 Okay, we're on air right now. Okay, so hello everybody and welcome back to the Latin American Webinars in Physics. My name is Joel Jones from the UCP in Peru and I'll be your host today. Today we're having our fourth colloquium with a very relevant topic for these days. It's the cost-benefit analysis of large research infrastructures, in particular the LHC. The colloquium is being given by Stefano Forte, who is a professor at the Physics Department at the Universitat di Milano in Italy. So Stefano doesn't really require an introduction. He's a very well-known in the particle physics community. Just to give you an idea to those that do not know him, he's the author of more than 100 published papers and has got an average of 97 citations per paper. So, well known, right? But still, I'll give it a try and introduce him to all of you. Okay, so Stefano carried out his PhD at MIT, which was followed by postdocs at Saclé and CERN. He's been part of the INFM in Turing and Rome, and in 2003 was named full professor at the University of Milan. In addition, he's been a visiting scholar in fancy places like the Ecole Polytechnique, Danielsburg Institute, Edinburgh University, the Université Pierre Marie Curie, and also in the University of Barcelona. And he's currently the recipient of an ERC Advanced Grant, which started in 2018. Okay, so before we begin, let me remind everybody that you can ask questions and comments via the YouTube live chat system. And this question will be passed on to Stefano at the end of his talk. So for some reason, my video has collapsed, but okay, that shouldn't be a problem because I'll send you to give you all Stefano. We're all yours. Okay, so thanks Joel, I'm almost embarrassed by this introduction. So this is also the first time in my life that I give a web seminar, so I hope I can do this right. So let me share my screen now. Right, and then let me get my slides on. Right, okay, so like Joel said, I'm going to talk about a topic which is somewhat unusual for everybody and also for me since I'm a high-energy physicist, like I guess the majority of the people who are attending the seminar, but I'm going to talk about the topic which is interdisciplinary between science, social sciences, and economics, and it's actually the topic of some work I did in collaboration with some colleagues from the economics department here in Milan and the consultancy and there is now a project which is partly financed by CERN to keep doing this work. So you can see that it's unusual. It's also unusual for me because since it's not a physics topic but interdisciplinary topic and the colloquium, instead of using latex as I always do for slides, I'm using PowerPoints lookalike from Linux. Okay, so like Joel said, the main topic of my talk is cost-benefit analysis. So I guess I should start by explaining what cost-benefit analysis is. That's something I learned myself when I started doing this work. Cost-benefit analysis was invented by an engineer, actually, a French engineer by the name of Jules Dupuis from a called Polytechnique who in the beginning, well half of the 19th century, as France was becoming slowly an industrial nation and it also was becoming a country where the government would invest in public goods. He asked the question, what's the value of the bridge if no toll is levied on the bridge? I mean up to that point infrastructures like bridges were done by some private guy who actually had people paying. Now it's the government and then you know how much it's worth. It's worth the amount of money you can make by attacking a toll on people who go on the bridge. But then he asked, okay, can we establish a scientific framework? The guy was an engineer. I mean a called Polytechnique is the alma mater of Poincaré. I mean, it's an institution which is for engineers but had a solid mathematical background. So he wanted to pose in a quantitative scientific way this question. Now the question if you think about it is a difficult one for two classes of reasons. I mean one reason is purely quantitative and has to do with the fact that the value of the bridge does not just depend on how many people will go over it but on many factors. What I'm showing here is a picture of the bridge in Mostar. This is a bridge which was destroyed in the Civil War in Bosnia in the late 90s and then it was reconstructed based on an international fundraiser. And you know it's a symbol of the union of a town which had a Muslim community and a Christian community and that was the reason of the Civil War. So at present on this bridge there is a yearly diving contest. So you know you could think what's the touristic value of people going to see the bridge or to see the diving context and that of course should also be counted in the benefits but then of course you can also say sure there are benefits that you can quantify like you know how many people will check in the hotels of this town to see the bridge but there are others that you cannot quantify like how much is the symbolic value of having a bridge which is the symbol of peace between two different communities and to which extent can we make that quantitative and can we keep the things separate. Now cost-benefit analysis. So you understand it's a relatively complicated subject but it's also a well-established one and in particular in Europe there is an institution which is called the European Investment Bank which is not really a bank. It's called bank but it's really an institution which gives loans to European governments. So for example if you are say the government the Czech Republic and you want to build a bridge then you may want to ask a loan at a reduced rate from these guys and before deciding whether they will give you the loan they will make a cost-benefit analysis and they do this based on the manual which has additions where there is a well-defined protocol and you know of course at the end of the day you may decide to give the loan based on political considerations but still you want to know how much they think costs and it turns out that the coordinator of the academic panel who established the methodology which is used in the last editions of this manual is a colleague from mine, Massimo Florio from the Economics Department of my university and therefore he thought that it might be interesting to develop and apply this set of methodologies to science. Now the reason why this is interesting is that the way the value of science is assessed is usually not very accurate so for example the Bataille Institute which is a consultancy supposed to be professional economists who are supposed to know what they do a few years ago they claimed that the Human Genome Project had a return of a factor 140 meaning if you put one dollar you'd get 140 dollars back after a 10 year or something which you know just think about it and make any sense Washington Post would call it a three Pinocchio's kind of claim I guess only Donald Trump manages to get more Pinocchio's than that you know if that was the case why would people invest in stocks and not in the Human Genome Project? On the other hand you know these things may this kind of exaggerated claims may work well or they may backfire in our field of high-energy physics we know of the case the LHC which was actually saved by the fact that in 2002 when CERN had a severe economic crisis the European Investment Bank which I mentioned already decided to make a loan of 300 million euro to CERN they didn't do the CDA because the protocol didn't exist you know based on the CERN prestige they decided to do it had they not done this probably the LHC would have gone down the same fate as the SSC which as the older among those who are listening shortly remember almost killed high-energy physics in the United States and you know I belong to the generation of those where postdocs when the SSC was closed and in fact of all my contemporaries at MIT essentially the only one was left in physics because I went back to Europe rather than staying in the United States so this work I'm gonna talk about started when the European Investment Bank actually rather the research institute of the European Investment Bank made a call for grants asking for people to make a proposal to develop a framework for doing cost-benefit analysis for research infrastructure and my colleague from economics said okay why don't we do a proposal we did it we got the grant and we put together a team which worked for three years 2013-2015 you can see here it had people well the head of the team was my economist colleague and then it had a steering committee including myself and him and responsible of a consultancy here in Milan and a number of people both from science and from statistics and from economics you can see the scientific committee James Sterling who's very well known in the high energy community and at the end of this project we produced two things we produced a model so that from now on there is something which in principle could be applied to do a cost-benefit analysis for research infrastructures not just CERN but whatever just physics also biology also the human genome project or a telescope or whatever and we also produced a case study which was published where we applied our model to the LHC we applied it to the LHC including some part of the model going into the past and some part going into the future the result was published in 2016 as I said and the analysis started with the beginning of the LHC which stretched all the way to 2025 now so let me get to so far I described how I got to do this and what are the motivations so now let me try to get to the content of doing cost-benefit analysis for research so why is it that sometimes there are these real claims like the ones concerning the human genome project that's mostly because people fall into the fallacy which I call the fallacy of the hole in the ground the idea is you pay someone to dig a hole in the ground and then to fill it again and since you're spending money you say well that's a benefit because I invested something in research now that's obviously silly and the reason to see that is silly I'm very sorry is to make a proper job and the proper job means that when computing costs and benefits you have to take the difference between how much money you spend and how much money you get back so if you just dig a hole and you're paying something there is full balance between how much you pay how much it costs and how much you get and as a matter of fact the balance is usually negative because then you have entropy, you have someone falling into the hole so the guiding principle of any well done analysis is that you have to determine the gain so you have to determine the difference between what you spend and what you get and this difference has to be done with respect to a counterfactual meaning say if I want to determine what's the benefit of building a telescope well you have to say what is the counterfactual what if I did not build the telescope I would still for example pay the salary of a bunch of people so it's only a difference that you have to account for and that's why the thing is not obvious so the model which was developed by essentially my economics colleagues is based on calculating what is called the net present value a net present value is a concept which is familiar to economists but not to me so I assume also not to you so let me explain so the idea is if I give you a thousand dollar now the present value is a thousand dollars but if I tell you I give you a thousand dollars in ten years the net present value of a thousand dollars in ten years is less than a thousand dollars because you are getting them in ten years not now and there is something which is called the discount rate which basically allows you to evolve back the value of something that happens in the future to what would be its present value namely the value of it happening now and this means that in order to determine the net present value of something so say accelerator like the LHC you have to determine a bunch of benefits be happening at time T and a bunch of costs also happening at time T and then you sum overall discrete times at which these benefits and costs occur and you evolve back to the present and the discount rate is something which you can you know it's estimated by governments government agencies and then you have some things which I'm going to explain which are called non-use benefits meaning that they are benefits which happen at all times they don't happen at a given time and you also have to account for those so what we did was to classify the benefits the use benefits which happen at times and the non-use benefits which are atemporal and then we applied the thing to the LHC so let me just summarize what we found for the LHC and then let me get explicit and explain what are the various benefits so in the case of the LHC we made a study which stretched over 33 years from the beginning of the design of the LHC 1993 when people started spending at CERN started spending money on the LHC to be predicted end of life of the LHC after that we now know will happen the high LHC but that's a separate thing we had to choose a counter factual the counter factual was CERN keeps running but with no LHC and then we had to make some assumptions like the discount rate and so on I can give you further technical details in the question session if you're interested let me just mention that costs are in principle triviality in the sense that they are an accounting problem but they are a difficult accounting problem so for example when you're estimating the cost of CERN you have to keep into account the fact that there are many people working on experiments as CERN whose salary is not paid by CERN but rather by their home institutions and of course they are not putting 100% of their time to CERN but they are maybe also teaching a class in their home university so you have to make a sophisticated piece of accounting which is to say that if you want to be quantitative rather than just talk about folklore you have to be careful but it can be done and now let me come to the benefits so we did this for a case study but we claim that this is a general applicability namely that there are four use benefits and two non use benefits one of which is zero so four use benefits means that there is some user who gets the benefit and I'm showing them here so one use benefit is papers that you published and beneficiaries or scientists the second use benefit is human capital affirmation so students and postdocs work at CERN and then move on either by staying at CERN or by going to industry or by going to education whatever but they get a benefit from their education technological spillovers so firms for example like the company that built magnets for CERN which you see the picture who get the benefit by developing know-how which they are then going to market and therefore the whole society will benefit from the fact that there is some company which developed something going to CERN and then cultural benefits which are those from people who come and visit say CERN or go and look on social media and so on and then and of course the users of the last cultural benefit are visitors and people who read the outreach material and so on now non use benefits non use benefits are a bit harder to explain let me try and do this so there are two one is the value of future discoveries this is called in economics a quasi-option value it's an option value because it's a future value so it's like when you buy an option on selling tobacco 10 years from now here it's a value of making a discovery but it's a quasi-option value because the discovery is not guaranteed now of course there's a sense in which this is the real value of science but unfortunately this is totally unpredictable so if you want to be quantitative you should set it to zero I'll comment again on this at the end of my talk but then to be conservative in a serious model in a serious CBA this is something you must set to zero because it's too uncertain the other thing is the existence value or as it is called also sometimes the public good value the public good value or the existence value is the value that you attribute to something because of the sheer fact of existing so I'm showing a picture of a panda because most people in the world would say that if the panda become extinct then humanity will lose something now there's nothing for me to gain by pandas living in China but still I would be willing to pay to make a donation to some fund in order to make sure that the panda does not become extinct and there are methods to estimate this for example using the techniques on environmental economics based on how much people are willing to say pay in order for these things to happen and this is a non-use value because there is no use to the panda but still I'd like the panda to exist there is no use in some sense to having a research institution but a research institution in that sense is like an opera house a library, a museum it's something that you want to keep so for CERN when we put everything together we discovered I mentioned this because again it's not universal it can be very different in many cases but there are some qualitative features which are maybe common so we discovered that costs and benefits but actually with the cost which is for CERN 13 billion euro in present 2016 value which is what we published our study and instead the benefits actually amounting to a difference of 11 billion euro more than the 13 you spent with the probability distribution so if you want to do things properly give probability to events the way this was done by running in Monte Carlo and saying ok we give a probability distribution for having students making so much money and therefore at the end of the day you can estimate a mean gain median gain a standard deviation so in our case we estimated that the probability of losing money from CERN was less than 10% and like I said I'm sorry I said the wrong number I mentioned 11 billion that was however the maximum gain the average gain would be about 3 billion so the idea is you invest 14 billion and you get 18 billion so you get 4 billion out of that that does not include of course the values of discoveries it does not include what is good say for us scientists but it's something you can tell a politician look if instead of giving that money to CERN you had to train scientists and you had to develop firms and so on you would have spent so much ok and what we discovered also was that as you may expect the value of scientific publications is completely negligible so I'm not going to talk about that any longer and otherwise in about equal proportions the benefit was in knowledge formation so sorry it was in human capital so the fact that you train people in technological spillovers and in putting together the cultural and the existence value which are actually close related closely related to each other so the benefits is really almost equally distributed among these three things so at this point I'm I guess two thirds of the way through to tell you how we estimated these benefits so that might be spend my last 10 minutes giving you an idea of how we actually managed to estimate that you can get benefits from these three major sources human capital, technological spillovers and cultural plus existence values I will explain this to you by trying to answer two objections so the first objection is ok you develop a model you get some numbers but how do you know the model is correct and so what I'm going to show you is that after publishing this three years ago me but mostly my economics colleague made a number of studies to try to do empirical analysis that give evidence that support the conclusion of the model and then the second objection is this was applied to something that already exists can you do it for something that doesn't exist as I mentioned we got some grants from CERN and my economics friends mostly did it for the high loomie LHC and now there is some effort ongoing for the future circular collider so let me come to the three main classes of benefits so the first one is technological spillovers and this in turn is subdivided into two sub categories so one is company builds something for CERN a magnet and in so doing they gain know-how so the whole society benefits from this because there is extra know-how from the society and this you can estimate by using procurement data in a way which I'm going to explain a little more in a second and the second is simply you develop something and make it and you give it for free so for example there is some software which was developed as CERN and then given for free and of course there are things like the worldwide web but we did not include that in the analysis because you know that's a one-off thing you cannot count of that happening so let me explain the first thing procurement activity at CERN for procurement is huge I mean there are 4,000 suppliers from almost 50 countries, tens of thousands of orders, billions of Swiss francs Swiss franc is more or less like one dollar so which is more or less like one euro so billions of dollars or euros you can have a timeline of these and then you can make an analysis which we did at the time based on a model but we subsequently did based on actually analyzing the procurement data which was given to us by CERN, by the CERN procurement office and there is some multivariate model which is basically a system of coupled equations when you take companies that work for CERN and you see you have a bunch of variables you like you know how much they spend in R&D, how many patents they made, what is their productivity, how much money they make we had all these data and then you analyze the time flow of this you solve for the model and in the end you can see you do it both for companies that did work for CERN and for companies that did not work for CERN and also you can classify companies that did work for CERN between high tech and non-high tech you know the guy who supplied super conducting magnets to CERN and the guy who provided paving the roads at CERN building the buildings and you can see that there is a strong positive correlation the firm who worked for CERN has a much higher R&D investment productivity, more patents more revenues but that only happens when it's a high tech firm so that's a quantitative evidence in favor of our model and there are many more studies which I don't have time to discuss which were done by my economist friends, I'm mentioning them in the slides, the slide will remain online so you can check out literature in particular one of these studies is based on taking one piece of code, root which is freely available and trying to estimate how much money society is gaining by having this piece of code freely available. Okay so second big class of benefits, human capital human capital means someone goes to CERN, they work at CERN and they are a student, a postdoc junior faculty and then they move on and they gain something, the society gain something because they've been trained at CERN. Now there are standard ways of estimating this, for example there is this website which you see in the slide called Payscale where you can see how much more money you would make if you got your masters or a PhD at Harvard in comparison to say the University of North Dakota so you can use the same sort of methodology to assess how much money people are making because they went to CERN and the difference of course is that if you want to go to Harvard as an undergraduate you have to spend a huge amount of money and get a loan for that, if you are going to CERN not only you don't spend money, often you even get paid for it so that's an investment that society is putting and for which it's getting a benefit. And again we had a model but this model was subsequently validated by study which you can see here in this slide, it's been published, it's been done by my colleagues who got the full set of data from CMS Experimental Collaboration which is a huge collaboration, 4,000 people and therefore they have a database where they know where people went after working for them and then you can ask questions to their former bosses at CERN you can ask questions to them and you can actually establish what is the salary premium how much more money you would make by having been a CERN postdoc as opposed by not having been a CERN postdoc so that was a second class and let me get to the third class as my time is approaching rapidly to the end, the third class is cultural benefits. How do you estimate cultural benefits? Well that's easy, I mean you know there are lots of people who go and visit say the NASA Goddard Space Center which you see pictures here there are fewer people who go and visit CERN but you can estimate how much money that's worth by just having you know if you went to Disneyland then you have to pay, if you go to CERN you don't have to pay and therefore that's a benefit and indeed you can classify these things by looking at numbers of visitors and this is now something which is recognized it's been strongly recognized recently by CERN who decided to where they decided to construct this new science gateway designed by Lorenzo Piano a star architect funded by external companies like Chrysler, Fiat and which is something which will attract people to CERN to learn about science and physics and like I said you can estimate that using techniques of tourists and then there is a famous public good value or existence value which I mentioned before which you can say well how do we estimate that? Well people have tried to estimate this for example when the Exxon Valdez oil spilled happened there was an international court that had to decide how much worth it was having killed whales in the ocean and you use standard techniques. Now the standard way of doing this is to use willingness to pay and in order so we had some estimate of this and this has been recently validated again by my colleagues who ran a very extensive survey to French taxpayers so the idea they've done it to French taxpayers because they wanted to limit it to one country to keep things under control but they did it to a large database of people using standard techniques and basically the idea is you ask people how much would you be willing to pay to keep CERN alive and you have to do it following some guidelines to tell people what CERN is about because most people never heard of CERN and you don't want to have a bias so you want to take the man in the street a cab driver whoever and explain to him we have this lab doing science and they discovered the Higgs boson and was the Higgs boson and so on you show them a short movie and then you ask them how much they are willing to pay and this was done in a fairly proper way with a sample of five samples of two hundred people each so it's a thousand people and then you have a bidding procedure whereby in order not to bias them you ask them you know what if I told you that you have to have a one euro tax increase because of CERN and so on and you know I skip the details again I can give them to you if you're interested I'm not an expert of this and basically the outcome of this was that if you do the most conservative estimate so you do a bounded conditional average willingness to pay bounded conditional average means that you for example correct for the fact that someone who's richer is willing to pay more and you don't want to assume that people are rich and therefore you really have to give something that is sustainable and believable and the average willingness to pay for French citizen was about four euro per year what CERN costs to French citizen is less than three euro per year so basically we are saying that the willingness to pay of the French citizen is actually more and therefore this is reasonably to be counted as a benefit okay so with this I'm getting to my conclusion like I said just to make sure that this cannot only be done for the past but also for the future it's been done for the Hailloumi LHC and you know Hailloumi LHC was piece of cake because it's similar to the LHC same categories it just came on different times here the counterfactual is easy so you can see here that counterfactual is the blue thing in the bottom right which means CERN without Hailloumi LHC and actually the positive thing is the pink thing and the conclusion is that there is very low probability that again the Hailloumi LHC will have a negative cost so with this I'm getting to my conclusion which is that we claim that we put the cost benefit analysis for fundamental research on a quantitative basis and we also claim that there is strong evidence that there are benefits to society even when the utility of future discoveries is unknown and therefore we estimated it to zero and this comes from technological spillovers human capital and cultural benefits in almost equal proportion and my colleague and friend Massimo Florio actually spent last year on sabbatical writing a book on this which is about to be published by MIT Press so here is the data of the book if you're interested in asking say your library to buy a copy and then so with this I'm coming to the end and let me conclude asking the question but is all this meaningful I mean you know often I've been in this kind of talks before on this work or you know I've been describing this work to physics colleagues and often they say yeah but why do we care but this is not this is not why we do science I mean what's this cost benefit analysis good for you know the there's a fellow who called who coined the term big science Alvin Weinberg he's no relation to Steve Weinberg he's just the same name he was the head of Oak Ridge laboratory in the US in the 1950s and you know when he coined the term big science in the paper he said the particle accelerators are the cathedrals of the 20th century so I'm I'd like to make the point that doing a cost benefit analysis for research infrastructure is like doing a cost benefit analysis of the cathedral I mean the people who built the cathedral like Charter Cathedral which you see in the picture in the middle ages probably thought that they were building it to save souls and you know to send people in paradise or something yet if you read any history book you'll see that cathedrals had a very important economic role in medieval Europe because cathedral cities were the center of markets and therefore around cathedral city the economy of Europe was basically rebuilt and that's where the renaissance come from and therefore you know whatever the reason why the men of the middle ages were building their cathedrals there are some benefits that we can estimate now and which remain so the idea is the following you know maybe we physicists are doing science for reasons which are you know we think probably that we are clever than the men in the middle ages and therefore we are really discovering truth or some fundamental laws of nature and we think that there is a benefit for that but on the other hand of course we have our responsibility to society who's paying for us for our salaries and research and therefore by doing this kind of analysis we are telling them look I mean whatever we do whatever our motivations even for getting the possibility that we may discover developed worldwide web or discover teleportation or the time machine or goodness knows what still you can make a decent case that the money you're giving to us is something you get something back for it's not just you know something you give us for free and that the society is getting this back in terms of education in terms of technology and also in terms of culture and things that the man in the street is reasonably willing to pay for thank you very much ok super can everybody hear me yes ok great super so thank you very much for the webinar it's been great so let's have now the questions with the listeners I have a quick question Stefano so normally like for example here in the states if you want to do a research like for your group you might apply to a grant let's say NSF and then you get the money but in South America there are cases and I'm speaking a little bit about Colombia where this is not like highly articulated like that so in the sense that you want to do your research so what you do is you apply to your own home university for a sort of grant and then you get the money but then sometimes it's very hard because the impact is not seen and in particular if you do theoretical physics it's very hard you do have a recommendation for those cases because for instance yeah like the CERN develops like jobs infrastructure technology but also behind that there is this all also pure science and mathematics that most of like way before so I'm asking you like if you can have an advice for those cases where we are just doing like pure theoretical physics in a small scenario in particular let's say in Colombia where you apply to your home university to get funding so sometimes it's quite hard to say why do you want to study geodesics of this particular space time or something like that where of course it's not going to be equally as nice as you show the data because that's like very theoretical and behind but it will for sure I'll believe it will impact at some point so what would you be the advice to get that funding and then just to show it like yeah that's a very difficult question of course I guess what I would say is two things I mean first of all I think one can make reasonably the case that theoretical physics is part of a greater endeavor which involves national and international labs you know all labs typically have a theory division or a bunch of theories and often for a country which has which does not have much money to invest in science investing in theoretical physics can be a very good investment because you know there is not enough money to say but theoretical physics is cheap on the other hand theoretical physics are needed so in fact you may know in Italy in Trieste there is an institution which is called the International Center for Theoretical Physics which was actually ICTP which was established by mostly by the late Abdu Salaam who had this idea which was precisely that countries with no much money should mostly invest in theory because it's more cost effective the other thing of course then this is subject to the objection but what do we care we want investment to get back to us I guess it can be reasonably argued that there is are we still online? because I saw something that says law physics interrupted the presentation but that's made it YouTube channel so I just I was going to say that some of the benefits I described I think can be reasonably argued to be something which exists even at a smaller scale specifically the human capital formation I mean I'm not familiar with the way universities are funded in Latin America whether students pay tuition or not I guess probably not or certainly to a lesser extent than the US but I would say at the end of the day universities a place which creates and society is actually getting some benefit from this knowledge which trickles over to society and you could try to estimate using this okay thank you very much Stefano and it was a very very nice colloquium so I saw a question briefly appearing on my screen but I couldn't yes I'll repeat this simultaneously so this question comes from our YouTube channel it it comes from Maravine Flores who first is giving us some context he's a post doc in South Africa and he was from the Philippines who they don't have strong high energy physics group there and he would like to know what advice can you give to third world countries who are planning to join huge collaborations like Atlas and CMS considering the financial entry fee that this entails so I look maybe I'm not the right person to ask that question in the sense that I'm a theoretical high energy so that's more a question really for experimental high energy physicists I can tell you what I know namely that both CERN as a lab and big experimental collaborations like Atlas and CMS are very aware of this problem there are people at CERN I think one person is Albert Deluc who was formerly the physics coordinator of CMS and I think now he's in charge of this for CERN were specifically in charge of trying to find ways to do this and I know for example that this guy is all the time traveling the world around trying to manage to get deals whereby people could get some training from CERN in various ways so honestly I cannot tell you what's the best way to do it but I would tell you what I could tell you is to try and be as proactive as possible in approaching CERN and this is actually true not only for CERN but I guess in general with other sort of science projects and since you are now working in South Africa I can add the following that one of my PhD students, one of the PhD students funded on my ERC grant who started one year ago actually comes from South Africa he's originally from Madagascar but he got his PhD in South Africa I know that he was a summer student at CERN by being a summer student at CERN he got some very good training he got some recognition in the community he's now doing his PhD here he's very good I have no doubts that he would make a good career as a postdoc I'm not sure about his future he's a young guy he may want to go back to his home country Madagascar or to South Africa he's a decent person so I would assume that he would want to do that and certainly by being proactive he found the fellowship by looking on the web I guess at the time when he went to CERN as a summer student he discovered that there was this option even though he's not from a CERN member state so my advice would be to try and look as much as possible for opportunities don't be shy and many international institutions and labs like CERN are very happy if you do this if I may add a comment on this based on our experience in Peru being part of ATLAS and CMS will also boost a university's status in the international rankings let's say when you're part of a collaboration you have access to your part of the university is part of the experiment so it might provide an incentive for a university to join of course it's a lot of work for the local scientists to achieve the standards that the experimental collaborations are asking for but if you manage to do it then you can easily persuade the university to pay the fee given the large boost that it will get on the rankings yeah and add to this that I like I was mentioning to Joel about a month ago I was at the LHCP conference which is the major conference involving LHC which this year took place in Puebla in Mexico and it was organized by faculty of the University of Puebla who are ATLAS members I guess and it was obviously a big deal for them at the opening ceremony not only the president of the university was there but also the minister of science and you know local authority so it was clearly a very significant boost for them having 600 or 700 people from the certain community from all over the world go to Mexico maybe Mexico is a more fortunate situation in terms of science funding I have no idea but it was clearly a lot of work for them but my impression was that they done something for them so are there any other questions from the listeners in the current audience I have a question let's say if I can say it because they are making works here if I have enough science so first of all Stefano I like it a lot your I would like to ask you for instance because most of this stuff most of the science funding depends a lot of the politicians that are in the congress in the government so from the activity that you were highlighted like human resources or outreach which are the one that in your opinion could help to increase this political willingness to increase the funding to science or to funding science because in Latin America most of the politicians they are very they don't like to fund to give money to science because they think they don't know what is the potentiality of that so in your case of course in the case of Italy or other European countries because also it's very hard to increase a little bit to increase the budget for science in European countries I don't know your opinion about that well I mean you know from of course I have been lucky or fortunate in the sense that I interacted with for example people at the European investment bank who funded our grant and who are obviously people who in some sense already have a positive prejudice and through CERN we did interact with politicians but then you know politicians who come to visit CERN again typically he has a positive attitude I would say that probably I think the three benefits I described can all be presented sold in some sense to politicians if presented in the right in the right way in the following sense the human capital formation I guess is obvious and I already mentioned that answering to previous question the technological spillovers you know of course I guess smaller country with a smaller research budget will probably not build superconducting magnets but I would imagine that if some company manages to participate in a tender even for some relatively smaller part of CERN I mean as I said procurement from CERN is huge and you know if you manage to get a contract I would imagine that could be a considerable boost both in terms of R&D and in terms of prestige so you know in terms of national pride in terms of being part of such a project I think that's definitely something that the politician will understand and that's also having a market value if you have some I mean you know I'm not familiar with the kind of industry that you have in your respective countries but for example I'm familiar with the fact that Italy is a country which is very divided in the sense that there is a developed north and a much poorer south yet I do know that even in the most depressed regions of the south of Italy the ones which are still a large area of culture occasionally you can find the small startup for example Sardinia which is an island of Italy which is one of the poorest region in Italy for some reason some of the strongest internet companies in Italy came from Sardinia I'm not even sure why but probably there was some local guy who had the idea so I'm sure something similar must be the case in a continent like South America there must be local groups that you can try and get in touch with the other thing the third thing is again the cultural benefit of course if you put it in terms of cultural benefits of willingness to pay it's difficult to explain to a politician on the other hand if you put it in terms of national pride it's much easier to explain after all national pride is just a rebranding of the same thing the thing which impresses us will be maybe different from the thing that impresses a politician but I remember when I went to visit CERN with my economics colleagues for the first time we had a tour from the then CMS spokesperson who was a good friend of mine and that was the time when actually the experiment was open because they were actually rebuilding so you can actually see part of the CMS detector and when you go into the detector there are all these parts of the detector which have little flags for the country where they were built and I remember economics colleague was very impressed by the fact that you could see one piece of the detector where there was the American flag and the Iranian flag actually next to each other because there was some piece of detector was built from a small experimental group from Iran so you know I can imagine I'm not sure whether having a piece next to the US flag would be good propaganda in Iran but you see what I mean you can say look we are part of one of the world's greatest projects and I guess that is something which politicians will understand okay any other questions let's see no questions on the YouTube channel so I had a question regarding the development of local communities around like at CERN you have San Juanide Puli which probably without CERN would have less than I don't know 500 people living there and now it's like boom larger so how does that kind of development enter this sort of analysis yeah that's a very good question in fact it's a funny thing because I still remember that at some point there was some data of the economic impact of CERN that we were trying to acquire and we realized that the people at CERN would not want to give it to us because if they gave it to us it would transpire from those data like you said that the local region had benefited a lot from the presence of CERN but then they did not want people to know that because if that was known other member states would say but then we want to rebate of our contribution to CERN because France is actually getting more benefits than others so so in the end partly because of this this benefit was not really included so you know it is really debatable whether such a benefit should be considered or not partly because this is a bit holding the ground story in other words it's true of course that if you put money in having a lab somewhere then the local community will get a boost from that but then you can say is this really a net benefit at the end of the day I'm paying a salary to someone this person is buying bread at the local supermarket and if he was elsewhere he would spend the money elsewhere so the society as a whole is not obviously having a benefit and therefore after lengthy discussions this should not be really counted as a net benefit because a net benefit is something such that the society as a whole is gaining something from having done this as compared to not having done it and therefore this would not really be it of course be careful that some of this is included in the tourism value that is a true value of course because if people spend money on hotels say in Geneva or in the surrounding of CERN that is counted as a benefit but it's not counted as a benefit because it means that the hotel owner makes money because after all if the person did not go to CERN maybe he would go to Ibiza or to a series or something it's counted as a benefit because it shows that that person is willing to put money in something which is part of his education and therefore he's he's manifesting it's as if he was paying CERN so the hotel is a proxy for the money that that person is putting in the whole project if there was no CERN he would not so it's like the ticket to Disneyland sometimes right, exactly that's what I was thinking right if you build a huge amusement park then the community is around will also benefit but that's not a benefit on the other hand the ticket on the amusement park is a benefit obviously to the owner of the amusement park right I don't know any other question let's check the last time for the last time the youtube channel no more questions anybody else in the audience yeah Roberto in the case of yeah let's say that the funding is more or less sure when a government or the agreement but in the case when there are the where economical crisis how CERN has to deal with that or future projects because a lot of this money is kind of promised money but it's not like hard backup money that the government has to put how this affects the economical crisis the funding of the future project in CERN yeah well again this is a difficult question and it's maybe a question that you should ask to someone who's part of the CERN director of management but I can tell you what I know namely that CERN in many cases had to somehow negotiate a course through rough seas so for example I mentioned this cost overrun so when there was this economic crisis which I mentioned was actually internal to CERN what happened was that in 2001 it was realized that actually mostly the construction of the LAC was overrunning in costs and there was a concrete danger that some member states would want to pull the plug and somehow the CERN management managed to take counter measures one of which was basically then director general who was Luciano Mayani was basically incapacitated he basically established some other board that would check on his spending and then they managed to get this loan from the European investment bank and as to reschedules the LAC was originally due to start operation I think in 2008 and by delaying it actually they managed to well of course spend less money per year in the end they spent more money but per year they spent less money because instead of building it in 5 years we built it in 10 years so we built it on a lower budget so somehow managed to survive but this is difficult I mean I read recently a book about the demise of the superconducting supercollider and tunnel visions and it's written by an American historian of science by the name of Ree Orden and you can see that the reason of the demise of the SSC was really a combination so of course it has cost of a run but the LAC also had the cost of a run but then somehow the SSC did not have it was a project a green field project they were going to build a new lab in a new site so it did not have a strong community CERN has this strong community behind it so it has there is a kind of generation of CERN bureaucrats who trained for 50 years now because CERN was born in the late 1950s who know how to deal with the situation and managed to keep the whole act together while there it was a bunch of people who probably did not know each other and there was a head of the lab who did not talk very well to the politician and who did not talk very well to what in this book is called the military industrial complex in the US so somehow it was much of a political problem rather than so in that sense big labs like CERN or Fermilab even national labs like we have Frascati in Italy or Daisy in Germany often they are more than just a research tradition they may have know how to deal with political accelerators but I think they have also know how in dealing with politicians and making a case for science there are people who have been there who have been lab directors and they started as postdocs working with someone who was a lab director before them and so at the end of the day I think for us scientists having these labs or these communities even theory theoretical field centers is really a capital of science but also in terms of politics often I have the feeling that some people underestimate this I mean you think well the guy went to politics he is no longer a scientist you treat him with contempt well I think many of these people should be treated with respect I mentioned earlier James Sterling who is a person who did important things in QCD and he established the ITPP in Durham in the UK which is now one of the biggest centers for particle physics in Europe and then he became the provost which is basically the actor the president of Imperial College and he was a great scientist I knew him personally but he obviously played a big role in promoting science in Europe and so I think people the choice should be respected and labs and groups that do this should take this tradition okay fantastic any other questions from the audience we're close to the end of our time but we might have one quick question okay I think I think we're done okay great so thank you so much Stefan it's been a great colloquium thank you I'm pretty sure that all the viewers enjoyed it very much so now the webinars go on a break a winter break here summer break in the northern hemisphere and we start again in September isn't it the 11th of September when I'm getting a birthday gift and starting the webinars that day okay great so okay thank you very much and see you everybody in September okay