Economists Need More Jiu Jitsu
My first day of Brazilian Jiu Jitsu (BJJ), I spent nearly the whole lesson with my face in a man's crotch. The few breaks I got consisted of having his face in mine.
Since that first day, I've realized that BJJ is a lot like YouTube, but neither is very much like being an economist.
The move we were learning was called a Triangle, which is when you pin your opponent's arm and head between your legs and choke him. Even though we were both in our first week of training, we could easily figure out if the move was working or not based on the other's reaction. Amazingly, the reaction wasn't the only signal. We were telling each other if it was working and suggesting adjustments to make the choke tighter.
Like BJJ, YouTube also gives a lot of feedback. Success on YouTube is a simple formula: get people to click on your video and watch as much of it as possible. Within hours of uploading a video, YouTube provides metrics on how well it is performing on these objectives. Like the Triangle, YouTube lets you make adjustments to see if the video can perform better.
Because BJJ and YouTube provide timely, relevant feedback, you can see measurable improvement. In those first few weeks of BJJ, I couldn't last more than 30 seconds in a match, and if I survived that long I had no idea what to even try to win. Now I compete in tournaments. My first year on YouTube, viewers spent 10,000 hours watching my videos. In the last year, viewers have watched 41,500 hours. These are tangible measures that show I am getting better made possible by rapid, relevant feedback.
Academic research is not like this.
Feedback in Economics
Feedback in research, at least for economics, takes ages. Earlier this year I had a big moment publishing the main work from my dissertation. When I was reviewing the files, I was disheartened to see that the first work on the project started in 2014. It took me 7 years to go from inception to completion. My main project right now started in 2017--I have pictures of me showing the first results to my son that are cute except for the fact that I'm still working on that project. And I have a friend who started a project at least 10 years ago but still hasn't been able to publish it despite multiple reviews from the top economics journals. Research feedback takes too long.
Even worse, research feedback is imprecise. Most of it is idiosyncratic, reflecting the peculiarities of whoever happens to come across it. This isn't an accusation against all the economists who have given me bad feedback. I've witnessed myself doing it. A journal asked me to review a paper, and I suggested that the way it was setup made it natural for a certain analysis. After submitting that feedback, I came across an older version of the same paper and discovered that version had the analysis I had requested. Clearly a previous referee had asked them to remove it. I could immediately picture their frustration that they now had a referee asking them to include it.
Idiosyncratic feedback impedes progress. When I learn why my choke fails, it reflects principles that will work regardless of my opponent. When I look at the click-through rate on my videos, it reflects thousands of viewers who have seen my title and thumbnail and choose whether to watch it or something else. The feedback is reliable and targeted at the exact areas I need to improve to achieve my goals. But in research, a referee might raise a point that has no impact on the research's success.
So how do we make being an economist more like jiu jitsu?
The Economics Gym
Create a research gym. Like in a jiu jitsu gym, in economics you can get the feedback you need by finding some training partners. Find some young economists who can point out where your work is confusing or could use some improvement. Then give them feedback. Make it a regular thing and get comfortable looking stupid in front of each other.
I've been experimenting with this and I found that the feedback we give each other is not just about research: we'll discuss teaching and career goals, or sometimes just express our general frustrations. It's been great.
But even more fundamentally, there is another step you need to take: we can't be afraid to look stupid. The key to improving is feedback, and we will never get feedback if we are afraid to look stupid. This principle holds in jiu jitsu. In fact, I have found that the most experienced grapplers are the ones who ask the most questions. They know when a move feels wrong. They know that if they can't execute in a practice where there is little resistance, they will never nail the move in an actual match where the opponent is trying to choke them.
The research gym won’t resolve the idiosyncratic feedback problem. Your friends will tell you to do some things that other economists might not like. But some feedback is better than no feedback, so take what you can get. Don’t be afraid to look stupid, ask for help.
Indeed, if you're afraid that showing your work to someone will make you look stupid, that's a sign that you need to show it to them! That's the moment when feedback is most crucial.