Location, Location, Location: The Role of Location-Specific Human - - PowerPoint PPT Presentation
Location, Location, Location: The Role of Location-Specific Human - - PowerPoint PPT Presentation
Location, Location, Location: The Role of Location-Specific Human Capital in Rural Firm Entry and Survival Georgeanne Artz Iowa State University June, 2015 (The 18th Uddevalla Symposium, Snderborg, Denmark Where is Iowa? A research agenda
Where is Iowa?
A research agenda on rural entrepreneurship
- Who chooses it and why?
– Is it taste (necessity) or productive?
- How can we facilitate it in a way that makes
sense?
Why should we care about firm location choices?
Local governments spend $80 billion per year on business incentives
http://www.nytimes.com/interactive/2012/12/01/us/government-incentives.html?_r=0#home
State and Federal Sources: $170 billion/year
High rates of business exit in first few years
Entrepreneurs consider “exit” at time of entry
Iowa North Carolina Exit within…. urban rural urban rural 2 years 35% 30% 37% 34% 6 years 61% 55% 65% 61%
Focus of this work
- Role of idiosyncratic location-specific factors
and complementarity between entrepreneurs skill and the location in choosing a site
- Role of density in new firm entry & survival
– The salvage value of the firm will be greater in denser markets
Idiosyncratic factors?
- Location choices of firms affected by market
factors, agglomeration measures, input prices
(Arauzo-Carod, Liviano-Solis & Manjon-Antolin, 2010)
- Amenities
- Other stuff – idiosyncratic factors
Idiosyncratic factors?
- Entrepreneurs tend to start their business where
they live (or have lived)
– 48% of firms begin in the home or garage (Shane, 2008) – Entrepreneurs more likely to be local than workers (Michelacci and Silva, 2007) – 72% of Portuguese entrepreneurs start firms in home region (Figueiredo, Guimaraes and Woodward (2002) – ½ of ISU rural entrepreneurs grew up in a rural place
Idiosyncratic factors?
- Location-specific factors that affect the
location choice of firms
– Social capital / ties
- Firms created by local firms are bigger, more capital
intensive and better funding (Michelacci and Silva, 2007)
– Industry or amenity specific knowledge
- Lowers cost of entry, cost of production, increases
- pportunity recognition, investment in location
relevant skills
– Taste / preference for location
Research Questions
- How important are (unobserved) idiosyncratic location-
specific factors and entrepreneur-location match in location choices of new firms?
– Are they more important in rural than urban markets?
- Are these idiosyncratic factors productive (or taste)?
- How much of the idiosyncratic component is tied to
the location (and therefore transferrable to another entrepreneur) and how much is tied to the founding entrepreneur (and therefore lost at exit)?
Data
- National Establishment Time Series (NETS)
Dataset: Universe of all new firm start-ups with a DUNS number in Iowa and North Carolina from 1992-2010
– Exclude agriculture, mining , government and nonprofits
Where is Iowa? Where is North Carolina?
Proportion of Rural vs Urban Entrants
0% 20% 40% 60% 80% 100%
Iowa North Carolina Rural Urban
283,721 entrants 889,533 entrants
Model of Firm Location Choice
- Firm i in industry k chooses location j to
maximize the present value of the venture at time t
– Present value of stream of profits conditional on success that depend on
- Location – specific attributes
- Industry – specific attributes in location j
- Prices (wages, rents)
- Idiosyncratic factors (error components)
Model of Firm Location Choice
- The unobserved component (error term) will
include location-specific factors and the entrepreneur-location match
– But anything that doesn’t vary across locations will fall out of the model
- Controls for entrepreneur’s ability for example
Model of Firm Location Choice
- At time of entry, the entrepreneur also has to
consider the present value of the firm if it succeeds, but also the value if it transfers to a successor (or salvage value of assets if it fails)
– In denser markets, more likely to find a potential successor, assets more general – More likely to capture the value of the firm even if transfer to successor, sell assets in rural markets
Iowa North Carolina
Urban Rural Urban Rural
Clusterjkt: Location quotient
0.77 0.57 0.98 0.97
Upstreamjkt: Relative number of upstream suppliers
in the county compared to the state average 0.63 0.11 0.32 0.06
Downstreamjkt: Relative number of downstream
customers in the county compared to the state average 0.71 0.10 0.48 0.08
College%jt: Percentage of the county population age
25 and over with at least some college education 0.50 0.40 0.49 0.38
Concentrationjt: Sum of squared employment shares
across all sectors in the county 0.18 0.21 0.17 0.19
Incomejt : Median household income (in thousands)
24.13 21.56 23.04 19.34
Populationjt (in thousands)
90.43 14.61 139.40 32.20
Variables used to predict firm location choice
Research Question #1
- How important are (unobserved) idiosyncratic
location-specific factors and entrepreneur- location match in location choices of new firms?
– Are they more important in rural than urban markets?
Match component is more important for location choice in rural counties
Research Question #2
- Are these idiosyncratic factors productive (or
taste)?
– Empirical strategy – embed the idiosyncratic match components in a firm survival analysis
- If they matter for survival productive
Firm Attributes for Survival Analysis
Iowa North Carolina
Firm Attributes: Urban Rural Urban Rural Small Entrant: Proportion of all entering firms ≤ 5 employees. 0.852 0.891 0.914 0.927 (0.355) (0.311) (0.281) (0.286) Medium Entrant: Proportion of all entering firms > 5 but ≤ 50 employees. 0.132 0.101 0.077 0.055 (0.338) (0.301) (0.267) (0.249) Large Entrant: Proportion of all entering firms >50 employees. 0.016 0.008 0.009 0.007 (0.126) (0.089) (0.094) (0.082) Branch : Establishment is part of a multi- establishment firm 0.099 0.073 0.075 0.064 (0.299) (0.260) (0.263) (0.245) Number of Firms 170,548 113,173 724,912 164,621
Survival Function Parameters
Dependent variable: Iowa North Carolina log(survival time) (A) (A) Match:
0.968*** 0.437*** (0.048) (0.028)
Medium Entrant
0.237*** 0.366*** (0.012) (0.006)
Large Entrant
0.096*** 0.341*** (0.035) (0.017)
Branch
0.338*** 0.215*** (0.011) (0.006)
Log likelihood
- 233290.59
- 700912.46
𝛾 𝜒 > 0 the match is productive, not taste
0,05 0,1 0,15 0,2 0,25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Density Predicted Hazard of Firm Exit, by Urban and Rural Market: Iowa
Firm Age Urban Rural
Research Question 3
- Q. How much of the idiosyncratic component is tied to
the location (and therefore transferrable to another entrepreneur) and how much is tied to the founding entrepreneur (and therefore lost at exit)?
– Empirical strategy – analysis of variance of the idiosyncratic component across counties – “Between” component captures systematic variation that can be priced at time of sale/salvage – “Within” captures the match-specific capital between the location and the entrepreneur that is lost at time of transfer
Analysis of Variance
Iowa North Carolina Variance due to Urban Rural Urban Rural Between: 0.90 0.04 0.93 0.10 Within: 0.10 0.96 0.07 0.90 Total 1.0 1.0 1.0 1.0
Nearly all the variation in urban markets is due to location-specific factors In rural markets, nearly all the variation is due to the entrepreneur-location match
Summary
- Location-specific agglomeration factors matter for
firm entry
– But, the decision is driven more by idiosyncratic factors than observed market measures
- These idiosyncratic factors are productive
– Firms with larger match at start-up survive longer
- In rural markets the match is largely tied to the
entrepreneur while in urban markets it is location- specific
– The implication is that firms in rural markets will face lower resale value than urban firms with similar profits
So what?
- Asset fixity/spatial fixity lowers the salvage value
- f rural firms
– No transition – Rural entrepreneurs hold on longer because of fixity
- Maybe need more attention to exit (transfer &