I'm a hater of long commutes (happily with a 1.5 mile one right now), but some of the analysis is flawed: 1) Commuting cost. The $0.17 as a minimum marginal co

I'm a hater of long commutes (happily with a 1.5 mile one right now), but some of the analysis is flawed: 1) Commuting cost. The $0.17 as a minimum marginal cost per mile is reasonable. The IRS' $0.51 really can't be used as a margnal rate. Simply put, the operating cost of a car is not a linear function of mileage. There are exceptionally high fixed costs, namely depreciation, insurance, (some) maintenance and registration. Personally, as I only drive 7000 miles a year (well below car average), my car ownership costs per mile slightly exceed the IRS reimbursement rate. But my marginal rate is still extremely low - if I drove another 1000 miles a year, it'd cost maybe $200 more ($0.2/mile). Attempting to average a minimum marginal cost and an optimistic estimate of car ownership per mile does not produce a valid estimate for average marginal cost. 2) Mileage to time conversion: I have no idea how he concludes that 1 mile = 6 minutes round trip. Living father away generally implies freeway utlization, where a marginal mile would be closer to 2 minutes round trip. (Anecdote: I live in SF. Thanks to the lack of freeway system within the city itself, a commute may be shorter (and use not much more gas) living 20 miles away from work in the suburbs than 7 miles away within the city borders). 3) Value of time. On a long commute, I'm not necessarily wasting time. Whenever I had one, I deferred actual reading I would otherwise do to podcasts. I've always found the main benefit of a short commute is flexibility (go home and back in the middle of the day) and reduction of stress (traffic sucks). It is awesome to have a shorter commute, but in no way is it worth spending nearly $250k more on a house solely to live 15 miles closer to work for a typical person.