Orange Silicon Valley isn’t just a crossroads for startups and partners locally. It often acts as a bridge from the Bay Area to other continents, and that was the case on May 10 when the University of California, Berkeley’s Center for Effective Global Action (CEGA) brought together thought leaders, innovators, and funders for its Smart Government: Harnessing Technology for Public Good conference.
Hosted at OSV’s San Francisco event space, the conference provided a window into research and initiatives across Africa, Southeast Asia, and South America. The lineup of speakers and presentations touched on two areas where OSV has in-house expertise: innovation in Africa and smart cities. Darren Sabo and Will Barkis, who oversee OSV efforts in those areas, helped bring the CEGA event to Spear Street and offered some insights into the discoveries that were on display.
Darren and Will explained how important these projects can be, as well as what kinds of opportunities emerged for startups and potential corporate partners.
Orange Silicon Valley: How did the Smart Government conference come together here at Orange Silicon Valley?
Darren Sabo: The Smart Government conference is part of an annual conference that CEGA puts together. It’s called an “Evidence to Action” event, and they do different themes from year to year. The theme this year happened to be smart government. I’ve been impressed with CEGA’s work for quite some time. Based on our interactions with the head of one of CEGA’s initiatives called the Digital Credit Observatory, we were made aware of this event. Upon hearing the theme, I thought there was particular relevance given Orange Silicon Valley’s interest in smart cities and Will’s activities in the civic tech space — and his relationship with the city of San Francisco [as Volunteer Technology Advisor to the S.F. Mayor’s Office of Civic Innovation].
— CEGA (@CEGA_UC) May 10, 2017
Will Barkis: CEGA is this amazing center at the U.C. Berkeley. They are bringing together a community of researchers doing excellent social and behavioral science research, and then taking the evidence generated by that research and using it to drive action in the public policy sphere. It’s going to be hugely impactful in the long run.
In that nebulous thing we call “innovation,” there is a certain piece that is about basic science and research. So relationships with various academic researcher communities helps us do our job. It helps us when we partner with startups. It helps us when we work with large corporate partners. And it helps us internally when we’re organizing initiatives, particularly stimulating growth around various African markets.
DS: The vast majority of startups and small businesses in Sub-Saharan Africa aren’t making data-driven decisions to launch their products or services. They’re making decisions based on what their neighbor is doing — or what they see happening within their community. They don’t necessarily understand the broader impact of a lot of the products and services that they’re developing. By potentially giving them access to a wider base of research or understanding of what’s happening in the markets around them, and disseminating that research through our extensive network within Africa’s startup ecosystem, I think it can add tremendous value to those companies. I saw a recent statistic from SARS South Africa that 75% of SMEs in Africa do not make it past the second year of operation and 95% fail before their fifth year. I really believe that access to credible research and better data in general can improve those numbers dramatically.
OSV: Who showed up at the event, and what did you learn from the stories being discussed?
WB: CEGA has a very interesting and well respected set of researchers that they engage. I didn’t realize this, but it’s sort of geographically limited to the West Coast — so, stretching up to Vancouver and all the way down to San Diego. But their academic bench is pretty deep. This was their annual community gathering, so all of the researchers who are in their network who they fund — or who are affiliated with their center — come to this to give talks and to hear talks, as well as to just network. So it’s a conference by academics for academics — with some perspectives from their partners in government, like the minister of IT for the Punjab region of Pakistan, as well as the partners who fund the research like the Gates Foundation.
DS: Also, Thiago [Marzagao, a data scientist from the Office of the Comptroller General in Brazil] really stood out for me. I thought it was fascinating, some of the work that his team did around modeling available information on politicians to try to fight corruption in Brazil. They used a government dataset combined with machine learning algorithms to literally create a corruption score for individual politicians. While this can’t legally be used to block or reprimand a politician yet, the tool’s potential is extremely encouraging.
WB: My biggest takeaway from the day, which ties into what I hope Smart Cities means in the really long-term, 50-year vision, is that evidence-based policy making really is possible. To hear all these cases studies from scientists studying elections in Sub-Saharan Africa or corruption in Brazil, it shows how far we need to go from the art of policy making to the science of policy making. But it also gives me hope that we are actually moving toward evidence-based policy making.
OSV: The subtitle of the conference was “Harnessing Technology for the Public Good.” What types of opportunities did you see emerging from the discussions there?
WB: Having impact and driving towards truly evidence-based policymaking is not going to be easy. Local political realities are always going to be a challenge for implementation. The good news is that there is a set of technologies that are making a lot of the foundational pieces significantly less expensive. Getting data about the world has been expensive in the past — or implementing programs, sharing knowledge, or connecting people. All of these things have been expensive in the past. But now that we’ve entered the era of smartphones, mobile data, and communications, the cloud and modern databases; you can see how we can afford to actually scale and I think the technology will eventually be an enabler to scale interventions that just were not possible 20 years ago.
— CEGA (@CEGA_UC) May 10, 2017
OSV: This seems to get back to what Darren was saying about startups in places where they’re relying on local experience to build models. They can benefit from having a third party come in to enable better decision-making.
DS: I have to assume local governments aren’t much different from startups in that regard.
WB: We are a little bit living in the future here in San Francisco with the startup culture, because when you look at where most companies are in their own journey of modernizing and digitizing you see that companies are slow to adopt new systems and new processes. But governments are even behind that.
DS: I think it’s an interesting opportunity to give a voice to populations and make them feel like their opinions do matter; things that are happening to them are being measured. It’s a way of engaging with the population.
WB: There were several examples that really blew my mind. One of them was the absenteeism of teachers in India. The absenteeism is insanely high. I want to say 30%, and I don’t know how you teach classes when the teachers aren’t there 30% of the time.
Another illustration of empowering people was a farming challenge. A lot of farmers need veterinarians to help with aspects of cattle breeding, and the state paid veterinarians to do an OK job. But they often don’t do a very good job. And it’s not because of their training per se. There are systemic issues like compensation, and of course it is complicated. But this researcher created a kind of Yelp app via text for people to give feedback. And it radically increased the results they were getting from their veterinarians. The very same ones. And it didn’t cost the government any more money. It was a nudge to the system that created this cool, emergent property. I think you see a lot of examples like that that are showing the improved effects on communities of empowering people through these new technologies.
OSV: You’re talking about some really big ideas there with increasing cattle production or improving education outcomes when teachers aren’t showing up. How achievable to you think the types of solutions that fall into this space really are? Is there a range? Are some wins easier than others?
WB: The hard part is really the social, cultural, and policy aspects and creating what I like to call “social infrastructure.” Proving any one of these use cases out to say, “Hey, we can do better than we’re currently doing,” is hard with all of the stakeholders involved, but it’s even harder to then implement it in all the other countries where you could. The technology is relatively easy — but scaling the solution requires people to know that it’s possible and to know that the solution is there somehow. CEGA’s goal is to be that locus. You put some stamp of approval essentially on a given kind of policy. That is going to take a while. It’s not like you’re going to have an app store where the government goes and says, “Oh, we need to improve the quality of our veterinarians or teachers. Let’s just download the app.” It’s a very manual kind of process right now.
— PunjabITBoard (@PITB_Official) May 11, 2017
OSV: What types of collaborations and knowledge-sharing solutions do you see closest on the horizon?
DS: When you talk about product or service adoption, in almost any market, it’s very rarely the case that the problem lies in the technology. The key challenges and hurdles are almost always behavioral or organizational in nature, especially when you’re talking about enterprise technology adoption.
WB: I think this is long-term, slow-burn change, because it’s a lot of internal, organizational culture change. You’re moving from doing business as usual to scientific, evidence-driven policy making. If the leadership doesn’t believe in it, than it’s difficult to make happen.
OSV: And what do you think the biggest need is to scratch there? Is it finding behavioral studies? Is it watching the right studies?
DS: To be honest, I think it’s the politicians’ understanding of how the information affects their ability to maintain a stable government and stay in office. The currency is votes. How does the adoption of new research or technology translate to votes? If it helps contributes to the prosperity and happiness of the constituency, and makes the politicians look better, I think you’ll see a much sharper adoption curve. If for some reason adoption sheds a not-so-positive light on the political incumbent, then you’ll see more resistance and downright obstructionism in some cases.
OSV: And in many cases there can be multiple layers of those entities to work through, right?
WB: One of the things I always say when I talk about smart cities is that it’s a locally defined idea — it is based on local needs and local political priorities. So it is a slow burn because it is culture change. That said, some of the idea of smart cities and smart government that’s exciting to me is that it’s data-driven at its core, so you’re wrestling power out of the hands and silos that have it. Because of data. If the data shows that what you’re doing is corrupt or wrong for the outcomes that you’re saying you’re trying to achieve, then you can no longer keep doing it that way. If there’s no data, then it’s easier to keep doing it that way.
OSV: What are you most optimistic about that you saw discussed at the event?
WB: I saw a lot of evidence that we can and are moving towards evidence-based policymaking. Technology is helping enable a lot of that. The way the smartphone was the culmination of a lot of technologies getting good enough to cross some threshold of usefulness and coming together to create a kind of totally new experience, I think smart cities are kind of that similar thing, where you have movements around Open Data and AI and machine learning and data analytics, modernizing databases, smartphones themselves actually and cheap sensors, and other things coming together to create the possibility for a smart city, a smart community, a smart government. A lot of those technologies are getting exponentially cheaper or exponentially better for the user experience. Whatever it is. Communications are getting better and more abundant and cheaper. Faster. And what we do as our core business is going to enable so many of these types of things to be possible at massive scale that wasn’t previously possible. Finally, though, beyond the technology, the creation of social infrastructure by organizations like CEGA and partnerships with companies like Orange makes me very optimistic.
DS: What I heard from multiple parties was that in order to bring a lot of the technologies that were studied to market — some of the challenges faced were that a lot of the studies were RCTs. You heard the phrase RCTs a lot — randomized control trials. This research methodology has roots in clinical trials, and is great at moving beyond simple correlations to uncover causal relationships. However, the challenge is that’s generally a small-scale or limited-focus style of doing research — and in many cases it RCTs don’t address the important “why?” question. Because of this, applying those findings to broader-based research or go-to-market strategies for businesses is a major challenge. Instead it’s not uncommon to hear, “Well, more research needs to be done.” How do you apply those findings to delivering those products and services at scale? This is very important in terms of delivering actual real world impact.
What I heard from investors, as well as others from the private sector, was a push for the researchers to think about the scalability of whatever they’re testing right from day one in the research design. So, when they eventually publish the findings, it’s actionable by private parties to actually take something to market instead of ending with “sorry, need more time and funding for that.”
Instead, it would be great if they finished with: “Here’s the minimum viable product, This is how it can scale, or These are technologies that you can leverage to achieve scale, starting from this concept.” That’s not being done currently, but just the fact that a lot of different parties were putting pressure on researchers to think in those terms, is a very positive thing, so that when they design the next wave of research — and some of these studies can take a year or multiple years to complete — my hope is that they’ll think about how those products and services can scale. And maybe commercial partners can be part of the research design, so they we [or they] can explain some of the challenges of scaling those technologies — or those services — that academic researchers may not otherwise be aware of.
At the end of the day, it comes down to behavior and organizations and mindsets — not the technology itself.