Troubled taxpayer-owned insurer AIG has announced that it has a new approach to bonus compensation. It will now grade all employees on their performance. The firm will then use those grades to figure out how much to distribute in bonuses to those employees. My first reaction is -- does this mean AIG is admitting that bonuses weren't based on performance before?

The NY Times reports:

The new plan is meant to quell criticism that financial firms routinely overpay their employees, even if the institutions perform poorly.



And here's how it would work:

Under the new system, employees will be ranked on a numerical scale from 1 to 4. Those ranked 1, a group expected to be no larger than 10 percent, will receive much more in annual bonus payments than their peers, according to an A.I.G. spokeswoman.


Those ranked 2 or 3 -- together about 70 percent of A.I.G. employees -- will be considered as having performed above or in line with expectations. Those ranked 4 will receive lower incentive pay. Unlike with previous plans at G.E., they will not be immediately pushed to leave the firm.



From what I can see, this fails to imply that AIG will no longer 'overpay their employees, even if the institution performs poorly.' It will just justify that pay with grades. Nothing about the new system implies that if the firm -- or even a group within the firm -- posts a loss, those bonuses would approach zero.

I've never worked for AIG, so I don't know how they paid out bonuses up to now. But I have worked at two other financial services firms, and they both had very well-developed performance ratings systems that were also based on numerical scores, 1 through 5. And they also provided employees bonuses, which had a strong correlation to those scores.

So such a grading system is hardly novel. I would be pretty surprised if there wasn't some sort of similar performance rating system in place at AIG already. So what I suspect this announcement amounts to is a PR move on the part of AIG.

At least, I hope it is. Otherwise, the obvious implication is that AIG hasn't been rating their employees and paying them accordingly. And that would be a shock. It definitely made some big mistakes in the past, but I find it unbelievable that AIG could be so dysfunctional an organization to not engage in as basic a management practice as to hold performance reviews. And if those results didn't reflect pay, then what was their purpose?

I'm also pretty unmoved by the description above about how these grades will affect pay. Clearly, AIG is grading on a curve: even if you get the worst grade, you'll still get a bonus -- just a smaller one. Again, I just don't understand why they bother even calling such pay bonuses, if they're paid out no matter how poorly someone performs. Instead, they should just give these employees a higher salary and actually base incentive pay on incentive.

But the idea of a strict grading system for determining bonus pay isn't bad. AIG's system just isn't a very good one. I'm going to give this some thought and post what I believe would be a reasonable incentive compensation system sometime in the next few days.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.