 Good afternoon everybody. Welcome to this webinar, which is part of the ThinkVider webinar series on domestic revenue mobilization. And today the actual topic of our webinar is on harnessing big data and ICT to boost revenues. My name is Jukka Birkele. I'm a professor of public economics at the University of Helsinki at Watkinsville for Economic Research in Helsinki, Finland. And I also serve as a non-resident researcher with UnionVider and I work with colleagues there on matters related to taxation, especially in many African countries. So the webinar is related to the UnionVider research work that we carry out pretty much using data from administrative sources, namely revenue authorities in African countries. So we do this together with our colleagues so they can come from research organizations and they are also coming from these revenue authorities in African countries. So we have research activities ongoing with the Uganda Revenue Authority, Rwanda Revenue Authority, Tanzania Revenue Authority and Zampien Revenue Authority. And in addition, UnionVider works together with the South Africans on research related to South African economic developments, also on the basis of tax administrative data. And really what we use the data for is for analysis of enforcement and compliance, how can we increase revenues. And second on looking at the implications of taxation on taxpayer behavior in terms of employment levels, investments etc. And then also, this is something that Amina May touch upon, is that the tax data is also useful for looking at various other things because it provides information about all formal sector income earners and firms. So one mention one more thing, so one stakeholder and contact in our work is also the people who do technical assistance collaboration with the revenue authorities more from the practical side. So this includes for example the Finnish and the Norwegian tax authorities who have engagements in Africa. So I'm very pleased that to get to today in the panel we have representatives from all these stakeholders. So we will be starting with the presentation by my colleague Amina Ibrahim, Dr. Ibrahim is a research fellow at UnionVider and is the lead for this research that we do. And we are also joined with Julia Maskakni and Dr. Maskakni is the research director at the International Centre for Tax and Development in the UK. And we have been doing joint work with Julia for some time now and we are very pleased that you will be able to join. Our second panelist is Alen Nasanga who is the assistant commissioner for research and innovations at UK under revenue authority. And we have been working together with Alen and her colleagues for many years in researching these areas. And finally we have Mr. Timo Lalkanen from the Finnish tax administration where he serves as the director of the strategy realisation office. And with Timo we have been talking and doing work on technical assistance and developing tax enforcement joint. So as I said we will be starting with the presentation by Amina and then the panel will follow. People in the audience, you can already during the panel think about questions to Amina or any of the panellists. You can write them into the chat and then in the Q&A after the panel I can raise some of the questions from the chat to be discussed by the panel. Or then after the panel you can also raise your hand and then you can just unmute yourself when I ask you and you can yourself then place the question to the panellist. I think that's more or less everything from my side so Amina please now the floor is yours. So I'm going to start with evidence-based policy making and apply it to our thinking at revenue authorities and data and harnessing big data. So what I want to start with is really thinking about revenue authorities, what their job is and the challenges around that. So revenue authorities have this challenging task of raising revenues to fund public spending. And in order to do this revenue authorities might want to know whether increasing the tax rate or enhancing compliance helps them to raise revenues or whether a combination of both gets them to their revenue targets. Now in order to answer these questions revenue authorities need evidence to decide on what works best. And good evidence helps to create good policy and in order to do so revenue authorities require good data. And in this little diagram I've got good data is actually one of the crucial components to evidence-based policy making. And with data and some variation revenue authorities can produce some much needed evidence actually to improve policy. And the cycle continues when new policies are made or policies are changed. This creates a new source of variation and ongoing and continuous data collection can actually assist in creating new evidence that then again informs policy, whether it be new policy or updating policies or improving policies. How does this apply to taxation in the African context? Well many African countries already collect taxpayer information and they do so on a regular basis in order to calculate taxpayer liability. So you can think about collecting personal income taxes on an annual basis. You can think about VAT collections on a monthly or quarterly basis. And in recent years taxpayer information has moved from paper based collections to a digital format. And this has come with the onset of e-filing for taxpayers. And in some countries the use of electronic fiscal devices provides transaction level data captured in real time to be used by the revenue authorities. But ICT and using ICT is not really a new phenomenon for revenue authorities. They've been using taxpayer information to assess, for example, the riskiness of taxpayers and think about who they should audit and using this measure of riskiness to determine who they should audit. And what we've seen in recent years goes a little bit beyond this where revenue authorities are actually now using the administrative data to conduct research to answer various questions about compliance and on ways in which today's revenues. So I'm going to provide three snapshots and examples of research concerned with raising revenues and compliance, actually using tax data in three of the countries that we work in. The first example is from Tanzania. So the Tanzanian Revenue Authority worked with the Finnish tax administration to develop an Excel sheet that was used to identify taxpayers to be examined. So previously taxpayers were selected for examination by tax officers. And so this is an example of using ICT to make a risk assessment of a taxpayer coming to the tax office. The pilot study was run in the Dar es Salaam region and tax officers were trained on how to use the Excel sheet appropriately when taxpayers came into the tax office. And so the outcome of the study actually finds that this risk-based tax examination led to some moderate increase in additional reported income. And this really means additional revenue for the revenue authority. And we know from other studies that it's not straightforward, not all respect selection automatically improve outcomes. And sometimes we see some immediate effect and over time the effect dissipates. In this case, we know the results hold immediately, but we're not sure maybe for one or two years after how the examination has panned out. And so more research is actually required in this area given some of the sort of mixed results. The second case is from Zambia. So in 2017 the Zambian Revenue Authority introduced a withholding VAT mechanism. This mechanism meant that some VAT agents would withhold and remit the full VAT to the Revenue Authority. And this change was based on this understanding that withholding is an approach to improve tax compliance. So the research study tries to assess the impact of introducing this withholding VAT agents on firm reporting decisions and on tax revenue. And what's transpired is that the data compiled by these withholding VAT agents actually serves as a check to prevent false claims. And the paper trail that's been left behind or generated by this withholding VAT agents have also compelled suppliers to file their returns. And because these records are now, or these claims are filed electronically, the Revenue Authority has been able to more easily verify the claims. The study is ongoing but the early results show that the reported sales and value added tax increase for withholding VAT agents after the reform. And then that has actually also been an improved VAT compliance due to this change in the reporting mechanism. The third example is from Uganda. So in 2015 the Uganda Revenue Authority introduced a new electronic system of e-filing for presumptive tax, which simplified the filing of small businesses. So this is an example of a change in the technology to capture the taxes of small businesses. The small picture on the right is a snapshot of the e-filing form that taxpayers could use on the URA website. So this was an additional or new way for small businesses to file their taxes. And while some are still using Excel sheets, the take up was quite high. And the study shows that the change led to an increase in the number of tax filers, although some simply report the minimum income, but that the intervention also helped raise revenues in a small way. So these are three examples from research that we've done to demonstrate how ICDB can be used to raise revenues. And all of these have been conducted in collaboration with the revenue authorities using tax administrative data. Now harnessing big data is not the easiest task, and many revenue authorities have gone through this process of tax modernization. And one outcome of this process we think is about data accessibility. So with the modernization process data are collected and are stored in organized manner. Electronic filing and further use of ICT also means that information about taxpayers are more easily collected. So e-filing can also mean that the data collected are more complete. Taxpayers have less chance of skipping fields on the tax form where they are mandatory. And fields could even be automatically populated with information provided previously to the revenue authority or from other sources. As revenue authorities introduce electronic collection of VAT, corporate income taxes or personal income taxes. With a single taxpayer identification number, tax authorities can actually view these various or can link these various tax submissions. Regular collection or regular electronic collection also means that you can now see how tax collection changes over time. And it responds to certain policies. And the data collected can be used to evaluate the effectiveness of these policy changes. And so now we're back to the cycle of evidence-based policymaking that I spoke about at the, or started off with. So in a unique collaboration with revenue authorities in Uganda and South Africa, anonymized tax data are now being accessed by researchers in a secure way. And the data available can further research on improving tax collection. But as you mentioned in the opening, the data can also be used to shed light on other areas of the economy where data is actually quite scarce. As always, ICT can help improve data collection and revenue collection, but it isn't perfect. And there are a few ways that this can still be improved. Data crosschecks and third party information are important also for improving tax compliance and helping to verify the data collected. More data could be used then again to train risk engines and identify risky taxpayers. And as technology improves, there are technical fixes to ICT systems that weren't available before. And you can even think about more accurate information sort of on the idea that comes to mind, example comes to mind is a location field. So previously, if they could think about something that was manually collected that was hard to read, then we moved to something that was collected and people typed in their address, but they made errors in typing the street name or the area. And now some of these fields can automatically pre-populate as you're typing. And so that's just an example that comes to mind. But importantly, harnessing big data actually requires technical capacity. And this means upskilling staff already in revenue authorities and looking to areas such as data science for lessons on how to improve data collection and capture and processing. In particular, I'm thinking about for research. Where does this leave us? The core function of the revenue authority remains to raise revenue and much of the work that revenue authorities are doing relates to finding ICT solutions to help them collect this revenue. And so I've given three examples of how ICT is assisted revenue authorities in the process of revenue collection and actually how we've used data from revenue authorities to enable these evaluation. So the last thing I want to say is research presented in a collaboration, our collaboration with revenue authorities is actually part of SDG 17, where the goal is to build capacity and support revenue authorities in finding ways to raise revenue. Thanks. Thank you so much, Amina, for setting the scene so nicely and also for outlining some of the future directions that this sort of should be going. Okay, so now we can move on to the discussion by the panelists. Maybe, Amina, if you... Yes, thank you. So just a reminder for those who are joining, maybe a little bit late. So we just heard from Amina Ebrahim, who's a research fellow at Union Wider, and now we move on to the policy panel discussion, and the panel consists of Julia Muskaakni, a research director in the National Center for Tax and Development in the UK, Allen Nasanga, assistant commissioner from the Uganda Revenue Authority and Tim Olaukaren, who's a director at the Finnish Tax Administration. And I just also want to remind everybody that you may already know, if you have feedback questions to Amina or any of the panelists, please feel free to use the chat for writing your feedback or comment. Okay, so moving on to the panel, so let us start with Julia. Julia, can you start kick off by sharing some of your experiences and working as a researcher using administrative tax data and looking at the ICT solutions? Sure, happy to do that. Thanks a lot, Ykka, for the invitation and for considering me to be part of this excellent panel of speakers. I think I will just take, my comments will just follow quite neatly from Amina's presentation because she already said very clearly how ICT and big data can be really important for tax administration. She detailed examples from Tanzania, Zambia, Uganda. So I think I want to follow up directly from that and talk a little bit about the practice and how those things then, what we learn from research on how these things pan out in practice. And of course, the research that we have on this big digitized data set and ICT solutions is that they are often not used to the full potential in the reality of tax administration in lower income countries. And Amina touched upon this a little bit as well in her presentation and I know this comes up from research not only done by myself but also by my colleagues at ICT, Fabrizio Santore is now online, Seidiman is also online from the ICT but also your research from wider shows that quite clearly. In my own experience analyzing administrative data from Ethiopia and Rwanda, what I've seen is that in the data that is already available, there are often quite a few discrepancies. So for example, the same taxpayer reports a different turnover in their income tax declaration as compared to their VAT tax declaration for the same period without there being any obvious explanation for that. That was a result that we got from Ethiopia. But also similarly, we have done some work with the administrative data from Rwanda with my colleague Fabrizio Santoro and Denis Mukama from the Rwanda Revenue Authority. And there we have found that those electronic fiscal devices capturing all the sales in real time, they often produce data that is actually not consistent with what is in taxpayers tax declarations for the same period for the same taxpayer. And I could give you a few more examples like that. So it's clear that those things are not yet fully exploited in practice. I think one of the lessons for me is that it's definitely not possible to parachute data, digitize data, digitization and IT solutions into tax administration and expect them to transform tax administration by themselves. In many ways, data and IT solutions are very much part of a system and that system has many different parts. And I'm sure Alen and Timo will speak to that a little bit. But you know all the traditional functions of revenue authorities, taxpayer services, enforcement, audit departments and so on. Technology and data need to work and interact with those traditional functions in a way that allow for their potential to be tapped. And some of these things again come from research. So there is some research for example from Ecuador but also from other countries showing that using third party information is not necessarily helpful to convince taxpayers to comply if there is no capacity to actually follow up with traditional enforcement. And similarly the communication infrastructure that we now have with SMS or emails and I've done some research on that as well in Rwanda and as you know it has its own limitations if it cannot be paired up with actual enforcement. So imagine you receive these messages from the tax authority about sanctions and audits and your data contains discrepancies but actually nothing happens after that in terms of real enforcement or real interaction. So not only these things can lose effectiveness when they're not working in tandem with traditional functions but they can also potentially affect the credibility of the institution itself. So the reality of tax administration needs to be considered quite carefully. And I could give you other examples around information, sensitization, education campaigns and there's huge potential for technology to help these things. But then again that needs to match the experience of taxpayers as they go through the tax paying process and as they interact with taxpayer services and call centers and so on. So in many ways for me one of the lessons is that IT solutions and data to a certain extent are only as good as the rest of the functions they're supposed to support. They can of course help to improve those functions within revenue administrations but they certainly cannot replace them. They need to work in tandem. And one can make very similar considerations about taxpayers because what I just said was very much focused on tax administrations but when you look at taxpayers as well digitized data, electronic fiscal devices could actually be very helpful for taxpayers to help improve the transparency of their records even give them some form of record that is already digitized by design. But of course depending on the type of taxpayers and their level of engagement with the digital world in general these things might end up being useful or not useful to them actually depending on their reality. So for some taxpayers these IT solutions are premature and if that is the case they might even come with additional compliance costs. So in many ways instead of helping they might actually represent more of a burden especially for those small taxpayers that might not have the necessary equipment or knowledge or skills to make them work properly. And again we do have a fair bit of research on these things. Some of it is specific to data and technology some of it is more general about tax administration tax compliance in low income countries but making sure that these things work together shouldn't really get our hopes down on the potential for data and technology but should really give us sort of guide us in the way that these things can be harnessed to their full potential. There's a lot more that I could talk about but I've been told that I have about seven minutes so I think I'm about up to my time. So I'll leave it there for now. Thank you Julia that was perfect and it was nice you were able to connect your work on voluntary compliance and matching the voluntary compliance and then the necessity also to follow up from the point of view of the tax administration in terms of then sort of a harder real enforcement as well. I'm wondering, I mean if you stop sharing the screen can we see the panellist as a bigger on that screen? You still? Yes thank you. Okay so now we can see more easily. Okay thank you Julia once more and maybe we move on then to hear from Alan and Alan can you share your experiences on how revenue authority can actually utilise research and how do you connect the information communications technology and research. Alan please. Thank you so much Juhal and Tim. Thank you for the invite. Yes picking on from our first Julia. Yes in a way technology enables the authorities to have visibility into who the taxpayers are and so can help us manage their compliance better. Yes it's true that the taxpayers are different levels of use of ICT but there are those taxpayers who once they know that the authority is aware the authority knows that I transacted with so and so the authority has all the data then they automatically come in and comply. However we have different levels of use of ICT for example in the country there are those persons who are very conversant with use of ICT but there are also those the small and medium who still need as an authority we need to sensitise them, we need to hand hold them into using many of the platforms we put out. Yes so you have a point in there it is helpful but because of the way our dynamics use of ICT is in the country even things like internet coverage that speeds the bandwidth away because of some of those factors they are not all at the same level but nonewithstanding technology is key and is helping us a lot as Uganda Revenue Authority for example we sometime back we developed an enterprise data warehouse and what this is we just pick data from different transaction systems within our authority and integrate it to create like a profile about a taxpayer then on top of this data we developed a risk engine which be using particular risk parameters we generate risk calls as well as financial importance calls for particular taxpayers then as an authority these help us focus our compliance actions specific to taxpayers where we are going to get slightly more revenue as well as those who are risking as you can imagine we are now about 3,000 people in Uganda Revenue Authority and the taxpayer population is so big so we use ICT and data analytics really to help us target our efforts to know which taxpayer should we go to which area where should we deploy more but as you said Julia we use the use of ICT is not detached from the other measures we still have enforcement measures and compliance measures that need to move hand in hand nonetheless ICT is very important and without our data there's very little we can do for example it's hard to know who is not registered we have sometimes initiatives like the downtown where they are I don't know how to describe downtown to you but there are many small businesses going on we have some initiatives that people go walking there and try and find who is not registered but over time we've realized that those are known as effective as comparing for example pulling the local authority data and mapping it against our register to identify who is missing partnerships where we collaborate with data we have local authorities like city council authorities they also register the small businesses and give them trading licenses so through a partnership we work with them and ask them to have our tax identification as part of the certificate so we map their dataset to our data warehouse and that is more efficient for example in finding who is known to our register also in terms of who is known to complying who is not paying exactly what they are supposed to do these are data initiatives of mapping our dataset to another organization dataset goes a long way in facilitating us and authority it's interesting but it's possible we have the national social security fund people make submissions there based on their income and it's possible that they submit differently to the social security fund compared to what they submit to URA so a mapping of our data here and there with the national social security and URA we are able to identify particular gaps also it's interesting within the URA we have different tax types as you are aware income taxes and we have VAT it's possible that sometimes somebody declares a different amount in VAT and a different amount in income tax so using our tools, our technology, our e-hub mapping these different datasets enables us realize the gaps and so which taxpayer to engage and rank them based on who is not exactly very truthful within their declaration this has really helped us a long way in trying to drive compliance we also use the data from the custom side we manage the imports and exports mapping the import and exports to what a particular person has declared also goes a long way in helping us but yes the taxpayer community out there is not everyone is abreast with use of technology we are not at the same level so partly as an authority our role is then to sensitize to hand hold them to show them the importance of this technology and there are also cases where the authority and government has gone ahead to like maybe subsidize some of the requirements for them to uptake the technology because you find like we introduced a digital tax stamps but at the beginning the stamps needed to be free of charge for people to take them up we also have now on board an electronic system but the government had looking for the gadgets for all the small persons to be able to comply to use the technology the taxpayers are small and medium people complain about the cost so again we have to come in and find a way of getting them to use the technology without incurring a cost yes Julia you are very right there is a cost to technology which probably we still need to handle this particular challenge either through subsidizing for the small and medium in the short run as they realize the benefits of use of technology technology is a way to go we really can't go back from use of technology and specifically using data the data matching here and there helps us a lot it's interesting but the same company can submit different declarations for one tax type and the other and they are all submitted to you and on the first it's possible to imagine that the officers your staff can easily map them but because of the number of staff vis-à-vis the population it's not possible in a short while to map them but through technology we can easily match this data and see who is not compliant who has not in our system people make a declaration first when you make a declaration you're declaring how your business has performed then after that you come in with a payment it's possible that I make a declaration when I pay much less than what I declared and all this technology that is helping us bridge the gaps one to know who hasn't declared who is not on the register apart from those who have declared have they really paid everything they've declared all those little gaps are really we manage them through use of our data through use of our data warehouse and of course through sensitizing the public out there we try and say that if you comply if you let us know we can go into arrangements like installment payments etc but if you stay out there and you wait for us to find you to enforce on you then it will be more expensive on you and true when the taxpayers are aware that the authority will discover me if they know that it's a matter of time you are able to discover that I have not paid this amount but my submissions are not very truthful then they will comply so in a way that way technology is really at the forefront in your area and it's really helping us manage risk, manage compliance as well as grow our register by mapping to the different local authorities that are also registering the same taxpayers thank you so much thank you so much Alen that all sounds very very impressive and indeed we are at the moment as we speak trying to look at some of the longer impacts of investment in technology in looking at for example the revenue impacts of tax or in Uganda alright now let us move on to the next panelist who is Timo Timo you have worked from the view of a revenue authority in the global north so please say about your experiences in working together with the revenue authorities in the developing countries what role technical assistance can play there over to you Timo thank you and nice to be with you here I've actually got three points on data and revenue authorities and I think the first point really came clearly up in Aminas presentation and also Julius and Alen's points so looking into the not so big data and seeing what's actually in front of your eyes so tax administrations really have lots of data available and it's good to look if you're really doing enough with that data there's a bit of an echo okay I'll try again so and after that when you've looked what data you already have so understanding which data is actually needed after that and trying to get better access to the data that is needed so third party information other government agencies private stakeholders and taking use and sharing that data sometimes it's a legislation issue so trying to influence the legislation if that poses an issue my second point is about customer and ecosystem understanding so using the data to actually understand your customers or taxpayers if you will so both sides of the story to them and their behavior but also the service needs and possibilities to tap into the natural processes of your customers so based on the understanding then trying to make it easy for the customers to comply and hard not to so incorporating taxation into the natural processes of companies or individuals whenever that's possible and aiming towards compliance by design where the customers actually cannot choose whether to pay their taxes or not or at least they have to make conscious decisions and actions not to pay them split payment for VAT would be an example where taxes are transferred to tax administration immediately when the transaction occurs so these kind of changes in the paradigm require understanding of the whole ecosystem around the citizens and companies not just thinking about the taxation all the stakeholders in the ecosystem and it requires active cooperation with public and private parties and I recommend looking at the OECD's tax administration 3.0 that provides useful outlines for this progress and the maturity model that is linked to the models that's a way to actually evaluate your current state and next steps if that seems too far ahead in the future so my third point is is more about the basic systems and processes in place and Julia mentioned the basic capacities needed so all the data in the world is not going to help you if you don't have the capability to process it first of all the analytical capability to make information, knowledge and wisdom out of the data so for example finding the risky customers or transactions and the best suited measures to deal with them but also the capability to use the data as part of the taxation processes so data and analytics need to feed into the taxation processes to actually improve compliance and increase revenue so this might require new systems that are capable of higher interoperability silo legacy systems are sometimes a problem might not sufficiently support this and sometimes tax administrations lack the needed in-house capacity to build the needed integrated systems from scratch so sometimes the best solution might be to look at what's available in the market because there are already made commercial off the shelf integrated taxation systems out there but in addition to the systems it usually also requires redesigning the business processes to best take advantage of the data and Julia touched upon this as well so if you just bring in new data the processes might suffocate and struggle to understand the new data and what to do with it or simply ignore the new data and continue as before and this better data new business processes also require new skills from the officers and citizens which in turn requires training and much more like Alan mentioned the hand holding part so even the best systems and the most wisely designed processes fail if people don't know what to do or don't want to act in a new way so change management is essential to any success and it might mean much more than just training people need to understand the reasons for changes and what's in it for them and they need constant support the one example that Amina mentioned of what we've done together with Tanzania Revenue Authority and UNU wider so the basic principle in technical assistance and capacity building is that you do no harm which is why we wanted to measure the effect of our intervention so not just how many meetings or trainings we've had or measuring something that we don't really have influence on so we designed a new pilot process for tax control for examination in Tanzania and also the support model to ensure the change so it was a bit more than just an Excel sheet and risk profiling it did include the risk profiling but also it started from the new instructions to officers and a totally new process flow so and with the help of the UNU wider team then we were able to measure the outcomes of the new process and the effects on revenue this kind of work has created lots of interest in also other tax administrations doing capacity building since measuring the outcomes is difficult and usually it's not done people are not able to do that so the point here is to also get data about your processes and their outcomes so you can make wiser decisions so three points looking into the not so big data getting customer and ecosystem understanding and having your basic systems and processes in place, thank you excellent, thank you so much Timo so indeed so this requires quite a thorough change sometimes I mean it's not always enough to have the technical capacity somehow the organization also needs to be functional and be available to benefit from the improvements thank you once again for panelists is there among the panelists or Amina any additional thoughts that may have now come up when you listen to the other panelists before we move on to the Q&A no direct need for intervention at this point so everybody happy so far so good, excellent we have some questions from the audience we have, do we still have Mr Kwami from Toko present are you online still would you like to ask your questions, unmute yourself and ask your question okay, thank you my question is to Amina I want to know how the regulation framework on the privacy as we are talking about data when we collect data from someone can use it maybe not to the objective you declare first so how is the regulation going how people or firms can be confident to use to boost revenue collection, thank you thank you, that's an excellent point and needs to be addressed Amina would you like to take that question sure, I think that's a really important point and thank you for raising it so my first comment would be that actually it's not the same in every country there is a lot of data regulation and protection laws and in each country that we work with we work with the local partners that then inform us of what the regulation is and how to interact with that I think our model specifically in terms of research and when we work with revenue authorities is to make sure that we work with anonymized data and the point really for research is not to identify it doesn't benefit us in any way we end up looking at more aggregate estimates of what's going on so anonymization being the basic methodology some tax authorities revenue authorities ensure that data is only accessed in a specific way so it's never accessed online some revenue authorities require to sign disclosure agreements and oaths of secrecy so there's several ways that they ensure that taxpayer identification is not leaked in the research process and even then once all the research is done often the results are checked to make sure that even by mistake nobody is being identified and so I think that's an important part of the process and I think it's taken some time to get there there are lots of ways in which revenue authorities also need to understand and respect some of it is personal information and not share those specific things so widely so there was actually I see that in one of the messages our colleague from SAAR South Africa Revenue Service Lillian asks how what are the implications of all this for sharing information across government institutions so maybe Amina can you perhaps start commenting on that and also maybe Alen how do you see the information flow between the various parts of the government and are there risks there as well from the privacy point of view maybe we can take a page out of the book from the Nordics on how this is collected so I think it takes a lot of effort to start working between different departments and managing to get data so you find that different departments collect different data in different ways and then they anonymize data in different ways so then it becomes very difficult to put those data sets together and at the minimum I think there should be a conversation about how they could be done and there are some efforts at least I know in South Africa the Department of Labor to look at unemployment insurance fund information where identification numbers can be anonymized in the same way and starting to match into the data but to go back to Kwame's point about the more data you start adding together maybe the increased level of sensitivity of the data and so you need to think quite hard about what are the regulations and how you want to maintain the sensitivity around the data way possible Ellen, do you have any other thoughts? Yeah, thank you so much it would just be sharing our case study in Uganda there are efforts as government to share data but they are not moving as fast as we would have loved simply because as government we are at different levels of automation so there are some areas that are so advanced and others are slightly lagging behind we currently share data for purposes of completing a single transaction for example within the URA we have a new way of registering taxpayers and they get their team instantly and this we do by sharing data with a registration authority with a person's registration authority NERA is called NERA in Uganda now what happens is that when a taxpayer comes to register as an individual we just ask them for their national registration ID number and then when we pick that ID number we already just confirm their details from the other agency and then we can issue them a registration a team registration very fast but in terms of sharing data analysis purposes as in you give us your database and we map on our database that happens on a case by case and we still have to go through issues of then we sign an MOU we have the same regulation for data protection but still every agency if I may say is still inward looking they collect data thinking about only their purpose and then when it comes to sharing we still although we are all government then we say you need to sign the MOU because of data protection and you would imagine that we should just share together so they still progress but I must say it's a bit slow thank you so much Thank you this may sound like the technicality of sharing data between various arms of the government but in fact it's related to one of the core messages in tax administration research which is that the possibility to use third party information requiring the tax pay and more easily this third party information flows in the better the tax administration will be and it keep also tax avoidance evasion okay so we still have some luckily some time this is a very short webinar only one hour so I'm certain that we could continue for quite some time but I see Laban has raised his hand Laban are you following? Yes Yuka thank you interesting composition from all the discussions very important points but when you hear from all of the presenters all four of them you can see an underlying issue with change when people do not follow the technology change that's happening one is bound to face problems and that's what we are beginning to see in many tax authorities especially in our part of the world so how can we ensure that the technological change we are seeking to make more efficient our tax administrations is matched by adaptation in our skills this point was raised earlier technology is only as good as the skills it supports and the persons it supports the processes it supports often times there has been under investment in the other components of an effective or efficient tax system much has been focused on investing in the technology and we are beginning to see the cost of that so it's a question to provoke thinking but it's something we've seen we've observed even here in Zambia how to ensure that our staff our taxpayers are carried along with the changes that are happening thank you Yuka thank you that's an excellent question Laban maybe can you still come back and introduce yourself for those who don't know you I'm from the Zambia Revenue Authority I've been a project manager for a data analytics project all of the discussion today is very relevant to us especially on that project but I also work with researchers in the department statistics in general and everything to do with tax Yuka I hope I've been confused present as thank you like all of us of course yes now back to the panelists would you like to perhaps respond to Laban Julia anything from your side or for the other things that we have touched on before we close I can just make a couple of remarks that really pick up from what Alan and Timo also said before which I found really interesting and one thing is how to bring citizens along there's definitely a big role for sensitization and training and taxpayer education and Alan has been talking about that I know the URA is very active there and other revenue authorities as well so it's important to build skills but also to build awareness of the importance of taxes more generally and the approach to tax collection and on the tax administration side to build those skills some basic functions but also make sure that this approach of customer orientation really filters through the whole organization down to the frontline workers because that is what tax payers see on their side they might not read the flagship reports but they will see how tax officials they deal with act and I think doing that will help build the credibility of tax administrations which is already very high in many contexts and trust with citizens who ultimately need to want to comply because there's no way any revenue authority in any country can actually check every single taxpayer so that element of credibility and trust is quite key in my view whether or not we speak about data and technology I think those considerations are probably broader Thank you Julia once more and still maybe we have around time because we started a little bit late so maybe we have one or two more extra minutes and Timo as well on these newer questions Alan over to you Thank you so much The role of technology and data analytics can't be underestimated in modern revenue administration Yes there are challenges of skills the staff skills as well as the citizenry so change management needs to is also key we need to sensitize we need to upgrade the skills, hand hold and help both staff and the public take up the different technologies Of course as the more data we use third party information can all help us into realizing any gaps within maybe compliance risks which can then be closed Yes the data or the results of the data that BI needs to fit into the work processes the workflow so that for example where you do some analysis it needs to go into the process, trigger some activity to a particular staff to take action and that way we can improve Thank you so much Thank you again Timo, any final thoughts? Managing the change part is one of my favorite topics so I could go on a bit here but I'll keep it short it's just like take a systematic approach and recognize that it actually needs resources it's in an ICT project so it's the same thing as testing you need people to do testing you do need people to do change management it's not something that you can do as an afterthought it needs to be planned systematically and the actions need to be taken to to actually get something done and ready Thank you, that's an excellent point where we now need to end Unfortunately our time is up Thank you once again so much All the panelists for joining Thank you for the audience for the questions Thank you Amina for the excellent presentation in the beginning and while we have all the revenue authorities and researchers have done a lot of advances in terms of using all this new excitement did develop there's quite a bit of still work ahead of us so stay tuned and follow up the new wider work in this area going forward as well Thank you once again and bye bye Thank you, bye Thank you, thank you so much