Given a plane with flaps that can be deployed a little bit, i.e. a few degrees, it is possible to increase the camber of the airfoil increasing relative lift at the price of drag and speed.

Given such a modified wing airfoil the optimum angle of attack of the wing will change.

From : Don Stackhouse

In addition, the extra camber will tend to increase the aerodynamic pitching moment of the airfoil.

Given that the optimum angle of attack has changed, then the decalage will need to be adjusted. This can be sort of achieved with a little trim adjustment.

It may be more than a little, depending on the pitching moment. In any case, it's not just a change in the angle of the tail, it's also a change in the LIFT of the tail. This means that at one or the other or both of the two conditions in question, the tail must make a finite amount of lift.

Any surface out in the wind that has a hinge line will be cleanest when the hinge is not flexed. Any flex introduces a bit of drag above and beyond that bit of aerodynamics cleaned up by trimming out the change in airfoil

Does this make sense so far?

All except that last part, plus those other factors I mentioned. A cambered airfoil can (in most cases) make more lift (with less drag) than a symmetrical one, all other things being equal. That's why we use them on wings. If the tail also must make lift, then a cambered airfoil also makes sense for it. Without the camber, the tail must have more surface area to compensate. If control authority is the question, a 2-piece stabilizer-elevator can be smaller than a 1-piece stabilator. This is true regardless of the tail type involved.

Among the various situations that require the tail to make lift, the abovementioned pitching moment due to flap deflection is one of them, and the need to generate force at the tail to accelerate and decelerate the mass of the aircraft about the pitch axis (such as pushing over at the top of zoom on launch, pulling into a pylon turn, etc.) is another. You may also need large amounts of lift from the tail during launch depending on the location of the towhook relative to the wing and C/G. A 2-piece tail is better at generating this lift. It is also usually structurally more efficient.

A 2-piece tail has lots of room for its spar, and the designer is free to locate the spar cap material in the most effective location, as far as possible from the neutral axis, spread over the outer surface of the tail panel. A pivoted all-flying surface must have a circular member somewhere in its load path, and in most of today's typical cases (i.e.: plug-in rods) smaller in diameter than the thickness of the surface. This reduces the structural efficiency of that member, which means that it must be thicker (with a correspondingly thicker airfoil in that part of the control surface), or made from a much stronger (and probably heavier) material, such as steel. Since this is the joint that also carries all of the primary flight loads into the tail cone, then it must also be very rigid and slop free, but still very low in friction. In addition, since all of the bending and shear loads form the tail are funneled down into this pivot, there is the issue of stress concentration on either side of this joint to deal with. Since the 1-piece tail might also be larger in span, the bending moments to be transmitted through this joint may be larger as well. All of this adds weight.

In the case of the 2-piece tail surface, this joint is not a moving part, so friction isn't an issue, and it's very easy to get a strong, rigid, but light attachment. It's the old "divide and conquer" strategy; the structure that has to transfer the accumulated lift of the tail is not the same one that has to provide for all of the movement. Individually those problems are usually easier to solve than when they're lumped together.

A final concern is that airfoil quirks (such as aerodynamic hysteresis) tend to cause more trouble on 1-piece control surfaces than on 2-piece control surfaces. The popular 8020 airfoil is an example of this; although it has run into hysteresis problems on all-flying tails, this is rarely a problem on 2-piece tail surfaces.

Given all of the above, maybe the question should be "why do folks continue to use all-flying tails?"

Don Stackhouse
DJ Aerotech