The Power of Public Data with Harvard Professor Archon Fung

You can easily see the amount of calories in your peanut butter and your car’s safety rating. But this publicly available data comes from hard-won battles — and the numbers behind many of our most crucial issues are either under lock and key — or they are simply not even measured at all. In this month’s podcast episode I interview Harvard University Professor Archon Fung PhD about what happens when we turn important data that is often hidden and contentious into a publicly available resource for the world to see.

Watch the interview on Youtube

Katie: Why is the public disclosure of data so important?

Archon: The disclosure of data is very important because we depend on data and information to make all sorts of important choices in our lives from where to send our children; which neighborhood to live in; what kinds of food to buy and eat; what kind of products to choose to buy; if we get sick, what doctor to go to; or where to send a loved one to get medical care. We can make these decisions well or poorly. A lot of those decisions depend on the quality of data and information that’s available to us.

Katie: The type of people who are going to be listening are people who work in government, not-for-profits, or are social change entrepreneurs who are trying to make change happen. What are the kind of problems that happen when you’re actually trying to change the world and there is no disclosure? There’s no public data, and there’s no regulation around making information public. What goes wrong?

Archon: One of the things that goes wrong for social change agents is that their job becomes a lot harder because it’s hard to tell the good guys from the bad guys. Say you’re working in pollution prevention, or an environmental area, and you want to reward companies that are good environmental stewards and citizens, and maybe punish companies either by singling them out and drawing attention to them. If you’re a consumer deciding to buy somebody else’s products, it might be helpful to have some information to be able to differentiate between companies that are doing a good job vis-a-vis the environment or society, or their workers and employees versus companies that are doing a bad job.

Katie: The most exciting example that’s come out of your work, or from what I could see, was the Toxics Release Inventory from the EPA. Can you talk a bit more about how that started and what results that had?

Archon: The Toxics Release Inventory is an interesting case. It became a law and regulation in the wake of a big disaster of a Union Carbide plant in India near a town called Bhopal. It released a large amount of gas, methyl isocyanate. A lot of people all around the world got concerned about pollution and toxins in the environment in the wake of that.

A little bit later, there was a law in the United States that was passed that created the Toxics Release Inventory. What that law does is it requires all companies that are running facilities above a certain size to report to the government the amount of toxins that they’re releasing into the environment, be that into the air, into the ground, or shipping offsite.

There’s a list of several hundred chemicals that are classified as toxins. They have to report what they’re releasing and what they’re shipping. Anybody who lives in the United States can get on the internet, on their web browser, type in their zip code, and a map will pop up with all of the facilities that are recorded in the Toxics Release Inventory. You can see in your neighborhood, town, city, or county who’s releasing what.

Over the many decades in which it’s operated, it’s had many effects. One of the surprising subjects is that companies and facilities managers have really paid attention to the list and tried to clean up their act. I think if you’re running a facility, it’s a bad mark if you’re revealed as a bad after on that list by being, you know, one of the top 10 polluters in the county, state, in a particular zip code, or whatever it is.

Katie: The numbers were huge! It decreased chemical use by about 45% in a pretty short period of time. Is that true?

Archon: I think there’s a lot going on in that period. Many things are happening in addition to the Toxics Release Inventory, but the Toxics Release Inventory, I think, did create a lot of incentives for many people to change their behavior. It was also used by environmental groups, journalists and lots of other people to identify companies that were releasing a lot of toxins into the environment. It creates a dynamic that some people call a “race to the top” or “race away from the bottom,” because you can begin to see and make visible what’s happening, which was invisible before the database was created.

Katie: What I’m really curious about with the Toxics Release Inventory is, how much of the change do you think comes from external pressures of the data? For example, consumer demand, government, or not-for-profits putting pressure on once these numbers are made available, which kind of makes sense to people. Once you put numbers out there, obviously it’ll be some sort of market pressure.

I’m curious also as to a different type of pressure that happens psychologically, internally. For example, if there’s no market pressure or no external pressure, just the companies seeing the numbers themselves, when their own employees see the ranks, do you think that has a psychological effect even if there isn’t a market pressure?

Archon: Well. I think that’s a really good question, and I think it’s hard to tell. It’s very difficult to tell how to break up the different components of what the Toxics Release Inventory is doing between external pressure versus internal desire of managers and companies to do the right thing.

I will say that it has to be both, and it has to be a mixed story, because if consumers and people in the public, journalists, and environmental activists really didn’t care then there’s probably no reason that a factory manager should care either. If nobody cares, then it’s just numbers on a sheet that have no particular meaning. Why should you manage to it if your performance on that criterion isn’t valued by anybody? It’s got to be both.

Now, I will say that I think people have different understandings of what actually happened, but my own understanding is that a lot of the dynamic did depend on internal decision-making among people in these companies. I expected to find quite a bit more external activism and campaigning on the part of environmental groups because I thought it was probably a pretty attractive source of data for these groups to identify who’s polluting a lot, and who’s quite green and not polluting very much at all.

Some environmental groups did use it, especially at the local level, but it got a lot less take-up than I thought it might. My own interpretation is because many environmental groups aren’t accustomed to using data in quite that way.

In the United States over this period, environmental groups are accustomed to fighting for tougher environmental regulations in Washington that environmental regulators would then enforce. That doesn’t require a lot of Toxics Release Inventory data. You can do that through lobbying and it requires other types of data.

If your organization, habits, skills, and what you do is accustomed to advocating in one kind of way that is fighting for better regulation, then I think you’re likely not to be attentive to an opportunity to use this data; So, I think that lest the data were used less and less effectively than they might have been by environmental advocates.

Katie: Yeah, that’s really interesting. I’ve been doing a lot of outreach for some of my own concepts recently to people who work in not-for-profits and in government, and I’ve noticed a definite bias in the way of thinking that everything needs to be done through lobbying and through top-down legislation.

I’ve been coming from a different angle. Let’s gather lots of data. Let’s make the data public. Let’s make it easy for people to find. Just through making this data public and beautifully designed, I think we can come from this different angle. I noticed it was this incredibly novel concept for them. They were very much used to thinking in that type of way that you just described.

Archon: I have to think about the Toxics Release Inventory alongside a very different transparency and disclosure regulation, which is not really about environment but about the city. This is a different transparency regulation called the Home Mortgage Disclosure Act.

This is a regulation that required banks to disclose to public authorities, who would then disclose to the public, who they were making loans to and who they were refusing to make loans to. This was an anti-redlining regulation.

I don’t know if your listeners are familiar with the term “red-lining,” but it was an accusation for a long time that banks were operating in racially discriminatory ways by not giving home loans to low-income minority borrowers.

Many banks denied this. Many other observers denied that it was happening, but then a regulation passed that created a lot of data to allow people to analyze who exactly was getting loans and who wasn’t getting loans. One result of this was that it turned out that people thought a fair amount of discrimination and home lending was occurring.

Once the data came out, it enabled a lot of advocacy organizations in cities, in particular National People’s Action fought for this regulation in Chicago. What they could then do is identify banks that were making loans in a non-discriminatory, fair way and separate those out from banks that were discriminating, then targeted those banks. They were saying, “Hey, you guys need to clean up your act. You need to make fair lending decisions.”

Then those advocacy groups are also able to advocate to regulators and say, “Hey, look! You need to regulate! Fair lending is an important part of our legal and regulatory structure, so you need to enforce on these banks.”

In the Home Mortgage Disclosure realm, there were many organizations who were accustomed to using the data, and they were the ones who demanded that government generate this data.

Katie: It’s really interesting how the government can come in from both ends of the data. In your book, you talk a lot about looking at disclosure as one type of regulatory framework versus a more descriptive command and control type of legislation.

What that means just for people listening, the government just says, “We’re not going to tell you what to do. We’re just going to make sure that you tell everybody the numbers, and then we’ll just be hands-off, and you do what you want.”

What I’ve also seen happening with all these new data startups that are coming up is they’re investigating more data of their own accord, not with the government telling them to do it at all, then that new emerging data is then influencing what government does. The data starts to tell a story.

For example, with air pollution, with urban heat island, you can start to see a lot more granular information with the new startups that are working in this space, and then that influences policy. You can kind of come from both ends once you start getting into the numbers, which I thought was a really interesting dynamic.

Archon: Yeah! I think that some efforts in data in your examples allow people to get a better picture of what the world is like, and then that will say, “Okay, well here’s some problems that we didn’t quite understand before.” Like where childhood asthma is occurring or some public health issue that we didn’t really understand what the patterns are, and now we can tackle that more effectively. I think there’s a whole lot of that happening and that’s really important work. That’s one way in which more effective generation and use of data can make the world a better place.

Another way is to generate data that helps people do the right thing more or less, or make the right decisions, or leverage pressure. This is kind of the more governmental use of data to achieve ends that it wants to achieve, like reducing pollution, increasing fair lending, or whatever.

As you were saying in the kind of top-down way of government action and regulation, we think of the government is one entity. It’s trying to find the bad guys, regulate them, and stop them from doing bad things. That’s what regulation is.

In the information domain, there’s another set of actors out there, and that’s society. Now, you have a three-actor game. You have government, you have some potentially bad actors, and then you have society. What transparency does is it uses government power to generate data for society to put pressure on the bad actors, as they did in the banks and they did a little bit with the Toxics Release Inventory. There’s all kinds of examples of that.

Often times for people in government, that’s a little bit hard to wrap their minds around that. What they’re really trying to do when they generate information is provide society with information to help basically a regulatory purpose, like fair lending, less pollution, or a better public health outcome.

Yet, when you go to the Starbucks… I don’t know what the regulations are in California but a lot of states and jurisdictions have regulations that require restaurants to put calorie levels or sugar levels on their menus. What that’s doing is the government requiring Starbucks or some other food purveyor to tell the consumers what kind of calories their food has.

It does a couple of things. It enables the consumer to eat in a more healthy way, but then it also encourages the restaurant, or people selling food, to put healthier food on their menus.

One of my favorite transparency regulations comes from Los Angeles County. I don’t know if you’ve eaten out in Los Angeles lately. It started there, and the government requires restaurants to put on their front door a letter grade that’s A, B, or C. The letter grade is the result of the health and safety inspection rating that they last had.

So, you walk by a restaurant and you can see if it’s a C restaurant, the kitchen’s not very clean and they’re hygienic practices aren’t very good. You can keep walking and you walk by an A restaurant, and the inspectors have determined that their health practices are really good, so that encourages people to eat in A restaurants and avoid C ones.

In turn, people who run restaurants start paying a whole lot more attention to their health and safety practices because losing a bunch of business from a bunch of folks is much more significant than getting slapped with a fine.

Katie: We actually have that disclosure of calories on donuts in Safeway. There’s actually a big warning sign because donuts are so high in calories. It’s really funny! I actually photographed it and put it on Facebook. It said, “Warning! Donuts are unusually high in calories and can be over 300 calories a donut.” I looked at them and said, “Wow, that’s so small!” It was different than any other food; It was only the donuts.

You explained the relationship really beautifully of the relationship with the government providing a minimal amount of legislation that allows the consumers or citizens to then create that change.

One thing I think that disclosure does, because I’m really interested in studying gamification and behavioral psychology, going back to how much of it is just an internal pressure on the discloser, like the restaurant or the corporation, how much of it is external… I think that part of the reason why it works is that we’re all fundamentally socially comparative creatures.

We have this concept of social norming where we all want to be kind of in the middle, or a little bit better than the middle, and nobody wants to be the worst. I think that’s really core to the human psyche, for us being accepted. Nobody wants to be the bad one who’s left out.

I think there’s these restaurant-grade cards, the Toxics Release Inventory, and all the other star ratings just has this incredibly deeply psychologically powerful effect. Even if nobody ever sees the restaurant rating that you mentioned, I still think as a restaurant owner when you get a C-grade, that makes you feel bad and that influences you to change so you can be in the upper half of the pack rather than the lower half.

Archon: Oh yeah! Even if you’re not afraid of losing business, you’re saying you’ll probably change anyway because you don’t want to run a restaurant with a dirty kitchen.

Katie: I just think people really don’t want to be doing worse than everybody else. I think it’s really easy to lean on market mechanisms for explanations for things, but I don’t think that’s the whole of what’s going on with people.

Archon: I agree. There’s a colleague of mine here at the Kennedy school named Todd Rogers. He’s a psychologist, an organizational behavior person, and he studies a lot of these mechanisms to try to get people to make better decisions. He studied things like in California in your energy bill now, a lot of people in a lot of states get a bar graph that says, “Here’s how much energy you used in the last month, and here’s how much energy your neighbors used.” You feel bad if your bar is way out higher than your neighbors for the kind of reasons that you outlined. It does encourage people to save energy, to use less electricity.

Katie: Yeah, it really does. All of the research around that is really interesting. I’m really curious to see how more of these data sets rolling out have this almost intrinsic motivation with people that is not pushed by these external forces, which brings me to the nexus of behavioral economics and disclosure, which you naturally moved on to. For anyone who’s listening that doesn’t know what behavioral economics is (behavioral sciences) it’s basically the behavior of how people function in terms of their purchases and actions.

It seems to me that it’s an incredibly natural continuum of disclosure to then move into the behavioral sciences. Once you put the data out there, you graphically design it so it works, and then you have to get people to do the thing, like not eat the doughnut or use less electricity. Have you seen these two professions starting to converge or emerge something new?

Archon: Yeah, I think this is a really interesting discussion. I think that there’s actually kind of a tension. Sometimes they go together, but sometimes they pull apart. Some of the design work that you’re very interested in, a bunch of the behavioral psychology and economics is, insofar as it intersects with policymaking and people, trying to make the world a better place. A lot of the work is in trying to design nudges.

Cass Sunstein, another colleague of mine, wrote a famous book with Richard Thaler called Nudge. For your listeners, if you’re interested in behavioral economics and designs to affect behavior, Danny Kahneman’s Thinking Fast and Slow is another book that they might be interested in.

A lot of nudges are designed to you get people to do things that are good for them. You go to cafeteria in a Harvard College dormitory, and the fruits and vegetables are in the front, but the chips are under the counter so that the kids will eat the fruits and vegetables because that’s more available. If you go to get a driver’s license aa the Motor Vehicles Division in many states, the organ donation is an opt-out rather than an opt-in, so the default if you don’t do anything, you’re on the organ donor registry. That’s a nudge because society is a better place if you donate your organs because somebody else can use them.

Now, transparency oftentimes operates in a much different way. Instead of nudging you, transparency, I believe, is oftentimes designed to get you to think more deeply about your choices.

Danny Kahneman, many psychologists, and many people in advertising differentiate between system 1 thinking and system 2 thinking. When you drove to your co-working space today, you probably didn’t think much about how to get there. You probably just took the route that you always do out of habit, and that’s system one thinking. You don’t really think about how to do it. It’s very efficient.

When you go to the grocery store and buy breakfast cereal, I bet you, me, and most of the listeners out there do one of two things. You buy the cereal that’s at eye level or you buy the cereal that you bought last time. People in grocery stores know this, and they charge manufacturers more to stock the cereal at the eye level because that’s a powerful nudge.

Whereas the whole point of putting the nutritional label on that cereal box is to try to get people to look at it and push them into what psychologist call system 2 thinking, which is the part of your brain that really thinks about things and weighs the cost and benefits and decides which decision to make based on a little bit more deliberation and a little bit more thought.

The last time you bought a car you used system two thinking rather than system one thinking. Maybe you shop around a little bit. Maybe you went for a test drive. Maybe you went to the Consumer Reports website and read about reviews. That’s system two thinking, which is totally different from buying breakfast cereal, which is system one thinking.

Katie: I’ve read those books and they were actually in my list of questions to ask you, so you’re ahead of me answering before I get to the questions. I’ve never heard it explained that way in terms of nudge being system one and disclosure being system two, if that’s the right way around. Did I get it the right way?

Archon: Yeah. I get it a little mixed up myself.

Katie: You don’t thinking disclosure is a nudge? I thought disclosure was a nudge.

Archon: It can be a nudge. Cass Sunstein and Richard Thaler in their book they call transparency a kind of nudge, but for the reason I gave, I think it oftentimes is not really a nudge in that way, and that it’s designed to engage your conscious, your more deliberate decision-making apparatus. The slower [thinking] that takes into account pluses and minuses rather than getting you to decide by putting you in a habit group.

Katie: That’s a really interesting way to explain that. One thing that they do have in common, which is the concept that’s brought up in the Nudge book, is of libertarian paternalism, which I had never heard of until I read that book.

Disclosure sort of fits into that a bit as well, which is all about keeping the government regulations on the small side of things, which is segueing from our conversation before. Having a lot of people who are involved in social environmental change. Being very legislation focused in their careers in their way of operating.

Taking those different lenses to keeping the legislation on the lowest possible end that we can possibly make optimizing for human psychology, then just letting people in organizations be free to do the right thing. How do you see this concept of libertarian paternalism evolving in this field?

Archon: I guess I think that you’re right that both transparency and many nudges have in common the desire to lighten the regulatory hand, so that governments a little less heavy-handed at least in some of these measures. I think that providing information has less of a paternalistic element than nudges do.

The cafeteria example that I gave before right of putting the fruits and vegetables in the front but hiding the chips and fries a little bit, so that people are more likely to buy the fruits and vegetables. That’s a paternalistic move because somebody, the person, the nutritionist, or the person laying out the cafeteria has decided that these students would be better off if they ate more fruits and vegetables and fewer potato chips. That’s a sense in which it’s paternalistic.

A transparency measure would be to put both the fruits and vegetables out there as well as the potato chips, and then provide some information about the health effects of each one, and then allow the person to decide.

There are mixed cases. I’ve seen an experiment in which they put up a sign that says, “If you eat this donut, you will have to run eight miles to work off that number of calories.” That provides information, but I regard that also as a little bit of a nudge. It’s a little heavy-handed, like who’s going to think about running eight miles because they ate a donut? I’m not sure eight miles is the right figure, but you get the idea.

Katie: Probably! Based on the very scary sign that was I saw at Safeway, I said, “We’re never buying these again.” I still bought my daughter one anyway, but she’s like only three. She can handle the calories.

Archon: I wanted to bring up a possibility that your readers may not have considered so much, which oftentimes is not intuitive to either scientists or policy makers, which is that sometimes transparency and information can create harmful effects.

When I was in undergrad, we had a technical and scientific background.

We were all, partly by culture and training, very enthusiastic about people having more data in information. Let me give you a couple of examples of when information can produce behaviors that some people might regard as undesirable, which is making the world a worse place rather than a better place.

One measure that we talk about in the book is Megan’s Laws. There probably is a Megan’s Law in California. These are laws that require people who’ve been convicted of sex crimes basically. The laws vary in different state, but if you’ve been convicted of sex crime, in many states you have to be in a criminal registry even after you’re released that reports your name, address, and sometimes even your face in a public database.

In the book, we come down on the side of thinking that Megan’s Laws are oftentimes not really a very good public policy because they provide information, but they don’t really guide useful action that much, and indeed may encourage harmful action, like vigilante law enforcement. People deciding on their own to punish these people that they see in the registry even though these people have done their time and served their penalty to society, etc.

That kind of information may enable and unintentionally encourage a kind of behavior that’s socially harmful.

Katie: Well, it seems that in that example, and the other examples they’ve been talking about, is the difference between the disclosure being around a human being and around an object, company, or particular metric, like temperature or chemicals. If it doesn’t seem to be around a human being, it seems to probably not cause that much of a problem. For example, this lake has chemicals in it, but if you were to attach the disclosure to a human…

Let’s say we made the disclosure how much energy or water do you use during a drought. If there was one house who really didn’t care about being responsible with water and wasting lots of water, you’re right! People might come and attack the property because it would be attaching the public disclosure to a human being. Do you think that’s what it is? The difference that makes it not work?

Archon: That’s interesting! I hadn’t thought about it that way. I think probably the risk increases when it identifies individuals as bad actors, and then the rest of society can pile on and sanction them or ostracize them in some way. That’s interesting. I hadn’t thought about it that way.

I guess the way I’ve been thinking about it is that for some kind of disclosure and transparency, you put the information out there in order to elicit an effect and a behavior, and therefore later on an effect. The effect could be a good effect, or it could be a bad effect depending on social consequences.

Another example that I think part of America would regard as a good effect, another part of America might regard as a bad effect, comes in immigration. This one is not about individuals.

In the initial months of the Trump Administration, the Department of Homeland Security released regular reports about which jurisdictions around the United States were not complying with their orders to hand over immigrants and cooperate with federal immigration authorities. This is kind of like the Toxics Release Inventory, but it’s identifying jurisdictions that the Trump Administration regards as bad actors because they’re not doing what the Federal Immigration Authority wants them to do, which is turn over immigrants that have been detained and maybe deport them.

There’s this list of bad actors. I bet San Francisco was on that list given the state’s and locality’s policies and general disposition with regard to federal immigration policy.

I think if you’re in many parts of California, you would regard that as a bad transparency regulation because it’s designed to basically punish non-cooperative localities. If you’re in the parts of America that supports tougher immigration laws and sanctions and more deportations, then you regard that as a good transparency regulation because it’s cracking down on you guys who aren’t cooperating with the right policy.

Data visualization

Katie: The core of what you’re saying is the design of the transparency system needs to be designed around eliciting an action, which I think another thing that comes up with having people who are designers, data scientists, engineers, the kind of people I hang out with, that I have noticed as well… basically, these are not the people who are to focus on policy. These are people who are very focused on data science… is collecting a lot of information and then maybe doing a really great project or really great map, then sort of forgetting about the bit where it has to actually influence a person and really tacking it on at the end.

I went to one workshop. It was to get feedback for a mapping project, and then they said, “Yeah, we’ve just spent basically a year and a few hundred-thousand dollars working on this really beautiful project and we’re going to put a call to action. We haven’t done it yet, but we’re going to put it on at the end.”

I see this happen a lot where people are very involved in a process and really forget the final phase, that you actually have to get a human being to do a thing. It’s kind of what you also bring out and making sure that it’s the right thing. Is it a bad thing that you’re going to accidentally get someone to do, but the thing is going to improve everybody’s lives?

Archon: I think this is a huge vulnerability on the part of engineers and engineering, and not design thinking but engineering thinking, in a way is that it’s so easy to get the data, process the data, and generate a beautiful map. You can do that within the skill set of a software developer, an engineer, and data scientist.

Then, as you said very well, there’s a whole other step of making something good in the world happen with that data, and that’s a whole other skill set that has oftentimes very little to do with engineering. It’s about organization, then it’s about public management, or it’s about activism, or it’s about a whole different set of skills.

To make it work, I think you need multiple skills that marry to one another, of which being a great engineer is only a half or a third. I think that’s oftentimes missed in these kinds of efforts.

Katie: That’s what I really want to see happening in the movement. I want to see all of the Silicon Valley type engineers and data scientists getting together with action designers, behavioral scientists, people who specialize in all these techniques of getting people to act, with not-for-profit people who are engaged in the causes, and then coming together with these other skills. I think that’s really where the magic is, or where the kind of cutting edge is of social change right now. It’s getting all these skills together that have been very separate.

Archon: Yeah, I agree.

Katie: I wanted to ask you about the difficulties in getting a good transparency legislation up and running. It seems to be that there’s a whole lot of low-hanging fruit in this space that I think could be really useful. I was talking to a gentleman a couple of weeks ago. He had a senior position for in a government energy authority in California. His whole history had been involved in energy efficiency legislation, and I asked, “Have you heard about this field of disclosure?

What I think you should do is don’t pass a law that mandates specific layers of insulation building codes. Pass a law that mandates the disclosure of the amount of kilowatt hours that each commercial building is using at any one time, so it’s not a secret. I have to put it out there.” He said, “That actually came up a few years ago.”

I tried to pitch it as hard as I could because I think this is a great thing that needs to happen. At the moment, it’s completely invisible what any different commercial building is doing. He said that there was a lot of pushback. He was like, “We’d get a lot of pushback to that. They do not want to disclose their energy consumption.”

Archon: The building managers, the real estate corporation, the people who run the buildings basically.

Katie: Yes! If we could basically have the success of what we had with the Toxics Release Inventory to commercial building, so groups like the Green Building Council can really get to work mapping out and looking at this data. How do we get past this block? How has it gotten passed before? How can we pitch it in a way that’s positive? That’s going to get all these building people excited instead of angry.

Archon: I say this with the utmost affection and respect. I think another blind spot for many people coming from an engineering and scientific background is that, and I myself am in this category, is that it’s a conflict averse kind of type. It’s a little bit of blindness to politics, and the kind of conflicts that can arise in politics.

I do think that transparency regulations and requirements that do a substantial amount of good in the world always hurt someone. Those people are oftentimes likely to recognize that beforehand and resist the disclosure.

Think about campaign finance disclosure, requiring politicians and candidates to tell the public and everybody else who they’re getting money from. We’ve got some of that, but expansions of direct disclosure regulation in politics have been resisted very strongly because both donors and politicians are hurt by the public disclosure of that information.

Similarly with the restaurant report card example that I gave, in a lot of cities that’s not really spreading because restaurant owners and restaurant associations are organized to fight that kind of regulation.

Similar with the building energy use disclosure that you were talking about and proposed. It doesn’t surprise me at all that people who run these buildings would resist that regulation.

Katie: There’s another really good example that wasn’t in your book that I’m interested in your thoughts on which is, in Australia, we have these star ratings on appliances. Have you seen them? They’re like a sticker and they’re quite bold. You get like a five-star or seven-star fridge. We have them for water appliances, gas, and electricity, and a standard disclosure model.

From what I remember, it came in sort of before my time, but there was a huge amount of pushback from the appliance manufacturers. They got it through, and then a few years later now, they all love it! If you tried to take it away, they now see it as one of their main marketing agendas.

I think the same thing happened with the catalytic converter that came in the 1950s when cars were really polluting. They would come out saying that enforcing the catalytic converter would be the death of the auto industry. Now, does any automatic manufacturer want to get rid of their catalytic converters? No.

I’m wondering how we can pitch the disclosure legislation to inspire the potential disclosures to have the same experience of what happened with the energy star ratings. To say, “It’s not going to be the end of the world. You don’t need to have an epic tantrum about it. This will be a much-loved marketing tool on the other side.”

Archon: There are several ways that have worked in the past, and we do see a lot of disclosure requirements and regulations out there, so it does happen, then your question is the right one. How does it happen? I think there’s several different pathways

One pathway is that these disclosure regulations oftentimes happen after some disaster or crisis happens. A bunch of years ago, late 1990s or early 2000, there were a series of rollovers from tire defects in autos. It became a big deal, and then one thing that grew out of that was the rollover rating of cars.

Now, new cars have a rollover rating that says the percent chance that your car will roll over if you lose control. That’s a disclosure regulation that’s really important. I think people pay attention to that, and cars are less likely to roll over because people have made manufacturing changes. That was made possible because people expected a response, even car manufacturers, I think, expected a response to this disaster that happened, so that’s one pathway.

Another pathway is to get some of the people on board that will have to disclose, and they’ll look good because they’re actually doing pretty well. On the energy efficiency ratings that you were talking about, I think a bunch of buildings in San Francisco, or in other places, would actually look pretty good because they’re green buildings, so maybe begin with the natural set of allies. Say, “Hey, you actually want this because once it happens, your hard work will be publicly recognized and rewarded.” That’s another pathway to getting some allies and some friends for the disclosure regulation.

Katie: I can see that if you could privately get the data somehow, and then take the best-performing 50%, befriend all of them, and get their buy-in. Push it through to not let the laggards in the industry not cause you too much trouble.

Archon: There’s a third path, which is that we live in a big country. There are 50 states. States do a lot of regulation. A lot of companies that operate at the national scale, it’s a nightmare for them if there are 50 different state-level versions of some disclosure regulation. They’d rather have one, obviously, that covers the whole country. If disclosure starts popping up at the state level, and it looks like different states are going to require different things, then some manufacturers and companies begin to favor a national regulations, but at least they’re all playing on one set of rules that’s much easier to manage.

Some of the nutritional labeling has this quality. I don’t know if you’ve been following this, but there’s been a long debate about whether to add one line to the nutritional label that says “added sugar” in addition to the amount of sugar that’s in there. It’s really added sugar that you care about. Bananas have a fair amount of sugar in it but you’re not too worried about that. It’s the chocolate chip cookie that is the added sugar. It’s been a big debate about whether to add that one line, “added sugar.”

Katie: I think we might have that in Australia. It feels familiar to me to differentiate the two. I can see how telling the concept of simplicity over complex, might be interpreted as draconian, legislation would be a really good sell.

I used to do LEED, or Green Star we call them in Australia, certifications on buildings, and even being a very pro-environmental person, working on those buildings often the environmental guidelines that were given to us by localities were just terrible. They were badly designed. We couldn’t design for them. Just trying to create a design framework for architects and engineers to make an environmentally-friendly building, it’s just incredibly hard to do a one-size-fits-all.

You’ve got planners trying to basically tell you that the window needs to be this high, the eaves need to be this long, the insulation has to be this thick and the building has to face this particular direction. It’s just crazy, even for a very pro-environmental person. It was actually technically impossible sometimes. You could imagine people who are not interested in the environment, they just hit the roof. That’s an absolute nightmare for them to get legislation that specific.

If all they have to do is just disclose the kilowatt hours in real time on some website, maybe some digital screen developers can figure out how to put it in there, and you sell them to it that way, it seems that that would be like a relief for them. They would say, “Yes, please. Let me just disclose the kilowatt hours and not have to have the exact width of my windows described to me to my architects that’s impossible to follow.”

Archon: Yes, absolutely. It may end up being more if it encourages a race to the top, which is the goal of a lot of these kind of comparative disclosure and information systems.

Katie: The race to the top. It’s a great way to put it. I love the idea of leaderboards and social comparison. People trying to get up a leaderboard in their performance, getting into this core of this social norming, social beings that we are.

I wanted to ask you about the concept of environmental literacy as well. It’s a phrase that’s not used very often, but it really fascinates me. Being an environmental engineer and being really obsessed with environmental data, basically it’s invisible. I have no idea how much trash I make. I have no idea how much water I use, how much energy I use. Kind of on the electricity bill, but not in any real time. The impact on how many chemicals I use.

Even if I try really hard to figure it out, it’s still really hard to figure out. How do you think we could increase our environmental literacy? Or how would that come together with the concept of disclosure and transparency to get us to be better citizens and drive change?

Archon: I don’t know who’s working on this.

Katie: Not enough! I am!

Archon: Imagine you had a Fitbit that was your carbon footprint for the day. That would be an amazing thing. I mean, it would be impossible to do. Well, maybe not with the amount of sensors and other things that are developing, but I think that kind of quantified-self, but this would be more a quantified-social-self, or a quantified environmental-self, would be extremely effective.

I think one problem with it is that it’s only people who care a lot about this sort of thing that would pay attention. It would take a while for it to diffuse more broadly as some of the health technology is only interesting to people who are already pretty healthy, so it doesn’t target the people who would most benefit. I think that those kinds of measures would be very effective, or I don’t know how effective they would be. Let me withdraw that a little bit.

I think it’s really worth investigating and developing. A friend of mine is involved in a company that is trying to develop social, environmental, and health metrics for different products, so what if on your baby shampoo, or whatever it is, you could tell very easily different health features like how they treated their workers or an impact on the environment. That might affect your decision-making around that that kind of thing. The very availability of the information might increase social literacy and environmental literacy as you’re putting it.

Katie: I hope to see the field evolve more. I’ve been working on a number of projects in this space at the moment and the hardware hurdle, it’s real! You have to actually build a sensor that’s going to collect the information out of electronic parts, and then get funding for it, then roll it out over a big industry. There’s a whole lot of potential things could be done.

For example, if you were trying to measure storm water drains in a community to figure out which drains were the worst, you would have to actually build a sensor that would sit in the drain, then it would have to work and not break, be attached to Wi-Fi and get electricity, and then send it out. You would have to have multiples of them, maybe thousands of them, across the city.

A total idea could be done but someone actually have to sit down and make it with a soldering iron, deploy it, fund it, and get it to work, but I’m hoping to encourage and inspire more ideas like that through these conversations to help bring this environmental literacy out into the open, and different types of startups or technology that we could build to encourage it.

Archon: Maybe you thought about this, but one thing that people have developed already is Nest does this thermostat. There’s appliances that give you some kind of feedback, that we didn’t have just five or ten years ago, that is designed to make you a more self-conscious user of electricity, air conditioning, or whatever it is. What are the next steps in that domain that are self-nudges in a way?

I do want to be more environmentally conscious and have a smaller carbon footprint. Just like I wear a Fitbit, I believe if I had more feedback that would help me accomplish that end.

Katie: There is a whole bunch of different papers get published in the Journal of Environmental Psychology. My favorite academic paper to read. There’s others as well. I’m not to say that I’m a complete comprehensive expert in it, but every single indicator is different.

Air pollution is one thing, and then air pollution is also made up of at least five, if not ten, different indicators. Water is different. Some of the other people I have been interviewing on the podcast are actually entrepreneurs who work in these companies.

Aclimar for air pollution. WaterSmart was the last podcast we just put out. Planet Labs getting satellite data and then using different image processing algorithms to actually turn a photograph from a satellite into data, so you can see that change. Another company called SkyTruth does this with fishing boats or with mining. You can get a satellite image of the mountaintop mine and then rather than just having human beings to look at the photos, they can actually build an algorithm that cuts up the image pixel by pixel and can tell whether it’s a tree or whether it’s a coal mine, and then it can tell how it’s growing bigger. In waste, there’s company that’s putting load sensors on the arm of the truck that I did some development for.

There’s no one-size-fits-all. Your environmental footprint is made up of so many different indicators from different industries, different economic drivers, so they’re all very different and require a completely different startup, different group of people, different funding mechanism. To put them all together, there’s not just one thing that describes the whole thing, but it’s a really fascinating space and I’m ever fascinated by it.

One other thing I wanted to ask you though about the design. When I event with your colleague Eleanor in Boston, she mentioned something to me that really resonated with me as a designer, which was when we’re looking at how we display information, break it up into three levels of easy, medium, and difficult. This is still falling on from this environmental literacy debate.

I think if you’re really into environmental data or you’re a designer, you likely want to design some crazy infographic. You can see these infographics now that just have like hundreds of little graphs, a little flow charts, and it looks like wallpaper. I’m like, “Well, that makes really cool wallpaper,” but it takes three hours to figure out what it means.

Creating elaborate graphs that are really advanced things like d3 and JavaScript. That kind of thing is cool in a studio in San Francisco, and again, completely forgetting that it has to actually influence a human being.

I fall into that category too. I tend to want to do complex things and have to pull myself back, and be like, “Make it simple.” She said just using things like display your information in A, B, or C, a star rating, just color, and not lots colors, maybe just three colors or five colors. If people want to know a bit more, then they can go to the next layer and they can learn a little bit more.

If they really want to learn a little more than that, then you can give them the full hog of all the technical data. What can we do to learn from this, to help make our disclosure work into these different levels of information design?

Archon: I think that’s a great point. First of all, there should be those different levels of data available at different levels of complexity, so I agree with what you said. There’s however many layers of complexity for users that are interested at different levels. Some people may just want a yes/no decision, a stoplight, or something like that, whereas other people may really want to drill down and see the individual data points. That’s the first point.

The second point is that the right way to think about these information systems is that there’s all kinds of organizations involved in producing the data and displaying the data. One challenge I think that faces people working in government when they’re putting the data out is that they feel like they are solely responsible for the soup to the nut, generating the data, making sure it’s high quality, to putting it out there, to making sure that the display that the end-user gets, that they designed that properly. I don’t think that’s how it works.

You probably looked at the weather yesterday or today, but you probably didn’t go to the website of the National Oceanic and Atmospheric Administration. You didn’t go to that agency. You went to or something like that because information systems are an ecology of different kinds of organization.

Oftentimes, a government that’s producing the data, but all sorts of intermediaries are packaging the data. It’s important because those intermediaries have a lot more information about what end-users are actually going to want. The FAA collects on time and delayed arrival data for every flight that’s out there. so if you’re flying to Boston on a United flight, you can find out the on-time percentage, but you’re not going to go to to find that data. You’re going to go to your flight tracking app or your airline website, because you go to your airline website to find all the other things that you need to know to get on that airplane and fly successfully.

That’s the second point is that it’s a whole ecosystem of all kinds of different organizations that are generating and packaging the information, so recognize and embrace the division of labor. Don’t try to squeeze it all into one organization. That’s too much, and it won’t work.

The third point is that as a designer of either policy or information, you should have many different kinds of users in mind. If you’re designing a piece of environmental data, like pollution data, like toxic use that we discussed before, one kind of user is the individual, like the homeowner, the person who’s thinking about buying a house in that neighborhood or whatever. Designers and policy makers have that kind of user in mind, an individual end-user, but for a bunch of these information transparency systems, some of the most important users are organizations with more sophisticated analytic capacity.

Think about financial regulation. These 10k forms and all of the corporate data and financial data that all corporations have to report and generate. Some of the main users of that data are professional financial analysts working for pension funds, investment companies, or whatever it is. For the whole system to work effectively, you have to provide data not just to the end-user but also really detailed disaggregated data for professionals and organizations that are going to crank it through.

Katie: There’s this technique called “user story mapping,” or behavior mapping. It’s used in software development. it’s really common if you are designing in the web application design world, but it’s one of those things that nobody’s ever heard of it anywhere else. It’s something that is just so critical in the process to actually just have a user, figure out who they are, and then map their daily life. How they move into contact with your particular software and how they’ll use it. When you do that process, it really unveils a whole lot about the design process.

Anyone listening look up “user story mapping.” “User behavior mapping.” It’s an incredibly useful tool that does not need to be trapped within the world of software development.

I really like what you said about the government not needing to do it all. What can totally happen with data is that the government can be involved in all the gathering of the data, can create a database in an API, and then independent developers can then build on top of that. You don’t need to just have one version of, for example, the weather or whatever environmental literacy app it is. You can have ten different competing application developers doing different variations on what it is they do, then they just need to call the data from a database that has that kind of data, and then you can have the creativity and the market mechanisms. Maybe they can host a design competition, give an award try and encourage different players who are really experienced in the front-end design calling just the database that the government is involved in funding. I think that’s the way that things should move toward.

Archon: I really agree. It’s been one of the real revolutions in this space, it’s probably a tiny little revolution, is in the transit space. The Muni in San Francisco, or the MBTA in Boston over the last five or ten years, many of these agencies now put the GPS feeds on trains and buses out so application developers can write apps that show you how many minutes till your bus get to your stop.

For a long time that innovation was really blocked for a whole bunch of kind of bureaucratic and organizational reasons people in the transit agencies didn’t want that information to really be out in the public. The GPS units were already on there. The actual cost of making it public was quite low, but I think they were probably concerned about all of the bad things that could happen in terms of public pressure, a loss of control of the information.

I think once a few transit agencies started doing it and people in the cities I think really liked it. When I ride the bus, I don’t care about the schedule anymore because I can see exactly how many minutes till the bus gets there, so I have a cup of coffee or whatever until it’s time to go. It’s an example of what you’re saying. The transit agency, all it does is flip on a switch to make the GPS locations public and then a dozen app developers in every city can write the app to say, “Where’s my bus?”

Katie: it’s a great example. It’s a way of thinking that really needs to spread more, and that’s exactly how I got into this space. I started to design apps that I

wanted to build. I wanted to design a thermometer that would go on your Facebook page and show you how much energy you are using in real time. It was my first idea like five years ago, then I realized that I couldn’t get it. I decided to go through all the different metrics, and I’m like, “I can’t get any data anywhere!”

It’s just really easy as a developer. You can do a request to an external database. It’s a simple thing to do, but I think the people who are involved in getting this data, if they realize that there’s this real hunger in the Silicon Valley startup world to do creative things like that, and if they make it available, there will be plenty of people like me and my friends who will want to start working on projects that use that data, so I want to see that happen.

That’s a big exciting change. It’s been done in the transport industry, you said, in some cities. Let’s hope it keeps happening.

if there was a utopian world, the best possible scenario of the world in 100 years, where all of these issues that you study felicitated as beautifully as possible, how would it look? How would it work?

Archon: I did write a paper a few years ago called Infotopia.

Katie: I read that! I couldn’t find it again. I didn’t want to ask you about it because I couldn’t remember it. There was something about future utopian government data…

Archon: Yeah, I think it’s really more for an academic audience in some ways rather than a popular reading audience. I think that how we should think about these things is in terms of the risks that we face and that are most important to us to live happy, fulfilling, and healthy lives, and then think about the kinds of information that we need to navigate these complex information scapes.

Also, in addition to decision making, there’s an accountability dimension. What kinds of organization are going to really help us live those healthy and fulfilling lives? Or if they act badly, create risk for us and create problems for living healthy and fulfilling lives? That should guide society’s efforts the public efforts, government’s effort, and corporate effort to make information available.

In the future world, it would be good if we had a shared sense that in order to hold companies and other actors accountable on one hand and to make the really important decisions in our lives, we need a lot of information, and we need to demand that information from governments and from companies alike.

It comes out in easy examples. I think food risk information. Every year, there’s information about contamination of some sort and a bunch of people get sick, or health problems in hospitals or medical care, or financial risks. We saw a big financial meltdown.

There’s these big actors out there who create risks for us. One way to control that risk is to have a lot more information from them about what they’re doing. In a positive future, we’d have all of that information, we’d have organizations, and as individuals, we’d be literate in using that information to drive these government organizations and private corporate organizations to higher levels of behavior in a race to the top that would help us live, rich, fulfilling, prosperous and healthy lives that we want to live.

Katie: The way I see this upward spiral of what you’ve mentioned is that once we can have better information and we can solve these difficult problems that really bring us down, we can almost move to another level of our own enlightenment or consciousness.

When people are almost moving up Maslow’s hierarchy into the self-actualization at the top, we tend to get, I think in social change, really focused on these polar bears dying and kids are sick. We’re very problem-focused, orientated people, but why solves the problems?

Well, we solve the problems so we can reach a new phase of our growth as a civilization. That’s what I see sort of like the next step, which is maybe if just a fancy way of saying what you said which is healthier and happier lives, but when you’ve solved all of the most significant problems in your life and in the world, that’s been sort of a trampoline to jump into a new phase of enlightenment in humanity.

Archon: Good! I really hope we get there sooner rather than later.

Katie: It’s a big fancy way of thinking about it.

When you wake up in the morning, if there’s one thing that gets you really excited or that keeps you awake at night in a positive way, like one particular angle that is really gets you about your work, how would you describe that?

Archon: I think all of my work in information, but also in kind of politics and policy space, is about trying to figure out ways to enable people to participate more fully and more effectively in determining their own fate. In the transparency and information world, we’ve just been talking about getting people information that they need to make better decisions to determine their own fate. In the politics and policy world, it’s about creating avenues of engagement so that people can make the rules of the game and organize their lives together in ways that are empowering and fulfilling.

That’s what really drives me is trying to be attentive to the possibilities and innovations all around the world and that are open to us, but that we haven’t quite grabbed on to yet that enable people to be more empowered both as individuals, but especially for me, socially and collectively. That’s what really gets me up in the morning is trying to find out what those opportunities are that we haven’t quite yet seized on to and try to give voice to them, draw attention to them, create conditions in which people can really understand, explore, and develop those opportunities.

Katie: It’s a beautiful way to look at it. The way you’ve described it as seeing that there are many opportunities to change out there that we may have not seen.

In neuroscience, they say that we only perceive a very small amount of the universe, or of all the information around us, through our senses, that our whole consciousness actually prunes off. It’s got a huge job pruning off information, so we can focus. I just thought that was a beautiful metaphor for looking for opportunities for social change.

Maybe it’s the same thing. Maybe we’re getting tunnel vision, and there are actually hundreds of different other opportunities that we are not seeing we read your book listen to what you’re doing, then we can start to see all of these things that we maybe aren’t seeing around us.

Archon: Yeah, and I think if more of us try to look for those things, we’ll find more of them.

Katie: What a beautiful sentiment to end on.

You can get a copy of his book. It is called Full Disclosure on Amazon.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

** Support the podcast on Patreon
** Download free tutorial guides on how to be a world-changing super star at
** Learn more about the video course “Save the World with Gamification” at

Follow me on the social medias
** Facebook
** Instagram
** Twitter
** Pinterest