Thursday, November 3, 2011

On Economic Data and Time Scales

I was looking at the excellent Bonddad blog today, and came across the post, "Uh oh: YoY gasoline usage down 5%, worst since October 2008."A small piece is quoted below.

US Gasoline demand
Something's happening here, but what it is ain't exactly clear. The most obvious candidates are:

1. demand destruction. But if so, why is consumer spending, as measured by the Gallup daily survey, holding up so well?
2. energy efficiency. But have we really bought so many hybrid vehicles to make that big a difference?
3. the weather. OK, we did have a strange Nor'easter that pummeled the northern and western suburbs of the Megalopolis, but that was one day only.
4. random stuff just happens. Always a possibility, but this seems unlikely given at least three weeks in a row of awful YoY comparisons.

 Now, I don't want to pick on Hale Stewart, the author, or his commentators, but I need to point something out here because this is a narrative I have heard many times in the last 3 weeks. Reduction in gasoline demand is not due to increased fuel efficiency. I don't doubt the fleet is getting more efficient, but this narrative completely ignores appropriate time periods and this is something I see quite often in the blogosphere. A reduction in demand due to fuel efficiency will be measured in decades, not months. If there is changes in the monthly numbers, it is either a different factor or noise, but the effect of increased fuel efficiency will not be measurable in the weekly data. Why? Because, at current sales rates and assuming no net growth in fleet, it would take roughly 23 years to turn over the US automotive fleet.

Auto sales as % of fleet, estimated fleet growth and replacement rates

Instead, I suggest we look for a more common-sense narrative and look at hotel occupancy, since Americans love to drive and, since we're talking gasoline sales, we need not worry about trucking.

Hotel Occupancy rate, via Calculated Risk

Hmmmmm. Notice anything? Like hotel occupancy and gasoline demand both coinciding in their drops?

Let's use some more common sense here guys and realize that different things happen in different time scales and if you are not paying attention to that, you are lying to yourself and making worse decisions as a result.



UPDATE-1: This is not a criticism of Hale! This is a a comment regarding the four commentators that attributed the drop to fuel efficiency. This is a narrative I have heard from many people in the last 3 weeks and it is WRONG.


  1. Oooo! This post is related to the Economic data and time scales, Yes learned about that on the economic subject. I am not an expert on economics, and I selected the Affordable assignment writing service for solving economic assignments plus this stage has served thousands of students globally.

  2. 스포츠토토티비

    Hi! This is my first visit to your blog! We are a team of volunteers and starting a new project
    in a community in the same niche. Your blog provided us useful
    information to work on. You have done a outstanding job!

  3. 스포츠토토

    Thanks designed for sharing such a nice thought, paragraph is nice, thats why i have read
    it entirely

  4. Its not my first time to visit this site,
    i am visiting this website dailly and take nice information from here all the time.


  5. Very fascinating story, it is difficult to find interesting and relevant content among blogs now. That is why I will definitely share your blog with my friends. I wish you success!
    Site: Wilsons Adventure Bound Leather Bomber Jacket


Do the right thing.