 Predicting migration and the implications on Europe. We first have two state speakers that work directly with policymakers. We first have Reiner Munds who's been working with the EU Commission for a long time, advising decision makers at the EU level. He's a migration expert with the Joint Research Center of the European Commission. His talk will focus on the future of migration to and within Europe. Next up, we have Holger Zahl, who is an analyst with Frontex, dealing with forecasting, early morning and scenario building, in his day-to-day work and advising EU member states on future trends. So we're interested to hear from him on the latest trends in future developments in Europe. Third, we have Tobias Heidland, who's a senior reger at the Kiel Institute for the World Economy and also an incoming professor at Kiel University, Professor of Economics. Tobias is involved in a range of research projects relating to forecasting, both using digital trace data, but also with traditional forecasting methods, and has recently also been involved in a forecasting project to project flows to Germany. So we're interested to hear from Tobias. And lastly, we have Peter Wendorf. Peter is a chair of the Expert Council of German Foundations on Integration and Migration a well-known and influential think tank and policy advisory board here in Germany. And they have last year released a large report on also migration from Africa. And Peter will compliment the panel really well because she will describe a cautionary tale about what implications faulty forecasts and predictions may have. So I'm looking forward to the discussion and thanks to all of you. A couple of housekeeping rules. If you do have a comment, please post it in the chat box. Please always make your affiliation known and address the speaker directly if you have a question specific to one of the presentations. To everyone else, please mute yourself when you're not talking. And then I will call on individual speakers to make their presentation. So with great pleasure, I'm handing off to the first speaker, Weiner Munds, for this presentation. Please be reminded to keep your presentations to 10 to 12 minutes so we can have ample discussions after your intervention and also after all the interventions have finished towards the end to allow for questions from the audience. So with that, I'm handing over to you, Reiner. Thank you very much. Good morning to my co-panelists, to you, Jasper, and to people listening. Video unfortunately we cannot meet in person in Vienna, but we do the best under these conditions. And so I'm going to share my presentation with you. So I'm just presenting a few scenarios. But in order to understand the scenarios, I think it's good to understand what is going on. So I'm going to use the status quo. And so these are the most important gates of entry and flows into the EU and into other countries on the OECD world as well. So we're having most prominently asylum. This peak, as we all know, in 2015-16, it was then came down back again and we have already seen these figures here when comparing them with the Google Trends thing that we were looking at just a second. Then we have labor migration. So labor migration was between 500 and 600,000 first permits issued in the early 2000s. This went dramatically down as a consequence of the financial crisis and the high numbers of unemployed people in EU countries. State at the lower level recovered a little bit over the last couple of years. Then we have the most important and most stable gate of entry, which is marriage, migration, and family reunion. So the EU has admitted in 2018 almost 800,000 people more than under any other category for marriage and family reunion reasons. And then we have educational migration. Now, what do we know when we speak about near-casting, about the immediate future? We know that this number was higher in 2019, about 650,000, so went back up again. And we'll be lower this year because of travel restrictions and other reasons. We see that the number of asylum requests so far until the end of September is about 36% lower in EU countries than it was a year ago. We can also assume that this number of work-related permits will go down this year just for the same reason as we had it 12 years ago during the financial crisis because unemployment is going up and the demand for additional labour is going down in Europe. I'm also showing these gates of entry because they have heavy implications for the integration of migrants. We know that people coming as labour migrants quickly find jobs or really have job offers, have the skills mix required to be economically active and successful. But we know that people coming as dependent family members, as newly wed spouses, and people coming as refugees take on average about 15 to 20 years till a cohort coming that way achieves the same labour market degree of labour market integration as natives do. The gate of entry has implications for the future of migration. When we look into the future, I think it's important in the context of what we are doing today to distinguish between forecasts and projections on the one side and foresight on the other hand. I'm just showing this the most trivial example. When you look at the Eurostats Europort projections, you see there is a baseline scenario with immigration and immigration taking place and there is a no migration scenario. So there's not much foresight behind this, but these are the two scenarios that are offered here. It's more an analytical scenario that would say what's the contribution of migration, but at least the baseline scenario assumes that the trends as we see them would continue until the mid of the 21st century. So there's two scenarios so that the projection is the forecasting element and the foresight is behind the scenarios. It's important to distinguish when we look at demographic projections. Same when we look at future age distribution, we see that the difference between the no migration and the with migration scenario is in age groups 0 to 50 because that's the years of the projection period and we see what difference it makes but there's no story behind beyond saying what if. Now the way forward in my view usually migration projections are based on net migrants but for the kind of analysis that we are doing here it would be very important to get away from net migration is to model but hard to predict and even less easy to understand because a net migrant does not exist, has no age no gender, no human capital. Technically speaking I think the way forward we would need scenarios and assumptions both on gross immigration and on immigration. Now I'm going to talk about five scenarios four of which have been developed and mandated by the so-called ESPAS process which is the European Strategy and Policy Analysis System which is a joint exercise between the Commission, the Parliament and the European Council and so let me just start with scenario number one. So scenario number one that we will go back to a situation that we had in the early 2000s we just remember the second slide that I have shown with the flows this would mean external borders are more or less under control mixed flows continue but there will be no dominance of asylum flows family reunion marriage migration and humanitarian admission in total would be more important labor migration but labor migration would still play an important role as an aging native workforce require the recruitment of labor and skills from third country no additional barriers to labor mobility within the EU would exist so we would have a continuation of the situation that we had since the year 2000 existing integration deficits would be likely to persist because labor migrants would not necessarily be selected but definitely the others coming to the EU there would be no labor market test or no test whether their skills would match the requirements of domestic EU member states labor markets now I have introduced a scenario number two that was not developed during the SBAS process in the past which relates to the a protracted corona crisis the first two assumptions about the future would remain the same that borders are more or less under control or even more closed than in the past and the mixed flows would continue but we would see that family reunion marriage migration and humanitarian admission would stay dominant but might be reduced because of increased border controls and other reasons that I'm happy to discuss here labor migration would be drastically reduced due to high unemployment in EU countries just as it has happened with this drastic reduction that we have seen between the years 2008 and 2012 when it went down from 600,000 first permits for employment and economic activities in the early 2000s to just 200,000 a year between 2012 and 2015 there would be travel warnings and selective border closures creating barriers to mobility within the EU just we are having it currently for example, Hungary not allowing its citizens to leave or when they leave not to allow to come back without going into quarantine border commuting for work almost impossible and the integration deficits would grow as people with migrant background are more likely to be unemployed and to be harder hit by a protracted corona crisis and we would see cases of immigration of people going back home including third country nationals a scenario number three would assume instability in the neighborhood in cases we have seen with the Syrian refugee crisis which would come from a violent conflict or in the future from extreme weather conditions producing large flows of people seeking protection in the EU this would add to existing integration challenges it might lead to a situation where there is negative sentiment with migrants leading to greater electoral scores for political parties advocating restrictive asylum and migration policies and large numbers of migrants could spend years in limbo in a situation where it's not decided whether they are allowed to stay or where it would be decided they have no right to stay but there might be no possibility for return and readmission scenario number four if EU member states would become more selective in their admission of migrants which would be a shift from present admission criteria with strong humanitarian elements to a stricter skill space selection of labour migrants as we know them in the Australian and the Canadian immigration and selection system based on points or a shift or more demand in the employer selection as it's existing in New Zealand and has already been tried in the past by Sweden this would also probably require a more restrictive handling of humanitarian admission as countries like Australia, Canada and New Zealand are selecting refugees that they are resettling onto their territory in contrast to the EU countries where you basically have spontaneous arrivals of asylum seekers the economic gains from migration under these conditions would be larger whereas integration challenges would be smaller and it would be easier to make a positive case about migration vis-à-vis domestic electorates a fifth scenario outgoing natives public opinion growing more skeptical or even hostile to the admission of foreigners a general political consensus effectively leading to much lower immigration higher return rates as well as higher return rates of already established migrants and the general climate that would under these conditions most likely also reduce intra-immobility and incentivise immigration at the larger scale and the main challenge here would be managing demographic aging gradual population decline and the shortage of skills without recurring on important migrant labour we know at least one developed economy that has decided to embark on this path and this is Japan so the five scenarios are not mutually exclusive EU member states can and will have different migration policies and trajectories as they are autonomous when it comes to the admission of third country nationals both in cases of asylum and in cases of labour migrants we will see probably diversity and so some countries can follow one scenario others more another and scenarios could materialise consecutively so scenario number two which is a kind of a near casting scenario assuming what will be the likely effects of the COVID crisis this year and next year maybe in the following two or three years this could end already in a couple of years and be replaced by another scenario so dear friends and colleagues this was it happy to discuss just this presentation and all other questions that might arise thank you very much Reiner for this great intervention I have one question for you discussion point for you before we move on to the next speaker actually two questions if you can answer them briefly so one is the scenarios that you presented are very interesting what are they based out of sort of where they based on discussions with decision makers, politics makers stakeholders or sort of how were they developed basically and the second question was on family family migration basically and I was wondering isn't this particular channel of migration easier to forecast to predict to the future because for people to be reunited you need somebody in Europe to apply for family unification and countries likely have information on that so I want to get your opinion or your expertise on whether family migration is being forecasted quite successfully and where the countries are doing that and whether it is easier than other forms of migration first question I have already said the scenario number two has been inserted by me just recently because when we were doing this developing these scenarios over the last five years for the SBAS report 2019 the COVID crisis was not at the horizon as we all know the other scenarios have been developed in the context of what is called the SBAS this is a foresight exercise that is a systematic foresight exercise that is done inside the commission and the parliament with some help of the European Council there is dedicated units that are working on that so that is based on expert discussions inside the commission with some feedback from external experts so I have developed these scenarios and they have been refined in expert discussions with colleagues okay so family reunion I think it is important to understand there are two completely different types of family reunion the one is the classical family reunion that emerged from the guest worker model you had one key person being allowed to work in Europe or recruited usually recruited and then some of these persons established themselves in Europe and then they were allowed to bring their dependent family members spouses and dependent children below the age of 18 with them as soon as they had established themselves as we have not been recruiting many labor migrants in the recent past this model today mainly applies to recognized refugees so if you come as a refugee by yourself and you have still family members living in camps in Lebanon or in Jordan or back in Turkey then under certain conditions you are allowed to reunite with them in the country that gave you a sign it's more difficult for people who only get subsidiary protection where in most countries that's more cumbersome but the most important form of family reunion today is marriage migration this marriage migration takes place between diaspora members mostly with EU citizenship living in Europe second and third generation longer and distant relatives or distant acquaintances in the region of origin of the grandparents or great grandparents so we have diaspora migration between certain parts of Turkey and Germany between certain parts of Morocco, Algeria and Tunisia to France to Spain to Belgium and this is not so easy to predict because the question whether this marriage takes place or not is one question and the other is where would the couple reunite and diaspora member marries in the country where he or she lives among peers or even into the mainstream of the dominant society or whether you get married to someone who is living in the ancestral land of your grandparents is a decision that is not easy to predict thank you very much Reiner if you do have additional questions for Reiner please just post them in the chat box we will be able to follow up on some of them later in the discussion or we'll also transfer questions to Reiner later thank you very much Reiner we now move on to our second speaker Holger from Frontex Holger over to you you've muted still I think there we go hello from Warsaw sorry I did not unmute myself there well I want to express first of all my thanks to the organizers for having me it's a real pleasure to be with you even in this virtual format for sure I would have loved to trade Vienna Warsaw for Vienna today as it is I want to talk to you about latest trends of immigration in Europe and I will now try and share my slides with you which I believe is hopefully a success yes we can see them well okay thank you so my presentation is less conceptual and maybe less technical and it's more or less about what went on this year in irregular migration the year I believe we could have all done without probably what I want to do is I want to take us back to the very beginning of the year and recall that the year started off with strong migratory pressure in comparison to the same months of last year and this is an almost all the major migratory routes to Europe so January and February saw almost 50% more detections of illegal border crossings reported compared to the same two months of 2019 illegal border crossings I know it's quite an extra audience but maybe for those who are not so familiar it's our key indicator for irregular arrivals at the land and sea external borders rather than about this time here at the beginning of the year I was working on the risk analysis for 2020 the annual risk analysis and I suggested for a variety of reasons such as for instance various destabilizing trends in key countries of origin factors like the persistent risk of mass drought flows from the Idlib region of Syria as well as for instance a slowing rate of decrease of migrant arrivals in recent years I suggested that 2020 could well see the turnaround in the number of arrivals of regular migrants after of course the peak of 2015-2016 we've seen falling numbers and little well the storm was already brewing we started to see on CNN how Wuhan region was closed down but things turned out quite differently as we can see here on the slide due to what I think most of us did not have on the immediate horizon due to the COVID-19 crisis for us the pandemic started more or less in March if you recall in Europe and the containment measures introduced by our member states but also third countries and most importantly key transit countries they led to a decrease in arrivals of a very drastic decrease of over 3 fourths in April compared to March into April until May all routes reported a sharp decrease in arrivals except the western African route so that's to the Canary Islands I'll come to that in a minute because this is a quite interesting effect in fact of the containment measures then the trend after partial and gradual listing was at the end of May we saw a slowly resuming flow reported to all borders now here the map some of you may well have seen different versions of this map it's a standard staple in a risk analysis presentation by the agency which shows the main migratory routes at the EU external borders rain of course land routes, blue to sea routes and here we have an overview of detections of of illegal border crossings beginning of the year to the 15th of September per routine comparison of the same period as 2019 if you would sum them up here we have seen approximately 63,450 detections and if you did the comparison to the same period of last year that is in fact a decrease of around 16% we do see here that's the point of the slide really we do see here an uneven development across the different routes and that's partial of course due to uneven restrictive measures across countries along these routes and what I really want to do here I want to focus on the overarching effects of the pandemic on irregular migration and all of these effects that's how it fits into the day of presentations all the effects we probably would not have seen foreseen in the beginning of the year the one route that did not see the described dip during the spring was as already mentioned the western African route that's migration to the Canary Islands by boat along ongoing phenomena at what counts towards our external borders and that's because containment measures by the Moroccan authorities have meant that migrants from mostly West Africa were unable to cross through Morocco or fly into Morocco to leave them in the north by boat via the sea or the straits of Grebaltar what we call the western Mediterranean route so some of the migratory flow has been in fact re-routed from the western Mediterranean to the western African route and that's why we have such a strong increase here in arrivals to the Canary Islands also a very widely seen sector at the external borders was a change in the modi operandi of people smugglers and of course the audience is well aware that a grand share of migration here happens what we call facilitated with the help and employment of people smugglers and this modi operandi changed to avoid detection and with that crucially a mandatory quarantine that was imposed almost throughout the external borders so for instance in the eastern Mediterranean we've seen an increased use of speed boats that can avoid interception speed boats often times used in other instances for drug smuggling the targeting of non-hotspot islands was a large phenomenon in the eastern Mediterranean again trying to avoid detection we also saw an increased use of leisure boats like sailing yachts which do in the summer traffic they melt in sort of the summer traffic in the Mediterranean in the western Balkans in this regard we saw an increase in the use of tunnels the use of false documents of course always a persistent phenomenon migrants hidden in vehicles clandestine entry attempts increasing as well another phenomenon really is a visible increase of economic migration from North Africa Tunisia and Algeria were propelled to second and first most common nationality of migrants so far this year this is of course an effect primarily of the tanking of the tourism industry in either of them during the summer it became a bit better towards the fall in terms of conditions there for main sources of income in Algeria and Tunisia there was also an interesting third effect of poor factors there was some stops in readmissions of certain nationalities by some member states and in general of course returns of migrants without a right to protection almost stopped during the first months of the COVID crisis and because of these last two factors the central Mediterranean saw the higher pressure as seen here on the slide then during the summer of 2019 now in the spirit of I really like this earlier remark of attempting to be vaguely right I want to say something about the short term future in the coming months we do believe migratory movements will be highly dependent of course on the evolution of COVID-19 containment measures in countries of origin transit and in member states themselves rising infection rates right now here is a relatively current ECDC map they may well signify return to thermal restrictions of movements and thus bring down numbers once more the effect will be different in different regions for instance in the western Balkans the share of transiting migrants staying in migrant camps is higher than in other regions and what we saw already earlier this year is that if these camps then are under restrictions are under lockdown the effect on attempts to cross the borders of the western Balkans were much more affected than elsewhere on the other hand as we approaching winter a decrease in migratory movements due to seasonality and weather conditions they can also be expected as always but it must be noted that any change in the migration policies of key transit countries could quickly override of course the seasonal pattern of arrivals we remember the events of late February as just a tale of caution here the attention to containment measures may well further exacerbate the determination of migrants to leave before the new restrictions are enacted smuggling networks as always will stand by and look to profit from this and use different solutions to avoid border controls as we do assess that migrants stuck in transit in certain countries have not yet abandoned the original travel and destination plans considering the numbers so far similar it would appear that the presumed backlog of would be migrants stuck still in transit has not yet been fully drawn down I think I'm sort of close to or maybe already exceeded my 10 minutes so given that we have a lot of experts here speaking about carrier building the long run I will probably stop here they will present much more informed prognosis about the longer term future than I could and with that I will yield the floor back to Jasper, thank you Jasper. Thank you Olga much appreciated also for sticking to the time there and thanks for the interesting presentation quick follow-up from my side from Frontex also in the business of forecasting let's say much of the data that you presented and our current now cast from this year and trends that will likely lead to future and that give us a lot of ideas about what is to be expected Frontex is also doing different ways of actually looking to the future using data and using experts to elaborate on the general approach that Frontex takes there and how EU member states are using those the products that come out of these forecasting attempts. Sure and thank you for the opportunity to switch from let's say the now casting to also tell you a little bit about the work of the agency in forecasting and foresight we do in fact both we have been doing a it was a trial in effect of forecasting it was a three month horizon per migratory route of course for an operational agency and for that we looked to the allocation of Bordeaux resources primarily in line with our operational mandate currently we're in the evaluation phase of that to see how well we did if we really used the resources wisely if we did enough to bring in experts from member states and so on then foresight is something we do in in fact it's mandated now by our new mandate you may know that the agency receives a new mandate at the end of last year we like to call it the 2.0 regulation of the important course card agency it is an exercise of foresight for a 10 year period and it is the beginning of a multi annual strategic policy cycle for integrated border management based on this analysis in terms of methodology it's very much what others have presented starts with a literature review goes into mega trends and semantic risk assessment scenario building at the result this product then goes to the commission to develop policy priorities and strategic guidelines for a 5 year strategic policy cycle on integrated border management and so we are now sort of at the stage where we have produced this product and it's now in the commission's court to start on the basis of our foresight strategic policy cycle very interesting insights there and more to context in those aspects if you do have additional questions for Holger please drop them in the chat section of this webinar or email us directly we'll be happy to get in touch with Holger again for any follow up questions I'm now handing over to our third speaker Tobias Heidland is talking about COVID-19 and future migration flaws Tobias over to you we can see the slides we can't hear you yet okay I think someone turned my mic on sorry the control panel has gone so when I went full screen so thank you thanks for the nice conference in my time input I'll start disentangling the most important impacts of COVID-19 on migration flaws I see them from a micro level and in this exercise my thinking is guided by a couple of the models that we've seen quantitative economic and other disciplines migration forecasting models and also a lot of work with individual level survey data but I'll do without any formula any numbers because I don't think that the quantitative models can at the moment give us a clear prediction about what's happening I think the models the more rigorous quantitative models I think so when thinking about migration flaws I think it's crucial not to consider these flaws as a force of nature I very much like the second point that we shouldn't be thinking about net flaws rather about gross inflows gross outflows because that's the decisions that people take migration is made up or migration flaws are made up by individuals who take decisions based on the circumstances they live in or the situation they have policy framework that restricts the choices and in this context where we can think about the impact of COVID-19 now I'll be thinking about the next years not so much about the very very short term that we just saw in the Frontex numbers March to June effect or the slowdown in mobility that Liz Collette mentioned in the morning so to start off let's talk about what I think is the most likely economic scenario in my view we'll see the situation where after the crisis the different policy responses towards the disease will have led to a highly heterogeneous economic damage across different countries and probably all countries will suffer somewhat but currently the economic damage differs quite substantially in the second quarter the annual life drop in GDP in the UK was about 20 percent was about 20 percent in Japan and a few countries are doing rather well and wonder for example still expecting positive growth this year and these economic aggregates hide the fact that within each country the impact of these losses differs a lot by profession by economic sector and so on so as you're probably all aware there's some industries in large part of the public sector where we haven't felt any risk effect we've worked longer hours than before but there are others especially the self employed especially informal workers in sectors such as hospitality who have been hit very hard and we should look at the likely effects within country as well when thinking about what's going to happen to migration flows from certain countries so it's the typical migrant from a country harder hit or less hard hit or did the burden of the shock just fall on those who typically would have stayed put anyway and we shouldn't see a response in the flows so some of this economic fallout will last for years and while in many areas we'll hopefully bounce back very quickly in economic terms probably where vaccines are available earlier that will go faster and other places will take longer my fear is that especially the poorer the not so well governed countries in the world will lose out and the health toll the larger economic damage and also specifically targeted bilateral immigration bans on these countries will persist for longer so to guide our thinking I find it helpful to consider four stylized kinds of households which take migration decisions and what happens with migration intentions and also the ability to migrate and what also matters for the migration decision is of course the demand for labor but as Raina Moon has just shown us in the initial graph is that labor migration is just one flow among many so let's consider households that are more or less affected by the economic shock and who live in richer and poorer countries so first let's start with high income countries here for people who are personally unaffected by the economic toll hardly anything changes beyond the very short term so people might be a bit unwilling to risk moving to another country in the health crisis but apart from that we shouldn't expect too much of a difference in chance for migration outcomes by contrast the household is economically severely hit and out of economic necessity they will have more reason to send labor migrants in the future and importantly if we're thinking about say EU countries then the preferred destination countries for these households are typically still in reach because there are few policy barriers in place to restrict their movement so while the preferred destination country may change people might not go to the UK and rather they're unlikely to be a really large reduction in migration flows from certain countries in the very short term or in the short and medium run apart from these effects like unemployment and lower demand for labor in the destination countries however what we can expect is probably that these divergent flows so people go into other countries than before the crisis may also create some persistence through network effects and that may make migration of the beaten path easier for others who follow these corridors in the future so the countries that get through the crisis well I expect to benefit in attracting more labor migrants than before the crisis as the economy goes back to normal okay so let's move to poorer countries origin if someone lives in such a country is personally unaffected we should expect the crisis in destination countries to reduce their incentive to migrate well off at home and migration becomes less attractive because the destination countries have lower demand for labor that could decrease flows and again there might be this diversion of flows to other destinations which got through the crisis better and a group which might be severely affected is households who are personally affected by COVID-19 for them there's more reason to migrate but credit constraints may become binding so migration may become impossible for them due to the economic and sorry what we could see is a greater inequality in migration outcomes depending on how different countries and that people got through the crisis that would be probably greater inequality between countries in terms of the economic outcomes in combination with very specifically targeted bilateral short-term immigration policies aimed at reducing the risk of contagion with COVID-19 this is I think likely to lead to greater inequality in migration outcomes as well and as we've seen the incentives can change both within countries and also between countries and the ability to migrate can develop differently for different countries so I see a particular risk for losing out on the gains of international migration for emerging and developing countries that are heavily affected by the corona shock and the most important factor in the medium run will then not be the individual incentives to migrate but policy changes to certain destination countries I think so yeah to give you a bit more detail on this point from quantitative models what you usually get out is that immigration policy is the most important lever when it comes to determining the size of flows given the size of the people the amount of people who want to migrate to a certain destination and migration policy is not a push factor as it's often presented in the media or discussed by politicians it's rather a very restrictive filter which decreases the number of people who get indicated here by these triangles so only a few percent of those who would like to move permanently to a country actually do so in a given year and the most important reason for that is restrictive migration policy or any forecasting exercise it's crucial to understand what future immigration policy would like and would look like and without this information or quantitative estimates of future flows are highly uncertain and what will happen to migration policy well what's realistic is that some of the trends from the literature will persist and what we know from research on public attitudes is that in economic crisis immigration is more restrictive and it's not only that those people who are personally affected who have more restrictive preferences in a crisis it's very important to realize that social economic social tropic concerns matter that preferences that are shaped by the expected effects of an immigrant inflow on others in persons in groups for example we're worrying about the competition effect of immigration on some co-nationals that I don't know personally even although I myself might even benefit from this immigration economically so the social tropic concerns tend to be more important than concerns about individuals own personal economic situation and the individual voters of course not the only important factor in determining policy what depends on political discourse, the media environment the political economy and respect of countries so my personal expectation is that there would be more polarization and public attitudes towards migration over the next years both again within countries and in regions such as the EU that might lead to changes in immigration policy that then again affect future flows so key points in a nutshell once again first from considering the effects of the 19 on migration flows we should consider both the between and the within country changes due to the crisis secondly in most areas of migration we should not expect less willingness to migrate because economic pressures will if anything increase incentives to migrate for part of the population and finally the most important determinant for the size of the flows is and will be in the future the migration policy of rich destination countries and this is also the greatest cause of uncertainty when we try to predict migration flows thanks and over to you again Yester thank you very much Tobias for this very interesting presentation again I take the privilege to launch one the question of my own first and then later we will discuss with the whole group but I saw sort of two points that you made quite strongly presentation hand it's important to look at the micro level so individual people making decisions that are driven by certain characteristics and now in most forecasting efforts that countries undertake them they usually do not deal with micro level data information about individuals often they deal with flows right so many people arrived at the border or so many people applied enrolled in a service or applied for asylum applied for work permit and so forth so we talk about most data that is used for forecasting actually at the micro level so aggregated data now I know you've been involved in different forecasting exercises can you give us a sense of how including the micro level perspective including additional information rather than just past flows in migration models can actually improve them and do you think this is the way we should invest further in the future? Yes excellent question thank you so there are different ways of creating micro founded migration models and at least for the for the short and medium run a way of a way forward that I like quite a lot is gravity style panel models so gravity models were mentioned in the very beginning as potential way forward so these are models which include structural explanatory factors such as the distance between countries shared languages these kinds of factors and these stay relatively fixed in the short term but they interact a lot with other drivers so models which come from this micro perspective think about the individual incentives for migration for example I as a potential migrant have just lost my job in the country of origin and now I'm considering the situation was to stay whether to migrate internally whether to migrate internationally where to go I think that's the way forward of modeling these in these short term forecasts and I think the way forward here is to combine surveys such as the Gallup Workflow where you have information about the the willingness to live in another country or also plans, concrete plans about migration with the flow perspective and what we know from for example Jacob Bidzak presentation about the issues in forecasting and what we should arrive at eventually in the next couple of years I think is a model where it's kind of model classes where we separately think about the incentives for individual migrants and then how these interact with the with the policy component and what is still missing to my knowledge from from modeling is really dealing with the destination country choice so how are flows diverted I think this is really the the thing to address in our modeling and here I think it's crucial to have the micro perspective in to think about individual language skills for example who can be diverted realistically from one country to another and who will be unresponsive to unemployment in a French speaking country because they really want to go from French speaking countries for example right thank you Tobias so a strong argument to go beyond just looking at who came in the past and extrapolating that trend into the future but actually taking into count various sources of data survey data and others also on the micro level so on the individual level decision making level let's say to arrive at better predictions for the future I wanted to ask one other follow-up question and yeah and like you answered fairly briefly you also the second strong point you made in your presentation was the significance of policies in their impact now there is a little bit of a controversial discussion around in what the policies actually do can border control policies restrictive policies actually affect flows or will flows just be diverted into other channels that might go that might not appear in official statistics right so some policy maker might be worried about the power of their own policies they're putting in place now you're suggesting that there is evidence saying that the policies are the strongest factor when it comes to flows and that looking to the future future policies are likely to shape most significantly the flows that you policy makers can expect here can you briefly comment on that on the power of policies and the confidence in the legislation that they could have placed well yeah so I think what is important for policy makers is coordination with other countries especially at the new level because that's the risk of diversion I absolutely see that and it's difficult to pin this down in quantitative models so far so a shout out to not only people at GMDAC but also maybe statistical offices policy makers who might be listening would be fantastic for us researchers for these kinds of modeling exercises would be monthly data which is standardised across different European countries so with this kind of data it would be much easier to hand code for example policy changes in a future research project at the monthly level as well and we could see these diversions of flows not only in yearly data where it's really difficult to claim causality but at a finer granularity and if Eurostat or any other institution could provide that in the future that would be great service to all migration forecasters and migration researchers thank you very much Tobias I hope many policy makers at home in the offices were listening and surely they also can get behind that goal of arriving at more fine-brained granularity data which is of course hard to combine thank you very much Tobias for this intervention we'll now transition to our last speaker in this panel Petra Benno from the SVR in Germany and Petra will provide a sort of complementary perspective on the use and relevance of these different approaches and the potential harms they can do so I'm excited for Petra's intervention Petra over to you if you're ready we can see your slides yet we can see your black screen very sorry can you hear me yes we can hear you well I'm sorry but you can't see me I don't know what happened actually well I just begin talking maybe you can see me afterwards again anybody can help me because they've come home jumped out of the no problem at all Lucas who is helping us with technical issues Lucas do you have Petra's slides on your computer as well and would be able to share them and then Petra you could tell Lucas to forward each slide when you're ready I don't know Lucas he is in the process of sharing his slides thank you so much Lucas can you see me now perfect Ben so dear colleagues thank you so much for having me here in this inside full conference and also in this very inside full panel what you see on the slide is actually my university affiliation so I'm here what we call in German with two heads on I'm also a university professor at Erlangen University and as such I can anticipate to you maybe and invite you to follow up on a scenario building which we are doing in my team at the university of Erlangen Nürnberg it's a scenario building that we are doing for the next few months on migration and integration amidst the COVID-19 mix with regard especially to German migration but also integration policy so it is very close to what Tobias and also Reina have just presented but we dive somehow more into the details of immigration in just one country that was also part of the ideas Tobias presented to drive more into each EU member state which is not what I've been asked to present right now I couldn't either because we're still in the process of doing so but I can invite you to read the publication that is due in January next week and supported by the Mercato Foundation in Germany and what has been asked is to take one step back and to have a sort of critical look on the role of science science can and should play in foresight and I will maybe start with what I've called Alamism following by the question what we really know how to deal with what we know and what we also don't know and what we should aim at so Lucas if you could present the next slide thank you so Alamist warnings in projections have lately gained certain popularity in Europe we're still talking about Europe and let me just give you the example raised by Stephen Smith two years ago it's quoted now in the English version of last year but the original stems from 2018 and then Smith warned that a quote young Africa would relentlessly make its way towards the quote the old continent that is Europe and that in just more than 30 years a fourth or even a third part of the European population would be of African descent if we were not able to fight poverty in absolute numbers this would mean that 150 to 200 million people would be living in Europe then in 30 years who had come from the African continent and this hypothesis stems from the idea that a big and increasing number of young people in their countries of origin in Africa do not find economic perspectives and therefore will forcibly make their ways towards the north alarming projections like this however are not new if you remember some of you might remember that by the end of the 1980s and beginning of the 1990s warnings of the potential new period of so called quote great migration have been issued already even then but particularly today I think we have to be very very careful with such alarmistic projections because against the backdrop from propagandists of the coming quote big population exchange a right wing populist or even extremist instrumentalization of such projections is realistic so we should be very careful and we carefully have to avoid parting from assumptions that the demographic explosion alone would be unavoidable of projecting that even if democracy develops as projected by Smith this development then would necessarily lead to more migration we also have to avoid to part from a somehow homogenous genius in any case underdeveloped Africa who's population would migrate directly towards Europe and masses neither does the biggest part of migrants come from the poorest countries since these lack the resources to even migrate let alone in masses nor do the people from countries with high birth rates automatically migrate towards those with lower birth rates from poorest towards the richest from the most populated countries towards the lesser populated regions from the tropics towards the climatically more moderated zones and finally from the younger to watch the older areas so this is something we have to bear in mind not only with regard to projections but also with regard to scenario buildings this of course does not mean that demography wouldn't matter but to remain within the example of the migration from Africa we would then have to state that on a global scale there's no such thing as an exodus from Africa the biggest share of African migrants remain on the same continent and also has to be distinguished between different African regions so if there is intercontinental migration at all Europe also is not the only destination among others something that maybe sometimes get out of sight and finally there's no not only a regular migration on flight but also migration into the educational and skilled labor systems to speak with François Errand is also quoted on the slide this projection, this myth projection includes a lot of economic speculation and to continue with Errand the task of scientists consists neither in alerting authorities and the public nor in appeasing or pleasing them scientific ethical responsibility rather consists in informing it has then to represent the complexity and the very different starting positions of migration movements and motives the different consequences for countries of origin and also for the receiving countries so this complexity also has to be considered in political consulting in order to show how political and legal steering potentials should react to it Lucas if you could change the slide please given the very different theoretical models and preconditions we have also seen here and given the different estimations and projections researchers help themselves often with scenario buildings and we have seen that in terms of uncertainty especially in this panel where we have talked about Europe and possible effects of the COVID-19 pandemic when it comes to migration from Africa to still remain in the same research area the joint research center of the European Commission use the topic of many more to come migration from and within Africa some two years ago and try to draft three scenarios of migration instead of projections and they part from four parameters identified as the main variables for migration of course there's also demographics but it includes also the socio economic conditions the ecological and climate induced factors and political stability while stating that a rise in migration is quite probable from Africa GRC states that it is impossible to predict if this rise will affect mainly the inter-African mobility and if so if it will also affect other regions and continents so in the end actually there's very little that we can say for sure and disappointingly enough this is of course particularly true for the unexpected events like COVID-19 can you pass on the slide and even more so the study shows that even polls and surveys that have also been several times mentioned here and which for quite a while have aimed at predicting the motivation to migrate in young Africans have proved to have only limited predictive credibility these surveys are actually as Ravenna sourced underlined in her introduction they are quite instructive when it helps to determining a feeling of those wanting to migrate however in no way can they be taken as a realistic source for the migration processes to come as an example we might quote the Gallup poll on migration intentions during the period of 2010 to 2015 and during this period around one-third and one-fourth of all Africans about 15 years that they wanted to migrate remember the Smith's projection finally what we saw was that only between four and seven percent of those who said they were motivated to migrate actually did so did make plans to do so even and only one percent actually undertook any action to actually migrate to another country leading to a zero point twelve percent who really migrated that is the intention to migrate uttered in the interviews does not give us a real hint as to whether this migration actually will take place so what do we really know about the motives that cause people to leave that country of origin decisions to migrate on the individual state as Tobias had tried to sum up not all just caused by demographic developments neither are they only the results of economic trends or only caused by climate induced or political developments it is in fact the entanglements of different motives that make individuals or groups decide to migrate or or rather to stay and this complexity makes it so difficult to reach to do research about migration courses and drivers or triggers and certainly makes projections for current early warning systems so controversial so I think we should also be very clear about what we do not know to show where we lack data away data is not regularly and systematically raised and to be sure there's a whole lot of data missing from migration within and from many reasons but even if we if our data are more often than not insufficient or inadequate we might analyze what we know and which consequences this may have but be clear then in highlighting a lack of knowledge and also uncertainty and disagreement among among researchers as Uzan Emelder has pointed out at the same time and as we have learned today we have to deal with a lot of uncertainties regarding the complexity of motives implicit assumptions and theoretical models insufficient data and future shocks like the one we are experiencing right now we also have to face different ways of interpreting the data we arrive at and we certainly have to face very different methods and disciplines so this is also true for me for the growing gap that I see between forecasts based on technical means and we have heard some of them in this morning and those that are based on empirical social sciences rarely do these to meet each other and help to understand different coins of maybe the same metal but different interpretations and also there's already a lot of early warning systems forecasts for sites and projections going on in Europe be it with the intention of providing better reception conditions as some of the models we see in Switzerland with the intention of financing political measures with regard to integration as in Sweden for instance. Then we do have early warning from Frontex we have seen one before from EASO from Broadford Development from UNHCR etc that is to say we do not lack research per se but what do we need then we certainly need better data and research relying on these data of course we need more networks among these different projects and existing databases and definitely more international interdisciplinary cooperation of which this meeting at least is a good start and this is why the German expert council that I'm chairing has supported in the German presidency the EU's initiative to build up a new network that coordinates and validates the existing research in Europe only then can we proceed to build up common and knowledge based policy recommendations thank you then for listening and keep updated with our scenario buildings and also with connecting different networks in existing in Europe and I give the floor back to Jasper then question back to you that I've also seen the comments made in the chat book so you were referencing the Smith book but there are also other economic studies and others that see or that estimate growing demographic migratory pressures among African countries that could at some point spill over to the EU even though currently it does so in a very limited way and I see in your argument as well that you know it's very difficult it's tricky or controversial to forecast these sort of patterns that are in the data because it could provide ground for populist movements that are spurring critical views on migrants possibly increasing the risk of xenophobia increasing the risk of anti-immigrant violence and so forth so the question behind that for me is in a way do we use bad predictions that we know are not the best or do we use no predictions right? Is it better to have an estimate that we know is not accurate or rather use no estimate at all because of course the policy world is placing a lot of demands for evidence-based policymaking the demand for forecasts the demand for data is there the data that is used is the officially when data in many of the projects so what do you what do you make of this the risk of having possibly faulty predictions on the one hand because there's so many unknowns while at the other hand scientists and researchers like yourself of course always urge policy makers to use the data that is out there and to base their decisions on facts what do you make of this dilemma? Of course it's a dilemma but what I was just trying to say is that it's a sort of call to be honest it's a call to be honest about what we know what we don't know about and especially to be honest about complexity of course we have to reduce somehow we have to reduce complexity but then be honest about if you use demographics as the only driver in projections this this would be to underestimate complexity of motives complexity of drivers complexity of the importance of policies as Tobias has just underlined etc so so it's basically a call for an ethical understanding of projections even scenario buildings and also of an ethical approach when it comes to political consulting especially I was still muted so thanks for this response appealing on that question how do you see also the role of media who's often sort of this interlocutor between policy makers and then researchers and researchers they release a study and then what happens to those results how they interpret it is often can go into different directions and how do you see the role of media here instead of researchers and policies policy makers do a better job at not misunderstanding each other of course it is very difficult to break down very complex highly complex facts and figures to to the media but then we do have very good media who try to be the brokers in that specialist journalists who have specialized on migration and there are also in Germany I don't know in other countries we do have something like a media deans integration that's a survey from the journalists to other journalists breaking down highly complex issues and findings from researchers so this could be a sort of have a broker in between and also we ourselves as researchers I think we can learn a lot still in breaking down our findings to make them comprehensible to the broader public and just to mention one possible issue one possible I was just told that my internet is down is it am I still worried okay perfect then I go I just go on and we have a network in Germany built up by different institutions that aims at networking between research institutes it is called FFVT in German and it also has the aim to make research more accessible to the broader public and also to the media so there's also from the side of the researchers themselves not only from the part of the media themselves but also from the part of the research themselves the idea to make ourselves more understood and more heard and internationally more connected I can post you the link afterwards thank you very much we would appreciate that I'm looking at the chat box and potential questions coming in Lukas and Karthik please let me if there's anything else coming from the floor otherwise I would in the remaining eight minutes you're open this topic to other panelists on this panel because it does of course affect all of us Rana Munzi worked with decision makers policy makers on a daily basis Holger you do too and Tobias in a lot of networks where you're trying to bridge insights from academia and make them sort of intelligible for policy makers and maybe Reiner if you're still with us how do you respond to that that sort of critique to say there is a risk in putting out projections there's a political risk and yeah in using using these data what would your response be yeah the the attention span of political decision makers is usually small because they have so many things on their plate so if it's not deemed to be really urgent or you're at risk of losing the next elections it's not always easy to capture but Petra has just I mean refer to one of the products that I have produced during my time at the commission which is this report many more to come with a big question mark and this was to show that the idea that millions of Africans would show up the next day in Europe is ridiculous but I've tried to underpin this with scientific arguments in the report but usually you only get the message through so the report serves as a scientific backup to show that it's serious what you're saying and you have good arguments but the argument that you can push through is something that maybe finds place on half of a page or something like that if you're directly talking so I think the important translation and I think Petra was just talking about this is between scientific results an honest way of using them in a transparent manner and then a few messages that you can distill. Occasionally there is a how you say one of the reasons why sometimes the dialogue between the scientific side and the decision makers is not so easy is that scientists need some kind of public they're publishing things but some politicians would be more happy if they would get the messages not I mean where a platform where it has already been published but if they would get them somehow in private would make use of them in a different way so I think there's a certain tension between scientists or I mean funded research that is made available generally and some preferences of policy makers who would like to get the information more exclusively I'm happy to discuss this further although we don't have enough time for that Thank you very much Ryan for this direct response running to you to be as a researcher have you ever come across an estimate of yourself where you're sort of having second thoughts what this estimate could do what reactions it could cause by the media, by different parties or the public and what is your thought process internally to think about whether to release something or not is it just purely what's in the data is good to go or are you thinking about political risks and forecasts and in foresight Well, yeah so whenever you you find something that is helpful especially extremists both in parliaments and say on social media I think you need to be very careful so in the past we've made the experience at our institute when it came to sanctions towards Russia that has been misused and they're kind of turned around in terms of the conclusions so we're definitely highly aware of these issues and I think in a field like migration research you need to be that shouldn't at all affect the kind of conclusions that you make but I think it should create an additional layer of safeguarding and thinking about the kind of interpretations and recommendations you make based on the findings so whereas probably in other disciplines especially in economics but also other sciences researchers can be a bit quick to throw out the results on say Twitter I think what we need to be aware of is that this stuff can be taken out of context can be used for for alarmist messages absolutely with Petra Bender in this respect and that basically increases the level of caution that we as researchers should have Thanks Tobias and big thanks to all of the speakers and thank you very much for being so disciplined with your time as well it was a very fascinating intervention today thanks to everyone who participated out there in this panel I hope you found it interesting if you have any follow-up questions you can still post them in the chat or contact us by letter we will be happy to recontact the speakers and try to get more responses for you we are now going into a one-hour break you can get your lunch and we would be very happy to rejoin here at 2 p.m. CET with the next session on the nexus between predictions and policy making and in this panel we have representatives from EU member states like Germany we have representatives from EU bodies like ASO and we have leading researchers that present different research on forecast methods so we would be happy to see you again in one hour and join your lunch break and thanks again to everybody involved see you soon