If you use Morningstar or any other data services provider, all returns are presented net of fees. So if Morningstar says, in 2011, fund A did 10% and Fund B did 7%, those are the numbers you want to compare.
And just to be even more explicit about it, if Fund A is your American Fund with a 2% expense ratio, and Fund B is your wife's Vanguard fund with a 0.2% expense ratio, that means, if there were no fees at all, your American fund would have returned 12% and your wife's Vanguard fund would have returned 7.2%.
The key is to make sure that you're comparing funds that have the same goals. Over the past 5 years, pretty much any stock fund, even with insane expenses, would have outperformed any bond fund with the lowest possible expenses. So if your wife's IRA is simply invested in different types of funds than your IRA, it doesn't make any sense to compare them. Instead, compare your American fund to a similar low-expense fund at Vanguard.
Even with that correct comparison, it's still entirely possible for high-expense funds to outperform low-expenses funds of the same type, even after expenses are included. That outperformance is unlikely to hold over the long term, but even then, is not impossible. And if that's true for your American fund, that's all the more reason to get out of it now and switch to a low-expense fund. You'd effectively be "selling high" before reversion-to-the-mean pulls it back down.