The bus vs. my car

I've just started riding the bus to work.  I'm not trying to save the environment or anything.  My car's in the shop, and I'm too cheap to spring for a rental.

The truth is, though, that I also like to ride the bus. 

When I was a grad student a few years ago, I lived in an apartment near MoPac and Ben White.  My wife and I both commuted.  She got to drive the '98 Civic and I got stuck with the '92 Cherokee with something like 200,000 miles on it.   One very hot May, the air conditioner went out in the Cherokee. I noticed then something that I really had not thought about before:  If you park your car in the Austin sun, you cannot drive without air conditioner.  It's not that it's too hot while you are driving -- once you get going, you can crank down the window to catch the breeze.  The problem is that if you park out in the sun, as we had to, you can't get in the car to begin with; there's just no way to cool the interior down from 130 degrees. 

So I started riding the #16 bus.  It was a long ride, about an hour to downtown.  But it was a relaxing ride.  The bus was never crowded.  It cut through a bunch of nice little neighborhoods.  I could just look out the window and zone out, or I could read a book or newspaper.  I kind of enjoyed it, and the extra time was not a big issue.

We bought a new car that summer, though, and I started driving again, since by driving I could cut a 60 minute ride down to 20 minutes.  I haven't ridden the bus since, at least until recently.

I take the #3 now, which stops on South Lamar right outside my neighborhood and drops me off three blocks from my office downtown. Once I'm on the bus, the travel time is virtually the same as driving.  The bus makes a lot of stops, but when I'm driving in rush hour traffic, I have to make a lot of stops anyway.  The bus takes maybe an extra 5-10 minutes.

It takes extra time, about 8-10 minutes, to walk to the bus stop from my house.  But that doesn't really bother me in the morning, when it's pleasant.  It's not as pleasant in the afternoon, but the fact that I'm going home takes the edge off.

I definitely prefer riding to driving.  I've always hated fighting my way through traffic to downtown when I'm still half asleep.  All I have to do on the bus is stare out the window. 

The bus is insanely cheap:  $10/month for all the rides you want. By comparison, I pay $100/month for parking at the office, $80/month or so for gas driving back and forth from the office, and another $40-$50/month for maintenance, taxes and insurance.   The car's paid for, but I "pay" at least another $200/month to cover the depreciation attributable to my commute.  When I add it all up, I figure I could take a $10 cab ride each way every day and come out basically even.  The bus is a real bargain.

Here are the two things, though, that really, really bug me about riding the bus. 

First is the discontinuity in travel time.  If I'm driving and leave my house at 8:20 a.m., the odds are that I'll get to work just a minute or two later than if I had left at 8:19 a.m.  I might hit an extra red light, but the difference in arrival time usually will be negligible.  That's not true with the bus.  If I walk out the door at 8:20, I might miss a bus that I could have caught if I had left at 8:19.  As a result, I won't get to work one minute later; I will get to work 20 minutes later.  Nothing irritates me more than seeing my bus rumble past the bus stop when I'm just a 100 yards away.

Second, I hate an uncertain wait.  When I get to the bus stop, I never know whether the next bus is 2 minutes away or 20 minutes away (unless I happen to see another one go by right before I get to the stop).  I don't mind the walk, and I sort of enjoy the ride.  But I hate not knowing how long the wait will be.

As I was waiting for the bus this morning, I was daydreaming about possible solutions.  The most obvious to me would be to track each bus using GPS and display the positions in real time on a map.  Riders could  check their bus's location on the internet before leaving home, and time it so they got to the stop just before the bus.  No more "just missing" the bus or sitting on the bench staring at your feet for 20 minutes.    

It turns out that this solution is so obvious that some other cities are already doing it. 

I can't speak for anyone else, but I'm pretty sure that I would give up commuting by car permanently if I knew exactly where my bus was before I left my house each morning. 

If CapMetro got, say, 25% of its revenue from fares rather than 8%, would we see more technological innovation?

Comments

[jmvc]

thanks for riding! CMTA is actually in the middle of a $13 Million Intelligent Transportation System implementation that will do almost exactly as you daydreamed..."using GPS and display the positions in real time on a map. Riders could check their bus's location on the internet before leaving home, and time it so they got to the stop just before the bus. No more "just missing" the bus or sitting on the bench staring at your feet for 20 minutes."

It sounds like you are already a pretty patient guy - you have to be to ride the bus regularly - so hopefully you will continue to patiently ride as CMTA works to improve the rider experience. - jmvc

[Tim]

Aweseome! I also take the #3. I start at Manchaca and Dittmar, and end downtown. It takes me about twice the time as driving, but I can work on my computer or read which I find much more enjoyable than driving. Because I'm at the end of the line I pretty much always know when my bus is coming, but I can definitely see how this could be useful.

[M1EK]

The #3 was one of two buses I used regularly commuting up from home to 183/Braker - I was never able to do any work on this bus - way too jerky. The 983 (which I took sometimes, but runs on very long headways), could sometimes work. I'm amazed you guys can do anything more than light reading without getting carsick on the #3.

JMVC, glad to see you're here - I was about to post the same thing after you referenced ITS work recently on my crackplog.

I face an upcoming physical office for the first time in years - and unfortunately it's outside CMTA's service area (near Westlake High - a bit long of a walk given my health and then two buses) so we may be ending our blissful 18 months of only one car. Too bad we didn't built light rail in 2000 - which, by now, might have started luring more employers back to the center-city (neither commuter rail, streetcar, nor the combination will be able to do jack squat here of course).

[M1EK]

Oh, and the true contrarian position doesn't buy the baloney that mile-based depreciation is worth worrying about. Consider this: if you leave a Honda Civic in the garage the whole year, are you really saying it's worth $2400 more at the end of the year than one that's driven the average amount? Repeat that over 5 years? Not likely - in the real world, most depreciation on most cars is a function primarily of time, not miles.

[AC]

Well, to some extent it does depend on milage. All else being equal, a 5 yr old car with 120k miles will be worth less than the same car with 60k.

But granting that depreciation is largely a function of time, how you apportion fixed cost among variable output (here, miles driven) is somewhat arbitrary. One way is to apportion it pro rata. Another is to assign all fixed costs to the first unit -- i.e., that first mile driven each month costs you $400, but the rest of them are really cheap. But the depreciation has to go somewhere.

[M1EK]

"Well, to some extent it does depend on milage. All else being equal, a 5 yr old car with 120k miles will be worth less than the same car with 60k."

A slight amount less, if you sell it yourself, and are a very good salesman. It's never been an issue when I've done a trade-in; and at best, it's a few hundred bucks difference. Take a look at bluebook (wildly optimistic) - even they see a difference of only $1600 on private-party values between high (120K) and low (40K) mileage on a 1998 Civic LX, all else being equal. That's about $200 per year, not per month.

I prefer rather than assigning to "first unit" to just treat as a different type of cost - time-based versus mileage-based, and thus assume no savings on mileage, because realistically that's how it works. If you leave your car at home one day next week but drive the other four, your depreciation savings are basically zero.

Of your $200/month estimate, for instance, maybe $10 or $20/month is truly mileage-based - the other $180 to $190 is only achievable if you don't have the car at all (a savings which we are enjoying right now, don't get me wrong, but to use it when you have a parked car is painting a very inaccurate picture).

This is now long enough that I should just crackplog it - maybe if I have time I'll clean it up and do it later.

[AC]

I see our disconnect now. I was assuming I'd get rid of my car, and was simply calculating the commute's share of the savings. That's why I also counted insurance savings. But I didn't disclose that assumption, I admit.

Otherwise, here's what we're arguing about. You're arguing that when deciding whether to drive at the margin -- whether to drive one more mile -- you should ignore depreciation because the added marginal cost is essentially zero. I agree. Ignore it.

On the other hand, when deciding whether to own the vehicle, you must take depreciation into account, just as any business must take fixed capital costs into account in addition to marginal costs. I'm sure you agree with that.

You can't separate that from miles driven, however, because the only utility you get from the car is the utility you get out of driving those miles. (Well, maybe you get some peace of mind in knowing you could drive somewhere, but I'm ignoring that because I think it's not worth much). Put differently, you have to recoup your fixed costs through miles driven. There's no obvious way to allocate that fixed cost among miles driven. But you're effectively doing it somehow, just as car rental companies recoup fixed costs by charging a price that includes a fixed cost component as well as marginal cost component. It's just as reasonable to allocate it pro rata per mile as to do it any other way.

[M1EK]

Well, we're still not on the same page - yes, in the marginal case, which is the one which makes the most sense in these "commute calculators" since very few of us can completely give up a car, I think we're now agreed that using depreciation at all is dumb.

However, even when considering the "own a car at all" case, you should allocate depreciation by time, rather than by miles, even though the IRS doesn't do it that way. Just because they can't make that model work for the "employee uses own car to do work business" case. Look at how they handle other equipment: We don't depreciate printers by number of pages printed, even if the cost of the ink is non-trivial. We don't depreciate computers by the number of bytes they process. Does the IRS depreciate business-owned cars by the mile, or do they do it on a 5-year schedule? (I can't tell).

If you're budgeting honestly, it still has to be a two-part equation: piece #1 is how much you have to spend to own the car at all, and then piece #2 is how much it costs for each mile after that. And piece #1 includes not only depreciation, but also insurance, registration, and some time-based maintenance.

[M1EK]

Oh, and I think the ability to drive somewhere is worth quite a bit - otherwise the car-sharing services wouldn't get any customers.

Upon rereading my comment and yours, it's not that you didn't accurately assess fixed versus variable, but rather that I disagree that there's any utility in just dividing the fixed cost over the variable cost and treating that as your per-mile cost. The IRS does that for your own car for business purposes because there's no other way to do it in their particular circumstance, not because it's the best way overall.

Part of the problem here is that people underestimate how much of the investment in an automobile is fixed - I don't want to make it any easier for them to do so by declaring fixed costs as variable ones - because then when people ride the bus a couple times a week and notice they aren't really saving any money, they're going to get pretty pissed.

[nigel]

Having used English buses for several decades the #3 feels like the ultimate in smooth rides. Dunno about the south leg, but I've ridden it from 30th to the Arboretum just about every week day for the last year. Make it run every ten minutes and lop off that weird Northcross detour on the way back and I'd marry it....

[M1EK]

Nigel,

You ought to try out the 982 or 983 (if it's running when you're going) - much quicker if you're going to the Arboretum, and far smoother, although if you think the #3 is already smooth, you might actually fall asleep on the express...