Report No. 22797-ME Mexico Technology, Wages and Employment (In Two Volumes) Volume II: Technical Papers December 13, 2001 Poverty Reduction and Economic Management Unit Latin America and the Caribbean Region Document of the World Bank ABBREVIATIONS AND ACRONYMS CNCM Computerized Numeric Controlled Machinery CONACYT National Science and Technology Council EIA Annual Industrial Survey ENESTYC National Survey of Employment, Wages, Technology and Training FDI Foreign Direct Investment GATT General Agreement on Tariffs and Trade GDP Gross Domestic Product INEGI National Institute of Statistics, Geography and Information Technology MNC Multinational Corporation NAFTA North American Free Trade Agreement NCM Numeric Controlled Machinery R&D Research and Development SBTC Skill-biased Technological Change S&T Science and Technology SST Stopler-Samuelson-Type TA Technology Adoption TFP Total Factor Productivity Vice President: David de Ferranti Country Director: Olivier Lafourcade Sector Director: Ernesto May Sector Manager: Mauricio Carrizosa Lead Economist: Marcelo Giugale Task Manager: Gladys Lopez-Acevedo Gladys Lopez-Acevedo (LCSPE) wrote this report; Joseph S. Shapiro and Erica Soler provided valuable research and editorial assistance. Volume I of this report summarizes the background papers in Volume 11. We want to express appreciation for the close collaboration and report comments offered by Miguel Cervera, Abigail Durin and Alejandro Cano (INEGI headquarters), the General Direction of Labor Statistics, Policies and Studies at the Ministry of Labor (STPS), the General Direction for Science and Technology Policy at the National Science and Technology Council (CONACYT), and the Direction of Technology and Quality at the Ministry of Economy (SE). We are grateful to William F. Maloney (LCSPP) and Hong W. Tan (WBIHD) for their comments as peer reviewers. The lead Economist for Mexico, Marcelo Giugale, provided overall guidance. This study was carried out under the general direction of Mr. Olivier Lafourcade (Director, LCC1 C). Table of Contents 1. Technology and Firm Performance in Mexico ....................................... 5 Gladys L6pez-Acevedo This paper investigates the relationship between a firm's adoption of new manufacturing technology and its performance. A panel database that identifies technological adoption and tracks firms over time allows us to use different measures of firm performance-wages, productivity, net employment growth, job creation, and job destruction. Results show that technology is associated with high firm performance in all these metrics. The effect of new technology on performance is larger for firms located in the North and Mexico City. This marginal value significantly increased after the 1994 crisis and NAFTA. Furthermore, technology increased the wage of semi-skilled workers relative to unskilled workers by about eleven percent over seven years. JEL Codes: L60, L20, J3 1, J38. 2. Technology and Skill Demand in Mexico ................................. 35 Gladys L6pez-Acevedo This paper investigates the effects of technology on the employment and wages of differently skilled Mexican manufacturing workers, using firm panel data for the 1992-99 period. First, we analyze the relationship between technology and skill demand. Findings support the skill-biased technical change hypothesis. Second, we examine the temporal relationship of technology adoption to firm productivity and worker wages. We find that skilled labor increases after technology adoption. One time period after technology adoption, wages of both skilled and semi-skilled workers exhibit markedly increased growth rates relative to the growth rate of low-skilled workers. Third, it is found that investment in human capital magnifies technology-driven productivity gains. JEL Codes: L60 ; L20 ; J31; J38. 3. Determinants of Technology Adoption in Mexico ....................................... 52 Gladys L6pez-Acevedo This paper seeks to identify the impact of firm-, region-, and industry-specific characteristics on technology adoption by Mexican firms. Cross-sectional and panel data for the 1992-99 period show that the firms most likely to adopt new technology are large, train workers, have highly skilled workers, are near the U.S. border, and are owned by foreign entities. Also, larger firms, firms with a large share of highly skilled workers, and firms that train workers, use intensively more complex technologies in their production process. JEL Codes: L60 ; L20 ; J31 and J38. 4. Volume II Appendices ....................................... 86 3 4 TECHNOLOGY AND FIRM PERFORMANCE IN MEXICO Mexico - Technology, Wages, and Employment Technical Paper #1 Gladys L6pez-Acevedol l This research was completed as part of the "Mexico - Technology Wages and Employment" study at the World Bank. We are grateful to the INEGI for providing us with the data. Joseph S. Shapiro and Erica Soler provided valuable research and editorial support. These are views of the author, and need not reflect those of the World Bank, its Executive Directors, or countries they represent. 5 1. Introduction In the last two decades, broad-based reforms at both the sectoral and macroeconomic levels have fundamentally restructured the economic and institutional framework in Mexico. In the mid- 1980s, Mexico began to shift from a state-interventionist system to a market-based economy. Reforms instituted a liberal trade regime, established capital-account convertibility, privatized public enterprises (including banks), and reduced government regulation of the financial, transportation, and utility sectors. At the macroeconomic level, fiscal discipline and structural reform brought about sharp decline in the fiscal deficit and inflation. The government first launched a radical program of policy reforms in 1989 aimed at reducing government regulation and liberalizing trade. Trade liberalization, which began in mid- 1985 and accelerated after Mexico joined the General Agreement on Tariffs and Trade in 1986, further intensified with the adoption of the North American Free Trade Agreement in 1994. Though the external openness of the Mexican economy has quickly expanded, internal reforms have been slower to materialize. The World Bank (1998a) indicated that the productivity difference between export and non-trade sectors reflects the difference in speed between international and internal regulatory reform. It is telling in relation to this that manufacturing, the most important trade sector, improved rapidly in the early 1990s while the service sector deteriorated. But manufacturing only accounts for 25 percent of Mexican gross domestic product, while services account for over 40 percent, which may explain the slow response of the Mexican economy to vigorous trade policy reforms (World Bank 1 998b). During this last decade of rapid development, Mexican wages have polarized. The World Bank (2000) contends that skill-biased technical change caused by trade liberalization explains best the increase in earnings inequality that Mexico has experienced. In this paper, we estimate the effect of new technology adoption (TA) on wage inequality using a rich panel database of manufacturing firms that identifies TA and tracks firms over time. Furthermore, we compare the performance of firms that adopt new technology to those that do not using three separate firm performance measures: the wages earned by workers, the productivity of a firm (output per 6 worker), and the annual growth in the number of employed workers; while other studies have tended to use a single measure of performance. Section 2 of this paper reviews relevant literature on firm performance and TA. Section 3 explains the data and our methodology. Section 4 discusses results for firm performance by time period, firm size, and firm location. Section 5 presents results of the TA determinants and wage performance joint estimation. Section 6 analyzes wage inequality. Section 7 offers conclusions. 2. Literature Review A. Performance Measures Studies measure firm performance in different ways, reflecting both the heterogeneity of the concept and the challenge of practically measuring it. In this paper we use five measures of firm performance-wages, productivity, net employment, job creation, and job destruction. These measures are proxies for a fairly amorphous concept. We want to understand how healthy a firm is, how likely it is to exist in the future, how much utility it creates for workers and consumers, and the contribution it makes to Mexico's development. Our measures by no means exhaustively cover these concepts, which collectively constitute firm performance, but a firm with high marks in these measures also has an exemplary performance. Employment growth is a prevalent measure of firm performance (Geroski 1995). Positive changes in employment represent superior performance; negative changes in employment represent inferior performance. As Caves (1998) documents in his exhaustive compilation, employment growth has been used in many types of studies as a measure of firm performance (Baldwin and Rafiqusszaman 1995; Audretsch 1995; Davis, Haltiwanger, and Schuh 1996b; Baldwin 1995). Employment growth is particularly important for policy makers who focus on job creation. As noted by Davis, Haltiwanger, and Schuh (1996a) job creation and destruction are part of a larger process determining changes in the number and mix of jobs. In this process, new businesses enter the market, some expand, others contract, and many disappear. Additionally, capital, workers, and jobs are continually relocated between different activities. 7 The creation and destruction of jobs requires workers to decide between employment and unemployment. As a result of these processes, some workers must suffer long unemployment spells or severe declines in their earnings. Others may retire early or change residence to find work. A second measure of firm performance is the wages that the firm pays to workers. A healthy firm may pay high efficiency wages, or it may simply maintain high quality of life for its workers by paying high wages. The wages paid by firms have been used as a measure of firm performance in numerous studies, including Aw and Batra (1999), Audretsch and others (2001), Bartel and Lichtenberg (1991), Berman, Bound, and Griliches (1994), Bernard and Jensen (1995), Brown and Medoff (1989), Dunne and Schmitz (1995), Doms, Dunne, and Troske (1997), and Oosterbeek and van Praag (1995). Another firm performance used in this paper is firm productivity. This measure has also been used in numerous studies, including Baldwin and Rafiquzzaman (1995), Baldwin (1995), Bartel and Lichtenberg (1991), Aw and Batra (1999), Baily, Bartelsman, and Haltiwanger (1996), and Baily, Hulten, and Campbell (1992). Higher productivity represents superior performance; lower productivity represents inferior performance. These measures of firm performance are non-identical; in cases they may be contradictory. For example, it is certainly feasible that a firm increases productivity by reducing employment (Baily, Bartelsman, and Haltiwanger 1996). In such an instance, productivity would indicate superior performance, while employment would suggest inferior performance. We try to interpret results in cases where the firm performance measures indicate similar performance patterns. When this similarity is absent from results, we either mention each metric separately or exclude the specific results from discussion. B. Linking Technological Adoption to Firm Performance Measures Some theoretical studies argue against stating unequivocal effects of TA on a developing country's labor force. Braverman (1974) contends that the introduction of advanced technology 8 results in a reduction of the average skill of workers. In this view, technology simply replaces skilled workers. Additionally, Rush and Ferraz (1993) find that technology improvements increase skills for some groups and leave others largely unaffected. A variety of studies link TA to firm performance. One is Doms, Dunne, and Roberts (1995), who examine the impact of advanced manufacturing technology on U.S. manufacturing firms. They use data from the 1988 Survey of Manufacturing Technology to identify the adoption by establishments of 17 different advanced production technologies. These technologies include such innovations as CAD/CAM systems, robots, computers, and networks. They find evidence that firms adopting technology exhibit superior performance. Another is Audretsch and others (2001), who use wages, productivity and employment as performance measures for a panel of firms in The Netherlands. They find that investments in research and development (R&D) and skilled labor improve firm performance. Aw and Batra (1999) provide evidence that technology (measured by R&D and worker training) has an impact on firm performance (measured by wages). This is consistent with the World Bank (1999), which also relates wages to technology (measured by R&D and technology acquisition). Several studies have confirmed the relationship between TA and firm size (Mansfield 1962; Davies 1979; Romeo 1975; and Globerman 1975). This is probably one of the most robust results among surveys analyzing determinants of TA (L6pez-Acevedo 2001). Others have found that firm size determines wages. As noted by Brown and Medoff (1989), other things being equal, large employers pay more than small employers. One way to explain this wage differential is through labor quality. Under this view, larger firms employ higher quality workers due to the greater capital intensity and capital-skill complementarity of larger establishments. Revenga (1995) analyzes the impact of trade liberalization on employment and wages on Mexican manufacturing using panel data of firms for the 1984-1990 period.2 She finds that tariff 2 The data used was drawn from the plant-level Annual Manufacturing Survey. 9 reductions correlate with average wage increases. The correlation may reflect simply an increase in productivity caused by a relative increase in the portion of skilled labor. In a related vein, Tan (2000) investigates manufacturing sector data for Malaysia, and finds that information and communication technology increases total factor productivity by 4 to 6 percent annually. Sargent and Matthews (1997) conclude that installing capital intensive, computer- controlled production machinery into a formerly manual Mexican plant does not impel a firm to train low skilled workers. If the adoption of advanced manufacturing technologies causes an increase in plant size, then it also increases the firm's skill development activity. However, they also find that productivity and skill development do not correlate with compensation. 3. Data and Methodology The data used in this paper comes from a panel of manufacturing firms created with data from the National Survey of Employment, Salaries, Technology, and Training (ENESTYC) and the Annual Industry Survey (EIA). The panel includes observations for 1992, 1995, and 1999.3 Our goal is to understand, for particular types of firms, how is technology related to each firm's performance measure. For this estimation, we use a similar specification for the different performance measures: log(Pi,) = A + AlX1t + fizadoptit + 6it (1) where: log(Pi,) = the logarithm of the performance measure; Xi, = a vector of firm characteristics; adoptit = a dummy variable indicating whether the firm adopted new technology; Ei, = normal regression error; i = refers to the firm being considered, and t = the time period. 3 For a description of these surveys and the panel see Appendix A and B in this volume. 10 For the productivity measure, we include a continuous variable for capital assets to control for correlation between capital and TA, since both influence productivity. Within each measure, for each time period, we restrict the sample only to firms of a particular size or location to estimate situation-specific effects. We do not present results by industry, nor for microenterprises, due to insufficient observations. We measure wages in real pesos, productivity as units of output divided by the number of workers, and net employment as the difference between new hires and dismissals for a given year. Since we have detailed plant level information, we measure net job creation using firm- level employment changes, rather than worker-level changes. 4. Results Several models were estimated. Only the results from the best models are discussed here. We estimated equation (1) using a fixed effects model specification.4 As an experiment, we also estimated a random effects model specification, however, the results were broadly similar, though the fixed effects model tended to yield more robust estimates of the TA paramneters of interest. Therefore, we only discuss the results of the fixed effects estimations for all the measures, organized by the sample universe (only small firms, only firms in the North, etc.), in Table 1.5 A. Overall On balance, firms that adopt new technology exhibit superior performance in all the metrics than those firms that did not adopt technology. Controlling for firm size, age, the skill level of workers, and firms in the maquila sector, firms that adopted new technology in the 1992- 99 sample are related with higher wages for workers of all skill levels. Controlling also for 4 The fixed effects model implements the first differencing approach that generates parameter estimates measured in terms of changes over time and, at the same time, eliminates any potential biases from unmeasured firm-level factors that may be correlated with included variables. 5 Tables Al . I-Al .38 show complete results of the fixed effects estimations for each finn performance measure. 11 capital assets, firms that adopted new technology in the same period are associated with a 26 percent higher productivity than firms that did not adopt technology. Table 1. Relation between Technology Adoption and Firm Performance Sample Measure 1992-95 1995-99 Diff. 1992-99 All Wages: Total 0.5058 ** 0.5594 ** 0.0536 1.2417 ** Highly skilled 0.2817 ** 0.5265 ** 0.2448 1.0614 ** Semi-skilled 0.4981 ** 0.5866 ** 0.0885 1.2722 ** Low skilled 0.2861 ** 0.4271 ** 0.1410 1.2529 ** Productivity 0.0549 ** 0.5360 ** 0.4811 0.2577 ** Net employment 0.3382 * 0.1130 j-0.2252 0.0011 Job creation 0.1846 ** 0.2189 ** 0.0343 1 0.0985 Job destruction 0.1040 ** 0277 -0.1317 0.0438 Small size Wages: Total 0.2284 ** 0.2756 * 0.0472 :1.9678 ** Highly skilled 0.1329 0.2506 0.1177 2.1315 ** Semi-skilled 0.2242 ** 0.2432 0.0190 1.9052 ** Low skilled 0.2393 ** 0.3264 * 0.0871 2.2553 ** Productivity 0.0773 ** 0.3747 0.2974 -0.0229 Net employment 0.1736 - - -0.1965 ................. ................... ............. .... . ........ . . .. . ... . . ... ...... ....... ..... .. .. ........... ...... ..... I..... . .. .... ......... .... ........ . ..... ...... .. .............. .... .......... ... Medium size Wages: Total 0.2711 ** 0.4696 ** 0.1985 1.6908 ** Highly skilled 0.3023 ** 0.4374 ** 0.1351 1.5258 ** Semi-skilled 0.2269 ** j 0.4664 ** 0.2395 1.6805 ** Low skilled 0.2145 * 0.3948 ** 0.1803 1.7769 ** Productivity 0.0839 0.3778 ** 0.2939 0.2025 ** Net employment 0.4949 ,-0.2620 -0.7569 -0.0021 Large size Wages: Total 0.3797 * 0.5302 ** 0.1505 1.4971 ** Highly skilled 0.5272 0.5526 ** 0.0254 1.3165 ** Semi-skilled 0.4442 0.4974 ** 0.0532 1.5389 ** Low skilled 0.0688 0.4242 * 0.3554 1.6095 ** Productivity -0.5443 0.4122 ** 0.9565 0.2271 ** Net employment 0.0238 0.2741 0.2503 0.4370 North Total wages 0.2580 ** 0.5831 * 0.3251 0.6985 Productivity -0.0368 0.7089 ** 0.7457 0.4051 ** .-..Net em. oyment 0.5536 0.0501 -0.5035 -0.0097 Center Totalwages 1.1191 ** 0.5582 ** '-0.5609 1.3955 ** Productivity 0.0947 ** 0.4634 ** 0.3687 0.2552 ** ...... ..Net employment -0.0467 0.3821 0.4288 0.822 South Total wages 2.1293 ** 0.4658 ** j-1.6635 I1.5689 Productivity 0.0017 0.4959 0.4942 0.1573 Net employment -1.6972 0.2310 1.4662 1.2882 Mexico City Total wages 0.3618 ** 0.6487 ** 0.2869 1.5586 ** Productivity 0.0375 0.4866 ** 0.4491 0.0923 Net employment 0.5631 * -0.2403 -0.8034 0.1190 * Significant at 10% level; ** Signifiant at 5% level. Note: Figures show regression coefficients for the TA indicator variable, which in these models can be interpreted as elasticities. 12 In the later period of 1995-99, firms adopting new technology are associated with 56 percent higher wages, and 54 percent higher productivity than firms that did not adopt technology. In the earlier period of 1992-95, firms that adopted new technology are related with 51 percent higher wages, 5 percent higher productivity, and employment of 34 percent more workers than firms that did not adopt technology. B. Time Period: 1992-95 versus 1995-99 For all the firm performance measures we find a marked change in the influence of technology between 1992-95 and 1995-99. Technology relation with wage and productivity performance is significantly larger in the latter period than in the earlier period. The only exceptions are for wage performance in firms located in the Center and South regions. Firms adopting technology are associated with 51 percent higher wages in the early period, and 56 percent higher wages in the later period, than firms that did not adopt technology. Although the net employment measurement for all firms appears to contradict this trend, net employment is not significant in the later period. The relation of technology with job creation, measured as the number of new hires in a given year, is higher for the 1995-99 period than for the 1992-95 period. Moreover, technology is positively associated with job destruction, measured as the numbers of dismissals in a given year, in the 1992-95 period, while there is no significant relation in the 1995-99 period. In only two statistically significant cases the relation of technology with firrns' performance was higher in the early period than the latter. In the Center and South of Mexico, technology was less effective in 1995-99 than in 1992-95. In the North, the change in the wage performance between time periods was 32 percent; in the Capital, the change was 29 percent; in the Center, the change was -56 percent, and in the South it was -166 percent. We should note that in both periods technology still is associated with higher wages, but in the Center and South technology is related to wages by a smaller percentage in the later period than in the earlier period. Much of Mexico's trade-dependent industry is in the North near the U.S. border and in 13 the Capital. It may be that these industries were more affected by liberalization and the 1994 crisis, and so the increased competition they experienced added to the value of technology for them. C. Firm Size Technology is associated with higher wages in all firm sizes, but for the 1992-95 and 1995-99 periods, the relation between technology and wage performance positively correlates with firm size. However, for the overall period (1992-99), the relation of technology with wage performance is larger for smaller firms. Closer analysis of firm size paints a different picture. We ran several regressions where the dependent variable, rather than total wages, was the wages of a particular skill group. These regressions clarify the relation of technology with different types of workers. We proceeded to conduct separate analyses for small firms only, for medium firms only, and for large firms only. These analyses suggest a robust conclusion for the 1992-99 period. For a worker of any single skill group, technology negatively correlates with firm size. For highly skilled workers, small technology firms are associated to a wage increase of 213 percent, medium technology firms of 153 percent, and large technology firms of 132 percent. For low skilled workers, small technology firms are related to a wage increase of 226 percent, medium technology firms are related to a wage increase of 178 percent, and large technology firms are related to a wage increase of 161 percent. Wages for semi-skilled workers experience similar differences. It appears that for large firms relative to small ones, technology increases employment to some extent but decreases wages. In absolute terms, technology increases wages and employment in both small and large firms, but its relative effect differs between firm sizes. The relation of technology with the performance of a firm's productivity also positively correlates with firm size. For medium-size firms, technology is associated with a 20 percent effect on productivity, while for large firms it is 23 percent. 14 D. Firm Location No regional relationship exists in the first time period, but in the later period, firms located in the Capital or close to the U.S. border, present the largest effect of technology on performance. In the 1995-99 period, technology firms in the North are associated to a 58 percent wage increase over their non-technology peers; firms in the Capital are associated to a 65 percent benefit, firms in the Center are associated to a 56 percent benefit, and firms in the South are associated to a 47 percent benefit. For productivity, Northern technology firms are related to a 71 percent benefit, Capital technology firms are related to a 49 percent benefit, and Central technology firms are related to a 46 percent benefit. However, in the earlier period, this trend was reversed: Northern technology firms were associated with a 26 percent wage benefit, Capital technology firms were associated with a 36 percent benefit, Central firms were associated with a 112 percent benefit, and Southern firms were associated with a 213 percent benefit. For the complete 1992-99 period, the highest relation between productivity and technology is for the Northern firms (40 percent), and the highest relation between technology and wages is for the Capital firms (156 percent). 5. A Joint Estimation for Wage Performance and Technology Adoption In addition to the association between TA and firm performance we took into account the causality between TA and firm performance. Therefore, we conducted a joint estimation for the TA and worker wages equations using a three-stage least squares method. Since this paper investigates the relation of technology with firm performance rather than the determinants of TA, we only show results for the regression with worker wages as dependent variable (Table 2). These results present expected findings. Technology is related to wages by quite large amounts in all three-time periods. However, surprisingly, we find that this relation is larger for the 1992-95 period than for the 1995-99 period. Larger firms paid higher wages than smaller firms in the later period, though in the first period (1992-95) smaller firms appeared to pay higher wages than large firms. 15 Table 2. Joint Estimation for Wage Performance and Technology Adoption Dependent Variable: 1992-95 1995-99 1992-99 Log(Total Wages) Coeff. Z-St. Coeff. Z-St. Coeff. Z-St. Firm size: Small -0.7557 ** -9.840 -0.6829 -1.528 -2.3483 ** -3.540 Medium -1.3218 ** -9.249 0.0216 0.048 -1.9225 ** -2.733 Large -2.7490 ** -11.724 0.9890 ** 2.113 -0.9771 -1.345 Age 0.0119 ** 6.878 0.0063 ** 5.830 0.0075 ** 5.474 Share of labor: Semi-skilled 0.0123 ** 3.911 0.0152 ** 3.319 0.0266 ** 3.445 Low skilled 0.0077 ** 2.516 0.0091 ** 2.012 0.0226 ** 2.923 Maquila 0.0055 0.094 0.0381 0.727 -0.0192 -0.319 Technology adoption 4.2530 ** 9.416 2.3211 ** 6.723 2.6770 ** 4.523 Year: 1995 -5.6625 ** -51.945 -5.9910 * -48.656 1999 0.2261 ** 2.462 -5.8385 ** -94.480 Constant 8.5051 ** 18.443 3.1167 ** 6.019 9.5515 ** 9.918 Number of obs. 6,425 3,388 3,141 R-sq (Technology adoption) 0.1455 0.1028 0.0770 R-sq (Wage Performance) 0.7244 0.0771 0.8449 * Significant at 10% level; **Significant at 5% level. 6. Wage Inequality To estimate the effect of TA on wage inequality, we estimate fixed effects models where the dependent variable is the logarithm of the wages of skilled workers divided by the wages of unskilled workers. We run two regressions: one for the logarithm of the ratio of highly skilled workers' wages to unskilled workers' wages, and another for the logarithm of the ratio of semi- skilled workers' wages to unskilled workers' wages. Table 3 shows that, controlling for relevant firm characteristics; technology has exacerbated the wage gap between semi-skilled and unskilled workers by about eleven percent in the seven years of our sample. Additionally, the higher the overall skill level of a firm, the larger the wage gap between skilled and unskilled workers. We also find that smaller firms have worse wage inequality than larger firms in the 1992-95 period. Results for wage inequality between highly skilled and unskilled workers appear in Table A2. 1. TA worsens wage inequality between highly skilled and unskilled workers in all three periods, but results are statistically insignificant. However, as in the first case, the higher the overall skill level of a firm, the larger the wage gap between highly skilled and unskilled workers. 16 Table 3. Wage Inequality Dependent variable: 1992-95 1995-99 1992-99 Log(semi-skilled/unskilled wages) Coeff. Z-St. Coeff. Z-St. Coeff. Z-St. Firm Characteristics Size: Small 3.3306 * 1.786 -0.8229 -1.069 -0.8450 -0.769 Medium 1.8374 1.137 -0.9877 -1.277 -0.8427 -0.762 Large 1.9115 ** -1.987 -0.9186 -1.186 -0.7914 -0.714 Age -0.0485 ** -5.125 0.0066 ** 2.516 -0.0061 ** -2.306 Employees: Highly skilled 0.0045 ** 2.197 0.0007 0.336 0.0049 ** 2.113 Semi-skilled 0.0015 ** 13.610 0.0025 ** 15.259 0.0022 ** 14.74 Unskilled -0.0016 ** -15.263 -0.0025 ** -15.607 -0.0025 ** -17.512 Maquila -0.1802 -1.428 0.0084 0.093 0.0776 0.872 Technology adoption 0.0059 0.136 0.1270 ** 2.064 0.1136 ** 2.145 Constant 0.5366 0.535 1.3566 * 1.748 1.7584 1.585 Number of obs. 5,733 3,075 2,910 R-sq: Within 0.1518 0.2985 0.2962 Between 0.0117 0.3432 0.3792 Overall 0.0127 0.3297 0.3436 * Significant at 10% level; **significant at 5% level. 7. Conclusions Using a panel of firms with observations in 1992, 1995, and 1999, this paper has sought to understand how new technology correlates with the performance of Mexican manufacturing firms, measured by wages, productivity, net employment, job creation, and job destruction. We use fixed effects models to estimate firm performance and determine wage inequality. Results suggest that controlling for relevant variables, technology is positively related to firm performance. Trade liberalization and the 1994 crisis magnified this relation. The effect of new technology on firm performance also correlates positively and strongly with firm size, and proximity to the U.S. border or location in Mexico City. Results present expected findings, that is, technology is correlated with higher wages in all time periods. In an analysis of the behavior of wages, TA improves the wages of both low-skill workers and high skill workers, although it improves the latter more. 17 References Audretsch, D. B. 1995. Innovation and Industry Evolution. Cambridge: MIT Press. Audretsch, D. B., G. van Leeuwen, B. Mendveld, and R. Thurik. 2001. "Market Dynamics in the Netherlands: Competition Policy and the Role of Small Firms." International Journal of Industrial Organization 19(2) :795-822. Aw, B. Y. and G. Batra. 1999. "Wages, Firm Size and Wage Inequality: How Much do Exports Matter?." In D. B. Audretsch and R. Thurik, eds., Innovation, Industry Evolution and Employment. Cambridge: Cambridge University Press :13-56. Baily, M. N., C. Hulten, and D. Campbell. 1992. "The Distribution of Productivity in Manufacturing Plants." Brookings Papers on Economic Activity: Microeconomics: 187-267. Baily, M. N., E. J. Bartelsman, and J. Haltiwanger. 1996. "Downsizing and Productivity Growth: Myth or Reality?." Small Business Economics 8(4) :259-78. Baldwin, J. R. 1995. The Dynamics of Industrial Competition. Cambridge: Cambridge University Press. Baldwin, J. R. and M. Rafiquzzaman. 1995. "Selection versus Evolutionary Adoption: Learning and Post- Entry Performance." International Journal of Industrial Organization 13(4):501-22. Bartel, A. P. and F. R. Lichtenberg. 1991. "The Age of Technology and Its Impact on Employee Wages." Economics of Innovation and New Technology 1(2) :215-31. Berman, E., J. Bound, and Z. Griliches. 1994. "Changes in the Demand for Skilled Labor within U.S. Manufacturing: Evidence from the Annual Survey of Manufactures." Quarterly Journal of Economics 109(2) :367-97. Bernard, A. B. and J. B. Jensen. 1995. "Exporters, Jobs, and Wages in U.S. Manufacturing: 1976-1987." Brookings Papers on Economic Activity: Microeconomics :67-112. Braverman, H. 1974. "Labor and Monopoly Capital: The Degradation of Work in the 20th Century." New York: Monthly Review Press. Brown, C. and J. Medoff. 1989. "The Employer Size-Wage Effect." Journal of Political Economy 97(5) :1027-59. Caves, R. E. 1998. "Industrial Organization and New Findings on the Turnover and Mobility of Firms." Journal of Economic Literature 36(4):1947-82. Davies, S. 1979. Diffusion of Process Innovations. Cambridge: Cambridge University Press. Davis, S. J., J. Haltiwanger, and S. Schuh. 1996a. Job Creation and Destruction. Cambridge: MIT Press. ---------. 1996b. "Small Business and Job Creation: Dissecting the Myth and Reassessing the Facts." Small Business Economics 8(4) :297-315. Doms, M., T. Dunne, and M. J. Roberts. 1995. "The Role of Technology Use in the survival and Growth of Manufacturing Plants." International Journal of Industrial Organization 13(4) :523-42. Doms, M., T. Dunne, and K. R. Troske. 1997. "Workers, Wages, and Technology." Quarterly Journal of Economics 112(1): 253-90. 18 Dunne, T. and J. A. Schmitz, Jr. 1995. "Wages, Employment Structure and Employer Size-Wage Premia: Their Relationship to Advanced-Technology Usage at U.S. Manufacturing Establishments." Economica 62(245): 89-107. Geroski, P. A. 1995. "What Do We Know About Entry?." International Journal of Industrial Organization 13(4) :421-40. Globerman, S. 1975. "Technological Diffusion in the Canadian Tool and Die Industry." Review of Economics and Statistics 57(2) :428-34. L6pez-Acevedo, G. 2001. "Determinants of Technology Adoption in Mexico." Technical Paper. World Bank. Mansfield, E. 1962. "Entry, Gibrat's Law, Innovation, and the Growth of Firms." American Economic Review 52(5) :1023-5 1. Oosterbeek, H. and M. van Praag. 1995. "Firm Size Wage Differentials in the Netherlands." Small Business Economics 7(3):173-82. Revenga, A. 1995. "Employment and Wage Effects of Trade Liberalization." Policy Research Working Paper 1524. World Bank. Romeo, A. 1975. "Inter-industry and Inter-firm Differences in the Rate of Diffusion of an Innovation." Review of Economics and Statistics 57(2) :311-19. Rush, H. and J. C. Ferraz. 1993. "Employment and Skills in Brazil: The Implications of New Technologies and Organizational Techniques." International Labour Review 132(1) :75-93. Sargent, J. and L. Matthews. 1997. "Skill Development and Integrated Manufacturing in Mexico." World Development 25(10): 1669-81. Tan, H. 2000. "Technological Change and Skills Demand: Panel Evidence from Malaysian Manufacturing." Working Paper. The World Bank Institute. World Bank. 1998a. "Mexico Country Economic Memorandum: Enhancing Total Factor Productivity Growth." Report No. 17392-ME (Gray Cover), August. ---------. 1998b. "Mexico: Strengthening Enterprise Finance: Recent Trends in Enterprise Financing." Report No. 17733-ME (Green Cover), September. ---------- 1999. "Mexican Labor Markets: New Views on Integration and Flexibility." Volume Two: Technical Papers. Poverty Reduction and Economic Management Unit, Mexico Department. ---------. 2000. "Earnings Inequality after Mexico's Economic and Educational Reforms." Report No. 19945-ME (Gray Cover), May. 19 ANNEX 1: Firm Performance Fixed Effects Estimations Table A1.1. Wage Performance of Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Size: Small -11.6052 ** -5.456 0.2905 0.672 1.9815 0.814 Medium -24.5036 ** -12.423 0.7041 1.631 2.3509 0.960 Large -50.4489 ** -31.028 1.1036 ** 2.559 2.3329 0.953 Age -2.0664 ** -143.300 0.0105 ** 6.150 -0.1000 ** -11.300 Employees: Highly skilled 0.0029 0.891 0.0054 ** 3.957 -0.0161 ** -2.005 Semi-skilled 0.0008 ** 5.068 0.0016 ** 17.057 0.0015 ** 3.870 Lowskilled 0.0008 ** 5.306 0.0011 ** 11.233 0.0031 ** 6.677 Maquila -0.0773 -0.396 -0.0188 -0.315 -0.2846 -0.914 Technology adoption 0.5058 ** 7.472 0.5594 ** 13.983 1.2417 ** 6.554 Constant 75.0886 ** 59.810 4.6725 ** 10.834 8.0708 ** 3.294 Number of obs. 6,425 3,403 3,184 R-sq: within 0.8865 0.3514 0.0959 between 0.0131 0.6732 0.0441 overall 0.0162 0.5944 0.0369 * Significant at I0/o level; ** Significant at 5% level. Table A1.2. Wage Performance of Highly Skilled Workers in Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Size: Small -10.4433 ** -8.442 -1.4432 ** -2.349 0.6486 0.218 Medium dropped -1.0629 * -1.727 0.8157 0.274 Large -37.1335 ** -24.185 -0.8533 -1.386 0.5432 0.183 Age -2.0898 ** -133.917 0.0100 ** 3.937 -0.1030 ** -10.616 Employees: Highly skilled 0.0367 ** 9.147 0.0212 ** 10.556 0.0026 0.285 Semi-skilled 0.0007 ** 2.902 0.0015 ** 8.833 0.0007 * 1.677 Low skilled 0.0006 ** 2.908 0.0010 ** 5.858 0.0026 ** 5.033 Maquila 0.1374 0.656 -0.0482 -0.536 -0.3140 -0.905 Technology adoption 0.2817 ** 3.870 0.5265 ** 8.732 1.0614 ** 4.951 Constant 68.7051 ** 100.679 4.3511 ** 7.084 8.0569 ** 2.709 Number of obs. 5,091 2,860 2,787 R-sq: within 0.9093 0.2645 0.0890 between 0.0007 0.3563 0.0095 overall 0.0015 0.3450 0.0200 * Significant at 100/ level; ** Significant at 5% level. 20 Table A1.3. Wage Performance of Semi-Skilled Workers in Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff t-Stat Coeff. t-Stat. CoeM t-Stat. semi-skilled workers) . . Firm Characteristics Size: Small -15.7456 ** -7.043 0.8519 1.300 2.6544 0.876 Medium -30.3552 ** -14.189 1.1929 * 1.813 3.0990 1.011 Large -51.0929 ** -30.787 1.6078 ** 2.440 3.1096 1.014 Age -2.0843 ** -138.594 0.0128 ** 6.341 -0.1010 ** -11.240 Employees: Highly skilled 0.0007 0.219 0.0024 1.495 -0.0190 ** -2.324 Semi-skilled 0.0012 ** 7.515 0.0021 ** 19.023 0.0018 ** 4.647 Low skilled 0.0004 ** 2.388 0.0005 ** 3.836 0.0025 ** 5.308 Maquila -0.0358 -0.178 -0.0429 -0.605 -0.2572 -0.814 Technology adoption 0.4981 ** 7.065 0.5866 ** 12.357 1.2722 ** 6.617 Constant 77.0467 ** 58.036 3.4413 ** 5.219 6.7411 ** 2.192 Number of obs. 6,230 3,380 3,177 R-sq: within 0.8845 0.3251 0.0919 between 0.0086 0.5735 0.0266 overall 0.0125 0.5150 0.0325 * Significant at 10% level; ** Significant at 5% level. Table A1.4. Wage Performance of Low Skilled Workers in Manufacturing Fir s Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for low Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. skilled workers) Firm Characteristics Size: Small -2.6941 -0.980 -0.0745 -0.121 1.8284 0.603 Medium 1.4820 0.622 0.4479 0.728 2.1268 0.700 Large -51.8257 ** -36.558 0.7639 1.241 2.1840 0.719 Age -2.1426 ** -155.520 0.0060 ** 2.539 -0.1122 ** -10.903 Employees: Highly skilled -0.0033 -1.089 0.0021 1.143 -0.0268 ** -2.954 Semi-skilled -0.0003 * -1.831 0.0001 0.375 0.0011 ** 1.957 Low skilled 0.0019 ** 12.157 0.0028 ** 19.271 0.0046 ** 8.182 Maquila 0.1674 0.900 -0.0392 -0.474 -0.3695 -1.069 Technology adoption 0.2861 ** 4.547 0.4271 ** 7.595 1.2529 ** 6.088 Constant 67.9290 ** 46.358 3.7585 ** 6.109 7.0532 ** 2.323 Number of obs. 5,896 3,095 2,916 R-sq: within 0.9131 0.2863 0.1108 between 0.0015 0.5918 0.0972 overall 0.0066 0.5154 0.0565 ' Significant at 10% level; ** Significant at 5% level. 21 Table A1.5. Productivity Performance of Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 Log(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Size: Small dropped 3.0028 ** 2.525 3.3980 ** 3.509 Medium -3.7131 ** -5.359 2.6162 ** 2.216 3.1127 ** 3.234 Large dropped 2.6404 ** 2.243 3.0252 ** 3.150 Age 0.1150 ** 19.084 0.0130 ** 4.012 0.0183 ** 8.095 Employees: Highly skilled 0.0006 0.508 0.0021 0.693 0.0059 ** 3.117 Semi-skilled -0.0003 ** -5.388 -0.0001 -0.317 0.0001 0.831 Low skilled -0.0004 ** -5.818 0.0001 0.366 -0.0003 ** -2.753 Maquila 0.0600 0.823 -0.2120 * -1.731 -0.0323 -0.398 Technology adoption 0.0549 ** 2.051 0.5360 ** 7.022 0.2577 ** 5.355 Capital assets 0.0000 -0.595 1.4e-06 ** 2.872 1.9e-06 ** 7.450 Constant 1.0192 ** 5.208 1.0563 0.902 0.4836 0.505 Number of obs. 3,894 2,101 2,714 R-sq: within 0.2182 0.1261 0.1083 between 0.0142 0.0953 0.1126 overall 0.0146 0.1079 0.1050 * Significant at 10% level; ** Significant at 5% level. Table A1.6. Net Employment Performance of Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Size: Small dropped 0.8657 0.750 -0.0065 -0.017 Medium dropped 1.5397 1.325 -0.1841 -0.915 Large dropped 1.5874 1.361 dropped Age 0.0139 0.353 -0.0053 -0.567 -0.0025 -0.337 Employees: Highly skilled -0.0078 -1.369 0.0172 ** 2.014 0.0104 ** 2.088 Semi-skilled 0.0013 * 1.909 0.0013 ** 2.580 0.0004 1.324 Low skilled 0.0017 ** 4.915 0.0006 1.028 0.0006 * 1.745 Maquila -1.2054 ** -2.132 0.2843 1.085 0.0446 0.201 Technology adoption 0.3382 * 1.838 0.1130 0.617 0.0011 0.007 Constant 1.6938 1.611 0.5863 0.494 2.3674 ** 7.164 Number of obs. 1,680 1,323 1,158 R-sq: within 0.1260 0.1016 0.0374 between 0.0742 0.1132 0.1313 overall 0.0726 0.1177 0.1117 * Significant at 10% level; * Significant at 5% level. 22 Table A1.7. Job Creation of Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(new hires) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Size: Small dropped 0.9377 0.885 dropped Medium -0.2618 -0.253 1.3384 1.261 0.3246 * 2.456 Large dropped 1.5009 1.414 0.4757 ** 0.187 Age -0.0360 ** -2.639 -0.0027 -0.772 0.0006 1.191 Employees: Highly skilled 0.0004 0.161 0.0023 0.912 0.0038 8.484 Semi-skilled 0.0015 ** 8.745 0.0013 ** 6.446 0.0013 ** 7.416 Low skilled 0.0014 ** 9.386 0.0009 ** 4.890 0.0013 ** 0.486 Maquila -0.0110 -0.056 -0.1353 -1.153 0.0559 1.334 Technology adoption 0.1846 ** 2.820 0.2189 ** 2.732 0.0985 11.942 Constant 3.5437 ** 8.555 1.4705 1.382 2.3608 ** 1.884 Number of obs. 4,262 2,714 2,494 R-sq: within 0.0961 0.0804 0.0814 between 0.1819 0.1506 0.1388 overall 0.1657 0.1426 0.1133 * Significant at 10% level; ** Significant at 5% level. Table A1.8. Job Destruction of Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1 1992-1999 log(laidoffs) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics I Size: Small 0.4891 0.499 0.5472 0.542 -1.0652 ** -6.302 Medium 1.3311 1.317 -0.3599 ** -3.933 Large -2.7041 -1.587 1.5890 1.571 dropped Age -0.0046 -0.408 -0.0021 -0.741 -0.0074 ** -2.552 Employees: Highly skilled 0.0007 0.283 0.0010 0.434 0.0016 0.551 Semi-skilled 0.0005 ** 3.092 0.0006 ** 3.349 0.0008 ** 5.329 Low skilled 0.0007 ** 5.573 0.0006 ** 3.459 0.0007 ** 4.411 Maquila -0.1039 -0.682 -0.1767 * -1.716 -0.0286 -0.275 Technology adoption 0.1040 ** 2.007 -0.0277 -0.396 -0.0438 -0.702 Constant 3.0692 ** 5.866 1.8903 * 1.867 3.6234 ** 27.782 Number of obs. 5,076 2,885 2,723 R-sq: within 0.0236 0.0594 0.0576 between 0.2641 0.1831 0.2337 overall 0.2013 0.1658 0.1710 * Significant at 10% level; ** Significant at 5% level. 23 Table Al.9. Wage Performance of Small Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.2342 ** -130.885 0.0115 1.345 -0.6836 ** -7.114 Employees: Highly skilled 0.0144 * 1.841 0.0509 1.305 -0.2212 -1.215 Semi-skilled 0.0035 ** 5.862 0.0230 ** 5.595 -0.0034 -0.351 Low skilled 0.0019 ** 4.538 0.0140 ** 3.107 -0.0033 -0.394 Maquila 0.0946 0.497 0.0003 0.001 0.5370 0.308 Technology adoption 0.2284 ** 2.970 0.2756 * 1.808 1.9678 ** 2.909 Constant 66.7124 ** 140.760 3.2216 ** 9.437 24.5766 ** 9.567 Number of obs. 2,187 311 206 R-sq: within 0.9526 0.4274 0.4523 between 0.0056 0.3573 0.0414 overall 0.0061 0.3508 0.0346 * Significant at 10% level; ** Significant at 5% level. Table Al.10. Wage Performance of Highly Skilled Workers in Small Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. highly skilled workers) Firm Characteristics Age -2.1817 ** -102.417 0.0158 1.172 -0.6266 ** -5.681 Employees: Highly skilled 0.0551 ** 6.100 0.2484 ** 3.904 -0.0201 -0.085 Semi-skilled 0.0015 * 1.954 0.0142 * 1.984 -0.0073 -0.686 Low skilled 0.0006 1.134 0.0060 0.835 -0.0082 -0.883 Maquila 0.2111 0.866 -0.4179 -1.204 0.5077 0.275 Technology adoption 0.1329 1.381 0.2506 0.876 2.1315 ** 2.606 Constant 64.5873 ** 107.615 1.4061 ** 2.572 21.5287 ** 7.044 Number of obs. 1,838 254 172 R-sq: within 0.9423 0.3667 0.4151 between 0.0004 0.3362 0.0326 overall 0.0032 0.3179 0.0268 * Significant at 10% level; ** Significant at 5% level. Table A1.11. Wage Performance of Semi-Skilled Workers in Small Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.2457 ** -127.415 0.0151 1.551 -0.6718 ** -6.664 Employees: Highly skilled -0.0084 -1.036 -0.0133 -0.296 -0.2687 -1.381 Semi-skilled 0.0076 ** 12.233 0.0361 ** 7.740 0.0036 0.364 Low skilled -0.0003 -0.628 0.0069 1.337 -0.0094 -1.098 Maquila 0.0902 0.460 0.1015 0.381 0.5639 0.313 Technology adoption 0.2242 ** 2.830 0.2432 1.406 1.9052 ** 2.726 Constant 66.3022 ** 135.379 2.2876 ** 5.882 23.6665 ** 8.832 Number of obs. 2,181 305 205 R-sq: within 0.9515 0.5331 0.4427 between 0.0135 0.4256 0.0427 overall 0.0038 0.4240 0.0345 * Significant at 100/o level; ** Significant at 5% level. 24 Table A1.12. Wage Performance of Low Skilled Workers in Small Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coeff. t-Stat. Coef£ t-Stat. low skilled workers)_ Firm Characteristics Age -2.2078 ** -108.324 0.0107 1.156 -1.1020 ** -10.732 Employees: Highly skilled 0.0284 ** 2.977 0.0556 1.083 -0.3039 -1.633 Semi-skilled -0.0032 ** -4.211 -0.0041 -0.804 -0.0051 -0.517 Low skilled 0.0053 ** 10.661 0.0303 ** 6.279 0.0112 1.094 Maquila 0.1784 0.764 -0.0177 -0.077 1.3936 0.908 Technology adoption 0.2393 ** 2.630 0.3264 * 1.846 2.2553 ** 3.838 Constant 64.9057 ** 114.613 2.1844 ** 5.847 33.0652 ** 12.470 Number of obs. 2,053 275 186 R-sq: within 0.9374 0.5372 0.6569 between 0.0085 0.2551 0.0964 overall 0.0134 0.2747 0.0647 * Significant at 10% level; ** Significant at 5% level. Table A1.13. Productivity Performance of Small Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.0966 ** 11.531 -0.0012 -0.029 0.0907 ** 3.441 Employees: Highly skilled 0.0066 * 1.954 -0.6451 * -2.241 0.0271 0.797 Semi-skilled -0.0014 ** -5.169 0.0134 0.877 -0.0066 ** -2.751 Low skilled -0.0011 ** -4.089 0.0287 1.752 -0.0015 -0.578 Maquila -0.0112 -0.136 -0.7320 -1.038 0.4251 1.048 Technology adoption 0.0773 ** 2.226 0.3747 0.900 -0.0229 -0.155 Capital assets 0.0000 -0.829 0.0000 -0.006 0.0000 -1.047 Constant 1.2966 ** 5.281 4.2832 1.200 2.3305 ** 3.250 Number of obs. 1,605 150 132 R-sq: within 0.2302 0.6767 0.4344 between 0.0058 0.0956 0.0018 overall 0.0069 0.0753 0.0029 * Significant at 10% level; ** Significant at 5% level. Table A1.14. Net Employment Performance of Small Manufacturing Firms Dependent variable: 1992-1995 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -0.1090 * -1.922 -0.0243 -0.160 Employees: Highly skilled 0.0424 1.308 -0.0788 -0.198 Semi-skilled 0.0016 0.951 0.0179 1.152 Low skilled 0.0031 ** 5.065 0 0158 1.437 Maquila 0.3343 0.523 dropped Technology adoption 0.1736 0.675 -0.1965 -0.264 Constant 4.0638 ** 2.889 1.1206 0.323 Number of obs. 585 64 R-sq: within 0.3118 0.3302 between 0.0580 0.3032 overall 0.0729 0.3490 Note: Estimation for 1995-1999 was not possible due to insufficient observations. * Significant at 10% level; ** Significant at 5% level. 25 Table A1.15. Wage Performance of Medium-size Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.2438 ** -99.936 0.0152 ** 5.117 -0.2996 ** -12.433 Employees: Highly skilled 0.0734 ** 2.698 0.0177 ** 2.749 -0.1136 ** -3.200 Semi-skilled 0.0088 ** 4.733 0.0075 ** 14.137 0.0072 ** 2.756 Low skilled 0.0072 ** 4.647 0.0065 ** 10.995 0.0014 0.514 Maquila 0.0037 0.014 -0.0414 -0.450 -0.1122 -0.196 Technology adoption 0.2711 ** 2.981 0.4696 ** 8.295 1.6908 ** 5.547 Constant 56.1743 ** 107.432 4.2907 ** 40.263 15.4095 ** 19.297 Number of obs. 1,139 1,524 1,298 R-sq: within 0.9587 0.4770 0.2217 between 0.0024 0.3347 0.0001 overall 0.0068 0.3487 0.0177 Significant at 10% level; ** Significant at 5% level. Table A1.16. Wage Performance of Highly Skilled Workers in Medium-sized Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 highl skilledworke Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.2079 ** -74.638 0.0153 ** 3.291 -0.2841 ** -10.799 Employees: Highly skilled 0.2540 ** 7.447 0.0866 ** 8.225 -0.0854 ** -2.144 Semi-skilled -0.0003 -0.111 0.0045 ** 5.417 0.0048 * 1.653 Low skilled 0.0048 ** 2.187 0.0053 ** 5.590 -0.0010 -0.318 Maquila 0.2786 0.775 -0.1394 -0.967 -0.2488 -0.388 Technology adoption 0.3023 ** 2.555 0.4374 ** 4.941 1.5258 ** 4.448 Constant 55.0889 ** 77.814 2.3202 ** 13.633 13.5562 ** 15.601 Number of obs. 894 1,285 1,133 R-sq: within 0.9502 0.3287 0.2003 between 0.0145 0.2144 0.0010 overall 0.0139 0.2318 0.0116 * Significant at 10% level; ** Significant at 5% level. Table A1.17. Wage Performance of Semi-Skilled Workers in Medium-sized Ma ufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. semi-skilled workers) Firm Characteristics Age -2.2533 ** -91.120 0.0165 ** 4.701 -0.3018 ** -12.375 Employees: Highly skilled 0.0236 0.791 -0.0040 -0.534 -0.1374 ** -3.825 Semi-skilled 0.0187 ** 8.892 0.0114 ** 18.195 0.0108 ** 4.054 Low skilled -0.0007 -0.404 0.0029 ** 4.114 -0.0023 -0.811 Maquila 0.0164 0.058 -0.1558 -1.438 -0.1340 -0.231 Technology adoption 0,2269 ** 2.263 0.4664 ** 6.954 1.6805 ** 5.448 Constant 55.6548 ** 96.611 3.5941 ** 28.372 14.8949 ** 18.426 Number of obs. 1,131 1,516 1,297 R-sq: within 0.9522 0.4960 0.2354 between 0.0021 0.4313 0.0004 overall 0.0063 0.4443 0.0212 * Significant at 10% level; ** Significant at 5% level. 26 Table A1.18. Wage Performance of Low Skilled Workers in Medium-size Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. low skilled wres __ Firm Characteristics Age -2.1985 ** -78.789 0.0057 1.562 -0.3750 ** -13.232 Employees: Highly skilled 0.0207 0.627 0.0101 1.152 -0.1133 ** -2.627 Semi-skilled 0.0020 0.734 0.0008 1.228 0.0058 ** 2.035 Low skilled 0.0238 ** 10.281 0.0144 ** 19.679 0.0086 ** 2.893 Maquila 0.1101 0.344 0.0070 0.063 -0.0661 -0.109 Technology adoption 0.2145 * 1.920 0.3948 ** 5.739 1.7769 ** 5.623 Constant 54.0983 ** 82.181 3.0757 ** 23.254 15.4893 ** 17.432 Number of obs. 1,042 1,404 1,196 R-sq: within 0.9443 0.5012 0.2632 between 0.0008 0.4903 0.0034 overall 0.0091 0.4860 0.0210 * Significant at 10% level; ** Significant at J% level. Table A1.19. Productivity Performance of Medium-size Manufacturing Firms Dependent variable: r 1992-1995 1995-1999 1992-1999 log(productivity) Coeff. t-Stat. Coeff. t-Stat. i Coeff. t-Stat. Firm Characteristics Age 0.0725 ** 3.330 0.0200 ** 3.508 0.0385 ** 8.168 Employees: Highly skilled -0.0184 -0.834 -0.0005 -0.042 0.0127 * 1.749 Semi-skilled -0.0035 ** -2.448 0.0018 * 1.895 -0.0002 -0.426 Low skilled -0.0029 ** -2.486 0.0029 ** 2.301 -0.0003 -0.447 Maquila 0.1047 0.389 -0.2818 -1.527 -0.1210 -1.000 Technology adoption 0.0839 0.998 0.3778 ** 3.604 0.2025 ** 3.202 Capital assets -2.1e-05 ** -2.111 0.0001 ** 6.928 0.0001 ** 12.179 Constant 2.3731 ** 4.468 2.7011 ** 12.591 2.5359 ** 15.750 Number of obs. 439 919 1,083 R-sq: within 0.1749 0.3774 0.3331 Between 0.0163 0.0239 0.1895 Overall 0.0047 0.0359 0.1659 * Significant at 10% level; ** Significant at 5% level. Table A1.20. Net Employment Performance of Medium-size Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -0.0849 -0.761 -0.0193 -0.594 0.0075 0.404 Employees: Highly skilled 0.5085 ** 2.694 -0.0041 -0.066 0.0114 0.349 Semi-skilled 0.0073 0.510 0.0077 ** 2.272 0.0023 1.027 Low skilled 0.0099 1.296 0.0078 ** 2.116 0.0031 * 1.775 Maquila -1.3010 -0.852 0.2369 0.493 0.5642 * 1.670 Technology adoption 0.4949 1.136 -0.2620 -0.800 -0.0021 -0.008 Constant | 1.7751 0.666 1.4846 * 1.728 1.1655 * 1.905 Number of obs. 267 573 449 R-sq: within 0.3957 0.1136 | 0.0708 between 0.0477 0.1014 0.0655 overall 0.0713 0.1029 0.0775 * Significant at 10% level; ** Significant at 5% level. 27 Table A1.21. Wage Performance of Large Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.2355 ** -44.339 0.0079 ** 3.314 -0.0742 ** -6.113 Employees: Highly skilled 0.1464 1.154 0.0037 * 1.770 -0.0112 -1.244 Semi-skilled 0.0077 0.616 0.0012 ** 12.535 0.0017 ** 3.995 Low skilled 0.0298 ** 2.158 0.0009 ** 8.512 0.0035 ** 6.869 Maquila -0.3268 -0.438 0.0089 0.092 -0.0182 -0.036 Technology adoption 0.3797 * 1.808 0.5302 ** 7.590 1.4971 ** 4.837 Constant 41.6300 ** 48.372 6.1945 ** 59.762 9.4357 ** 16.680 Number of obs. 581 1,560 1,560 R-sq: within 0.9024 0.3194 0.3194 between 0.0023 0.5163 0.5163 overall 0.0037 0.4812 0.4812 * Significant at 10% level; ** Significant at 5% level. Table A1.22. Wage Performance of Highly Skilled Workers in Large Manufact ring Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coef£ t-Stat. Coeff. t-Stat. Coef£ t-Stat. highly skilled workers) Firm Characteristics Age -2.2503 ** -24.594 0.0070 ** 1.968 -0.0775 ** -5.947 Employees: Highly skilled 0.3149 1.560 0.0310 ** 8.194 0.0090 0.903 Semi-skilled 0.0011 0.078 0.0012 ** 6.963 0.0010 ** 2.073 Low skilled 0.0125 0.742 0.0008 ** 4.537 0.0031 ** 5.582 Maquila -0.6012 -0.457 -0.1295 -0.855 0.0385 0.069 Technology adoption 0.5272 1.507 0.5526 ** 5.137 1.3165 ** 3.826 Constant 43.4315 ** 27.187 3.6018 ** 23.058 7.6658 ** 12.304 Number of obs. 261 1,316 1,480 R-sq: within 0.9299 0.3222 0.1016 between 0.0008 0.2641 0.0304 overall 0.0040 0.2827 0.0421 * Significant at 10% level; * * Significant at 5% level. Table A1.23. Wage Performance of Semi-Skilled Workers in Large Manufactur ing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coef£ t-Stat. Coef£ t-Stat. semi-skilled workers) Firm Characteristics Age -2.2117 ** -32.023 0.0087 ** 3.105 -0.0741 ** -6.053 Employees: Highly skilled -0.2247 -1.340 0.0002 0.085 -0.0134 -1.480 Semi-skilled 0.1459 ** 5.411 0.0016 ** 14.254 0.0019 ** 4.509 Low skilled 0.0136 0.868 0.0003 ** 2.904 0.0030 ** 5.893 Maquila 0.2785 0.283 0.0049 0.043 -0.0312 -0.061 Technology adoption 0.4442 1.626 0.4974 ** 6.044 1.5389 ** 4.925 Constant 38.1168 ** 31.355 5.6114 ** 45.972 8.8774 ** 15.551 Number of obs. 408 1,554 1,672 R-sq: within 0.8967 0.3222 0.0963 between 0.0104 0.4492 0.0284 overall 0.0155 0.4290 0.0428 * Significant at 10%/. level; ** Signifimnt at 5% level. 28 Table A1.24. Wage Performance of Low Skilled Wor ers in Large Manufacturi ng Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(wages for Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. low skilled workers) Firm Characteristics Age -2.2318 ** -34.646 0.0049 1.375 -0.0809 ** -5.869 Employees: Highly skilled -0.2194 -1.346 0.0036 1.085 -0.0222 ** -2.194 Semi-skilled 0.0325 0.994 0.0000 0.059 0.0015 ** 2.448 Low skilled 0.0707 ** 4.388 0.0022 ** 13.583 0.0048 ** 7.704 Maquila -1.1808 -0.994 0.0876 0.590 0.1566 0.278 Technology adoption 0.0688 0.273 0.4242 ** 3.910 1.6095 ** 4.788 Constant 40.1928 ** 38.168 4.8480 ** 30.899 7.8986 ** 12.351 Number of obs. 417 1,411 1,532 R-sq: within 0.9085 0.2958 0.1240 between 0.0018 0.5015 0.1068 overall 0.0090 0.4555 0.0791 * Significant at 10% level; ** Significant at 5% level. Table A1.25. Productivity Performance of Large Ma ufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.6864 0.621 0.0100 ** 2.225 0.0147 ** 5.045 Employees: Highly skilled 0.5336 0.861 0.0033 0.874 0.0053 ** 2.649 Semi-skilled 0.0895 0.378 0.0002 0.806 0.0000 -0.045 Low skilled -0.1694 -0.401 0.0002 0.828 -0.0004 ** -3.319 Maquila dropped -0.2318 -1.126 -0.0467 -0.392 Technology adoption -0.5443 -0.638 0.4122 ** 3.124 0.2271 ** 3.188 Capital assets 0.0069 0.223 9.6e-07 * 1.835 0.0000 ** 6.767 Constant -11.0303 -0.291 3.8258 ** 18.679 3.7894 ** 28.237 Number of obs. 14 1,031 1,498 R-sq: within 0.9583 0.0838 0.1216 between 0.4637 0.1333 0.0995 overall 0.3386 0.1351 0.1364 * Significant at 10% level; ** Significant at 5% level. Table A1.26. Net Employment Performance of Large Manufacturing Firms Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.0210 0.100 -0.0030 -0.211 -0.0116 -1.015 Employees: Highly skilled 0.0361 0.067 0.0152 * 1.764 0.0122 ** 2.141 Semi-skilled 0.1342 1.205 0.0005 1.003 0.0002 0.591 Low skilled 0.1779 1.877 0.0003 0.391 0.0003 0.641 Maquila dropped 0.3862 0.881 -0.6237 -1.575 Technology adoption 0.0238 0.023 0.2741 0.889 -0.4370 -1.459 Constant -2.2565 -0.658 2.4349 ** 4.351 3.5170 ** 6.645 Number of obs. 53 647 645 R-sq: within 0.8803 0.0741 0.0783 between 0.1151 0.0437 0.0084 overall 0.1402 0.0487 0.0229 * Significant at 10% level; ** Significant at 5% level. 29 Table A1.27. Wage Performance of Manufacturing Firms in the North Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.1823 ** -116.448 0.0086 ** 2.165 -0.1282 * -5.388 Employees: Highly skilled -0.0041 -1.152 0.0124 ** 3.536 -0.0080 -0.559 Semi-skilled 0.0007 ** 4.172 0.0019 ** 9.730 0.0013 1.337 Low skilled 0.0005 ** 3.868 0.0008 ** 4.823 0.0022 ** 2.131 Maquila -0.0744 -0.297 -0.0636 -0.493 -0.2001 -0.307 Technology adoption 0.2580 ** 2.938 0.5831 ** 6.038 0.6985 1.567 Constant 57.7898 ** 127.590 5.5342 ** 40.187 11.8802 ** 12.658 Number of obs. 1,733 800 630 R-sq: within 0.9467 0.3195 0.0819 Between 0.0110 0.6364 0.0007 Overall 0.0007 0.5662 0.0082 * Significant at 10% level; ** Significant at 5% level. Table A1.28. Productivity Performance of Manufacturing Firms in the North Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.0695 ** 4.691 0.0060 0.759 0.0118 ** 2.280 Employees: Highly skilled -0.0022 -0.496 -0.0018 -0.345 -0.0005 -0.147 Semi-skilled -0.0005 ** -2.597 0.0003 0.599 0.0000 -0.046 Low skilled -0.0003 ** -2.562 0.0002 0.365 -0.0005 ** -2.240 Maquila -0.0822 -0.458 -0.1802 -0.656 -0.1089 -0.710 Technology adoption -0.0368 -0.535 0.7089 ** 3.958 0.4051 ** 3.891 Capital assets 0.0000 0.511 2.1e-06 ** 2.267 2.4e-06 ** 5.659 Constant 2.1514 ** 4.598 3.6255 ** 11.915 3.7589 ** 17.786 Number of obs. 740 454 550 R-sq: within 0.1280 0.2178 0.1700 Between 0.0000 0.1035 0.2205 Overall 0.0002 0.1306 0.1841 * Significant at 10% level; ** Significnt at 5% level. Table A1.29. Net Employm ent Performance of Manu acturing Firms in the North Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.0777 1.106 -0.0147 -0.543 0.0033 0.144 Employees: Highly skilled -0.0087 -1.389 0.0853 ** 2.095 0.0131 1.215 Semi-skilled 0.0014 1.404 0.0011 0.923 0.0002 0.327 Low skilled 0.0014 ** 3.282 0.0010 1.006 0.0011 1.290 Maquila 1.4577 0.828 0.6246 1.201 0.6038 1.158 Technology adoption 0.5536 1.636 0.0501 0.109 -0.0097 -0.023 Constant -0.5939 -0.357 1.9476 ** 2.134 2.0121 ** 2.372 Number of obs. 521 318 211 R-sq: within 0.1967 0.2368 0.0893 between 0.2021 0.0697 0.1108 overall 0.2014 0.0766 0.1251 * Significant at 10% level; ** Significant at 5% level. 30 Table A1.30. Wage Performance of Manufacturing Firms in the Central Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -1.7333 ** -58.310 0.0169 ** 6.179 -0.1269 ** -8.762 Employees: Highly skilled 0.0126 * 1.801 0.0017 1.032 -0.0351 ** -2.446 Semi-skilled 0.0028 ** 5.976 0.0022 ** 13.508 0.0026 ** 3.517 Low skilled 0.0027 ** 5.906 0.0022 ** 12.070 0.0042 ** 6.105 Maquila 0.0003 0.001 -0.0290 -0.337 0.0076 0.017 Technology adoption 1.1191 ** 7.611 0.5582 ** 10.105 1.3955 ** 5.212 Constant 52.3282 ** 63.035 5.1170 ** 55.399 10.4436 ** 20.247 Number of obs. 2,884 1,510 1,612 R-sq: within 0.7655 0.3794 0.1230 Between 0.0667 0.5754 0.1065 overall 0.0009 0.5222 0.0565 * Significant at 10% level; ** Significant at 5% level. Table A1.31. Productivity Performance of Manufacturing Firms in the Central Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.1274 ** 16.716 0.0127 ** 2.269 0.0230 ** 6.204 Employees: Highly skilled 0.0011 0.803 0.0035 0.409 0.0124 ** 3.390 Semi-skilled -0.0004 ** -3.944 0.0004 1.127 -0.0001 -0.431 Low skilled -0.0004 ** -4.591 0.0004 0.879 -0.0006 ** -3.161 Maquila 0.1187 1.285 -0.3592 ** -1.973 -0.2428 ** -1.987 Technology adoption 0.0947 ** 2.769 0.4634 ** 4.072 0.2552 ** 3.610 Capital assets 0.0000 -0.956 0.0000 -0.745 1.2e-06 ** 2.863 Constant 0.4943 ** 2.147 3.7676 ** 19.398 3.5785 ** 26.728 Number of obs. 1,981 951 1,353 R-sq: within 0.2862 0.0796 0.0892 Between 0.0075 0.0871 0.0616 overall 0.0099 0.0776 0.0692 * Significant at 10% level; ** Significant at 5% level. Table Al.32. Net Employment Performance of Manufacturing Firms in the Central Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) L Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -0.0545 -0.792 -0.0010 -0.062 -0.0023 -0.169 Employees: Highly skilled 0.0328 1.259 0.0257 1.314 0.0087 1.045 Semi-skilled 0.0011 0.672 0.0020 ** 2.776 0.0003 0.452 Low skilled 0.0019 * 1.948 0.0023 ** 2.194 0.0009 1.494 Maquila -2.2254 ** -2.763 0.2977 0.758 -0.4759 -1.400 Technology adoption -0.0467 -0.153 0.3821 1.503 0.1822 0.799 Constant 3.6011 * 1.814 1.2197 ** 2.556 2.1821 ** 4.741 Number of obs. 717 594 594 R-sq: within 0.1561 0.1382 0.1382 Between 0.0398 0.1335 0.1335 overall 0.0436 0.1429 0.1429 * Significant at 10% level; ** Significant at 5% level. 31 Table A1.33. Wage Performance of Manufacturing Firms in the South Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -1.2012 ** -15.913 0.0203 ** 2.507 -0.0773 ** -2.254 Employees: Highly skilled 0.0256 0.904 0.0116 0.783 0.0144 0.199 Semi-skilled 0.0036 ** 2.089 0.0017 ** 3.281 0.0003 0.102 Low skilled 0.0084 ** 4.103 0.0015 ** 3.579 0.0040 1.327 Maquila -1.9240 -1.324 0.5831 ** 2.402 -0.7356 -0.377 Technology adoption 2.1293 ** 4.602 0.4658 ** 2.596 1.5689 1.598 Constant 38.8099 ** 17.518 4.9897 ** 16.494 9.6685 ** 5.893 Number of obs. 391 218 151 R-sq: within 0.6829 0.3296 0.1005 Between 0.0708 0.4689 0.0125 overall 0.0056 0.4395 0.0142 * Significant at 10% level; ** Significant at 5% level. Table A1.34. Productivity Performance of Manufact ring Firms in the South Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.0599 ** 2.011 0.0178 * 1.858 0.0210 ** 3.265 Employees: Highly skilled 0.0009 0.123 -0.0157 -0.741 -0.0172 -1.327 Semi-skilled -0.0007 * -1.718 -0.0010 -1.079 -0.0015 ** -2.684 Low skilled -0.0016 ** -2.682 0.0001 0.102 -0.0004 -0.619 Maquila -0.1024 -0.247 0.2675 0.423 0.3057 0.799 Technology adoption 0.0017 0.011 0.4959 1.553 0.1573 0.835 Capital assets 0.0000 0.182 2.0e-05 ** 2.220 2.2e-05 ** 4.681 Constant 2.4305 ** 2.344 3.2620 ** 6.591 3.5023 ** 10.629 Number of obs. 222 112 118 R-sq: within 0.1771 0.3938 0.4045 Between 0.0012 0.0898 0.0146 Overall j 0.0027 0.1260 0.0918 * Significant at 10% level; ** Significant at 5% level. Table A1.35. Net Employment Performance of Manufacturing Firms in the South Region Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.0131 0.068 -0.0996 -0.782 -0.0709 -1.416 Employees: Highly skilled 0.1076 1.038 -0.0655 -0.245 0.0764 0.917 Semi-skilled 0.0029 0.337 -0.0010 -0.351 0.0010 0.445 Low skilled 0.0074 1.684 0.0002 0.025 0.0062 1.445 Maquila -1.0224 -0.731 2.5018 0.813 1.4192 0.818 Technology adoption -1.6972 -1.509 -0.2310 -0.193 -1.2882 -1.418 Constant 0.7254 0.136 5.5111 ** 2.622 3.8439 * 2.115 Number of obs. 78 81 53 R-sq: within 0.7152 0.4436 0.5148 between 0.1544 0.0102 0.1389 overall 0.1469 0.0078 0.1866 * Significant at 10% level; ** Significant at 5% level. 32 Table A1.36. Wage Performance of Manufacturing Firms in Mexico City Dependent variable: 1992-1995 1995-1999 1992-1999 log(total wages) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -2.2507 ** -104.274 0.0035 1.157 -0.1019 ** -5.665 Employees: Highly skilled 0.0148 ** 2.267 0.0106 ** 3.126 -0.0175 -1.091 Semi-skilled 0.0004 ** 2.190 0.0013 ** 8.206 0.0014 ** 2.124 Low skilled 0.0006 * 1.818 0.0010 ** 4.865 0.0023 ** 2.447 Maquila 0.1028 0.381 -0.0928 -0.686 -1.1305 -1.621 Technology adoption 0.3618 ** 3.882 0.6487 ** 7.693 1.5586 ** 4.089 Constant 76.7338 ** 111.880 5.6920 ** 45.075 11.0466 ** 14.089 Number of obs. 1,417 860 748 R-sq: within 0.9509 0.2884 0.1103 between 0.2344 0.5283 0.0011 overall | 0.0212 0.4698 0.0236 * Significant at 10% level; ** Significant at 5% level. Table A1.37. Productivity Performance of Manufacturing Firms in Mexico City Dependent variable: 1992-1995 1995-1999 1992-1999 _ og(productivity) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age 0.1126 ** 8.402 0.0072 1.396 0.0095 ** 2.189 Employees: Highly skilled 0.0008 0.231 -0.0021 -0.474 0.0031 0.843 Semi-skilled -0.0002 * -1.756 0.0002 0.751 0.0001 0.906 Low skilled -0.0002 -1.446 0.0008 ** 2.172 0.0000 -0.229 Maquila 0.1331 0.847 0.1816 0.806 0.2152 1.231 Technology adoption 0.0375 0.678 0.4866 ** 3.718 0.0923 1.011 Capital assets 3.3e-06 0.881 3.3e-06 ** 3.152 5.5e-06 ** 5.443 Constant 0.0745 0.159 3.6537 ** 17.183 3.6820 ** 19.604 Number of obs. 951 578 652 R-sq: within 0.1915 0.1507 0.0980 Between 0.0504 0.1470 0.2049 overall 0.0476 0.1527 0.1446 * Significant at 10% level; ** Significant at 5% level. Table A1.38. Net Employment Performance of Manufacturing Firms in Mexico City Dependent variable: 1992-1995 1995-1999 1992-1999 log(net employment) Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Firm Characteristics Age -0.0475 -0.667 0.0003 0.025 0.0044 0.352 Employees: Highly skilled 0.0143 0.426 0.0128 1.326 0.0088 0.909 Semi-skilled 0.0005 0.382 0.0020 * 2.002 0.0009 1.643 Low skilled 0.0043 ** 2.799 -0.0009 -0.826 0.0003 0.493 Maquila -0.9553 -0.667 -0.5789 -1.037 0.4879 1.100 Technology adoption 0.5631 * 1.824 -0.2403 -0.693 0.1190 0.388 Constant 2.7202 1.192 2.3452 ** 3.814 1.7872 ** 3.148 Number of obs. 364 320 271 R-sq: within 0.2809 0.1411 0.0884 between 0.0082 0.0477 0.0394 overall 0.0146 0.0564 0.0394 * Significant at 10% level; ** Significant at 5% level. 33 Annex 2: Wage Inequality Table A2.1. Wage Inequality Dependent variable: 1992-1995 1995-1999 1992-1999 Log(highly skilled/unskilled wages) Coeff. Z-St. Coeff. Z-St. Coeff. Z-St. Firm characteristics Size: Small 2.6615 1.595 -1.3684 * -1.718 -1.1810 -1.386 Medium 0.7362 0.537 -1.4635 * -1.826 -1.3873 -1.625 Large Dropped -1.5938 ** -1.984 -1.4417 * -1.689 Age 0.0313 ** 2.742 0.0045 1.464 -0.0019 -0.593 Employees: Highly skilled 0.0361 ** 12.726 0.0206 ** 8.614 0.0416 ** 14.628 Semi-skilled 0.0010 ** 5211 0.0016 ** 7.168 0.0005 ** 2.581 Low skilled -0.0022 ** -14.093 -0.0023 ** -11.471 -0.0027 ** -15.89 Maquila -0.1107 -0.705 -0.0382 -0.354 0.1591 1.498 Technology adoption 0.0275 0.518 0.1128 1.529 0.0534 0.819 Constant -2.7354 ** -3.260 0.5792 0.724 0.7860 0.922 Number of obs. 4,744 2,616 2,564 R-sq: Within 0.1785 0.2274 0.2480 Between 0.0905 0.2676 0.3063 Overall 0.0870 0.2604 0.2952 * Significant at 10% level; "*Significant at 5% level, 34 TECHNOLOGY AND SKILL DEMAND IN MEXICO Mexico - Technology, Wages, and Employment Technical Paper #2 Gladys L6pez-Acevedol 'This research was completed as part of the "Mexico - Technology Wages and Employment" study at the World Bank. We are grateful to the INEGI for providing us with the data. Joseph S. Shapiro and Erica Soler provided valuable research and editorial support. These are views of the author, and need not reflect those of the World Bank, its Executive Directors, or countries they represent. 35 1. Introduction In the mid-1980s, Mexico began a process of structural reforn that fundamentally changed the economic environment facing productive enterprises. Liberalization began in 1984 and accelerated when Mexico joined the General Agreement on Tariffs and Trade in 1986. In 1989, the government began radical policy reformns to reduce government regulation and liberalize trade. The adoption of the North American Free Trade Agreement with the U.S. and Canada in 1994 intensified liberalization. These reforms replaced quotas with smaller tariffs, eliminated price ceilings and floors, simplified trade regulation, eased foreign direct investment restriction, and privatized state-owned enterprises. Because of this international openness, technology now lies at the heart of Mexican economic activity. Globalization-induced competition has made firms increase the speed and efficiency of new technology adoption (TA). It has also inspired firms to increase research and development (R&D) budgets (OECD 2000). However, liberalization has worsened Mexican wage equality (World Bank 2000). The Gini coefficient, which measures income inequality and is especially sensitive to changes around the population median, rose from 0.47 in 1984 to 0.53 in 1998 (Table 1). While the poorest quintile lost 1.3 percent of its income during this period, the richest decile increased its wealth by 7.7 percent. In relative terms, all strata except the richest lost income during this period (Figure 1). Table 1. Inequality in Mexico, Measured by Gini Coefficient Year National Urban Rural 1984 0.47 0.44 0.45 1989 0.52 0.50 0.44 1992 0.53 0.50 0.43 1994 0.53 0.51 0.42 1996 0.52 0.49 0.45 1998 0.53 0.49i 0.48 Source: Author's calculations based on ENIGH. 36 Figure 1. Wealth Distribution among Mexican Population 45 X _ 40 - X 30 -|-- X Bottom20% < 25 - }+1anwer40% 25 | Middle 30% o 20 - - -- -4--Top 10% 5 -4------ 0- 1984 1989 1992 1994 1996 1998 Source: Autior's calculations based on ENIGH. At the height of liberalization, Mexico underwent a severe currency crisis. In December 1994, Mexico sharply devalued the peso causing a deep economic recession. In 1995, real gross domestic product (GDP) fell by 6.2 percent, demand fell by 14.4 percent, and investment fell by 43 percent. This market contraction shifted trade as exports increased by almost 31 percent and imports fell by 9 percent. Nevertheless, a cheap peso bolstered exports and offered new markets to firms whose domestic sales had collapsed. Domestic demand quickly recovered, and by 1997 real GDP had returned to its pre-crisis level. This paper investigates the skill-biased technological change (SBTC) hypothesis for Mexico, using panel data from the National Survey of Employment, Salaries, Technology, and Training (ENESTYC) and the Annual Industry Survey (EIA). The panel has observations for 1992, 1995, and 1999 (for a description of these surveys and the panel see Appendix A and B, and for a list of variables Appendix C). Section 2 reviews Mexico-specific circumstances. Section 3 asks whether technological change is biased towards particular skill groups. Section 4 analyzes the relationship of skilled labor employment to the use of technology. Section 5 discusses productivity gains from TA and training. Section 6 presents conclusions. 37 2. Literature Review A key implication of technology diffusion is its impact on the labor market. Several studies (Davis and Haltiwanger 1991; Krueger 1993, and Mincer 1991) find that technology has raised the relative demand for more skilled workers and, consequently, reduced the demand for manual labor. Katz and Murphy (1992) use a basic supply and demand approach to show that relative labor demand shifts come from intra-industry changes (such as factor non-neutral technological change, changes in prices of non-labor inputs, and outsourcing) and inter-industry changes (such as shifts in market-wide product demand, sector differences in factor-neutral total factor productivity (TFP) growth, and trade shifts that change the domestic share of output at fixed relative wages). Aw and Batra (1999) also provide evidence that technology (measured by R&D and by worker training) has an impact on firm performance (measured by wages). This finding coheres with the World Bank (1999) "Mexican Labor Markets: New Views on Integration and Flexibility," which relates wages to technology (measured in terms of R&D and technology acquisition). In the last decade, Mexican wage inequality has increased sharply. Three theories - the liberalization hypothesis, the labor institution hypothesis, and the SBTC hypothesis - can explain this increase in earnings inequality in Mexico.2 The liberalization hypothesis argues that reduced trade barriers put an economy under competitive pressure to specialize along its lines of comparative advantage. A developed country with high levels of human capital will shift to advanced industry and services as low-skilled industries feel competition from countries with abundant low-wage workers. In a test of liberalization theory, Hanson and Harrison (1995) examine the impact of Mexican trade reform on wage structure using firm-level data. They test whether trade reform shifted employment toward skilled labor-intensive industries (the Stolper-Samuelson-Type [SST] effect). They conclude that changes within industries and firms caused the wage gap, and 2 See, for example, the "Symposium on Wage Inequality" (1997) and the "Symposium on How International Exchange, Technology and Institutions Affect Workers" (1997). 38 that the SST effect provides a poor explanation. Thus the increase in wage inequality was due to other factors. Burfisher and others (1993) also examine the SST effect under NAFTA. However, the liberalization hypothesis seems ill-fitted to the Mexican experience. Mexico began lowering trade barriers after 1984, particularly for American and Canadian commerce, whose share of merchandise imports from Mexico increased from 68 percent in 1985 to 78 percent in 1996. Since Mexico has an abundant supply of low skilled labor relative to its northern neighbors, the liberalization hypothesis predicts that lowered trade barriers would raise both the wages and the demand for unskilled labor in Mexico. However, as trade barriers fell, low-skilled labor in Mexico experienced a real wage decrease while skilled labor had a real wage increase, resulting in worsened wage inequality. The trade theory can be adjusted to fit Mexico better: while trade liberalization pushes a country towards its comparative advantage, it also facilitates the transfer of ideas and technology. This seems to provide a better explanation of earnings inequality. Feenstra and Hanson (1996) put forward a variant of the liberalization hypothesis that involves outsourcing whereby multinational enterprises in the developed country relocate their less skill-intensive activities to the less skill-abundant developed countries. However, what is referred to as a low- skill activity in the United States may be a high-skill activity in Mexico, which could explain the similarity in earnings inequality between the two countries. The institutional hypothesis considers broad changes such as minimum wage reductions and weakened trade unions. A binding minimum wage truncates the lower end of the wage distribution. As the real minimum wage decreases due to inflation, its effect on industry also decreases. This translates into increased wage and earnings dispersion. Since the early 1980s, institutional developments have exerted insignificant influence on earnings distribution. The distribution of real wages, for example, reveals little distortion due to the minimum wage, which suggests that the minimum wage is low enough to affect industry minimally. Similarly, union wages in Mexico are similar to nonunion wages, controlling for education differences. This data suggests that unions have little or no impact on earnings. Although we cannot entirely reject the 39 institutional hypothesis, it does not appear to be the principal explanation for Mexican earnings inequality. The SBTC hypothesis links earnings inequality to technology that raises relative demand for skilled labor. Cragg and Epelbaum (1996) find that, given different labor supply elasticities, the primary source of Mexican wage inequality is a skill-biased demand shift rather than skill- uniform demand growth. Meza (1999) also investigates the Mexican labor market, and finds that intra-industry demand shifts toward a more educated labor force explain unskilled labor wage decreases better than interindustry demand shifts. The World Bank (2000) also shows that demand increases for a more educated labor force within the economic sectors explain the increase in their premium when compared to the demand shifts for less educated workers between economic sectors. Acosta and Montes (2001) show that there is a constant increase on skill premia during the 1987-1993 period, but there is a deceleration during the 1994-99 period and a decline after 1997. These authors contend that the former may be caused by between industry differences. Batra and Tan (1997) study the SBTC hypothesis as a plausible explanation of wage inequality using firm-level data for Colombia, Mexico, and Taiwan (China). They find that technology's impact on wages is greatest for skilled workers and lowest for unskilled workers, thereby supporting the SBTC hypothesis. In this hypothesis, skilled workers become more efficient in jobs traditionally performed by unskilled workers (Johnson 1997). 3. Technology's influence on wages and employment To determine whether the technology of Mexican firms influences their skill demand, we estimate determinants of each skill group's share of total wages. In this fixed-effects model, we use a modified first-differencing to eliminate firm-specific error. For a cost minimizing firm, the following equation models technology's influence on wages: Wj,= ,iA ln VA,, + ,l2 lnK,, + 63 In Ri, +34 In TFP,, + /5 ln UR, + gi, (1) where: 40 W, = - W,, or the wages of firm i at time t minus the average wages of firm i over all time periods; Wii = the wage share of skill-groupj in the total wages of firm i; VA = value-added (calculated with INEGI's methodology, i.e. the difference between the value of the production of the firm and its expenditure in materials, water, energy and electricity) in real 1997 pesos; K = capital assets not deflated;3 R = the relative wage of skilled workers in relation to unskilled workers; TEP = total factor productivity, a measure of technological change; UR = the unemployment rate, included as a control for macroeconomic shocks, and £ = the normal regression error. A positive A3i parameter in equation (1) indicates that growing industries are more likely to increase the wages of skill groupj. A positive A2 parameter indicates that capital and skills are complementary inputs in the production process; a negative parameter indicates that capital and skills are substitutes. The /33 parameter indicates how changes in relative wages affect the wages of skill group s. We expect this parameter to be negative, since a rise in the relative wages of one skill group leads a cost-minimizing employer to substitute towards other groups. The 84 parameter of the TFP shows the extent to which technological change is skill-biased. If TFP is skill-neutral, it should not impact skill mix. A positive 34 parameter implies that technological change is skill-using or skill-biased, while a negative 84 indicates that technology is skill- replacing. Finally, /5 will indicate us how the employment of the different skill groups varies as unemployment changes. We estimate TFP as the residual in a production function. The specification of the Cobb- Douglas production function is: ln(VA)11 = ,8iln(K)11 + fi21n(L)i, + ei' (2) where: VA = value-added, in real 1997 Mexican pesos; K = fixed capital assets, not deflated; L = labor inputs, total hours worked per annum; Eit t = the normal regression error; i = the individual plant being analyzed, and No deflator is available for fixed assets, due to this we assume that the flow of capital services is adequately reflected in its book value. 41 t = the time period. We estimate equation (1) for 1992-99 manufacturing firms panel using a fixed-effects model. Tables 2 and 3 give separate results for wages and employment shares of skill groups, so we can use both to measure skill demand. Table 2. Wage Determinants, by Employee Skill Group Dependent variable: Highly skilled Semi-skilled Low skilled Share of wages Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. Value-added -0.0647 ** -4.161 0.0281 ** 1.966 0.0365 ** 3.241 Capital assets 0.0479 ** 5.794 -0.0401 ** -5.270 -0.0078 -1.301 Wage ratio s/us 0.0082 ** 2.061 0.1204 ** 32.757 -0.1286 ** -44.456 TFP 0.0232 ** 3.864 -0.0148 ** -2.681 -0.0084 * -1.926 Unemployment rate -0.0225 ** -2.636 0.0146 * 1.850 0.0080 1.285 Overall R2 0.0315 0.4550 0.5508 * Significant at 10% level; ** Significant at 5% level. Notes: 1. Number of observations= 1,185. 2. Regressions included indicator variables for 9 manufacturing divisions. 3. Share of wages is the share of the firn's total wages paid to workers of a parficular skill level. Table 3. Employment De terminants, by Employee Skill Group Dependent variable: Highly Skilled Semi-skilled Low skilled Share of Employment Coeff. t-Stat. Coeff. t-Stat. Coeff. t-Stat. All EmDlovees Value-added -0.8129 ** -2.487 1.1669 0.687 -0.3552 -0.210 Capital assets 0.7198 ** 4.140 -1.1942 -1.321 0.4785 0.531 Wage ratio s/us 0.0704 0.837 14.9349 ** 34.190 -15.0057 ** -34.455 TFP 0.3072 ** 2.428 -0.5728 -0.871 0.2653 0.404 Unemployment rate -0.0458 -0.255 3.3348 ** 3.566 -3.2807 ** -3.518 Overall R2 0.0037 0.5869 0.5887 Male Employees Value-added -0.0076 * -1.895 0.0412 * 1.735 -0.0210 -0.825 Capital assets 0.0077 ** 2.826 -0.0277 * -1.714 0.0088 0.510 Wage ratio s/us 0.0007 0.671 0.1335 ** 20.184 -0.1392 ** -19.679 TFP 0.0042 ** 2.946 -0.0216 ** -2.527 0.0105 1.147 Unemployment rate 0.0008 0.084 0.0447 0.819 -0.0713 -1.221 Overall R2 0.0081 0.4319 0.3658 Female Emplovees Value-added -0.0013 -1.193 -0.0340 ** -2.045 0.0226 1.499 Capital assets 0.0011 1.541 0.0252 ** 2.234 -0.0151 -1.480 Wage ratio s/us 0.0003 1.051 0.0275 ** 5.945 -0.0228 ** -5.435 TFP 0.0007 * 1.831 0.0094 1.565 -0.0032 -0.588 Unemployment rate -0.0022 -0.900 0.0341 0.894 -0.0061 -0.175 Overall R2 0.0145 0.0057 0.0353 * Significant at I00/o level; ** Significant at 5% level. Notes: I. Number of observations = 1,185. 2. Regressions included indicator variables for 9 manufacturing divisions. 3. Share of employrent is the share of the firm's total employment held by workers of a particular skill level. The effects of output growth vary between skill groups. It appears that growing firms are less likely to hire highly skilled male employees. In contrast, for the semi-skilled group and the 42 low skilled group, the value-added parameters are positive and significant in the wage share estimation and insignificant in the employment share specification for all workers. With reference to gender, the value-added parameter is positive for the semi-skilled male workers and negative for the semi-skilled female workers, this suggests that growing firms tend to increase male rather than female production workers. It is interesting to note that the unemployment control is positive for the semi-skilled group and negative or neutral for the highly skilled and low-skilled groups. This suggests that employers increase the relative share of production workers as overall unemployment rises. We find that capital and highly skilled workers are complements while capital and semi- skilled workers are substitutes. This result is consistent with Tan (2000). Also, capital and semi- skilled male production workers are substitutes while capital and semi-skilled female production workers are complements. Finally, technology is biased toward the use of highly skilled workers. This implies that technology is skill-using for highly skilled workers but skill-replacing for semi-skilled and unskilled workers. Furthermore, technical change is skill-using for male and female highly skilled workers and skill-replacing for the semi-skilled male production workers. Tan (2000) finds that, in Malaysia, technical change is skill-using only for male highly skilled workers. 4. Time of benefit from technology Probit models of technology adoption are informative but ill-suited to dynamic processes (L6pez-Acevedo 2001). To analyze the temporal relationship of labor demand and TA, we use event analysis. As Tan (2000) suggests, comparing the period of TA to skill-mix changes in previous and consecutive time periods facilitates effective event analysis. The variable r represents the time period relative to TA; the period of adoption is T= 0 , the period preceding adoption is T= -1, and the period following adoption is r= 1. Since companies adopt technology at different times, the period in which T= 0 differs between firms. Also, we define that r = 0 in all periods for a firm that never adopts technology. Since we only have point data for 1992, 1995, and 1999, rranges from -2 to 2. Using information on rfrom all firms that adopt technology, we 43 estimate the r-profile of skill shares, or the relationship of when a firm hires skilled labor to when it adopts technology. We measure the T-profile relative to a= 0 to allow comparison with firms that adopted technology in the same period, and with firms that never adopted technology. We want to know how employers vary skill-mix in the years preceding and following TA. A regression model relating skill shares to rmay be written as follows: gut= +o 8,Xt + fl2 3 Zyt + A3T, + f41nUR + Et (3) where: SOt = the share of workers of skill groupj in firm i during period t; X = a vector of firm and industry characteristics; Z = a vector of dummy variables for each rbetween -2 and +2; T, = a time trend term, and UR = the unemployment rate, included as a control for macroeconomic changes. Here, the f2 coefficients trace the r-profile of skill share j relative to z- = 0. We use models of similar specification to examine how productivity and wages vary with T. Table 4 reports the estimated coefficients of r; in relation to r = 0. These z-profiles appear graphically in Figures 2 through 8. We fit these graphs with cubic spline to show the underlying trends in these variables over r more effectively. In essence, they relate firm hiring patterns to TA timing. Table 4. Estimated Coefficients of rfor Skill S hares, Producti vity and Wages of TA Firms Skill Shares Waees by Skill GrouD r Highly skilled semiskilled Unskilled Productivity Highly skilled semi-skilled Unskilled -2 -0.7193 * -2.8499 3.8857 -1.0721 ** -1.3990 ** -1.1136** -0.9067** -1 -0.5144 * -7.6369** 9.0491** -0.5273 ** -0.7277 ** -0.7224** -0.3480** 0 1 0.2992 * 3.2059** -3.0640** 0.5745 ** 0.7215 ** 0.6064** 0.5846** 2 0.5262 ** 3.0331** -4.5258** 1.0297 ** 1.4301 ** 1.1697** 1.1401** * Significant at 10% level; ** Significant at 5% level. Notes: Regressions included dunmy variables for 9 mnufacturing divisions, indicators for small, medium and large firms, dummy variables for foreign firms and joint-ventures, and a control for macroeconomic changes. 44 Figure 2-4. Skill Shares Pre- and Post-TA Relative to Non-TA Firms 15,9368 n I I I I 1:18489- -2 l 2 bm- SHARE OF HIGHLY SKILLED WORKERS AND ADOPTION 71 09672 -2 2 time SHARE OF SKILLED AND SEMI-SKILLED WORKERS AND ADOPTION 85,9857- V8 I I 20.3028 -2 1 1 2 time SHARE OF UNSKILLED WORKERS AND ADOPTION 45 Figure 2 and Table 4 show that prior to TA, the share of highly skilled workers changes little relative to the time of TA. The share of highly skilled workers among total firm employment is fairly small. Although we observe an increase in the number of highly skilled workers hired after TA, the increase is also small relative to the total number of workers in the firm. Figure 3 shows that the share of skilled and semi-skilled production workers increases before and after TA. By the second period after TA, the share of skilled and semi-skilled workers has risen considerably. Figure 4 shows that the share of less-skilled workers is actually larger before TA, but falls post- adoption. The v-profile for productivity appears in Figure 5. Productivity of technology firms is relatively low before TA. But after TA, productivity of technology firms increases. Although the y-axis of the graph represents the log of productivity, the scale is linear and not logarithmic; hence, the productivity of technology firms increases both before and after TA. However, one period after TA, the rate of productivity growth jumps. Figure 5. Productivity Pre- and Post-TA 6.32951- 2.41272 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~I I -2 2 time PRODUCTIVITY AND ADOPTION Technology increases firm productivity and consequently firm profits. But how do worker wage changes relate to firm profit increases? To address this question, we estimate separate equations for each skill group, with the dependent variable being the logarithm of wages of the skill group. The resulting r-profiles of three skill groups - highly skilled, semi-skilled, and low skilled workers - appear in Figures 6-8. 46 Figure 6-8. Wage Premiums by Skill Group Pre- and Post-TA 6.90231i E 2 1.18366-1 -2 ~~~~~~~~~~~~~~~~~~~~~2 HIGHLY SKILLED WORKERS WAGES AND ADOPTION 7.66017 IZ [ Z 3.50662 -2 2 tirn. SKILLED AND SEMI-SKILLED WORKERS WAGES AND ADOPTION time UNSKILLED WORKERS WAGES AND ADOPTION 47 Wages show similar trends to productivity, as it appears that employers share productivity gains from technology with all skills of employees. However, skilled employees receive significantly more benefit than unskilled employees. Wages of all skills of employees begin rising two periods before TA, and continue rising until two periods after TA. Furthermore, the growth rates of the wages of all skill groups jump after TA. 5. Technology Adoption and Productivity Growth As Tan (2000) argues, a production function including variables for technology type and experience with technology is the best approach for investigating formally the productivity gains from introduction of new technology. The production function also lets us test whether the benefits of experience with technology increase when combined with employee training: ln(VA),, = /,3 ln(K,, ) + ,B2 ln(L,, ) +/3 !3X,TAi, + E, + A45Z, + g,, (4) where: VA = VAi, - VAi, or the difference of VA at time t and the average of VA at all times for firm i. VA = value-added (calculated with INEGI's methodology, i.e. the difference between the value of the production of the firm and its expenditure in materials, water, energy and electricity) in real 1997 pesos; K = capital assets (not deflated);4 L = labor input; EjTA = a vector of interactive indicator variables for thej types of technology adopted;5 Er = a vector of indicator variables for technology experience; Z = a vector of firm and industry characteristics; i = firm under analysis; t = time period; e = normal regression error. To find the value-added of training when new technology is adopted, we include interaction terms for technology type and training. Robust results in Table 5 show that 4 No deflator is available for fixed assets, due to this we assume that the flow of capital services is adequately reflected in its book value. 5 This is a dummy variable that gets assigned a "I" if the firm adopts new technology and provides training, an a "O" otherwise. 48 combining training with TA increases productivity in three of the four categories of technology. On average, when training increases technology's productivity the benefit is more than six percent. These results indicate the importance of complementary investments in worker training to realize the productivity potential of technology. Training has the greatest effect when a firm adopts machinery tools, and the least effect when a firm adopts robots. Table 5. The value-added of training with technology Dependent variable: 1992-1999 Log(value-added) Coeff. t-Stat. Production Function Constant 9.1594 ** 27.833 Log(capital) 0.0948 ** 3.843 Log(total hours worked) 0.1378 ** 5.224 Types of TA and Training Automatic equipment 0.0635 * 1.742 Machinery tools 0.0789 ** 2.593 Computerized machinery 0.0671 * 1.684 Robots -0.0041 0.976 TA Experience 1 -0.1029 ** -3.864 2 ~~~~~~~ ~ ~~~~-0.0081 0.817 ...........~.... .......... .I.. .... ^ ........................ . ............. ..... ..... . ... -^. , ... ......-............. Number of ohs. 2,733 Overall R2 0.4143 * Significant at 10% level; ** Significant at 5% level. Note: Regressions included dummy variables for 9 manufacturing divisions, foreign firms and joint-ventures. 6. Conclusions This paper attempts to understand how TA affects demand for highly skilled, skilled and semi-skilled, and unskilled workers. Both in terms of wages and employment, with TA firms tend to hire more high skilled workers and pay them better. We also analyze the relative times of TA and demand shifts for a given firm. We find that demand for highly skilled workers increases after the adoption of technology, but not before. By the second period after TA, demand for highly skilled, skilled, and semi-skilled workers has risen notably, while the share of employees made up of unskilled workers tends to diminish after the adoption of technology. Additionally, training and increases in human capital magnify technology-driven productivity gains. 49 References Acosta, P. and G. Montes. 2001. "Trade Reform, Technical Change, and Inequality: The Case of Mexico and Argentina in the 1990s." Processed. Aw, B. Y. and G. Batra. 1999. "Wages, Firm Sizt' and Wage Inequality: How Much do Exports Matter?." In D. B. Audretsch and R. Thurik, ed3., Innovation, Industry Evolution and Employment. Cambridge: Cambridge University Press :13 -56. Batra, G. and H. Tan. 1997. "Technology and Firm Size-Wage Differentials in Colombia, Mexico, and Taiwan (China)." World Bank Economic Review 11(1) :59-83. Burfisher, M., Robinson, Sherman and Thierfelcier. 1993. "Wage Changes in a US-Mexico Free Trade Area: Migration Versus Stolper-Samuelson Effects." Working Paper 645. University of California, Berkeley, Department of Agricultural and Resources Economics. Cragg, M. I. and M. Epelbaum. 1996. "Why Has Wage Dispersion Grown in Mexico? Is It the Incidence of Reforms or the Growing Demand for Skills?." Journal of Development Economics 51(1) :99-116. Davis, S. J. and J. Haltiwanger. 1991. "Wage Dispersion between and within U.S. Manufacturing Plants, 1963-86." Brookings Papers on Economic A.,tivity: Microeconomics :115-80. Feenstra R. and G. Hanson. 1996. "Globalization, Outsourcing, and Wage Inequality." American Economic Review 86(2) :240-5. Hanson, G. and A. Harrison. 1995. "Trade, Technology, and Wage Inequality." Working Paper 5110. National Bureau of Economic Research. Johnson, G. 1997. "Changes in Earnings Inequality: The Role of Demand Shifts." Journal of Economic Perspectives 11(2) :41-54. Katz, L. and K. Murphy. 1992. "Changes in Relative Wages, 1963-1987: Supply and Demand Factors." Quarterly Journal of Economics 107(1) :35-'8. Krueger, A. B. 1993. "How Computers Have Changed the Wage Structure: Evidence from Microdata, 1984-1989." Quarterly Journal of Economic., 108(1) :33-60. L6pez-Acevedo, G. 2001. "Absorptive Capacity and Technological Adoption: Panel Evidence from Mexican Manufacturing." Paper presented at the Economists Forum 2001. World Bank. Meza, L. 1999. "Cambios en la estructura salarial en Mexico en el periodo 1988-1993 y el aumento en el rendimiento de la educaci6n superior." Trimestre Econ6mico 49(196) :813-60. Mincer, J. 1991. "Human Capital, Technology, arld the Wage Structure: What Do Time Series Show?." Working Paper 3581. National Bureau of Economic Research. Tan, H. 2000. "Technological Change and Skills Demand: Panel Evidence from Malaysian Manufacturing." Working Paper. The World Bank Institute. 50 World Bank. 1999. "Mexican Labor Markets: New Views on Integration and Flexibility." Volume Two: Technical Papers. Poverty Reduction and Economic Management Unit, Mexico Department. ---------. 2000. "Earnings Inequality after Mexico's Economic and Educational Reforms." Report No. 19945-ME (Gray Cover). May. Mexico Department. 51 DETERMINANTS OF TECHNOLOGY ADOPTION IN MEXICO Mexico - Technology, Wages, and Employment Technical Paper #3 Gladys 1,6pez-Acevedol This research was completed as part of the "Mexico - Technology Wages and Employment" study at the World Bank. We are grateful to the INEGI for providing us with the data. Joseph S. Shapiro and Erica Soler provided valuable research and editorial support. These are views of the author, and need not reflect those of the World Bank, its Executive Directors, or countries they represent. 52 1. Introduction The creation of new knowledge will not necessarily be at the core of Mexico's technological evolution, but rather the adoption and application of existing technological knowledge developed abroad (Bell and Pavitt 1992). In turn, the country's ability to adopt and apply technological knowledge will be crucially shaped by its absorptive capacity, defined as its capability to learn and adopt knowledge developed abroad (Audretsch 1995; Cohen and Levinthal 1989). Moreover, a threshold level of knowledge and technological achievement are prerequisites to developing significant absorptive capacity (Dosi, Pavitt, and Soete 1990; OECD 1997). Technology diffuses through many channels. Most technology research in Mexico takes place on university campuses and receives government guidance (OECD 2000). Therefore, it is important to measure the impact of government initiatives on new technology adoption (TA) by finns. This paper seeks to identify the impact that TA policies, firm- and industry-specific factors have on TA by Mexican firms. In this effort, we offer two innovations. First, we present detailed analysis of the heterogeneity involved in TA. Most work uses limited measures of technology, such as research and development (R&D). Our rich data not only allows us to examine the effect of R&D spending, but also to look at different types of technology. Second, we measure the effect of policy measures designed to promote TA. By combining location-specific data of firm characteristics with data on government policies and regional structure, we understand better how exogenously determined factors affect TA. We present this paper in six parts, beginning with the Introduction. Section 2 reviews literature on determinants of TA; Section 3 describes TA patterns in Mexico; Section 4 presents cross-sectional results; Section 5 presents panel data results, and Section 6 offers conclusions. 53 2 Determinants of Technology Adoption Technology diffusion occurs when a user adopts technology that the user did not invent. Whenever a firm uses a technology developed by another company, the firm experiences the benefits of technology diffusion. In Mexico, where private R&D investment is very low compared to other countries and new technologies are costly, effective technology diffusion is crucial. In essence, technology diffusion is just a special case of TA. In an analysis of Mexico, Blomstrom, Kokko, and Zejan (1992) investigate what policy measures encourage multinational corporations (MNCs) to bring technologies into countries outside their headquarters. Since much technology used in Mexico is developed outside of Mexico, such encouragement can be crucial. They find that local competition most encourages imports of technology by MNC subsidiaries. Thus, a way to maximize the inflow of modem technology is to create a competitive environment in which firms must operate. Foreign direct investment (FDI) is the major technology diffusion channel in developing countries like Mexico (OECD 2000). Since MNCs undertake much of international R&D, they also determine international technology flows. Most developing countries have limited R&D budgets and heavily depend on foreign-developed technology. It may be that MNC subsidiaries use advanced technologies in international markets and provide technical training to local firms through FDI and worker training, thereby adding to the country's human capital base and increasing technology through FDI. Others doubt that the technology of MNCs benefits local producers (Blomstrom and Kokko 1998). Some authors argue that MNCs keep technology within their own control, so geographic technology transfer occurs but technology diffusion to other companies does not (Cantwell 1989; Haddad and Harrison 1993; Aitken and Harrison 1992). Blomstrom (1989) suggests that technology transfers to domestic firms come from spillovers rather than from formal transactions (Caves 1974; Globerman 1979; Blomstrom and Persson 1983; Blomstrom and Wolff 1994). 54 Besides investigating what kind of environment encourages TA, research also examines what firn-type is most likely to adopt technology. Firm size is a widely-recognized determinant of TA. Larger firms tend to support the high costs of new technology, and may find a broader range of technologies that meet their needs. The larger a firm's size, the more technology it adopts. Several studies have confirmed this finding (Mansfield 1961; Davies 1979; Romeo 1975; Globerman 1975), and it may be the most robust determinant of TA. Trade liberalization is a major tool of increasing competition, and in support of Blomstrom and others (1992), several studies correlate liberalization with technology diffusion (Grossman and Helpman 1991, Romer 1994, and Young 1991, find that trade liberalization contributes to economic growth through productivity growth). Liberalization increases the variety of intermediate inputs to manufacturing, facilitates knowledge-transfers, amplifies learning-by-doing effects, and increases the size of consumer markets. These changes encourage diversification within a firm and, correspondingly, TA. Romer (1994) argues that trade distortions may effect productive efficiency by preventing the implementation of new technology, and by limiting the incentive to develop new products. Iscan (1998) finds that after liberalization, total factor productivity (TFP) in Mexican manufacturing increased in conjunction with firm exports. Firms that export might face more competition abroad and so feel pressure to adopt technology. A recent World Bank and INEGI study (2000) tested the direction of causality between exporting and enterprise performance. The study attempted to relate exports and inter- firm linkages to TA and technical training, but found that the relationship was statistically insignificant. Another determinant of TA is the availability of appropriately complex technology. Like other variables, its effect is unclear from a theoretical perspective. Some authors argue that the advanced technology used by many MNCs is too complex to improve basic manufacturing in developing countries (Lapan and Bardhan 1973; Cantwell 1989; Haddad and Harrison 1993). Others argue that some technical gap between the host country and the MNC is necessary for the host country to receive any benefit, so spillovers grow proportionally with the technology gap (Blomstrom and Wang 1992; Blomstrom and Wolff 1994). Torres (2000) provides an interesting result using state-level basic factor analysis. The author finds that science ability, represented by variables like public expenditure on R&D, number of universities, number of published articles, and number of researchers, significantly influences technology diffusion. In sum, the literature finds that variables such as FDI, local competition, complexity of available technology, trade liberalization, foreign capital in a firm, and regional science ability influence a firm's use of technology. To paint a full picture of technology diffusion, we include a broad array of variables representing all of these factors. 3. Patterns of Technology Adoption: Descriptive Statistics In this section we discuss what kinds of firms most frequently adopt new technology. We use data from the National Survey of Employment, Salaries, Technology, and Training (ENESTYC) for 1992 and 1999. Since this section uses simple statistical averages rather than regressions controlling for relevant variables, the reader should not interpret cause (size encourages adoption) from correlation (larger firms adopt more). Studies on technology diffusion in North America and Europe identify firm size as a significant determinant of TA; Figure 1 shows similar results for Mexico. In 1999, while more than 40 percent of all Mexican manufacturing firms adopted some type of new technology, the exact share increases in conjunction with firm size. While only 38 percent of microenterprises adopted new technology, 78 percent of small enterprises, 87 percent of medium-size enterprises, and 93 percent of large enterprises adopted new technology in 1999. The relationship between TA and firm size did not change between 1992 and 1999. But TA was considerably lower in 1992 for all firrn sizes than it was in 1999. Prior to trade liberalization, less than 30 percent of firms had engaged in TA-microenterprises had a TA rate 56 of just under 25 percent, small enterprises had a TA rate of 53 percent, medium-size enterprises had a TA rate of 71 percent, and large enterprises had a TA rate of 80 percent. Figure 1. TA by Firm Size, 1992 versus 1999 100 o 9F 0 _ _ _ ____199_1 goI Ta0 l 1_ _ shw htTAi199vaiscnidrbybtwe9ifrettpso 40 - 830 750 610 20 percent for manual equipment to 0.8 percent for computerized numeric controlled machinery (CNCM). Large firms are the most likely to adopt robots. However, adoption rates for numeric controlled machinery (NCM) and CNCM vary considerably across firm size. Only 0.6 percent of microenterprises had adopted numeric controlled or computerized technology. However, 10.4 percent of small enterprises, 16.6 percent of medium-size enterprises, and 21 percent of large enterprises adopted this kind of technology. Thus, larger firms seem to adopt more complex technology than their smaller counterparts. Table 1. TA by Technology Type and Firm Size, 1999 Percent of firms that adopt technology Type of technology . Micro Small Medium Large All Manual equipment 20.0 18.2 12.8 6.5 19.7 Automatic equipment 10.6 22.4 18.7 16.4 11.5 Machinery tools 7.2 26.7 38.6 47.3 9.3 NCM 0.5 4.4 4.9 5.5 0.9 CNCM 0.1 6.0 11.7 15.5 0.8 Robots 0.0 0.0 0.7 1.5 0.0 Source: Author's calculations based on ENESTYC 99. 2 Annex I shows descriptive statistics for 1992. 57 The propensity of TA for firms engaged in training their workers also increases in conjunction with firm size. While TA rates and reliance on worker training positively correlate, the links between them vary considerably across firm size. Figure 2 shows that only 36 percent of microenterprises without training adopted technology, but 67 percent of the microenterprises that did provide training adopted technology. This difference in TA by training condition is less noticeable for large firms, where 88 percent of the firms without training adopted technology versus 93 percent of the firms that train workers. Figure 3 shows that export-oriented firms have higher TA rates (83 percent) than non- export oriented firms (41 percent). This rate varies by firm size, but as in the training case, the difference in TA between export-oriented firms and other firms is much higher for microenterprises than for other firm sizes. Figure 2. TA by Training and Firm Size, 1999 Figure 3. TA by Export Status and Firm Size, 1999 > ! S! L t ]1 = ~~~~~ ~ ~ ~ ~ ~~~~~~~90 = l Micro Small Medium Large All Micro Small Medium Larg All | ClFjnes that raOjn wahrksr * Oswer fine3] | Mport onenrco firnn s *0 erirtnj Source: Author's calculations based on ENESTYC 99. Source: Author's calculations based on ENESTYC 99. Figure 4 shows that TA rates vary between different manufacturing industries. TA is relatively high in basic metal industries (81 percent), chemical products, oil derivatives, and coal (67 percent), and metal products, machinery and equipment (61 percent). Just as TA varies across industries, it also varies for each firm size within an industry (Table 2). 58 Figure 4. TA by Industry, 1999 90- ~70 50 ~20 - 0- M PlD £ LIM 1K .0 0- a -a - a a- SouJrce: Author's calculations based on ENESTYC 99. Table 2. TA by Industry and Firm Size, 1999 Percent of firms that adopt technology Division ~Micro Small Medium Large All Fo, beverage, tobacco 32.1 66.2 86.3 91.6 33.9 Textiles, clothing, leather 26.2 84.6 86.1 92.4 34.2 Wood, wood products 42.5 78.5 77.4 85.9 44.3 Paper, paper products 47.6 81.9 79.1 89.7 51.8 Chemical products, oil derivatives, coal 55.6 77.6 91.8 94.6 66.6 Non-metallic minerals 27.0 65.5 86.3 88.0 28.8 Basic metal industries 61.2 89.8 87.6 98.0 81.1 Metal products, machinery, equipment 58.4 83.7 91.4 94.0 61.5 Other manufacturing industries 43.6 62.4 89.1 97.1 46.2 Source: Author's calculations based on ENESTYC 99. TA also varies within regions, with firmns located in the North having the highest adoption rates. As Figure 5 shows, in 1999, 53 percent of Northern firms adopted technology, 43 percent of firms in the Capital adopted technology, 41 percent of firmns in the Center region adopted technology, and only 32 percent of Southern firms adopted technology. We also observe that TA increased within each. region between 1992 and 1999. Figure 6 shows the adoption of different technology types by region. We observe that more than 25 percent of firms in the Central region adopt more complex technology (NCM, 59 CNCM, and robots). In the South only 15 percent of firms adopt more complex technology. The most prevalent technology in all regions is machinery tools. Figure 5. TA by Region, 1992 versus 1999 60 1 60 40 3S40 - 1992 *1999 20 lo - 0 U.. North Center South Capital Source: Author's calculations based on ENESTYC 92 and 99. Figure 6. TA by Region and Technology Type, 1992-1999 North Center XW~j~6 .p At6.1i, 6qwp.ff o 141%hlo 320 66o6b6*cloyt. I O ~ ~ ~ ~ ~ ~~~.h~yol M..'aI6q.n CNCU M6t . qi,,6 -1 534% 36.9% t s 8 6.3% "'41 06%~~~~~~~~~~~~~~~~3 CaPital South 515% ~~~~~~~~~~~~~~~64% M 00.2 .qW6 Inel.n 6649.0 Z.s -w- cMwi mmoslzsM cmaye1 Netechlol4gy ~ ~ ~ ~ ~ ~ ~ lo 41 R6_% __ 2 Nn.,I..I. 336 153%. 51 0*%M Source: Author's calculations based on the ENESTYC 1992-99 panel. 60 4. Technological Adoption: The Cross-Sectional Approach 4. 1. Methodology As Kokko (1994) notes, empirical approaches to technology transfer generally model labor productivity of local firms as a function of the market share of multinational subsidiaries and of the other variables described in section 2 (Caves 1974; Globerman 1979; Blomstrom and Persson 1983). If evidence shows that the presence of multinational corporations (MNCs) increased the labor productivity of local firms, a study concludes that spillovers took place. Grether (1999) uses a different production function to examine Mexico.3 The author first estimates a firm's multifactor productivity. She uses this measure rather than labor productivity because multifactor productivity is not biased by changes in the input mix. In her view, an increase in multifactor productivity may reflect the acquisition of an additional input, so smaller efficiency dispersion suggests superior technology diffusion. She then computes a multivariate regression using plant-specific, sector-specific, and location-specific variables as possible determinants of technology diffusion. Since the ENESTYC has information on the different types of technology that firms adopt, we need not assume (as we would have to in a production function approach) that a new technology input increases productivity. This section estimates TA likelihood using cross-sectional logits for 1992 and 1999. The advantage of the cross-sectional models is that they enable comparison between TA changed in the early 1990s to TA later in the decade. Since Mexico experienced exogenous shocks during our sample period from the North American Free Trade Agreement (NAFTA) and the 1994 crisis, a sudden structural change took place during our survey. Therefore we expect that patterns from the first part of the 1990s might not continue through the rest of the decade. 61 In order to understand this change thoroughly, we use three categories of absorptive capacity - firm-, industry-, and region-specific factors. We measure firm-specific factors by including dummy variables for firm sizes, firm age, labor skill level, R&D spending, foreign ownership of the firm, subsidiary firm, export tendency, formal training, maquila status, and union status. We measure firm age in years after startup. Literature provides strong evidence that TA likelihood positively relates to firm age. We know considerably less about the impact of foreign ownership on the likelihood of TA. The absorptive capacity may be greater and the cost of adopting technology lower if a firm has strong links to foreign enterprises. This would suggest a positive relationship, thus the share of capital in the firm held by foreign owners positively correlates with TA. A skilled and educated workforce also enhances the absorptive capacity of a firm (Cohen and Levinthal 1989). We measure the shares of the firm employment accounted for by highly skilled, semi-skilled, and less-skilled workers. Since highly skilled workers enhance a firm's absorptive capacity, we expect that TA likelihood increases with a firm's skill base. We also expect that worker training to increase human capital increases absorptive capacity. We include a dummy variable reflecting if the trainer that the firm hires comes from the public sector. Another dimension of absorptive capacity is R&D. Investment in R&D, measured as the share of firm expenditures accounted for by R&D spending, should increase TA likelihood. But this measure does not include investment destined for technological transfer or equipment acquisition. To account for this exclusion, we include a variable that exclusively measures firm's expenditures on technology transfer or equipment acquisition. We expect this variable to positively correlate with TA. Since R&D from different sources may have different impacts on absorptive capacity, we include dummy variables to reflect five different R&D sources: consulting firms, public research institutions, private research institutions, the non-R&D department of the firm, and the R&D 3 Alternative methodologies, such as the framework that Fare, Grosskopf, and Lovell (1994) use, allow us to estimate productive efficiency 62 department of the firm. Clearly, R&D more closely linked to the firm's production will have greater influence on the firm's absorptive capacity. Since maquilas are foreign assembly plants with distinct industry groups and policy regulation, we measure whether maquilas are more likely to adopt technology. We measure industry-specific factors by including dummy variables for each of the nine major manufacturing sectors: (1) food, beverages, and tobacco; (2) textiles, clothing, and leather; (3) wood and wood products; (4) paper and paper products; (5) chemicals, oil derivatives, and coal; (6) non-metallic mineral products; (7) basic metal industries; (8) metal products, machinery, and equipment; and (9) other manufacturing industries. To determine the effect of location, we include three measures of regional science capacity, which we hypothesize to correlate with absorptive capacity: individuals with a science degree, government expenditure on R&D, and researchers registered in the federal or state systems. We measure these as percentages of the population by state. To develop a fuller understanding of technology, we distinguish between five types of technology that a manufacturing firm may adopt: (1) any type of technology; (2) automatic equipment; (3) machinery tools; (4) CNCM, and (5) robots. 4.2. Empirical evidence Although we have results of varying significance for all firm, industry, and region specific factors, we only discuss statistically significant results. A tabular presentation of the results from the best logit model appears in Tables A2.1 and A2.2. A. Any type of technology Firm factors. Firm size strongly correlates with TA likelihood, as microenterprises are less likely to adopt technology than their larger counterparts. On the whole, TA likelihood 63 positively relates to firm size, even after controlling for factors that could cause bias. In 1999, a firm's age negatively relates to TA likelihood, while there is no significant relationship in 1992. A strong positive relationship between the shares of different skills labor and TA likelihood appears in both years. Firms providing formal training are also more likely to adopt technology. In 1999, public training is positively correlated with the likelihood of TA. TA likelihood correlates both with R&D intensity and with transferred technology. Firm R&D investment promoted TA in both years. In 1999, it appears that R&D from public research institutions, from a firm's own research department or from another department encouraged TA, while in 1992 only R&D from public research institutions encouraged TA. A firm's status as a maquila related negatively to TA likelihood in 1992, but related positively in 1999; the same results appear in subsidiary firms. Union presence positively relates to TA likelihood. Industry factors. Controlling for firm-specific characteristics, TA likelihood is significantly lower in textiles, clothing, leather, wood, wood products, and in non-metallic mineral product industries than in other industries. Understandably, these are three industries where technology plays little role in production. In contrast, frmns in basic metal industries, and firms producing goods that do not fit in one of the eight standard categories exhibit a markedly higher likelihood of TA. Regional Factors. Science graduates and researchers per capita are positively correlated with the likelihood of TA. But, surprisingly, there is a negative relation with public expenditure in R&D and TA likelihood. B. Automatic equipment Firm factors. Small firms have a higher likelihood of adopting automatic equipment than other firms in 1999. The effect of firm age in the likelihood of TA shifted from positive in 1992 to negative in 1999. Foreign ownership firms are less likely to adopt automatic equipment than other firms. Subsidiary firms increase the likelihood of adopting automatic equipment in 1999, but in 1992 subsidiary firms are less likely to adopt automatic equipment. The shares of semi- skilled and low skilled workers positively relate to the adoption of automatic equipment. In 1999, 64 formal training correlates positively and strongly with the likelihood of adopting automatic equipment, but hiring a public trainer correlates negatively with the adoption of this technology. A firm's R&D investment is only significant in 1992. In this year, a firm's investment in technology transfer increases its likelihood of adopting automatic equipment, while the firm's investment in R&D has the opposite effect. Also in 1992, R&D from public institutions increased the likelihood of adopting automatic equipment. Export-oriented firms are more likely to adopt automatic equipment than other firms in 1999, while in 1992 the presence of a union increased a firm's likelihood of adopting automatic equipment. Industry factors. Adoption of automatic equipment is quite likely in all industries in 1999, but most likely in basic metal industries. Regional factors. The frequency of science graduates seems to lower the likelihood of adopting automatic equipment in 1999, but it increased the likelihood in 1992. The effect of location changed markedly from 1992 to 1999. In 1992, firms in the Central region were the most likely to adopt automatic equipment and firms in the Capital region were the least likely to adopt automatic equipment. However, in 1999, firms in the Capital region were the most likely to adopt automatic equipment, and firms in the South were the least likely to adopt automatic equipment. C. Machinery tools Firm factors. The adoption of machinery tools strongly correlates with firm size, as large firms are more likely to adopt machinery tools than medium firms, which in turn are more likely to adopt machinery tools than small firms. Firms age reduces the likelihood of adopting machinery tools. High shares of semi-skilled and low skilled workers encourage adoption of machinery tools. Formal training strongly and positively correlates with the likelihood of adopting machinery tools, and public training also increases the likelihood of adopting machinery tools. A firm's investment in R&D and technology transfer increases the likelihood of adopting machinery tools. In 1999, R&D from public institutions and a firm's department other than R&D represented a higher likelihood of adopting machinery tools. Maquila firms are less 65 likely to adopt machinery tools than other firms. In 1992, export-oriented firms were less likely to adopt machinery tools than other firms, while in 1999 firms with a union were more likely to adopt machinery tools. Industry factors. In 1992, the food, beverages, and tobacco industry was the least likely to adopt machinery tools. In 1999, paper and paper products firms were the most likely to adopt machinery tools. Regional factors. In 1992, prevalence science graduates and researchers decreased the likelihood of adopting machinery tools, but in 1999 their prevalence had no significant effect on the adoption of machinery tools. Firms located in the Capital region were the most likely to adopt machinery tools in 1992. In 1999, a firm's location in the Capital region did not affect its TA likelihood, but firms in the Central and South regions were still less likely to adopt machinery tools than firms in the North. D. Computerized numeric controlled machinery Firm factors. The adoption of CNCM also strongly correlates with firm size. Formal training strongty increases the likelihood of adopting CNCM. A firm's investment in R&D positively correlates with the adoption of CNCM in 1999; the same is true of R&D from a firm's department other than R&D. Industry factors. In 1992, the paper, paper products, and chemical products industries were the most likely to adopt CNCM. In 1999, the metal products, machinery, equipment, paper and paper product industries were the most likely to adopt CNCM. Regional factors. In 1992, the nearby presence of science graduates and researchers increased a firm's likelihood of adopting CNCM, while public expenditure in R&D reduced the likelihood of adopting this technology type. Firms located in the Capital region were the least likely to adopt CNCM in 1992. 66 E. Robots Results for robots are insignificant due to insufficient observations. The proportion of firrns that reported adopting robots is very small. Nevertheless, there is a strong correlation between firm size and robot adoption appears despite the small sample. To relate a variable to technologies of different complexity, we estimated multinomial logit models for 1992 and 1999. These estimations included six types of technology: manual equipment, automatic equipment, machinery tools, NCM, CNCM, and robots. We can order these technology types from simple manual technology to highly complex CNCM and robots. The results appear in Tables A2.3 and A2.4. An increasing importance of firm size and skilled workers with the likelihood of adopting more complex technology can be appreciated. 5. Determinants of Technology Adoption: Panel Estimation 5.1. Methodology The cross-sectional approach gives us a photo album of single-year snapshots that show TA patterns at different times. Panel data gives us real time video showing how patterns change. To put it in another way, cross-sectional data gives us points on a curve. Panel estimation shows curve slope at different points, but we need both levels and rates of change to fully understand TA. Another benefit of using both types of analysis is that, while results for a particular category may be insignificant for cross-sectional estimation, the results may become significant in panel estimation. In order to understand TA determinants better, we use the following probit model (following Tan (2000)):4 Pr(Adopt),, /Aj + 8,Xt, + /J2Share, + f3ARegion, + Ei, (1) where: Adopt, = a dummy variable denoting technology adoption in period t for firm i; 67 X6t, = a vector of firm attributes; 5 Share, = the rate of technology adoption at time period t, differentiated by technology type; Region, = the geographical region where the firm is located at time period t; a,i, = normal regression error. 5.2. Empirical Evidence Results from the best probit random effects model are shown in Tables A3. I -A3.3. These tables summarize estimations for the 1992-95, 1995-99 and 1992-99 periods. The results distinguish five types of technology: (1) any technology type; (2) automatic equipment; (3) machinery tools; (4) CNCM, and (5) robots. A. Firm Factors Size. For the 1992-99 period, medium and large firms, 57 percent and 66 percent respectively, are more likely than micro and small firms (the omitted category) to adopt some type of technology. Large firms are 43 percent more likely to adopt machinery tools and 93 percent more likely to adopt CNCM. Medium-size firms are also more likely than micro and small firms to adopt machinery tools and CNCM. For the 1992-95 period, firm size negatively correlates with TA, while in the 1995-99 period the correlation is positive. This result may come about because NAFTA and the economic crisis of 1994 encouraged medium and large manufacturing firms to acquire technology to compete globally. Age. Adoption of machinery tools positively relates to firm age, while adoption of automatic equipment negatively relates to firm age. One reason for this is that our sample only includes firms that survived through the entire period. Surviving small firms may be more flexible than surviving large firms. 4 To control for persistent technology adoption, equation (1) was also estimated following Wooldridge's (2001) methodology. The parameter estimates remain robust to those shown in Annex 3. 68 Foreign ownership. For the 1992-99 period, foreign ownership increased the probability of adopting automatic equipment and TA overall by 23 percent. For the 1995-99 period, foreign ownership increased the probability of acquiring robots by 42 percent. For the 1992-95 period, foreign ownership had an important influence on the adoption of CNCM. Skill. Skilled workers and human capital tend to enhance the absorptive capacity of firms (Cohen and Levinthal 1989). We expect that the presence of skilled labor will encourage TA, but that unskilled labor will negatively relate to TA. Results show that, for the 1992-95 period, the number of highly skilled employees positively correlates with TA, and for the 1992-99 period, it positively correlates with the adoption of automatic equipment. The number of professionals, technical employees, managerial employees, and semi-skilled workers shapes a firrn's TA patterns. Training. Intuitively, training workers should enhance a firm's absorptive capacity. The positive and significant coefficient of the training variable in our regression suggests that training does indeed enhance absorptive capacity. For the 1995-99 and 1992-99 periods, training positively correlates with TA. For the 1992-99 period, a firm that trains workers is 20 percent more likely to adopt some type of technology, and 41 percent more likely to adopt CNCM than a firm that does not provide training. For the 1992-95 and 1995-99 periods, training positively correlates with the adoption of more complex technology, like CNCM. R&D. Firn investment in R&D is also positively related to the adoption of complex technology. Moreover, a firm's investment in technology transfer increases the probability of adopting CNCM in the 1992-95 and 1992-99 periods, and increases the probability of adopting robots in the 1995-99 period. Maquila. The technology performance of maquila firms differed between 1992-95 and 1995-99. We find that the probability of TA for maquila firms in the earlier period is 10 percent 'We lag skill shares by one period (to the previous period's levels) to preclude bias from skill changes that occur during the period of technology adoption. The use of lagged skill shares measures restricts the sample of firms. 69 higher than for non-maquila firms. However, in the later period, it is 32 percent lower than for other firms. Exports. For the 1992-95 period, a firm's status as an export-oriented firm positively correlated with TA, and specifically with automatic equipment. However, we observe a negative relationship between exports and the adoption of machinery tools for the 1995-99 period. The 1992-99 period had no significant relations between export-oriented firms and the adoption of different types of technology. B. Regional Factors For the 1992-99 period, firms in Mexico City seemed less likely than firms in the North (the omitted category) to adopt some type of technology. There was a similarly strong relationship for the adoption of machinery tools. Surprisingly, for the 1995-99 period, firms in the Central and South regions were more likely to adopt CNCM than firms in the North. For the 1992-95 period, firms in the South and Capital regions were less likely to adopt technology than firms in the North. We can conclude that, in general, firms located in the North are more likely to adopt technology than firms located in other regions. C. Technology Diffusion Rate Significant limitations only allow us to estimate technology diffusion for the 1992-99 period. The effect of the technology diffusion rate is positive for the adoption of any type of technology, automatic equipment, and machinery tools, suggesting that a firm is more likely to adopt a technology if other firms are using the technology. 5.3 Technological Intensity Another measure of TA involves not just whether a firm adopts technology, but the degree to which it uses this technology. We refer to this degree as the intensity in the use of new technology. We measure technological intensity as the share of production equipment that the 70 technology accounts for. Tables A3.4-A3.6 compare technological intensity for six types of technology: manual equipment, automatic equipment, machinery tools, NCM, CNCM, and robots. From the panel tobit estimations for the 1995-99, and 1992-99 periods, we find that intensity in the use of more complex technologies is positively correlated with firm size. However, for the 1992-95 period, we find that the opposite relation: larger firms are negatively correlated with the intensity in the use of more complex technologies. We also find that the share of semi-skilled and unskilled workers, for the 1992-99 period, reduces the intensity in the use of NCM, for the 1995-99 period, reduces the intensity in the use of manual equipment, and for the 1992-95 period increases the intensity in the use of machinery tools. For the 1992-99 period, training increases the intensity in the use of automatic equipment and CNCM, for the 1995-99 period. For the 1992-95 and 1995-99 periods, training is positively correlated with the use of more complex technologies, while it reduces the use of more simple technologies. We find the same patterns with investments in R&D, for all the three periods, R&D increases the intensity in the use of more complex technologies, while it reduces the intensity in the use of more simple technologies such as manual equipment and machinery tools. For the three periods, export oriented firms are positively correlated with the intensity in the use of robots, while the presence of a union reduces the intensity in the use of machinery tools. The fact that a firm has joint activities has no effect in the technological intensity. Finally, firms located in the North are related with more intensively use of automatic equipment, and less intensively use of machinery tools, than the other regions. 6. Prolonged Activity We want to be able to predict the likelihood of a firm's activity at time t+l by knowing its actions at time t. To summarize this analysis, Table A4.1 shows conditional means for certain types of activity in 1999, given the activities in 1992. 71 We find consistency in export, training, and technology activities over time. In other words, a firm that receives over half its sales from exports, trains its workers, or adopts technology in 1992 is quite likely to do so again in 1999. In addition worker training precedes and strongly correlates with TA. A firm that did not adopt technology but did train its workers in 1992 has an 89 percent likelihood of adopting technology in 1999. The same phenomenon appears with exports: a firm that exported but did not adopt technology in 1992 has a 79 percent chance of adopting technology in 1999. Although part of this increase in technology adoption between periods is exogenous-TA rates for all firms increased by five to twenty-five percent between 1992 and 1999 (Figure 2)-the exogenous effect cannot explain all of the increase for non-adopting export and training firms. 7. Conclusions Two main conclusions appear from this paper. First, we can generally predict a firm's TA likelihood by knowing a few of its characteristics. TA likelihood increases with firm size. Firms that train workers, have a high share of skilled labor, have a high proportion of foreign capital, have large R&D budgets, and are located in the North are most likely to adopt new technologies. Moreover, subsidiary firms and firms with a union strongly increase the likelihood of TA in 1999. Firms adopting the most complex technologies are large, train workers, and have large R&D budgets. There is an increasing amount of skilled workers with the likelihood of adopting more complex technology. Also, larger finns, firms with a large share of highly skilled workers, and firms that train workers use intensively more complex technologies in their production process. Second, public policy can influence TA patterns in two main ways. The first is direct-by sponsoring formal training, funding R&D, or facilitating the formation of clusters or backward linkages, for example. The second mechanism is broader and develops through changes in the external environment. NAFTA, for example, appears to have significantly increased TA likelihood. Overall, TA rates increased considerably between 1992 and 1999. 72 The OECD (2000) emphasizes that governments can improve the effectiveness of R&D expenditure by supporting proliferation of venture capital and credit institutions. Public/private partnerships with selective participation also maximize the value of government R&D expenditure. These partnerships could take the forrn of shared seminars, working groups, or regular discussion meetings. Inviting private sector representatives to policy planning meetings offers a good way of integrating public priorities with private needs. Additionally, public research expenditure should focus on basic knowledge and broad findings that can aid a wide variety of industries. 73 References Aitken, B. and A. Harrison. 1992. "Does Proximity to Foreign Firms Induce Technology Spillovers? Evidence from Panel Data." MIT, Cambridge. Processed. Bell, M. and K. Pavitt. 1992. "Accumulating Technological Capability in Developing Countries." Proceedings of the World Bank Annual Conference on Development Economics, Supplement to The World Bank Economic Review and The World Bank Research Observer, Washington D.C. Blomstrom, M. 1989. Foreign Investment and Spillovers. London, Routledge. Blomstrom, M. and A. Kokko. 1998. "Multinational Corporations and Spillovers." Journal of Economic Survey 12(3) :247-77. Blomstrom, M., A. Kokko and M. Zejan. 1992. "Host Country Competition and Technology Transfer by Multinationals." Working Paper 4131. National Bureau of Economic Research. Blomstrom, M. and E. N. Wolff. 1994. "Multinational Corporations and Productivity Convergence in Mexico." In W. Baumol, R. Nelson and E. Wolff, eds., Convergence of Productivity: Cross-National Studies and Historical Evidence. New York: Oxford University Press :263-84. Blomstrom, M. and H. Persson. 1983. "Foreign Investment and Spillover Efficiency in an Underdeveloped Economy: Evidence from the Mexican Manufacturing Industry." World Development 11(6) :493-501. Blomstrom, M. and J. Y. Wang. 1992. "Foreign Investment and Technology Transfer: A Simple Model." European Economic Review 36(1):137-55. Cantwell, J. 1989. Technological Innovation and Multinational Corporations. Oxford: Basil Blackwell. Caves, R. E. 1974. "Multinational Firrns, Competition, and Productivity in Host-Country Markets." Economica 41(162):176-93. Cohen, W. M. and D. A. Levinthal. 1989. "Innovation and Learning: The Two Faces of R&D." Economic Journal 99(397) :569-96. Davies, S. 1979. Diffusion of Process Innovations. Cambridge: Cambridge University Press. Dosi, G., K. Pavitt and L. Soete. 1990. The Economics of Technical Change and International Trade. New York: New York University Press. Globerman, S. 1975. "Technological Diffusion in the Canadian Tool and Die Industry." Review of Economics and Statistics 57(4) :428-34. ---------. 1979. "Foreign Direct Investment and Spillover Efficiency Benefits in Canadian Manufacturing Industries." Canadian Journal of Economics 12(1) :42-56. Fare, R., S. Grosskopf, and C. A. Lovell. 1994. Production Frontiers. Cambridge University Press. Grether, J.M. 1999. "Determinants of Technological Diff-usion in Mexican Manufacturing: A Plant-Level Analysis." World Development 27(7) :1287-98. Grossman, G. and E. Helpman. 1991. Innovation and Growth in the Global Economy. Cambridge MA: MIT Press. 74 Haddad, M. and A. Harrison. 1993. "Are there Positive Spillovers From Foreign Direct Investmnent? Evidence From Panel Data for Morocco." Journal of Development Economics 42(1) :51-74. Iscan, T. 1998. "Trade Liberalisation and Productivity: A Panel Study of the Mexican Manufacturing Industry." Journal of Development Studies 34(5) :123-48. Lapan, H. and P. K. Bardhan. 1973. "Localized Technical Progress and Transfer of Technology and Economic Development." Journal ofEconomic Theory 6(6) :585-95. Mansfield, E. 1961. "The Speed of Response of Firms to New Techniques." Quarterly Journal of Economics 77. OECD (Organisation for Economic Co-operation and Development). 1997. Diffusing Technology to Industry.- Government Policies and Programmes. Paris. ---------. 2000. Science, Technology and Industry Outlook. Paris. Romeo, A. 1975. "Inter-industry and Inter-firm Differences in the Rate of Diffusion of an Innovation." Review of Economics and Statistics 57(2) :311-319. Romer, P. 1994. "New Goods, Old Theory, and the Welfare Costs of Trade Restrictions." Journal of Development Economics 43(1) :5-38. Tan, H. 2000. "Technological Change and Skills Demand: Panel Evidence from Malaysian Manufacturing." Working Paper. The World Bank Institute. Torres, E. 2000. "Diferencias Estatales en la Especializaci6n Tecnologica y Cientifica de Mexico." CONACYT (Consejo Nacional de Ciencia y Tecnologia). Processed. Wooldridge, J. M. 2001. "The Initial Conditions Problem in Dynamic, Nonlinear Panel Data Models with Unobserved Heterogeneity." Michigan State University, Department of Economics. Processed. World Bank. 1998a. "Enhancing Total Factor Productivity Growth." Report No. 17392-ME (Gray Cover). ---------. 1998b. "Knowledge and Innovation Project." PAD, Report No. 17896. ---------. 2000. "Export Dynamics and Productivity: Analysis of Mexican Manufacturing in the 1990s." Report No. 19864-ME (Green Cover), September. Young, A. 1991. "Learning by Doing and the Dynamic Effects of International Trade." Quarterly Journal of Economics 106(2) :369-405. 75 Annex 1: Technology Adoption Descriptive Statistics for 1992 Table A1.1. TA by Technology Type and Firm Size, 1992 Percent of firms that adopt technology Type of technology Micro Small Medium Large All Manual equipment 6.4 6.6 7.7 4.1 6.4 Automatic equipment 10.4 22.9 23.0 19.2 11.9 Machinery tools 6.6 14.5 25.3 35.6 8.1 NCM 0.2 3.1 6.2 5.0 0.7 CNCM 1.0 5.6 8.1 15.1 1.8 Robots 0.0 0.0 0.3 0.7 0.0 Source: Author's calculations based on ENESTYC 92. Table A1.2. T by Training and Export Status by Firm Size, 1992 Provides formal training Export oriented Firm size Yes No Yes No Micro 64.4 21.9 32.2 24.6 Small 63.8 40.8 45.1 53.0 Medium 72.3 65.4 78.7 70.1 Large 81.3 65.1 77.3 79.8 All L 66.7 23.2 47.8 28.8 Note: Figures refer to the percentof fimns that adopt tedmology. Source: Author's calculations based on ENESTYC 92. Table A1.3. TA by Industry and Firm Size, 1992 Percent of firms that adopt technology D)ivision Micro Small Medium Large All Food, beverage, tobacco 26.9 56.0 69.1 77.7 29.0 Textiles, clothing, leather 10.5 39.7 70.6 78.9 19.1 Wood, wood products 13.1 52.3 67.6 64.4 16.8 Paper, paper products 40.8 71.4 67.6 83.8 45.4 Chemical products, oil derivatives, coal 40.7 58.4 73.4 80.2 52.3 Non-metallic minerals 9.7 38.7 61.7 72.0 12.1 Basic metal industries 70.0 51.3 49.2 74.6 64.3 Metal products, machinery, equipment 34.2 56.9 73.7 82.8 39.1 Other manufacturing industries 68.0 56.3 72.2 86.7 66.8 Source: Author's calculations based on ENESTYC 92. 76 Annex 2: Cross-Section Estimations Table A2.1. Likelihood of TA for Manufacturing Firms, 1992 Any type of Automatic Machincry tools CNCM Robots Explanatory Variables technology equipment Coeff. Marg. Ef. Coeff. Marg. Ef. Coeff. Marg. Ef. Coeff. Marg. Ef. Coeff. Marg. Ef. Firm-specific Size: Small 0.452 ** 0.0851 0.131 0.0101 0.369 * 0.0163 1.565 ** 0.0033 22.061 0.0000 Medium 1.068 *# 0.2383 -0.058 -0.0045 0.936 ** 0.0589 2.089 * 0.0071 24.745 0.0000 Large 1.585 * 0.3651 -0.295 -0.0210 1.495 ** 0.1228 2.982 ** 0.0187 25.076 0.0000 Age 0.000 0.0000 0.011 * 0.0009 -0.020 * -0.0008 -0.013 0.0000 0.014 0.0000 Foreign ownership -0.166 -0.0298 -0.361 -0.0251 0.099 0.0043 0.135 0.0002 0.475 0.0000 Subsidiary -1.031 ** -0.1412 -1.020 ** -0.0535 -0.097 -0.0037 -0.306 -0.0003 0.771 0.0000 Share of labor: Semi-skilled 0.007 * 0.0013 0.008 ** 0.0007 0.013 ** 0.0006 -0.002 0.0000 -0.128 0.0000 Lowskilled 0.005 * 0.0010 0.007 * 0.0006 0.009 ** 0.0004 -0.009 0.0000 -0.124 0.0000 Training 1.334 ** 0.2658 0.531 ** 0.0452 1.048 ** 0.0581 1.431 ** 0.0026 Publictraining 0.058 0.0103 -0.238 -0.0163 0.111 0.0045 -0.428 -0.0003 0.737 0.0000 R&D 0.065 ** 0.0126 -0.031 ** -0.0025 0.045 ** 0.0019 0.096 ** 0.0001 0.080 0.0000 Technology transfer 0.018 * 0.0034 0.016 ** 0.0013 0.005 * 0.0002 0.007 0.0000 0.012 0.0000 Source of R&D: Public Institutions 1.025 ** 0.2297 1.240 * 0.1606 0.389 0.0191 -1.014 -0.0007 Private Institutions 0.856 0.1908 0.213 0.0189 0.938 0.0607 -0.547 -0.0004 Other firm's department -0.238 -0.0422 0.062 0.0051 -0.808 -0.0236 1.670 * 0.0045 2.454 0.0000 Own firm's R&D dept. -0.446 -0.0747 0.503 0.0498 -0.562 -0.0182 0.532 0.0007 0.868 0.0000 Maquila -0.692 * -0.0957 0.034 0.0024 -1.142 * -0.0268 0.349 0.0004 -0.624 0.0000 Exportoriented -0.266 -0.0468 0.203 0.0178 -1.353 ** -0.0319 -0.501 -0.0004 0.707 0.0000 Union 0.409 ** 0.0712 0.558 ** 0.0467 0.300 0.0120 0.244 0.0002 0.140 0.0000 Industry-specific Textiles, clothing, leather -0.853 *t -0.1158 -0.308 * -0.0197 -0.786 ** -0.0214 1.055 * 0.0016 Wood,woodproducts -0.814 ** -0.1122 -0.168 -0.0114 -2.313 ** -0.0379 0.082 0.0001 Paper, paper products 0.466 ** 0.0920 0.364 * 0.0322 -0.839 ** -0.0235 3.241 ** 0.0212 Chemical products 0.003 0.0006 -0.699 ** -0.0417 0.216 0.0096 1.776 ** 0.0048 17.097 0.0000 Non-metallic minerals -1.380 ** -0.1646 -0.694 ** -0.0393 -1.878 ** -0.0353 -0.635 -0.0005 Basic metal industries 1.443 ** 0.3336 0.015 0.0012 -0.508 -0.0169 0.370 0.0005 Metal products, machinery 0.296 ** 0.0483 0.915 ** 0.0818 -1.187 * -0.0266 0.423 0.0004 17.540 0.0000 Othermanufacturingind. 1.843 * 0.4244 2.246 ** 0.3859 -1.237 * -0.0305 1.000 0.0018 Region-specific Science grad/cap 0.017 0.0032 0.452 ** 0.0443 -1.495 ** -0.0336 3.300 ** 0.0270 -1.795 0.0000 Researchers/cap 1.846 ** 0.4293 0.341 0.0319 -1.184 -0.0299 7.324 * 0.6167 -2.964 0.0000 Public exp. in R&D -0.053 ** -0.0099 0.013 0.0011 0.014 0.0006 -0.272 ** -0.0003 0.130 0.0000 Region: Central -0.304 * -0.0341 0.606 ** 0.0356 -1.444 * -0.0254 0.567 0.0004 -0.986 0.0000 South -0.285 * -0.0365 0.188 0.0115 -1.710 ** -0.0303 -0.738 -0.0004 Capital -3.640 ** -0.2393 -3.540 ** -0.0845 3.885 ** 0.5049 -9.318 ** -0.0011 2.173 0.0000 Constant -1.452 *-4.276 **0.519 -11.2160 ** 33,957 Number of obs. 5,071 5,071 5071 . 5,071 2,022 Log likelihood -2414.8016 -1598.7678 -1136.7526 -225.359 -27.2859 Pseudo R2 0.2095 0.1379 0.2065 0.5118 0.3280 * Significant at 10% level; **Significant at 5% level. Note: Dependent variable = I if firm adopted any type of technology, 0 otherwise. 77 Table A2.2. Likelihood of TA for Manufacturing Firms, 1999 Any type of Automatic Machinery tools CNCM Robots Explanatory Variables technology equipment Cocff. Marg. Ef. Coeff. Marg. Ef. Coeff. Marg. EL Coeff. Marg. Ef Coeff. Marg. Ef. Firm-specific Size: Small 1.100 * 0.0002 0.570 ** 0.0491 0.941 ** 0.0691 2.894 ** 0.0126 14.937 ** 0.0000 Medium 1.311 ** 0.0002 0.290 0.0239 1.319 ** 0.1233 3.575 ** 0.0294 16.861 0.0003 Large 1.760 ** 0.0002 -0.029 -0.0021 1.717 ** 0.1893 3.789 * 0.0365 17.058 ** 0.0004 Age -0.040 * 0.0000 -0.015 * -0.0011 -0.015 * -0.0008 -0.008 0.0000 0.000 0.0000 Foreign ownership -0.694 -0.0003 -1.346 ** -0.0578 0.426 0.0270 -0.199 -0.0002 0.657 0.0000 Subsidiary 0.541 ** 0.0001 0.525 ** 0.0445 -0.220 -0.0099 0.459 0.0005 0.608 0.0000 Share of labor: Semi-skilled 0.005 ** 0.0000 0.013 ** 0.0010 0.008 ** 0.0004 0.008 0.0000 0.082 0.0000 Low skilled 0.008 * 0.0000 0.007 ** 0.0005 0.007 * 0.0004 0.001 0.0000 0.086 0.0000 Training 0.469 ** 0.0001 0.457 ** 0.0347 0.405 ** 0.0219 1.236 ** 0.0018 0.558 0.0000 Public training 0.402 ** 0.0001 -0.555 ** -0.0308 0.463 ** 0.0277 0.378 0.0004 0.303 0.0000 R&D 2.854 ** 0.0003 0.029 t* 0.0021 0.043 ** 0.0023 0.015 0.0000 0.029 0.0000 Technology transfer 0.371 ** 0.0001 -0.056 -0.0040 0.179 * 0.0103 -0.021 0.0000 0.167 0.0000 Source of R&D: Public Institutions 2.602 ** 0.0003 -0.474 -0.0286 1.439 ** 0.1430 -0.830 -0.0005 -0.083 0.0000 Private Institutions -0.252 -0.0001 -1.507 -0.0431 -0.424 -0.0003 Otherfirm'sdepartment 3.989 ** 0.0003 1.038 0.1170 1.238 * 0.1135 -1.230 -0.0006 Own firm's R&D dept. 3.197 ** 0.0003 0.710 * 0.0701 -0.155 -0.0076 0.396 0.0004 0.860 0.0000 Maquila 0.356 ** 0.0001 0.178 0.0129 -0.417 '* -0.0171 -0.020 0.0000 0.026 0.0000 Export oriented 0.527 0.0001 0.654 ** 0.0620 0.180 0.0101 -0.283 -0.0002 0.115 0.0000 Union 0.568 ** 0.0001 -0.018 -0.0012 0.487 ** 0.0288 -0.505 -0.0003 0.788 0.0000 Industry-specific Textiles, clothing, leather 0.204 * 0.0000 0.744 ** 0.0623 -0.558 * -0.0207 0.016 0.0000 -0.418 0.0000 Wood, woodproducts 0.355 ** 0.0001 1.615 ** 0.1811 -1.664 * -0.0423 -1.651 -0.0007 -0.300 0.0000 Paper, paper products 0.915 ** 0.0002 0.465 * 0.0389 0.602 * 0.0388 1.016 * 0.0015 -0.032 0.0000 Chemical products 0.675 ** 0.0001 0.940 ** 0.0982 0.257 0.0147 0.393 0.0004 -0.831 0.0000 Non-metallic minerals 0.099 0.0000 0.102 0.0071 -1.487 ** -0.0409 -0.882 -0.0005 0.434 0.0000 Basic metal industries 1.452 * 0.0002 2.139 ** 0.3442 -0.405 -0.0179 -0.535 -0.0004 0.038 0.0000 Metal products, machinery 0.776 ** 0.0001 1.852 ** 0.2034 -1.195 * -0.0342 1.196 ** 0.0015 0.804 0.0000 Other manufacturing ind. 0.960 ** 0.0002 1.381 ** 0.1730 -0.646 -0.0256 -1.056 -0.0006 0.404 0.0000 Region-specific Sciencegradlcap 0.304 ** 0.0001 -0.244 -0.0163 -0.074 -0.0038 0.117 0.0001 0.069 0.0000 Researchers/cap 1.071 ** 0.0002 -0.314 -0.0203 0.518 0.0347 -0.586 -0.0004 -0.640 0.0000 Public exp. in R&D -0.046 ** 0.0000 -0.009 -0.0006 -0.026 -0.0014 0.006 0.0000 0.009 0.0000 Region: Central 0.068 0.0000 0.117 0.0039 -0.391 ** -0.0084 0.199 0.0002 0.589 0.0000 South 0.164 0.0000 -0.439 ** -0.0233 -0.325 -0.0128 -0.234 -0.0002 2.082 0.0000 Capital -0.732 -0.0003 2.291 * 0.3249 0.214 0.0111 2.959 0.0119 2.392 0.0000 Constant -2.192 ** -3.337 * -2.382 ** -7.898 ** -32.705 Number of obs. 7,220 7,207 7,220 7,220 7,165 Log likelihood -2629.3164 -2216.7957 -1786.7886 -205.736 -9.8271 Pseudo R2 0.4654 0.1424 0.1998 0.4059 0.4412 * Significant at 10% level; *"Significant at 5% level. Note: Dependent variable = I if firn adopted any type of technology, 0 otherwise. 78 Table A2.3. Multinomial Logit Results for Technology Adoption in Manufacturing Firms, 1992 Manual Automatic Machinery tools NCM CNCM Robots Explanatory Variables equipment equipment Coeff. Z~-St. Coeff. Z-St. Cueff Z-St. Coeff. Z,-St. Coeff. Z,-St. Coeff. Z-St. Firm-specific Size: Small 0.089 0.35 0.317* 1.90 0.378 * 1.86 1.452 2.56 1.128 ~'3.14 25.556 0.00 Medium 0.577 1.22 0.520 1.61 1.213 ** 3.54 2.168 2.97 1.965 '~3.59 28.803 0.00 Large 0.116 0.16 0.742 * 1.69 1.893 4.46 2.175 10243 3.052 5.00 29.690 0.00 Age 0.005 1.10 0.010 2.90 -0.010 -2.00 0.015 1.14 -0.012 -1.06 -0.003 -0.04 Foreign ownership 0.376 0.63 -0.140 -0.32 -0.079 -0.18 0 511 0.71 -0.613 -1.00 0.967 0.37 Subsidiary -1.061 -2.42 -1.088 * 4.39 -0.599 -2.19 -1.369 -2.03 -1.113 -2.27 0.108 0.04 Semi-skilled workers -0.003 -1.09 0.008 *0 4.11 0.012 *0 4.67 0.017 1.00 0.004 0.61 -0132 -0.36 Lest skilled workers 0.003 1.46 0.005 *0 2.59 0.010 4.19 0.009 0.55 -0.003 -0.46 -0.136 -0.38 Training 0.989 3.29 0.782 4.27 1.388 *0 6.28 2.705 0* 5.05 2.575 00 7.56 19.443 0.00 Public training 1.008 0* 3.10 0.101 0.46 0.559 2.34 -0.475 -1.09 -0.306 -0.92 0.939 0.34 Technology transfer 0.026 6.35 0.029 0* .96 0.029 *0 8.40 0.019 * 1.75 0.032 5.72 0.046 1.17 Maquila -1.077 0 -4.46 -0.114 -0.84 -1.097 -5.37 0.123 0.30 0.866 3.28 -0.714 -0.27 Export oriented 0.776 1.54 0 163 0.39 -1.384 *0-2.34 -0.122 -0.12 -0.577 -0.66 0.678 0.23 Union -0.363 -1.61 0.614 0* 4.22 0.398 *0 2.17 0.424 0.86 -0.054 -0.17 0.453 0.15 Region-specirie Science girad/cap 0.861 0* 2.87 0.737 0* 3.13 -0.976 0* 3.82 0.678 0.61 2.756 4.10 -1.244 -0.22 Researchers/cap 0.768 0.71 1.201 0* 2.25 0.009 0.01 0.707 0.33 7.100 *0 7.35 -2.313 -0.16 Plublic exp. in R&D -0.060 0 -1.74 -0.021 -1.08 -0.021 -0.82 0.020 0.25 -0.233 0*-5.89 0.112 0.22 Region: Central 0.227 0.92 0.385 0* 2.26 -1.471 *0-8.31 0.050 0.07 -0.017 -0.04 -1.254 -0.36 South 0.989 00 3.44 0.038 0.17 -1.443 -6.78 -1.027 -0.72 -1.771 -1.54 -32.089 0.00 Capital 1.595 0.89 -4.416 00 -6.19 1.449 1.10 -5.427 * -1.87 -11.706 -9.07 0.138 0.01 Constant -3.893 -7.90, -4.053 -10.26, -0.679 * -1.75 -9.238 0*-4.03, -8.983 00-7.26 -38.897 - nb, 4 :t ' i Lc-A ie'h-. -3! 32 S "s P"eud-' R2 'I I-'5 *Significant at 1O% level; "0Significant at 5% level. Note: Dependent variable =I if firm adopted manual equipment, 2 if firm adopted automatic equipment, 3 if firm adopted machinery tools, 4 if firm adopted NCM, 5 if firmia adopted CNCM, 6 if firm adopted robots, and 0 otherwise. The comparison groLp is no adoption of new technology. Table A2.4. Multinomial Logit Rsults for Technolog Adoption in MnfcuigFirms, 1999 Manual Automatic Machinery NCMCNMRbt Explanatory Varibles equipment equipment tools ____________________ Coeff. Z,-St. Coeff Z,-St. Coeff Z,-St. coeff Z,-St. Coeff Z,-St. Coeff 7,-St. Firm-specific Size: Small 0.469 so 2.74 1.281 00 7.61 1.509 * * 8.87 1.869 *0* 4.81 3.607 * * 7.64 39.461 1.16 Medium 0.396 0.84 1.392 * 3.15 1.968 * * 4.83 1.686 * * 2.23 4.439 *0* 6.40 41.490 1.20 Large 0.142 0.20 1.678 0* 2.72 2.556 *0* 4.56 2.334 002.64 5.089 0* 6.18 42.186 1.20 Age -0.056 0* -12.87 -0.042 0* -8.72 -0040 *0-7.92 -0.049 00-3.47 -0G43 *0-3.72 -0.036 -0.74 Foreign ownership -0.793 -1.22 -1.016 0 -1.64 -0.308 -0.56 -0.880 -0.95 -0.265 -0.37 0.450 0.23 Subsidiary 0.189 1.37 0.239 1.55 0.128 0.75 0.500 1.28 0.580 1.53 1.055 0.55 Semi-akilled workers 0.000 0.31 0.013 * 9.97 0.012 00 7.61 0.001 0.24 0.012 1.17 0.074 0.21 Less akilled workers 0.006 00 6.23 0.008 00 5.42 0.009 00 4.93 0.003 0.58 0.003 0.29 0.075 0.22 Training 0.632 * 4,26 0.971 5 6.31 0.900 0 5.32 2.334 5.82 1.941 * 4.16 1.334 0.35 Public training 0.422 ~s 2.06 0.075 0.35 0.831 *0 3.99 0.882 so 2.32 0.903 *0 2.33 0 950 0.46 Technology transfer 0.740 0* 5.20 0.706 0* 4.89 0.740 00 5.19 0.564 1.55 0.727 *0 3.92 0.959 * 1.75 Maquila 0.065 0,54 0.079 0.54 -0.328 0 -1.84 -0.399 -0.88 -0.222 -0.53 -0.365 -0.19 Export oriented 0.871 *0 2.53 1.194 00 3.38 0.575 1.51 0.597 0.91 0A498 0.78 1.029 0.48 Union 0.680 *0 4.78 0.281 * 1.70 0.650 *0 3.94 0.531 1.37 -0.046 -0.13 1.276 0.52 Region-specific Science grad/cap 0.206 1.62 -0.170 -1.10 -0.248 -1.39 -1.165 *0-2.08 0.174 0.25 -0.046 -001l Researchers/cap -0.186 -0.46 -1.033 * -1.89 -0556 -0.98 -5.451 0*-3.00 -2.081 -1.02 -1.936 -0.19 Public exp. in R&D 0.014 0.98 0.019 1.02 0.013 0.68 0.190 0* 3.18 0.055 0.81 0.044 0.13 Region: Central -0.111 -1.11 -0.083 -0.70 -0.582 00-4.53 -1.023 0*-2.53 -0.045 -0.09 0.178 0.07 South -0.182 -1.24 -0.456 -2.501 -0.545 00-2.76 -0.152 -0.27 -0.548 -0.55 1.459 0.41 Capital -1.399 *0 -2.35 2.430 0* 2.761 1.093 1.24 5.211 * 1.76 4.415 1.30 4.441 0.25 Conatant ~~~-1.190 *5 -5.21 -1.716 *0 -6.141 -1.84 ~o-.7 -.9 *-.1 -.4 0-.9-54.322 Number of ohs. = 7,220 Log likelihood -7638.0338 Pseudo R.2 = 0. I1113 *Significant at 10% level; "*Significan~t at 5% level. Note: Dependent variable =1 if firm adopted manual equipment, 2 if firm adopted automatic equipment, 3 if firm adopted machinery tools, 4 if firm adopted NCM, 5 if firma adopted CNCM, 6 if firm adopted robots, and 0 otherwise. The comparison groip is no adoption of new technology. 79 Annex 3: Panel and Intensity Estimations Table A3.1. Probit Results for TA in Manufacturing Firms, 1992-1995 Any type of Automatic Machinery tools CNCM Robots Explanatory technology equipment Variables Coeff. Marg. Ef. Coeff. Marg. Ef. CoeffE Marg. EL. Coet!. Marg. Ef. Coeff. Marg. Ef. Firm-spccific Size: Medium -0.4654 * -0.0948 -0.1252 -0.0192 -0.3380 * -0.0504 -0.5987 -0.0639 -0.4280 -0.0049 Large -0.9472 ** -0.1994 -0.4574 ** -0.0701 -0.8626 **-0.1211 -1.1654 **-0.1118 -1.8429 -0.0166 Age -0.0008 -0.0002 -0.0036 ' -0.0007 0.0051 **0.0010 -0.0031 -0.0005 -0.0198 -0.0003 Foreign ownership 0.3595 ** 0.0750 0.1187 .0.0195 0.0139 0.0023 0.2407 0.0327 0.6557 0.0118 Labor: Highly skilled 0.0076 * 0.0019 0.0037 0.0007 0.0041 0.0008 -0.0056 -0.0008 0.0115 0.0002 Semi-skilled 0.0001 0.0000 -0.0001 0.0000 0.0001 0.0000 0.0001 0.0000 0.0000 0.0000 Low skilled 0.0002 ** 0.0001 -0.0001 0.000 0.0002 * 0.0000 0.0001 0.0000 0.0000 0.0000 Training 0.0800 0.0065 -0.0704 -0.0043 0.0934 0.0059 0.1244 * 0.0062 0.2719 0.4876 R&D 0.0045 ** 0.0011 0.0009 0.0002 0.0032 0.0006 0.0044 0.0007 0.0026 0.0000 Technology transfer 0.4148 ** 0.1023 -0.0251 -0.0048 0.0855 * 0.0169 0.2501 0.0410 -0.4019 -0.0054 Maquila 0.0977 * 0.0183 0.0469 0.0068 0.0672 0.0099 0.1196 0.0140 -0.2756 0.4876 Export oriented 0.2516 ** 0.0601 0.3583 0.0709 0.0809 0.0153 0.0300 0.0044 0.1758 0.0030 Union 0.0452 0.0028 -0.0565 -0.0027 0.0265 0.0013 0.1183 0.0046 0.0440 0.0002 Region: Central -0.0828 -0.0115 -0.0187 -0.0020 -0.0782 -0.0083 0.0217 0.0018 0.5478 0.0058 South -0.1736 * -0.0403 -0.2052 -0.0348 0.0157 0.0028 -0.1653 -0.0223 -7.6690 -0.0166 Capital -0.1532 -0.0299 -0.0082 -0.0012 -0.2397 **-0.0347 0.1317 0.0161 0.1930 0.0026 Constant -0.0658 -0.8221 **-1.1248 **-1.4916 -2.7130 keglikehood -23 .26 .......6 . -1361.4194 -14TR2.3422 ---- _ -965.6919 -1------- 1 .916 .. .. *Significant at 1 0%/ level; **Significant at 5%1/ level. Notes: 1. Dependent variable 1lif the firm adopted any type ofnew technology,O0otherwise. 2. Technology diffusion rate was dropped due to coIllinearity. 3. Skill shares are lagged one period. 4. Number of observations = 3,293; number of groups = 3,293. Table A3.2. Probit Results for TA in Manufacturing Frms, 1995-1999_______ Any type of Automatic Machinery tools CNCM Robots Explanatory technology equipment Variables Coeff. Marg. EfL Coeff. Marg. Ef. Coeff. Marg. EfL CoefT. Marg. Ef. Coeff. Marg. Ef. Firm-specific Size: Medium 0.4670 * 0.0427 0.0234 0.0024 0.3846 * 0.0531 0.3166* 0.0338 0.0047 0.0001 Large 0.6922 ** 0.0599 -0.0830 -0.0082 0.5115 **0.0691 0.4888 **0.0522 0.0429 0.0007 Age -0.0031 -0.0005 -0.0045 * -0.0008 0.0025 0.0006 -0.0004 -0.0001 -0.0032 -0.0001 Foreign ownership 0.1491 0.0222 -0.0827 -0.0130 -0.0010 -0.0002 0.1257 0.0208 0.4161 * 0.0137 Labor: Highly skilled 0.0078 0.0014 0.0028 0.0005 0.0041 0.0010 0.0020 0.0004 -0.0305 -0.0010 Semi-skilled 0.0003 0.0000 -0.0006 ** -0.0001 0.0002 0.0000 0.0002 0.0000 0.0000 0.0000 Low skilled -0.0001 0.0000 0.0000 0.0000 -0.0001 0.0000 -0.0001 0.0000 0.0003 0.0000 Training 0.3956 ** 0.0084 -0.1443 -0.0032 0.0416 0.0013 0.7484 0.0172 4.4212 0.0221 Technology transfer -0.0585 -0.0106 -0.1000 -0.0179 -0.0972 -0.0240 0.1147 0.0218 0.1836 **0.0064 Maquila -0.3195 ** -0.0491 -0.1085 -0.0158 -0.0492 -0.0099 0.0517 0.0078 -0.4858 -0.0105 Exportoriented 0.0349 0.0056 0.0480 0.0081 -0.1747 *-0.0392 0.0374 0.0064 0.2439 0.0079 Union -0.1668 -0.0034 -0.1807 -0.0037 -0.0505 -0.0014 -0.0543 -0.0011 4.4508 0.0205 Region: Central -0.0087 -0.0009 -0.0751 -0.0076 -0.1640 **-0.0226 0.2613 * 0.0278 0.1181 0.0022 South 0.0580 0.0095 -0.3375 * -0.0533 -0.0439 -0.0102 0.2899 * 0.0535 0.1829 0.0059 Capital -0.1519 -0.0207 -0.0628 -0.0085 -0.0847 -0.0156 0.1895 0.0270 0.3241 0.0086 Constant 0.2471 -0.5688 **-0.5561 **-2.4005 **-11I.0014 --L__og li_kel 1i-hoo d- -598.6809 -5 .......3... -114..92.-79.884-95.0515 *Significant at I10%/ level; "*Significant at 5% level. NVotes: 1. Dependent variable = I if the finTn adopted any type of new technology, 0 otherwise. 2. Technology diffusion rate was dropped due to collinearity. 3. Skill shares are lagged one period. 4. Numnber of observations =1,702; number of groups =1,702. 80 Table A3.3. Probit Results for TA in Manufacturing Firms, 1992-1999 Any type of Automatic Machinery tools CNCM Robots Explanatory Variables technology equipment Coeff_. Marg. Ef. Coeff. Marg. Ef Coeff. MAE& Ef. Coeff. Marg. Ef Coeff. Marg. EfL Firm-specific Size: Medium 0.5677 ** 0.0626 0.2373 0.0351 0.2987 ** 0.0610 0.6368 ** 0.0712 -0.1330 -0-0017 Large 0.6572 ** 0.0583 0.0310 0.0037 0.4350 ** 0.0682 0.9297 * 0.0843 -0.2672 -0.0026 Age -0.0009 -0.0001 -0.0037 * -0.0009 0.0041 ** 0.0010 -0.0017 -0.0003 -0,0081 -0.0002 Foreign ownership 0.2797 ** 0.0387 0.2369 ** 0.0502 -0.0750 0.0000 0.0566 0.0076 0.2455 0.0055 Labor: Highly skilled 0.0109 * 0.0019 0.0063 * 0.0015 0.0022 0.0191 -0.0045 -0.0008 0.0112 0.0003 Semi-skilled 0.0003 0.0001 -0.0004 * -0.0001 0.0001 0.0187 0.0001 0.0000 0.0003 * 0.0000 Low skilled 0.0004 * 0.0001 0.0002 0.0000 0.0001 0.0186 0.0001 0.0000 0.0000 0.0000 Training 0.2106 ** 0.0058 -0.0678 -0.0025 0.1057 0.0221 0.4147 ** 0.0103 4.3972 0.0197 R&D 0.0029 0.0006 0.0013 0.0003 0.0014 0.0190 0.0021 0.0004 0.0035 0.0001 Technology transfer 0.0774 0.0139 -0.1456 -0.0366 0.0138 0.0209 0.1262 * 0.0221 0.0289 0.0007 Maquila -0.0944 -0.0114 -0.0803 -0.0160 0.0280 0.0257 0.0957 0.0142 -0.3871 -0.0061 Export oriented 0.0631 0.0106 -0.0993 -0.0234 -0.1565 -0.0159 0.2328 0.0396 -0.0895 -0.0017 Union -0.1130 -0.0019 -0.1685 -0.0038 -0.1420 0.0155 -0.0275 -0.0005 4.9396 0.0123 Region: Central -0.1264 -0.0112 -0.0158 -0.0018 -0.2335 ** -0.0077 0.1500 0.0129 0.1647 0.0018 South -0.2131 -0.0387 -0.1804 -0.0421 -0.1246 -0.0077 0.0351 0.0063 -5.6082 -0.0226 Capital -0.2012 * -0.0310 0.0218 0.0042 -0.3380 ** -0.0403 0.2027 0.0277 0.3089 0.0057 Technology diffusion rate 0.0407 ** 0.0087 0.5701 ** 0.1411 0.0476 * 0.0300 0.0194 0.0034 -0.4093 -0.0081 Constant -2.3784 **-9.2729 *4 -1.9412 ** -2.6713 ** -11.3325 - ... '- ......_ _......... ....I. I. ... ............... ........... . ...... .... ........ . .......... .............. ... . ....... ......... ...... . . ...... j .4 .i .... :............. Log likelihood 1016.1305 -837.4122 -1253.5804 -833.0133 -84.3751 * Significant at 10% level; **Significant at 5% level. Notes: 1. Dependent variable = I if the firm adopted any type of new technology, 0 otherwise. 2. Skill shares are lagged one period. 3. Number of observations = 2,089; number of groups = 1,066. 81 Table A3.4. Tobit Results for TI in Manufacturing Firms, 1992-1995 Manual Automatic Machinery NCM CNCM Robots Explanatory equipment equipment tools Variables Coeff. Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-St. Firm-specific Size: Medium 5.101 ** 5.02 -5.891 4* -5.15 7.446 ** 5.45 -1.774 ** -2.88 -2.867 ** -5.03 -0.478 ** -2.30 Large 14.941 ** 9.69 -10.576 ** -6.14 3.093 1.51 -5.231 * -5.61 -5.928 ** -6.92 -0.719 ** -2.29 Age -0.019 -0.79 0.018 0.66 0.009 0.28 0.015 1.04 -0.028 ** -2.06 -0.004 -0.89 Foreign ownership 1.714 1.53 2.825 ** 2.23 -8.055 ** -5.32 -0.350 -0.51 1.269 ** 2.01 0.314 1.37 Semi-skilledworkers -0.015 -0.50 -0.002 -0.06 0.118 ** 3.18 -0.017 -0.98 0.024 1.53 0.003 0.59 Less skilled workers -0.019 -0.64 0.004 0.11 0.150 ** 4.07 -0.023 -1.35 -0.003 -0.17 0.003 0.49 Training 0.885 1.08 1.902 ** 2.20 -2.342 ** -2.30 0.653 1.36 -0.439 -1.03 -0.037 -0.23 R&D -0.066 ** -2.89 0.125 ** 5.26 -0.067 ** -2.37 0.002 0.15 0.077 ** 6.53 -0.007 -1.51 Technology transfer -2.166 ** -2.11 1.368 1.24 -1.947 -1.49 0.768 1.26 1.885 ** 3.45 -0.129 -0.63 Maquila 1.139 1.23 -2.126 ** -2.05 -0.649 -0.53 -0.024 -0.04 0.447 0.87 0.114 0.60 Export oriented -0.325 -0.17 -2.542 -1.24 1.650 0.67 0.268 0.24 0.634 0.62 0.692 1.84 Union -0.568 -0.56 2.771 ** 2.50 -2.537 * -1.92 0.432 0.71 -0.588 -1.07 -0.209 -1.03 Joint activities 0.036 0.62 -0.003 -0.05 -0.026 -0.36 0.048 1.41 -0.025 -0.81 0.001 0.07 Industry specific Food, beverages, tobacco -1.225 -0.95 4.310 ** 2.96 -0.983 -0.57 -2.216 ** -2.83 0.075 0.10 -0.368 -1.39 Textiles, clothing, leathe -0.466 -0.35 -6.890 ** 4.63 8.701 *4 4.89 -0.449 -0.56 0.297 0.40 -0.246 -0.91 Wood,woodproducts 0.010 0.01 -11.397 ** -5.08 18.706 ** 6.98 -2.379 ** -1.97 -1.255 -1.12 -0.350 -0.86 Paper, paper products -8.078 ** -4.76 -3.102 -1.62 1.027 0.45 4.778 ** 4.64 4.727 ** 4.96 -0.092 -0.27 Non-metallicminerals 2.137 1.11 -0.784 -0.36 -4.324 * -1.67 -2.937 *4 -2.52 0.632 0.58 0.045 0.11 Basicmetalindustries 1.513 0.67 -3.788 -1.49 6.047 4* 1.99 -2.544 * -1.86 1.410 1.11 -0.147 -0.32 Metal prod., machinery 1.648 1.30 -10.233 ** -7.24 10.021 * 5.94 -0.643 -0.84 2.324 ** 3.31 0.546 4' 2.12 Othermanufacturingind. -2.131 -0.70 -1.974 -0.58 5.967 1.46 0.222 0.12 -0.491 -0.29 0.260 0.42 Exports -0.071 ** -3.05 -0.047 ** -2.02 0.076 ** 2.77 -0.015 -1.09 -0.052 ** -4.52 -0.009 *4 -2.10 Region: Central -0.973 -1.02 -1.797 * -1.67 1.048 0.81 0.122 0.21 0.890 * 1.66 0.047 0.24 South 5.522 ** 3.42 -3.939 ** -2.16 -2.741 -1.26 -0.396 -0.40 -1.101 -1.21 -0.245 -0.74 Capital -1.934 * -1.71 -3.935 ** -3.08 5.012 4' 3.28 0.287 0.42 0.312 0.49 -0.378 -1.63 Constant 22.391 ** 6.87 26.997 ** 7.65 20.481 ** 4.90 8.831 ** 4.55 5.541 ** 3.17 1.042 1.61 Log likelihood 31526.285 -31924.904 -33021.72 -28043 063 -27286.233 -20792.54 * Significant at 10% level; "4Significant at 5% level. Notes: I. Dependent variable = share of technology use in the production process. 2. Number of observations= 6,586; number of groups = 3,293. 82 Table A3.5. Tobit Results for TI in Manufacturinn Firms, 1995-1999 Manual Automatic Machinery NCM CNCM Robots Explanatory equipment equipment tools Variables Coeff. Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-St. Firm-specific Size: Medium -2.779 -1.62 4.157 1.92 -2.804 -1.28 -0.737 -0.76 1.044 0.88 0.098 0.25 Large -5.122 * -2.91 8.350 * 3.75 -8.558 ' -3.83 0.634 0.64 3.433 * 2.81 0.578 1.44 Age -0.006 -0.23 0.008 0.23 0.009 0.25 0.012 0.81 -0.016 -0.82 -0.008 -1.35 Foreign ownership -0.332 -0.22 -0.906 -0.47 -1.561 -0.80 -1.732 * -2.03 0.816 0.76 1.779 ** 5.10 Semi-skilled workers -0.118 * -2.24 0.100 1.51 0.017 0.26 -0.038 -1.26 0.037 1.01 -0.006 -0.46 Less skilled workers -0.112 * -2.16 0.066 1.02 0.087 1.30 -0.042 -1.41 -0.010 -0.27 -0.003 -0.27 Training -3.171 * -2.14 3.844 W 2.05 -4.337 * -2.28 0.281 0.33 2.743 * 2.66 0.193 0.56 R&D -0.073 * -3.03 -0.012 -0.40 -0.031 -1.00 0.044 ** 3.15 0.068 ** 4.08 -0.001 -0.19 Technology transfer -0.492 -0.61 -0.787 -0.77 0.089 0.09 0.723 1.56 -0.331 -0.59 0.208 1.10 Maquila 0.085 0.07 0.064 0.04 -0.805 -0.51 0.139 0.20 0.809 0.94 -0.257 -0.91 Export oriented 3.297 * 1.86 -0.842 -0.38 -2.160 -0.96 -0.541 -0.55 0.424 0.34 0.825 ** 2.03 Union -0.084 -0.05 4.099 * 2.02 4.772 * -2.35 2.863 * 3.21 -0.930 -0.83 -0.204 -0.56 Joint activities -0.009 -0.15 0.063 0.80 -0.085 -1.06 -0.012 -0.33 0.023 0.52 0.041 ** 2.83 Industry specific Food, beverages, tobaw -0.586 -0.31 5.664 5* 2.34 -3.795 -1.59 -1.732 * -1.66 -0.310 -0.23 0.178 0.42 Textiles, clothing, leathe 0.500 0.29 -0.537 -0.25 0.866 0.40 -2.021 ** -2.13 0.524 0.44 0.163 0.42 Wood, wood products 6.598 t 2.23 -7.672 *o -2.05 5.367 1.45 -2.670 * -1.65 -1.430 -0.70 -0.399 -0.61 Paper, paper products -10.169 ' -4.38 1.704 0.58 -1.135 -0.39 0.507 0.40 6.311 ** 3.91 0.569 1.10 Non-metallic minerals -0.884 -0.33 -3.956 -1.16 1.162 0.35 -2.964 ** -2.02 3.329 * 1.78 1.473 ** 2.46 Basic metal industries -0.583 -0.18 -2.155 -0.51 3.983 0.96 -1.341 -0.74 -0.384 -0.17 -0.546 -0.73 Other manufacturing in -0.268 -0.07 -1.971 -0.38 2.846 0.55 4.078 t 1.81 -3.637 -1.27 -0.489 -0.53 Exports -0.031 -0.79 -0.202 -4.14 0.179 3.72 0.016 0.77 0.015 0.54 0.020 ** 2.36 Region: Central -0.908 -0.69 -2.1S8 -1.32 1.963 1.20 0.497 0.70 -0.365 -0.40 0.519 * 1.78 South 3.120 1.40 -7.987 * -2.82 6.432 * 2.31 -0.804 -0.66 -4.269 5* -2.75 0.050 0.10 Capital -2.612 -1.74 -2.198 -1.16 3.226 1.72 0.711 0.87 0.006 0.01 -0.073 -0.22 Constant 40.224 7.50 17.318 * 2.56 35.211 ** 5.15 6.380 ** 2.11 2.260 _ 0.61 0.107 009 Log likelihood -16135.422 -16935.163 -16993.677 -14227.418 -14895.218 - 1167.327 * Significant at 10%/a level; *Significant at 5% level. Notes: 1. Dependent variable - share of technology use in the production process. 2. Metal products, machinery, and equipment industry was dropped due to collinearity. 3. Number of observations = 3,419; number of groups = 1,717. 83 Table A3.6. Tobit Results for TI in Manufacturing Firms, 1992-1999 Manual Automatic Machinery tools NCM CNCM Robots Explanatory equipment equipment Variables Coeff. Z-St. Coeff Z-St. Coeff Z-St. Coeff Z-SL Coeff Z-St. Coeff Z-St. Firm-specific Size: Medium -5.623 * -2.88 4.068 1.56 -1.832 -0.69 0.964 0.77 2.580 ** 1.97 -0.270 -0.61 Large -6.505 * -3.27 7.775 ** 2.92 -6.624 ** -2.45 2.100 * 1.65 3.603 ** 2.69 0.176 0.39 Age 0.001 0.05 0.024 0.62 0.011 0.28 0.015 0.84 -0.032 * -1.65 0.002 0.24 Foreign ownership 2.421 1.63 -1.552 -0.78 -0.958 -0.47 -1.227 -1.30 -0.275 -0.27 0.541 1.61 Semi-skilled workers -0.041 -0.57 0.092 0.96 -0.038 -0.39 -0.106 ** -2.23 0.026 0.53 0.014 0.84 Less skilled workers -0.045 -0.63 0.081 0.85 0.009 0.09 -0.102 ** -2.16 -0.008 -0.17 0.009 0.56 Training -0.037 -0.03 4.470 ** 2.50 -2.116 -1.16 -0.992 -1.13 1.602 * 1.78 -0.253 -0.81 R&D -0.033 -1.27 0.055 1.62 -0.066 * -1.88 0.014 0.80 0.100 ** 5.86 -0.005 -0.85 Technology transfer -1.549 -1.43 -1.222 -0.85 -0.201 -0.14 0.115 0.16 2.074 * 2.88 -0.136 -0.54 Maquila -0.032 -0.03 -1.687 -1.02 2.581 1.53 -0.164 -0.21 -0.048 -0.06 -0.075 -0.27 Export oriented -1.536 -0.76 0.171 0.06 -0.856 -0.31 -0.844 -0.65 1.858 1.37 1.784 * 3.86 Union 4.854 * 2.82 -1.898 -0.82 -5.930 ** -2.53 3.051 ** 2.77 0.063 0.05 -0.373 -0.95 Joint activities -0.009 -0.14 0.082 0.97 -0.088 -1.02 -0.010 -0.24 0.048 1.14 0.045 ** 3.06 Industry specific Food, beverages, tobacco 1.139 0.61 6.082 ** 2.41 -3.699 -1.46 -2.707 * -2.32 -0.111 -0.09 -0.109 -0.26 Textiles, clothing, leather -0.047 -0.03 -1.249 -0.55 4.730 ** 2.06 -0.372 -0.35 -0.474 -0.42 -0.414 -1.10 Wood, wood products 7.418 ** 2.62 -6.518 * -1.69 6.420 * 1.65 -1.783 -1.00 -1.786 -0.93 -0.542 -0.85 Paper, paper products -7.701 * -3.45 -1.888 -0.62 -0.067 -0.02 0.564 0.40 7.818 ** 5.16 0.253 0.50 Non-metallic minerals -0.213 -0.08 4.105 1.12 -3.277 -0.89 -4.801 ** -2.84 2.057 1.12 0.158 0.26 Basic metal industries 2.517 0.86 -4.407 -1.11 6.577 1.64 -2.287 -1.23 -0.966 -0.49 -0.430 -0.65 Othermanufacturingind. -4.275 -1.19 12.622 ** 2.59 -2.560 -0.52 -1.618 -0.71 -3.892 -1.60 -0.887 -1.09 Exports 0.016 0.44 -0.217 ** 4.33 0.199 * 3.95 0.002 0.07 0.024 0.98 0.008 0.94 Region: Central -2.607 ** -1.98 -2.408 -1.35 4.232 * 2.35 -0.051 -0.06 0.261 0.29 0.205 0.69 South 0.937 0.38 -9.704 ** -2.90 7.343 ** 2.18 -0.257 -0.17 -1.273 -0.76 0.050 0.09 Capital -1.174 -0.77 -4.237 ** -2.04 4.606 * 2.21 -0.062 -0.06 0.436 0.42 -0.196 -0.57 Constant 25 029 ** 3.49 20.594 * 2.18 37.265 ** 3.85 14.040_** 3.01 0.639 0.13 -0.028 -0.02 Log likelihood -14657.138 15518.031 -15599.636 -13326.971 -13362.579 -10037.887 * Significant at 10% level; **Significant at 5% level. Notes: 1. Dependent variable = share of technology use in the production process. 2. Metal products, machinery, and equipment industry was dropped due to collinearity. 3. Number of observations= 3,155; number of groups = 1,066. 84 Annex 4: Transition Matrix Table A4.1. Transition Matrix for Firms in Year t and t+n, 1992-1999 1992 1999 Activity % of firms Tdeohpntiolng Training % Exports % from the total . Continue Stop Continue Stop Continue Stop No Technology Adoption 27.00 70.28 29.72 Technology Adoption 73.17 78.27 21.73 No Training 18.00 60.78 39.22 Training 82.00 93.07 6.93 No Exports 95.40 3.64 96.36 Exports 4.60 71.43 28.57 Technology Adoption Given no training 6.80 65.52 34.48 Given training 41.20 67.80 32.20 Given no training and no exports 6.47 65.94 34.00 Given training and no exports 34.00 67.66 32.34 Given exports but no training 0.42 57.14 42.86 Given training and exports 1.30 71.43 28.57 No Technology Adoption Given no training 2.00 28.00 72.00 Given training 6.70 51.41 48.59 Given no training and no exports 1.90 26.83 73.17 Given training and no exports 6.30 49.63 50.37 Given exports but no training 0.09 50.00 50.00 Given training and exports 0.32 85.71 14.29 Training Given no adoption and no exports 1.90 88.89 11.11 Given adoption and no exports 0.00 Given exports but no adoption 0.32 0.00 100.00 Given training and exports 1.30 92.86 7.14 Exports Given no adoption and no training 0.00 Given no adoption and training 0.32 78.57 21.43 Given adoption and training 1.31 50.00 50.00 Given adoption 1.60 77.14 22.86 _ Source: Author's calculations based on ENESTYC 92 and 99. 85 VOL. II APPENDIX A INEGI has compiled the National Survey of Employment, Salaries, Technology, and Training (ENESTYC). The Ministry of Labor co-designed the questionnaire, which gathered rich information on training, technology, wages, employment, forms of labor contracting, and internal plant organization of Mexican manufacturing firms. The government conducted the survey in 1992, 1995, and 1999, but its questions on technology ask whether the firm adopted technology in the periods 1989-1992, 1994-1995, or 1997-1999, respectively. Our references to the time of technology adoption mention only the final year of the period (e.g. 1992 rather than 1989-1992). Data from the 1992 survey includes 5,071 firms, from the 1995 survey includes 5,242 firms, and from the 1999 survey includes 7,429 firms. A valuable feature of ENESTYC is that it allows us to identify the same firm in 1992, 1995, and 1999. Nonetheless, we should qualify our estimations with survivor bias. Only firms that exist in all three years can be included in the panel database. As Audretsch (1995) shows, survival likelihood is strikingly low for small and new enterprises and increases with firm size and age. Thus, the panel includes an unrepresentatively high number of large and mature firms. While random observation selection should not cause bias in our resulting estimations, surviving firms are not randomly selected. Darwinian selection of extant firms means that the firms in our sample tend to be more efficient and have better performance than an average Mexican firm. Another advantage of this database is the broad spectrum of firm sizes included by industry, shown in tables B.1-B.3. The rich information available in ENESTYC allows us to distinguish technology diffusion policies for firms of different size and character. INEGI also conducts the Annual Industrial Survey (EIA). The survey covers 6,500 manufacturing plants throughout Mexico that account for 80 percent of production in each industry group. Since the survey attempts to cover the majority of manufacturing production but not a majority of plants in all categories, our sample includes all large plants and most medium- sized scale plants, but few small-scale plants and very few microenterpise plants. 86 We link the ENESTYC panels to firms in the EIA. This allows us to combine EIA data on productivity, labor, value-added, and capital with ENESTYC variables for the plants common to both surveys. The panels also include some regional variables using the Indicators of Scientific and Technology Activity in Mexico from the National Council of Science and Technology (CONACYT). A description of the variables in the panels appears in the Appendix. The 1992-95 panel has 3,293 firms, the 1995-99 panel has 1,717 firms, and the 1992-99 panel has 1,066 firns. The information on individual establishments that INEGI gathers through its questionnaires (which law requires firms to answer) is legally confidential, and INEGI is unable to give the raw data to outside agencies. Therefore, we followed an established procedure in which most data analysis was done in INEGI's Aguascalientes headquarters with the support of INEGI personnel. Nevertheless, the reader should bear in mind the limitations on data analysis imposed by this institutional arrangement. 87 VOL. 11 APPENDIX B Table B.1. Manufacturing Firms in the 1992-1995 Panel by Industry and Size Size Division All Large Medium Small Micro Total 3,293 352 576 1,099 1,266 Food, beverage and tobacco 669 105 114 163 287 Textiles, clothing, leather 551 36 93 231 191 Wood and wood products 149 28 42 61 18 Paper and paper products 219 16 31 103 69 Chemical products 494 40 94 185 175 Non-metallic minerals 161 45 31 25 60 Basic metal industries 102 13 13 39 37 Metal products, machinery 897 65 147 272 413 Other manufacturing industries 51 4 11 20 16 Source: 1992-95 ENESTYC Panel. Table B.2. Manufacturing Firms in the 1995-1999 Panel by Industry and Size Size Division All Large Medium Small Micro Total 1,717 829 737 145 6 Food, beverage and tobacco 372 232 114 26 Textiles, clothing, leather 273 133 113 23 4 Wood and wood products 57 19 32 6 Paper and paper products 146 54 83 9 Chemical products 306 126 153 26 1 Non-metallic minerals 75 32 33 10 Basic metal industries 41 21 15 5 Metal products, machinery 419 198 183 37 1 Other manufacturing industries 28 14 11 3 Source: 1995-99 ENESTYC Panel. Table B.3. Manufacturing Firms in the 1992-1999 Panel by Industry and Size Size Division All Large Medium Small Micro Total 1,066 554 439 72 1 Food, beverage and tobacco 227 154 63 10 Textiles, clothing, leather 162 70 80 12 Wood and wood products 36 9 19 8 Paper and paper products 95 36 52 7 Chemical products 190 86 87 16 1 Non-metallic minerals 46 34 10 2 Basic metal industries 36 18 18 Metal products, machinery 257 138 102 17 Other manufacturing industries 17 9 8 Source: 1992-99 ENESTYC Panel. 88 VOL. 11 APPENDIX C 1992-99 Panel Variables Description Variable Description Value From the ENESTYC Firm size according to the number of workers: Micro I - 15 Dummy for each size Firm size Small 16 - 100 1= if the firm belongs to a certain size Medium 101 -250 0= otherwise. Large 250 - more Manufacturing industries: I) Food, beverages, and tobacco 2) Textiles, clothing, and leather 3) Wood and wood products Dummy for each industry Division 4) Paper, pals products, printing, and publishing 1= if the firm belongs to a certain industry 5) Chemicals, oil derivatives, and coal 0= otherwise. 6) Non-metallic mineral products 7) Basic metallic industries 8) Metallic products, machinery, and equipment 9) Other manufacturing industries Total workers Number of workers in the firm. Continuous Regions: Dummies Includes the states of Baja Califomia, Baja North California Sur, Coahuila, Chihuahua, Durango, I= if the firm is located in the North, Nuevo Le6n, Sinaloa, Sonora, Tamaulipas, and 0= otherwise. Zacatecas. Includes the states of: Aguascalientes, Colima, Center Guanajuato, Hidalgo, Jalisco, Mexico, I= if the firm is located in the Center, Michoacan, Morelos, Nayarit, Puebla, Queretaro, 0= otherwise. San Luis Potosi, and Tlaxcala. Includes the states of Campeche, Chiapas, I= if the firm is located in the South, South Guerrero, Oaxaca, Quintana Roo, Tabasco, 0= otherwise. Veracruz, and Yucatan. 1= if the firm is located in the Capital, Capital Distrito Federal 0= otherwise. Years Firm's age. Continuous Dummy Foreign capital Percentage of foreign capital in the firm. I= if foreign capital in the firm > 50%/o, O0= otherwise. Dummy Subsidiary Subsidiary firm. I= if the firm is a subsidiary, 0= otherwise. Joint activities Number of firms with joint activities. Continuous R&D Firm's investment in R&D (it does not include Continuous technology transfer or equipment acquisition). Technology transfer Firm's investment in technological transfer. Continuous Categorical 0= No technology Type of technology that the firm adopts: 1= Manual equipment 1) Manual equipment 2= Automatic equipment 2) Automatic equipment 3= Machinery tools Technology type 3) Machinery tools 4= Numeric controlled machinery 4) Numeric controlled machinery 5= Computerized numeric controlled machinery 5) Computerized numeric controlled machinery 6= Robots 6) Robots Dummies 1= if the firm adopts a certain type of technology, 0= otherwise. 89 Dummy Technological Intensity in the use of a certain type of I= if the use in production of a certain type of intensity technology. technology > 40%/o, 0= otherwise. Highly skilled workers Number of executives and managers in the firm. Continuous Semi-skilled workers Number of production workers in the firm. Continuous Unskilled workers Number of general workers in the firm. Continuous Share of highly skilled Share of highly skilled workers from the total of Ranks between 0-100 workers workers in the firm. Share of semi-skilled Share of semi-skilled workers from the total of Ranks between 0-100 workers workers in the firm. .._ ____ Share of unskilled Share of unskilled workers from the total of Ranks between 0-100 workers workers in the firm. Total hours Total number of hours worked in the firm. Continuous New hires New hires. Continuous Laid offs Dismissals. Continuous Net employment New hires less dismissals. Continuous Total wages Total wages in real pesos paid in the firm. Continuous Highly skilled wages Total wages in real pesos paid to the highly Continuous _______shlled_wages skilled workers in the firm. Contmuous Semi-skilled wages Total wages in real pesos paid to the semi-skilled Continuous workers in the firm. Unskilled wages Total wages in real pesos paid to the unskilled Continuous ______ _____ _____ workers in the firm. Share of highly skilled Share of the highly skilled workers wages from Ranks between 0-100 wages the firm's total wages. Share of semi-skilled Share of the semi-skilled workers wages from the Ranks between 0-100 wages firm's total wages. Share of unskilled Share of the unskilled workers wages from the Ranks between 0-100 wages firm's total wages. Dummy Training Training for workers. 1= if firm provides training, . , .______________ 0= otherwise. Dummy Source of training Source of the training that the firm provides. 1= if the training comes from the public sector, 0= otherwise. Dummy Union Existence of a union in the firm. 1= if a union exists, .___ ____ ____ _ 0= otherwise. Source of R&D: 1) Consulting firms 2) Public institutions Dummy for each source Source of R&D 3) Private institutions 1= if the firm's R&D is from a certain source, 4) A firm's department otha than R&D 5) Own firm's R&D department _ Sales Firm's total sales in current pesos. Continuous Dummy Exports Firm's market orientation I= if foreign sales > 50%, 0= otherwise. Dummy Maquila Firms dedicated to maquila activities. I= if maquila 0= otherwise. Productivity Firm's productivity measured as output per Continuous ___________________ worker. Contmuous Time Indicator variable of the timing of technology Ranks between -2 and +2 _____ ____ ____ ____ adoption. Technology diffusion Proportion of firms that adopted technology in a Ranks between 0-100 rate given year. From the EM Value added Firm's value added. Continuous Capital assets Firm's capital: fixed assets, not deflated. Continuous 90 TFP Firn's total factor productivity. Continuous Industry R&D Percentage of R&D expenditure from total Ranks between 0-100 expenditures, by industry. _____________________ Industry exports Percentage of exports from total sales, by Ranks between 0-100 industry. Regional variables from CONACYT Science graduates Percentage of individuals with a degree in Science Continuous from the total population, by state. Graduates Percentage of individuals that got a degree from Continuous the total population, by state. Percentage of researchers registered in both Researchers federal and state systems from the total Continuous population, by state. Public R&D per capita Federal expenditure in R&D per capita, by state. Continuous Public R&D Percentage of the federal expenditure in R&D Continuous from the total federal expenditure, by state. 91