It removes the risk for the developer to just 'build' (which is ultimately what they enjoy doing) and also shifts the key responsibility for the initial marketing / customer acquisition away from the developer (which is often what they either don't enjoy doing or suck at).
After this phase, either the idea is a success, the developer gets paid and a mvp is now a business hopefully with profit (for both parties involved) or the idea was not a success and the developer still gets paid some money.
If you have a world changing idea and you've quit your job with a plan to start a start up - that's fine. This probably isn't the right opportunity for you - try Y Combinator or go in search of real funding once you've built a self financed MVP.
However, if you're a developer with a niggling idea that you keep meaning to work on because you know it could turn into something bigger or you just don't feel confident in your own marketing / customer (or user) acquisition skills then this is a really great opportunity!
I've literally missed so many opportunities before, where I've had an idea - not executed it due to no short term monetary reward or no confidence that I could market it correctly and then watched as a start up is formed a year or two later by someone else which was essentially that idea!