- Global Reach Blog - http://globalreach.blogs.census.gov -

A Mountain of Data

Posted By rosannatorres On February 2, 2010 @ 6:00 am In Foreign Trade Data | 2 Comments

By: David Johnson

Many of you know how important the big 2010 Decennial Census [1] is, and how it’s just around the corner. The data collected will be used to make projections and aid budget decisions for the next ten years. Sometimes, it can be hard to understand how data collected in 2000 is still relevant in 2009 but the task of collecting information on every man woman and child and then processing it is so monumental it couldn’t be done more than once every ten years.

Mountain of Data [2]Collecting Trade Data

The situation is the same for collecting foreign trade data, though we in the Foreign Trade Division are fortunate enough to have many systems in place that allow easy submission of shipment data through the Automated Export System, Automated Broker Interface, and automated review programs that simplify the process. The question remains though, what is going on in those 40-45 days ( release schedule [3]), from the end of a statistical month until the data is ready for the public?

So Much Data, So Little Time

Just to give you an idea on volume, the port of Los Angeles, the United States’ busiest container port, handled 7.85 million TEU [4]‘s (twenty-foot equivalent units) in 2008 (a list of the world’s busiest ports [5] on wikipedia). That’s over half a million units per day at one port! Now, if each of those containers had just one import or export document filed for it, we would still have a mountain of data, but often there are multiple commodities and parties involved for each container. Extrapolate this to all of the other ports in the U.S., and you start to get an idea of how much data we are talking about and that’s just for ocean shipments!

We Check then Double Check and Even Triple Check

The amount of the data that we receive at the Foreign Trade Division, and all of the Census Bureau for that matter, is mind boggling. Even if 99% of all the data came in completely correct, there would still be billions of dollars in shipments that are being filed with errors. These errors can range from incorrect commodity classifications to submitting the value of a shipment in foreign currency to any number of problems. One of the major reason’s why this month’s data won’t be available for another 40 some days is that we are busy using automated techniques as well as personally contacting filers to verify the accuracy of the data we receive.

It’s our Job to Produce the Best

Every day, analysts assigned to specific groups of commodities, import metals for instance, comb through datasets that are produced from the master set of all of the month’s trade data. These sets contain records that may or may not be correct, but have tripped up the electronic flag that says: “maybe we should take a closer look at this one”. In the end, what this means for the trade community is that even though the data isn’t coming hot off the press the first day of a new month, it has been thoroughly reviewed, approved, and contains some of the most accurate trade data in the world.


(Visited 630 times, 1 visits today)

2 Comments (Open | Close)

2 Comments To "A Mountain of Data"

#1 Comment By Ellen On February 3, 2010 @ 5:53 am

I work in a library that gets a lot of trade questions and one that periodically comes up is: how data is reported dollars v. measure (tonnage, etc.). Other than at the commodity level, does Census (or anyone) report higher level data by the measure as opposed to dollars?

#2 Comment By Global Reach David On February 3, 2010 @ 9:17 am

“I work in a library that gets a lot of trade questions and one that periodically comes up is: how data is reported dollars v. measure (tonnage, etc.). Other than at the commodity level, does Census (or anyone) report higher level data by the measure as opposed to dollars?”
Each 10 digit commodity code is assigned a unit of measure (unit of quantity) that will be most informative to the industry when it is aggregated for the month or year etc. This can be inconvenient at times when a product is being invoiced by piece but the unit of quantity is in kg however, a box of bolts from one company may be 10 times the size of a box of bolts from another company.
Commodity headings and subheadings may contain various units of quantity at the ten-digit level and will not allow for a simple additive summary. My best suggestion would be to use a resource like USATRADONLINE.GOV or DATAWEB.USITC.GOV and extract the data at the ten digit level and summarize it in a program like Excel, making sure to account for differences in quantity.
I hope this answers your question and I am sorry it isn’t more convenient to get the higher level quantity information. Also, there may be a third party company that provides this service, but I am personally unaware of one.

Article printed from Global Reach Blog: http://globalreach.blogs.census.gov

URL to article: http://globalreach.blogs.census.gov/2010/02/02/a-mountain-of-data/

URLs in this post:

[1] 2010 Decennial Census: http://2010.census.gov/2010census/

[2] Image: http://globalreach.blogs.census.gov/files/2012/04/6a0120a61b56ed970c0120a826053a970b-pi.jpg

[3] release schedule: http://www.census.gov/foreign-trade/faq/gen/gen0007.html

[4] TEU: http://en.wikipedia.org/wiki/Twenty-foot_equivalent_unit

[5] list of the world’s busiest ports: http://en.wikipedia.org/wiki/List_of_world%27s_busiest_container_ports