Marchetti's Constant and World City Average Commute Times

[coatimundi]

Came across this BBC article the other day that talked commute times, and mentioned Marchetti's Constant (30 minutes), which I had never heard of.

http://www.bbc.com/future/story/20161202-what-your-commute-looks-like

Marchetti's idea is that people have a daily travel budget of around an hour that they choose to spend in different ways, picking transport options that fill up that time. If we live close to work, we might walk or cycle. If roads or public transport improve, we might move further away. But whether we drive, cycle, walk or take public transport we will spend roughly the same amount of time doing it.

The graph was interesting to me: according to it, the top 3 US cities for commute times are New York, LA and Detroit, in that order. I guess Detroit makes sense because the majority of its white collar population seems to live well outside the city.

[sparker]

Came across this BBC article the other day that talked commute times, and mentioned Marchetti's Constant (30 minutes), which I had never heard of.

http://www.bbc.com/future/story/20161202-what-your-commute-looks-like

Marchetti's idea is that people have a daily travel budget of around an hour that they choose to spend in different ways, picking transport options that fill up that time. If we live close to work, we might walk or cycle. If roads or public transport improve, we might move further away. But whether we drive, cycle, walk or take public transport we will spend roughly the same amount of time doing it.

The graph was interesting to me: according to it, the top 3 US cities for commute times are New York, LA and Detroit, in that order. I guess Detroit makes sense because the majority of its white collar population seems to live well outside the city.

Seems to me that Marchetti's constant has come to be more of an idealized reference point; many of us out here in CA would consider 30 minutes each way on the road to be a desirable goal, but one which may not be practically achievable.  I think the concept itself is valid, but a more robust study should be done as a follow-up -- one than controls for (a) varying property values in center cities and suburbs (b) regional income differentials (c) extensiveness -- and efficiency -- of public transit in the study region, and (d) levels of automotive ownership in that same region.  Couple that with a few surveys (each with a decent-sized n) about localized commute attitudes, and you may well have an updated take on the general Marchetti concept.

[Duke87]

Seems to me that Marchetti's constant has come to be more of an idealized reference point; many of us out here in CA would consider 30 minutes each way on the road to be a desirable goal, but one which may not be practically achievable.

And the same thing is seen in the Tri-State area around NY, where the size of the metro area combined with the impracticality of driving into the central business district makes a commute of 30 minutes or less difficult for most people to achieve.

Something that NYC and Los Angeles do have in common, though, is that both are perceived as special places where many people want to live simply for the sake of living there. So perhaps a better way of looking at this is that while the growth of any generic city may be capped by 30 minute commute availability, people will accept longer if they feel they get something in exchange for it - or, alternatively, helplessly surrender to it if they happen to be born in one of these metro areas and are not able to relocate.

[GaryV]

The graph was interesting to me: according to it, the top 3 US cities for commute times are New York, LA and Detroit, in that order. I guess Detroit makes sense because the majority of its white collar population seems to live well outside the city.

I'm not sure that was intended to be a list of the highest, or just an example.

Most jobs in metro Detroit are not in Detroit proper.  Most commutes are suburb to suburb.

[hotdogPi]

The graph was interesting to me: according to it, the top 3 US cities for commute times are New York, LA and Detroit, in that order. I guess Detroit makes sense because the majority of its white collar population seems to live well outside the city.

I'm not sure that was intended to be a list of the highest, or just an example.

Most jobs in metro Detroit are not in Detroit proper.  Most commutes are suburb to suburb.

Definitely not a list of the highest. If it was, the "constant" would be an upper bound, not an average.

[coatimundi]

Came across this BBC article the other day that talked commute times, and mentioned Marchetti's Constant (30 minutes), which I had never heard of.

http://www.bbc.com/future/story/20161202-what-your-commute-looks-like

Marchetti's idea is that people have a daily travel budget of around an hour that they choose to spend in different ways, picking transport options that fill up that time. If we live close to work, we might walk or cycle. If roads or public transport improve, we might move further away. But whether we drive, cycle, walk or take public transport we will spend roughly the same amount of time doing it.

The graph was interesting to me: according to it, the top 3 US cities for commute times are New York, LA and Detroit, in that order. I guess Detroit makes sense because the majority of its white collar population seems to live well outside the city.

Seems to me that Marchetti's constant has come to be more of an idealized reference point; many of us out here in CA would consider 30 minutes each way on the road to be a desirable goal, but one which may not be practically achievable.  I think the concept itself is valid, but a more robust study should be done as a follow-up -- one than controls for (a) varying property values in center cities and suburbs (b) regional income differentials (c) extensiveness -- and efficiency -- of public transit in the study region, and (d) levels of automotive ownership in that same region.  Couple that with a few surveys (each with a decent-sized n) about localized commute attitudes, and you may well have an updated take on the general Marchetti concept.

This is averages though. In the Bay Area, there are people that spend well over an hour one-way on their commute, but the average commute will always driven down by those who live close to work and spend just a few minutes on a bus, bike, walking, or even driving. I had a job in Tucson that was 7 minutes from my house at the time. But most people seemed to spend 20-30 minutes on the road there.
But I think the article is more pointing out the concept of those averages hovering around 30 minutes and not trying to purport yet another commuter study. If you want numbers, just Google it, because there are a lot of different ones, but they tend to hover around the same levels in that chart. And that's the point of the Marchetti Constant: the averages will hover around 30.
Now though, if we were to look at the median commute time, then I think this idea would fall flat since the Bay Area would likely be much higher. However, the Bay Area would also fall into the desirability bias, like New York and LA, where people are willing to tolerate a longer commute for the sake of living in a more desirable region with better jobs, better activities, better weather, etc.

I'm not trying to defend this guy's idea, but I think you're getting caught up too much in the numbers on the chart.

It does make more sense that they just threw up some cities though. On these lists, you typically see Chicago, Atlanta, DC and the Bay Area, and those missing probably makes it look a little off. I don't know why they would, of all cities, put up Detroit.