Senior Spotlight: Pedro Farias
What’s your thesis about and what were your findings?
I think the simplest way to understand my thesis is it’s asking a question that has two parts to it. Does public transportation help save lives and help local businesses? That’s basically it. Let’s start with the save lives part. So, one of the questions that the thesis asks is does public transportation reduce the number of car crashes, and by doing so, since car crashes are fatal especially in Massachusetts which has a higher proportion of fatal crashes than most other states, then you would be saving lives by reducing the number of car crashes. And a bonus is that you also save costs because car crashes end up being very expensive. That was one thing I was looking at, whether this increase in public transportation essentially led to a lower number of car crashes.
On the business side, I was looking at whether businesses benefited from this increase in public transportation, whether they found an increase in business activity as well. For that, I had to get a bit creative, so I used Yelp and Twitter data to see whether the businesses that were close to the increase in public transportation supply actually saw an increase in Twitter and Yelp usage. Those are the two main components of my analysis. We’ve seen on the business side that the number of tweets and Yelp reviews went up--for tweets it went up about 4% and for Yelp it went up between 2-8%, and those are statistically significant, which means it's not likely for this to just be coincidence.
For car crashes, we see that there’s an average decline of about 4%. That’s statistically insignificant because when we looked at it on a more granular level, so looking at one square kilometer blocks within the city of Boston, you actually see that the areas with the most minorities and young individuals benefited a lot more from car crash reduction, so there’s actually a 12% car crash reduction in those areas, and that reduction is statistically significant. It’s actually really interesting because it’s what we call a heterogeneous effect wherein we have certain areas that benefited a lot more. And those areas are where you have more young individuals and more minorities. So you can think of a way where your increasing public transportation in the areas where there’s the highest benefit is where you’re going to focus your efforts, essentially.
The actual increase in public transportation that I’m looking at is when the MBTA instituted late night service from 2014 to 2016. I’m looking at that service which extended hours for certain stops, not all the stops on the MBTA, into the evening. So with that, I’m able to use a few different empirical methods and economics to estimate basically the impact of that increase in supply of the late night service.
What made you interested in this topic?
The original thought was when I interned in Boston last summer, so I was going to work around 8:45 on the very cramped MBTA subway. And I thought, this is ridiculous, there’s no way the subway ever has to be this full. And so I thought first to design something called the linear program which would try to optimize the MBTA scheduling. But I talked to my advisor Ed Glaeser about that and we found that that would be number one, hard to get data, and number two, it’s probably not something that would produce empirical results. He said maybe this idea of optimizing the scheduling won’t work, but late night service is something the MBTA has been exploring recently.
I thought then that I could look at the history of late night service in general to get some of the answers to these questions with a lot more robust data. One of the good things about looking at the period from 2014 to 2016 is that in economics it's good to have these exogenous events--events where something happens, essentially, that isn’t necessarily caused by other things. In the case of late night service, we had two of those things: when it started, and when it ended. That way, I can test not only did car crashes go down at the beginning of the service, but also did they go up when the service ended. And same but the opposite effect with Twitter and Yelp--did they go up when the service began and down when it ended. So that would make my results even more robust. That way we’re evaluating it twice instead of just once. It turns out that both of those are true. We do see a downward trend in Yelp and Twitter after the conclusion of the service and an upward trend in crashes.
Does winning the Hoopes make you more interested in pursuing your field or topic of research?
Winning the Hoopes made me very, very happy and motivated me a lot. I am very likely to go to graduate school. And I wasn’t sure if that would be something like an MBA/MPP or something like trying to get a PhD. Winning the Hoopes definitely does make me think more about the PhD. I’m still not 100% sure because the PhD is a very significant commitment that requires a lot of effort as well, but it does make me think more carefully about that option.
What are you thinking about virtual Commencement?
It’s not ideal, but it’s very nice that the college is doing what they can to give us something during this crazy time. I’m excited to see everyone and get some kind of closure, at least in the shorter term, because I think that’s something my class is going to miss is the actual closure, the Commencement ceremony we would have had since we all left campus so abruptly. It’s going to be cool to see how this is going to play out. The one benefit I was thinking about is that Commencement is usually very hard to get tickets for all the events, so you have to be very careful about who you invite, but now I can invite basically anyone because it’s all livestreamed so that allowed me to invite a lot more people and recognize a lot more people who were involved in my academic journey.
What are post-graduation plans, and do they include Brazil?
In the short term, I’m not going back to Brazil. I’m going to be in Boston working for McKinsey and Company for about two or three years, we’ll see. And then my plan is to go to graduate school from there, TBD what program. But after graduate school, my idea is to move back to Brazil and use what I learned at McKinsey and whatever grad school to go back home. My vision right now is, when I’m back in Brazil, to focus on companies that are struggling and then improving those companies. That’s where the McKinsey part comes in because I want to look at processes. McKinsey has a very robust way of what they call RTS (Recovery and Transformation Services). So I want to look at those processes there. And post doing that for a while, I would definitely consider going into politics in Brazil.
Interview conducted by Kathleen Barrow, Brazil Office Fellow