I submit that we need to frame the argument properly.

There are two problems being discussed, and sometimes people
confuse the two.

The first problem is the generation of energy.

The second is the storage of energy.

I would further submit that the matter at hand has to do with
the storage of energy rather than its generation.

Fossil fuel powered vehicles (i.e., gasoline and diesel) use
energy that is directly generated via internal combustion (leaving
aside the energy required to obtain and refine the fuel, which
doesn't actually put any energy into the fuel that wasn't latent
in the petroleum in the first place).

Most alternative energy sources for vehicles, however, involve
energy which is stored by the vehicle, not generated by it.

This is really what you're doing when you transform water into
its component parts of hydrogen and oxygen. You put energy in,
you get those two elements out. Practically speaking, you can
toss off the oxygen because there's no need to carry it with you
in the vehicle.

Why do that?

Well, perhaps because you're now generating your energy in the form
of electricity which comes from a resource that's difficult to
carry around in a vehicle. There are engineering obstacles to
getting adequate solar power, wind power, hydro-electric power,
or nuclear power built into the family flivver. And storage
batteries have some real down-sides to them as well--- they are
heavy, take an unacceptable period of time to charge, etc.

So if you use a renewable source of energy to by one means or
another produce hydrogen from water, you end up with a somewhat
easily transportable (and rechargeable) source of energy.

The question at hand is the efficiency of that transformation.

In other words, how many ergs does it take to produce 1 erg of
energy to move the vehicle. Both transformations must be
considered...to hydrogen and from hydrogen.

The other piece of that is "how expensive (both in dollars and
in environmental impact) is the energy required to accomplish
that?"

If electricity is cheap enough (on both of the above indices)
you may not care that the overall transformation is only 30%
efficient (hypothetically).

If we find that electricity can be made without limit and/or
much expense, and that water is plentiful (and doesn't get used
up by the process because it comes out the exhaust of a hydrogen
vehicle) and that the engineering needed to transport and distribute
H isn't too heavy, cumbersome, or expensive, then we have the wave
of the future for powering transport.

Another thought problem--- depleted fuel storage issues aside, is
solar, hydro-electric, and wind power less contributory to global
warming than nuclear? Does nuclear power put heat into the earth's
eco-system which otherwise would not exist? Whereas solar, hydro,
and wind power simply divert energy that is inherent in the
environment in the first place to other uses?