The Effects of Pollution on Consumer Spending

The Effects of Pollution on Consumer Spending

The United Nations Global Pulse, an innovation initiative on big data and data science, and Western Digital recently announced the winners of the Data for Climate Action Challenge (D4CA) at the Data Innovation: Generating Climate Solutions event during the United Nations climate change conference (COP23) in Bonn, Germany. A winner from the Thematic category, a team of students from Yale University, investigated the relationship between air quality and consumer spending.

An unprecedented open innovation challenge to harness data science and big data from the private sector to fight climate change, D4CA was launched earlier this year and called on innovators, scientists, and climate experts to use data to accelerate climate solutions. Access to large amounts of data – anonymized and aggregated to protect privacy – accelerates the ability to spot connections, gain insight and develop predictive algorithms that can provide more precise direction and decisions. The Data for Climate Action Challenge demonstrates what’s possible when public and private sector organizations partner for social good.

The Effects of Pollution on Consumer Spending
Caption: The team from Yale accepts their award. Left to right: Dave Tang (Western Digital); Sophie Janaskie, Matt Moroney, John Brandt, Angel Hsu (Yale); Robert Kirkpatrick (UN Global Pulse)
WIRED Brand Lab: Tell us about your project

Matt Moroney: Our project is on air pollution and how it affects consumer spending. Traditionally when the cost of air pollution to society is estimated, it usually only looks at healthcare costs. We were wondering if there was an impact on how much people spent when it came to negative changes in air quality. So we wondered if people spent less on days when air pollution was high, and what effects that might have on local businesses and ultimately governments via taxes gleaned from consumer spending.

WBL: What dataset did you use?

John Brandt: We used a dataset from BBVA, which is a Spanish-based bank. It gave us access to its API – which provided us with completely anonymised data that told us essentially where people were spending, by how much and how old these people were, the transaction price and how many transactions there were.

We were given that for 170 postal codes in Spain, which represented about 150 million transactions over a two-year period from 2014-2016. We used the European commission’s air quality network, which had air monitoring stations within the post postcodes we had the data for. In total, we had access to 180 air quality stations, which gave hourly and daily readouts of air pollution.

WBL: What did you find?

John: For a 10 per cent increase in particulate matter, consumers in Spain spent between 20-30 million euros less every day. For every 10 per cent increase in ozone concentrations, consumers in that same area spent between 25-41 million euros less per day.

Data Driven Yale
Data researched included area of 170 postal codes in Spain.
WBL: Is that the result of people having access to air quality reporting, or is there something else going on? 

John: We’ve been thinking about that. Madrid does have forecasts that factor in air quality and air pollution, and the government actively urges people not to go outside or to drive. So for Madrid we can see a stronger relationship with spending and air pollution. But outside Madrid, where there wasn’t as much information, people appeared to be unconsciously making these decisions on their own.

Matt: The truth is, if you can’t see across the street, are you going to take your children outside or walk to the shops? Intuitively that made sense to us.

WBL: That’s pretty significant. What else did you find?

Sophie Janaskie: Sometimes as we looked through the data, we could see big anomalies on certain days. So we’d google what happened on those days and it’d be things like Spain won a World Cup match or a national celebration.

Matt: There was also one where a bunch of people in the government got arrested for corruption and there were a series of protests, and no one spent money because things were closed. On New Year’s Eve we’d see a huge spike in air pollution because people were letting off fireworks.

John: You could tell when Spain won soccer matches because spending went way up that night.

Data Driven Yale

WBL: Before you started, was this relationship between air quality and consumer spending already established? 

John: The World Bank has a report on air quality impacts on general health, where they focus on asthma, respiratory illnesses, death and displacement, but they don’t look at consumer spending.

WBL: Where would you like to take this method of looking at anonymised credit card data?

Matt: We’d like to scale up to a city that historically has always had bad air, like Mexico City or Shanghai or Singapore. We’re in preliminary discussions to work on a project like this in Mexico City.

John: Our vision is to make a live, updating map that tracks spending data with air pollution and give a real-time account of what’s going on. We’d then like to have an increasing number tallying the cost to economies worldwide as a direct result of not doing more to improve air quality.

WBL: Are private businesses aware of how much can be deduced from their datasets?

Matt: Not that we’ve seen. I think that’s going to be one of the biggest challenges of this century. The 20th century was about collecting data, this century is about how are we going to use that data to make better changes.

This is just one example of another cost that wasn’t included in the conversation about climate change. Many people have argued cleaning up pollution is too expensive, but this proves that there are unseen costs that we need to look at more closely at before deciding we can’t clean up the environment.

This content is produced by WIRED Brand Lab in collaboration with Western Digital Corporation.