At my software development internship last Summer that I have written on a few times I participated in a Triangular Principal-Agent relationship. I had a manager, who was the manager of the overall software team, as well as a mentor who was a normal member of the team that was tasked with being in charge of me. I did most of my work with the mentor, and I received tasks from both the mentor and the manager. Most of the time these tasks lined up roughly, in a way that I could complete everything for both of them in the given time. There were a few weeks, though, in which the tasks did not line up well and I did not have enough time to finish them.
I think hypothetically it would have made more sense to complete the tasks given by the manager rather than the mentor, as he was the one with the overall vision of where the project was going. In reality, though, I found myself completing the mentor tasks first. I felt like he was more in tune with my exact role in the project and what I was actually researching and prototyping, so his tasks seemed to be more efficient. Happily enough, this did not seem to step on any toes. There was good communication on the team and since the management was made up of people that had done software for years they understood that some problems took longer than expected. (A common saying at meetings was that the last 10% of an issue took 90% of the time). They were all very accepting of me delaying some deliverables a week or so, as I still showed that I was completing work.
In a theoretical situation where my mentor and the manager were at odds with each other on what I should be working on I think it probably would be politically smarter to just go with whatever the manager said. This would decrease pressure on myself as well as my mentor, even if it resulted in slower overall progress. Of course, I would try to explain how my mentor's ideas could be more applicable to the problem we are solving, but in the end, if there was still tension I think following the manager would be the way to go.
Another potential solution for this problem could be working more hours to satisfy both the manager and mentor. From what I have heard from friends that start off in top companies and adults that have been around the block, out of college a lot of young adults work crazy hours to try to differentiate themselves from the others. This would not be optimal for a work-life balance, but the experience could certainly be good for my resume for a kick start to my career.
In the given example with the lawyer and the firm I think that it would be more useful for the underling to act in accordance with the lawyer above him. I would guess that any sort of performance review would be based upon how the lawyer perceives him, rather than the abstract thought of the firm. Thinking more on this point, if one of the two principals in the model is close to the underling and the other is more of a distant figure, it seems like it would be in the best interest of the agent to act in accordance with that nearer model. It seems like that pricipal's thought of the agent would probably matter more to the organization than that of the top agent.
I think hypothetically it would have made more sense to complete the tasks given by the manager rather than the mentor, as he was the one with the overall vision of where the project was going. In reality, though, I found myself completing the mentor tasks first. I felt like he was more in tune with my exact role in the project and what I was actually researching and prototyping, so his tasks seemed to be more efficient. Happily enough, this did not seem to step on any toes. There was good communication on the team and since the management was made up of people that had done software for years they understood that some problems took longer than expected. (A common saying at meetings was that the last 10% of an issue took 90% of the time). They were all very accepting of me delaying some deliverables a week or so, as I still showed that I was completing work.
In a theoretical situation where my mentor and the manager were at odds with each other on what I should be working on I think it probably would be politically smarter to just go with whatever the manager said. This would decrease pressure on myself as well as my mentor, even if it resulted in slower overall progress. Of course, I would try to explain how my mentor's ideas could be more applicable to the problem we are solving, but in the end, if there was still tension I think following the manager would be the way to go.
Another potential solution for this problem could be working more hours to satisfy both the manager and mentor. From what I have heard from friends that start off in top companies and adults that have been around the block, out of college a lot of young adults work crazy hours to try to differentiate themselves from the others. This would not be optimal for a work-life balance, but the experience could certainly be good for my resume for a kick start to my career.
In the given example with the lawyer and the firm I think that it would be more useful for the underling to act in accordance with the lawyer above him. I would guess that any sort of performance review would be based upon how the lawyer perceives him, rather than the abstract thought of the firm. Thinking more on this point, if one of the two principals in the model is close to the underling and the other is more of a distant figure, it seems like it would be in the best interest of the agent to act in accordance with that nearer model. It seems like that pricipal's thought of the agent would probably matter more to the organization than that of the top agent.
Here are a couple of thoughts based on your idea of closeness to the underling. Can you draw an org chart that includes the manager and the mentor in it? Was the relationship between the two of them vertical or horizontal? If it was vertical, I'd have called the mentor the middle manager, in this case in the middle between you and the manager. If the relationship was horizontal, then that is closer to the two principals problem as envisioned.
ReplyDeleteIn the vertical case, the question is why the manager gave you any direct tasks at all rather than giving everything to the middle manager who would task you. I can think of one possible reason. If ultimately the question is whether to offer you a permanent job and if that decision resides with the manager, then the manager might want some direct evidence of how you performed on the tasks he gave you. Do note that in general, if the relationship is vertical and the manager is tasking people who report to the middle manager, that is a no-no and is considered micromanaging.
If the relationship between the mentor and the manager was horizontal and if both them got to weigh in on whether to offer you a job or not based on your internship, then I'm afraid the structure you described is a necessary consequence of internships serving dual purposes - doing the work at hand and evaluating the intern. This, then, creates structures that shouldn't be replicated with permanent employees.
Hi Professor,
DeleteThe case was vertical. I think the reason they both gave some tasks was that I was the only software intern so most of the time the mentor was not assigning any tasks to anyone. The software team was small enough (~5 ish people) that it seemed like a lot of decision-making was group made. The manager still had a firm grip over everything, but everyone on the team was privy to most aspects of the project and had been on it a while so all of their input was valuable. This had the consequence of making a very strict manager underling relationship break down.