OLMIS

by Pat O'Connor
Published Jan-30-2012

Introduction GDP Growth in Oregon¿s Metropolitan Areas

The U.S. Bureau of Economic Analysis (BEA) produced experimental estimates of GDP for the nation's 366 metro areas late in 2007. BEA has long produced estimates of state and national gross domestic product (GDP), but estimates for metro areas are a fairly new experiment.

GDP measures the market value of final goods and services produced within an area in a particular period of time. It is a widely used measure of economic activity at the state and national level. GDP is BEA's preferred and most comprehensive measure of economic activity.

Metropolitan statistical areas, defined by the U.S. Office of Management and Budget, are standardized county-based areas having at least one urbanized area of 50,000 or more population, plus adjacent territory that has a high degree of social and economic integration with the core, as measured by commuting ties. Oregon has six metropolitan statistical areas.

BEA's statistics on GDP by metropolitan area can be used to answer a variety of questions, and also to inform business and policy decisions. However, users should be cautious because the estimates are experimental. Some of the questions the GDP data can answer are:

• What is the size of a metropolitan area's economy?
  
• Is a metropolitan area's economy growing or declining?
  
• What industries are driving a metropolitan area's economic growth?
  
• How does the growth in one metropolitan area's economy compare with growth in other metropolitan areas, with the state, with the nation?
  
• What has been the trend in economic growth over time in a metropolitan area's economy?

Table 1 shows how Oregon's metro areas rank nationally in terms of GDP size among the 366 metropolitan areas in the U.S. The Portland metro area is the only area in Oregon that ranks in the top 100 metro areas. Salem and Eugene are both towards the middle of the pack, ranking 153rd and 157th, respectively. Medford, Bend, and Corvallis all rank in the lower half of the nation's metro areas.

Table 2 shows that the 366 U.S. metropolitan areas as a whole expanded 17 percent between 2001 and 2008. All of Oregon's metropolitan areas exceeded that growth. The slowest growing metropolitan areas in Oregon were Salem and Medford which both had 18 percent growth. The fastest growing GDP in Oregon was the Corvallis MSA which showed 90 percent growth from 2001 to 2008. In fact, the Corvallis MSA had the fastest GDP growth of all 366 metropolitan areas in the U.S. between 2001 and 2008. This must have been a real surprise to Benton County residents. It certainly was a surprise for yours truly, who covers the Corvallis MSA economy quite closely. Unfortunately for the Corvallis MSA, its remarkable GDP growth highlights a flaw in BEA's methodology rather than break-neck GDP growth in the county. We need to look closer at BEA's methodology to find out why the Corvallis MSA's estimated GDP grew so rapidly.

In the December 2007 edition of the Portland Metro Labor Trends, Amy Vander Vliet cautioned readers of potential problems with BEA's methodology and its ability to accurately reflect GDP growth in metropolitan areas. GDP data is collected at the state level, not the metro level. The method BEA devised to allocate a state's GDP among its metro areas uses two data sets: statewide GDP by industry and county-level earnings by industry. The state's GDP within each industry is allocated to counties based on county earnings data for the given industry.

One of the components of GDP is investment in capital equipment; two examples would be a new building or new machinery. Some industries, such as manufacturing, spend heavily in capital investments. This is especially true for high-tech manufacturers, auto makers, airplane manufacturers, and oil refineries. An example in Oregon was in 2002 and 2003 when Intel spent nearly $2 billion to build and equip a plant in Hillsboro.

This is where we start to see the problem for Oregon's metro area GDP estimates, in particular the Corvallis MSA. BEA admits that there is a weak correlation between earnings and economic output for some capital-intensive industries. Let's take a real life example; let's say capital spending in high-tech manufacturing increased about $500 million in Oregon in 2004 mainly due to capital spending in the Portland area. In the same year the Portland area showed a decline in workers' earnings in high tech. Meanwhile, Corvallis didn't experience significant capital spending but it did see an increase in its high-tech earnings. With Corvallis's earnings growing and Portland's shrinking, Corvallis receives a larger slice of the state's GDP for high tech manufacturing. However, the capital spending that occurred almost exclusively in Portland is spread around the state based on county earnings in high-tech manufacturing. The end result in this example is that Portland's GDP growth in high-tech manufacturing is understated while Corvallis's GDP in high-tech manufacturing is overstated. It shows up much more dramatically in Corvallis's total GDP because high-tech manufacturing in the Corvallis MSA comprises a larger percentage of the area's total GDP compared with the Portland area.

Although this example illustrates a shortcoming in BEA's methodology, sub-state GDP estimates can be useful for many states in the U.S. and their metropolitan areas. For most industries total earnings in an area is a good proxy for economic output. However, we have seen that this is not always the case in manufacturing.

Oregon has a relatively large manufacturing sector compared to most states in the U.S. Due to Oregon's large manufacturing sector the potential for misallocations, as illustrated in the example above, is larger in Oregon than for other states. BEA's new data series will likely be useful for many metro areas in the U.S., but it appears to present some challenges for analyzing Oregon's metropolitan areas.