The game is afoot: call it the “Mystery of the Teleporting Commuter”. Starting in October 2016, the amount of gasoline supplied to the US market started to decline on a year over year basis. This negative trend accelerated in January, leaving both energy analysts and macroeconomic pundits to wonder if the US has entered a stealth recession.
We regularly look at US gasoline production in relation to the Department of Transportation’s “Miles Driven” data, and when you add that variable to the mix the mystery starts to clear. Even though gasoline supplied was down 1.6% in October and -0.3% in November, the DOT data (November last month available) shows miles driven up 1.6% and 4.3%, respectively. By this math, imputed fuel efficiency for the US fleet is improving noticeably compared to historical trends that date back to the 1970s – a trend that is worth watching in 2017.
And while that’s an important piece of the puzzle, the real upshot is the gasoline data alone is not enough to conclude that the US economy has grabbed a lower gear or slipped into reverse. The miles driven data shows we can (and probably are) still cruising along.
Are you an American male between the ages of 35-54? If so, congratulations (of a sort). According to the US Department of Transportation’s Federal Highway Administration you drive more than any other age cohort/gender demographic. A lot more, as it turns out: you clock an average of 18,858 miles/year, some 14% more than the national average of 16,550 miles/year. The group least likely to get behind the wheel: women over the age of 65, who drive an average of 4,785 miles/year. If you are in neither camp, you can see how your demographic stacks up here: https://www.fhwa.dot.gov/ohim/onh00/bar8.htm
Regardless of where you are on that continuum, Americans do a lot of driving. In fact, over the last 12 months ending November 2016, we put a collective 3,215 billion miles on the odometer. That is the equivalent of driving back and forth to the Sun every 3 weeks. And since most Americans (86% according to a 2009 Census study) commute to work alone in their car, taking an average of 25 minutes door-to-door, driving is closely correlated with employment levels. All those miles behind the wheel are therefore an important economic indicator.
It is for this reason that a Goldman Sachs note out this week about the decline of gasoline supplied to the US market caught a lot of attention. Less gas consumption should, logically, mean fewer people doing the daily commute. Not to mention shopping, going out to eat, and everything else Americans like to do.
- Zerohedge had a note out this morning that several clients/contacts mentioned to me – see here: http://www.zerohedge.com/news/2017-02-07/goldman-stunned-collapse-gasol…
- CNBC featured it in their afternoon reader email and on the website here: http://www.cnbc.com/2017/02/08/gasoline-demand-recession-indicator-spru…
- If you are in the mood to data hound, the raw data from the Energy Information Association is here (under “Finished Motor Gasoline) : http://www.eia.gov/dnav/pet/pet_sum_sndw_dcus_nus_w.htm and longer time series is here: http://www.eia.gov/totalenergy/data/monthly/#petroleum
The upshot is this: implied gasoline supply to the US based on data from the EIA was down 5.2% in January. These are levels consistent with a recession, even though Goldman is careful to point out that they don’t expect one is in the works. So what’s going on?
There is another data set to consider: the Department of Transportation’s “Traffic Volume Trends”, which estimates the total miles driven on American roads through 4,900 sensors placed on both rural and urban arterial roads. Squaring that information against the gasoline supplied data from the EIA yields some useful observations:
- Gasoline supplied – the number everyone is worried about – has actually been declining on a year over year basis since October 2016. The comps are: October (-1.6%), November (-0.3%), and December (-2.4%). We use the EIA’s monthly data, which is less noisy than the weekly data that is getting all the attention, so January 2017 is not available yet.
- The DOT’s “Miles Driven” data tells a different story. October 2016 miles driven were up 1.6%, and November’s comp to prior year was +4.3%. You can see all the data on the DOT website here: https://www.fhwa.dot.gov/policyinformation/travel_monitoring/tvt.cfm. &…; Given that the National Retail Federation estimated Holiday 2016 at a 4% comparison and employment growth was positive in the month, we doubt miles driven in December 2016 was lower than the same month in 2015.
- Put the two items together, and you get a noticeable increase in implied fuel economy for the US fleet over the last 120 days. We’ve tracked this statistic back to 1973, and average US fleet miles per gallon has in fact been increasing gently over the years. It now sits at 23 MPG, where it was just 13 MPG back in the 1970s. The increase comes from Federal regulation of Corporate Average Fuel Economy, which requires auto manufacturers to sell a product mix with a specific miles per gallon goal. CAFE standards have risen over the years, resulting in better average economy.
Long term trends to better fuel economy aside, the recent drop in gasoline consumption alongside ever-greater “Miles driven” is a notable trend to follow in the coming months. The US vehicle fleet (cars, pickups, minivans, SUVs, etc.) totals over 250 million units, and (at most) 17 million clunkers scrap out every year, to be replaced with 17 million new (more efficient) vehicles. Now, we have had several years of very good vehicle sales during a time when CAFE requirements were rising quickly. Incremental fuel efficiency is therefore certainly a part of the story. As for how much, we will have to monitor the data through 2017 to see how this develops.
The bottom line, however, is that the gasoline supply data is only part of the story, and the DOT statistics tell the rest. Miles driven can climb even when gasoline shipments decline. It happened as recently as Q4 2016. It probably happened in January 2017. Sure, there will be a recession at some point. But it probably didn’t start in January.