Some years ago my dad, who is an avid blogger, wrote a piece about a plague of disorderly advertising that was chocking Kyiv. Notices promoting everything under the sun were plastered on the walls, fences, lamp-posts, in metro cars, on the bus stops, even on the pavement. Our own building’s entrance was a sad sight - always covered with debris of paper and glue.
Our neighbor, and elderly lady, laboured heroically to wash off the wall and the door frame every now and then, but it was a Sisyphean task. The clean wall only served as a magnet for new advertisements, and the cycle continued.
One day my father decided to do something different. With a piece of plywood and a few planks he put together a makeshift notice board and hung it outside the building entrance. A few days later something almost magical happened – the first notices appeared on the board, but the wall and the door were clean…
My dad is a linguist, and his acquaintance with economic theory goes as far as reading Karl Marx’s “Capital” in his early student years. But this time he indeed acted as an economist – he solved a problem in mechanism design, a problem which belongs to the class of coordination games in economics. The notice board provided a focal point which allowed the “players” – the building residents and the advertisers – to coordinate on a mutually beneficial equilibrium. At least for our building, the notice board was an easy and smart alternative to the laborious task of hunting down and punishing the “ghost” advertisers.
This winter when I was visiting my parents I was again reminded of the power of mechanism design to solve public nuisance problems. Dog ownership is very popular these days, and the path around our neighborhood lake serves as the main dog promenade. The problem is that dog owners rarely bothered to pick up waste after their pets, creating a major inconvenience, not to mention a serious public health hazard. This winter, however, the area around the lake looked different. It was much cleaner than I had ever seen it before. There also appeared green posts with metal bins and colorful signs attached to them. The bins were of two kinds – one contained a bag dispenser and another served as a waste basket. The sign on the post provided instructions on how to use the contraption, also reminding the dog owners to be responsible and community conscious.
I can tell you I was overjoyed more than I could express in words – both as an economist and a citizen. As an economist I witnessed another elegant solution to the common coordination problem. As a citizen, I was reminded that only a small action is sometimes required to transform the face of our community for the better. As human beings we may be instinctively drawn to beauty and order, but most of the time we are too lazy to make an effort to pursue them.
Nowadays I am back in Tbilisi, taking daily walks in Vake and Vera. Both are middle-class neighborhoods, boasting high rates of dog ownership, and, predictably, plagued by the same dog waste problem. I am of course irritated by the dog owners who don’t bother to carry plastic bags or clean up after their pets, but I also realize that policing and prosecuting such behavior is probably not the most efficient way to deal with the problem. Perhaps “pet stations” will provide the solution? They might! I am only certain of one thing – it will be useless to wait for the government or the city hall to take the first steps. Ultimately everything depends on the citizens who should take responsibility for their community and their environment. Taking responsibility may be hard, but try it – it is a powerful feeling.
Comments
It is so encouraging to see how smart common men defeat the unpleasant dominant strategies, prevalent in the society!
Nice post but the economics is misleading. It would seem that the solutions in the post highlight not a successful application of mechanism design but the essential role of human virtues that are ignored in classical mechanism design problems. There is no coordination problem in the post, as the term is classically understood in economics. Rather it seems like a problem of the commons. This particularly seems to be true of the dog poo problem. The common resource is a clean environment. However, dog owners act as free riders and misuse the commons by leaving dog poo. How do you stop them? Neoclassical economics (of which mechanism design is a part) would advocate stricter enforcement (higher penalties, greater vigilance). But this is not the solution that Yasya highlights as being successful. In the successful solution a direct public appeal to dog owners worked. This is a form of moral suasion that works because humans have nice traits too. Indeed if one reads Herbert Gintiss Bounds of Reason (which should be in the ISET library), a strong case is made for the incorporation of human virtues into classical game theory. I deal with these issues in detail in my forthcoming book.
There is certainly a coordination problem - and its solution - in the first example (that of Yasyas father).
The dog poo is also a problem of coordination. It does not make sense to clean after your dog in Georgia because nobody else does so. You look like an idiot (as many commentators on my FB post indicated). Conversely, you look ridiculous (or worse) if you dont clean after your dog in Germany (and now also in some parts of Ukraine).
This is not to deny the role of human virtues in resolving coordination problems. In the poo example, the bag dispensers were an effective coordination device because there must have been sufficiently many virtuous dog owners to start using bags before everybody else. Once a critical mass of virtuous guys switched to cleaning after they dogs, the same virtuous behavior became imperative for all other dog owners as well.
This is what I said: There is no coordination problem in the post, as the term is classically understood in economics [please note this part].” But the term “coordination problem” could also have other literal meanings that might be useful for people to think about problems- but I was not writing about these alternative usages.
In coordination problems people take actions (typically simultaneously) and there are multiple equilibria. It may be difficult to coordinate on equilibria that increase everyone’s payoffs (coordination failure). There is no strategic interaction between dog owners and non-dog owners unless non-dog owners can somehow influence the payoffs of dog owners- I do not see how. In most cases it is very hard to know whose dog was responsible for the mess that you see. For this reason, dog owners may take their dogs out before anyone wakes up to observe them. I jog in the mornings and often see dog poo, but I have yet to catch anyone in the act. You can try writing down a few games to see the point of what I am saying. So certainly not a coordination game in the classical sense.
For the same reason, the example with Yasya’s Dad is not a coordination game as understood by economists.
In each example it is the case that the actions by one group of people influence the payoffs of others; and others cannot always monitor and enforce. This is true of common resource problems. Hence, it was the example I chose. But the prediction here is that common resources will be fully exploited (in the first example the common resource is a clean wall and in the second, a dog-poo free environment). Since Yasya gives compelling examples to show that the predicted outcome from common resource problems does not hold in this case, so one or more of the assumptions of the classical framework must be violated. My case is that it is the absence of human virtues assumption that does not hold in this case. I could give many examples of this backed by strong experimental and field evidence; details in my book.
Problems of the commons represent coordination failures, according to your own (standard) definition, Sanjit:
A. There are multiple equilibrium:
1. each person has 1 cow, all cows have sufficient fodder, people are happy;
2. each person has multiple cows; cows die of starvation; people are miserable.
B. It is difficult to coordinate on the good outcome (1) because ... it requires coordination. Individuals dont have any incentives to reduce the number of cows in their ownership to the sustainable equilibrium level unless others do so at the same time.
In the poo example, everybody are better off in a no-poo equilibrium, those who own dogs and those dont. Nobody enjoys walking his/her dog in a poo minefield. As a dog owner I have plenty of field (or, rather, street) evidence to support this claim. Even in a street exclusively populated by dog owners you would have exactly the same multiple equilibria:
(1) people clean the poo after their dogs and enjoy the safety of walking in a clean street (or field)
(2) people dont clean their own poo and from to time step into a poo minefield.
Both equilibria can be shown to exist and be stable. There is very little joy in the poo equilibrium relative to the no-poo equilibrium but it is difficult to coordinate on the better outcome because everybody (or at least a critical mass of dog owners) have to start cleaning after their dogs at the same time.
Virtue (care for others) can play a role in helping arrive at the no-poo equilibrium, as in Yasyas example in which the better equilibrium is achieved with the help of a carrot (a free bag + information campaign). The same result could be achieved with the help of steep fines and monitoring. In this latter case, virtue plays not role. People start cleaning because of purely egoistic considerations (they dont want to be fined).
Eric, I think you missed a critical sentence in my post: There is no strategic interaction between dog owners and non-dog owners unless non-dog owners can somehow influence the payoffs of dog owners- I do not see how. A very small percentage of the population is dog owners but all others (who have no meaningful action in this game) suffer from the dog-poo just as in the commons problems (the commons here being a dog-poo free environment). So unless you just wish to focus on a game between dog owners alone (which is not Yasyas stated problem), you would struggle to set this up as a coordination game. I have already shown you a clean solution to both of Yasyas examples but you may or may not agree with my solution. Without the role of virtues you could never have any equilibrium in which dog poo is ever cleaned- so they play a critical role. Of course you may try to formalize all this as a coordination game if it helps you to understand the problem better. Perhaps you could discuss this further with the local game theorists and write a paper on it.
I get your point and better understand the source of disagreement among us.
In my view, however, non-dog owners (ie PEOPLE) can and do strongly affect the behavior of other people who stray from socially accepted norms. They do so through public shaming of (yelling at and even beating up, sometimes to death) those who physically abuse weaker people, steal, jump queues, do not clean after their dogs, etc. When rollerblading through Berlin a few years ago, I got yelled at when crossing the road on a red light. This was quite effective.
Of course, for public shaming to be effective in deterring anti-social behavior, those who stray from socially acceptable norms should be a small minority (a redundant qualification since a norm is something that is accepted by a large majority). Hence, the problem of coordination.
Yes, as I said (in your terminology) unless PEOPLE can influence the players (dog owners) who create negative externalities on others. Such public enforcement varies between activities and cultures. The fact that we observe so much dog poo in Tblisi and even here (to a lesser extent) suggests that public enforcement is weak. Even if one wishes to defend a story where norms play a critical role in ensuring that dog-poo owners always clean-up, still the role of virtues has to be a central part of the story- while this is missing in classical game theory. Only a virtuous set of people would have an incentive to obey a norm that cannot be effectively policed publicly. On this you might wish to see Parts II (Social Preferences) and IV (Behavioral Game Theory) of my book when it comes out.
In any case, I find myself in disagreement with the message of Yasyas blog that classical mechanism design solutions work well in this case. Quite to the contrary. There is growing work in behavioural mechanism design that Yasyas interesting examples might well feed into in the future. But this is exactly what is the purpose of these blogs: to facilitate meaningful debate on issues and enlighten us all. For this we must thank Yasya and you, Eric.
I cannot help but say to everyone- it is almost impossible to understand the economic world around you unless you take a serious study in behavioural economics. What other economics is out there that applies well to humans?
Nice post! While noticing the same problem in Tbilisi, I was thinking that the government should take some measures (like fines) to solve the problem. But I guess intervention of a social planner is not the only solution. People can also do that. The only problem is that someone should make the first step, like it was in case of putting a notice board. However, usually we dont think that someone could be us and so we still see waste of dogs on the streets.
Yes, Greta, I think you are right.I was thinking about the first steps as well. I think that neighborhood residents have to first show that they care a lot about the issue. There needs to be a grassroots campaign, perhaps social advertising, etc. After that introducing the bag dispensers would make the transition much easier - people would already know they are doing something the public approves of and cares about.
One may think that people already know what is the right thing to do, but in fact it is not clear...There is indeed a problem of coordination on a socially acceptable equilibrium.
A person who voluntarily picks up after their pet when no one else does runs a risk of being branded as a snob by dog owners and non dog owners alike.
To give you an example, a friend of mine who is Georgian and a very socially responsible dog owner, was told by a neighbor (who did not own a dog, I believe) something like who do you think you are - do you think you are better than everyone else? Of course some other people have since expressed their wholehearted approval of her behavior, but this example illustrates that the socially accepted equilibrium is not yet clearly defined, or at least not clearly communicated.