On Yamaha's website, the specs for both the PSR-E453 and the PSR-E463 state that the power consumption for the keyboard is 8 watts when using the PA-150 adaptor. Yet, the specs also rate the amplifiers' output as 6 watts + 6 watts (6 watts for the left channel, and 6 watts for the right), for a total of 12 watts. So..... Do these keyboards have solar panels to supply the extra 4 watts of power to the amplifiers? And that's not even counting the power required for the sound generating circuitry, not to mention the fact that if an amplifier is outputting 6 watts, then it is requiring more power than that to run it because the amp won't be 100% efficient.
What is likely going on is that the 6 watts per channel rating is a peak value, not a continuous RMS value. The true continuous power is probably about half of that, but that is just speculation. Additionally, when you double the power of an amplifier, all other things being equal, the resultant increase in volume is only about 3 dB, so it won't sound dramatically louder. But you should get less distortion at higher volumes.
Now, when I look at the specs of my PSR-E433, it states that the power consumption is 16 watts when using the same PA-150 adaptor, with amplifier ratings of 2.5 watts per channel (5 watts total). So, it's a bit of a mystery. Is the E433 rating for the amps a true RMS continuous rating? Why does the E433 require twice as much power to run it? Less efficient circuitry all around?
Something similar happens with car stereo, when the head unit (radio) is advertised as "50 watts per channel", but then if you want to run a subwoofer and you buy a 50 watt per channel amp for it, the amp is a much larger and heavier piece of equipment than the radio. It is because the radio is rated in peak power (the actual RMS power usually being 10-15 watts per channel), while the amp is rated in true RMS power.