Fourth Summer School in Trade, Industrialisation, and Development 2005 Gargnano, Italy Trade, Innovation, and Technology Di usion: Implications for Developing Countries Lecture 1: Basic Measures of Innovation, Di usion, and Trade September 2005
How should we proceed? A desire is that theory and measurement progress in concert Two failures: 1. Keynesian macroeconomic forecasting models: data without theory. 2. Factor Endowments: Elegant theory that sits awkwardly with the facts
More successful: Growth economics.
I. The International Trade of Countries: The Gravity Equation X ni X n Y i ni X ni = X ny i ni Notation purchases by n from i total purchases by n total production by i distance measure constant (unit dependent)
Evidence 1. Total Imports and Market Size: 1970-1972 vs. 1995-1997
1000 100 imports, I_n ($ billions) 10 1.1 GER BEL NET CAN ITA UNK FRA JAP SWE NOR DEN AUT SOU SPA AUL MEX BRA SINIRN VEN NIALG ISR INDARG INO KOR GREFIN HUN CZE MAY IRE POR POL THA PHI NZE CHN TUR SAU LIY CHI PAN MOR EGY COL KUW PAKPER LEB JAM IRQ SUD GHA COTKEN MOZ ECU CAM TUN COS ELS GUA TRI CYPSRI DOM PAP SYR BOL SEN MAD JOR HONIC ICE BAN MAT ETH URU BAB FIJ UGA GABSIEBEN MAW MAS SOMTOG PAR NEP GUN USA.01 ZIM.1 1 10 100 1000 10000 total absorption, X_n ($ billions) Figure 1: Imports and Market Size, 1970-72
1000 USA imports, I_n ($ billions) 100 10 1.1 GER FRA UNK NET CANITACHN KOR MAY THA TAI SPA SWE AUT SWI AUL BRA DEN IRE NOR ISR SAUTURINO PHI POR POL GREFIN SOUARGIND HUNCHI PAN NZE EGY VEN COL KUW ALG TUN MOR PAK ROM IRN PER OMA DOM NIA CYPAR LEB ECU COS URU ELS JOR GHA JAM COT ANG MAS ZIM GUA BAN SYR MATSRI BUL BAR HON ICE TRI BOL KEN PAP YEM GAB ALB NIC CAB CAM SEN BEN BAB SUD MOZ TOGMAD MAU MAI FIJ ZAM NEP UGA MAW LAO GUN AFG NIG BEZ DJI SEY MON BUK COM SIECEN CHA RWA BUR JAP BHU.01.1 1 10 100 1000 10000 total absorption, X_n ($ billions) Figure 2: Imports and Market Size, 1995-97
2. Bilateral Exports and Production
stat stat2 1 market share of i in country n, X_ni/X_n.1.01.001.0001.00001 1.0e-06 1.0e-07 1.0e-08.1 1 10 100 1000 gross production of exporter, Y_i ($ billions) Figure 3: Bilateral Exports and Production, 1995-97
3. Bilateral Trade and Distance:1970-1972 vs. 1995-1997 X ni X in X nn X ii vs. distance
1 bilateral trade index, sqrt((xni*xin)/(xii*xnn)).1.01.001.0001.00001 1.0e-06 1.0e-07 100 1000 10000 distance between countries (kilometers) Figure 4: Bilateral Trade and Distance, 1970-72
1 bilateral trade index, sqrt((xni*xin)/(xii*xnn)).1.01.001.0001.00001 1.0e-06 1.0e-07 100 1000 10000 distance between countries (kilometers) Figure 5: Bilateral Trade and Distance, 1995-97
II. The International Trade of Firms 1. Markets and French Firm Penetration
estimate of firms entering market (thousands) 10000 1000 100 10 1 SIE GER AUL UNK SWI CHN BRA AUT TAI GEE ITA FRA DENBEL NET SPA NZE BUL NOR CZE FIN ROM SWE YUG ISR SOU ARG MEXKOR IND MAY IRE GRE VIE HOK HUN VEN CHIPOR SINSAU CUB TUR IRN COL ALG EGY ZIM SUD ECU CAM COT PER INO ALB MOR TAN COS PAN SYRPHI TRI URU PAK TUN BUK SEN GUA ETH ELS DOM JORKUW THA SRI BOLPAR HON JAM ZAI IRQ PAP OMA BAN NIA TOG ANG SOM MAS LIY CHA MAL KEN RWA BEN MAD NEP UGA NIC BUR NIG MOZ CEN ZAMGHA LIB MAU AFG MAW.01.1 1 10 100 1000 10000 market size ($ billions) Figure 7: Entry and Market Size CAN JAP USR USA
2. French Firms and Market Penetration
firms penetrating at least that many markets 100000 10000 1000 100 10 1 2 4 8 16 32 64 113 number of markets penetrated Figure 6: Frequency of Selling in Multiple Markets
3. Market Penetration and Size in France
average sales in French market ($ millions) 1000 100 10 1 10 100 1000 10000 100000 firms selling to k or more markets Figure 8: Firm Size and Frequency of Multiple Markets
1. Innovation Measures III. Innovation Around the World
Greece New Zealand Iceland Mexico Slovak Republic Hungary Portugal 50 45 40 35 30 25 20 15 10 5 0 Figure 8: Industry Financed Business Enterprise R&D Canada Italy Sweden Netherlands Belgium Australia Spain Finland Denmark Austria Norway Czech Republic Ireland Turkey Poland France Korea United Kingdom United States Japan Germany percent of OECD total
TABLE 2 Business Sector Research Scientists (per 1000 Industrial Workers) COUNTRY Scientists Income Population Finland 12.2 69 5176 United States 10.2 100 275423 Japan 9.8 73 126919 Sweden 7.7 69 8871 Luxembourg 6.8 138 441 Russia 6.6 28 145555 Belgium 6.2 70 10254 Norway 6.0 90 4491 Canada 5.9 81 30750 Germany 5.5 67 82168 Singapore 5.3 80 4018 France 5.1 66 60431 Denmark 4.5 80 5338 Ireland 4.4 76 3787 Korea 4.2 42 47275 United Kingdom 4.2 68 59756 Taiwan 4.2 55 21777 Austria 3.9 70 8110 Netherlands 3.6 72 15920 Australia 2.4 76 19157 Slovenia 2.0 48 1988 Spain 1.8 53 39927 New Zealand 1.7 56 3831 Italy 1.6 64 57728 Slovak Republic 1.6 35 5401 Czech Republic 1.4 42 10272 Hungary 1.4 31 10024 Romania 1.4 14 22435 Poland 0.8 27 38646 Portugal 0.7 48 10005 China 0.7 11 1258821 Greece 0.5 44 10558 Turkey 0.2 21 66835 Mexico 0.1 27 97221 Data are for 2000 or the previous available year Income is relative to the United States (100) Population is in 1000's Sources: OECD (2004) and Heston, Summers, and Aten (2002).
2. Patenting Measures
100000 10000 1000 100 10 1 Figure 9: R&D and Patents 100 1000 10000 100000 R&D expenditure patents granted by the United States Patent Office
Spain Norway Singapore Denmark China, Hong Kong Austria 50 45 40 35 30 25 20 15 10 5 0 Figure 10: Foreign Patenting in the United States Korea Italy Sweden Rest of World Switzerland Netherlands Israel Belgium-Lux. Australia Finland Canada United Kingdom France Taiwan Germany Japan US patents granted to residents in 2000 (% of total foreign)
DE 10000 JP US patenting in Germany (DE) 1000 100 10 BG TR SK LI PT PL CZ GR LU HU ES NO IE NZ ZA RU BR CN IN AT FI BE DK AU IL CH NL IT SE FR GB KR CA 1 RO ARMX 1 10 100 1000 10000 100000 patenting in the United States (US) Figure 11: Sources of Patents
DE patenting by Germany (DE) 10000 1000 100 10 MD BY AL EE UZ MK FR IT GB AT CHES NL BE SE JP DK PT GR IE LUCN AU FI KR CA CZ RU MX PL HU MC NO NZ SITR SKUA IL LT AR BG LV RO KZ VN GE US AM LSZ SDMW UG KE 1 SG BA IS TJKG MN 1 10 100 1000 10000 100000 patenting by the United States (US) Figure 12: Markets for Patents
3. Bilateral Patenting and Distance: P ni P in P nn P ii vs. distance
bilateral patenting index sqrt((gni*gin)/(gii*gnn)) 1.1.01.001.0001 100 1000 10000 distance between countries (km) Figure 13: Bilateral Patenting and Distance
IV. The Evolution of Productivity: Growth as a Global Phenomenon
Figure 14: Evolution of Productivity in High Productivity Countries 100,000 10,000 GDP per capita 1,000 100 1870 1890 1910 1930 1950 1970 1990 year United States Norway Ireland Denmark Canada Switzerand Australia Netherlands France Belgium Japan Sweden Finland Austria United Kingdom 13 small Italy Germany Taiwan New Zealand Spain South Korea Portugal Greece
Figure 15: Evolution of Productivity in Low Productivity Countries 100,000 10,000 GDP per capita 1,000 100 1870 1890 1910 1930 1950 1970 1990 year United States Chile Czechoslovakia Venezuela Argentina Malaysia Uruguay Hungary Mexico Costa Rica Turkey Brazil Colombia USSR Yugoslavia Peru China Sri Lanka Guatemala Indonesia El Salvador Cuba Philippines Honduras India Nicaragua
Challenges for Theory 1. Accommodating Market Segmentation and Geography 2. Reconciling Aggregate and Producer-Level Data 3. Modeling Specialization in Innovation and Technology Di usion 4. Explaining the Geography of Economic Growth