The Brookings Institute and JPMorgan Chase carried out this joint project because the statistical data from the US Census Bureau led to unsatisfactory conclusions about U.S. exports.
Export Nation data, used in this study, determines export estimates based on the origin of production. In contrast, the Census Bureau overestimates the importance of certain port cities, which simply serve as the last stop for goods before they are exported. This study defines this choice of measurement as “origin of movement” to paint a contrast from their “origin of production”. Another important distinction is that Export Nation is more meticulous when it comes to metro-level services export data. The Bureau of Economic Analysis (BEA) produces service industry export data primarily at the national level.
The Administration’s goal is to double exports in five years. From any method of measurement, this goal does not seem feasible. According to this study, specialization is highly correlated with export growth rates. However, extreme specialization also reveals vulnerabilities and these cities may experience plunges in export rates when the economy worsens. Some industries, and thus some cities, suffer more than others, dependent on what they specialize in. For these reasons, I think it might be interesting to conduct another study and control for 2009 as a crisis year.
The study offers various tables, some I like more than others. One of the charts of these is titled “Largest Metro Area Exporters by Volume, 2012”. My criticism is that they group metropolitan regions in creative ways. For example the New York Metropolitan region includes (parts of?) New Jersey and Pennsylvania. I understand including parts of NJ in the New York Metropolitan region, but the PA extensions exaggerate the results. In all of the other tables, when New York is listed, it is labeled simply as “New York”. Other conglomerates such as Boston-NH also exist in this same table. Interestingly, cities in Texas and California are listed as distinct locations, even if they are geographically close. Why is there not a single term for the San José, San Diego, and Los Angeles Metropolitan region?
Another complaint is the mention of “global export intensity levels” which translates into meaning the export share of GDP, which in turn is also completely absent of the use of firm-level export data. When cities are ranked as being below average global export intensity levels, it means nothing. Sure these cities may have “high” infrastructure and innovation endowments, but the size of the domestic market is the variable that should be by far the most important. It is wonderful that the capital cities of Singapore and Copenhagen might have high export intensity levels. However, U.S. cities can thrive by diverting large flows of goods directly to the domestic market. Engagement amongst U.S. cities is every bit as dynamic, just other variables become more important than those traditionally studied in other projects. The correct approach would be to stratify countries and regions into different classes and then compare them across the class in which they belong. I have little patience for studies that attempt to make a “serious” cross analysis of countries that are as diverse as Brazil, Rwanda, Bhutan and France. Maybe Bhutan does not need the same institutions as France to run efficiently? Maybe Rwanda does not have a port? Maybe history does not predispose Brazil to trade with countries on the other side of the continental divide?
The table of page 7 is labeled “Most Export-Intensive Metro Areas, 2012”. The cities included were not what I expected to see. Greenville, SC rounds out tenth place. I think this table would be more informative if the percentages of export intensity were converted into percentages of growth of export intensity for a particular city over the entire level of federal export intensity growth for that time period. It is very easy for a city that captures .003% of the national export volume to quadruple its share of national exports in a short amount of time. It is much harder for major cities with large export volumes to make large export percentage increases. This is another example of comparing the raisins to the watermelon.
There is a table on page 8 labeled “Metro Areas with Highest Growth in Exports as a Share of GDP, 2003-2012”. I view this table as a more realistic depiction of reality. Since 2003, there have been large percentage increases in export shares of GDP for particular cities. Salt Lake City and Ogden, Utah top the list. Provo, Utah also ranks seventh. The other interesting grouping is New Orleans and Baton Rouge as Louisiana cities ranking third and fourth. Also, nearby Houston, TX and Little Rock, AR are fifth and ninth respectively. Lastly, southern cities such as Raleigh, NC and Charleston, SC place sixth and eighth. The only liberal area that makes the cut is Seattle, Washington coming in tenth. Considering that this study includes services as an export and that it is a nine-year snapshot of local level data, the results are concerning for more traditional export hubs of the country.
Some observations I want to make about other sectorial export trade figures are the following: I wonder if the funds that the Federal Highway Administration received ended up facilitating manufacturing export growth. What happened to leftover materials that suppliers did not funnel to rebuild highways and infrastructure? Were they exported? Also, food products as a percentage of export growth for US total export growth declined, as indicated on page 10. This decline was both in numerical and percentage terms. I wonder what countries captured the U.S.’ share of food product exports and which export markets were “lost”.
In addition broader trends indicate that in the post-recession, the service sector is growing less than it did before. Manufacturing and Engineering sectors are up, the latter being the fastest-growing export sector in the U.S. The cities in which the services sectors grew the most from 2009 to 2012 include some traditional cities together with some new ones. Three Floridian cities make the list, Orlando, Cape Coral, Miami and Jacksonville. Also Honolulu in Hawaii makes fifth place.
On the same page there is a table labeled “Major Industries with Fastest Export Growth Rate, Top 100 Metro Areas, Pre- and Post-Recession”. I find these results extremely disturbing. It seems as if the categories have almost been “refined” to measure the post-recession data. I would have liked to see a total list of industries used and to ensure that their labeling and the components of each label remained consistent from 2003 until 2012. The petroleum and coal products industries are the only two that make the top-10 list for each interval. One last remark would be that reanalyzing the data in 2014 might provide better results. Certain sectors such as financial services, insurance services, and management & legal services may require longer term contracts, and in effect, income derived from projects would not be immediate.
Again, the study in its entirety can be found here.