MIT Sloan professor Roberto Rigobon and Harvard University professor Alberto Cavallo are the co-founders of the Billion Prices Project. The project collects prices — it has around 15 million prices today — provided by online retailers around the world. The public data is used to conduct research in macroeconomics and international economics. The professors say there are many ways to measure life, and they’re helping fellow researchers access the data to do it.
Why online data is important to your work
Rigobon — “I think that we measure life in practice very badly. I mean, when you think about our relationships, how do you know that a relationship is in bad shape? Usually when you break up. How do you know that a lake is polluted? Usually when you lose a lake. How do you know that society, like right now in Venezuela [this podcast was recorded in 2017], my country, is upset? Well, the day they show up and riot in the middle of the street. So, sadness, we measure sadness too late. It's usually when it reaches depression.”
Data that matters
Cavallo — “I started collecting these larger data sets in Argentina and producing an inflation index that was trying to sort of compare to the government. And then over time, I eventually wrote a paper that was trying to show that the [Argentinian] government was lying. In that paper I tried lots of different alternatives of ways that you could think the government might have been producing their index. I focused on goods that had the lowest exposed price increase. I looked at the goods that were in the price controls. It always getting higher numbers and I realized that the only way to get an approximation to what the government was doing was simply to take the real inflation rate and divide it by three. So, it wasn't a particularly sophisticated algorithm.
“But eventually I think this sort of work made us realize there was a potential of using online data. Not only in a country that lied about inflation but potentially in other countries that have goods ... where the traditional data collection method has some limitations and, you know, this work eventually led to the creation of the Billion Prices Project that we started together with Roberto. Ever since then we've been exploring ways in which online data can be useful for this type of measurement.”“Now we're working a lot on not only inflation measurement, but building baskets of identical goods that are priced across countries, and that's what we called PPP indicators or related price indicators.
Rigobon — “But this, as you can tell, it's a very unsexy project. These are economists trying to measure inflation and, you know, nothing can be more boring than that. Then we're doing marginal improvements. It's kind of a freak of a project because it's a project about economics that is truly relying on computer science tools to be able to do it and is based on statistics from statistical offices and procedure.
“So, it has been interesting when you go and explain this to people because they first look at you saying like ‘why on Earth are you doing this?’ So it was interesting to understand the impact. That's why I think it's interesting; it touches so many different audiences. People from big data are interested in understanding how we do this. People from economics like the outcome. People that use the statistics want to understand how to use it, but it took a while.”