Diego Restuccia. University of Toronto and NBER

Transcription

1 Discussion of Are There Too Many Farms in the World? Labor-Market Transaction Costs, Machine Capacities and Optimal Farm Size, by Foster and Rosenzweig Diego Restuccia University of Toronto and NBER NBER Summer Institute 2018 Productivity, Development, and Entrepreneurship Workshop Cambridge July 18, 2018 Restuccia Too Many Farms? NBER SI / 11

2 Average Farm Size across Countries Corr = 0.61 AUS Log of Average Farm Size PRY BRA GER DNK FRA LUX 3 PER COL IRL AUT PRI ESP PAN ISR BEL NLD HND BHS CHE PRT NOR 2 IRN ITA WSM TUR FJI SVN BFA PAK GRC THA CYP 1 NAM UGA PHL DMA KNA LCA INDGIN LSO VCT BRB 0 GNB KOR JPN ETH NPL IDNEGY MWI GRD ZAR VNM Log of 1990 Real GDP per Capita ARG CAN USA GBR FIN Source: Adamopoulos and Restuccia (AER 2014) Restuccia Too Many Farms? NBER SI / 11

3 Size Distribution of Farms across Countries Small (< 5 Ha) Large (> 20 Ha) Source: Adamopoulos and Restuccia (AER 2014) Restuccia Too Many Farms? NBER SI / 11

4 Land Productivity in Agriculture Actual Yield (log, Int.$ per Hectare) ZAR AFG BELARE KWT EGY GRC IRL LUX SLB UZB MYS JAM KOR GBR JPN GER NZL CHE FJI TKM CHN COL CRI MLT FRA ISR DNK AUT GUY IDN LBN TWN ITA PNG CHL PRI PER SUR SVN CZE OMN SWESGP TJK VNM USA SWZ BGD COG LKA MNE DOM BLZ BRA CYPESP NOR GHA CIV CUBHRV LAO PAK PRY KGZJOR THA VEN POL HUN GTM VUT URY SVK ARG SAU FIN QAT BEN SYR CMR PHL ALB AZE ECUTUR PANGAB LTU BHS BRN MDG BIH LVA BLR IND MKD PRT AUS CAN UGA KHM BOL SLV UKR ROM ZAF BGR MEX LBR HND ARM IRN EST MWI NPL BDI GMB NICMDA CAF TGO ZMBNGA AGO GIN GNQ LBY TZA GEO ZWE RUS GNB RWA MLI SLE MOZ SEN BTN DZA HTI KEN ETH KAZ BFA YEMTCD TUN LSO MAR ATG SDNMNG IRQ NAM SOM ERI MRT NER BWA NLD ρ = , N = Real GDP per Capita (log, 2000) Source: Adamopoulos and Restuccia (NBER WP 2018) Restuccia Too Many Farms? NBER SI / 11

5 This Paper Focuses on documenting U-shape relationship between farm size and productivity using micro data from India Transaction costs in the labor market and machine capacity are key channels There is little to disagree with their analysis and conclusions: the world is littered with policies attempting to exploit the inverse size-productivity relationship with dismal results My discussion: provide a macro perspective into the main factors driving small farm sizes in poor countries Restuccia Too Many Farms? NBER SI / 11

6 (1) Productivity and Farm Size Small farm size is a symptom of low productivity in poor countries Simple framework: Standard two-sector model of agriculture and non-agriculture Non-homothetic preferences: minimum consumption requirement of agricultural goods ā, and only care to consume ā, hence, c a = ā Normalize population and productive time to 1 L units of total land Aggregate production function: Y a = AκL γ N 1 γ a Restuccia Too Many Farms? NBER SI / 11

7 (1) Productivity and Farm Size Implications: c a = Y a = ā Solve for labor allocation in agriculture N a = ( ā ) 1 1 γ AκL γ Average farm size is roughly proportional to L N a Low average farm size is a productivity problem in poor countries Quantitatively, differences in aggregate factors (A, L) only account for small portion of farm size differences between rich and poor countries (Adamopoulos and Restuccia, AER 2014) Restuccia Too Many Farms? NBER SI / 11

8 (2) Distortions, Productivity, and Farm Size Farm size and productivity are deeply confounded by distortions Back to simple framework: Incorporate production heterogeneity Build on Lucas-Hopenhayn establishment-size theory Production unit is a farm: requires the labor and managerial ability of farmer and land input y i = As 1 γ i l γ i where l i is operated land and s i idiosyncratic productivity Assume farmer operates facing idiosyncratic output distortions τ i in competitive markets (catch all to rationalize operational scales, convenient for counterfactuals) Restuccia Too Many Farms? NBER SI / 11

9 (2) Distortions, Productivity, and Farm Size Farmer problem From FOC, farm size is max (1 τ i)as 1 γ i l γ i ql i l i 0 l i = [ ] γa(1 τ i )s 1 γ 1 1 γ i q Farms of same size could have very different productivities In this context, farm size is just not useful for categorization Distortions lower productivity in agriculture (misallocation), implying small farm sizes in poor countries Restuccia Too Many Farms? NBER SI / 11

10 (3) Yield and Productivity Yield not necessarily a useful measure of productivity at the farm level, relationship affected by distortions and technology Back to simple framework: From FOC, yield is y i = q 1 l i γ (1 τ i ) With no distortions, yield should be the same across farmers with different productivities s i Yields track productivity well with extreme distortions, e.g. when l i fixed and equal across farmers Restuccia Too Many Farms? NBER SI / 11

11 (3) Yield and Productivity Technology can confound yield and productivity relationship CES technology, capital and land more substitutable than Cobb-Douglas, land augmenting productivity Even without distortions, yield higher in small farms where productivity is low However, this property of technology not key for cross-country farm-size implications (Adamopoulos and Restuccia, AER 2014) Restuccia Too Many Farms? NBER SI / 11