'Remote work across jobs, companies, and space' with Dr Bledi Taska

good afternoon everyone and thank you very much for for joining this presentation I and thank you for very well and welcome to the Oxford Market school and tell us I lead a group on technological and economic change here at the Martin school and it's a great pleasure this afternoon to have with us Dr lady Tasca who will help us think about the topic about remote work across firms across space and across jobs which are very much the topics that relate very much to the bridges that we do here at the Martin School and I hope that Bloody will help us and guide us through all these interesting questions it is a global phenomenon it is becoming uh very uh pervasive these days so we look forward to the explanations and the and the data that Bloody will give us I'm also delighted that we have people both joining us here and and online thank you very much for joining and also keep an eye on the events that the obstacle management school has in the coming weeks the structure of this event is that lady will give us his insights and presentation of his research and then there's going to be a q a session uh where I hope everyone will get his chance to inheritance to uh ask questions about this research a little bit of uh background bloody Taska is the Executive Vice President livecast and chief Economist uh and he is a labor Economist who leads a group of economists and data scientists and they look into labor market data real-time labor market data they look into uh interesting questions around skills mismatch label shortages effects of automation reskilling of workers and and other areas his work has appeared in many popular media Outlets as well as top economics journals and above and beyond all I think that is a very generous very curious very intelligent person and we are delighted to have him here and the way he thinks about this complex issues combining these detailed data is um is what we we really appreciate so armed with all these impressive data about the labor markets uh lady will talk today about the more work and over to you bloody thank you thank you so much mandelier thank you everyone for for the invitation for being a great hosts uh it's it's a real pleasure to be here uh so I'm as at least I am I'm the chief Economist I like us which is a real-time level Market analytics firm but I'm also a labor Economist by training which uh at some point went to the dark side uh and now I'm doing research for fun uh so today I'm going to talk about you know something that all of us are you know been aware of for the past couple of years which is a remote work and sort of show you some of the trends and what the data says about it especially sort of from the demand side and the supply side but most importantly I want to motivate a discussion with around two topics that are very uh specific and very Central to weather for Martin does which is the future of work and future development and you'll see that essentially remote work can potentially be very essential about those two and the quick uniforms essentially in terms of a discussion that we can have okay so today this is going to be what I'm going to present in is part of two two strands of work one or more about uh you know Academic Year one done with Stephen Hansen civil Davis Peter lamborghin where we use online postings to understand the demand and Trends across the globe and there's also a more recent one that we're doing with rebellio labs it's a bit more of a policy uh work which essentially will look at the supply side but my goal here today is is to be able to connect these two together and sort of like what what are the potential implications about you know future future work and future of of development so just a brief summary in a more you know academic style what are some of the things that we learn uh well and surprisingly we did learn that remote work has become you know established part of the higher strategy we've seen it increased by 3x in us and actually by way more in UK and also if you have any insights as to why UKC in this I would love to talk to you later but there's also some you know surprising things that we do we do learn uh and one of them is that the share of remote workers that we estimate is is very different essentially that others that have been estimated in the literature so for example if you go in U.S or even in Germany there's around 25 percent of people actually working remotely and with or 30 to 35 depending on the country right but what we find is is a little bit different and that is because a posting is a promise of the employer to uh essentially hire you for the next couple of years right so if if I'm hiring you uh and if I tell you you're going to work remotely I can take you back right but if I tell you well you can go and you know today you can work him up I can say in in three in three months you can you can come back so there's there's this you know version commitment that we're seeing in the posting data that actually brings to it you know lower estimates right and obviously we find that you know remote work uh varies greatly across occupation Industries and cities and and again unsurprisingly you find that it is correlated with computer use you know education and learnings but I think the some of the interesting Partners do come from when we're seeing some of you know some of the results for the cities and again here is the cities that are a little bit more expensive that we're seeing you know immense growth in remote work and finally you know this uh growth in sort of like in remote work originally starts you know from the local level right and so like either you can think in us or in UK first you know employers are looking for opportunities outside because a they are losing people but also there's a lot of you know cheap Talent essentially to look for so if I'm an employer in UK I'm starting looking in in other places essentially but you know think a bit in terms of more globally and what this is going to show is that you know there are now Arbitrage opportunities in terms of wages that are happening first locally but also in other places so if I'm an employer in UK now maybe I can hire a software developer you know uh somewhere in southern Italy and then in a couple of years or even now as I will show some data Maybe hire somebody uh in Africa and again so from a Global Perspective now you know Kobe the remote work showed us that there is actually a work remote work can be done which was a little bit of a taboo and there are a lot of you know Thailand opportunities out there and this is I'll try to show that it is very different than you know what essentially we've been doing uh for for outsourcing which has you know has been around for for 20 30 years okay so why are we doing this well first one the first motivation was because it's confusing right if you go to the media today and you see essentially what some of the information you get confused some of the media are saying that your employers want people to go back and then you hear what people don't want to go back and then there's discussion between hybrid and there's discussion with the remote so what is really happening right and there's a lot of you know surveys but most of them have a significant lag right either in U.S or in other places it's not easy to understand almost in real time and this is where the data helps because you know we've developed also a sort of a dashboard that everybody can go and access online it's called working from home map and you can see at least for those countries every month we update the data and you can see what happens to remote the remote work and so first goal for us is to measure Trends in remote work in almost real time and sort of you know solve this inefficiency that might exist with survey data okay how do we do that so first on the demand side uh which is the most established for moment we use online job vacancies data from lycast uh where essentially we're going to aggregate for many many sources to be able to make sure that this is capturing is a good proxy for labor demand more than 50 000 sources overall which you know we duplicate the data and then we structure it and for this project we use data for us UK Canada Australia and New Zealand and now we're working with expanding this in Continental Europe there are some language uh challenges there because again this is a machine learning model that has to be run and trained on Text data but you know there are some you know we can discuss more some ways actually to get around it in a sample period you know it's 14 to 23 uh and for each online job vacancy we have both the unstructured text and structure covariates we use the instructor text to understand if this is a remote job or not I want to show you how and then we use these different uh variables to understand how the remote work varies across occupation across employer and location You Know sample sizes uh differ significantly by by region also you know they are correlated with population but in U.S for example they give you an idea we have around 40 million observations per year uh very large sample uh and also to give you some context that you know the employee population is is 150 million so you get approximately you know one job posting one position per three uh people working uh in in by country and there's also a very large You Know sample of employers 3.5 million us and then as you go the sample gets reduced now I you know it would be a missed opportunity to present this in UK and not to talk about one uh but essentially this is a machine learning algorithms what what we call work from home algorithmic measure right so this is in a way to make it simpler one of these large language models uh it's called tilbert it's a cheaper version of a bird and it runs more efficiently but what we do is that we get you know humans to go and check 30 000 postings and you know say is this a remote job or not right so we create this training Set uh by you know having human classification of forty thousand uh postings and then we use that to train a model to be able to understand For the Rest of millions and millions of observation if this is it will be extremely again uh complicated and costly if we're to humanly classify all those you know 200 and something million uh observations by hand uh so the model is you know delivers more almost you know near Human Performance and and actually I think what we're going to show is that there's been we're not the first to try this online job postings but you'll see that there's so many nuances that this model essentially uh gets a lot better than than more more you know string matches or you know rule based what we call models and so these are some examples I'm not going to go through all of them but again in the in the left hand side you have you know this false positives uh which you think it is work from home but if you read it very carefully it is not uh and then on the right hand you have false negatives but some of these could be potentially uh you know alleviated by some more sophisticated rules so for example the first two say you will work from home care facility right so this is not a you know work from home this is working at a home care facility but if you're only looking for like a specific string right you you you you'll not be able to get it the second the same work remotely but it has a no so if you were to use some you know neighborhood rules or or or or negation rules you could potentially alleviate some of this but the other two are more complicated right it says previous experience working in remote locations this is very hard to to get away with a real based same the other one May respond to managerial or home office requests right same idea if you're working from a home office so there's a lot of those examples but again and there's a lot of analysis within the paper if you're interested you can go and see it but the idea is that it's not it's not easy in many cases to be able just from you know specific keywords uh to understand if this is a remote worker or not position and then the idea is that again the dictionary view only looks at very specific words but the the the the llm actually looks at the whole context and actually puts weights on the remote work but it also puts ways on the other okay now to the interesting stuff uh let me show you uh how results look like and what we learn from this so unsurprisingly we did see that you know remote work increased uh you know significantly uh postcovid but I personally did learn something even just by looking at this specific graph first of all what I found surprising is that you know remote work was not zero it pre-covered there was actually some good percentage of of jobs that were being there remotely and in US was around four percent then in UK it was around lingering around three percent so that was kind of an interesting thing because again there are people that have been working on this for a long time and they've been discussing it but the idea is that you know coffee though you know changed everything and he did uh and then actually the other fact that UK is significantly higher than the rest uh it was also an interesting example and then the fact that again would be hearing that you know remote work is decreasing significantly we're going back to the office this also includes the past two three months of 23 the first two months of 23 and we we do see a little bit of a drop but we're not seeing this Dramatical drop that you you hear in in the media it seems like we're again there might be some I I still think we're not in equilibrium but again we're not all the way back one thing to explain about this graph is that this graph in simple ways is actually taking into account the composition of of demand so it is controlling for for occupation distribution because occupations are very different in in in in this country so this is imposing the same distribution of occupation and making sure that this is Apples to Apples and only the variation is coming from the within occupation variation not across occupation similarly when you look at occupations you find some interesting stuff and you find some more interesting stuff like I think the interesting but not as surprising is that you know we're seeing the concentration and the growth happening a lot in tech jobs but there are also other type of occupations that potentially earn maybe or not they're like you know legal or architecture and engineering or you know art design and and entertainment that seem to appear uh and actually have significant growth uh then this will not like your traditional occupations essentially uh that you've seen before so this this graph is is trying to to do too much uh it has a lot of information in a way I think that the main point that it's trying to show here is that if you draw 90 degree line the majority of the occupation are above that and essentially shows that after 19 the majority of them actually uh grew some other interesting thing is that it also tried to compare with another known paper which is the Bengal and human and you can see in their cases some of the occupations essentially that they classified as as non-pillar workable which is like travel agents or advertising cell agent they are actually some of the most teleworkable when when you see see the data I think this for me starts to you know to bring some of personally the most interesting insights is that when you look at cities and regions that are growing the most right essentially these are the most uh kind of expensive cities and there's a lot of things that are playing here uh I mean obviously the one the first one is the composition again of the occupations and demand in a lot of the Cities you have more you know remortable occupations than than in other places as we will see but that's not the only thing that is happening right these are also more expensive so they are you know seen brain drained during period of covet and they're also because they're expensive they can also find more opportunities like again from a personal experience you know when I'm in Boston for me being able to look in other in other regions made it a lot easier not cheaper to be able to find you know uh talent that potentially just if I would focus only specifically in in the in the region and here you can see a little bit more clearly because again this is doing for for us but it's doing all of the all of the regions all of the you know States you can see here that the composition essentially of demand you know is playing oh that's like you know the ones that are growing the most are are ones that essentially are both most expensive but also ones that uh you know have more remotable jobs but then on the other side hand the ones that did not have as much and are not growing as much are ones that you know where the industries are such that these are not so much of remortable jobs in there and similar graph here and finally from the demand side again there's a lot of here that is you know firm specific uh and it's driven you know uh by fermentation and we're still looking into this and so like what is it driving and I think you know in this case you know we're conditioning on the same occupation group than the same Industries and we see that there are significant differences and I think when Kyle first posted this they called it is the Tesla effect or the Ellen effect and you can see this essentially that you can see SpaceX and test lab being some of uh some of the lowest uh you know essentially anymore jobs both in terms of levels and actually they didn't grow at all right again there is also potentially some noise in our data as well but but you you can see that there's there's some variation happening okay now this is what we're learning from from the demand side and again there's a website you can all go and play around the data gets updated every every month we're trying to uh again expand this in in Europe first and then next hopefully we're going to try to expand this globally as well in in Latin America and in other places as well but recently we also tried to look also the supply side what happens to people again this is what companies say they want but we don't always know if this is happening at what level is is this happening right so for this case we used online social profiles from from revenue Labs where we have more than 500 million profiles globally and and more than 120 million uh profiles in the US remember these are you know profiles that exist online so we we have the latest work history in the latest job of an individual but this is a panel document so we can see for at least when they put in there for all of the history of their work and all of their education history but also also the location of where they've been in previous jobs and also where they are currently and we're using this to Define remote which in this case is is a fairly lower Bound in the sense that these are fully remote individuals like we can we are not able to identify for this at least part of the project if you are you know living in Oxford and then you come into the office two three days a week but then the rest you're working from home but essentially if you are let's say in us and you are you know your office is in Boston and you you are living in in New York then you are able to to call that individual uh remote in this case and also if you are living in Latin America and you're working with somebody either in Europe and or in your us we're able to call that you know personal remote again there's a lot of work to get the levels correct depending on the question that you want to ask but mostly the thing that we are focusing here is sort of the trends and what we are seeing and how that might be important for policy implications and so what are we seeing again this is just the first level in terms of understanding when you look at us and different states and where the individual lives you see almost a sort of a doubling even from 2020 and again in 2020 not a lot of this fully remote had happened so I think it's fine to use that as a benchmark because people has not still started you know uh leaving their places but that was sort of The Benchmark and then when you're looking in 23 you actually see that you know people are you know compared at least the growth in terms of people working in different locations for the companies is actually increasing a similar you know picture concern if you look at India or or other places now what we're interested in this case is to understand okay but what does this mean for for employers we saw this from a personal also experience that again this first happened within regions within United States like there was this wage Arbitrage opportunity appearing like you know companies San Francisco realized that well you know I can look for people in cheaper places companies in you know in in London probably did the same companies uh in Munich probably did the same right uh and you know I was meaning today they said yes we were actually working with Vienna in other places and so like this is happening it's not just you know uh something that you know potentially of of interest and but we've seen some specific pairs appearing in the data right so not we're seeing North America with Central America and the Caribbean uh or we see in southern Europe with sub-Saharan Africa showing up in terms of uh in terms of growth with what has happened in the in the past couple of years so that tells us that there is something here is happening in the past couple of years that is related with companies being able to you know to find talent that is potentially cheaper in in some of those places and when we look at the cost differences right and then we're going to correlate the two and this is showing essentially sort of like the discount that a company can take by looking for uh workers somewhere else so if I'm in North America and I look again this is average across all occupations but if I'm in North America and I'm looking in South America I get almost an 80 discount uh in in terms of the workers that potentially will be able to get now the discount is not the only thing right that there's also a lot of other things at play that you know create sort of this premium the first is the time zone right right so the time zone matters a lot because either you're in the same time zone or you have like you know a couple of hours difference that that is fairly doable and so especially in a sort of more asynchronous way that we are working now but if you're somewhere where you have 12 hours difference that that makes it a lot harder right and so that partially where you're going to see some of these pairs is going to be because of a similar time zone and us and and and South America have a very you know very similar time zone the other is the language right uh in in places not only Talent matters but if the talent also has a language and is able to essentially not have those barriers like for example Chile a lot of people there uh speak English very well that makes it also a lot more attractive so time zone you know but then finally sort of the you know the infrastructure and how easy or difficulties also for the company so and I think this is something very important also from a policy perspective for Europe now I don't know what how things might have changed but you know last year Where I Was Here In Summer there was a lot of discussion that you know people were not able to work and leave essentially in different countries so if I'm working for the company in UK it was kind of difficult at least to work for more than some specific time to live you know in in Germany on other places right because there are a lot of tax considerations now there might be some validity to why companies might be doing this or a location I've been doing this but if at the end of the day what will happen is that for people it is easier to work with a company in U.S versus in Europe that might imply a very significant you know brain drain for for Europe in the years to come right because I cannot imagine why somebody would not want to go and live in Greece work somewhere you know with us get best of the Two Worlds whatever the weather and I get and I get the salary right and and if Europe is not you know allow this sort of digital mobility within there's going to have significant they're going to be significant problems in terms of uh you know you know digital brain drain so you know when we correlate these two we find that there is some correlation uh the axis might not be the the most might have to do it in absolute terms but the idea is that you know the more the the more premium the more discount you get essentially which is to be an absolute the more we see the growth happening in sort of this region pairs so you know between North America and Central America where we're seeing you know or southern Europe sub-Saharan Africa we're seeing some of the highest discounts that's where we're seeing uh also a lot of growth so this is a little bit of a you know waving hands and and but the idea is that it is to be you know it's the motivate for future research but the thing is that we personally believe that this time is different right and again we've been doing that sourcing for for 20 30 years so this is not the first time that doing remote work uh you know we reinvented remote work remote work has been there but Outsourcing is focused on very entry-level jobs or very specific tasks of sort of you know the tech jobs like you know it's very concrete you know uh things to people it's like you do a b c and d and not only you do absad but you do a and if a then B then B and C is like very very specific task and what are we seeing at least right now is that this time is different like we've seen an expansion of the remote work that goes beyond this entry level or this entry or this basic sort of uh you know tasks that that may be Outsourcing uh you know is doing right so again this is a little bit of a same data but different you know accent but if you just look at the occupations with the highest reality mode it's the web ones and and the tech ones but if you look at the ones that are are growing the most those are not the tech ones all right so marketing for example or engineering or manufacturing have also seen a lot of growth and again this is not the the Outsourcing sorry this is within us but still we're seeing a lot of this growth happening in areas that you know essentially are not that can take for example here is is is is sixth and the other thing is that when we look at jobs that are remote and only remote and we compare in terms of skills how similar they are you know we find it actually fairly similar there are some differences especially those are like communication skills and Leadership skills that are that are needed for the remote work to be effective but when you look at the technical skills they are they are fairly so here we're doing similarity index it's like just trying to see like that how skills are similar between the two jobs and we find that they're fairly similar so when employers are looking for a remote job versus a remote job they are looking for Fairly the same person these are not you know different individuals that they are looking for at least in the past couple of years right and finally again when we're looking at some type of specialization right we're seeing that there are specific opportunities in different areas of the world where you know for for employers becomes more uh you know uh more prominent to be able and Define so for example you know if you're looking for a graphic designer Bangladesh is uh is uh it's an obvious place that potentially might might be able to to do so essentially again some of these uh you know results are trying to motivate the fact that this time is kind of different and it might have implications for sort of for both employers and uh and and essentially for employees but also most importantly for for the future cities so let me give me a couple of ideas of what I think those might be so first of all there are some conclusions for for the future of work and those are both from the employer side but also from the employee side right so for for the employer again now we've discovered that remote work can is is effective although there's still debates if you know how much are not is that and there's there's a lot saying that we we cannot do it but essentially employers are realizing that this is giving them an advantage you know the idea that you can find the best talent irrespectively of the borders is is very strong and you know we we're thinking in today's terms right we're saying well you know but you're here in JP Morgan or other places or you know Tesla saying that what people have to come back and I I think I have two ideas about that first I don't think all jobs can be done remotely but we're also thinking in sort of short terms now think about five to ten years from now and and ignore some of a lot of the sun calls that exist because a lot of the discussion right now is because essentially you know we have invested a lot of money in places and in offices and those leases you know are going on for 10 years but what happens when new startups and new companies are coming around they want to be the most cost effective will they invest in an office will they look for you know Talent everywhere they can respectively my bet it is that they they will right and so in a way this idea that you know going and finding the people where they are and not being concerned by by a border or by office I think it's too powerful for the market to ignore in the long run I think there's going to be pushback into short-term but I don't think we'll be able to ignore the first time and then how do you compete with that startup if they have the cheaper talent and they are not having the cost of the office right I think you you have to to adjust right so I I think there are significant implications for employers not in the in the immediate children but but definitely in the long run and yeah on the other hand you know this is also a Market Force because you know employees have realized that they they at least some of them depending also on the agent also on the errands that they have to run that you know having the flexibility is is a good thing but if one of the employers is offering this for the other one even if they don't want to they need to take this into consideration because now there's Market competition right and so it becomes a significant force and it becomes an amenity that a lot of employers are offering and you as an employer might or might not have the choice because if you decide to not hear the market then you are going to be losing uh you're going to be losing Talent so the point I'm trying to make is that in the next couple of years I don't think this is you know done and we're going to see a lot of sort of equilibrium effects happening in in the near future that I'm going to be pushed by what the market essentially dictates but the things that excite me even more is what this means for for ages right and I think there's there's two implications here there's a local and there's the sort of the international implication locally or semi-locally you know we've seen a lot of Workforce Development you know policies right happening before and you know regions that wanted to to essentially develop they had this you know original policy where they had to attract a company uh to come there they would they would provide incentives for the company uh lots of them a lot of times and a lot of the times that would actually be fairly ineffective depending you know uh on the investment of the firm or not but now this decouples that now for me to be you know to develop and have a lot of people and to be able to capture this uh it it doesn't mean that I actually need the company I need the people and instead of providing those incentives to the company I could provide those incentives to individuals and you know this is happening somehow either with the Nomad visas or either specific subsidies that uh regions are are doing across the world for for people to come it's still an experiment but as an idea it's powerful but again because it gives a new tool to your post-economic development uh you know agents to be able to work that does not include essentially the you know uh does not involve the company itself but it also Cuts both ways like for Regions that are expensive and humanities see for example San Francisco maybe uh you know it it gives less incentive for people to be there if now you know I don't have to be here because I don't I don't need to work I don't need to be here to work with with the company and I can be in another place and I don't have any amenity for it uh you know it's not a nice place to live then there's no reason for me to stay there anymore so it does create this sort of you know imbalance in terms of like what will happen in the next few years for ages and this is the part where I would like I love to hear your thoughts because it's a bit kind of a bit more far-fetched but but they're also you know important implications for for immigration right and there are important implications for immigration because we could lead to some type of work I will you know call digital immigration but most importantly it gives the incentives to region to invest even more in the education and risking and upskilling of their individuals right now if somebody from you know any developing country has the opportunity to potentially work with a company in develop the world they don't they hey they don't need to live physically but they think about it before right say you're you're to get somebody from a developing country and say you have the opportunity to go to U.S or to London then all you have to do is to study hard and to be the best software developer in the world like do you think they would take it like I'm I'm sure I bet that they would but that was an impossibility right because it depended a lot on on other things you know it depends a lot on getting the Visa and all of those things but now that is again they saw you know zero one is they you know that they disappeared but the barriers became significantly lower that person does now have some opportunity to be able to work potentially with a company in the developer and that also creates incentive both for at the individual level but also for for Regions themselves to be able to educate and train you know big masses of their population and I'm very excited about that in some ways because again it might mean that we might see some type of of convergence or digital conversions potentially across all of these regions that promote work memory and I'm going to stop there and I'm going to open it for discussion [Applause] well hey let's start from one is that from you here Pizza I can't hear you online yes bloody thank you very much super interesting um I was triggered by the first graph that you showed actually where you see that the percentage of of jobs or or vacancies in the U.S that you could do remotely was higher than the ones in the UK and then suddenly after covet UK just shoots higher than the US right yes you have any explanation for this particular pattern because the order sort of seems to be the same except these two countries I do not uh that that is actually something that I I would also like to to understand better and if somebody has any intuition as to why that might happen uh but you know I would like to know as well but I do not my thought could be so I was also thinking about it that a lot of people in the UK suddenly managed to find a place in the US to work I don't know I mean I I think this I mean if I had to do you know I can take white white guesses and and this might be also you know an interesting research question but potentially you know brexit might have played a role again I think a big part of why remote work you know uh survived and Will Survive was because of Labor shortages right then labor shortages do play an important role to like you know the bargaining power that the employee has versus the employer uh and UK has been you know famously known for the past couple of years that had a significant labor shortages U.S as well but I think labor sources here might have been even more severe because again you got two things happening at the same time right I think the one thing was the economy yeah and labor shortages were kind of similar everywhere but that together was sort of brexit in the past couple of years it might have created an additional uh UK effect and so in that case you know my guess would be that potential labor soldiers created sort of this additional uh you know bargaining power potentially for for employees here as well I'm just going to jump in because there's a question online similar to that it might even be exactly the same why does the UK have such a high percentage of remote work when other countries have a more scattered population yeah I mean I don't know about the this kind of population I think that the main difference here is between you know UK and nus uh and and essentially you know what what that may mean uh I mean in terms of scattered yes us is is is is larger but to be honest I don't have a great answer okay just related to that very briefly have you tried to drop London from the UK and surrounding ERS and see what's happened because I would suspect it's just because the UK happens to have one Hub where a lot of people in Professional Services work but that's super expensive and a lot of people live outside it could be yeah and if you drop like New York in the UK Aus it probably doesn't make a difference right yeah I mean I think that there could be definitely sort of a you know if you look at the concentration of jobs between their Hub and overall I think that potentially is bringing a big part of that and so actually that you know the comment about scatterness I I actually although country it really it probably works on the opposite right because you have different hubs but you also have a different a lot of other places that are bringing the average down right in the in in UK you have London and some of the other places but London is a bigger part of the economy than potentially you know San Francisco New York and and the rest are uh and so like what we know and and you know you know New York is definitely big but you know California is the largest but then you have Texas and Florida right essentially which are are also you know some of the top three in in us and a lot of those might have jobs that are not being done remotely and a lot of that is potentially bringing uh bringing the economy down sorry I'm just going to jump in sorry if I should again JW said maybe there has always been an overconstration of job opportunities in London and Southeast and that might be why the UK has yeah um well I guess there's differences between UK was very interesting I mean I remember when you first also kind of talked about it on Twitter is kind of this discussion kind of broke um something about UK is that is uh different from any other country they have a very high share of gig economy and just increasing so something that I wonder is if some of this gig basically worker that are self-employed actually transit to the remote or for other company because they could kind of like equip their flexibility so basically from the self-employed to working remotely for other countries sorry other companies and to kind of deviate a little bit from this discussion I'm generally thinking about your thought about the kind of extent that remote work kind of could um you know like expand the uh boundary that company could kind of hire workers based on the cultural differences so if the cultural differences between two country area even cities could kind of limit the um let's say Outreach of the remote work to towards some regions I mean again I I can I can speculate but I it's it's it's a research question in the sense that I don't I don't know I mean in the sense like I don't know how much culturally similar you know India and us were all this time but the U.S Outsource a lot of very specific jobs in the effort for a very long time probably not only us so I think cultural difference definitely uh you know are some of some barriers but as with everything well as you kind of try to alleviate some of this potentially and so there is a way uh you know I mean for for remote work to work it can't just be it's not just the concept of I'm doing this this online and that's it like there's a lot into this asynchronous uh you know communication and working together that needs to be in place and so like for example you know the reason why I work potentially in India even with before was because there were specific centers there and there were people there locally that would be managing like so the management aspect would not fully happen you know from uh you know from you from us and similarly you know for remote work to work there's got I mean again it will be also a little bit of a an experiment going forward because one of the reasons why it worked is because some of the things were already in place a lot of the teams were in place would be a little would be interesting to see how this works when completely new teams are being created from scratch remotely right when you already have the con you know when people already know the work when the managers are already in place it might be a little bit easier or potentially like to hire people remotely because they can train the people they can communicate with the people they can try to share that you know the company culture potentially that uh but if you were to hire everyone from scratch and creating a remote culture even within a I think that that will definitely be challenge yeah so in developing countries what are the incentives for public authorities and private companies to pay The Upfront costs of Education and Training if an increasing increasing proportion of the people with Education and Training uh then work remotely for the developed countries and don't contribute directly to economic growth and development in the developing countries so that will depend on how they you know the policy that they will have to be able to capture some of this benefits right and and I I think the idea here is very it's not any different there are emittances right if these people would live and they would go to a developed country and they would send money back right there this sort of incentive because you know the the country you know the origin countries is capturing some of these economic benefits I think it's the same thing as long as the country is able to capture in terms of taxes so they might not be able to tax the company but they are able to tax the revenue from the individuals and there is significantly higher economic I think they're they're significant incentives for them actually to keep the people there and be able to bring all of this Revenue in the country I think that's easier said than done because again with some discussion I had last year uh with policy makers from Latin America they are already facing this issue like they already had significant problems of tax evasion and now there's like we have both now brain drained and tax evasion because a lot of people are working with companies in uh you know in us and we're not being able to capture any of this Revenue in terms of taxes so I think there are significantly you know barriers to to pass there but if the country is able to capture that I don't see it as any different aromittances coming to the country but in this cases perhaps people actually don't have to move hi sorry to go back to to set and maybe we are missing brexit effect here because many we were leaving the pandemic joined with a brexit effect and many EU citizens that they are not uh UK Nationals or they they have problems with the residency is a way of as well capturing these workers that they may live otherwise yes no I think brexit definitely playing a role I don't know the institutional aspect of that for example again some of you I might know better but like if even if that was possible to work remotely for for a long time so for example it breaks to happen and people had to go back to their home country could they in theory continue in working with you know from a tax perspective because I know there are some tax issues and tax treaties I think but definitely brexit is sort of the you know the hidden candy here foreign side of things um first of all there's been a long uh like more than a decade where there's been a lot of Outsourcing and platforms where you could find Freelancers doing this work so perhaps the direct effect uh after covet might have been that these people turned into something more formal where they work with those firms but the second is that I mean traditionally we think about um like the governments making decisions about reskilling upskilling people and how they should you know turn their universities into you know places where people get all the digital skills but in in today's world all these things have changed and also you can get those skills through other means so essentially if the demand in the market for you know this type of skills is high and needs to be adapted to specific places uh in the world I think that we might not even need to go through the discussion about whether the local universities or the local authorities have to do something if there is a demand which what we mostly see is that demand from from the labor side of things uh I don't know how that that would fit in in in this discussion no I fully agree I mean but you if I had to guess I think both of it will happen I mean I think yes the fact that so much information is free and you can actually access it online like somebody can go and get MIT classes by then it's open right and into any child from everywhere in the world can go and study at MIT quote-unquote I think that's that's amazing and so definitely it's very different than what was you know 20 years ago so I think yes you are absolutely right that from an individual perspective there's less you know barriers to to you know to learning and a lot of that will happen at an individual level but governments are also doing a lot of this I mean I think they have realized that I mean there's at least personally with almost every country that have been in you know in contact there are discussion about obscillary skill and so you know I think a lot are using for example Singapore as the sort of the you know uh The Benchmark which essentially has some very uh effective and you know active uh reskilling uh programs but there's you know from countries to you know Latin America to Asia everywhere they are talking about this so it's not something that is like you know second third in their mind there are uh significant you know Europe is has this five-year in your skills and is the is the year of skills this year so um so we have um timatayo online has um said that if the jobs aren't advertised as remote working um can they transition into remote working jobs and do does your data collect that information if they're advertising as a remote or non-remodal if they're not advertised as remote working can they become a remote working job and does your data um capture that I mean usually from my experience that is not the possibility I mean if what they might be is that there might be some noise in the data so there are some companies where potentially they might have a sort of a full blank statement like Airbnb and I think most of the time those are actually also transition periods Airbnb for some times was saying that all of their jobs are remote but they were not really advertised like if you go to the job posting they're not advertising but that was more of a transitory period and then eventually they did they did add it to every job posting so I think there's that's a measurement noise but in in aerial terms for a Renault remote job to actually become remote in the job posting I I or overall as a phenomenon I find that very difficult I think what has happened from a personal experience is that especially with workers that potentially might be experienced there are you know individual discussions with managers and if you see that for somebody you know being remote is is very important and your decision is between losing that individual uh or they become remote I think in that case you know that that's an individual decision of the manager and and they do allow an easier a lot easier right now than potentially covet because you know we realize but uh for entry-level jobs you know there are non-remort that become remote I would say that that would be hard hi I'm wondering how the competition across region would look like in the future and if you have some early indication from the data so suppose if um uh farms in the high paying region are trying to hire from low cost region then what would be the response of the Employers in the locals region so is there a tiered responsible or lag response you see in the data so what we have seen both from the data and and again by talking to a lot of individuals that there is some type of a you know wage convergence in in a way uh and that actually has been slower than it should in the sense that for the regions that were fairly you know cheaper uh you know workers realized very fast that I don't have to work with the employer in this region I can actually work with somebody you know you know in in the area that pays a lot better uh and vice versa right so for a lot of places you know there were more expensive they realized they can find talent that that is either cheaper or equally costly but potentially it's a lot easier for them and so like we see this sort of you know potentially wages going yeah upper for the cheaper and going down for for the most expensive so we're going to see some type of a you know way lock you know within States uh you know wage conversion potentially and then put even globally uh with you know the very cheaper ones actually to grow up now I do think that there is also some type of limitations at least from a regional perspective because what happened you know especially when it comes to the discussion uh from you know from from a city perspective is that what we did see parallel is that a lot of people actually move to places that were were cheaper and that they had great amenities so in U.S specific we saw two regions After People moved a lot of them moved from San Francisco to Texas to Austin and a lot from New York moved to Florida or you know from around the area because it was warm there was no reason to move to Florida before because it besides the weather which is great but you know in terms of job opportunities was was very bad right then suddenly now everybody's you know uh from the East Coast moving to Florida and a lot of people from the West Coast are moving uh to Texas but there is there is some type of equilibrium effect there that you know is related especially with housing prices and rents because if everybody's moving out Florida then prices are jumping up uh so they can only reach a limit before people start living right so I think there are these two equilibrium effects that are happening both in the wages and in the housing prices that will you know that will interact with each other in terms of like you know maybe suddenly everybody goes to South you know south southern Italy right because it's great there and you know you can work with you know with London and and Germany and other places but again if some Southern University stars become you know everybody goes there they will become very expensive I was wondering um you showed a very interesting slide that in Egypt is sort of the highest of accountancy uh in the U.S you had a high number of Engineers and uh I think in Pakistan it was graphic designers and I was sort of wondering is that a reflection of the existing skills in the sort of pre-remote working age and also um to what extent is that becoming a self enforcing or reinforcing phenomenon I mean we know about um virtual assistants in the Philippines for example you know going forwards are they going to carry on as those Trends or are they're becoming sort of global um freelance occupations if I had to bet uh I I would bet that this will change significantly because this is based on the current sort of also industry you know infrastructure yeah that and distribution that all of this region has so it's sort of like a pre-remote uh and pretty remote plus some some Outsourcing you know related so if I had to guess a lot of this will change I I think one of the things that I did not discuss is that you know especially uh if you were also to thinking combine a remote work with uh you know with all of this charging video open AI type of stuff it also then like you can see that for example some of these jobs become even more easier to do like why would you have you know some level some some Basic Marketing right in us you know in theory yeah English now is not a barrier so everybody can write essentially great you know and then you have you know mid-journey creating amazing Graphics you could in theory you know get a small you know marketing firm almost everywhere in the world so there's some level of some of these jobs are actually becoming a lot easier both to do and also potentially you know to do it remotely some of these jobs are also been especially the low pain are disappearing uh you know especially if that's sort of customer service jobs that we had that sort for all these years I don't expect to see a lot of them in the next five to ten years especially I think they're going to be replaced by uh by generative AI so to your question I do think this will change significantly and I do think that some of this will actually increase and some of this will disappear um so given that what we've seen about both discipline they demand for uh remote work what is the main argument for on the firm side that makes economic sense apart from the real estate that you know you're still paying for offices and you want people to feel them and you know this is an issue there I mean is there any anything else underpinning the sort of the things we often read in the press that you know people might want to bring on back because you know things change or is there any evidence in the data that uh um you know that you collect that there is a pushback a specific reason for this pushback because we see the economic reasons for you know using remote work and you know not just within a country but across across different places but is it sort of the regulatory is it the tax or is it something else that uh you know possibly we're missing here so there a lot of things that potentially we're missing here I think this is most of this work is descriptive in showing what has happened what potentially may even can happen it's not discussing if what has happened or not happened actually is is Good from a firm level or even an individual level right it's speculating some of the benefits but there's a lot of other I think there's a lot of questions that are open uh that will need future research so for example this remote work there's a lot of discussion that remote work is not as productive of no remote work we we don't know we don't have any real evidence right yet and I think it's too early uh is it for the individual is it uh the same easy to to be promoted if they're working in a remote job versus a non-remote job it's something that we don't know so the benefit from a company level and the benefit or cost from a company level or benefit because besides just the the basic cost these are still open sort of research questions that we we don't have you know a great answer okay have some speculations but we don't we don't really know is the mobility of workers the same are is it able for example for a company to to access a more diverse pool of of individuals if they you know if they do remotely or not we can speculate yes but but there's uh you know there's no no real data that that we have seen today so the discussion of the day is basically focused on cost efficiencies but there's other type of uh you know cost and benefits to the company identity individual which I think there are open research questions they're a tough one we need to wait a little bit you know we need some lag for this to happen and see what is really going on um so for the final question Lily online has asked do you have any insights on the demographics of remote workers I.E are there more men or women working remotely not you would not be very difficult to get but I don't have a clear answer to that specifically now based on the data that I do have the thing that we do know at least from from some data is that remote work has increase the server participation of people that were not able to work as easily before right and so like either people that had disability like we've seen that there's been an increase in participation of people that could not go to a physical space that now they can work remotely or for example specific seminal population that was very difficult for them like you know we had like women's veterans that had to move all the time and they were not able for example to find a very specific job so they have a specific segment of population that were was very hard for them to be able to find a permanent job and in an office and we've seen that there's been an increase in their participation in the labor force thank you thank you very much Glenn thank you everyone for joining us today please keep an eye on the Oxford modern school events and yeah thank you everyone online thanks thank you

Channel: Oxford Martin School

Views: 332

Keywords: oxfordmartin

Id: qFkrRlfWyVk

Length: 58min 10sec (3490 seconds)

Published: Thu May 11 2023