An interview about a neat thesis on an issue that many of us from the Bay Area and similar tech-crazed regions are familiar with! Natanya is an Economics major with Business and Math minors. You can check out her thesis here if you want to know more!
What’s your thesis about?
How tech workers impact median household prices. I looked at this from an econometric standpoint, collected data from the Census Bureau, and then used a fixed effects model to analyze whether [tech workers] have an impact, adjusting for variables like entity and time and other variables that may affect median household prices.
Okay, cool! So, what made you decide to choose this topic, and is this the topic you started with?
I always had this topic in mind, I just wanted to go about it in a different way. I wanted to use a traditional hedonic regression analysis…
Can you describe what that is?
Yeah, hedonic regression is just looking at supply and demand side factors that affect median household prices, but [it looks] very specifically at parcel data, or zipcode data. So, very individualized data. And that’s actually why it didn’t work, because it’s impossible to get, which is super unfortunate. But… that method really helps look at the intrametropolitan patterns that may affect median household prices rather than just the larger picture. So that’s pretty unfortunate, but it’s fine [laughs].
So, from that, what was the process of changing your topic to something else?
Sure, so I just kind of had to change the question a little bit. So, originally it was: “how do tech workers in San Francisco/Bay Area affect median household prices?” And so, since I had to use a different model… I basically had to adjust for all the counties in my model, so therefore I had to broaden my question to: “how do tech workers affect median household prices?”
Gotcha. So, I was going to ask you what the biggest challenges were but… is that the biggest one? [Laughs] Or was there something else that was a surprising challenge?
Yeah, it was mainly just collecting data, finding it, changing the question, and then probably manipulating the data so it actually reflected my question a little better.
Cool. And, did anything in the results end up surprising you, or anything you found in the process?
Yeah, definitely. My results weren’t as significant as I thought they would be. What is basically found is that, for one added tech worker, there’s about a $4 increase in median household prices. So, of course, the bigger the tech worker number is, it seems a little more significant… But, that [lack of significance] is probably due to the fact that it’s not individualized data, so we’re not really seeing those patterns.
Well, if you were to build on your thesis and keep working on it, what would you pursue related to this topic?
Well, I’m actually going to do this in another class. I’m going to basically look at the Seattle market, so I’ll just be looking at other markets experiencing similar trends to San Francisco, or to California in general… Unfortunately, I wouldn’t be able to compare those models necessarily, because the variables would have to be the exact same… So I would probably just try to find individualized data.
Cool. Well, what advice do you have for students as they think about their theses next fall and… get excited? [Laughs]
Look for data before you try to find a question. Or just start looking in advance, because [data] is always going to be the hardest thing to find.
True. Awesome- thanks, Natanya!