If neutral is connected to ground and current desires ground then I would think it would go to ground. But my understanding is that it is kept off of ground until it's "upstream" of the breaker box, or unless a malfunctioning device shorts hot to ground. So what "prevents" the electricity from going to the ground that the neutral is connected to?
The short answer is resistance. Electricity will find the easiest path to ground. At the neutral-ground bonding in the breaker panel is a connection to the ground wire that goes to the actual ground. It's a heavy gauge, and connected to grounding bars deeply embedded in the ground to offer the least possible resistance.
For current to travel from the neutral back to the ground wires to where it can electrocute you, it needs:
As long as any one of those three conditions doesn't apply, you should not see any current on the ground wire.
A great question and it seems odd to have a ground and a neutral. We have a neutral to keep the ground from being loaded. We dont want the ground to be load, it doesnt have the capacity and its for back-up.
In AC systems, there is current on the "hot" leg and on the "reference" leg. We use a neutral leg for reference. In countries with 240v systems, there is no neutral, but two opposite legs of 120v that provide a reference for each other. You see in AC systems, the voltages are not constant - it alternates back forth +X volts and -X volts. + and - from what? voltage is just potential differences from the reference. In 120v systems, on leg will be -60v from the neutral and then switch to +60v from neutral 60 times a second. In a 240v systems with out neutral wire there is a perceived reference between the two 120v legs that are off opposite phase.
So, yes in theory, you could get the potential difference from a 120v "hot" leg and ground, BUT then your would be loading the ground and it would no longer serve a safety function. In practice when you load the ground, appliance panels and other things that are grounded out become will be loaded and can shock.
I think the answers here are looking too deep into this. The simple answer is, that it's not a complete circuit.
Electricity only flows in a complete circuit, and always want's to return to the "source". If there is no complete circuit, electricity simply will not flow.
Looking at a simple diagram of the secondary side of the distribution transformer, it's easy to see the complete circuits.
You can easily see the 240V circuit.
And the two 120V circuits.
But if you look at the grounding conductors, there is no complete circuit.
Because ground is not a return unless you bridge across from neutral, hot, or have a component short to frame ground.
It is a free standing wire/conduit network connected to earth that isn't a part of a circuit till a fault occurs.
The ground wire is for safety, it should be very close to the same potential as Earth. To that end, there is a legal limit of how much resistance there is from the ground at the service entrance to dirt outside. In the US it's 25Ω, or you can get away with just using two ground rods 6ft (or more) away from each other. (It's cheaper to install two ground rods than spend time and money on properly testing the resistance)
The reason the neutral and ground wires are connected at the service entrance, is to help keep the voltage potential of the neutral to Earth, as low as possible by keeping the resistance to Earth as low as possible.
To have absolutely no current on the grounding wire going to the ground rods, the nearest transformer would have to have equal current on both hot conductors. Since the transformer typically has it's neutral (center tap) grounded at the the same location as the higher voltage side, there would also have to have to be equal current on all of conductors on the high side as well.
So basically there is always some current going through the grounding system, at the service entrance. The main thing, is that the voltage is kept as low as possible; by keeping the resistance as low as possible.
If you want proof that there is some voltage, and therefore current, on the grounding system, measure the voltage from the grounding rod to a point in the dirt a few feet away.
If you measure any voltage with a high-end multimeter of even a tenth of a volt (AC); then there is current on your grounding system.
Even if you turn off your power, you will still see some voltage with both of these tests. Assuming there is someone else connected to the same transformer as you, and your neutral and ground wires are still wired properly. The amount of wire the current would have to travel through before it got to your premises would be greater, so the amount of current/voltage would be greatly lessened.