where you're given a choice to make it feel like you've got a choice, but really you haven't got a choice at all. I'll give you a few examples. "Still or sparkling?" in a restaurant. Brilliant and highly profitable phrase, because once you say "still or sparkling," it's much harder for the person to say "no tap." Okay? Um, "red or white?" would, which basically means you can get to drink a drink you may not like very much, but you get to choose what color it is. Okay? Kings, Liam said "red or white" are the three most depressing words in the English language. What you're actually doing is you're giving the appearance of giving someone a choice. So it's plain and spicy popadoms may come into the same category. I'm not sure. Um, you're giving the appearance of giving someone a choice while actually closing choice down. Because what you're really saying when you say "red or white" is not, "I'm going to give you a lovely choice of alcoholic beverages." It's, "Don't get any fantasy ideas about G&T, mate, okay? Because I don't stretch to that, you bastard." Okay?
Now, what fascinates me, and I, I intend to illustrate, actually, I think we're really, really bad at choice. And one of the reasons I suspect we don't have free will as a species is that if we did have free will, I think we would have got better at exercising it. Um, but I tend to illustrate this fact with the M20 heading, uh, west, uh, from Dover towards the M25. And as in, anybody by the way, missed the M25 turnoff when coming? Yeah? Okay, five, six, seven, eight people. You, you have as well. Yeah. By the way, I got one hand up in the audience when I was in Johannesburg, so this is obviously not a rare problem. Okay? Now, I, I got a flattened deal, and after I had missed the turnoff for the fifth time, I thought, "This is a bit weird." Okay? You may be a total idiot, but this is more than coincidence. Now, I don't generally miss motorway turnoffs, and this one was proving to be just an absolutely recurrent nightmare. So I hadn't got a clue why it was happening. So I went to Google Street View and I kind of, you know, in a kind of CSI-style reconstruction, I went back over the road. Now, this is the first sign you get. Notice there's no distance indicator. Doesn't tell you where this junction is going to occur. And it also says "London" on both sides of the road. And that's the first clue you got that you may have to do something. You can't, you can't make any sense of the bloody thing because "Leam and Swanley." Swanley is principally famous as the name of the interchange that you're heading towards. So which direction that's going to be in is a bit of a mystery. Utterly useless information. But you got, "I don't understand that bloody sign at all," you think subconsciously. But I can see another sign on the horizon. You can probably just see it, um, against the sky. So what I'll do is I'll drive towards that sign and I'll make a decision when I get there, when I've got some more information which I can actually act on. So you go, "Thank God for that, there's another sign coming." And as you drive towards it, this happens. Before you can actually read the sign, the two lanes diverge. So the, the fact that you've made this mistake is entirely excusable because here, the only way of now correcting your mistake based on the information provided by the sign is by performing an illegal maneuver at extraordinary risk to yourself, by swerving at high speed across the painted section of the motorway, which I must admit I've done once. Okay?
Now, what fascinates me about this, what it reveals, I think, about the human brain is that when we're presented with shit choice, we don't actually notice. You see, I spent 10 years of my life studying choice architecture, and it still took me seven or eight minutes on Google Street View to go, "Hold on a second, this is bonkers." Nobody actually drives through that and goes, "This sign is stupid, it's in the wrong place." They just make a bad decision. And when we're presented with bad choices, we don't go, "That is a bad choice." We go, "It must be some evolutionary thing." That most of the decisions on the Savannah were kind of real-time decisions, and we just have to make the best of the present opportunities available to us. Hypothesizing other possibilities, like moving the signage, wasn't really a mental strength to be cultivated a million years ago. And I've got a contention that there's a kind of weird movement I'd like to start, which is like proactive libertarianism, which isn't just promoting freedom of choice, it's actually creating new choices for people to make.
Now, I'll give you just a little example of this. I personally think that Britain should move to a four-day week. In fact, I wrote in The Spectator saying, "Anybody who wants to win the election outright could just propose a four-day week and they'd win." It's only because the every government politician is in the grip of St. Carn ofite economic forces, to be honest, obsessed with growing GDP, that this obviously leisurely and pleasant solution can't be adopted. Um, but why don't we do it anyway? And the answer is partly, by the way, a really simple one: no one ever asked. We, there is never a point in our life in which we are actually presented with the option of having more leisure instead of more money. So I want to do an experiment at Ogilvy, which is my employer, which is just to set a rule, which is everybody who's worked there for two years or more, every time they're offered a pay rise, they must be offered the option of taking half of it as money, half of it as leisure. And I want to see what happens. Because the reason nobody's doing that, maybe the reason nobody's doing it is you never asked. That's the oldest explanation, uh, for why humans fail to do things.
And what fascinates me about a lot of these solutions is they, what I call "obvious only in retrospect." I spent years accumulating more and more electronic devices so that when I travel, mobility is the most nonsensical thing. It means basically every time Apple or somebody invents a new mobile device, it means that my, the bag I carry around with me gets even heavier, to be honest. Um, but every time a new one of these things came out, I bought a new adapter plug for foreign, you know, for backward countries that don't have the proper British plug system. And, um, I suddenly had about five of these. And after I accumulated five of them, it suddenly occurred to me, "No, no, no, you only need one." And you carry a four-gang extension lead with you in your suitcase. Now, what's interesting is I wrote about that in The Spectator. I got about 100 emails from people going, "Never occurred to me. Why did I never think of that?" It's an idea that's kind of obvious in retrospect. Now I've told you, you'll go, "It's perfect," because most hotel rooms only have one damn socket anyway. So actually, it solves that problem where you're going around sort of unplugging the television in order to charge your phone. Um, but it's one of those things which is only obvious in retrospect.
And these things fascinate me. First of all, choices just disappear. Probably one of the greatest alcoholic drinks you can buy in terms of pleasure per pound, sherry. Nobody has ever thought of buying. Has anybody here thought of when we've gone into the last wine, much even contemplated buying sherry? Probably about two people in the room. Britain would have drunk about 100 million bottles of the stuff every year, only about 30 or 40 years ago. Why did it disappear? I think the consumption occasion just disappeared. It's not that the pleasure got any worse, it just got kind of lost to mental saliency. Um, other things in terms of choices that we only make when we know we can make them. Uh, this thing, the Orange Line, what kids call the Ginger Line, uh, technically called the Overground line. Most of that line has existed for years and years and years, okay? But North Londoners cannot mentally function except by using the tube map. Have you noticed this? Because North Londoners don't understand South London at all. Because they, I don't understand how people get around. Truth is that South London has lots and lots of trains which don't appear on the map. And the Overground line is a perfect case of brilliantly creating economic value simply by making people aware of a choice they didn't know they had. So when they actually put it on the tube map, colored it orange, and called it the Overground, they did a bit of fancy stuff with the new rolling stock and extended a bit, but most of it had existed for 20 years. There's a thing called Silver Link Metro for people who have been trained, okay? When they put it on the tube map and called it the Overground, usage tripled. The reason people were using it, weren't using it, it was a choice they didn't know they had.
And this strikes me as a fascinating area of just general explanation. Because this is the most beautiful one I can give in the medical field. For years and years, doctors, GPs, are pretty intelligent people by and large, I think we can assume, can't we? We better say that because we're actually in the Royal College of General Practitioners, aren't we? Had a bit of trouble getting here, by the way. I rang this morning and asked for directions. They told me to ring back tomorrow at 8:00 AM. Joke. Okay. Anyway. Um, okay, anyway. Um, pretty bright people, okay? But they assumed that when a patient came in and said, "Can I have antibiotics for an upper respiratory tract infection?" they just assumed there was a choice they didn't know they had. They just assumed there were two choices: give them a prescription for antibiotics, don't give them a prescription for antibiotics. Someone, and I can't find out who, but it seems to have been comparatively recently invented, the deferred prescription. You give them a prescription, it becomes live in, let's say, four days' time. And the doc simply says, "Take this prescription away. If you're still feeling crook, if you're not feeling any better, it doesn't have to be an Australian doctor. If you, if you're still feeling a bit ill in four days' time, take this into the chemist, okay, and get some antibiotics." Now, what's so important about that? If you give people an immediate prescription, about 98% of people take antibiotics. Give people no prescription, I think it's still about 15% because they come back and get a second, second appointment. If you give people a deferred prescription, it's about 33%. Now, if you factor in the fact that actually, what percentage of those prescriptions are unnecessary, you've probably reduced the unnecessary prescription level from about 80% to 16%.
Now, what you only realize, I'll be very quick here, is that what nobody realized when they thought those were the only two choices, an immediate prescription for antibiotics contains within it a biased choice architecture. And the reason is, okay, think about it like this. You're feeling ill, you get up to go to the doctor's, all right? And you're still feeling a bit crap. And so you go to the doctor, and the doc gives you a prescription for antibiotics. Now, when you leave the doctor's, what immediately is outside the doctor's surgery? A bloody pharmacy. So you go, "Well, I might as well cash this in now, okay?" Or you go to Boots next door because you don't want to make a second trip because you're planning to wrap up with a bit of Lemsip and a duvet and watch Trisha and Pointless and so forth. So you cash it in. You pay £8.10 for a prescription charge. Once you've paid for the damn pills, sunk cost bias, you're going to take them, aren't you? That's why 98% is so amazingly high. So you only realize the implicit bias in the choice architecture when you realize that there is another choice.
Now, I'll give you an example of this very quickly. Um, when you go to a restaurant, none of you are aware to the extent to which you're being manipulated. Okay? You arrive at the restaurant, you sit at a table. The restaurant wants to sell you wine because wine has no known price point, so you can mark it up like a bastard, right? People kind of know what Johnny Walker Red costs, so there's a limit to how much markup you can change. Buy a bottle of Chablis for six quid and charge 30 for it, and actually people will just rave about how marvelous it is and what good value it is because they haven't got a clue. Now, you arrive, they want you to buy wine. Now, what they do, they arrive, you sit down at the table, and there are already wine glasses on the table. Okay? And then they bring you a drinks list, but it's not called the drinks list, it's called the wine list. And if you notice the choice architecture of a wine list, there are about nine pages of gratuitously wide selection of wine. And then for the deviants, perverts, and transgressives, they only hand out one wine list. There's only one alcoholic drink where one person can choose for the whole table. So the second that bugger with the wine list says to you, "Red or white?" it's game over for anybody who wants to drink gin. So what you, you, you're sitting there happily thinking, "Gosh, isn't it marvelous to live in a free market with all this freedom of choice?" And meanwhile, the restaurant's done you over like a bastard, frankly.
Now, that's the point. Marketing, very quickly, I'll wrap up now. The things flashing, another opportunity. It's assumed by economists that you can't get people to make more efficient use of GPs' time without financial incentives. And because you can't charge for a visit to the GP, how do you get people who are free all day to take appointments in the middle of the day, and people who have to work to go in the morning or the afternoon? Ocado's already solved that problem quite often. Just look to consumer capitalism. Consumer capitalism is like the Galapagos. It's got all these foibles and absurdities. And if you actually ask questions about them, all they do is, when there's already a delivery in the area, they don't reduce the price of delivery at all. They just make the van green. And people who are indifferent go, "Well, I'm not really bothered whether it's 9:00 in the morning or 11:00, so I might as well choose the green one because it just seems a bit nicer." Ocado wouldn't tell me the formal results, they just said, "We could not believe how effective that nudge was." Do that with GPs' appointments. You can yield manage GPs' appointments.
I've mentioned the, the red pills. Don't give people 24 white pills if you want them to finish the antibiotics. Give them 20 white ones, four red ones, and say, "When you finish the white pills, take the red ones." That makes it easier to do the right thing in System 1 than it is to do the wrong thing. Designing for our instinctive, evolved psychology. Um, also, totally stupid things. Um, the forces of reason are actually quite, uh, dangerous. Okay? People have been trying to compete with Coca-Cola. I'll only be one minute more now. If I go back one more. People have been trying to compete with Coca-Cola for about 150 years, right? And if you went to an economist, they'd say, "Well, there's one way to beat Coca-Cola in the soft drinks market: you've got to produce a nice tasting drink that costs less and comes in a really large container, so people get value for money." The only person who successfully competed with Coca-Cola are these people. It tastes disgusting, costs a fortune, and comes in a tiny can. Um, now, the reasons for that are deep in, deep in human psychology, by the way. Um, if we want to believe a drink is efficacious in some way, it has to taste a bit yucky. I was talking to someone who works for a household cleaning product. It's perfectly possible to manufacture a fly spray that smells beautiful. No one will buy it because they don't believe it. It's too, basically, cognitively confusing. Basically, if fly spray has to smell a bit crap, because I go, "Yeah, well, if it's a bit crap to me, it must be a bastard to the fly." That's designing for System 1, okay? And economists don't do that. Economists design the world for System 2, as though we're reasonable.
Finally, I think this is my last slide. Can I just have one minute? Okay. When people design booze and diet programs, okay? You go to an epidemiologist and you say, "Okay, how many units should people drink, uh, you know, every week?" And suspiciously, they always come up with a multiple of seven. Um, I mean, which suggests that the science isn't entirely robust. Um, but their idea is the amount you should drink. They look at some epidemiological study and say, "Well, people who drink less than this don't seem to come into any harm generally, so we'll make it that." And that's the rational solution. The System 1 solution to behavioral change is probably different. It's probably more akin to the 5:2 diet, for example. Do not drink for three days every week, preferably consecutive days. Then don't worry about what you drink the other four. Similarly with dieting. Um, religions basically solved this problem. They called it fasting, okay? Now, the interesting difference there is, you go, "But that's not rational because it doesn't have a clear kind of mathematical, um, you know, point of origin." The, the logic chain from epidemiological study to "don't drink three days a week" actually gets broken. Doesn't mean it's not a better idea. Let me explain. If you practice, say, a 5:2 diet, or if you practice not drinking three days a week, there are a lot of ways in which it's far better than counting units. It's far harder to delude yourself. Emanuel Kant said, "It is easier to fast than it is to diet." Can't deluded himself. He limited himself to one pipe of tobacco a day after breakfast, but by the end of his, his life, his pipes were absolutely bloody enormous, right? So if the world's foremost ethical philosopher can con himself, there's no chance for the rest of us. Um, we can delude ourselves that that vast sort of quarter-liter bottle of Chilean Merlot at 15.8% is one unit, okay? When it's abstinence, you can't con yourself in the same way. Secondly, okay, units require you to stop drinking when you're already a bit pissed. "Oh, I've just reached my limit. Not going to happen." Right? Thirdly, if you look at man as a social animal, you realize that, um, fasting or two days a week off or three days a week off is socially reinforcing. If you and your partner choose the same two days, you can help keep each other honest by going to the cinema, not going to the pub. Units will never be socially reinforcing. We all have good friends, but nobody has a friend who's such a good friend that you go out to a party and they go, "I've actually got nine units left this week, but I know you've only got two, so I'll just have two drinks to keep you company." Not going to happen, right? So in all manner of other complex ways, religion came up with a better heuristic, System 1 answer than science.
Now, I think I'll end here. The final thing is, as I mentioned with, with Red Bull, one of the marvelous things about capitalism, okay? I know, I always wind up academics with this, is that in order to succeed in academia, you have to be really intelligent. But the brilliant thing about business is you can succeed in business by being stupid and lucky. Now, that's, it's much more democratic. The fact that it's convex to luck. I know academics hate business people because they go, "But they're really stupid." I know, that's what's brilliant about it. Okay? All right, I'm being really mischievous here, but seriously, being stupid in business can be a huge advantage because you do the things that no sensible person would do, and sometimes they work, and you get rewarded for it. So I'll just end up with this little thing of built-in randomness, which I think is a, is a final little tip. Everybody in the rational world of science believes there's a single right answer, and they, you know, it's occasionally called best practice or the optimal solution, okay, or maximum efficiency. Nature doesn't do it that way. It actually builds in a degree of inefficiency or short-term efficiency. A certain significant percentage of bees ignore the waggle dance. And people looked at that and thought, "That's a bit strange. You'd think that to maximize pollen collection efficiency, that evolution would have created a higher level of compliance." Then they worked it out mathematically and discovered without the random bees, the hive will get trapped in a local maximum. And when the, uh, layout of the availability of pollen changed, the hive would get stuck and starve to death. There is a degree of healthy variation and inefficiency in any system. And there's a degree of randomness in any system which is necessary in order for you to get lucky. And so that quest, which I think science always looks for, which is the single right thing to which all people must adhere, it seems wonderful, but it also comes at a cost because you never find anything out by accident. That's a bit of a reading list. That's me. Thank you very much indeed. Closure. You want to ask questions? I've overrun. Brilliant as ever, Rory. I just got one quick question with like a 20-second slot for an answer. Um, okay, we're about to hear from the National Clinical Director for Innovation for the NHS in England. Is there one suggestion, one nudge, you think he should take away and go and implement? I'll give one example which I didn't have time to talk about. One of the things I've been a passionate, don't discount the value of luck in the discovery process. One of the things that most fascinated me was the hostility to e-cigarettes because we don't know if they're good or not, okay? Let's be honest about this. We don't know for sure. But it's possible that completely unintentionally, actually, as the work of weird entrepreneurs in China, not as the product of, you know, a pharmaceutical company, someone's come up with the best cessation device in a 100 years. So the fact that people rejected that idea, not because it didn't work, in fact, they contrived all manner of ludicrous post-nationalizations to ban them, like, you know, "it promotes the image of," or whatever. But because it came from a source where solutions weren't supposed to come from, struck me as really, really alarming. In other words, had a pharmaceutical company come up with the e-cig, it would have been heralded as a magnificent, you know, improvement and a fantastic advance. Okay? But because it came from left field, the instinctive reaction was to throw it out. That's really scary because hell of a lot of, first of all, a hell of a lot of improvement in life is just damage limitation. You know, it's just making things less bad, okay? That's the first thing. And a hell of a lot of it's actually the product of just, uh, you know, selection acting on random forces. So stay, make it possible for yourselves to be lucky. That's the, that's the final bit of advice. Thank you, Rory. Pleasure, always. Thank you very much.