New Evidence on the Host Country Effects of Japanese Multinationals (Tables)

Table A-1 Average Annual Compound Growth Rates by Industry: Japanese and Domestic Startups, 1978-88 (National Panel Data 1978-88) SIC Industry Domestic # Growth Plants Rate Japanese # Growth Plants Rate Distribution by Industry Domestic Japanese 20 Food Products 22 Textiles 23 Apparel 24 Lumber 25 Furniture 26 Paper 27 Printing 28 Chemicals 30 Rubber and Plastic 32 Stone, Clay, Glass 33 Primary Metals 34 Fabricated Metal 35 Non-Electrical equip. 36 Electrical equip. 37 Transportation 38 Instruments 39 Miscellaneous 3551 824 1537 1340 830 1202 2989 2843 1729 1761 1171 2946 4321 2911 1326 1406 854 4% 7% 6% 8% 7% 5% 7% 4% 10% 5% 6% 7% 6% 8% 9% 7% 5% 11 1 1 1 3 1 0 5 8 2 6 5 11 26 16 6 3 11% 60% 0% 6% 35% 0% 19% 54% 18% 20% 50% 53% 28% 28% 13% 23% 10.59% 2.46% 4.58% 4.00% 2.47% 3.58% 8.91% 8.48% 5.15% 5.25% 3.49% 8.78% 12.88% 8.68% 3.95% 4.19% 2.55% 10.38% 0.94% 0.94% 0.94% 2.83% 0.94% 0.00% 4.72% 7.55% 1.89% 5.66% 4.72% 10.38% 24.53% 15.09% 5.66% 2.83% Overall 2-Digit Average 33541 6% 106 29% 100.00% 100.00% 209 Canned Fruit/Veg 380 7% 307 Plastic Products n.e.c. 1384 10% 331 Steel 371 7% 357 Computers, Office Equip 941 8% 354 Machine Tools 636 5% 365 Radio/TV Receivers 109 7% 367 Electrical Components 945 11% 371 Auto Assembly/Parts 713 8% 6 5 5 3 3 7 12 13 17% 22% 13% 66% 25% 37% 19% 19% Sources: USEEM database; authors' survey Table 1 Description of Startup Plants in the Sample SIC State 30* 30* 30* 30* 30* 30* 30* 30* 30* 30 30 30 30 30 30 30 35* 35* 35* 35* 35* 35* 35* 35* 35* 35 35 35 35 35 35 GA GA GA KY KY KY NJ NJ NY GA GA GA KY KY MA NJ GA GA GA KY KY KY MA NJ NJ GA GA KY KY MA MA Product misc plastic products tv frames gaskets industrial belts rubber components auto dashboards plastic labels plastic labels/seal auto parts plastic parts auto molding plastic food packaging plastic parts plastic parts misc. plastic products plastic bottles PC monitors construction equip. construction equip. machine tools cutting tools metal prod centers personal computers textile machinery computer peripherals industrial saws motors motor brushes precision mach. parts industrial equip. industrial equip. Number of Workers Annual Avg. Start Employment Year Growth (%) 129 300 375 120 100 340 110 39 600 105 175 216 77 131 50 200 411 52 15 275 21 25 788 27 60 250 350 64 70 65 160 1980 1989 1980 1988 1988 1988 1985 1986 1979 1983 1985 1978 1987 1978 1978 1978 1985 1988 1987 1982 1989 1987 1983 1988 1990 1978 1983 1984 1984 1990 1978 9.6 26.0 11.6 18.9 42.5 8.0 27.6 14.7 31.4 9.1 -1.9 8.2 17.8 4.6 1.5 -5.0 22.4 24.0 -27.9 23.1 51.4 3.8 16.5 22.0 7.1 2.6 12.6 4.2 24.1 -16.2 9.7 Table 1 (cont.) SIC State 36* 36* 36* 36* 36* 36* 36* 36* 36* 36* 36 36 36 36 36 36 36 GA GA GA GA KY KY KY MA MA NJ GA GA GA KY KY MA MA Product Number of Production Workers auto & consumer elec. 931 videocassettes 300 compact disks 300 tvs, phones 676 electronic auto parts 534 wire assembly 1800 electronic auto parts 125 micro diskettes 215 micro diskettes 350 elect. optical instru. 44 circuit boards 350 elect. auto part 440 wire harnesses 350 truck wire harnesses 134 air condition 300 circuit boards 104 computer equipment 54 __________________ * indicates Japanese ownership Start Year Annual Avg. Employment Growth (%) 1987 1980 1987 1984 1985 1988 1989 1989 1989 1982 1989 1978 1978 1985 1978 1987 1982 20.3 8.8 37.7 17.6 19.9 34.5 46.2 6.1 36.8 7.2 -11.2 9.7 7.6 4.3 11.0 26.8 0.0 Table 2 Adoption Rates of Compensation Practices: Japanese-owned and Domestic Startups Compensation Practice All Plants U.S. Plants Japanese Plants 100.0% 45.0 25.0 15.0 100.0% 46.4 17.9 28.6 Earnings Above Average Wages Wages in Top Quintile Profit Sharing Group Bonuses 100.0% 45.8 20.8 22.9 Non-Pecuniary Job Benefits Employee Voice Job Security 75.0 27.1 40.0 5.0 100.0** 42.9** Source: Authors’ Establishment Survey Significance Test: Significance of Difference Test using Pearson Chi Square with Yates continuity correction. When expected frequency is too small, Fisher's Exact Test is used. Significance of differences is measured between U.S.-owned and Japanese-owned plants. * = 0.05; ** = 0.01 Table 3 Adoption Rates of Selected Management Practices: Japanese-owned and Domestic Manufacturing Plants Practice All Plants U.S. Plants Japanese Plants 79.2% 31.3 60.0& 15.0 92.9%** 42.9* 41.7% 27.1 35.0% 15.0 46.4% 35.7 52.1 35.0 64.3* 58.3 20.0 50.0** 48 20 Training Intensive Entry Training Multi-skill Training High Performance Management Production Teams Job Rotation Problem- solving (Quality Circles) Quality Control by Production Workers N= 28 Source: Authors’ Establishment Survey Significance Test: Significance of Difference Test using Pearson Chi Square with Yates continuity correction. When expected frequency is too small, Fisher's Exact Test is used. Significance of differences is measured between U.S.-owned and Japanese-owned plants. * = 0.05; ** = 0.01 Figure 1 Illustrative Ramp-up of a High Performance Startup Output, Employment, Output/worker Output Target Output/worker Actual Output/worker Employment First shift Second shift Third shift Plant Expansion Figure 2 Alternative Ramp-up Paths Employment High Performing Plant Plant Expansion Average Performing Plant Low Performing Plant Time Table 4 Definitions of Variables Used in the Ramp-up Models RAMPUPi = Average annual compound growth rate of employment in establishment i from year of startup to final year (1991-92 for case study sample, 1988 for national sample) HIPERFORM = Number of high performance management practices adopted by establishment i [daily and weekly meetings with employees, employee involvement in managerial decisions (quality circles, peer supervision, quality control by production workers), flexible work practices (teams, job rotation) NATJi = Binary variable; 1 if establishment is Japanese-owned. STARTSIZEi = Log of employment in the startup year of establishment i. INDi = Vector of binary variables for the two-digit SIC of establishment i. REGi = Vector of eight binary variables for regional location of establishment i. STARTYEARi = Vector of binary variables for two-year cohort in which establishment i began production [1988-89 is omitted cohort from case study analysis, 1986-87 is omitted cohort from national panel data] . AUTOi = Binary variable for plant i being in a 4-digit SIC industry supply the automotive industry AUTO i * NATJi = Binary variable for Japanese–owned plant i being in a 4-digit SIC supplying the automotive industry NATJ i * INDj = Vector of binary variables for Japanese-owned plant i being in the twodigit SIC industry j. u = Random error term Table 5 Estimated Effects of High Performance Practices on Ramp-up Rates: Case Study Sample (absolute value of t-statistics in parentheses) Dependent Variable: RAMPUP Independent Variable Constant HIPERFORM (1) (2) Without NATJ With NATJ 44.80** 36.88** (3.66) (2.95) 6.07** (2.91) NATJ STARTSIZE PLAS MACH 4.77* (2.25) 10.33* (1.95) -6.92** (3.40) -6.43** (3.25) -4.82 (0.91) -8.94 (1.58) -5.04 (0.98) -8.68 (1.59) -9.84 (1.55) -15.83 (1.71) -11.13 (1.58) -6.22 (0.97) -16.81** (2.53) -3.44 (0.50) -17.17 (1.92) -8.07 (1.16) -3.34 (0.53) -14.81* (2.28) Starting year cohorts 1978 1980 1982 1984 1986 # of plants 48 Adjusted R-square 0.68 0.72 3.70** 3.95** F value ________________________________ * significant at the .05 level ** significant at the .01 level Source: Authors' establishment survey. 48 Table 6 Estimated Effects of Organizational Regimes On Ramp-up Rates: National Panel Data 1978-88 (absolute value of t statistics in parentheses) Dependent Variable: RAMPUP Basic Model (1) Basic model with auto interactions (2) Model with sectoral interactions (3)______________ Variable Constant NATJ STARTSIZE 0.20** (25.32) 0.24** (7.54) -0.04** (38.96) 0.20** (25.35) 0.21** (5.54) -0.04** (38.96) 0.02** (2.66) 0.11 (1.52) -0.01 (0.81) -0.01 (0.80) 0.04** (3.59) 0.04** (3.56) 0.03** (2.94) 0.03** (2.84) 0.03** (2.60) 0.03** (2.61) 0.02 (1.34) 0.02 (1.26) AUTO AUTO*NATJ 0.20** (25.34) -0.04** (38.96) Industries SIC20 NATJ*SIC20 SIC22 NATJ*SIC22 SIC23 NATJ*SIC23 SIC24 NATJ*SIC24 SIC25 NATJ*SIC25 0.00 (0.75) 0.08 (0.85) 0.04** (3.57) 0.52 (1.69) 0.03** (2.97) -0.06 (0.18) 0.03** (2.62) 0.05 (0.17) 0.02 (1.31) 0.35* (1.97) Table 6 (cont) Industries SIC26 Basic Model (1) Basic model with auto interactions (2) 0.02 (1.47) 0.02 (1.48) -0.02* (2.36) -0.02** (2.44) 0.05** (5.37) 0.05** (5.13) -0.01 (1.19) -0.01 (1.34) 0.02* (2.02) 0.02* (1.98) 0.02* (2.42) 0.01 (1.79) 0.00 (0.15) 0.00 (0.14) 0.04 (5.11) 0.04** (4.89) 0.06** (5.54) 0.05** (4.56) NATJ*SIC26 SIC28 NATJ*SIC28 SIC30 NATJ*SIC30 SIC32 NATJ*SIC32 SIC33 NATJ*SIC33 SIC34 NATJ*SIC34 SIC35 NATJ*SIC35 SIC36 NATJ*SIC36 SIC37 NATJ*SIC37 Model with sectoral interactions (3)_____________ 0.02 (1.49) 0.06 (0.20) -0.02* (2.34) 0.16 (1.00) 0.05** (5.26) 0.49** (4.11) -0.01 (1.18) 0.16 (0.74) 0.02* (2.09) 0.11 (0.88) 0.02* (2.39) 0.44** (3.13) 0.00 (0.07) 0.49** (5.00) 0.04** (5.15) 0.19** (2.79) 0.06** (5.52) 0.25** (2.84) Industries SIC38 Table 6 (cont) Basic model with auto Basic Model interactions (1) (2) 0.02 (1.56) 0.02 (1.54) -0.01 (0.87) -0.01 (0.87) 0.01 (0.57) -0.01 (0.80) 0.00 (0.19) 0.00 (0.65) 0.00 (0.24) 0.03** (3.98) 0.02** (3.83) 0.01 (0.56) -0.01 (0.81) 0.00 (0.34) 0.00 (0.68) 0.00 (0.25) 0.03** (3.89) 0.02** (3.80) 0.00 (0.54) -0.01 (0.84) 0.00 (0.25) 0.00 (0.68) 0.00 (0.28) 0.03** (3.88) 0.02** (3.76) -0.04** (6.08) -0.02** (3.91) -0.03** (5.16) -0.02** (4.62) -0.03** (6.07) -0.02** (3.88) -0.03** (5.14) -0.02** (4.60) -0.03** (6.06) -0.02* (3.84) -0.03** (5.13) -0.02** (4.63) 33647 0.05 61.79** 33647 0.05 8.15** 33647 0.05 41.23** NATJ*SIC38 SIC39 Model with sectoral interactions (3)_______________ NATJ*SIC39 0.02 (1.62) 0.07 (0.52) -0.01 (0.86) 0.18 (1.00) Regions MT WNC ENC NNE WSC ESC SAT Starting year cohorts 1978 1980 1982 1984 # plants Adjusted R-square F value * significant at the .05 level ** significant at the .01 level