Why Solar Adoption Can Be Contagious

In a new study, F&ES Professor Kenneth Gillingham found that solar adoption rates depend less on wealth or population size, and more on whether your neighbors decide to go solar.

Note: Yale School of the Environment (YSE) was formerly known as the Yale School of Forestry & Environmental Studies (F&ES). News articles and events posted prior to July 1, 2020 refer to the School's name at that time.

residential solar PV
Many Americans might not think of Connecticut as a hotbed for solar energy generation, but this New England state is actually well suited for photovoltaic (PV) solar installations. In addition to its high electricity prices, dispersed population with many suitable rooftops, and lack of other renewable sources, the state has also created policies to promote solar adoption. But adoption rates have varied significantly across the state — and they do not necessarily follow patterns of income or population.
In a recent paper published in the Journal of Economic Geography, Kenneth Gillingham of the Yale School of Forestry & Environmental Studies and Marcello Graziano of the University of Connecticut demonstrate that the most important factor influencing solar adoption may just be the choice of your neighbors.
In an interview, Gillingham describes the adoption patterns they discovered, why clusters of solar power tend to occur in certain areas, and what can be learned from these trends.

Q: Many people might be surprised that there really is a robust solar market in Connecticut. What is it about the state that makes PV solar an attractive option? 

Well, the economics of solar boil down to the price of the solar system, electricity rates, the amount of sun you get, and the incentives. Across the country there are similar prices for the solar systems themselves. But Connecticut has high electricity rates, a reasonable amount of sun, and reasonable incentives. 

Q:  But as you report, the adoption rates are not the same across the state. What are some of the patterns you discovered?

It’s very interesting. We actually find that the most populated areas in Connecticut are not necessarily the areas of highest adoption. In fact, neither are the wealthiest. It appears that areas that are suburban or rural tend to be some of the hotbeds of adoption of solar PV. Meanwhile, the lowest rates occur in the cities. 

Q:  Why is this? 

Some of this has to do with the fact that the homes in suburban or rural areas are owner-occupied. And some of it has to do with people’s preferences. People seem to get more excited about solar in suburban and rural towns. 

Q:  You suggest there’s a social element to it: People are more likely to adopt solar if their neighbor has made the switch. How important are these behavioral pressures?

It’s one of the main points of our paper. I had done earlier work in California that showed that if a person’s neighbor adopts solar in their zip code, or on their street, they’re more likely to adopt. This paper focuses on Connecticut, and it does it on a much more refined level. We look at adoption rates both over time and space.
We find that if a neighbor close to you installs solar, you’re much more likely to install than if a neighbor four miles away installs. That makes a big difference. Also, you’re much more likely to install soon after that neighbor installs, but there’s a dissipating effect over time. 

Q:  What were the sources of data for the paper? 

We worked with Connecticut Clean Energy Finance Investment Authority, which has data for all installations in the state that have received a rebate. There’s a financial incentive program, so if you’re a homeowner and you purchase solar you can receive a rebate in Connecticut. And we received all the data from all those installations — it’s anonymous of course —and we matched that up with census demographics and maps of Connecticut. 

Q:  Do these local incentives vary across the state or are they consistent? 

Well, they’re pretty consistent across the state. But there are also programs called “Solarize” programs that are very localized within a municipality or a coalition of municipalities. They are grassroots marketing efforts aiming to get the word out about solar. And they’ve been enormously successful, as we’ve seen in our results.

Q:  Do national data suggest that these patterns are consistent elsewhere in the country?

Well, no one has done a national-level study of this. But, they are quite consistent with the previous work I did in California. And interestingly, they are consistent with some work done in the UK. 

Q:  How can these insights be used to change behavior, particularly in places where adoption isn’t as great right now?

One way is through programs like the Solarize program. These programs — which are sponsored by the Connecticut state government and clean energy task forces in towns — are designed explicitly to leverage these behavioral influences. They designate ambassadors for solar in the communities who go around town and tell their neighbors about the benefits of solar, how it works, the economics of it. And through this campaign, they effectively leverage these “neighbor” effects to lead to more installations in those municipalities. 
Gillingham is also working on a Yale-based project, Solar Energy Evolution and Diffusion Seeds (SEEDS), which is researching strategies and messages that are most conducive to solar adoption.