How to build a new social safety net in two weeks

The Innovation Dividend Podcast, EP 2

The Innovation Dividend explores how innovation in society and government are paying off. Over the coming weeks, we’ll be exploring how ‘policy frontliners’ are innovating in real time in the COVID-19 pandemic and asking which of these changes and “raw learnings” might become part of our new normal. You can see the backstory by Kal Joffres here, and the Podcast EP 1 here.

In this episode, we interview Anir Chowdhury, who is at the frontlines of Bangladesh’s COVID-19 response. He tells us how government gets things done fast, how some pre-COVID investments by government are turning out to be a lifeline for deploying national programmes, and how the Bangladesh government is assembling a safety net for 5 million people in two weeks.

• You can subscribe to the podcast on Spotify or Apple.

[00:00:00] Anir Chowdhury: The short-term needs to happen right now within like days. And the long-term has to happen within months. I mean, long-term now means months. It does not mean a year.

[00:00:13] Kal Joffres: Hi and welcome to The Innovation Dividend, the podcast that explores how innovation in society and government are paying off. I’m your host Kal Joffres and over the next few weeks we’ll be exploring how the COVID-19 pandemic is unleashing some interesting and unexpected sources of innovation across Asia.

[00:00:29] We’ve been following Anir Chowdhury. He’s at the front lines of the COVID-19 response with the government of Bangladesh, and for the last 13 years he’s been leading the formation of an innovation ecosystem in Bangladesh, institutional reform, and a service innovation fund. He’s also a member of the Prime Minister’s National Digital Task Force.

[00:00:48] Anir, welcome back!

[00:00:49] Anir Chowdhury: Kal, thank you very much for having me again.

[00:00:51] Kal Joffres: How has the situation evolved over the past week? What have been some of the new developments that you’ve been dealing with.

[00:00:57] Anir Chowdhury: Things have gone south as we had anticipated, not in a very happy way obviously. If you look at the health side, the number of infected went from a bit over 300 to a bit over 2,100 — so seven times increase in a week. The number of deaths went from about 21 to 84 — so four times, and this will rapidly rise. That was what we had been discussing with our Ministry of Health for a long time. But I think, what was happening is that we were, in the wishful thinking zone that, because of certain expectations that heat may be a COVID killer. Also, most of the countries actually had the BCG vaccine, which the tuberculosis vaccines about 80% of the country.

[00:01:44] Response relation and had that vaccine. So there is a discussion, I think from many countries that BCG may protect us, but obviously the number keeps rising.

[00:01:55] The government has announced subsidies, a new form of social safety net for the short-term, to be delivered in the next, one week or so for about 5 million people. So these 5 million people have never been, in the social safety net. So they are not classified as poor or hardcore poor.

[00:02:16] They’re classified as lower middle class, and even some people belong to the middle class. So we’re seeing a new cadre of people who will need subsidy from the government.

[00:02:28] It happened in a few hours. So the government has announced subsidy for about 5 million people. The safety nets have to be done in a certain way. We cannot leave out people 333 trainees to be repurposed again to make sure nobody’s left behind, they’re able to call into that system and, represent their needs.

[00:02:50] The private sector has also joined hands with the government. NGOs have joined hands, and they’re also doing their own way of providing subsidies.

Setting up an architecture of solutions through cross-sector collaboration

[00:02:58] Kal Joffres: Could you give us a sense of what was the timeline for that? How long has this been in the works and in and how much time is it going to be deployed?

[00:03:06] Anir Chowdhury: So this has been in the works for about a week or so. So the Prime Minister, after a lot of economic analysis, decided that, the new needy must be helped, and how to find them is the biggest challenge. How to get the subsidies to them is the next challenge. So all those things are being worked out as we speak.

[00:03:26] So teams of companies, data scientists, the government, officials are working together to make this happen. So it’s a combination of ministries that would take weeks to set up a meeting are taking maybe hours to set up a meeting. So as soon as this was declared by the Prime Minister, last week, and then there was a next announcement yesterday with the Prime Minister, we met this morning, within a few hours.

[00:03:51] So then the announcement and we worked out, a simple technology: a couple of the field officials already tested out parts of the technology in the last one week, without even involving technology companies. So they did their own thing and we getting a lot of very interesting proofs of concept from these trials.

[00:04:10] And these proofs of concept are really informing us to design the short-term architecture. And the long-term architecture is also being informed by these decisions. So how do we get from point A to point B? It’s really a war situation right now. So we’re at war footing. There is no time to do procurement.

[00:04:28] There is no time to do a long-term design where the architecture would actually, would stand the test of time. Whether it could actually take calls from a million people, if they call it the same time, we don’t know, but we have to just move ahead. We need to distribute, food. We need to distribute cash.

[00:04:48] And if we cannot come up with the technology in time, it will be done regardless. It’ll be done in an analog way, perhaps much less effectively, much less safely. And it may take much longer and we’ll be making a lot more mistakes and we’ll probably be leaving out a lot of people. So the combination of technologies would allow us to be safe and would allow us to be inclusive.

[00:05:10] Just to give you one example, how inclusive we are trying to make this is that the 333 number I mentioned was repurposed, to have people call into it and said that they need food and they need a cash subsidy. And we verified that. If we did not have that number, the 12,000 or so people that we helped out in the last one week, we would not have been able to help because we would not know who they are.

[00:05:37] So this 12,000 is just a small number because we didn’t, we were just testing the system. Now that we have tested it, it seems to be working well. We’ll scale that up in the next one week. So we’ll probably be getting hundreds of thousands of calls and we’ll be finding maybe tens of thousands of people there who could not be found through the normal listing process because they probably have been living, these families probably haven’t been living for the last few years in the capital city of Dhaka.

[00:06:04] Now they have moved back to their village home because they cannot afford living in Dhaka because they have no income. But they are now calling into 333 and saying, we need help.

[00:06:13] So maybe the neighbors don’t know them. Maybe the, the listing process for the social safety net, will not find them, but they are there. They’re coming up spontaneously and letting us know that they need help.

[00:06:26] From a technology standpoint, we’re trying to help the analog process that it existed for a long time because we know that it would go ahead without technology if we cannot come up with the help at the right time. So time is of the essence, not the technology. Helping people is of the essence, not the technology. Technology is trying to catch up with what needs to be done, what must be done.

[00:06:47] Another thing that has happened, interestingly, is that the garment workers, of which we have close to 4 million, used to get their salaries, most of them in cash from the factories.

[00:06:57] Obviously, some have been, some of this salary distribution has been converted to electronic money. So essentially mobile financial services, because we have a fairly rich infrastructure of mobile financial services, but not all. In one week. In the last week, a 1.9 million accounts have been created in the mobile financial services system, based on national ID, based on birth registration for these garment workers.

[00:07:24] Because we made a decision nationally that transferring cash payment as salaries after these workers is not only time-consuming, expensive, but also, unsafe because as we know that the virus can actually stay on the paper note for a few days. So now these 1.9 million workers will get their salaries.

[00:07:50] So mobile financial services through basically mobile wallets and previously there were about maybe a million plus that had that system. So we’ve been discussing it for maybe two years, and we got about a million or so, a million plus who have been converted to cash from cash to electronic living. Now in one fell swoop in one week, another close to 2 million people converted to electronic cash. So that was a positive thing but are born out of necessity.

[00:08:19] Kal Joffres: You know, in the United States, we hear that there’s no national registry of ventilators. At the national level, there’s no awareness about where ventilators are in different parts of the country. What is the data infrastructure for Bangladesh right now and how are you filling that in?

[00:08:37] Anir Chowdhury: It’s a very good question. I mean, you compare with the United States. Obviously, you know, the States has a much better data infrastructure, but, the advantage that we have is that, we have a monolithic, data infrastructure within the health ecosystem. So what is within the government is actually in one data repository.

[00:09:00] What is outside of the government is a different matter, and I’ll come to that. So within the government, that data is kept, pretty well as far as the government health institutions are concerned. So we have at the rural level about 14,000 community clinics. We have in each sub-district level about 492 at the moment hospitals, and we have tertiary installations at the district level. A few hundred of them, I don’t remember the exact number.

[00:09:30] So this is the government facilities. And data is kept fairly well because I would say that in terms of use of technology, health system within the government is probably the pioneer in that.

[00:09:43] They’ve been using information technology for over a decade, and they’ve done a really good job in terms of keeping data intact and updated. Now, if you look at the private sector, it’s all fragmented. Some do really well and some don’t in terms of keeping data, but it’s completely not integrated with the government.

[00:10:03] So you have a big Island — the government — and you have many small islands within the private sector, and we don’t know how data will be integrated. We have probably twice as many doctors in the private sector as in the government. So it’s crucial that we also try to integrate with the private sector ecosystem of which we know very little.

[00:10:25] In terms of non-healthcare is something I also want to highlight is that a food distribution and a subsidy distribution, that’s become a big issue in the last one week. It was an issue even before that, but it’s become a critical issue in the last one week because people don’t have jobs.

[00:10:45] They don’t have income, especially the people who used to live on a daily income and now the new poor who are seasonal workers in the agriculture, economies, in the agri-economy. So we have the new, new needy. If you may want to use that word, the new poor. They may be temporarily poor. They may be poor for a while. We don’t know. We just don’t know right now.

[00:11:10] So what is the database of that? And that’s where we’re really struggling. So we have been trying to develop a national register of people. So it’s a citizens register, basically citizens’ civil registry, and it’s been going on for a while. We have different databases, all electronic.

[00:11:26] So we have a database of national ID, which is quite robust and dependable, but that only covers people above the age of 18, which is, potentially about a hundred plus million people. But then we have another, 70 million people to account for. Now they exist in the birth registration database, but birth registration database is not very dependable because you can do duplicates, duplicate entries because there is no biometric.

[00:11:55] National ID is the only biomedically verifiable database and then there are education databases. So we have probably half a dozen or so electronic databases that are not linked, and then there is the household survey that was conducted recently. But again, the data still remains on paper because it was collected on paper and has not been entered in a digital database.

[00:12:18] So now we’re at a juncture. Well, we have all these databases, but they’re not linked. And, we don’t know how to figure out who the needy are. So there’d be people who want to take advantage of this opportunity and who’ll come to the location where relief has been distributed, or will ask for a food and a cash subsidy multiple times, and then there’ll be people who will not be counted, who completely remain out of this whole ecosystem and will be completely deprived.

[00:12:50] They’ll be deprived because we cannot reach them. They’ll be deprived because they feel hesitant to ask for subsidy, but they need that subsidy and we must give them that subsidy. So in terms of resource allocation, so to just to extend the question that you have on the healthcare side, I wanted to also bring up, bring out this very important issue of national register. Countries that have developed a robust and updated national register with wealth information with some income information.

[00:13:19] I think I probably in a position to respond much better as far as subsidies are concerned. We’re struggling right now. So just this morning we had a meeting with the Minister of the Ministry of Disaster Management and Relief and the Ministry of ICT.

[00:13:33] So we had a meeting with the two ministers and we’re trying to figure out exactly how to take a short-term measure and how to take a long-term measure. So this crisis has really brought the sense of urgency up. It jumps off taking the long-term measure as well. So short-term measure, we have to figure out a way of sort of doing a band-aid job of cobbling together these two, these many databases, but then the long-term will be to have an integrated approach of developing databases for national registry of people, but also other things as well.

[00:14:08] Kal Joffres: And so are you implementing both of those at the same time? What happens to the long-term track of work?

[00:14:14] Anir Chowdhury: Yes. We decided this morning to implement both of those at the same time. So there is the short-term SWAT team, so a few companies that are very proficient in terms of coming up with a very quick solutions. They won’t be looking at long-term architecture, so they’ll just give very localized solutions.

[00:14:32] And we’ll help the field the administration who are completely panicking in terms of relief distribution. So we have a fairly large field administration team across the entire country. So we have head of district, head of sub-districts of 64 districts, 490 sub-districts, and then 4,500 what are called a union councils, which are at the lowest level of the government.

[00:14:55] So all of these will be activated at the same time. So this entire chain of command will be activated at the same time, and they have to distribute cash. They have to distribute food at the same time. So they need a short-term solution. And we are using, different combinations of technology from just Excel to Excel sheets to dashboards.

[00:15:15] We’re repurposing our 333, which I talked about last week of 333 is a national hotline where people would call for getting government service information. Not service, but information. So where do I go to get the passport? How much does it cost? What documents do I need? That kind of thing.

[00:15:32] So there are, this, this has been an operation for the last two years and this has helped the people in many ways. Now we repurposed that. So a couple of weeks ago, we repurposed that for health. So this has become the largest telemedicine system in the country with 4,000 doctors connected to it.

[00:15:49] Now last week, we repurposed it for food distribution. So now people needing food or needing a subsidy may call in to that and we verify that information and then forward that information to the local administrator and the local administrator then takes action. So that’s been moderately successful.

[00:16:09] We have not announced it in a big way because we’re still testing out the system. So we have seen that in a week or so. We have gotten over 50,000 calls and, about 12 or 13,000 people have received a subsidy.

[00:16:23] The short-term needs to happen right now within like days. And the long-term has to happen within months. I mean, long-term now means months. It does not mean a year. And we are completely bypassing the procurement process.

[00:16:35] Kal Joffres: You’re bypassing it by purposing existing contracts and assets.

[00:16:40] Anir Chowdhury: The two things that keep driving all the work, all my action right now: one is repurposing. What can we repurpose? That’s what we do in our personal life when there is a crisis, when we cannot get to something that we need, but we are not able to get through to the point.

[00:16:57] So we repurpose something. We repurpose every object in our household to fixing a crisis, to doing a band-aid. So that’s what we’re trying to do but the longterm is also important. So, we are looking at the longterm. So repurposing does not go through any kind of procurement but also the company that are coming together right now.

[00:17:19] We’re bringing them to some kind of a collaborative framework. How we legalize this collaborative framework in terms of public procurement, is a question that I’m able to answer in future in a future discussion. We haven’t figured that out yet. Some actually are under that type of contracts.

[00:17:37] We may repurpose those contracts. We may use those contracts for something totally different and that will probably be legalized and we can legalize that but some of these companies that are working with us have no contracts. So assigning NDAs just to do a data protection, but we may actually provide contract with them in the future and how we’ll do that move.

[00:17:59] We’ll figure that out. Some companies may not ever ask for contractors. They’re just doing this as a contribution to the national mission that put place.

Cross-sector information management

[00:18:08] Kal Joffres: For a lot of this to happen, information sharing between agencies, between ministries probably has to change. Could you give me a few examples of how information sharing has changed during the epidemic.

[00:18:23] Anir Chowdhury: Information sharing between health and non-health institution — I’ll start with that. Health information was supposedly guarded for privacy reasons and also because of the silo-centric thinking that we have within the government. But now we are getting information from telcos. We are getting information from health.

[00:18:41] We’re getting information from a number of technology companies. We are now going to get information from social safety nets. We’re going to get information from e-commerce. we’re going to get information from Finance Department. So all those different information on coming together in some form.

[00:18:59] Not that all of them will immediately start on each other, but they will at some point. So that crisis response started with information sharing across health but now the crisis response has become economic crisis. So safety nets, different kinds of income generation activities, potentially education, how we will reopen education.

[00:19:21] So all those will come in. So this is one example of where people are not thinking about their egos and their departmental boundaries. And sometimes we’re probably making some mistakes in terms of data privacy. But those we’ll figure out. But the good thing is that we are now working together, at least in the digital form.

Evolving work of government

[00:19:42] Kal Joffres: So the work of government has really, pushed into overdrive and while you’re essentially designing new public services, coordinating and in very different ways, you also have this constraint of social distancing. How does the day to day business of government look like today, what does a day in your life look like?

[00:20:04] Anir Chowdhury: There was an article I was reading from China, which said that I used to work 9–9/6 which meant I used to work from nine to nine six days a week, and now I work 0–0/7 which means that I work midnight to midnight, seven days a week. So that’s my personal day has become like. So there is no boundary between work and life.

[00:20:31] It’s one seamless continuum, it feels. So there is no travel to an office. No need for a haggling with the traffic because you’re not moving. So a lot of people within the government have found themselves in that position people who are able to use the video conferencing facilities, email, social media to communicate and coordinate and collaborate.

[00:21:05] So that’s the plight of people who can maneuver that digital ecosystem to get work done and still get work done with other people. But then there is a lot of people who feel completely left out because they may use phones for calling people but again calling is not enough.

[00:21:23] You need to be able to converse and share documents and sign things and collaborate and draw things on the whiteboard. So they are finding this extremely difficult, extremely challenging. We have put in place a decision support system that’s been going on for the last several years, four or five years now, that is gaining in popularity now.

[00:21:47] So that program used to be in 8,000 offices, only partially used. But it’s being more and more used by a lot of the government officials and the target was to go to 19,000 offices by the end of next year. That may actually happen earlier because they are depending on that electronic decision support system to move decisions, move files.

[00:22:10] There is an electronic signature, embedded in it. So a lot of decisions are getting done that way. So we are contemplating that there must be some kind of capacity building for accepting and adapting to this digital lifestyle, for government officials. And we are trying to design that.

[00:22:27] We have not done it yet. So people who feel left out, who feel challenged in this digital environment to getting things done, we must bring them up to speed. That is an essential thing.

[00:22:38] Kal Joffres: How has the day to day work of government changed? I mean, it sounds like there’s a lot of interdepartmental, inter-ministry work that’s happening. You mentioned people working in teams. Can you tell me a little bit about that?

[00:22:52] Anir Chowdhury: So I’ll give you an example from just this morning. We made decisions of how things would be done. We brought in people from the field administration because they didn’t have to travel to Dhaka.

[00:23:03] They could participate from wherever they were. They could provide their field level experience that this is going to work and this is not going to work. This is the technology that will help us. This is the technology that will not help us so don’t try that right now. In the short-term, the minister’s made the decisions.

[00:23:18] We made a very quick, analysis of the situation. We did a quick PowerPoint, a presentation that was developed a half an hour before the meeting, within probably an hour before the meeting. And decisions were done, minutes will be developed for that meeting within the next few hours. And the government group will be probably done by the end of the day.

[00:23:41] So this is an example of speed that can be harnessed. And this is something that I hope not in the feverish pitch. But I hope that that continues the sense of urgency, this sense to help people, this sense to bring people together. This has to collaborate and make decisions across ministries, which also is a difficult thing sometimes in our silo centric world.

[00:24:04] Kal Joffres: What do you think we would need to do to keep some of this, the good parts of this new way of working post-COVID?

[00:24:11] Anir Chowdhury: I think the first thing that could drive this kind of collaboration and speed is the intense feeling that we must be citizen-centric, that we must be driven by citizens needs. And that’s what we’re seeing when we do this kind of meeting, that we must respond. We must be able to help.

[00:24:32] We must be able to be there when the citizens need us. So I think that’s the first thing if we can take this to our heart. I’m told that it takes about 21 days to form habits. So I’m sure we’ll have 21 days of this.

[00:24:45] So the people who have done this for 21 days, I think they’ve formed a habit. So that’s one thing. So citizen centricity, citizen centric thinking, how we integrate, how we collaborate, how we act together, and not in silos. The second thing is I think, thinking of, and this is a fabric from a technology standpoint, think of the government as a platform.

[00:25:07] So not thinking of government as buildings or databases or different types of mental and physical silos. So think of government as a cloud. So we use a different type of Google services. We use Google search, we use Gmail, we use Google Assistant, we use Google Docs.

[00:25:29] So we use all kinds of Google services more and more sometimes without even realizing that we’re using Google services. So Google is a platform and there is a device that connects us to the platform. We don’t even know what the different groups within Google that have created this or different technologies that Google has bought to make this happen.

[00:25:51] So there is a seamless process that brings all these different services together into one platform. So that’s what the citizens need though. If we can turn the whole government into a platform, connect all the different pieces, and they can remain in multiple buildings, they can remain in multiple silos.

[00:26:10] But as long as we can connect them digitally under one framework, a common protocol perhaps that would sustain this work. And I think it is possible. I’m seeing a light at the end of the tunnel after having done this for many years now that this urgency, this need to be citizen-centric, this I think demonstration of possibilities is what is going to make this happen.

[00:26:34] And the third thing is collaboration with the private sector. And this is still, something that we’re not doing as well as we could, but more than before. the government has always felt that the private sector is not as pervasive as the government obviously, other than maybe a couple of large end users like BRAC.

[00:26:54] And a couple of large, business conglomerates like the ones who use pharmaceutical distribution and who do fast moving consumer goods. Other than those, most businesses are assigned those, but everybody has a role to play in this whole crisis response. So if we can, again, create a framework for collaboration, and again, that’s not technology.

[00:27:16] That’s really how we think and how we act. If we can create a framework for collaboration that would also, work really well for us. So those three things. The first one is a citizen centric view. The second one is thinking of government as a platform. And third is a collaboration framework. Government and non-government institutions, and perhaps looking at all service providers as a platform. That’s what citizens need. That’s what they, that’s what they want.

Dealing with backlash against public health measures

[00:27:44] Kal Joffres: In the US we’re starting to see protests come up around the lockdown and of course you’ve mentioned earlier in our conversation, the garment factories wanting to get their workers back online. How is government keeping its ear to the ground about popular sentiment? And how are you now thinking about how to deal with the potential backlash against some of the public health measures.

[00:28:11] Anir Chowdhury: It’s a very timely question, Kal. So I’ll probably answer it in two different ways, but the answer may not be complete, but I’ll share what I have right now. So one resistance is from the people obviously. They’re not coming out in the streets and doing public protests yet. because the government has made a timely decision to provide social safety net.

[00:28:31] It may not be a complete replacement of the lost income, but it will be something that we’ll provide food and provide some cash to the hungry and the needy families, the new needy, as I call it. They were not needy before because they had regular income, even if that income was on a daily basis, but they had it, now they don’t.

[00:28:52] So we’re not seeing public protests on that because the government has taken a timely measure on that. Now, how long this can be continued with limited resources that the government has, would potentially not a lot of external help because every country is facing this. So we don’t expect a lot of help from big donors coming in right now.

[00:29:14] So we have to fend for ourselves. So we are sort of on our own right now. So how long this will continue, we don’t know. How long this a social safety net must be continued, we don’t know. If it’s for many months, then we’ll be at a problem. We also have buffer stock and buffer harvest, which needs to be brought to home. So that’s in the field.

[00:29:33] We have to figure out maintaining the safety measures — safety, health measures, so how we do that. But that will be respite for the hungry. There is another kind of backlash that’s coming from the industry, specifically from the manufacturing industry, and the SMEs, the small and micro enterprises, because they are not generating income for themselves.

[00:29:52] So these enterprises, small and large, they are not getting the subsidy that the individual needy are getting. The government has announced about $9 billion dollars in terms of loans and some small grants also, but mostly loans for the industries, for the manufacturing industry and for the small and medium enterprises.

[00:30:12] How this will be handled, how this will be actually doled out, who gets what share and who benefits and who doesn’t. That’s still to be worked out. So for the time being, both backlashes are contained, both backlashes are being addressed by the government. But again, the duration of this crisis and the continuity of these subsidies for the individuals and loans for the industry, we’ll have to see.

[00:30:37] Kal Joffres: Anir, thank you so much for joining us again this week and really looking forward to speaking with you again next week to understand things on the ground have changed, and what are some of the new developments and innovations that you’re working on.

[00:30:51] Anir Chowdhury: Great. Thank you very much, Kal, for taking the time again to have me on this program. Thank you.

[00:30:57] Kal Joffres: That was Anir Chowdhury. He’s a policy advisor with UNDP Bangladesh and the government of Bangladesh as well as a member of the Prime Minister’s National Digital Task Force. We’ll be following Anir over the next couple of weeks to understand what the situation is, how government services are evolving, and what are some of the different challenges and opportunities around that.

[00:31:18] So stay tuned and we’ll be releasing podcasts regularly on this topic.

The Innovation Dividend explores how innovation in society and government are paying off. Over the coming weeks, we’ll be exploring how ‘policy frontliners’ are innovating in real time in the COVID-19 pandemic and asking which of these changes and “raw learnings” might become part of our new normal.

You can subscribe to the podcast on Sound Cloud here ,Spotify here or Apple here.