Negotiation engineering | Michael Ambühl | TEDxZurich

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
Oh when I was in high school I had a teacher who wanted to convince us of the importance of mathematics in these days everything can be computed he said making the following example you have to engage you have to hire a new person in your company say you have n persons waiting outside your office and randomly you interview one person after the other each time you have to say yes or no if you say no you can't go back and if you say yes you can no more take one of the waiting lists outside you see the dilemma if you select very early a candidate the best one may come much later and if you wait till the very end the best one might have passed already long time ago so what is the best strategy to select the best of n candidates applied mathematics gives you a nice and surprising result namely you select you test the first 37% but you reject them namely n / e e being the so called Euler's number and then you select the first one which is better than the rejected ones n over e a nice piece of mathematics isn't it impressed by it I decided in high school to study at ETH Applied Mathematics the discipline which gives such nice results the example I presented here is commonly known as the secretary problem but as you could easily see it could also be called the marriage problem however when I was a young student I had to realize that the formula n ovary might not really work to find a spouse because you don't know n in advance and because life is just a bit more complicated my future wife brought me to new ideas anyway applied mathematics remains an interesting discipline and later in my function as a negotiator I often used mathematical tools now as a professor at the ETH we try to conceptualize these experiences and we call this conceptualization negotiation engineering by negotiation engineering we understand the division of a complex problem into subproblems and the application of mathematical tools in subproblems let's talk now about two real cases case number one free movement of persons migration is as we all know a complex challenge facing society last year Swiss citizens voted on a constitutional article mandating the government to regulate migration however this article is in contradiction with the free movement of person agreement that Switzerland concluded earlier with the European Union while that new article foresees quotas the EU wishes no quotas using negotiation engineering we propose a model that would make it possible to preserve the principle of free movement of persons what's the basic idea for a solution in exceptional migration cases exceptional measures should be possible but what does exceptional mean let me make a simple example with our four neighboring countries we take the net migration in percentage of the population we calculate the mean value and the standard the creation and if the Swiss rate would be much higher then a special regime would allow to bring the rate down to a predefined threshold now we extend the exercise to the 32 UF de countries and we check how their migration rates are distributed around the mainland in case of the normal distribution one could reasonably argue that the migration rate which is higher than the mean value plus two standard deviation you have an exceptional case only in these cases temporary limited.if measures would be possible the parameters and the modalities of the internal measures unilateral or bilateral ones would have to be discussed in negotiations that's it the take-home message of this case is mathematical methods can contribute to an object if occasion of a sensitive question now I go over to the second case the Iranian nuclear program a case I have been dealing with in my former life for the past decade the five permanent members of the UN Security Council plus Germany the so-called p5 plus 1 and Iran had a dispute over the Iranian nuclear program why Iran constructed centrifuges to enrich uranium and the crucial question was whether this program is in accordance with the nuclear non-proliferation treaty when the two main parties the United States and Iran reached the deadlock the negotiation stopped there was mistrust and non-constructive rhetoric the u.s. called for regime change and wanted all nuclear activities to cease while Iran was naturally against the regime change and wanted guarantees for enrichment Switzerland as a traditionally neutral state promised in 2007 a way to promote the dialogue with steps that included confidence-building measures and of course with a nice little mathematical model from us the proposals laid the groundwork for the first high-level meeting in Geneva in 2008 this was the first meeting where all the seven delegation p5 plus 1 and Iran were participating however the key players were not ready to start the negotiation was our math too complicated probably not time was politically just not right to start the negotiations and the key players went into a not very rational continuation of the confrontation let's all try to analyze why we do it with game theory whose founder was by the way the ETH graduate John von Neumann the situation that prevailed can be described with the so-called prisoner's dilemma this can be represented nicely in this matrix you have the US on one side Iran on the other side both have two strategies being flexible meaning willing to negotiate or inflexible not willing to negotiate in the situation that existed the payoffs these ones and let me now go rapidly into this matrix let's assume just for a second that both players would say let's be flexible however Iran would immediately realize that changing from flexible to inflexible it would gain one point so it has no incentive to stay flexible the same logic goes for the u.s. they would have no incentive to stay flexible choosing both inflexible they land in the stable strategy combination of in flexibility which is a stable combination of strategy and it is a so-called Nash equilibrium as a result both are inflexible there are no negotiations and we have a typical lose-lose situation what happened now no negotiation one side installed more centrifuges and the other decided on more sanctions it started with you and it went up this way from 200 centrifuges to 20,000 centrifuges from for sanctions to 80 sanctions so both lost only external factors changed the game at the outset there were political changes from President Bush to President Obama from President Ahmadinejad to President rouhani now we leave these smiling pictures and go back to our sober matrix here we are again for other factors change the game change the perceptions and the payoffs of the two players Iranian star to feel the effect of sanctions and that made them a bit more flexible in our model of course only we remove one piece from right to left the Americans on the other hand became concerned about the nuclear the size of the nuclear program of Iran and that made them of course only here in this model a bit more flexible we remove one piece from the lower to the upper box then Iran realized that it could lift the life of a regional power only with a minimum understanding with the big powers having a sort of a modus vivendi and the Americans realized that the certain political involvement of Iran would be helpful in order to reach one day as a stability in the region these changes transitioned the game from a prisoner's dilemma toward a cohort game we have with flexible flexible combination a stable situation we have here the new Nash equilibrium flexible meaning willing to negotiate the real negotiation could start and the result is a good one today we can be satisfied when both became flexible a good solution was possible I wonder however if they would have become flexible earlier perhaps an even better outcome would have been possible so the take-home message from this case is be open for negotiations coming back to the start of my presentation the usefulness of mathematics as a negotiator I found that there is hope for mathematical tools while they can certainly not so of all the problems of the world of course not they can help within certain limits to the emotional eyes a problem and to better understand the process whether it is about finding a spouse regulating migration or restricting a nuclear arsenal so my teacher was probably not totally wrong math can be useful and I would add also in negotiations thank you very much
Info
Channel: TEDx Talks
Views: 26,382
Rating: 4.7882352 out of 5
Keywords: TEDxTalks, English, Switzerland, Social Science, Behavior, Communication, Global issues, Globalization, Science, War
Id: TYVC7TyGNWo
Channel Id: undefined
Length: 14min 59sec (899 seconds)
Published: Wed Jan 06 2016
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.