SUMMARY: Why businesses locate where? ECON 376 Kilkenny

1. Immobile endowments: Comparative Advantage (Ricardo (1800); Heckscher-Ohlin-Samuelson (1930-50))

- an a-spatial explanation

Claim: “A region tends to specialize in and export the goods that employ their relatively abundant factors relatively intensively.”

Empirical application: Measure the endowments (not employment) of the primary factors of production in each region. Compare the ratio of factor endowments within one region to another region’s factor endowment ratio (e.g.: L_Iowa/K_Iowa < L_California/K_California). Identify products that employ those factors intensively (e.g.: L_hogs/K_hogs < L_software/K_software Š California has a comparative advantage in software; Iowa has a comparative advantage in hogs.)

2. Transportation Costs (Varignon, b.1654 ; Weber, 1909; NUG30-group, 2000)

Claim: “All else equal, an establishment’s profit-maximizing location is the one that minimizes transport costs.”

Empirical Application: Measure transport costs per unit distance for inputs and outputs (t_i,t_o). Measure the units or inputs needed per unit output (q_i). Compare t_iq_i to t_o. Choose the i-site if t_iq_i > t_o. to reduce t_iq_id_isQ to 0 (avoid all input transport costs). Otherwise, if t_iq_i < t_o, choose the market site to reduce t_od_smQ to 0 (avoid all output transport costs).

3. Competition (Hotelling, 1929; Christaller, 1933)

Claim: Businesses that buy/sell the same thing disperse spatially to avoid competition; each monopolizes nearby suppliers/customers rather than raise/lower their prices.

Empirical application: Shippers: Measure all costs of production (K, AC), transport costs (t), and the delivered prices of alternative supplies (DP_alt). Find the market area (m) at which your delivered price just equals the alternative supplier’s: m = (DP_alt-AC)/t. Verify: is this market large enough to cover costs and return to lowest acceptable income to the owner? Shoppers: Measure the average consumer’s budget or outlay on the item(s), “E”; transport costs (t), average cost of production (or mill price). Find the extent of the market area (m) beyond which customers will not come to patronize your shop: m = (E-C)/t. Count the number of customers and total quantity demanded within this market area, given data $E/person. Verify: is this market is large enough to cover costs and return to lowest acceptable income to the owner?

4. Agglomeration economies of scale (Smith, 1776; Marshall, 1920; Arrow, 1962; Romer, 1979, Jacobs, 1980…)

Claim: Businesses concentrate spatially because costs are lower, or revenues are higher the more activity there is in the same place.

Empirical application: Measure fixed, variable, and average costs of production (K,vQ,C) in detail: input prices (p_i), input:output ratios (q_i), local wages (W), and labor per unit output (q_L). Measure the level of activity in each location “r” and industry “i” at time “T” by the number of employees (E_irT). Measure concentrations of businesses by location quotients (LQs) and Hirschman-Herfindahl (HHI) indexes. Measure concentrations of population using census data. Statistically analyze cross-section data to identify static economies of scale. Analyze time-series data to identify dynamic economies of scale.