Hacker News new | past | comments | ask | show | jobs | submit login

In efficient markets, like sports teams, even average team members often make more than the manager (or coach). This also happens in some financial organizations that reward profit and loss on individual levels (it's very rare, and not as efficient as sports teams).

It's not particularly interesting to observe that outside of efficient businesses, managerial layers have a lot of room for rent seeking to ensure manager pay is higher than team contributor pay.

It's more interesting to try to understand why it persists, and what can be done to make employment more efficient in general. Or at least help people see which employers / fields are relatively more or less efficient, so you can decide if being paid way less than someone demonstrably reducing productivity (it happens with a lot of managers) is an ok tradeoff for you.

You can be sure it's a whole lot of Hansonian status signaling. For example, the well-documented resistance of project leaders to agree to be paid according to prediction markets about the expected project outcomes, delivery deadlines, etc. Even when experiments with such things led to better project outcomes, the managers then refused to agree to use the idea again or even to be part of, owing largely to the fact that their ability to manipulate things politically, and thus ensure higher pay regardless of the measured outcomes, was removed by the function of the prediction market. Without that angle to politically manipulate, they outright refused to work!




For example, the well-documented resistance of project leaders to agree to be paid according to prediction markets about the expected project outcomes

Is that a serious proposal? I've never heard of such an idea.

Prediction markets failed spectacularly to predict the outcome of the recent UK referendum despite an abundance of polling data and the only major uncertainty being statistical (not many people were changing their minds at the last moment, I guess). Most obviously because prediction markets tend to be quite shallow, and the types of people who use them bring their own mental baggage to the table. Why would a project leader be willing to submit to such a flaky and moreover gameable scheme?


You seem unfamiliar with the long and extremely formal research and experimental evidence supporting the use of prediction markets. It might be interesting to read a more detailed popular account [0] and then dive from there into some of the more specific discussions about what happened in the DARPA experiments.

> Prediction markets failed spectacularly to predict the outcome of the recent UK referendum despite an abundance of polling data and the only major uncertainty being statistical (not many people were changing their minds at the last moment, I guess).

Well, you are right that markets are not perfect and are susceptible to aggregated biases of the participants, for example just as how mispricing persists in the stock market.

I don't see how this offers a reason to feel that markets aren't the best that can be done, however. If you can propose a better mechanism by which to get info ahead of time, I (and millions of people) would be very interested.

> Most obviously because prediction markets tend to be quite shallow, and the types of people who use them bring their own mental baggage to the table. Why would a project leader be willing to submit to such a flaky and moreover gameable scheme?

This is bizarre and unsubstantiated. What are you talking about "mental baggage" -- it sounds like you have a personal grudge against market thinking and prefer to attack it, but without basis.

> Why would a project leader be willing to submit to such a flaky and moreover gameable scheme?

Actually, much of the research on prediction markets has focused on showing the extent to which they, mathematically, are not gameable [1]. It's bizarre that you feel the existing management mechanism, overt office politics, is somehow "less gameable" than holding people accountable for direct, specific, quantitative predictions. Even if markets were "gameable" you surely can't seriously be suggesting that they are somehow more gameable than overt office politics?

[0] < http://mason.gmu.edu/~rhanson/innovations.pdf >

[1] < http://mason.gmu.edu/~rhanson/biashelp.pdf >


There are markets, and then there are prediction markets. The two are not quite the same thing.

There are a lot of assumptions in your argument, like the number of participants in a prediction market being meaningfully higher and meaningfully more interested than actual project stakeholders. You're also assuming that such people are less biased and more neutral than other people who have to suffer office politics. Everyone has mental baggage and taking part in a "market" doesn't change that.


The reason why players are more valuable than managers on a sports team isn't because the market is efficient. Players put butts in seats. Nobody goes to a game to see the manager, even if he's the guy running the show. People like big name stars. They sell tickets, so they are more valuable. Big name actors can make more than directors. Same reason. They sell tickets. Most us us don't work in a market that works remotely like that.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: