[eddieplan9]

It's important to not use the IRS Mileage Rate to calculate real cost. (The linked study does not.) The IRS rate is designed for the worst case scenario for the purpose of tax filing (coz you don't want to penalize someone who has to drive a F-150 in a high-gas-price state in a year where gas price spikes to $5/gallon). Even if we don't consider the fact that the cost of vehicle depreciation, maintenance and insurance is an "almost-fixed" or marginal cost because you are going to own the vehicle and pay for the insurance any way, and let's be generous about estimating the cost here:

At IRS rate of $0.54 a mile, let's say we buy a Prius (which is the most popular car used for ridesharing), for $23,000, and drive it for 100,000 miles and throw it away.

At IRS rate, that's a cost of $54,000.

Gas cost is about $8,000 (50MPG [1], $4/gallon). Insurance $1500 x 5 years ($1200 base insurance + $300 extra for rideshare add-on [2]). That still leaves us $15,500 (28.7% of estimated cost) to cover repair deductible etc. That's almost enough to buy another Prius.

Edits: I mistakenly said the study used IRS Mileage Rate. It does not.

[1] http://mikes-review.com/does-a-toyota-prius-really-get-50-mi...

[2] https://www.nerdwallet.com/blog/insurance/best-ridesharing-i...

[ghaff]

The paper actually uses a $0.30/mile figure. "A Median driver generates $0.59 per mile of driving, and incurs costs of $0.30 per mile."

The paper actually seems to be trying to make the point that Uber drivers are gaming the tax system because their actual costs are lower than the IRS allowance.

$0.30 seems like a fairly reasonable number and it also matches up well with renting a car that's intended to be driven for a week like this and just buying gas for it. (There's also some cost associated with deadheading to the next pickup.)

The $0.59/mile revenue does seem a bit low. UberX or Lyft would seem to typically have pricing in the neighborhood of $1.00 to $1.50 per mile of which about 75% goes to the driver. So $1/mile would seem likely to be a better revenue number in which case the net is more like $0.70/mile and I would have to believe most drivers, even in a larger city, are driving 10 miles or more in a typical hour.

I doubt driving for these services is a huge win once all costs are taken into account but it seems closer to being a minimum wage job than a total bust.

[eddieplan9]

Thanks for pointing out my mistake. You are right. I edited my comment.

To be fair, if the point is to compare profit, then we also need to consider that a traditional worker making minimum wage also needs to commute to work. The average commute of a worker is 30 miles round trip [1]. The cost of that will eat into their earning as well and cannot be deducted from their taxes.

[1] https://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/pu...

[ghaff]

It's sort of a weird point which, if it belongs anywhere at all, probably belongs in a different paper. It's a reasonable question to ask whether IRS guidelines that were mostly designed in the context of standard business deductions are really appropriate for driving services where most drivers are buying cars that are optimized for the role.

But it's also pretty much a totally orthogonal point to one about how little Uber drivers make.

[mhneu]

Disagree. First, time value of money. The Prius costs $24k that you don't get to invest over the time you're driving. Let's call that $1-2k. Plus, most drivers will pay interest on financing the car. ($1-5k depending on credit rating). Then add: Car washing/cleaning (and/or time spent doing so). Repair (as you noted). Maintenance: oil changes, wiper replacement, tire changes. (In the opposite direction, depreciation is overestimated because you're throwing the car away after 5 years).

That gets your $15.5k down much closer to zero.

The article's bottom line is that Uber/Lyft drivers are getting paid way, way below minimum wage when all expenses are taken into account. That's left intact by the above calulcations, even if the numbers vary by 20% high or low.

[sokoloff]

> First, time value of money. The Prius costs $24k that you don't get to invest over the time you're driving. Let's call that $1-2k. Plus, most drivers will pay interest on financing the car. ($1-5k depending on credit rating)

You can't assume that the owner has paid cash for the car (forgoing investing the $24K) AND assume they finance the $24K and ding their P&L with finance charges. (It's double counting.)

[jogjayr]

> Even if we don't consider the fact that the cost of vehicle depreciation, maintenance and insurance is an "almost-fixed" or marginal cost because you are going to own the vehicle and pay for the insurance any way,

But all of these costs go up if you drive more miles. Your vehicle depreciates more if you put more miles on it (but not if it's parked in your garage), you need more frequent oil and tire changes, and your insurance rates go up because you're more likely to make a claim the more miles you drive annually. And rideshare insurance is more expensive.