Assignment (+link to assignment page)
||assignment posted by
||date due (tentative)
||Unit(s) of analysis
||suggested page length
||percent of grade
|2. Industry Analysis
||understand the structure and distribution of local economic sectors
||location quotients, shift-share analysis
||several data tables, sources, several paragraphs
What to turn in on Nov 22 (see below)
This assignment consists of three parts. (Be sure to read the instructions carefully and complete all parts, including Parts 1a, 1b, 2a, 2b, 3)
Sources of Data
You are to choose your own sources of data. Please download (or manually input) data directly from their sources (generally government sources). Here is a link to some commonly used sources (though you may use others).
Unit of Analysis
Sectors at generally a city, county (province) or region. Ideally you should use the geographic areas for your final project (though if the data is not available you can choose another location). If your final project area is at a small geographic scale, you may not find data at that geographic scale (e.g., at the neighborhood level). You might therefore choose a larger geographic unit that includes your final project subject area. (e.g., if your final project is on two Detroit neighborhoods and you can’t find economic data at that level, then you might use the city of Detroit, Wayne County, the metro area, etc.).
Part 1: Location Quotients
1a. location quotient data table
Calculate location quotients for at least 6-10 sectors [no need to do more] for your two places. For each sector, estimate the basic and non-basic employment levels. Be sure to include in the table the following information: sector employment, total employment, the calculated location quotient, and the estimated “surplus” (i.e., export) employment.
- What level of detail should you use? You have flexibility here, and can use either the 2-digit level, more detailed/disaggregated levels, or a combination of levels (in one table).
- For examples of a data table format, see the Isserman reading, p. 185.
- Be sure to include the year and source of the data, with complete titles and labels.
- Here is an example of a good (and bad) data table and general graphic design ideas.
1b. In a paragraph, comment briefly which sectors have high or low location quotients and the implications. (e.g., do these sectors represent unusually beneficial or problematic economic activities, e.g., future growth, wages, skill level, multipliers, etc.)
Part 2: Shift-Share Analysis
2a. Shift-Share Analysis Table
Compute the components of shift-share analysis for at least 6 sectors in each of your two places (ideally use the same sectors as in Part 1). Select two time points (two years: x and y).
Be sure to include in the table the following information:
- sector employment and total employment in years x and y (both for your places and the nation) and the resulting growth rates.
- the national growth effect, the industry mix share, and the competitive share for each sector.
Also include the years and sources of the data with complete titles and labels. For examples of data table formats, see the shift-share readings.
2b. In a paragraph, discuss the implications of your results.
Note: you may notice that various authors use different terminology for the various components of shift-share analysis. This may lead to some confusion, though the underlying concepts of the basic shift-share model should be the same. (i.e., explaining regional employment changes due to (a) changes in national employment; (b) the region’s relative concentration (industry mix) in slow or fast-growing sectors; and (c) the relative competitiveness of a region’s sector as compared to the sector nationwide.)
Part 3: Comparison of Methods and Outcomes
In a paragraph, compare the results from Parts 1 and 2. How might you use the combination of location quotients and shift-share analysis data to understand the local/regional dynamics of the selected sectors?
WHAT TO TURN IN on Feb 22:
• a paper copy of your assignment (one per group)
-- bring to class.
• upload an electronic version of your assignment to your ctools drop box (one per group) before the start of class-- pdf format preferred.
File name (important): please use the following file name: g[group number]assign2.pdf (e.g., g4assign2.pdf). (Note: I will post your group number here.)
February 22: SHORT PRESENTATIONS
The written assignments are due Feb 22. Please also be ready to briefly discuss (ca. 5 minutes plus 3 minutes Q&A = 8 minutes/group) your two cases on this day, including:
- 2-3 main conclusions about your two cities/regions based on your results.
- comments / advice about methodology (e.g., finding data, calculations, working with Excel, dealing with suppressed data, picking the appropriate NAICS level of detail, etc.)
- a quick answer (in a few sentences) for Part 3 (above).
- note: you do not need to create a powerpoint presentation. We can simply project parts of your pdf file (e.g., your data tables) in class.
- ORDER of presentations: we will go in reverse order, starting with group 11 and ending with group 1.
Here are some general comments on Assignment 2.
As you may discover, there are shortcomings with some of the local & metro economic data. e.g.,
1. for confidentiality reasons, specific numbers are not released for sectors where there are just a few establishments. Instead, a range (e.g., 2500 - 4999 or a letter corresponding to a range) is shown. If you need to calculate a location quotient using a range, you can either (a) calculate the LQ for the range (e.g., for the lower number and for the higher number) and simply report the range OR (b) use the numerical midpoint (in this case, 3750) for your calculation. In both cases, simply note your use of range data and how you calculated the LQ.
more on disclosure:
Data can be withheld for disclosure reasons if the geographic and/or sector-scale is so fine-grained that the data would reveal information about a specific firm or small set of firms (e.g., if a city had only one steel mill, that data about that NAICS code would be withheld).
(a) go to a higher geographic scale: e.g., move from city (place) to county or metropolitan area (or even state)
(b) go to a higher industrial classification scale: e.g., use 2-digit industry codes if the 3-digit leads to withheld data. (Yes, 2 digit level can seem a bit coarse for shift-share analysis, but you do the best you can!)
(c) use a combination of (a) and (b).
(d) calculate using the range [this option is fine for the assignment]
(Yes, this can be frustrating, but you learn to deal with it. If you really need employment data on a specific sector in a specific place, you might use other data sources. see, for example, the discussion in the Virginia Carlson reading. But that is not necessary for this assignment. Simply do the best with the data available.)
2. Certain kinds of employment (e.g., government employment) are excluded from County Business Patterns, etc. This is frustrating. Again, note this shortcoming of the data in your analysis and briefly discuss the ramifications of the missing data.
The County Business Patterns (CBP) covers all NAICS industries except crop and animal production; rail transportation; National Postal Service; pension, health, welfare, and vacation funds; trusts, estates, and agency accounts; private households; and public administration. The CBP also excludes most government employees.
7. CBP excludes government employment and payroll. Where can I get this information?
Employment and payroll for state and local governments are collected in the Federal, State, & Local Governments section of the Census Bureau. There are no counts of government establishments, since governments data are reported by jurisdiction, not by establishment or physical location.
there is a separate count of government employees: link
I wrote the US Census with a question on this topic, and here is their brief answer:
County Business Patterns does provide a total line that includes all business sectors, even those, like mining, not shown for your metropolitan area. (for metro areas, see http://censtats.census.gov/cgi-bin/msanaic/msasect.pl?Sic=&MSA=41860)
On the other hand, you should recognize that some employment, most particularly government employment, is not included in either County Business Patterns or the Economic Census.
The Local Employment Dynamics program at http://lehd.did.census.gov/led/datatools/qwiapp.html provides a somewhat broader measure of employment, although even that does not include federal employees.
Overall advice: do the best with the data available. Keep your work simple, transparent, logical. Clearly label and title your tables. List all sources. (No need for charts in this assignment; several tables will do.) Be concise. No need in this assignment to go into great detail (you can do that for your final project: Assignment 5).
Note on Location Quotients:
I am having problems with determining the number of locally serving jobs. For a few of my sectors, this number is actually greater than the number of jobs available at the city level in the sector. This also leaves me with a negative number for the export based jobs. It might just be because of the data I have to use, but could you please let me know what I should do about this or if there is some mistake I have made?
if your LQ<1, then you assume that the local sector is not large enough to produce enough to meet local demand. They have to IMPORT some to meet local demand. (you don't have to worry about calculating how much outside employment is needed for these imports.)
if a town's LQ for bicycles is 3.0 and there are 300 local employees in the sector, then you would assume that the first 100 employees are producing for local demand. The output from the additional 200 employees is creating a surplus of bikes (i.e., more bikes that local demand can handle) -- and this surplus is being exported.
So: 100 locally-serving jobs + 200 export jobs = 300 total jobs in the bicycle sector
if that same town's LQ for motorcycles is 0.5 and there are 25 local employees in that sector, then you would assume that ALL of the 25 jobs are producing for local demand. Since demand for motorcycles locally is greater than what the local employment is producing, you assume that the town has to import motorcycles to meet local demand (note my simple assumptions -- since you assume homogenous consumption patterns between the locality and the nation),
So: 25 locally serving jobs + 0 export jobs = 25 total jobs in the motorcycle sector
[in this case, there are employees OUTSIDE the town producing motorcycles that are then imported into the town -- you would assume an additional 25 employees elsewhere, but you don't have to calculate that.]
You can see here that the SIMPLE ASSUMPTION (not always true, but a good starting point) is:
1. if LQ<1, all jobs are locally-serving (and the town needs to import to meet the demand that exceeds local production)
2. if LQ=1, then there is a balance between local supply and demand in a sector (no imports, no exports). all jobs are locally-serving.
3. if LQ>1, then all those jobs up to a level of LQ=1 serve local demand, and the surplus produces for the export market. (so you have both locally-serving and export jobs).
So: if you are getting a negative number for export jobs, that means the town is importing products in that sector: i.e., the number of jobs held by outsiders that make the products imported into your town. (You can't have a negative number of jobs.)
As you calculate your location quotients, remember that an export-based strategy can emphasize not just goods, but also services. Read this interesting op-ed piece in the NY Times:
(and click on the nice graphic)
IMPORTANT FINAL WORD: Use complete and correct citations (really small footnotes or references fine -- or perhaps use footnotes on one page and have a separate "sources" page). Refer to all sources used (including data, maps, images, tables, graphs, course readings and materials found on the Internet). Please familiarize yourself with standard practice of academic integrity in coursework. --> See this link for complete information.