24th Annual Financial Markets Conference - Mapping the Financial Frontier: What Does the Next Decade Hold? - May 19–21, 2019

Banner image for 2019 Financial Markets Conference: Mapping the Financial Frontier

Interview with Sudheer Chava and Warren Weber

Sudheer Chava, professor of finance at the Georgia Institute of Technology, and Warren Weber, visiting scholar at the Atlanta Fed, discussed the Blockchains in Finance sessions of the 2019 Financial Markets Conference. Brian Robertson, quantitative financial markets adviser at the Atlanta Fed, conducted the interview.



Brian Robertson: Hi, my name is Brian Robertson. I'm the quantitative financial markets adviser here at the Atlanta Fed. I'm joined today by two Bitcoin and blockchain experts: Warren Weber, who is visiting scholar here at the Atlanta Fed, and Sudheer Chava, professor of finance at the Georgia Institute of Technology.

Gentlemen, thank you for joining me today.

Sudheer Chava: Thank you.

Warren Weber: It's my pleasure.

Robertson: So we heard two sessions today about blockchain, and we've covered quite a bit of ground. Overall, though, how did you think the sessions went?

Weber: I think the sessions went extremely well. For people who didn't know much about blockchains, cryptocurrencies, digital currencies—I think they got a tremendous introduction from some people who really know what is going on, who've worked in the industry and have been doing a lot of deep thinking about the issues that surround them—the advantages, the disadvantages.

Chava: Thank you, Brian. I enjoyed the sessions a lot. Actually, I've been teaching a class on fintech for around five or six years, but our administration has actually got new perspectives, some of which I haven't thought about. For example, Eric's session on the limits of Bitcoin, and how we formalize that—that's a new insight which I haven't thought about, so they're very interesting and very informative.

Robertson: The first session this morning talked about the advantages, or potential advantages, we could see from these blockchain and Bitcoin-style technologies, whereas the second session focused solely on the Bitcoin technology itself. It seemed that there was a bit of a distinction between the two in that there are these permissioned blockchains versus permissionless blockchains, and I was just wondering if you guys might be able to provide some insight on what that distinction is, and how we might understand why that's important.

Chava: So I think most of the attention has been...or at least, the original purpose of Bitcoin, and the blockchain came out of the technology which is behind Bitcoin—the initial phase, I think people looked at Bitcoin and then afterwards, people thought about the blockchain as the real innovation behind that. It's a public blockchain, so one way to think about it is a database system where there's open read and write, so everyone could write into this one, as compared to a traditional database system, where there are a limited set of actors which do that. But where the permissioned blockchain gets back to a more limited database, is where there are players—which might be centralized parties, there might be trusted parties—which can actually have the write permissions.

For example, in the context of Wall Street, it might be DTCC [the Depository Trust & Clearing Corporation] which might actually do that. Another way to think about it, it's a limited ecosystem. Who gets into the system is limited in a permissioned blockchain, whereas the permissionless, the open blockchain, which underpins Bitcoin—it's anyone can enter into the ecosystem.

That's one way to think about it. But with both, the idea is that both have this immutability in it. Again, there are some advantages and disadvantages.

Weber: I think the one part I took away from the presentations this morning is that the so-called permissioned blockchains are really no different from the usual information management, data management systems. Maybe there's a little difference in how everything is organized, but, really, they all require some kind of trusted third-party agent to be involved in that, whereas the whole idea of the permissionless blockchain—and there are many of them out there, there's not just the Bitcoin blockchain; I think there's maybe too much focus just on the Bitcoin blockchain, but that's just my personal opinion—the permissionless blockchain is really about trying to get trust in something where there's no trusted third party involved in that.

All of the discussion in the second session was about one method of trying to do that, which is called "proof of work." And proof of work is very expensive. It's very costly. It was very good for the discussion to bring out how costly it really is to generate trust there, whereas if you have a permissioned blockchain, you have the trust because you have the trusted third party right there at the beginning.

Robertson: Again, we covered a lot of ground between these two sessions. With these distinctions in mind, what do you feel are the key points that we might want to take away from either the sessions overall, or from one session in particular? What do you think was kind of the golden nugget?

Weber: I think there were several golden nuggets that came out. The first one, that came out in the first session, was, don't think that just because it's on a blockchain—and I don't care if it's permissioned or permissionless—that it's somehow trusted, because it's data being entered, and so you have to trust the entering of the data. There's that first level of trust, before you get to the actual trust of the accounting system itself.

The other thing that came out—and I'm not sure where we ended up in the discussion—was about whether or not the immutability of the Bitcoin blockchain, or the permissionless blockchain, is actually good or bad. Are there are circumstances in which you might want to change things? For example, if somebody hacked a particular block in a particular way, would you want to undo that—and can you undo it?

The other thing that I think was interesting was the discussion about regulation, and the fact that current regulation, especially around things like blockchains, really reflects maybe more what was going on in the 1930s or 1940s rather than the realities of today. Maybe regulatory agencies ought to be rethinking how they regulate this particular area.

Chava: There are a lot of interesting insights that came about, and there are some which I have been thinking about for some time, and many new perspectives which I haven't thought about. The first thing is in terms of even regulators trying to stay ahead of it—and I think it's a very nice thing, with this innovation. So I think it's always...ahead of regulators, so I think it's important that the regulators are ahead of them, and following that and trying to incorporate them.

That's one. The second thing is in terms of blockchain or Bitcoin, or other innovations—a healthy dose of skepticism. I think it's important to think about this as a real financial innovation. There's a huge potential—I think that's what many people in the industry are motivated by—but there might be some cause for skepticism also. I think that's one thing which I took away.

Another way of thinking about it is, an often-attributed quote to Bill Gates, is, we always overestimate what can happen in the next two years, and underestimate what can happen in 10 years. So I think about blockchain in that context. There's a lot of hype and a lot of energy, which many people in the industry think is going to change the world in the next few years. That I'm not sure about. But there might be improvements and there might be more technological innovations which can make it a better tool maybe going forward.

And the third is in the current context, some of the limitations of Bitcoin, or even blockchains, which I've pointed out—one is, many companies, if you have this proof of concept—there are other implementations—where one is trying to impose this blockchain on the existing processes. I think that's a big issue, because just by putting, as somebody has pointed out, garbage into a system, again, it's garbage out.

Another way of thinking about it is, if existing systems are not optimized—that is, you just basically impose the blockchain on an existing system, an existing set of processes—then it's going to be very inefficient, and it actually might not scale well and might do much worse. So the way out is probably just going back to the drawing board, and going through the system and saying, what are the processes? Where is value being added? Where is efficiency that can come from a blockchain? Which ones do I want to put in a blockchain? Which ones don't I want to? Which data do I need? How does it scale? Is it appropriate for this particular situation? Is the solution in search of a problem, or a problem in search of a solution?

So to think about these broad issues, and have this kind of a healthy discussion—where people see the potential but also the issues, so that they can be addressed in the future—I think will be very interesting.

Weber: I think you made a great point, which is, one of the panelists this morning said, "When you're thinking about this, think of what the problem is that needs to be solved or that you're wanting to solve, and that the blockchain may not be the solution to every problem that's out there."

Robertson: Were there any issues that you are aware of around this field that perhaps we didn't get a chance to talk about here today?

Weber: Well, the one issue that came up—but this is a session on financial institutions—there's the more general issue of how a blockchain can be used to help people create their identity—who they are, what they've done—and the potential for, if you can get stronger identities, what that could do in terms of unlocking the possibilities for people getting credit and what that could do for the standard of living across the world.

Chava: I think that's a great point. I think in terms of both financial inclusion, where there's a segment of population—even in the U.S., I think there is 20 to 25 percent which is underbanked—which partly relies on our identity. Can blockchain be used for an immutable identity card? I think that's interesting. There's some work being done in that area, but I think a lot more needs to be done.

Again, all the issues with security will come in, all the issues with blockchain would come into that issue, but I think it's a great point, because if it can work—identity, I think that's the key to financial inclusion. I think that would be great.

I think we haven't talked as much about what's the initial coin offerings [ICO], and some others—because probably the time has passed for that. Last year I think was...[laughter]  Many of them I think almost fail. There are a lot of scams.

Weber: If you read the white papers, some of them sound like, "What are you trying to do?"

Chava: Right. Some of them actually are the straight copy of other white papers [laughter]. I have, actually, a PhD student who is looking at it in their research into the ICOs. But in general, the token economics—that, I think, is interesting.

Weber: I wanted to go back to the identity thing. Again, if you're entering things—like records, like marriage records or birth certificates, things like that—do you trust the person who is entering them? Or think about tokenizing mortgages: do you trust the person who is entering the mortgage, or someone who is doing the title search—do you trust all the entries that they're looking through?

So I think one of the things I really took away is, the first level of trust is the entering of the data. Just because it's data doesn't mean it's trustworthy.

Chava: One more point, which probably hasn't been covered as much, but I think which might be very interesting, is from a privacy angle. I think in a blockchain, people are thinking about differential privacy. So I think when all the information is out there, when everyone can read it, how do you maintain privacy? I think that, again, there are ways of doing it, but that's another interesting angle—privacy and I think...are something which I think would be nice to see.

Robertson: Absolutely. And from my perspective, one of the things that we didn't really get a chance to broach is this notion of, "So we have data on a ledger, and that represents something that could potentially represent something physical." Just because I trade the data on an exchange or through a ledger doesn't mean that asset actually gets transferred from one party to the next. So how does that get facilitated? How does that get done in a way that is in a trustworthy process—not unlike what we see. Think about buying a home. Just because you've transferred the funds for it doesn't mean you've actually received the title. So how do we make sure that sort of thing happens?

There's a lot of ground to cover in this topic. It's an exciting field to research, from my end, and to talk about with these gentlemen here today. I thank everyone for viewing in. We'll be doing other interviews throughout the conference, so make sure you check those out.

Again, I'm Brian Robertson, quantitative financial markets adviser, here at the Financial Markets Conference in Amelia Island, Florida. Thanks.