Updating benefit calculations and drive alone cost per mile

[Stan Suchan]

Which option from the table below do you think works best for our achievements calculation update?

At the last regional administrator's meeting, we decided to update the factors the system uses to calculate benefits like gallons of fuel saved, $ saved, etc. These factors affect the results that appear on dashboards and in reports. They were last comprehensively updated in 2011. During the discussion it was noted that people have very different perspectives, sometimes quite passionate, on this topic. To address these differences, we provide a link to a fact sheet that explains the factors. This allows people to reinterpret them to better align wtih their beliefs.

During the meeting much of the discussion foused on one factor: drive-alone cost per mile. During this discussion we confirmed that we would use 2013 AAA Cost of Driving medium sedan data for this purpose. The group wanted to consider a few options: 10,000 annual miles per vehicle vs. 15,000 annual miles per vehicle and using a factor that omits sunk costs like insurance, taxes, depreciation, etc. Below you'll find a table that presents these options. For reference, we are currently using a factor that aligns with option B in the table below. In 2011, this totalled $0.3383 per mile.

Please note that updates will be prospective; they will only affect future achievement calculations and totals.

   

 

 

 

 

 

 

 

 

[Stan Suchan]

We could calculate that number.

[Keith Cotton]

I like Option B. Driving a car, even if it isn't mine, has costs associated with its daily use -- taxes, license, insurance, registration. You could argue those are operational costs because they need to be paid for the car to be driven legally.

Interesting that these rates are for a new car. If there aren't average car rates then the explanation just needs to be very clear that this is based on a new car and a used car will have higher costs.

I don't mind D or E either. AAA's annual cost of owning and operating a vehicle study could also be used. It includes depreciation and finance. Maybe just tying to the "standard" AAA for the costs of a vehicle would make the most sense for the average user? Simple may be the best!

Parking is big expense for some users, and not having to pay parking is a big incentive for some. What if a user could enter their daily parking cost, and the program adds that in as a separate line item? It wouldn't require including parking into the per-mile calculation. It's custom to the user and the calculation can be added on by the system totaling a count of days not driven alone. Or something like that.

[Zach Shaner]

I think you should use the highest number that includes all inevitable operational costs shared by the highest number of users. To my mind, that would be B. Every single driver accrues costs related to fuel, maintenance, tires, insurance, license, registration, and taxes. On the contrary, depreciation/finance/parking are highly variable person to person. People can intuitively add the cost of a car payment or a monthly parking pass to the cost that RSO decides to list, but I've been turned off in the past by estimates from AAA and others that try to incentivize alternatives by overshooting their estimates, assuming everyone is financing a brand-new, high-value car and driving it every day into a high-cost parking garage. Only the exceedingly well-off do that, and even then only in Downtown Seattle. 

And I prefer B to C because regionwide VMT continues to decline, so I prefer the 10,000 estimate.

[Philip Warnock]

I'm inclined to fall into the "B" camp.  In part because I'm thinking of the carpool cost for non-drivers.  This would be about the rate I'd recommend using to calculate a fair-share contribution to the driver.  Recognizing that this is a little different from the cost savings calculation for the dashboard, the logic works for me backing into it for this purpose. 

[Victoria Tobin]

I met with my team on Monday to discuss you proposal of updating the benefit calculations and drive alone CPM and wanted to share their input. The majority of the team suggested staying with our current calculations - which I think is 'B' in the table above -  keeping with the industry standard makes sense - AAA is a known industry professional/market leader and a trusted, outside source – changing the calculations four years into the program does not seem like a reasonable move. What might this do to our reporting?

other comments from my group: If we add too many variables then we aren’t calculating equally, leaving room for misinterpretation and more end user questions - Not a hill to die on and to be spending money on additional coding. - If specific calculations are needed to justify funding, then perhaps there can be a configurable report or chosen specifics for the agency or group that needs those specific calculations – perhaps program RSO for the ability to pick and choose what networks need what numbers – this way the majority of folks can go with the average standard. - There seemed to be not enough consensus in either direction to change or not to change…the numbers are just an average, a guideline to go by so getting into specifics   

[Jef Lucero]

While I see merits in all of the points thus far made stating the case for options A–C, the one thing that I think we still need to consider (although Noel touched on it) is distinguishing between car ownership and simply choosing not to drive. I think we begin to confound the matter (and thus, perhaps, the data) using any model that at least gives the appearance that it fails to account for the fact that many if not most people that tend to commute via transit, bicycle or another mode still own cars but choose to leave them at home for commuting purposes. Maybe I'm somehow suffering from a disconnect or a blind spot, but from where I'm sitting those people are still paying for tabs, insurance, taxes and the like. I know some folks (Zach Shaner chief among them) that flourish car-free. But plenty of the rest of us that only occasionally cycle or walk to work see negligible savings in the additional areas expanded in options B and C.

I will say that the varying responses here speak (to me, anyway) to the import of specificity and customization to better define the individual's savings. If we're able to dedicate resources that would allow our customers to tailor their savings calculator from a menu of factors, I think the resulting data (and sense of ownership) would be much more meaningful to them.

I shared this discussion with one of our division's data specialists, and he emailed back this response:

"I really like the points that both Anne and Victoria make. $0.21 is too low and changing it at this point in the program is problematic for a variety of reasons.

My first thought went to evaluating the B and C options. I don’t like the differences here...It would seem that driving 50% more in a year only increases my costs by 30%. If fuel, maintenance, tires, insurance, license, registration and taxes are the only elements in that calculation then how can the rate per mile be different? Insurance discount for driving more? No. Discounts on license, registration, taxes, fuel or maintenance? Nope, I don’t get it. Perhaps I am missing something."

So many perspectives! Good talk, everybody. Happy Valentine's Day!

[Tracy Ellis]

You're welcome Stan for the can of worms.  But it's been a great discussion!

[Mark Hallenbeck]

So is there an option that allows a user to see two numbers?  Or choose between two numbers?  That is, tell them "you save this much in gas and maintenance" - but you could save THIS much if you consider how much longer your car will last if you don't drive it every day."

Another option would be to give a range of savings rather than one number.  In my experience, people generally react well to ranges.  (The people who want to see the small number, take the smaller end of teh range, and people who want to see a bigger number pick the higher end of the range.)  When presenting the number, you could say "you save between $0.21 and $0.37 per mile (or whatever) depending on the age and type of car you drive, how you pay for car insurance, and whether we consider the extra years your car will last if you don't drive it every day."  (Of course, you could shorten that sentence considerably too. :-)  ) 

[Stan Suchan]

Below you’ll find a spreadsheet that highlights final draft recommendations for changes to savings explained. As you’ll see, the changes are minimal. The most notable change is an increase in the drive alone cost per mile from $0.3383 to $0.37 (option B).

[Tracy Ellis]

On another note, can you explain how compressed work week stats are computed?  If someone works 4/10s and logs a CWW on M, T, W, Th; it's the fifth day - not logged, that actually is an eliminated trip.  Or, is it expected that the user log CWW on the one day when they don't work?

[AnneBruskland]

Compressed work week should be logged on the day that they don't work.  Zero fuel, emissions and cost attributed because this mode eliminates trip entirely.   In your example, they should log the mode and trip details used for the 4 days, and use compressed workweek option for the 5th day, when they do not work.