FIT IN Initiative Conference | Toulouse, May 2 & 3, 2024

[Music] welcome everyone I am so I'm Milo bian I'm a professor here at TC and I'm the director of the ptin initiative so as you as you know the initiative was launched about three years ago a bit more than three years ago and the one of the main objective of the initiative was to uh research to to promote research on digital payments in developing countries and the issue of competition and and and regulation in these markets and uh the idea was to have a mix of papers uh which were both Theory and empirical which were bringing Insight from AO from Finance from development and uh especially to address policy relevant questions so today I'm very happy there's lot of uh this work which is represented in the in in the conference I I really hope you will enjoy uh listening to this work and and learning about this work as much as we did in in cutting out the initiative and and interacting with all these nice colleagues which were involved in the in the initiative uh so let me thank the the other two organizers uh of the conference Emanuel orol will chair a wonderful panel uh later today uh from 5 to 6:30 and uh who has been the scientific director of the of the initiative and Matthew bouv uh who been the coordinator of the of the request for proposal and he will also present uh his his own work uh later today so let me thank all the all the speakers the panelist the the discussion all the all the participant who have traveled uh quite far sometimes to come to come here uh let me thank uh set guards especially and and the Gat foundation for for making this all this all this uh all this [Music] [Music] possible time for the nice panel thanks for sharing this so well we are missing one panelist ah the panelist that was missing yes it's very sportive and everything so that's coming right from Washington so we are very fortunate tonight to have this panel uh I must say that was my dream panel and well you are it's always difficult to have this um very good people that are very busy but here they are so it's really a privilege to to to listen to them um so the the first speaker will be uh Set uh G that is from the bate foundation and that will give us some perspective uh on the uh initiative and the other initiative uh is supervising and uh he set all the PhD from berley so he's a a colleague really and understand the economics very very well um and so it was a really a pleasure to work with him um and then uh I will give uh the floor to Toma who is coming from the oecd and uh um that will give us some um perspective some first of all General descriptive statistics on financial inclusion and then we'll talk about consumer and uh in particular education of consumer so this is directly related to the last paper we had consumer might be U bias in B in term of behavorial and so we will have some information about that and also about some protection of consumer so education can help and that's something we will uh we will have so that's uh the the first um order for the speaker then we will have Xavier uh Jin or Jin gav because there is no accent soav that will present a study that it also did than to set uh with the b m foundation on transaction cost so this is also about demand and what might prevent people and this is also related to the last paper that was about all these transaction cost and then we will move towards more uh regulatory issues uh we are very fortunate to have Mark buau with us so I will talk really about uh regulation Mark worked on this issue and uh is really one of the best specialist I believe in France on on that so we are very happy to to to have you there and last last but not least when I asked about this panel to one of my very good friend who is um the vi governor of B France who I should in Target for regulation uh in Mobile payment she said Brazil they are really the best owners on that they are really at the Forefront and guess what I manag to convince thanks to renat Gomez that is a very good colleague of us and that is working at the Central Brazil I managed to convince um so oh sorry uh Carlos yes from joining us and they will make a beautiful presentation so that's really and it will be very stret of op because this is a success story and we want to make that so please that uh the floor is yours thank you so much um and thank you for motivating this panel it's super exciting to be with these uh other um contributors um so as Emma mentioned from the Bill millind Gates Foundation and from the inclusive Financial systems team um and instead of diving into particular research findings on Pol relevant to policy regulation which I think many folks on this panel will do I wanted to kind of motivate a little bit why at the foundation we even are thinking about this and how from a research perspective we think this initiative in particular is is a valuable contribution in the context of other research that's going on um and so you know in particular if you think of uh a foundation supporting kind of Innovations and financial systems um it's to to reduce poverty it's not kind of the first sector that you go to when you think about trying to impact poverty right often donors bilateral donors or or private foundations are going to intervene where there are much clearer market failures or roles for public interventions whether it's health education social protection right and so um in so much as now we're intervening in a place where markets are incredibly active um we have to think a little bit about where where we can intervene and what regulators and policy makers need and my colleague Ari Adney and I AR adne's here um AR Adney focuses on regulation and I focus on research for our teams we were kind of joking that when she walked in she was she's not often in the academic context she's often working with Regulators in a much more kind of regulator Centric context and how in those context Regulators tend to present but not to have too much substance that they're presenting that's very in some sense subject to the politics of what it means to present when you're representing a government versus what academics present which is quite the opposite presenting theoretical models sometimes or empirical models related to theory that may be completely divorced um hopefully not completely but somewhat divorced from from the realities of a given situation right and so um in some respects that's part of why this initiative is really exciting so the to to reveal my kind of secret motivation which um I think Christof actually I have to thank you for having a very early conversation with me years and years ago on how I should consider engaging um kind of industrial organization experts in thinking about these problems and originally turned me on to the the the idea um so you know on the surface of it we're talking about interoperability of digital payments and the foundation has certain beliefs about why we think interoperability should benefit low-income consumers some of which uh the variety of papers have challenged recently in particular the last the last paper certainly if we don't also have consumer protection but we have certain beliefs and and so we've over the past few hours and tomorrow we'll also hear some of the findings that come out of the investment in this in this type of research and obviously that research for us is meant to help enable Regulators to make good decisions good design decisions um if they first of all decide encouraging them to implement interoperable payments Solutions but then to make good design decisions when it comes to that implementation that's very straightforward and things like that but I had a kind of I have to admit that I have a a secondary objective in doing this which was something that I think saan you're probably experiencing now as lead Economist in the world bank's infrastructure team which is the the real lack of industrial organization insight and knowledge in the development sector so I think uh Penny Goldberg many years ago who used to be the lead Economist of the World Bank obviously esteemed uh trade researcher at yel once mentioned to me how few if any uh iio experts there were within the World Bank many years ago um which is kind of funny when we think about you know the fact that um what we we think about kind of um some of the failures that lead to uh sustained poverty or lack of growth in lower income countries probably have a lot to do with IO related issues now my claim is that this is even more pronounced in the context of finance and in particular that there are some weaknesses many weaknesses in the way that development e economists in particular including myself um and maybe other colleagues on this panel can can speak to this have a limited tool set when it comes to interrogating the F market failures in finance in low-income countries right so broadly speaking development economists are using applied micro methods often looking at um uh not looking at General equilibrium outcomes but looking at um you know partial equilibrium outcomes and so we can say something about the context of individual interventions in in individual samples in smaller communities but rarely are we going to have the type of data that will allow us to say something about General equilibrium outcomes on the level of a market right and when we're suddenly thinking about tweaking Market rules in the context of finance and we're trying to understand the welfare implications of those it makes it very hard to reconciling that that broader motivation with the limited tools that that development economists often bring which is why I think thinking about IO which considers kind of Market level Dynamics also the competitive Dynamics and also Dynamics at all um I think is really important as we take seriously how research should or should not inform policy decisions and what I really have appreciated here is the marrying of both Theory and empirics um both in this form but more broadly in the research agenda and um putting both development economists and iio experts in the same room right and so you know just to to to to kind of make that practical when we as a foundation are trying to bring solutions to Regulators or ideas to Regulators I think the challenge but also the opportunity is for us to provide intuition from a theoretical perspective so Frameworks for thinking about the choices that should or could be made the tradeoffs that could or should be considered and then some of the empirical evidence to back up um you know uh some of those uh decisions right um it would be complete we'd be completely uh irresponsible if we didn't provide data and evidence to try try and back some of the decisions or recommendations at the same time I think we'd be irresponsible in suggesting that The Regulators can only be driven by data because some of the questions that we're going to ask just can never be captured right I think we heard earlier the lack of data to actually test a theoretical model because by construct we're thinking on the level of economies on the level of financial systems and we rarely have the de the data that we'd ideally have to empirically test some of our theoretical models so that's just a little bit lifting the the kind of curtains or pulling back the curtains to speak on Dual motivations of supporting this kind of work um and really uh by extension really appreciating the willingness of the researchers in in this program um um you know at tsse and Beyond to lean into the challenge of trying to get both um types of thinking in the room when it comes to this and just to as a final context to finally set the context this is not the only kind of work that we support um some of the other research that we support is much more traditional development Economist kind of Applied micro stuff um we are looking to um uh create better ways of um setting up exper so there's evidence there a couple there are a couple places where we know that mobile money is particularly impactful right so on smoothing consumption across shocks um um and also in terms of remittances right and so um you know part of the work that we're trying to do is to improve the external validity of some of those findings um and in in parallel create new methodologies for better are thinking about how to set up RCT and empirical projects more generally for improved external validity so multicountry experiments using very intentional sampling derived from structural models a group from NYU is doing that in South Asia for example we have an initiative at uh Edwards um Professor asu's university university of Ghana that's focused on interrogating agent infrastructure so the retail distribution infrastructure related to mobile money and digital Finance more broadly um we have an initiative at the the the B the um the um bra University that's looking at the impacts of digital Finance on women um in particular and gender more generally um so taking seriously the fact that women and men have different experiences when it comes to the financial systems in developing countries um zachi spoke a little bit about a very important population of women in uh in urban Bangladesh for example and then chabi um know well our work in consumer protection I'm supporting IPA consumer protection initiative um a very under researched aspect of the financial system which I think you know the last Model that we saw really motivated um you know the success or failure of interoperability to to to generate welfare benefits can very much depend on the presence or absence of sufficient consumer protection so that's just a little bit of a teaser of some of the work we've done some of the motivation behind uh this initiative and our support forward and more generally just an expression of appreciation for all the work that's been contributed so far thank you very very much set I think yes it was very clear so now CD point of view no I'm kidding please thank you thank you Emanuel and thank you for for the invitation it's really pleasure to to be here so I would just like to give you a a more broader point of view on this on this issue of financial inclusion financial consumation and financial education I'm not going to equational model but more telling you what we are what we are doing at the at the on this in this respect um so yes the the these three topics are usually what we call kind of Trilogy so we we we we we discuss them me they completely interrelated um we we discuss them through three main uh bodies so two bodies are more G2 bodies so dealing with financial inclusion is the global partnership for for financial inclusion and um for financial consumer protection it's uh following the two 2008 crisis the G20 created the task force and financial consumer protection and the third body is the financial Network for financial education which is not linked to uh the G20 but it's much Bader with more than 250 members and and setting some standards on financial education so the the these three main bodies are are setting let's say high level rules with regard to financial consumer protection Financial inclusion and um Financial education um I can maybe give you later if you want some some some more insights on this high level regulation which then let's say translate into all the jurisdiction all about the world and especially in the EU um so yeah just a few a few a few statistic or a few data that that that um that we have on on financial inclusion and of course thanks to the to the World Bank which is really at the Forefront on this on this matter with the thing that [Music] publication and so on so I mean we can basically split between yeah let's say de developing countries and and developed countries in term of financial inclusion we have seen with the the covid-19 pandemic that it has been Financial inclusion has been rising tremendously in developing countries um and now in developing in developed countries uh it's almost Universal um so we still have this two uh let's say uh streams in terms of finish inclusion um figures um but then when we discuss and when we gather all Regulators around the world the challenges broadly remain the same whether you come from a developing country or developed countries um so one of the the the the biggest issue that is raised by regulator and it's the the topic of many reports that we publish is um is about consumer empowerment and about the lack of financial literacy and digital literacy uh of consumers um be it in developing country or developed countries and um I will talk a bit more in depth about this literacy issues later but this is really one of the the the the the key Trend that is that is um identified um in most of the jurisdiction of the world uh and also the issue of financial resilience and financial wellbeing with the exposure to um frauds and scams which are also linked um to financial education so we conducted a survey last year about financial front and St and in more than 60 country basically the the the average percentage of people falling victim in of financial scam is 15% so of course there are many variation between the countries I remember the highest was in the Philippines and Indonesia where the number were really impressive and then you have also of course other country where this this rates are lower but all in all uh these figures are are quite worrisome and um so the numbers of SC and the number of people exposed to financial stum have raised um especially following the covid-19 pandemic um the sophistication of for and scam also increase tremendously so it's not anymore the only uh let's say easy scam on email but we have really R range of of scams especially with the um the the let's say the progress in terms of social engineering and the tactics that are used by fers uh to to towards U especially V groups um and the amount of loss of of each people being scammed also increased tremendously so this um these aspects um make the the this context makes the um importance of uming consumers uh through financial education really on the top of the agenda of many Regulators around the world so Financial education um if I can just give the Rd definition of it and which is now broadly accepted is really a combination of knowledge so the understanding and awareness of concepts of products uh available in the market for example um but Al thanks but also uh behaviors so the skills that people people put in place uh to apply for example this knowledge uh and also U the attitudes so let's say uh the state of mind and the of people and for example the way um they see their situation in the Fe the the future how they position them themselves towards future um challenges um so this is the the definition of fin education and when you mix that with digital literacy which is really uh how to behave on a digital scale you mix it with digital Financial uh literacy which is really um how you apply um educ financial literacy and financial education concept really into the digital world so digital payments uh crypto uh and all these issues um so this is a bit of a new topic now let's say after the the the covid-19 pandemic and it starts to really get into the the the regulations um and it uh apply to regulation structure regulation Frameworks through the applic application of uh National strategies um so that's really something we Advocate at the OCD we we we support members to to to to to set this this National strategies so there are now more than 90 National strategies around the world focusing on financial literacy issues and more and more countries more and more countries now are thinking about um launching some digital financial literacy strategies so really targeting some of these specific issues um so yeah maybe in the later stage I could I could elaborate more on on this issue how we measure it uh how we implement it but uh yeah this was really um how I wanted to to set the SC it's very interesting I think it's very complimentary to what we had theoretically before I believe the one of the many paper of Milo is relevant so now um I would like zav to talk uh to explain us the the transaction cost which is also relevant to me paper exactly following with the with a theme so um yeah so so uh the idea um was basically to to understand the cost of doing a transaction with a mobile money provider and so um now we can you know we can go to online and look for the price uh that's posted on the on the websites you know 70% you find the listed price of how much it's going to cost to do any transaction you know a cash out cash in so on and so forth um but basically you know there might be a difference between the listed price and what the actual agent charges you on the ground you know when you're in the field and in fact um the monetary cost or the the cost of a transaction uh you know the the fees that that the that the that the agent charges you may only be a part of the of the cost right and so here I'm thinking of course there's a there's a shetter cost of actually going to the uh to the uh to the agent and we we talked about Transportation cost but also you know you go to the agent the agent may not be there even though even during business hours regular business hours right I mean the guy might might have closed for uh uh you know temporarily my you know might might you know for whatever reason now even if the guy's there uh the guy May refuse to do the transaction because maybe they don't have flow they don't have cash depending on what kind of transaction you want to do right and so so the idea was basically to to document all that um and so we did that uh you know using different methods of data collection okay so um so we did consumer intercepts that means basically you know uh parking an enumerator nearby an agent and just you know waiting people to to to do a transaction and then you you kind of you know you intercept them at at the you know at the at the agent and then you ask about the last transaction that they did right uh um you can also do sorry another thing that we did was an audit study okay and so an audit study means basically recruiting people uh these could be professional people or this could be local people uh and then you give them a list of agents and you send them out to do transactions right and then you record for every transaction that they that they did that they attempted whether or not the agent was there and whether if the person was there whether the transaction was successful or not right now you you also keep track of the time that you know that it took to get them there the time time that they had to wait in line before they were being served the time that the the transaction uh uh to you know the time of the transaction of course whether it was successful or not and then how expensive it was right and so remember that you know and and you know you you guys of so the work of Francis Anan was was mentioned there uh uh so you know it's not only the listed price but actually these agents sometimes overcharge uh you know over and Beyond the the the you know the official the official listed price right and so um so again we we looked at um um and sorry so you can use professional professional Auditors these are folks that by by by Design will will have set foot into these agents for the first time so these are going to be first time interactions right now the typical uh the typical trans interaction with an agent is actually uh you know it's a repeated interaction where there's some there's some Report with the agent and so and so maybe you know maybe uh you know a dealing with a known agent may be different from a dealing with a new agent right and so and so we had then um we also recruited local local people that that were you know uh that that were that were incentivized to do to do transactions and again we send them to to the regular agents and we sent them to new agents to basically see whether there was any any sort of difference there and then of course uh we also asked the agents themselves right so you know uh do you overcharge do you close uh unexpectedly uh do you uh do you you know do you refuse transactions right um and of course you know you know we could you know so you can expect that these guys you know had in incentives to to basically not not lie I mean don't don't don't appear uh you know don't don't look bad you know in in in on you know when when when asked but but but you know you get a sense of of of you know different ways of collecting the same sort of data and whether and and of course you know uh um you know you can also uh look at you know uh how you know how important these these cost are relatively right and so um and so what we find basically is that actually the extent of overcharging sorry uh we should tell you where we did this uh so we we focused on Tanzania in Uganda and then in Bangladesh uh now there's a reason why we chose these countries because you know except for Kenya which is you know mobile money is prevalent these actually these are countries where where there's you know substant you know where money money money um mobile money is actually quite quite important in terms of you know the share of people that have accounts the share of people that use digital payments and and actually the share of people more recently that are using uh mobile money as a form of savings uh um so it's it's quite you know it's quite it's quite important um and so so what we find basically is that you know uh agents are typically in you know in the uh in the shops around 80% of the time and so and this varies uh by rural urban varies a bit by country um you know unconditionally the probability that you actually um that you actually do a trans action is about uh 60 to 70% okay so that means basically that you'll have to you know go to another agent or you'll have to visit this agent again uh in the you know to to to you know to to to complete a transaction and so what that means basically if you you know if you with some assumptions if you convert that into money basically it's around you know for a $10 transaction uh you spend about 70 cents just doing that you and that's just time cost okay converted into local wages whereas the monetary cost the fees including overcharging is only seven cents so 70 cents versus seven cents right so the bulk of the cost actually is non monetary and it has to do with time cost okay now you can compare about methods and there's some interesting stuff you know just just just so that we're clear actually when you compare uh transactions to an to a new agent versus a versus your known agent actually known agents are you know treat you better uh now that might be sorry treat you better meaning it's cheaper to them you know it's you know you're going to be more likely to complete a transaction that's maybe because you know when the guy is open you know maybe when the guy has has cash or has uh e money to to do the transaction and so that you know that would you know that as as opposed to you know just visiting one guy randomly which is what an enumerator uh would would do um and of course actually you know when you when you compare that to what agents say uh so agents of course uh actually they they they do report uh being being missing from the shop uh they do report actually being um being uh uh you know being able to complete a transaction uh you know a bit more than than than than the actual truth but actually you know we also looked at uh uh transparency uh uh measures such as whether or not you post the the least price in in your shop and whether you disclose the the the the costs uh after after a transaction now everybody says that they do disclose the cost when you ask uh you know Auditors uh it's about 10% right so basically they do the transaction they don't they don't really tell you and and also you know back to the back to the the last paper by by by Milo which is a you know it's becoming a seminal paper uh um so most most most guys most guys don't don't know the transactions I mean you know uh the transaction I mean we were we were asking about a cash out transaction and and most people don't and so so there's definitely scope for uh to to to to improve there I think I'll I'll stop here but but yeah than thank you very very much uh so now Mark will will will give us his a light so thank you so much thank you for the invitation and so it's a pleasure as always to be here and and to talk about this topic so I I'm think I think I'm the L expert here in this round table so I basically I'm here I guess because I worked some years ago uh about mobile payments and interoperability of Mobile payment platforms so it was in 2016 so I worked uh for the bank of mamb and uh with a colleague Sten hernik from Nova business school and so what I'm going to to say today is basically based on this past experience that we had with Stefan so so so we we've discussed a lot mobile payments mobile money today and uh so there are very important benefits to be expected so in the short run so and in particular it's uh about the diffusion of financial services but also to reduce the Reliance on cash which is a something which is very costly both in developing countries and developed countries so and but there there are also potential benefit in in the longer term so we can think that also what can happen is that the the division of these platforms of Mile money would stimulate uh innovation in complementary Ser financial services so if uh the diffusion of services is widespread and therefore there is really an important policy goal to increase the adoption of mobile money in in developing countries and uh so the question is how and then we can we have to ask ourselves so what drives the adoption of mobile Money Services in these countries uh so typically so it might depend on some uh local context so like uh the strength of the standard financial sector whether people are in banked or not uh in these countries it might also also clearly depend on the attractiveness of uh the services that are offered uh to Consumers for mobile money so in terms of levels of prices quality of service also the the variety of services not only the ab the ability to uh to uh exchange funds with PE but also maybe to to do some savings uh as was discussed and uh and finally what matters and what was uh discussed today in particular is the size of the network because typically you you want to join a mobile money platform if you have a large number of opportunities to to send and receive uh money from uh from your family and friends typically and um and so here comes the the the question of interoperability because uh we might expect that with with interoperability thank you um uh then we different Networks are going to become shared and therefore Network effects instead of being at uh the provider level they are at the market LEL and then with larger Network effects mobile Money Services might be more attractive uh for consumers um so let me um so let me think uh so why does it matter so it matters because so if networks at the the different Network at the beginning are too small then maybe each Network might be below what is called in network economics the critical mass and therefore services are not going to catch up so bringing them together with interoperability will lead to a larger Network and maybe allow to um pass this critical mass and then uh the networks are going to to to ignite basically and secondly also in some markets U there might be some dominant player for historical reason for example and uh an interoperability might be a way to LEL the playing field between this dominant player and the smaller player and to increase competition and if we believe the last paper of Milow might not always be the case um so so one question which uh is important perhaps is how interoperability can be achieved so whether it can be achieved through the market uh or whether there is some kind of regular intervention that is warranted and this this was a question what which was asked to us actually by the bank of Mambi so years ago and uh and so here I think so what we can do is uh think in terms of basic Network economics uh which are going to tell us uh the following so that it depends so which is a stand answer of an economist um but and it depends on what it depends on perhap the maturity of the market so typically what we can think is that at an early stage of the market firms are going to compete for the markets are going to try to build their network of user but also of agents for example and so this kind of Market Force might be beneficial and might allow the market to grow uh and that perhaps at this stage at this early stage imposing interoperability is not such a good idea and um and you you don't want to to to impos it as a regulator or as a bank as a central bank uh maybe at a more mature stage of the market if intervality does not emerge from the market players themselves because it could in principal because they might see an interest in having a larger Network after all um so if it does not then maybe some inter intervention is necess is necessary um so here I think what also is um the implementation uh details um so I'm talking about that also because I'm working currently on the digital markets act uh for the European commission and on interoperability of messenging services and here this is a topic where the details matter but uh I think it's it's the case also for mobile payments and uh we' seen that for example in terms of consumer protection information of consumers things like that can can be important uh so in the digital Market side there is something also that is discussed and that might be also something relevant I think for mobile money which is the security of transaction and how interoperability can affect the security so for example for messenging Services there is this concern that if you impose interoperability it's going to reduce the security of uh messages so ofet messages are less secure than onnet messages and we might wonder whether it's the case also for mobile payments for example or mobile transfers uh whether uh when you interconnect different networks uh different mobile money platforms it can raise uh security concerns and even increase uh the amount of scams uh which could be u a problem um and um and um so I think so standardization has a role here and um so I think so in the context of the dma in Europe this is also a concern um okay so the last thing I wanted to say so it's about multihoming so so I I I've seen that MTI homing something wi Sprite for mobile money platforms with the use of multiple seams by uh uh by users and therefore here we can think that interoperability can affect effect also MTI homing and so since MTI homing is a way for people to actually make different networks interoperable without interoperability imposing interoperability may have an adverse effect on on MTI homing reducing the amount of MTI homing and um and so I have a paper another theory paper D and and in this paper with myor Yan what we argue show is that in this type of context when you impose inability and there was multi homing actually you can favor uh the dominant players because if uh as a user I can um send phones and receive phones from any network so maybe it's just better to be on the main Network the focal Network the big Network and and and and not on a small Network because then I can interact with everyone so maybe interality May favor the dominant players which would go cter to the objective of increasing competition through interoperability thank you thank you so much thank you very much it was a very very clear very interesting to and um now last but not least pigs thank you thank you gas thank you man thank you for the invitation I'm very very glad to be here and share our experience developed PS I hope to bring very insightful things for you today and as said mentioned before I think this joint comparation between regulators and Academia is very fruitful so uh let me see let me see I think yeah okay thank you a lot so before I starting starting talking about PX itself I think uh I need to to bring more details about uh the context of pigs because people may think that we launched the fast payments and in one day and the following everything was working perfectly but the PS was only the missing part uh everything started maybe 10 years ago when we build a h a legal framework create a new type of of Institutions uh with a light the presedential requirement proportional to the risk uh it was the origin of the payment institutions in Brazil pay institions Brazil they cannot Grant loans but they can offer prepaid and credit cards and it was important because at that time one diagnostic that we have that we had in Brazil is that people needed needed digital payments uh 15 years 12 years ago in Brazil people use a lot cash and we we needed to to change that so how we do would do that we should create new institutions uh with a very very light Caper requirement and very simple license process because with that uh settled we could have a very a lot of Institutions providing these means of payment uh so these rules they were limits to their exposition to risk and this tutions they became very popular in Brazil notably by offering free of charge accounts and userfriendly technology and all service are offered through smartphones with no need need to visit any physical branches and it was very important in Brazil because we have uh regions in Brazil that are in the middle of the forest so it's impossible to have ATMs or even U branches so uh as those uh institutions became so popular that even the incumbents they responded offering similar products so when pix was launched uh the use of cell phones for financial transactions uh was a red WID spread well and unfortunately another factor that uh was happening when pix was launched was the pandemic uh pix was launched in November 2020 and we were in the middle of the pandemic and and digitalization imposed social and the digitalization uh imposed by social dist distance policy have changed the habits of non Brazilians and they had to incorporate uh mobile technolog so uh was very was a real need for people to make transactions Via Mobile and in addition to that the Brazilian government paid during the pandemic uh an emergent Aid through a j account A J account provided by a state own bank and in that time only three months uh about four 14 million unbanking people were inserted in in our financial uh system and now of that with digital accounts so people really had that in mind it was people were were familiar with smartphones doing transaction with smartphone and one of our worries at that time is that big taxs were arising uh offering many closed loops and we needed to to do something to avoid that friction so what we wanted to do is leverage this mobile widespread adoption uh and through a public policy provide something that could be open um for uh a big Tech or any big player offers something closed and with all the frictions that would come with that uh and even at that time uh we were talking with the industry we were uh the moment that we launched uh payment institutions that we changed the Congress changed the the law giving the Mandate the banks to the central bank to do all all these reforms we were talking with the industry why why do they not offer instant payments why do they not offer something that could overcome those barriers of course it's understandable because they they do not enjoy the incentive as the payment revenues would be jized and moreover developer from scratch uh not on not only instant payments but any kind of of uh feature that empowers the consumer they need need they need a complex coordination and started setting activities uh and the need for an open architecture fully interruptable and C G from bias governance is very very difficult to flourish from private provision so uh to finally go to the paks what considering all this framework we we posed that PS from from the beginning should be open accessible simple and safe and here I bring in 10 features that we consider that really underpins uh this directive to make pics open accessible and simple and safe and I I I would start with uh the first feature that we say that's uh having a single name a single brand and standardization why uh because we want to create this a simple payment journey and we establish a single logo that's stated in small letters besides the the image powered by Bank Central think you can see the logo has Part B Central and Conant we prepar the uh user experience manual to standardize the offering of PS by different institutions every institution must follow these rules when offer picks uh the idea here is to make sure that consumers understand that there's exist a public in infrastructure behind the instant payment and this is very important because uh uh pix wall was launched in November 2020 but one year after we launch the the logo was the first time a central bank was uh making uh was marketing a product so every journalist every news was talking about pics and since the beginning as the central bank was the first one to to offer paks and say that PS would be a commodity so it's just a payment it's not a product nobody's going to sell you a better P than other pics so it just commod every journalist and every news was was talking about it so it's like creating a a a fear of missing out effect because since from that day the private SE sector would be engaged they know that they should offer that and this is where what I I enter the next steps that is governance mandatory participation and scheme openness because by Design This is another important thing no one would need to download any apps or register on your websites to use PS all the consumers they they could use the apps or of their own account providers uh which were already known and trusted in other words PX was born integrated to the day Liv of of people and companies consequently uh being uh offered PS been offered within the own applications uh people see PS inheriting all the application security system everything that have been tested so you don't when marketing picks we were not marketing a product where people should trust that was safe or it wasn't uh available in a a new and totally different way no you should open your app you see the same logo in an app that you have that's what why the standardization standardization manual was important and I remember that I was uh doing development of the this user experience manual I stopped myself many time thinking about and I'm and I'm being so intrusive because the manual says uh institution you must offer pics in the first page when people uh click on the pics the next page will will present those following Fields so we did everything for uh to make a really standard but this is again this is what this is a very important issue because when we we started with uh pics people was able to to testes it was free of charge people could send pics for your mother father your neighbor just to test so this was very important well it and I forgot to say that's why I have my notes Here not forget uh pix has many attractive features for consumers because it provides a comprehensive and costree and non bureaucratic payment mechanism for both payers and P that's why if you travel to Brazil you can see that guy selling coconut water on the beach having a QR code printed so you just scan that QR code and pay for a coconut water you don't need to carry money and more uh we have some some figures showing uh the the amount of with draws within the the Brazilian system and the level is very very lower than it was in the beginning so people are usually not using cash anymore of course cash is still very used is not will still be very used in the future but uh people are uh step by step using more digital means of payment well and now it's important because it's very related to to men many issues that I saw here during some models that were presented because once the supply strategy that I was was describing here was Define The Next Step was think about distribution so not playing the the role because the Central Bank of Brazil has a twofold role we are the scheme owner but we are also the regulator of the financial system so using this role of regulator uh the Central Bank determined that the large institutions in terms of number of transactional accounts must offer pics so with the largest institutions offering pics the other institutions fall followed not to lose customers so as a result from day one virtually every Brazilian could make a make or receive a pics I remember I remember that in the lunch of of pics we had almost 800 institutions offering pics so basically every regulated entity was offering pics from the day one and the other supporting pillar of PS is it's its infrastructure so the goal was not to have it the goal was to have it accessible neutral on competitive grounds and low cost that's why the Central Bank of Brazil plays this role offering uh this decentralized infrastructure uh we adopted cost recovery approach we charge only one cent of Brazilian H for 10 pixes transactions to the receive institutions it's about uh one cent of dollar for every 50 transaction this is the cost of PS and the same time as the scheme owner we determine that all payment soltions uh that all the transactions between uh individuals would be free of charge even when an individual send a p to a m the individual does not pay and it is important because uh if PX aims to to to Aims to play the of cash people must feel that have the same structure if you don't pay when use cash you shouldn't pay when using pics if you don't pay when making transaction with cards you shouldn't pay it pay using pigs this is the the structure so um this is Fe structure uh where we we turn very very attractive for for individuals um make sure that consumers would be quickly on board on pigs thus uh PS would be a most adopt for firms consequently this strategy solve this coordination problem in here to Industry that exip Network effects so this is the the context of of picks here I bring just a a figure to show what I mentioned before the Central Bank plays uh plays the role of settlement infrastru structure and also prox databases because to make a px you you only need to to insert a key course uh datas of of pix maybe your um phone number ID email or a random key you can choose each one and you can uh choose in different situations you can you can search your uh if you have more than one account you have you may have one key and one your May one your cell phone in order and it was interesting because when the industry uh saw that having the main key register every institution started to to have this key before so the the only participants of the market of the industry uh started to marketing that your customer should register the the key before on that stuation so so also helped us a lot and here have some figures uh about the broad adoption we can see here on the left that we have PX transactions in every state of Brazil and if you can see I'm finish it mainly on the the north parts of Brazil uh p is very important you can see that as on those lowincome States uh we have we have more uh more difficult in providing physical infrastructure so people they don't have much ATMs they don't have much branches so they use pic a lot on the right side we we cross reference uh the alas information with the me of Labor employers uh data and we can see that every worker they use pics so if you see workers that receive up to one minimum wage about 11% has its Keys register and if you go to the others you have up to 9% this data is from uh December 2022 so we uh probably now we have more data and here uh what we have here uh we are we are comparing transactions you can see when P was launched the adoption was so fast that three years after we have almost the same transactions uh using PX with cards uh CS depth and prepaid cards and if you can see the with draw uh is red is falling um in terms of of volume um we have of course PX is not largest one we have the electron transfers electron transfers is is the way uh we have interbank transfers that's why the volume is are so high and is interesting I was talking with Milo early in this morning we already had the electronic transfer since 2002 and people could make transfer electron transfers uh but it was only during business hours and from Monday to Friday but was very expensive and and that times only banks had could offer these electron transfers non bank soltions they did not offer electron transfers and consequently Banks never offer that it was on it's on app so it's never worked well only now with that it's something that we learned when we did this the standard protocols because we we had that experience with Theon transfer nowadays you can see of course when pix was launched main transactions was P top uh but nowadays I I going to show the cost in the following slides we have more p2m uh transactions almost half of the transactions this data is from sep October now it's about 4 n billion transactions per month as you can see here one of the advantages for for the misss is that the cost of p is much lower than debit and credit cards so it was instantaneous accepted for misss and here the last data about 85% of the ad population Brazil has one key register uh with pics and if you consider 77% of the companies already have uh a uh a key uh res res with PS and to finish uh we have when we were designning pics we we had uh fourman drivers to to guide to guide our public policy and I think that we we are proud I think we we have reached that because we want to make something to ease the life of payers and payes and because very not is not bureaucratic is very easy uh we want to make new business model this is something that we always think about when you provide a public public infrastructure it's only going to go forward if the private sector can offer products and service on top of it so bu so pix is very flexible to receive that so we have companies offering by now pay later or offering and service for that and we want we wanted to decrease the costs and Empower consumers and one of the benefits of the the paks is that it practically eliminated the switching costs uh it's very common in Brazil to have more than one account people have two accounts and with this uh very or with these no switching costs and with those light Vehicles you you also have business model in Brazil that are very niched very focused for example uh I have a a banking account where I receed my salary but to use my credit card to make a crossb transactions very expensive they charge me a lot of fees some hidden fees and I know that there are some Payment Solutions that work only offer offering adap car to make uh crossb transactions so as as uh I was coming here to tulus last week I open an account at the same time I receive a digital credit card that I I put in my wallet my digital wallet and three days after I received the plask so I could do and and I can send money using pigs so if I want to make a purchase uh I any time I send a pig to this account automatically convert in euros and they can spend with the the card so uh it so having this kind of tool that's very cheap you you leverage business case so I'm finishing here sorry for I put more than but you you were invited to exactly to do what what you did thank you thank you very much Carlos um so we have uh yes I think uh uh so first of all it's uh I believe it's a really unique experience on Earth and a very successful one and it Liv at least it Empower a lot of poor people so that's really exactly why we are here and I believe it was very well designed presumably because of our good colleague rato Gomez um so we have 15 minutes left so if if uh you want to to ask some question please raise your end uh we'll be happy to uh share the mic but um if if nobody has question I have some so oh you have fantastic so so my question is to uh Carlos and maybe chav can also give some insight so you know when this whole fintech thing came everybody was like okay so the P digital payments are the most disrupted part of you know Financial system and but few I mean there been quite a few years since since all all this started but we see that the incumbents like Visa or Mastercard they are still pretty strong like their their profit margins are like in 50% or 45 like really high compared to any big digital or big Tech firm we know right so in in Brazil for example when when PX came what did you see more uh competition and and you know if there the merchant fee also went down or or you know just just if you could provide some insight on that thank you okay so very good question uh in Brazil we have uh very high MDR fees The Interchange fee even after the launch of PS didn't go down that's why we implemented gap on interchange fees on depb and we are planning to to do the same for credit cards so it's very very hard when we we talk about card industry we have dual dual poly between Vis and master competition doesn't work with that industry uh competition only gives more High inter interchange feed and really demands intervention so we did intervention with d cards prepaid cards and probably next next year we're going to implement credit cards so hanat is helping us a lot because he's the one who when he arrived in Central Bank say getting his background here from Academia he said he's a typical case for Implement cap and at the beginning as Regulators we we said are you sure because you're going to to make a really impact on on that model because uh the card industry they played an important role in Brazil because MDR is high okay but the acarus they went to the last mile when we launched paks we had POS on the most decent place so there's someone in a motorcycle taking a POS to a very small grocery shopping down for us so they charge for that and but there's important role so you always were thinking about can we or can we not intervene because if we push a lot the cap on that invers it the revenues will go down and maybe those parts of the of the country wouldn't would lack some payment options but now after some times we we yeah we concluded that we need to do that cap uh we already have other uh means of payments we have Peaks or the debt cards are working proper with uh lower uh interchange fees so we are going to to implement this cap okay um zavier do you want to add something or no but I have a question for car okay good body has a question for cars you are right to come here you see so can I yeah so uh so what are the um and sorry this is maybe a bit off off topic for the conference but but I think important nonetheless what are the consumer protection concerns that that you see in Peaks that you know the major consumer protection concerns is it is it fraud is it you know over you know excess credit as as fintech might might use the pig data to to give credit yeah so what are the you know what are the what are the big you know what what keeps you awake at night as a as a regulator I guess yeah good question we were also discussing that I think the first one is fraud we have many uh scammers using social manipulation so those we have very interest uh creative frauds in Brazil people that he call you saying that works in your in your bank and you should uh and someone's trying to to to transfer money and you need to do something to stop to stop please give me your your credentials and people get that or even someone call you with a very noisy phone call say oh I kidnapped someone oh my daughter yes I kidnpped your daughter give money so but uh car is okay oh car yes no keep that car so this social manipulation is something bad and what makes us let us awake at night is probably related to know your client problems because we had a WID widespread adoption of Payment Solutions and we are tightening the process of opening account because if you have more more if you have better Pro to identify who is opening an account you you make it more difficult because the money is within the financial sy so when you make a pixel you make a PIX to another account of course after that after that you make a Cascade going to WR going to many accounts but if if you have control of that you you may you may try to stop that so now within the the PX infrastructure we're launching probably at the end of the year a tracking tool so we're going to be able to track the P transactions up to five accounts of course we're not publish that be five accounts because the fraud but we're going to to to track the money we we going to anybody that says oh I was uh I was caught in a fraud the payment can freeze that money up up to five and also we make some experience using AI in order to identify who is potential uh ID or potential fake account sorry so so just to be clear the concern here is that you know when you're a victim of a fraud what the what the fraudster do does basically once once they get money they of course dispers I mean that that that account is basically you know the money is moved and and so that's that's what you're trying to prep gotta thank you very much uh I know V has a question but I would like to ask Tom very quick for this type of f what type of education you can because this is what you you are saying initially that solution is to educate and that's true we can educate people against uh scum but I was a bit curious you were a bit quick on that and I believe it's a very important topic so if you have a few example or something to say yes no thank you I mean I totally see what you mean we we hear that every day um I mean most of the fraud they don't come from malwares or troan horse like I mean it's there is already a human factor and someone vulnerable that that that that gives some personal data or something like that so I mean there there are there are different uh different um solution I mean it could from it could come from the the regulation I mean you can make PSP responsible for for fra laws and something like that um so this is this is another point it should come from the law um there are all the supervisor actions from the from the supervisor supervisor authorities so mainly based on the on um on complaints like complaint handling and you make some some consumer turn risks and you you you go to make some uh some on-site inspection or offsite inspection um and from the your question from the financial education point of view so the the the biggest um the biggest type of initiatives in terms of number it's really the kind of awareness campaigns I mean warnings and awareness campaign and and to to to kind of force consumer to to check the Black List the white list of of of providers on the on the website of the authorities so this is one of the big action um then awareness campaign so it can take can take many forms it can I mean most of the time this days they are on social media so with the use for example of of influencers or of or really celebrities to to to to make people aware of the these techniques which are often quite simple I mean fishing is quite simple you just give your your information by yourself so this this are really the let's say the ad initiative um and then in the longer term I mean it has to come from uh education I mean in a school context uh for for for children so from a young age but also at high school and students so um there are many many things happening the best of it is to to to have some some some courses about that in the as part of the school curriculum and uh to to to to provide some information through and not only information about knowledge but also some real case studies and how to avoid the scam and show the the steps um so yeah I would say these are the two main types of of of channels like really provision of information adoc warnings and then some more in-depth uh behavioral U type of um initiatives thank you very much l yeah I just wanted to mention two other kind of approaches that we are supporting that complement some of the customer facing education um one is you know in comparison the Brazil there a number of lower income countries that don't have the same capacity Among The Regulators uh in technological capacity and so we're supporting investment in in open- Source Solution that's kind of a rule-based um basically software to um uh enable Regulators to kind of like freeze transactions that appear to be fraudulent and things like that and the point being that um you know you mentioned AI but even just what we think of as like very basic filters for example are not adopted necessarily in in many of the lower middle-income countries um and then the second solution is one that came out of the Innovation for poverty actions consumer protection uh initiative which shavei knows well which was professor Matthew bird collaborated with IPA working with the Ugandan Communication Commission and the Ugandan Communication Commission has authority to demand what they call Customer Care records complaints from the mobile money companies so something happens to me I complain to my mobile money provider and then the regulator can demand those and so they have hundreds of thousands if not more of complaints and so what the the researcher did was use kind of natural language processing to look um and and kind of machine learning algorithms to to look at the C consumer complaints to create actually a predictive model and one of the conclusions of this paper which is online you could look at is that you know you actually can predict to some extent um who will likely be the targets and that can be really helpful not just for preventing or looking at those transactions but proactively reaching to those folks I think it in this case it was like men uh in a rural context I think at a certain time of day I can't remember what time of day that we're most likely to kind of complain about about fraud and so you could Target people in a much more you know sophisticated way as opposed to general education providing targeted messages to those Folks at the moment when they're most vulnerable according to the the analytics thank thank you very much I think it's super interesting Milo had a question maybe on your paper you want which one um what should I say no one thing I was um you know the the initiative is trying to understand the relation between uh competition and consumer welfare and one thing that was puzzling and striking in in in Sav work was that you this correlation across countries in which fees tend to be lower in in countries which more do which were more concentrated in terms of Market structure you know so one maybe naively could could could just conclude from that that competition is uh is not helping reducing fees and I wanted to to hear your thought and you know other panelists have have worked on on this complex issue of of competition in this type of Industry uh so I want to have a sense of uh whether you you know you had some ideas on on that pattern and how you would interpret it and so yeah it's sorry it's still it's still work work in progress um so you know there's there's various correlations here so there's correlations between the listed price the official price or the official fees and what they actually charge um um and uh and then you know when we talk about competition do we talk about number of operators do we talk about number of agents in a given Market um and so and so these are these are things that we can we can we can start to explore um you know also looking at sort of the quality of you know where this data is coming from you're looking at data that you um that you you know uh data from the ground right I mean so are you using data from the audit studies are you using data from the the agent reports are you using you know what what you using another another interesting correlation that you can look at is you know the cost or or reliability based on the number of customer so do customers know where you know where who the most reliable agents are who the cheapest agents are even even you know you know so just locally in a market right I mean so are the most popular agents the ones that are more reliable the ones that are cheaper is there a correlation between reliability and and and and and cost right so so all of these you know so yes you can look at you know sort of the number of Agents as a measure of competition uh you can look at um you know the number of clients that they that they kind of have uh as well and so sorry all of that you know we're exploring these things I don't have you know headline results for you to you know for you to you know for you now but uh but yeah no these are yeah I mean it's yeah but but in any case I mean the relationship between between you know competition and fees or reliability is by means you know by no means uh straightforward uh yeah I think Stefan has a question I would like just to make a quick comment on what Mark was explaining I believe um this parallel I mean this surprising fact that interoperability can be bad uh in the sense that it can make the Monopoly bigger the dominant it remind me of Roads if you have a very rural area and you want to connect it to the big city the op that it will revive no it might empty it and that's that's something we know about roads they can do exactly the reverse of what you had in mind saying I'm connecting this PO place to help it to thrive and everybody say oh now I can go to the big city and so maybe interoperability some do that just say okay now I can go to the big city and I don't need small so it might be against competition thank you uh I'm just a comment maybe it's not really a question but circling back to the initial uh um sort by set on on on you know bridging research and because it's a little bit what I'm trying to do now also and I can connect to the difficulty of that so I agree that you know bringing more IO but not only IO actually we need also more macro we need um more economists in general you know a lot of people have the title Economist outside in in the policy world but they don't really do economics actually in the way we would think about it here uh but that being said I just wanted to say I'm sorry if I sound a bit negative but I'm I'm really struggling to understand the big picture here I understand a lot of I see a lot of case studies and and papers about including what we we we've done right on on specific cases but I mean one one is a you know we have a government digital platform and it relies on online banking or there is mobile money or there is maybe you know car the car industry I mean I have the impression that there are all these things um and I don't see a general framework to think about it uh I'm I'm really struggling about that something I see at the World Bank as well is really difficult people don't have a general view about what is the economics of digital in general I think digital seems to be a general purpose technology that's useful for a lot of things and works well when other things are there but doesn't really make a difference if if other complimentary inputs are not present um and that's one angle it might be useful to think about it but I just wanted to convey that you know I've listened all day uh trying to say you know focused on on a lot of very interesting thing but I really missed the big picture and and it's not only about this conference it's in general about these events about digital you know everybody's super excited or AI is going to solve all the problems of the world you know but what's the economics of AI I have no clue to be honest I mean so it's is just a we all going to commit suicide thank you Stefan to me I ask Carlos to end up this talk because it's a success story and work well it's a developing country and here you are okay I just maybe you know I'm kidding we need some of that right sorry I I jumped that gun sorry uh so just to be clear so if we're talking about digital payments uh um and um and you know the fact that you can use you know whenever you transact digitally you leave a paper trail and these can be used as as uh as uh you know there's definitely links between you know use of digital payments or uh um to to consumption smoothing to uh to to to say you know so improving the safety net to access to credit and to actual real outcomes like you know productivity and sales and so on so forth so so these these linkages that that's true I mean consumption smoothing the a lot of evidence uh jobs and macroeconomic activity I mean I haven't seen so much but you maybe I missed it but yeah yeah well I mean this is also you know I guess you know people are working on this so so maybe we'll we'll see we'll see that but but uh so there's definitely you know and and people are trying to make these you know this connection I mean ultimately you know how does that affect welfare and you know what's the distributional consequences of of these right but uh but yeah sorry s that would be the world of the end also so that's look at that you are the perfect panelist okay the only thing that is missing yes is long air and maybe feminine genene somewhere but that's why I go you put it together that's all I pick the best um yeah no so I think it's a really important question um so as as chabby mentioned there's some very clear I mean so it a little depends on your perspective right so if you're perspective is how do I create a stable Financial system then you're worried about the risks that digital introduces um so for if I'm focus on credential regulation I say okay let's make sure the digital doesn't screw up my credential uh system and oversight if you look at it from the perspective of alleviating poverty you say okay there's really impressive evidence that um shocks household shocks are a major cause of P poverty traps and subsequent evidence that um digital Finance enables folks to much more quickly access emergency funds in a moment of shock so how do we build on that poverty alleviation intervention or in some sense poverty prevention intervention um in a meaningful way right if your perspective is economic growth there's a notion which I struggle to sometimes find evidence to articulate that there are credit limitations in access to credit which basic in developing countries which basically everybody that lives in developing country knows and everybody who seems to work on development country knows but people struggle to actually create evidence to demonstrate nevertheless there seems to be some kind of credit market failures that are associated with that limited access to credit and so you say okay we've got now this like compelling Reon to we have digital Finance in the form of mobile money or whatever it looks like in any given country and you say it feels a lot like that could be the start of opening up credit markets in much more productive ways right and so I think the problem and the reason that there's not a convenient framework to think about all these thoughts is because Finance is kind of everything right like the interest rate is the shadow value of you know uh in the economic model it's the most basic construct of kind of Commerce in some respects um it's not fin for you know so we can think of Finance as kind of capturing everything all the time and so in some sense that's exactly why thinking about finance and how finances related to development is really challenging and why I kind of appreciate this willingness to think about it from so many different perspectives um I think that the way to um to better wrap our heads around it is to start with the problem that we're trying to solve and so at the bank if you're trying to solell for poverty uh in the poverty group for example the thing about it in terms of how to promote consumption smoothing in a more effective way or how to Pro provide um you know if you're thinking of it in terms of economic growth thinking of it in terms of credit markets and better Insurance um and then if you're thinking about it in terms of just delivering uh government services more effectively acknowledging that the ability to send cash transfers instantly in a very targeted way in a safe way um in ways that could then subsequently be passed on to family members that need it most through digital kind of infrastructure you know those are four different ways that they about the problem all of which the bank someone at the bank is thinking about for example at any given time but just to think about Finance Rate large I agree is is is very difficult which is why I really and this will be the concluding thought is why I really appreciate frankly the way that academics and economists come to some of these problems it's not to look at the solution but it's looking at the fundamental question that we're trying to answer and I think that's always what we have to go back to so end with them well I think indeed it's a beautiful word of the end I would like uh you and me joining to applaud them because some of them come from