***Update 4/24, 4:00 p.m.:The original version of this post was wrong.

***Update 4/30: I wrote a very important addendumto this post.

heavy trafficThere’s a lot of confusion about the VMT provision in Washington’s new “Climate Action and Green Jobs” legislation (HB 2815). That’s to be expected because that section of the bill is rather confusing.

It calls for reducing per capita vehicle miles traveled. But reduce them from what level? And, is that a reduction in total VMTs, given population growth?

Well, readers, you’re in luck. Armed with my readin’ and countin’ skills, I’m going to try to figure out what the new law means. Buckle your seat belts.

First off, here’s the aim:

  • decrease the annual per capita vehicle miles traveled by 18 percent by 2020;
  • decrease the annual per capita vehicle miles traveled by 30 percent by 2035; and
  • decrease the annual per capita vehicle miles traveled by 50 percent by 2050.

But reduce VMTs from what baseline? The answer, says the bill, is 75 billion, minus the miles from trucks, which are apparently exempted. But wait: 75 billion is not a per capita figure! (At least, I don’t know anyone who drives 75 billion miles a year.) So what does this mean?

Here’s my best guess, based on the rumors one hears from policy wonks: the figure is connected to 2020. Mind you, the law never actually says this, but it happens that 75 billion is roughly the total VMTs that the state forecasts for 2020 (which, by the way, representsa 33 percent increase from where we are now). So, apparently, the “reductions” are to be from that 33 percent predicted increase.

Confused yet?

It gets weirder because, as Clark has astutely observed, the increase isn’t actually materializing. So the reductions come from a projection that we may never reach. My head is spinning.

Okay, back to business: what’s the baseline? The bill says 75 billion, but that’s nonsensical since it’s not a per capita figure.   

So who knows what the 75 billion is supposed to reference. But I’m going out on a limb here, and assuming that the intended baseline must be for the per capita figure that’s forecast for 2020 at which point the state says we’ll have 75 billion total VMTs. In 2020, we’re officially projected to drive, on average, nearly 9,800 miles apiece, as opposed to around 8,700 now. So I’m going to assume that the per capita reductions must come from that projected 9,800-ish figure. 

It’s a big assumption, but I’m stumped otherwise.

Moving on. Let’s say that the state achieves the per capita reductions—the ones relative to projected per capita VMTs in 2020. Would that be an absolute reduction, or not?

  • It would not be, at least not in the near future. At least not using the most reliable estimates for future population—estimates which are themselves notoriously unreliable. But by 2041, we’d finally start seeing reductions from today’s level.

    For the following calculations, I decided to use VMTs for 2006 as a baseline because they were the most recent figures for the entire state that’s been reported to the feds (and this suggests to me that they’re the most recent accurate figures). Translating the per capita VMT targets into projected total VMT targets, we’d get something like this:

    • increase the annual vehicle miles traveled by 10 percent by 2020;
    • increase the annual vehicle miles traveled by 7 percent by 2035; and
    • decrease the annual vehicle miles traveled by 15 percent by 2050.

    All of this referenced to VMTs in 2006.

    And here’s the curve of what total VMTs might look like:

     vmts - redux

    The blue line is today’s VMTs; the red line is the total VMTs that the new legislation appears to target.

    Okay, I’m asking for help. What the heck is going on?

    Please let me know about questions—and corrections!—if you see anything murky or unfounded here. And if you think the stuff above was messy, check out the rest of the methodology…

    Some additional confusions:

    • The VMT provision is not a strict mandate. Quoting from the bill’s “final report,” it’s a “collaborative process” to “make recommendations” that should include “a set of tools and best practices” for “making progress toward achieving these benchmarks.” That should be fun.
    • Another wrinkle: The bill says that it should be “consistent with the stated goals of executive order 07-02.” That executive order uses 2004 as a benchmark in several places for establishing climate targets. And while the order never mentions anything about VMTs, it’s curious that the VMT benchmark would be referenced to some predicted scenario rather than 2004. Wouldn’t a 2004 benchmark have been a lot simpler?
    • Yet another wrinkle: the bill says the baseline is 75 billion minus the mileage from trucks. But what’s that figure? Anyone? Bueller? WSDOT doesn’t appear to have a projection for that.
    • Finally — is anyone still reading? — there’s another way one could understand the law. The “baseline” of 75 billion could be the baseline for just the first target, which is set for 2020; or it could be the baseline for all of the target years. Should we to caculate per capita VMT reductions from the projections for each of the target years, meaning reductions from the forecasts for 2035 and 2050? Or should all of the reductions be referenced the the 2020 baseline projection, or what?

    Methodology:I assumed a straightline constant reduction in VMTs in each year between the legislated targets. There’s no particular reason to assume this would be the case, but it makes the analysis easier. “Today’s VMTs” means VMTs in 2006. For population, I used projection figures from the state’s Office of Financial Managment through 2030, the furthest year for which the state makes official projections. On the advice of an expert demographer, for 2031 to 2050, I assumed that Washington maintains a constant share of total US population, based on its average projected share from 2020 to 2030 of 2.32%. For US projected population, I used the most recent Census Bureau projections available. Importantly, population projections are very imprecise, especially at smaller geograp
    hic scales and especially several decades away. So my graph shouldn’t be taken to imply a huge amount of precision in the outlying years: it’s really more of an educated guess. All the legislative language is