Miss America to see mo coming Shari allora benvenuti at the Scott ethically Barry Eichengreen you know the important historical economy un protagonist Adele debated economic oman yalla yes oh no stefano lepree editorialist economic la stampa di Torino Cobra angry Najib is PA Guerra Kanan appropriate okay el mundo global it Zito Sturdevant and pupae a tradition family bloody McQueen - Enid and Nevada from misogynist Americana Thomas Friedman oh poor a pupae color sista recurrence and una manera particular RL we estudia taco apropos student market to the level you de la prima Vida videre k i hanged in in general put up solution Italian only Brazil national economy rápida study economic developer parentage incurred it was really important in on Lozano ambitious effort America son Seto italiano deliberate robot encuesta interests and profit for suprema illiberal de soie Madre Lucille una Soga Vasudha D Auschwitz and a liberal Toto County could do a liberal important excreta Barry Eichengreen so no an ephah golden feathers will gold standard alacrity monetary Italianate rained impure giant a villa committee shoot'em Altos de Gama Hall of Mirrors down a story a comparator della kris EK abbiamo the suitor a sentiment a key for selenium Cora finita a telegram decrees italiana trained marine animal to present-- in the budget economical current so La Traviata soup projects indicate the sitter internet they principally commendatori mondiale e li chen kun evolution italiano saline glycerol team article Oh sick Debbie Toshi name is akia Frontera LaMontagne Debbie Tichina Jim peyote settimana fact masturbation English hello Cepeda probable KCMO in Italia in Europe we considered alleged La Traviata sooo internet own article oak escrito soo bahk sale is seated at the Economist Eola pay I propose autocracy delicate insieme own not economist a franchisor shall v+ quest Osito una irani can see me informative Zima contra storia della chris d'elia connoisseur be puta Lavery Maputo Therese all the neuro of molto meglio D Camela result that Kemal Tomino Don me see fra lot Rocco did result in una parola ill whatever it takes the draggy Cavallaro invasion in questa contra storia lupron chatter jean-claude Trichet inanimate supreme la casa su and that malt-o-meal Barrera at the capacity the she's a prima Supra to talk reading people castle italia serie berries permeate a Maltese of forensic mean to do G a major commercial evolute Internacional infer on carefully making zhenya demented against any Berkeley in California for seviche Tara ankle it's a novel Ebro the Michael Lewis okay mr. Harris Errol belphin Legrand escameca mint Otto professor on a dog a queen Salah track to the quill film attracted only Brody Michael Lewis on Bravo's honest economic American okay Pony escrito not focusing on flash boys a cappella Punto the quest the calving fibra tikka Kevin go Stacey parent is apparel Massimo uncle millisecond delay operas on nanosecond in me the opportunity balls oh gee professor I can be speaker echo-y opened the German Chateau de Giovanni for a journalist economic offensive on okay el mercado de la volute Internacional era dominant oh that in your misery youngmee perkins beats enemy bang kiri did you rig obviamente una cosa que buddy I can envision it Zurich on encanta nope you tanto peril Cooney phenomena phenomena Kelly estudiar but we wish a Singaporean thermal to be important in su d go she's an ultra strong a modification in a commercial turn that's another level ooh okay veramente valley latina deke appear a proper cake imagine a sorbet una cosa gun the quantities all say una cosa cozy importante namer catty do g trial tomandamanda commission on postal inspectors who exposes us a laundry should mr. Callen drew notation trip importante que cosa so Charis Alondra no syrup unilag immunity up a US equities our aquatic influenza come on quite a skull camara Casa de Barry Eichengreen Stefano thank you for the kind introduction I think you covered most of the important points of the paper already we can move directly to the Q&A but you neglected to mention one thing which is for how long should I talk how long does this session go 45 minutes okay very good this paper is a little bit different from the research that I've done in the past and I will come back to why momentarily it is a collaboration with a couple of economists at the European Central Bank it has another title which may have been part of what lured you all in here this afternoon cables sharks and servers depicted respectively left center and right and I'll describe their relevance to the to the geography of the foreign exchange market momentarily but I chose this paper for this afternoon because I think it does relate to the theme of the festival this year which has to do with the geography or or the location of economic growth economic activity and in in in in my case service sector activity in particular the point of departure is the familiar observation that we are living through a technological revolution the cost of everything related to information and communications technology is falling like a stone and the implications of that for our lives for the economy for the location of economic activity our question this long weekend is quite unclear and that pertains in particular to them what is now the most important part of our economy the service sector which accounts for depending on the advanced economy one is concerned with 50 60 70 80 % of employment so I don't know what that number would be for Italy but I conjecture it would be something on the order of 70% of employment in in Italy is on the service sector so whether information and communications technology will favour service sector employment in in in Italy or dis dis favorite.i is potentially a first order question as with all interesting questions economic there are different points of view different hypotheses different conjectures so one view is that cheap information technology will significantly reduce the importance or or or the effect of distance on the location of economic activity other other factors and and and and not the costs of communicating across distance will determine where I act activity occurs so there some kinds of economic activities like what we in California do in Silicon Valley where agglomeration economies are very important where activity will concentrate in a few major centers like Silicon Valley and then because a distance is very unimportant for distribution whatever is being the knowledge and other things that are being produced in those few centers will be distributed worldwide then there are other activities where agglomeration effects are unimportant I was one pondering this afternoon what and what an illustration of that might be call centers I suppose would be an example where it's not important for for everyone who answers the telephone when you call your airline to be located in the same place instead cheap and efficient information technology may allow those activities which were previously concentrated in a few places to be dispersed across space and undertaken where the cost of the relevant inputs is leased so the airline that I fly has recently disbanded its call sectors and given its employees computers that they can work with sitting sitting in their bathrobes at home so whenever I call my airline to rebook my my ticket I get get someone in a different US city depending on the relevant time zone and the and the relative cost of labor and and and and and so forth a change in the in the location of economic activity that has been very much driven by advances in information technology so either way you cut it in in both of these cases which are rather different in their implications the first one implies that a lot of the high-value economic activity will be concentrated in a few places the other one and implies that activity will be dispersed widely across space distance is no longer the the cost of distribution of the service of knowledge or or or whatever is no longer the binding constraint there is at the same time another view which is that all this talk about a flat world Thomas Friedman's phrase is exaggerated and that distance still matters importantly for economic activity there are a variety of activities a variety of service sectors in in which local activity and jobs are still protected or still shielded by some combination of political cultural linguistic and regulatory frictions that even keep very cheap information and communications technology cannot erase that local knowledge still matters for local production physical proximity is still important for coordinating different stages of the production process for coordinating related activities and that limits the scope for outsourcing activities too far distant locations where the cost of the relevant inputs is less if you look at the empirical literature in International Economics we don't have a lot of empirical regularities that are robust and that really stand up to different empirical techniques and different datasets but one that does really stand up is the importance of distance for trade two economies that are close to one another create a lot more bilaterally than do economies that are far distant and distance still appears to matter empirically for the location of production of a variety of different kinds of goods Helen ray and rich report us to economists in London argue that it similarly still matters importantly for the location of financial service production financial intermediation so what my co-authors and I do in this research project is use the foreign exchange could as a case study with which to test these competing use why is this an appealing case number one it's one of the largest markets period one of the largest financial markets service sectors in the world it is purely image immaterial it's a service sector in which services for annexing buyers and sellers of foreign exchange and not the production of physical merchandise takes place it's been fundamentally transformed since the late 1980s by information and communications technology as I will describe to you and usefully for my purposes because I'm an economist and I've been trained to worry about identification and exhaustion 80 it can be argued I will argue that the about the transformation in in information communications technology affecting this market occurred for reasons unrelated to financial services for reflecting very different considerations imperatives that were essentially exogenous to the foreign exchange markets so here is mr. Friedman's the New York Times columnist view we call the first hypothesis that location distance and other aspects of geography no longer matter or they no longer matter as much in a global foreign exchange market where currencies can be created 24 hours a day seven days a week as they did before after Friedman we call this the flat world hypothesis the argument being that in a world of high speed communication sales and purchases of foreign exchange can take place anywhere and and we should expect to observe those transactions in a growing number of different locations reflecting the availability of local knowledge about what the Banco de Italia is is doing foreign exchange will be created in Italy people who worry about what the Fed is doing will want to trade me in in in New York so forth and so on and because of cheap information and communications technology you can find if you're a buyer you can find a matching seller easily enough wherever you're you're located courtesy of cheap communications the other view the movie that some people saw yesterday was the Big Short which was Michael Lewis's previous book subsequent to that he he wrote out our neighbor Michael Lewis who lives up the street from us in in in Berkeley wrote a book about information and communications technology as well in which he argued that colocation being close to where the electronic platforms on which born exchange is traded was critically important if you could respond to a buy order more quickly than your competitor you could get a slightly better price and if you were hooked up to a high speed cable and co-located only a few feet away ideally in the same building as the matching servers of Thomson Reuters in Trenton New Jersey Trenton is named after yep so if you know if you were closely nearby where the electronic platforms were were located you had an advantage and there was therefore an incentive for transactions to concentrate in a few places in Greater New York in London and in Tokyo where Thomson Reuters and electronic brokerage services the two big electronic craters locate there their computers you want it to be there or you want it to have a direct cable link to them very different implication of what the foreign exchange market would look like in the 21st century this this view would say it will tend to be flashy rather than flat it will tend to be concentrated in in a very few specific locations so that's the question we ask in this paper has cheap information and communications technology led to a greater concentration of sales and purchases of foreign exchange in that small handful of financial centers or has it enhanced the competitiveness of suppliers of these services in a bunch of other places you can buy and sell foreign exchange more conveniently in Milan using local knowledge about the Italian market but you can communicate easily with London because Italy is now connected by high-speed IT to the relevant financial center which in your case until after breakfast would be London so that's those are that are that's the question we ask here as the market grown flatter or flat year these kind of questions are not straightforward for economists to answer because in the real world everything depends on everything else everything is endogenous and and and we have an an identification problem in this context as well as others the state of technology can affect where economic activity takes place that's the causal connection that I am interested in here but conversely but at the same time changes in where economic activity takes place can also prompt changes in investments in technology the technology will depend on the geography of economic activity as well as the other way around identification is rendered even more difficult because in the real world most things move slowly together at the same time and that makes it even harder to figure out what is causing what instances where there are sharp changes in technology for reasons unrelated to the relevant activity that affects different market participants at different points in time I've just described to you kind of the ideal natural experiment where it would be possible to identify the effect of technology though that kind of natural experiment is not very common so the argument here is that in the foreign exchange market indeed we have just such a source of exogenous change in technology and indeed spatial variability different national markets were affected at very different points in time for reasons unrelated to financial services and those exogenous changes were the laying of subsurface submarine meaning you know on the ocean floor fiber optic cables starting in the late 1980s cables not related laid for purposes related to electronic foreign exchange trading rather they were laid by telephone companies they were laid by telecom companies for purposes of capturing telephone calls that students studying abroad would make make to their parents back at home or or for a variety of other purposes eventually over time with with a conservative sitter will lack the utility of those cables for electronic trading a foreign exchange was discovered but and and but only slowly over time and only slowly over time was that global network of submarine fiber-optic cables what we now think of as the Internet backbone was was that built out globally so here is the Internet backbone these are the cables that I will analyze in the paper starting with the cables that go across the North Atlantic which are highly visible right in this bigger that comes from a source called telegraphy from which our data on when different cable links went into service one of the things I will point out momentarily is that this cable network which dates from a couple of years ago looks virtually identical to what the cable telegraphic network would have looked like a hundred and twenty years ago and I will argue that that's no coincidence either which is a further fact that is helpful for identifying what is exogenous of what is endogenous so I fear that this figure is not terribly visible for the people way up at the top of the room but it starts in the late 1980s on horizontal axis and extends basically up to today and it has a variety of different countries indicated along the vertical axis and the little circle is the first year in which the country in question was connected to London or New York or Tokyo and you can see that there is a lot of variability over time you know France gets hooked up to London at the beginning other countries Iceland for reasons relevant my story it's a very hard place to land a cable because of the shoreline geography gets hooked up to London in New York only in 2004 here's here the Norman gnomes of Zurich now the gnomes of Zurich used to be as important in the foreign exchange market almost as the gnomes of Singapore but what would the analogous phrase be back in 1995 they were but come 2013 Singapore is more than twice as important in terms of foreign exchange market turnover as Zurich an incredibly important fornix foreign exchange trading center historically why this district distant difference well Singapore has it's an island basically it has sandy shorelines if you've been scuba diving there you know that shoreline is kind of rise gradually to the shore it's a very easy place to land to hook up the submarine cable to the land-based telecom network and Singapore maybe if we go back a couple of slides you can you can see a cluster of points down there that's Singapore with the big cluster because it's a logical hub for submarine fiber optic cables just now Zurich is getting a cable they had to dig it incredible expense a trench basically from Zurich to Frankfurt which hooks them up to the global network but that was expensive and that took time and here you see the time series for Zurich and Singapore they moved together for a long time then comes the cable network to Singapore then comes the growth of electronic trading and all of a sudden Singapore is twice as important as Surak as a foreign exchange trading center so as I've said our empirical strategy relies on the cable network and it relies on the fact that people who want to trade for an exchange at the best possible price also want to do that in London or Tokyo or in New York where the matching servers the electronic platforms that consecrate a bunch of transactions buying sell transactions are located so that's what we basically look at do you have a cable connection to London Tokyo or New York if you have one for how long have you had it with what impact on the geography so you might think I think a lot of people's intuition my intuition immediately went this oughta lead all the business to migrate New York and Tokyo that turns out not exactly to be the case there are two offsetting effects both operating in this domain so the first one is the direct effect of the cable if you have a if there's a cable connection running from Milan to London the direct effect of that is actually to encourage trading in Milan because traitor because foreign exchange trade providers the bank that you would go to if you wanted to do a foreign exchange grade in Milan or some other Italian city can use its local information about what's going to happen to Italian interest rates or whatever and then it can book the trade for you at a very low price in London using its cable connection but the way people think about where the money is made by the service providers it's made by the desk at which the sale takes place where you go in to the bank in Milan and say I want to sell euros and buy dollars the direct effect of a cable is to increase the share of currency trades occurring onshore in my example in Milan but there's also an indirect effect which reflects the fact that there are other frictions that tend to segment markets if you are far distant from London New York or Tokyo if you're in a bad time zone like New Zealand or India most of you a lot of your foreign exchange trading occurs onshore because if you wanted to trade in Tokyo or London they would be sleeping when you were awake and vice versa distance is a cost it's inconvenient from the point of view of this market if your local financial market is very liquid it'll be cheap and convenient to trade locally rather than offshore conversely if the local market is narrow and illiquid you would rather create in London and if their capital controls that might keep you on shore rather than then going off the indirect effect of a cable connection is to erode the effect of those frictions so again the direct effect makes it cheaper to create locally the indirect effect means that anything that was keeping business onshore previously like the inconvenience of distance and being on the road in the wrong time zone matters less when you have a cable connection yeah it turns out in the end that the second indirect effect dominates and on balance we do see that the world grows flashier and that there is more concentration of foreign exchange grading in the nature centers although that effect is not universally seen so I told you a little bit I guess I can be relatively brief here I've told you what we need to know about electronic broking before these electronic platforms if you wanted to trade foreign exchange you would ask your bank to place a crate for urine you and your bank specialist would get on the telephone and call a larger specialist who would be aggregating these calls over something called a swap box so they would have a speakerphone on on their desk and a bunch of people would say I want to buy euros at 103 you know 103.5 whatever and and and eventually a transaction would take place with electronic platforms with computer systems all of that became much more efficient but this migration to electronic transactions took time in particular it was necessary to link up other banks in other financial centers to those electronic platforms and to build out submarine fiber-optic cable networks so let me tell you a little more about these cables they go back more than 150 years so if you remember Morse code you remember Samuel Morse who was the pioneer in laying a submarine cable he established that you could lay a copper cable underwater in New York Harbor in 1842 and if you insulated it adequately from the water he used tarde M and that had been dipped in tar he could transmit Morse code signals telegraphic Lee and then they they replaced our camp with gutta-percha the gum from Malaysian got a person agrees virtually all of which were working cut down in order to obtain the gum to lay the global telegraph network over the subsequent half-century so we were in London free in the UK for a year and I had the opportunity to visit the City Museum of London where you can see a piece of the original cable that this one that ran from Great Britain to the European continent in the 1850s the first transatlantic cable was completed in 1858 it didn't work never worked very well it took six and a half hours to transmit one paragraph message from Queen Victoria to President Buchanan the cable failed completely after six weeks because of the currents that were dislodging it from its location on the ocean floor the creek was to lay the the cable deep enough that it wouldn't be interfered with by icebergs but shallow enough that equipment of time could actually carry the cable to the mid-atlantic they succeeded for the first time in 1858 they celebrated in New York with a fireworks display that burned down City Hall in New York finally in 1866 here's an illustration of the technologies they needed to carry this thick vast amounts of thick cable out to the Mid Atlantic copper cables surrounded by rubber or gutta-percha surrounded by an hour outer layer of iron the problem being that these cables were not very efficient not very reliable the signal would dissipate with distance so it had to be transmitted at a very high voltage in the absence of a amplifier along the way that created distortion that limited carrying capacity that heightened the risk of short circuiting technology marches on so you got repeater amplifiers devices that could be installed along the cable you got advances in knowledge about long distance telephonic communication and transmitting at higher frequencies but it took until after World War two for that new technology to actually be implemented because of the disruptions of World Wars 1 & 2 and the Great Depression when not very much investment was done stage who has been a coaxial cable so if you're old enough to have a traditional stereo system you will recall coaxial cables solid copper now insulated by polyurethane rather than got a person using vacuum tubes to amplify the signal the first one is right here laid between Scotland and Newfoundland in 1955 really a technological revolution that at the time so this cable was made famous because you could not only transmit Morse but you could hear Paul Robeson performing a concert to an audience sitting in New York because the US authorities have deceived his passport Robeson having famously demonstrated you use his prominence to demonstrate for civil rights but he could still with the cable perform to an audience sitting in st. Pancras Town Hall in London so let's see if this works Kable isn't working I'm looking for a YouTube video that is supposed to all right so I will I'm gonna come back to this once we have the technology after coaxial cables of course come finally our fiber-optic cables where fiber-optic means that data can be transmitted at the speed of light in glass which this means very very fast and the cable network then the first submarine fiber optic cable is completed and enters into service at the end of 1988 so here again is my critical fact that it was an investment project of American Telegraph and telephone company French telecom and British Telecom which they invested in in in order to be able to compete with other telephone companies to transmit voice traffic the problem being that that this cable not unlike its earlier predecessors had reliability problems the problem with fiber optics transmitting all that electric signal through glass is of course sharks that sharks are as you all know are electro sensitive they have little sensors in their mouths that sense electricity which tells them food is nearby and they go into feeding frenzies and they attack the cables so in order to make these cables reliable you have to achieve them in Kevlar you know very very move it from from the battlefield to to the ocean floor so there's some another video here which come back to as well as soon as the cabling is is complete shark bites fiber-optic cables under sea so it won't be a surprise what this one is when we when we get to it so if I come ten years ago ninety nine percent of international communications traffic was carried by these submarine cables everybody in this room has a cell phone and you're used to talking to one another by satellite the critical point being that that's a relatively recent development for five and more years from the mid for much of the first decade of the 20th century cables were much more important than satellite for communicating even voice traffic now voice has moved to satellite and data is what we communicate by cable so for those of us who follow the continuing development of the global submarine cable network cables are now being invested in and built by Google and such companies Google just completed one that runs from Florida to Brazil and another one that runs from California just a little bit north of where I live to Tokyo and from Tokyo to the rest of Asia this is the relevant one now right all right so we'll make that the conclusion of the of the lecture and finally in the last year there have been cables laid through purposes expressly related foreign exchange trading so this one Hibernia express and in the dark blue line you can see it is a more direct shot between London and New York traders can cut you know five milliseconds off the time involved in a trade but importantly in the purpose in the period that I'm analyzing there were no cables lated laid for foreign exchange trading okay so if I go back was it was worth waiting for all right so we now we get the point let me um you see if I can all right I will reward your patience one more and then we'll go on so for this we need this piece of metal at the bottom of the sea it means the world all ropes and once traveled the globe an international superstar performing to sellout audiences across America and the way film theater song or speech ropes and make people want to listen no they can't the US State Department have revoked his passport because voice I think this is above my paygrade clearly so so again you get the point there is a youtube video that can be consulted later and professors say we will leave that part for a homework assignment alright so there there is actually empirical work in this paper so let me describe that briefly what we do is we combine data on on submarine fiber-optic cable network with data on the location of foreign exchange trading which is gathered every three years by the Bank for International Settlements so they gather data for 55 different currencies and for each currency they ask people who create that currency do you create that currency locally or do you create it offshore and and we have those data for seven years every three years starting in 1995 so you can look and see how the share of foreign exchange trades that is occurring offshore in other words in a location other than the concrete issuing that currency has been changing over time and the answer is it's been going up reassuringly for me it's above 50% so there there are two currencies involved in every foreign exchange transaction that by definition one of them is being created outside the country of issuance in each case so the here starts out and up at about what 73% and has gone up now to closer to 76 77 78 percent maybe information technology is making the market flash ear and here you see the share of foreign exchange crates that occur in Tokyo in London at the bottom in New York in the middle and in Tokyo at the top so brexit really is an issue not only for London but for the world that that's another in that maybe the market is growing last year here you can see that there is a lot of heterogeneity lot of big differences among currencies in in terms of whether they are created onshore or offshore so the Korean Won is mainly created in Seoul polish zloty and and the euro are mainly traded in Poland and in euro zone and a lot of other currencies are arrayed in between so an economist would be interested in understanding why we see that heterogeneity most currencies in this 55 currency sample are created heavily off shore our hypothesis then as I said before is that a cable connection reduces the impact of other financial and policy frictions that may have segmented the global foreign exchange market in in the past it may also enhance the attractions of transacting in a location that is separate from one of those centers but now possesses the high-speed cable connection to it and which one of the countervailing effects dominates is not clear a priority a priority so we might think that our customer in Milan would now contact a sales desk in London and ask the sales desk in London to book her foreign exchange grade but communicating between Milan and London and small located them communicating between Milan and Rome different time zones different languages all that so maybe the services provided by the sales desk in Milan could become more competitive rather than less as a result of Milan having a high-speed connection to clear the transaction that it's sold to a local customer formerly we model the share of each of the 55 currencies created offshore as a function of three frictions distance where the relevant measure of distance here it turns out is timezone difference distance the example I described to you before we also look at physical distance and and some other alternatives but it's really time zones that seem to matter for foreign exchange traders whether the local financial market is liquid or illiquid whether there are capital controls limiting the ability of residents to transact offshore or not whether there is cable connection either directly or indirectly from concreate Korea to either Tokyo London or or New York and we look both at direct cable connections and indirect ones up through eight other countries up to the ninth order finally we interact the whether there's a cable connection or not with these financial frictions so very quickly let me say a couple of other things about exhaustion ad assumption that the cable is exogenous in this context as I've already stated in a heavy-handed way the geography of the cable network is heavily influenced by sea bed topography it's also heavily influenced by earlier Telegraph and coaxial networks they figured out in the 19th and early 20th century where the currents were favourable to laying a and remaining in operation for an extended period and that's where they put the fiber-optic cables a century later they figured out where the the convenient landing points were back in the days of coaxial and copper cables that's where they put the landing points in the main for fiber optic cables so on the left is the British Eastern Telegraph system on the right is the modern submarine fiber-optic cable system they of course look identical almost and there are some illustrations here that I will skip in the topographical considerations Iceland was connected to the UK finally only in 2004 because they're not very many convenient landing points for a cable we're interested in understanding the share of one exchange crates in a currency that occur offshore they depend on information asymmetries which means how far away timezone wise are you from one of these centers they depend on whether you have a liquid domestic financial market and they depend on whether you South Korea capital controls or not we include whether you have a cable connection as described before and we interact look at the shorter of the two arrows here whether you have a cable connection or not with three financial frictions and we include as economists our crane to do a variety of other control variables like how open to trade how and financially you are what your exchange rate regime is what your interest differential visa etc etc so what are the key findings the key findings are well if you have a cable the share of your currency that is traded offshore goes down the coefficient on the cable variable is negative that's the direct effect that I described to you before that having a cable connection actually enhances the competitiveness of the local sales desk in Milan easily London New York and Tokyo because the local sales desk in Milan speaks Italian it has local knowledge valuable to the Italian customer it can provide the service get the Commission and then clear the transaction using the cable and the electronic platform in London here are the interactive effects and they're all positive they're all opposite in sign to the direct effects up above in the same column so where it says time zone difference you've got a negative coefficient if you're in a bad time zone you create mainly locally a smaller share offshore but interacted with the cable variable offsets that friction and the interaction terms for the domestic market liquidity and capital controls are similarly off-setting so maybe here I will skip relatively quickly and and conclude what I'm showing you here is how having a cable of influences the impact of timezone difference on whether a currency is traded onshore or offshore so if you're in the same time zone as New York if you were Canada if you're Corrado your currency is traded mainly in New York it's convenient transact in the big center because the timezone is the same you you're awake at the same time as those guys in New York the dotted line toward the very left of the figure is timezone difference equals zero currency has mainly created very high share of the currency is traded offshore and then that line declines over time as you move one two three four time zones away from one of those big centers more of the trading is cooped up at home the solid line is the effect when you have a cable when you have a cable you create more onshore because of that convenience factor I described to you before but the effect of distances then attenuated the slope of this relationship is much less in the case of the solid line and in the case of dotted one and you can tell similar stories for and and which effect dominates on on balance if you're in the same time zone as one of the one of the main centers if you're Canada the main thing that having a cable connection does is to bring some of that business back onshore but if your New Zealand and your far distant from Tokyo or London or New York if you're on the right-hand side of this figure the main thing that having a cable connection does is to drive business officer one can tell the same kind of story for a financial market liquidity I will spare you one can tell the same story for the effect capital controls finally this is the last point you can conduct kind of a counterfactual exercise and compare the share of foreign exchange grades that's that this model predicts will occur offshore if you have a cable versus if you don't have a cable shares offshore are generally higher if you have a cable the indirect effects dominate the direct ones and we see this most dramatically to the left of this figure in cases like the New Zealand dollar and the in Ruby where countries that are in very bad time zones where badness is measured in terms of proximity to London New York etc the Canadian dollar and the Korean Won are the contrast in cases where they are in the same time zone as New York or they are in the same time zone as Tokyo so the main effect is to advantage the local sales desk the share offshore therefore it declines with a cable connection those are the cases where wait to the right the figure so that's what we find we find that cable connections make it more convenient to create currencies locally and they enhance the competitiveness of the local sales desk other things equal they lead to a flatter financial world if you will but they also attenuate other frictions like timezone distance local market liquidity the restrictive capital controls and regulation that would otherwise prevent transactions from moving offshore to the major financial centers and our estimates suggest that on balance the second effect dominates the burst and that this sector in particular is becoming fleshy or more concentrated rather than flatter as a result of the impact of to see Abby's Tacoma comedy endogenous scientifica posea supportive our conclusion is our intent a peregrine part the country intuitive like on Padilla mosa qualcuno the further demand sir thank you very much for your talk I was wondering if you can say something regulation of those of that small number of large foreign exchange markets so I think that in fact is part of the debate over the implications of brexit for the city of London and geography of financial services that the European Central Bank may be reluctant to provide euro denominated credit in large amounts to big financial players in a country outside the European Union so one implication is that that could make euro denominated transactions in London more more of a potential source of financial instability of crisis and it may create an incentive for regulators EU regulators to try to move that market back into the EU but the more general answer to your your question is that we you know our fate will be tied even more strongly financially to the wisdom of regulation in that small handful of major hello you talk about baton privilege of the dollar being the very first reserve currency right now in Europe we are talking a lot about breaking up leaving and so on don't you think that despite the free trade despite we used globally that's widely used internationally the United States can borrow internationally more cheaply as a result of the fact that every central banks around the world want our dollars and we could otherwise more importantly in my view the United States also enjoys an automatic insurance policy that whenever a bad thing happens financially the dollar actually strengthened and sent our financial markets are stabilized so in September of 2008 a bad thing happened financially Lehman Brothers failed everyone used as a benchmark and central banks held as reserves around the world if the euro was comparably important international and reserve currency the eurozone would derive benefits to that the other aspect of the answer is that I think the world would benefit from the euro being equally important international and reserve currency to the dollar that there is a tension between having a multipolar global economy and a dollar dominated monetary and financial system the United States is not going to be able to provide safe and liquid assets to the entire world all by itself indefinitely that there's a logic to economic convergence this is our growth economic growth and geography theme again emerging markets will eventually get their act together they there's reason that they should grow more quickly than the United States the US will account for a shrinking share of the global economy over over time we simply won't have the capacity to provide safe and liquid assets to the entire world all by ourselves indefinitely and that's why other big economies where there are really only two candidates the eurozone and China will have two their currencies will have to become a consequential globally as well from that point of view yes I think people underestimate the advantages of the euro yes thank you thank you for an excellent lecture I would like to take a different view on the question even though I like this question a lot also my my question was based on you showed us the history of the infrastructure or the tables in and I was thinking if you could comment more on the investment I mean these were all more or less made these crossed Atlantic cables before the telecom companies were privatized if I remember in years do you see any or did you look at the differences between when you make these investments now when fiber optics in more private setting compared to when the original cable investment were made more from state right so a number of the quite a number of the early fiber optic cables were co-financed by state-owned or and or operated telecom companies again I think even in those cases the motivation which is what's important for my argument was to provide a vehicle for telephonic communication for the public more broadly and and not to enable financial markets not not to advantage local financial markets or something like that but I agree with you that it it would it it's a good idea for us to look a little bit more closely at that than we have that solution a Sumatra demand of Osama ring Graziella Barry Eichengreen eradicate Vanga Trento and cannot observe all have you ever kissed a second