Is
it better to drive, fly, take the train, or take a
bus?
We compare which is
cheapest, along with the environmental
impact
You have a lot of choices when traveling
about how to get there. And each one offers a
different amount of speed, safety, comfort, cost,
and environmental impact. How could you begin to
compare all these different variables? Well, you
don't have to, because we've done it for you.
Summary of choices
It would be mean to make you read
the whole analysis just to get the punch line,
so here's the summary:
*Car's env. impact per person
is less when multiple people travel,
perhaps better than the train, but never as good
as the bus. (more...)
*Car's env. impact per person
is less when multiple people travel,
perhaps better than the train, but never as good
as the bus. (more...)
If you want to see how we got those
results, just keep reading.
Where to find these choices
Airplane. Try Travelocity
or Orbitz
to find the best fares for the U.S. Outside the
U.S. see our international
fares page.
Train. The U.S. has one main carrier,
Amtrak,
though you can also go from New York to New
Jersey on Septa,
and from New Jersey to Philadelphia on NJ
Transit.
Bus. A few choices:
Fastest:
Bus or train (short trips),
Plane (long
trips)
Flying doesn't save you any time
when the trip is short. First off, you're
supposed to arrive at the airport 60-90 minutes
before your flight, and it will probably take
you at least 30 minutes each way to drive, taxi,
or bus to the airport. That's 2 to 2.5 hours
right there, not counting flight time. You could
drive from NYC to Philly in less time than that!
Flying is only faster when you're taking a
decent-sized trip.
Remember that driving time is "wasted
time" -- you can't do anything useful while
driving. But when you take a plane, train,
or bus, you can read, study, work, or sleep.
That's especially true on a train where you can
get an upgrade to a "roomette" cabin which
includes a bed.
Overnight trains are often more convenient
than flying. That's because the time you
spend sleeping on the train isn't wasted time. A
plane might take 3 hours while a train takes 11,
but if you spend 8 hours of that sleeping in a
nice bed on the train, then the train is a
better option. You won't have to hassle with
airport security, you might not have to check
your baggage, your mobile phone will likely work
throughout the trip, and you won't have the cost
of a hotel room.
We'll use three sample trips:
- Short: New York to Philadelphia
(102 miles)
- Medium: New York to Detroit
(509-615 miles)
- Long: New York to Los Angeles
(2467-2791 miles)
The mileage is given as a range because air
miles are usually less than driving miles.
|
Speed of Short Trip (NYC to
Philly), One-way
|
|
|
To & From
the Terminal
|
Waiting time
at terminal
|
Travel Time
|
Total Time
|
|
Airplane
|
1:00
|
1:00
|
0:30
|
2:30
|
|
Car
|
|
|
2:30
|
2:30
|
|
Train
|
1:00
|
0:20
|
1:30
|
2:50
|
|
Bus
|
1:00
|
0:20
|
2:00
|
3:20
|
|
Speed of Medium Trip (NYC to
Detroit), One-way
|
|
|
To & From
the Terminal
|
Waiting time
at terminal
|
Travel Time
|
Total Time
|
|
Airplane
|
1:00
|
1:00
|
2:30
|
4:30
|
|
Car
|
|
|
9:30
|
9:30
|
|
Bus
|
1:00
|
1:20
|
12:30
|
14:50
|
|
Train
|
1:00
|
0:35
|
15:00
|
16:35
|
|
Speed of Long Trip (NYC to
L.A.), One-way
|
|
|
To & From
the Terminal
|
Waiting time
at terminal
|
Travel Time
|
Total Time
|
|
Airplane
|
1:00
|
1:30
|
6:00
|
8:30
|
|
Car
|
|
|
1 day, 19 hours
|
1 day, 19 hours
|
|
Bus
|
1:00
|
2:50
|
2 days, 11 hours
|
2 days, 15
hours
|
|
Train
|
1:00
|
5:35
|
2 days, 11 hours
|
2 days, 18
hours
|
Waiting time at the terminal includes
layovers.
Cheapest
Cost: Bus or
Car (short
trips), Plane or
Train (long
trips)
While airfare is relatively cheap,
you have to buy well in advance to get the best
fare. A couple of years ago when I
repeatedly checked the price for a particular
flight, gradually increasing the lead time on
each search, the fare got progressively cheaper
up to 30 days -- at which point I stopped
checking. (I'd run the same test again to find
the maximum practical lead time, but I'm writing
this in October wit the holidays around the
corner, and holiday fares are unusual so I'll
have to wait a couple of months before I can try
this test again.) Anyway, the point is, while
airfare can be cheap, you have to buy well in
advance to get the best fare. The penalty for
buying at the last minute on a train or bus is
generally much lighter by comparison.
Another thing to consider is that when
multiple people travel, the price per person
gets cheaper when you go by car. There's
usually no price break for multiple people on a
plane, train, or bus, since each person has to
buy their own ticket. But in a car you can
divide the cost by the number of people. (An
exception is that Greyhound sometimes offers a
2-for-1 "friend rides free" promotion, so if two
people are traveling that's worth checking
into.)
I include a little calculator below so you
can figure the cost for your own particular
trip.
|
Cost of trip depending how
you go (one-way)
|
|
|
Short Trip
(NYC to
Philly)
|
Medium Trip
(NYC to
Detroit)
|
Long Trip
(NYC to
L.A.)
|
|
Airplane
|
$66
|
$86
|
$151
|
|
Car*
|
$19
|
$115
|
$524
|
|
Train
|
$45
|
$75
|
$184
|
|
Bus
|
$26
|
$80
|
$211
|
*Car costs assume 22.6 mpg,
$3/gal., 5.5¢/mile maint.
Travel fares were checked in October 2008.
Airfares from Orbitz,
though I also like Travelocity
Safest:
Plane
Flying is much safer than
driving. If 140 people want to take a
cross-country trip of 3000 miles, they could do
that in a single plane, or in 70 cars, each with
two people. That trip would be only 3000
plane-miles in the plane, but it would be a
whopping 210,000 car-miles by driving. So the
exposure in traveling by car is 70x higher per
person. So even though cars produce fewer deaths
per mile, there are more total deaths because
there are so many more miles driven.
But while it's true that flying is safer
than driving, traveling by train is safest of
all. That's exactly what you'd expect, since
they travel on their own tracks. Anyway, here's
how all the modes stack up.
|
Fatality rate per billion
passenger miles traveled
|
|
Car (most
dangerous)
|
7.2
|
|
Airplane
|
2.3
|
|
Bus
|
2.0
|
|
Train
(safest)
|
0.5
|
See sources
&
references.
If you're skeptical than passenger-miles is
the right way to do the comparison, see the
explanation "Why
Passenger-Miles?"
Comfort:
Train
Trains are easily the most
comfortable way to travel. Compared to
flying. the train offers plenty of legroom, no
restrictions on the use of electronic devices,
generous free baggage allowances (and you can
even bring a bicycle), a dining car serving
freshly-prepared meals, and most importantly,
the option to upgrade to a "roomette" a small
room with two beds. That's available for an
extra $294 on a trip from Chicago to L.A.
Electrical outlets for planes and trains are
fairly similar: Always available in first or
business class, sometimes available in coach
class.
We'll rank cars as second-most comfortable,
because you have the ability to stop whenever
and wherever you want. The plane, of course,
ranks dead last.
|
Comfort by mode of
travel
|
|
Train
(most)
|
|
Car
|
|
Bus
|
|
Airplane
(least)
|
Least Environmental
impact:
Bus or Train
Flying is usually worse than
driving. Planes get a respectable 43
passenger miles per gallon (pMPG), but they
cause 1.9 times as much climate change per
gallon of fuel burned, by virtue of their being
up in the sky. So when we consider climate
change, the plane's efficiency is more like 22.6
pMPG.
Let's say you have a 23 mpg car. That
already beats the plane at 22.6 mpg. But what if
two people are traveling? On the plane it's
still 22.6 pMPG per person, since we're already
accounting for the fact that the plane carries
multiple people. But put two people in a 23 mpg
car and suddenly we're getting 46 pMPG.
The most efficient way to travel in the
U.S. is by bus. Inter-city buses get a
whopping 125 pMPG. By contrtast, Amtrak trains
get only 45 pMPG. (Though that could be because
U.S. trains often run fairly empty.)
Now let's put all this together:
|
Passenger
MPG
|
Transportation
Mode
|
|
125.0
|
Bus
(best)
|
|
81.6
|
Typical U.S. car, 4
people
|
|
61.2
|
Typical U.S. car, 3
people
|
|
45.0
|
Train (Amtrak)
|
|
43.0
|
Airplane,
w/o considering extra climate change
effect
|
|
40.8
|
Typical U.S. car, 2
people
|
|
22.6
|
Airplane after considering extra
climate change effect
|
|
20.4
|
Typical U.S. car, 1
person
|
I have more on the environmental
effects of flying, as well as a carbon
footprint calculator that lets you compare
the impact of driving vs. flying.
Why use Passenger Miles to
compare risk?
Some people (like De
Clarke) say that it's wrong to use
passenger-miles to compare risk, because if
the ridership gets cut in half then the accident
rate will appear to double. That's true, but
we're not looking at vehicle crashes,
we're looking at fatalities, which is a
different metric. If the accident rate stays the
same but half as many people ride, then half as
many people will die, so the death rate per
passenger mile is still the same. What this
means is that it's proper to use passenger-miles
to look at death risk, no matter how full or
empty the planes run.
If you're skeptical, just set up a
spreadsheet to compute the death rate per
passenger miles using any values you like.
Then change the average "fullness" of the
vehicle (e.g., 25%, 50%, 75%), and you'll see
that the death rate per passenger mile stays the
same.
Here's another way to look at it.
Let's say that we live in a really dangerous
world and planes kill 22 people on average for
every 1000 miles the plane flies, and 1 person
on average for every 1000 miles of driving. This
22:1 ratio is in fact the actual ratio in real
life, since planes do kill a lot more people per
mile traveled than cars do. Now let's say you're
contemplating a 1,000-mile trip. You take a car,
and bring along a friend. In our pretend-world
cars kill 1 person every 1000 miles, so one of
you will probably die, and there's a 50% chance
it will be you rather than your friend. Ouch.
But what if you take the plane? The plane kills
22 people every 1000 miles, so on a 1000-mile
trip we expect 22 people to die. But there are
140 people on the plane on average, so it's
unlikely you'll be among the 11 dead. Your
chances are 11/140, or 15.7%. That means that
flying is 50% ÷ 15.7% = 3.2 times safer
than driving.
Now, you might argue that if the plane
crashes then generally everyone dies, not
just 15.7% of the passengers. That's true, but
the example is still good, because the numbers
work the same way once we're talking about
thousands of flights and billions of miles. That
is, the overwhelming majority of flights have no
trouble at all, while on a few rare flights
everyone will die, and once that's averaged out
then flying still comes out to be 3.2x safer
than driving. In some years there are
zero deaths on American commercial
aircraft, while around 40,000 people die on
America's roadways every year, pretty
consistently. In short, flying is a hell of a
lot safer than driving.
Sources & References
- Airfares from
Orbitz.
Flight times courtesy OAG.
- Train fares & schedules
from Amtrak.
- Bus fares & schedules
from Greyhound.
- Driving miles between cities
courtesy of Google
Maps.
- All prices were checked in
October 2008.
Safety:
- 1949
air fatalities and 6 billion miles flown from
1988-2006 (NTSB,
"Aviation Accident Statistics", Table 6, U.S.
Air Carriers operating under 14 CFR 121). That's
103 deaths per year, or 90 deaths per year if we
exclude the passengers on the 9/11
flights.
- 848
billion passenger-miles flown per
year (Bureau of
Transportation Statistics, June 2007-May
2008)
- More
detail about U.S. air
crashes from NTSB,
including city, carrier, and type of
aircraft.
- Fatalities
by air carrier,
including international operators.
- 21,491 avg. car occupant
deaths and avg. 1,439,628 million passenger
vehicle miles traveled from 1988-2006, from
Traffic Safety Facts 2005
(PDF) and
2006
(PDF),
NHTSA.
- 718 avg. bus deaths from
1988-2006 ( avg.), from Traffic Safety Facts
2005
(PDF) and
2006
(PDF),
NHTSA.
- 330,137
million bus-miles
traveled from
1990-2006 (19,420 avg.), from BTS.
- 83
passenger rail deaths
from 1994-07 and 168,783
million passenger
miles from 1994-2006
(5.9 deaths and 12,983 million miles avg.) (BTS,
Table 2-38: Railroad Passenger Safety Data). If
we include data from 1990-1993 the fatality rate
is much higher (8.6 deaths/yr.), because of a
large Amtrak crash in 1993 that killed 42
people.
Environmental:
Last update: October
2008
|
