I don't know about "always".  I remember in the 1970s they were giving
away white vehicles.  Vehicle colours have more to do with public mood
and temperament than anything else, and white comes and goes out of
style.

HR.

All of your replies make entirely too much sense!  I thought it was strictly
because of cost on the fleet side, but thought it might do better with
scratches (and repainting) like silver, but never thought of the fact it
stays cool in the summer sun AND looks good, too.  Probably helps keep
things cool under the hood too.

It's going to be white or silver for me.  I have an '89 Tempo beater with
202k and it still doesn't look -that- bad even though it's rusty and has
been backed into three times

I've been hit 4 times in my truck.  3 of them were very minor taps on
my trailer hitch with no damage to my truck.  The last one was a
doozy..  Guy rear ended me at a stoplight.  He hit me doing 45,
destroyed his min-van and pushed me 20 feet uphill into another car
(Honda Civic) and then pushed the both of us another 20 feet  or so.

My truck lost a bumper and the trailer hitch and slightly spread the
"C" channel of the frame - from twisting the hitch under the truck.

Front end got a small ding in the bumper.  My light bar crumpled the
hell out of the trunk deck of the car in front of me.

No sheet metal damage at all to my truck.

My truck is lifted with big tires.  I asked the guy how he didn't see
the 22 foot long, 8 foot tall bright white truck and the 15 or so cars
stopped in both lanes in front of me.  "I don't know"

Oh, his van was blue.  Mabe I should trade this one in on a silver
one?

These studies are bunk because they suggest a causal relationship
between color and accident rate, ticket rate, mortality, or whatever
else is measured.  That is ludicrous.  You may as well throw in teen
pregnancy rate, because it's just that meaningless.

At the very best, it indicates which car colors are more popular
(thereby increasing the frequency of accidents, because there are more
of them around), and even a further stretch would be to say it
-implies- a -weak- correlation between car color and driving habits.
But, people don't always buy, inherit or steal the color of car they
want, and other random or incidental events introduce for too much
error.

Red cars are said to get more tickets.  Tell me that a color blind cop
gives out fewer tickets to red cars, ceteris paribus, than a non color
blind cop, and I'll tell you that you can never hold everything else
constant.

Some of the least safe cars are green?  I have two of them, should I
sell my wife's Saturn and my restored '67 Mustang, or I am just foolish
for having bought a green car and having '67 in painted in Dark
Highland Green?  What have I done!?  Am I supposed to suddenly feel
safer in my crappy old silver Tempo in driving rain and heavy traffic
with old wiper blades, a binding left front brake caliper and a drunk
driver swerving at me, than in my Mustang on a clear day and an open
road?.  Are you beginning to see some of the problems here?

I'm sorry, I know it sounds interesting, but I just cannot put any
stock in these "studies", and I cannot imagine a serious academic
wanting his or her reputation marred by being associated with the
publication of them.

--
Wound Up

After everyone's recommendations, and looking at them, I bought a white
Ranger three days ago.