Ed Leamer is chief economist at the Ceridian-UCLA Pulse of Commerce Index and a faculty member at UCLA. Leamer took the time recently to discuss with HAI Managing Editor Drew Voros the Pulse of Commerce Index, which debuted in February 2010. The index provides real-time economic data based on diesel fuel consumption from over-the-road trucking and can serve as an early indicator of macroeconomic developments because of its instantaneous data collection, according to Leamer.
Hard Assets Investor: What does the Pulse of Commerce Index provide that was missing from other economic data points?
Ed Leamer: We just don’t have much in the way of evidence about trucking activities. For the U.S. or for any country, think of [goods and products] inventory as a big reservoir. It has a certain level in it. And there’s a flow out of it, which is sales, and there’s a flow into it as well, which is the replenishment activities. This inventory has primarily been measured in terms of the height of the reservoir, how much inventories are. And that’s done inaccurately. The small wiggles in inventories are pretty hard to measure.
HAI: Can you talk about the methodology and how the data is collected?
Leamer: This index is meant to be a measure of inventories in motion, inventories destined for some new location. What’s actually measured is the volume of diesel purchases. Whenever a truck, whose driver has a Comdata credit card, purchases diesel fuel, that credit card transaction is recorded within a half second on Ceridian computers in Nashville. That’s the database.
We use the volume of diesel fuel, aggregate that to the level of the country, and that’s our index. There are 7,000 U.S. truck stops that are being monitored by this index.
HAI: What does the Pulse of Commerce Index tell an investor?
Leamer: For example, investors want to know if we are going to go into a recession right now. Everybody is scared to death in the equity markets. It’s a crazy time. Investors need to be able to sort out what’s coming clearly with regard to the economy, because that’s critical in their short-term portfolio outcome. If you decompose the index of leading indicators, there are only three or four components that are really very good for telling you where we are and where we’re going.
HAI: What are those components?
Leamer: Well, the slope of the [Treasury] yield curve is important. Normally you would get an inverted yield curve going into a recession, which we haven’t had. We think manufacturing hours are important because you usually have manufacturers cutting overtime, and therefore having a decline in weekly hours, which we haven’t had.
And then the other important component is building permits. If you asked me if I were going to be cast away on a desert island and I wanted to have three things that warned me that a recession was coming to the United States, those would be the three.
The fourth would be this tracking index.
HAI: And the difference with this tracking index is that it’s simultaneous data?
Leamer: Exactly. These are real transactions. The slope of the yield curve is similar in the sense we know a lot about Wall Street on a second-by-second basis in terms of actual transactions. We know what the bond prices were yesterday. We know it on a second-by-second basis. So that’s a sense in which that measure is analogous. But we don’t know much about Main Street; you know building permits and weekly hours of manufacturing. That always comes out after the fact.
HAI: What’s the easiest place to find this index for someone?
Leamer: Go to www.ceridianindex.com. There are all kinds of data there about the index and what it implies.
HAI: Does anybody use it for actual trading purposes? For example, has anyone constructed an ETF off the index?
Leamer: Not that I’m aware of. That would be interesting. The data sets are extremely rich, both geographically and timewise, because you’ve got purchases at 7,000 truck stops on a second-by-second basis. My understanding is that some investment banks are making use of this information.
HAI: Can you kind of give me an example of how the index might have showed a trend or something missing from the other economic data points?
Leamer: I’m a forecaster and the biggest mistake I made was early in 2008. I said we’re not really in a recession yet. And the GDP number for the second quarter was positive. That’s one thing. But also the initial payroll numbers that came from the BLS [Bureau of Labor Statistics], they had declining payrolls, but at about 50,000 per month, whereas a real recession is more like 150,000 to 200,000 a month. So I say, well, this isn’t a real recession; it’s going to have to get worse in order for it to be classified as a real recession.
Wouldn’t you know, they revised the darn numbers. We didn’t have it at that time, but our index was declining very substantially. It was absolutely clear we were in big trouble second quarter of 2008. There was absolutely no ambiguity.
HAI: So the Pulse of Commerce Index tells you more about the movement of goods?
Leamer: Correct. The energy guys think it’s an energy index, but we don’t think of it that way. We think of it as inventories in motion. In other words, in order to get the product from point A to point B, you’ve got to buy the diesel fuel.
HAI: What’s the index telling you now?
Leamer: That August was down 1.4 percent and July was down 0.4 percent. So we’re in a situation that looks like a double dip. We’re having enough of a decline that if you didn’t know anything else, you’d be worried about a double dip. But I’m not worried about a double dip.
HAI: Why is that?
Leamer: All the other indicators that I look at don’t indicate that it’s a double dip. I interpret it as a period of very slow growth, not a real decline. For it to be a real recession, the index would have to continue its decline for several more months.
HAI: Does the index give you a sense of what gas prices are?
Leamer: Yes, it does that. When there have been big swings in diesel prices, I’ll put in the monthly report a paragraph that shows how diesel prices have changed. I was talking with the Department of Transportation this morning, who was keenly interested in this data set, too.
HAI: How does the truck industry use this information?
Leamer: Both Wall Street and truckers have a tendency to focus on year-to-year comparisons. If you’re an asset manager, the year-to-year can be totally misleading, because it could have been a huge up followed by a huge down and it made it look like you were OK year-to-year, but the last half of the year was really down.
It’s kind of interesting that trucking is much less on the weekends than it is on the weekdays. The peak day of trucking is Wednesday. Saturday and Sunday are about half the volume of Wednesday. And the reason that’s important is the number of Saturdays and Sundays in any given month varies from year to year. So if you don’t make a weekday correction, you get misled about the month-to-month changes.
A data set, because of daily data, allows us to do that. It allows us to correct for wandering holidays. With our daily data, we can make those kinds of corrections, and/or we can focus on the periods of time that are not affected by the holiday. You go plus or minus four or five days and the holiday affect is gone. Because we have daily data, we can do it that way. It’s a much richer data set for determining where the economy actually is.
HAI: On the same note, what season is the heaviest volume for fuel sales?
Leamer: October and August. October is the big month. By big, I mean average daily deliveries. Because there are days in early December that are actually the peak days. The second half of December has low volume because of the two holidays. So December as a month is not the peak. But if you ask me when the peak week was, it would probably be the first/second week of December. The peak month is October. And then August is the next one.