A Blockchain for Farmers and Co-ops
An interview with Maomao Hu, Co-founder and COO of financial inclusion startup Kora Network
This interview is the first in a series of articles on the intersection of blockchain and the *real* sharing economy. By “sharing economy”, I do not mean the Uber/ AirBnb sharing economy which is nothing more than a euphemism for Captialism’s newest strategy for generating monopolies and extracting wealth & disempowering those who create real value. The *real* sharing economy is one in which is open, regenerative, equitable, and seeks to empower those left behind by the existing economic paradigm.
This was originally published in an abridged form as a Q&A on Shareable with a creative commons license and is reprinted with permission. Since Shareable readers and Medium readers constitute different audiences, I will be posting the full-length interviews for anyone who wants more information, including the slightly more technical questions I may ask.
From farmers in Nigeria to coffee producers in Peru, Kora, a startup with a global team, is already on the ground, working with groups across the globe to provide financial services, via the blockchain, to those who stand to benefit immensely from this technology. Approximately two billion adults are still unbanked and millions more pay high interest and fees to access basic financial services, but mobile technology and blockchain both offer a way to serve the unbanked and underbanked.
In order to maximize the ability for underserved populations to reap the benefits of blockchain, Kora enables users to interact with this technology using smartphones, basic cell phones and text messages — not just desktop computers. We spoke with Maomao Hu, co-founder and chief operating officer of Kora, who told us a little about the company’s work and about how the blockchain can significantly increase rates of financial inclusion by lowering costs, providing better data, and increasing transparency and speed. Armed with a background that ranges from graphic design, finance, and artificial intelligence, and influenced by the Occupy Movement, Hu sees the co-op as a crucial organizational structure, and one that stands to benefit immensely from recent developments in blockchain technology.
Aaron Fernando, Shareable: I want to focus on how you’re driving user adoption among those who might not be crypto-literate and also how using a blockchain addresses that last mile problem (getting technology down to the end user) for financial services. Can you talk a little bit about how Kora has focused on designing a platform that can be as accessible as possible?
Maomao Hu, Kora: Yeah, I feel like that’s an issue we get pretty often, like how to get more people on the blockchain. I mean our philosophy is not thinking about blockchain. We just think about what users need, and then blockchain just happens to be a nice way to get there. Blockchains are pretty well designed for financial services. And so, our platform, what we’re doing is just financial services for underserved areas.
One really important thing we are focusing on right now is payments. And how do we take payments into places like Nigeria or Ghana or Bangladesh? Well, we look at what people like — , which is usually either cash or mobile wallets or bank accounts (if they have one) and we figure, ok, how do we get there? We either partner with a bank, or with a mobile wallet provider, or figure out how to get to the last mile mom and pop shops and it’s like, basically we just work backwards, same as you would for any business.
In this particular case, blockchain happens to be quite useful because it’s a very useful way of transferring money — especially in underserved corridors, we can get a better conversion rate a lot of the times, and if you just went through traditional FX markets or a bank. Blockchains are also a nice way for us to verify identity. Because, if we were to just take a central identity provider, there’s all kinds of risks associated with that. Not to mention, how do we know that we’re 100% accurate, with the information we’re storing? Whereas if we were to set up a blockchain, then the liability on our end, as well as any risk to the user from our end just goes away, unless basically the blockchain’s been compromised. And so it’s just really, really cool.
Could you talk a little bit about what is eFiat and how you’re using that to make your services more accessible to people?
eFiat is just an accounting for fiat currencies on the blockchain. That’s all it really is. So we’re going through pretty conventional channels to get the money to people, usually. It’s just an accounting tool for us to keep track of how much money there is and how much there is.
It’s almost like an accounting tool, for us to keep track of where the money is. It’s not something like Tether, where we’re releasing it and it’s being used as part of crypto exchanges, which is even at this point like a replacement for USD in most places. This is very much focused on our own value chain, if that makes sense.
When you’re doing USSD messages, are you working directly with the service providers in, wherever — Nigeria, to start?
We don’t need to always, because sometimes the telecoms locally will provide API’s. In Nigeria, we actually work with two parties, neither one is telecom. One is an [INTO], so there’s licensed provider that has an agency banking license as well as a USSD license. So, you know. And they got that USSD use license from telecom. So we work with them and they work with telecom. Our backup, there, in case their connectivity goes down is an API aggregator. So we would call an activist talking. They went out and they built out a bunch of endpoints that actually connect to all the — there’s like 30 telecoms in Nigeria. They connect all of them and we just go right to their API.
You also mentioned identity. It seemed like you’re using uPort, based on something I saw on your website. Can you kind of explain, I guess in simple terms, what’s the problem around identity and the hurdle of it when you’re dealing with financial services — and how you’re kind of going about that and, potentially, if you are doing anything: how do you assure that fraudulent accounts don’t get created and stuff like that?
Creating a fraudulent account is kind of counterproductive here because you’re trying to create a credit history. I guess if you’re trying to create a fraudulent identity, that’s another thing, so let’s talk about that for a second. uPort is really a model that we’re drawing on. It’s a really interesting model for providing a full identity solution on the blockchain.
One really big problem that it solves is recoverability. So, what happens when someone steals your identity? How do you reset that? Basically, it’s a kind of system where you can have a trusted quorum and if 3 out of 5 family members — say if your identity was stolen, you can reset it. And it’s on the blockchain. So that’s the model that we drew inspiration from. For smart contracts, other than this point all said is just living internally in our testnet. So that’s what we’re doing with uPort.
To us, identity is kind of everything, in the sense that connecting to the right kinds of identities is how you get payments going in the first place. And that [question] “Where does the user’s identity actually reside?” Is it in their bank account? Is it a form of cash? Is it like — do they have some other way that they receive funds — like where does their KYC data reside? It’s really finding where that data is and piping it into the blockchain is how we can connect to the user. So that’s kind of how we think of identity. Instead of being a general framework for “This is how people should represent themselves”, it’s like “Ok, we need to send $50 to this person. Where are they?”
As far as anti-fraud — usually when we deal with these kinds of identities that are exclusively on the blockchain, they’re usually receivers. You kind of need to look for fraud more on the sender side more than on the receiver side. So it’s a bit of a non-issue for us.
When you’re positioning yourself, are you sort of, as a business more like a fintech firm that’s sort of bridging financial services with the customers that are there but don’t have (for whatever reason) physical access or software access to those banks, and then are kind of letting the banks handle the legality of whatever KYC and AML?
I mean everyone has to be responsible for KYC and AML, but let me put it this way, the kind of biggest bright line for us is that we’re not licensed as an operator, although we are actually in the process of getting a license for payments only. So one thing that we don’t have, is for example, is a US money transfer license, which takes like two years to get. Everything else is easy — building the software is easy, connecting to people is easy, going out there and doing business development with partners is our “problem”. Getting licensed is simply the biggest barrier.
Trust question with mobile banking, ends at Going through all you stuff, it seems like trust is going to and probably does already play a big part in this, and in a way, there’s almost like a trust broker aspect between the banks and the local ad-hoc networks. Can you speak to this a little? How are you ensuring that the end user can trust the entirety of the system?
Why does the end user need to trust the entirety of the system? Why do they need to do that? Why do you say that trust is a big deal here?
Because they’re entrusting their money with a new system that they might not be so familiar with.
Yeah, I mean this is not that unfamiliar of a system, honestly — it’s why we built it this way. Agency banking, mobile money has existed for a pretty long time. Mobile money has been around since 2006, so it’s been twelve years now. Agency banking, for way longer than that. I mean, that’s been the easiest thing for us. We just go to the end user and say, “Hey, now you know how to get your money from there, instead of going over there.” And they’re like, “Oh, ok, just like VTN, who’s our partner there.”
I guess to answer your question more directly, they’re getting money from people they trust already, which is like the local gas station. And the local gas station — we also have a licensed partner in Nigeria — so the gas station is trusting them. At no point do you really have to trust the system. We’re just kind of using the system to link everyone together.
In one of the podcasts that you spoke on, you mentioned that since the whole system that you’re creating is open source, you’re not trying to make any money by charging people fees for using the network. Instead, you’re making money by offering money transfer services. And in a way, it made me think that you’re almost building an open-access public utility and then using the audience that you gain from that to drive the real profitability of the company. Can you speak a little bit about how you thought this through and how you arrived at it?
Yeah, that’s very perceptive, what you said. We totally saw the Kora Network as, basically, a utility. It’s virtually impossible to make money off blockchains — the operation of blockchains as a protocol provider — for example if you make it open source, someone can just copy it and take out the part where it says you have to pay any money. And if you don’t make it open source, no one’s going to trust that you’re not just stealing all their money at any point in time.
So we thought about that for a really long time and eventually we settled on, “Ok, fine guys, we’re not going to make any money off the Kora Network” We’re just going to use it as, essentially, a customer acquisition funnel. And use it as a utility which really just provides the foundational services — mostly identity, as well as access to everything else it’s providing on the network.
There are a lot of financial services that have some level of fixed costs that they provide, as well as some level of liability. So creating software for identity is fine, but attesting to a regulator for that someone’s KYC information is correct: that makes you liable. And you can’t do that as a decentralized network. It has to be some entity that has an office and has a license and has people connected to it that could be fined for that to happen in a regulated context. So we’re essentially playing the role for that entity for a lot of services that you can’t just do by just sticking it on the blockchain, such as money transfer. That’s basically how we see our business: we profit off providing financial services and, of course we’re also incurring some level of cost and some level of risk for, essentially, taking on that liability.
As far as other startups in this space, it’s not really common as a business model. I think the most common business model out there is, “I’m going to make a token and I’m going to build a business and that’s going to make the token go up in price and then we’re all going to make money.”
Who know — obviously it worked really well for ConsensSys and some of these really big protocols. I’m really not sold that it’s going to work for everyone, but we’ll see.
There was something interesting that I wanted to clarify: you can interact with — your system is designed to be able to interact with multiple financial institutions and multiple individuals rather quickly using a Tedermint proof-of-stake kind of thing that enables people to… when it comes to the visibility of the data, it enables people to have their data to be auditable if necessary but protected from the general public. Can you speak a little bit about how that works?
When you say protected from the general public, what do you mean by that?
As in it’s not a totally public blockchain — I can’t just look up an account and see all the transactions that are going through with any account associated with this person’s identity.
Gotcha, yeah all Tendermint gives us is just scaleablility. Tendermint is just a really fast consensus algorithm. And so throughput is what it actually gives us. So we’re using Tendermint at the bottom (at the root) of our entire blockchain. That’s a public blockchain, you could just look up any address and say that these things happened from this address, but the only data we’re running through this Tendermint blockchain is a bunch of hashes. If you’re trying to do that through a generic address, you can’t even see which address and what hash.
The idea there is that we’re basically using Tendermint and the public blockchain as another accounting system — so called n+1 entry accounting. And then all of the actual consensus of which transactions happen, happen above the public blockchain, so to speak. So if I am sending money to you and I bank at Bank of America and you bank at Wells Fargo, and let’s say, for a second, that all of us are on the blockchain. Then the four of us (as well as whatever, Wells Fargo wants to bring in KPMG as an auditor and Bank of America wants to bring in PWC as an auditor and I want to bring in my local community bank as a watchdog — whatever, only the interested and consenting parties are put into a tiny, private blockchain (except it’s not really a blockchain, it’s just a tiny private consensus round) and we all agree on the state of my transaction to you. Which is by the way what happens in the banking world right now; it goes to whoever is responsible and everyone gets a copy of your data. (Most places, it’s required by law, so it’s not even really their fault.) And then that transaction gets stored on a personal data server. In this case, let’s say everyone’s running a full node and everyone’s running their own storage. Each person keeps a copy, sticks it into their hard drive or the cloud or whatever and then encrypts with their private key.
And so you better get two layers of protection at that point: one, you shouldn’t be giving just anyone access to your cloud anyway and second, you encrypted it so only a private key holder can access it. And then for balances, you would actually have to build them up from scratch. So, that’s kind of how Bitcoin works: you start from zero and only when you get a transaction to you, that’s when your balance goes up. For Bitcoin wallets, all you do is count all the unspent transactions under your name and so that’s how we keep balances and transactions private. And then the transactions are actually get hashed, which means that the data is used as a seed to generate a random string of numbers and letters. So there’s no data that’s immediately available to anyone from the hash. Unless you have the raw data. So that gets stuck onto the public blockchain and if there’s ever a dispute between these seven or however many end parties, any one of them could always look up the hash on the mainnet and say “Hey, my data matches what was written here and got your private key on it.” So that’s just a way — that’s why we call it n+1 entry accounting and that’s how you essentially resolve disputes.
And so just to be clear, you’re creating your own blockchain with its own native token, it’s not on top of Ethereum.
Nope it’s not on Ethererum, it’s our own blockchain, Kora.
Can you talk a little bit about how you recently started a beta project in Nigeria with farmers — demographics, what are you doing, what have you learned, if you can share anything about that?
Yeah, so that’s a lot of fun. We’re working with an aggregator that has about 25,000 farmers in all, but we’re not working with all of them. So one client or our partner, is an aggregator called Thrive Agric. We work closely with them. The other group that we work closely with is called VTN. They’re an agency bank out in Nigeria, and our end users are the farmers.
What we’re doing there is simple. It’s really just that Thrive Agric and their bank is in the city and the farmers are not, and it’s like, “How do they get money out to the villages instead of having their clients drive in?” We’re working with VTN. They have a USSD license and an agency banking license. They provide regulatory cover for us to move stuff around and cash out locally, in cash.
Yeah, that’s basically what we do. It’s coming along nicely and it’s a really good partnership because Thrive Agric wants to be a agricultural startup. They don’t want to be a financial services startup. But they start scaling really quickly and they’re like “Oh crap, now I’m managing a general ledger for 25,000 people and, there’s [pressure] for getting that money on time so they can produce crops for me, and I don’t want to do this anymore.” So that’s where this partnership started.
It’s been a really great learning experience for us and it’s been a lot of fun working with Thrive and VTN, and hopefully the first of many.
What specifically are the services that the farmers are receiving currently? Is it money transfer, loans?
Just payments right now, quasi-identity.
And you’re focusing equally on the technology as well as user feedback for design purposes?
I mean we focus on helping our users, so whatever needs to be done to do that.
What region of Nigeria is it in?
It’s around the village of Kaduna, the farmers are there; Thrive is around Abuja [the capital].
What other projects are you involved in? It seems like you were in Colombia, Bangladesh, and other places as well. Are those active or future plans?
We didn’t end up in Colombia. We actually ended up in Peru. But that whole space is everything that we’re doing around coffee and so we might actually end up in Ghana as well, doing a similar project. That’s working with coffee exporters, also around payments.
In this this case, it’s two payments: one from the buyer to the exporter and one from the exporter to the farmer. And obviously from the farmer to the picker as well. You see the money kind of flowing down through the ecosystem.
That was an interesting one, because, first of all in Nigeria the amount of people is higher but the total amount is much lower. In Peru, when we make one payment to one farmer, it’s like over a million dollars because it’s high-end coffee. But it’s also — for them it happens only once every four years. And then obviously that’s the pay, so it’s a very complex value chain.
In Bangladesh, it’s also very different because we’re also working with a co-op there. In Nigeria, it’s more an aggregator — which I guess is also kind of a co-op but they’re only domestic. In Peru it’s an exporter so it’s only cross-border. In Bangladesh, they have a lot more skills and a lot more focus on… they’ve been running for way longer. Thrive (our partner in Nigeria) has been around for like 9 months; our partner in Bangladesh has been around for 17 years. They started with just someone keeping a notebook of who has how much money and now they have 35,000 people and there still keeping notebooks of who has how much money.
So we’re moving their core systems to the blockchain, as well as doing all kinds of crazy stuff with them there — they’ve really embraced the things you can do with technology and financial services, in general. I can’t talk about some of the stuff we’re doing there but there’s a pretty big vacuum as far as stuff like more sophisticated financial services. It’s also important to remember that financial services is actually — if used the right way, you can actually help people (laughs). It feels weird to say.
It’s that investment banking is useful for raising money and it’s useful for price discovery. Those are the kinds of services, where if you apply them in the right way, they really have a close focus to where the value is coming from. You can unlock lots of illiquid value and scale. So we’re doing stuff like that in Bangladesh.
Did kind of co-op is it?
It’s everything. Cheese, rice, chicken. They actually run their own soup markets, which is super cool. We’re probably going to do a blockchain-based voucher system for the supermarkets.
That’s really cool. If I missed anything else, especially regarding innovative stuff that you’re doing around driving adoption, tell me more about that.
I’ll say that the co-op is really becoming the main thing — even if you read our whitepaper, it all leads up to these things called CVNs (Community Value Networks) and everywhere we go, the co-op has become a centerpiece. You can’t even avoid them if you want to. They’re really powerful — that’s a really powerful structure for reaching economies of scale locally. And if you read a lot of the more academic work that’s been done on co-ops — because you’ve always had the shadow of the tragedy of the commons hanging out when you’re in a co-op. Or just plain old corruption or embezzlement where it’s like, “Here’s a everyone’s money in cash; I’m going to take a few dollars to use it to go buy, you know, whatever.” The blockchain is actually this really powerful tool that almost overlaps, word for word, with some of the research that’s been done on co-ops.
If you’ve read Governing the Commons by Elinor Ostrom, you probably know that some of the principles for effectively governing the commons come down to transparency and consent. Consent about simple rules; transparency about who’s breaking those rules, and then being able to enforce them. The blockchain is this incredibly powerful tool for that. The blockchain itself, and the fact that it runs at all, means that a bunch of people agreed on a protocol and if you violate a protocol, you get penalized very quickly. Blockchain itself is based on The Commons that Ostrom set down, and so the same deal: in the co-ops that we’re building, there’s complete transparency about the funds that are being spent; in a digital currency context, we can do things where if people don’t agree on a potential payment, they can say, “No, you can’t make this payment. You can’t take this money out.”
Now in a cash system, it’s a bit trickier. What are you going to do, put like 15 locks on it? But in a digital currency system, it’s easy. If we haven’t reached this threshold, I’m not going to dispense the money. That’s what it says in the system. Financial services in emerging markets is incredibly competitive and if there’s a lack of services, it’s not because someone said, “Oh, fuck these poor people,” or “How about we don’t bank those people.” It’s usually that someone’s thought about it, did some research into it, and probably tried it and realized it just didn’t work.
I think the blockchain, for co-ops is really unlocking value in a way that has never been done before. I think it’s really cool, what we’re doing there. It’s becoming the centerpiece for trust on either end. We can make systems where if people trust that if I put the money in, I can’t take it out unless X happens, or if I say something is true that no one can disprove that, which is a kind of trust. But with co-ops it’s stuff like “How do I know that you know how to farm? What do you know about farming in Bangladesh?” Those are the trust problems that co-ops solve. The trust problems we solve for co-ops are like basic coordination problems — like here’s how we can prevent people from improperly using shared resources. So I think that’s a really cool piece.
That would be a real focus on the transparency side of the blockchain, but is it also the potential for governance and cost reduction of operation? What would you really ascribe that the blockchain really is doing that specifically benefits the co-op?
All three. It’s getting to transparency in how to use funds. It’s giving the ability for people, at scale, to control and govern how those funds are being used. It’s a lot cheaper. Like, who makes software for co-ops? Who makes banking software for co-ops? No one does. So it’s like providing tools, at all. The alternative to using the blockchain is not to just not use the blockchain, it’s what they’re doing in Bangladesh: let’s write everything down in big-ass notebooks and spend three months or longer every year just trying to audit all of them. Because they have basically a quasi-financial services group; they have files they have to send up to the government as well. It’s also a huge reduction in cost increase in convenience for them as well. So it’s everything, really.
