There are two main potential problems with this metric:
You are comparing a total obtained over a date range ("Return") over a point-in-time measurement ("Invested Capital") Invested Capital can change over the time period in which you're calculating return, so what value do you use? IC at the beginning of the return period? The end? The middle? An average? None is fundamentally correct. Suppose you have a stock offering on Jan 2 which increases your IC. Obviously the beginning IC measure would be inaccurate because you had more capital for the other 364 days that you had on day 1. The average would also be somewhat inaccurate because it's not like you start seeing financial returns immediately when you raise capital - it takes time to buy assets and see the benefits of those additional assets.
There is no fundamental definition of "Invested Capital". A very rough estimate would be "Liabilities + Equity" Since that measures capital raised through both bonds and stocks, but that includes things that aren't really "capital" like short-term debt and goodwill. So you'll see more complicated measurements that include more detail.
there is no debate on ROA/ROE
Sure there is - you have the same problem there. Dividing a total measured over a date range (return) by a point-in-time measurement (Assets/Equity). Do you use beginning/end assets? Average? Do you include unusual items like goodwill in equity?
The main point of these ratios is that they give you a very quick measure that might be misleading. But they can be used as a screen or flag that encourages you to dig further. If a company has a low ROIC, then you need to ask "why?" Is it because of a fundamental problem with the operations? Or because of accounting anomalies? They can never be taken as absolute truth - you often have to look deeper if you want to better understand the metrics of a company.