Lecture1_2019-3 - Lecture 1 Agglomeration Review of Empirical Evidence Econ 280D Cecile Gaubert U.C Berkeley Thanks to Gilles Duranton David Nagy

Lecture1_2019-3 - Lecture 1 Agglomeration Review of...

This preview shows page 1 - 9 out of 61 pages.

Lecture 1 Agglomeration: Review of Empirical Evidence. Econ 280D Cecile Gaubert U.C. Berkeley Thanks to Gilles Duranton, David Nagy and Esteban Rossi-Hansberg for sharing their slides. Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 1 / 59
The US at night Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 2 / 59
Why is economic activity so agglomerated? I Four broad categories of explanations: 1. Follows production input ( exogenous source of concentration ) I Natural resources (eg. mine, climate) 2. Production agglomerates endogenously. I Some spillovers depend on proximity I Can explain Silicon Valley, Detroit, Boston biotech I What is called ”agglomeration externality” 3. Some consumption amenity is concentrated (exogenously) I Climate, scenery, nice places to live 4. Some consumption amenity is endogenously agglomerated. I People like to live near each other I Higher density allows for more variety in consumption (more restaurants, opera) Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 3 / 59
I Another question : why isn’t economic activity more agglomerated? I Congestion forces / dispersion forces can be market forces or external I Cost of dense housing (inelastic housing supply, skyscrapers) I Traffic, Pollution ( air, noise, visual) I Access to agricultural hinterland I Heterogenous preference for different places I The literature: I Quantify ”agglomeration externalities” on the production side (extensive lit.) I Identify the channel(s) through which they percolate (limited) I Some about benefits on the consumption side (more to be done) I Little systematic research on congestion Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 4 / 59
Today, overview of evidence on 1. Magnitude of agglomeration economies 2. Single vs multiple spatial equilibria 3. Channels of agglomeration economies 4. Consumption side Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 5 / 59
Descriptive evidence on city-size premia Wage and tfp data for French employment areas, Combes et al (2010) 0.2 0.1 0 0.1 0.2 0.3 3 2 1 0 1 2 3 4 5 6 7 ln wages ln density 0.2 0.1 0 0.1 0.2 0.3 3 2 1 0 1 2 3 4 5 6 7 ln TFP ln density (a) Wages and employment density (b) tfp (Olley-Pakes) and employment density ( 306 employment areas, 1976 - 1996 average) ( 306 employment areas, 1994 - 2002 average) urce : dads , brn , rsi , siren and authors’ calculations. All variables are centred around their mean. The I Not causal I Slope about 0.05 Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 6 / 59
I Similar evidence from many other countries I Similar evidence for many other proxies for productivity (eg output per worker, etc) I A lot of evidence for many other (indirect) proxies for productivity such as firm creation, rents, etc Econ 280D. Spring 2019. C. Gaubert Lecture 1 Agglomeration: Evidence 7 / 59
Estimating agglomeration externalities With agglomeration economies on the production side, models imply: log w ic = α log dens c + η c + u i + ε ic I Wage as dependent variable but can use tfp , or indirect measures of agglomeration benefits: firm creation, etc I

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture