Shift share is a standard regional analysis method that attempts to determine how much of regional job growth can be attributed to national trends and how much is due to unique regional factors. Shift share helps to answer the question “Why is employment growing or declining in this regional industry, cluster, or occupation?”
Shift share compares this picture of “expected” change to the region’s actual change during that time. The difference between the two is one measure of regional performance. Shift share analysis is practical because it provides context for regional job growth.
Shift Share Components
- (Date Range) Change– The difference between the number of jobs within the selected date range.
- EITHER Ind. mix effect OR Occ. Mix effect– If we look only at nationwide trends this number is what we predict the (date range’s) change to be
- National growth effect– If we looked solely at the growth of the economy as a whole this number is what we predict the (date range’s) change to be
- Expected change– combines the (Ind./occ.) mix effect & nation growth to show what this region should look like if it was completely statistically average. Change above this level is credited to the region’s competitive effect.
- Competitive effect– difference between the actual change and the expected change. Provides a numerical value for how unique the areas is
Shift Share Components
To help make the following explanation clearer and more concrete, we’ll assume the following facts as the basis of a shift share scenario:
- The National economy grew by 4% (total employment) in the given timeframe.
- The Employment activities industry grew by 15% nationally, and by 350 jobs regionally. It had 1000 total jobs regionally at the beginning of the given timeframe.
- The Apparel Manufacturing industry declined by 5% nationally and by 80 jobs regionally. It had 200 total regional jobs at the beginning of the given timeframe.
The National Growth Effect
The national growth effect explains how much of the regional industry’s growth is explained by the overall health of the national economy: if the nation’s whole economy is growing, you would generally expect to see some positive change in each industry in your local region (the proverbial “a rising tide lifts all boats” analogy).
So if the entire national economy grew at a rate of 4%, we might have expected the regional Employment activities industry would also grow by 4%, or 0.04 * 1000 = 40 jobs. These 40 jobs are the national growth effect for Employment Services. For Apparel Manufacturing, the national growth effect is 0.04 * 200 = 8 jobs, meaning that we might have expected it to grow by 8 jobs over the time period simply because of general economic growth.
The Industrial Mix Effect
The industrial mix effect represents the share of regional growth explained by that industry’s growth at the national level. To arrive at this number, the national growth rate of the total economy is subtracted from the national growth rate of the specific industry, and this growth percentage is applied to the regional jobs in that industry.
In our example, Employment activities grew by 15% nationally, but we subtract the 4% growth of the national economy to arrive at a national industry-specific 11% growth rate for Employment activities (the industry’s national growth that exceeded overall trends). Applied to the regional industry, we would thus have expected Employment activities to grow by (0.11 * 1000) = 110 jobs due to industry-specific trends at the national level. Similarly, we get a national industry-specific relative growth rate of (-5% – 4%) = -9% for Apparel Manufacturing (i.e., the industry not only declined 5% nationally but failed to grow 4% with the rest of the nation), meaning we would have expected a regional loss of (0.09 * 200) = 18 jobs due to national industry-specific trends.
The Regional Competitiveness Effect
The regional competitiveness effect is the most important of the three indicators, as it explains how much of the change in a given industry is due to some unique competitive advantage that the region possesses, because the growth cannot be explained by national trends in that industry or the economy as whole. This effect is calculated by taking the total regional growth and subtracting the national growth and industrial mix effects. Note that this effect can be higher than actual job growth if national and/or industry mix effects are negative while regional growth is positive. This is because the regional competitiveness effect accounts for jobs “saved” from declining national trends as well as new jobs created.
So in our example, Employment activities grew by 350 jobs regionally, but 40 of those jobs might have been expected due to national trends in the economy as a whole, while 110 jobs might have been expected due to national trends in Employment activities specifically. This makes a total of 150 jobs expected from national trends. Since the actual growth was 350 jobs, (350 – 150) = 200 jobs cannot be explained by national trends, and so they must be attributed to unique conditions and advantages that the region possesses which contribute to the growth of this specific industry.
For Apparel Manufacturing, we might have expected a net change of (8 + (-18)) = -10 jobs regionally, while in fact there was a regional change of -80 jobs. The regional competitiveness effect is thus (-80 – 10) = -90 jobs, indicating that it fell short of the expected change by 90 jobs due to some specific conditions in the region, such as the closing of a factory.