The objective of this assignment is to estimate demand for a proposed new retail store and determine its economic feasibility.  This is a very common problem faced by retail consultants, and by chains interested in expanding their number of outlets.  We have made this problem much simpler than is typically the case in reality, where an investment decision over the economic life of a store would need to be evaluated (e.g., looking at say a 10-year projection of demand/revenues and costs).  Nonetheless, the basic geographic principles underlying the more complex problem are the same as those for this simplified model.

We'll go over the exercise in the Geography Collaboratory, and then you'll form small groups to collaborate in setting up the scenarios and running the exercise.  You'll hand in individual reports, however, based on the work you did in the group and on your on.

Imagine that you are hired as a consultant to do a feasibility study of a particular location in Seattle (46th Street and Fremont) for a client who wants to open a retail outlet (e.g., in the food/beverage sector, a supermarket, natural food store, bakery, ethnic food grocery, etc.;  or a drug store, or a furniture store, etc.).  The client has hired you to study the location and has provided you with a store concept (you will actually need to develop, and briefly explain in your write-up, the store concept yourself).  Based on this "store concept," you will determine the feasibility of the proposed store given your estimates of potential demand.  The store concept will determine the likely size of your market area and the degree of external demand, etc.  For example, for a small, convenience store, your market area will be small; for a supermarket, your market area will be larger; for a specialty foods store, you will draw a lot of consumers from outside of your immediate market area.  Your demand estimates are calculated using a simple spatial interaction model.  The factors influencing the feasibility of your store location, given your store concept, are:

(1) the share of personal income spent on this good/service
(2) the size of the the market area
(3) demand or sales leakages from this market area
(4) the demand for the good or service generated from outside the market area, or external demand
(5) an expected market share
(6) the friction of distance, expressed as an exponent of distance, facing customers who would travel to the store
(7) the size of the store (e.g., square footage), given that the larger the store, more sales (demand) is needed to cover costs

1.  Share of personal income:  What percentage of personal income do people in the market area spend on your product?  The data in Table A provide some of this information for the State of Washington and the City of Seattle.  As indicated in the data, Seattle residents spend on average 7.75% of their personal income on food items.  Base your estimate of food expenditures on these numbers.  Remember that the category is for all of a household's food expenditures; for example, if you are only selling baked goods, you will need to scale down this number (you can go look it up in Government Documents, or make a guess that seems reasonable).
 

 Table A
Composition of Retail Sales in Seattle and Washington State

RETAIL CATEGORY	PERCENT FOR  SEATTLE	PERCENT FOR  WASHINGTON STATE	MINIMUM SALES REQUIRED PER SQ. FT.
Building Materials	1.40 %	2.34 %	$100
General Merchandise	4.03	4.98	$200
Food and Beverages	7.75	7.48	$350
Furniture	2.85	1.95	$200
Drugs and Sundry goods	1.57	1.39	$100

Sources:  U.S. Census of Retail Trade; U.S. Local Area Personal Income

2.  Market area:  How large of a market area will your store or outlet service?  Which census tracts will your area draw customers from?  Select the tracts that seem the most reasonable given your store concept from the attached map (you will enter them as instructed below).

3.  Leakages from market area:  Not all consumers living in the area you have defined will shop there.  Some will shop in other market areas.  For example, those working in downtown will probably buy some foodstuffs from the Pike Place Market.  So, while most demand for groceries is quite localized, it is not totally localized.  Select some percentage which you think is reasonable for your selection of census tracts.  So, the more census tracts you decide to include in your market area, the less will be the likely leakages of spending.

Suggestion:  make this value no larger than 25% if you only pick 3-6 nearby tracts, or as little as 10% if you pick all tracts for which data are provided.

5.  Expected market share:  There may be many stores in the census tracts included in the problem.  Let's assume that there is one store of your type per census tract.  If we just considered the four adjacent to the *'d location on the map, it would be optimistic for a new seller to think that he could get more than, say 25% of the market, if his/her store size and qualities were equivalent to the existing competitors.  As the number of census tracts increases, the number of competitors increases, and so the expected market share would decrease.  You need to consider a market share which is "realistic" given your store concept (e.g., a very specialized store or a steep discounter might have higher-than-average market shares).  So, market share is dependent on i) the number of census tracts you think are in your market area, and ii) your store concept.

6.  Friction of distance:  The friction of distance is expressed by the exponent applied to the distance variable (distance from the center of each census tract to your store location).  It is set at 2.0 for this exercise.  You may reduce it to 1.0 by planning a delivery service to your customers.  However, you must then multiply the minimal sales per square foot (for your store type -- see Table A) by 2.5, to recover the higher costs borne by the store for the delivery service.

7.  Size of store facility and sales productivity:  Ultimately, the viability of the proposed store is a function of its profitability per square foot of retail space.  Let us assume that the threshold values listed in Table A are the break even points.  Let us also assume that the decision maker involved in the location of this store is considering 30,000 square foot facility.  You may change these figures, but you should justify the changes in the write-up; assume a minimum of 5,000 square feet and a maximum of 100,000 square feet. The square footage is the basis for "rental" cost of the property;  it's not used as a measure of "attraction" here as found in other potential demand models.

Click here, 207assign2.xls, to use access the spreadsheet.  You will see a dialog box that gives you the choice of opening the spreadsheet or saving it to a disk.  Choose "open" and click OK.  The Excel program will start and the spreadsheet will appear after a few moments.  From here, you can save the spreadsheet to your disk, and then re-save it as you make changes.

The first thing you will see is Table 1, with the census tract data.  The spreadsheet program is very simple, with the locations where you can put in data highlighted on the screen in yellow.  It is programmed to produce results driven by the data inputs you select (e.g. the entries in the yellow fields).

Census Tracts:  The attached map of north Seattle shows census tracts in the vicinity of the proposed new store, which is identified by a "*" at 46th Street & Fremont Avenue.  Each census tract probably has a store of your type, so there is lots of potential competition.  You need to decide which of these census tracts should be included in your analysis.  Essentially, you will need to decide:

 (i)  your product (food, furniture, etc.) and store concept (specialized? neighborhood?)
 (ii)  which tracts to include in your market area
 (iii)  some parameters governing demand at the hypothetical store site (leakage, market share, etc.).

Table 1 in the spreadsheet shows the number of households, income per household, and an estimate of distance from each census tract to the proposed store site.  These data, combined with the information you provide on market characteristics, will ultimately determine the feasibility of the proposed new store.

You may include all of the census tracts, or just some of them; you choose, based on the concept you've developed for your store.  Once you have decided which tracts you're going to include in your market area, go the the Table1 area in the spreadsheet and simply delete the figures for "number of households" and "income per household" for the tracts you do not want to include in your market area.  All the tract data are already entered -- you narrow down the market area by deleting the figures for "number of households" and "income per household" by highlighting those cells and pressing the "delete" key.  Leave the distances as they are (the distance cells are write-protectected, so you can't change them).  If you make a mistake and delete some you do want, you can copy them from the data listing to the right of the table.

Now you must enter the parameters which influence the demand estimates for your store:

1. Expenditures percentage:  This is the share of personal consumption expenditures spent on your product type;  enter the appropriate number from Table A, above (DO NOT ENTER A % SIGN).

2. Leakages percentage:  this will decrease potential local demand by the % you put in.  The initial value is set at 10 percent. (DO NOT ENTER A % SIGN)

3. External demand percentage:  this augments potential local demand by the % you put in.  The initial value is set at 10 percent. (DO NOT ENTER A % SIGN)

4.  Friction of distance:  this determines the level of demand as distance increases.  Because this is an exponent, the larger the number the more impact distance will have on demand.   It is set at 2.0 for this exercise.  You may reduce it to 1.0 by planning a delivery service to your customers.  However, you must then multiply the minimal sales per square foot (for your store type -- see Table A) by 2.5, to recover the higher costs borne by the store for the delivery service.

5. Market share:  This is your estimate of the market share you think this facility should get, given your estimates of the "position" of this new store in the market.  Guideline:  if you pick only three or four census tracts in which you think there is little competition, then maybe 25% to 33% of the total demand is "optimistic."  But, if you think the market area consists of 10 census tracts, and most of them have a competing store, then only 10% of the demand may be optimistic.  The value you pick for this variable will have a big impact on the "bottom line."  (You can experiment with different values in the yellow-highlighted cells to see the impact of changes in parameters).  The initial value is set at 10 percent. (DO NOT ENTER A % SIGN)

6.  Size of store facility (total lot size):  Stores in the categories being evaluated here range from mega-stores like Larry's in Bellevue or Kirkland with 100,000 square feet plus to more modest stores with ca. 20,000-25,000 square feet.  Earlier it was suggested that a 30,000 square foot store be evaluated, but you could select another size if you want.  A small bakery might be 5,000 square feet (including parking).  This does not measure "attraction" in this model; it's used only for calculating the "bottom line" feasibility.

Printing your results.  Once you have set up your spreadsheet, you can get print a copy.  Use the mouse to go to the File menu, pull the cursor down to Print, let go of the mouse button, and a print command will come up, click OK.  Note:  It can take a few minutes for the files to make their way through the computer system to the laser printer.

Alternative Estimates.  After your first estimate of demand per square foot, change some of the parameters above (with a good story as to why you're making the changes -- new concept?  larger store?  more specialized?  delivery service?) to improve the sales/sq.ft. numbers that result.  To do this, change the input values in the yellow highlighted fields, and print this alternative estimate.  If neither of these estimates reaches the minimum sales/sq.ft. required for your type of store (and for any delivery service option you establish), please explain why you think your store type and concept are infeasible in this location.  Save the spreadsheet to your own diskette, maybe under different names for each alternative estimate, in case you need to use it again.

Once you have developed your spreadsheet(s), please write-up a short interpretive report (ca. 2 double-spaced pages).  Phrase your write-up as if you are a consultant giving a brief report to your client before doing a more extensive analysis (reporting your store concept and scenarios).  Please include results of your spreadsheet run(s).  Please save your word processing file, so that after we read your write-up if it needs to be edited you can do so with minimal effort.  You will need to take a floppy disk with you to save your write-ups in case you need to rewrite the assignment.