There's a big difference between believing "kids will do well if they can" and "kids will do well when they want to".
If we believe kids do well when they want to, we will likely engage in manipulative schemes filled with rewards and punishments. The problem with rewards and punishments is that the kids who have the most troubling behaviours are those who have already had their share of bribes and threats. It should be more clear now than ever that doubling the dose of carrots and sticks isn't working.
In short, this form of discipline is broken.
If we believe kids do well if they can, then we understand this isn't a motivation problem -- it's a lagging skills problem. When a child has trouble reading, we engage in teaching them the reading skills they need in order to read -- and so, when a child has trouble behaving, we engage in teaching them the emotional, social and thinking skills they need in order to behave.
If a kid is illiterate, we figure out what skills they are lagging and we teach them those skills. We don't see illiteracy as a motivational problem, we see it as a lagging skills problem. And yet, when we see a kid misbehaving, conventional wisdom tells us to bring in the reinforcements in an effort to spur on their motivation to behave.
It's time we taught behaviour like we teach literacy. It's time we saw misbehaviour for what it truly is -- a lagging skills problem.
Kids don't go bad -- they get lost.
And it's our job to help them find themselves.
For more on this, I totally recommend Ross Greene's book Lost at School.