Polls confirm that those individuals who are downloading illegally online are buying less. That illegal downloading is decreasing sales is probably not a surprise to anyone.
Such a common, simple, wrong assumption at work here. A decrease in sales and an increase in music downloading have *not* been shown to be related.
For anyone wondering how to categorize the RIAA's fallacy here, it is "post hoc, ergo propter hoc" -- it happened thereafter, thus it happened because. In other words, the RIAA uses the sneaky implication that since use of P2P is correlated with decreasing sales, the decreasing sales are concluded to be caused by the P2P applications. Unfortunately, correlation does not imply causation, it simply suggests it.

Such propaganda tactics are tricky to pick out of a smooth-sounding line of reasoning, but invalidate the argument nonetheless.