- A second complication when doing the overall simulation is related to the fact that the biadic innovations of the firms in our sample are only 58% of the total biadic innovations in 1997--2011. This is because we do not have weights for all the firms that do relevant automation or machinery innovation in our sample period. Therefore, the number of inventions in a country in a given year consists of an in-sample count plus an outof -sample count. We make the assumption that the firms not in our sample respond in the same way as the firms in our sample. Hence, when computing the countrylevel innovation counts by assigning simulated innovations to countries using the firmsâ inventor weights, we assume that the ratio of the in-sample count to the out-of-sample count stays constant. That way if the in-sample simulated count increases by, say, 5% the entire count would increase by the same amount.
Paper not yet in RePEc: Add citation now
Abrams, D., Akcigit, U., and Grennan, J. (2018). Patent value and citations: Creative destruction or strategic disruption? NBER Working Paper 19647.
Acemoglu, D. and Autor, D. (2011). Skills, Tasks and Technologies: Implications for Employment and Earnings. Handbook of Labor Economics, 4(B).
Acemoglu, D. and Finkelstein, A. (2008). Input and technology choices in regulated industries: Evidence from the health care sector. Journal of Political Economy, 116(5):837â880.
Acemoglu, D. and Linn, J. (2004). Market size in innovation: Theory and evidence from the pharmaceutical industry. Quarterly Journal of Economics, 119(3):1049â1090.
Acemoglu, D. and Restrepo, P. (2017). Robots and jobs: Evidence from US labor markets. NBER Working Paper 23285.
Acemoglu, D. and Restrepo, P. (2018a). Demographics and automation. NBER Working Paper 24421.
Acemoglu, D. and Restrepo, P. (2018b). The race between machine and man: Implications of technology for growth, factor shares and employment. American Economic Review, 108(6):1488â1542.
Acemoglu, D. U. (2010). When Does Labor Scarcity Encourage Innovation? Journal of Political Economy, 118(6):1037â1078.
AdaÌo, R., KolesaÌr, M., and Morales, E. (2019). Shift-share designs: Theory and inference.
Aghion, P., BeÌnabou, R., Martin, R., and Roulet, A. (2019a). Environmental preferences and technological choices: Is market competition clean or dirty? Aghion, P., Bergeaud, A., Boppart, T., Klenow, P., and Li, H. (2019b). A theory of falling growth and rising rents.
Aghion, P., DechezlepreÌtre, A., HeÌmous, D., Martin, R., and Van Reenen, J. (2016). Carbon taxes, path dependency, and directed technical change: Evidence from the auto industry. Journal of Political Economy, 214(1):1â51.
Aghion, P., Jones, B., and Jones, C. (2017). Artificial intelligence and economic growth. NBER Working Paper 23928.
Akicigit, U. and Ates, S. (2019). Ten facts on declining business dynamism and lessons from endogenous growth theory. NBER working paper 25755.
Alesina, A., Battisti, M., and Zeira, J. (2018). Technology and labor regulations: theory and evidence. Journal of Economic Growth, 23(1):41â78.
- Aschhoff, B., Crass, D., Cremers, K., Grimpe, C., Rammer, C., Brandes, F., DiazLopez, F., Klein Woolthuis, R., Mayer, M., and Montalvo, C. (2010). European competitiveness in key enabling technologies. Technical report, Centre for European Economic Research (ZEW) and TNO.
Paper not yet in RePEc: Add citation now
Autor, D., Katz, L., and Krueger, A. (1998). Computing Inequality: Have Computers Changed the Labor Market? Quarterly Journal of Economics, 113(4):1169â1213.
Autor, D., Levy, F., and Murnane, R. (2003). The Skill Content of Recent Technological Change: An Empirical Exploration. The Quarterly Journal of Economics, 118(4):1279â1333.
Bartel, A., Ichniowski, C., and Shaw, K. (2007). How Does Information Technology Really Affect Productivity? Plant-Level Comparisons of Product Innovation, Process Improvement and Worker Skills. Quarterly Journal of Economics, 122(4):1721â1758.
Bartik, T. (1991). Who Benefits from State and Local Economic Development Policies? W.E. Upjohn Institute.
- Bena, J. and Simintzi, E. (2019). Machines could not compete with Chinese labor: Evidence from U.S. firmsâ innovation. mimeo.
Paper not yet in RePEc: Add citation now
- Bessen, J. (2016). How computer automation affects occpations: Technology, jobs and skills.
Paper not yet in RePEc: Add citation now
Blanas, S., Gancia, G., and Lee, S. Y. (2018). Who is afraid of machines? Blundell, R., Griffith, R., and Van Reenen, J. (1999). Market share, market value and innovation in a panel of British manufacturing firms. The Review of Economic Studies, 66(3):529â554.
Burda, M. and Seele, S. (2016). No role for the Hartz reforms? Demand and supply factors in the German labor market, 1993-2014. SFB 649 Discussion Paper 2016-010.
Calel, R. and DechezlepreÌtre, A. (2016). Environmental policy and directed technical change: Evidence from the European carbon market. The Review of Economics and Statistics, 98(1).
Clemens, M., Lewis, E., and Postel, H. (2018). Immigration restrictions as active labor market policy: Evidence from the Mexican bracero exclusion. American Economic Review, 108(6):1468â87.
- Coefficient: 0.56 â.1 â.05 0 .05 .1 Lowâskill wages (patents based residuals) â.1 â.05 0 .05 .1 Lowâskill wages (exports based residuals) (b) Residualized low-skill wages Figure B.XI: Foreign low-skill wages for each country computed either with patent-based weights or with trade-based weights. Wages are computed for the years 1995-2009.
Paper not yet in RePEc: Add citation now
Coelli, F., Moxnes, A., and Ulltveit-Moe, K. (2016). Better, faster, stronger: Global innovation and trade liberalization. NBER Working Paper 22647.
Cohen, W., Nelson, R., and Walsh, J. (2000). Protecting their intellectual assets: Appropriability conditions and why U.S. manufacturing firms patent or not. NBER Working Paper 7552.
de Rassenfosse, G., Dernis, H., Guellec, D., Picci, L., and van Pottelsberghe de la Potterie, B. (2013). The worldwide count of priority patents: A new indicator of inventive activity. Research Policy, 42(3):720â737.
DechezlepreÌtre, A., Glachant, M., Hascic, I., Johnstone, N., and MeÌnieÌre, Y. (2011). Invention and transfer of climate change-mitigation technologies: a global analysis. Review of Environmental Economics and Policy, 5:109â130.
DechezlepreÌtre, A., MeÌnieÌre, Y., and Mohnen, M. (2017). International patent families: from application strategies to statistical indicators. Scientometrics, 111(2):793â828.
- Dependent Variable Auto95 Domestic + Foreign Foreign (1) (2) (3) (4) (5) (6) Low-skill wage-0.6971-1.3102-2.2759 1.4820 2.4995 0.9875 (1.3783) (1.2749) (1.6483) (2.0809) (2.2640) (2.2871) High-skill wage 0.7190-0.5961-0.3672-4.0906-2.3516-4.4106* (1.3025) (1.7655) (1.3453) (2.5322) (2.5102) (2.6414) GDP gap 0.0407 0.0427 0.0385 0.0529 0.0481 0.0532 (0.0374) (0.0376) (0.0377) (0.0559) (0.0555) (0.0559) Labor productivity 3.0468-2.6255 (1.9773) (1.6419) GDP per capita 3.9620** 0.9242 (1.9791) (1.9811) Control variables stock + spill stock + spill stock + spill stock + spill stock + spill stock + spill Fixed effects F + CY F + CY F + CY F + CY F + CY F + CY Observations 61170 61170 61170 61170 61170 61170 Firms 4187 4187 4187 4187 4187 4187 Panel B: Independent variables are lagged by 15 years.
Paper not yet in RePEc: Add citation now
- Dernis, H. and Guellec, D. (2001). Using patent counts for cross-country comparisons of technology output. STI-Science Technology Industry Review, 27:129â146.
Paper not yet in RePEc: Add citation now
Dickens, R., Machin, S., and Manning, A. (1999). The effects of minimum wages on employment: Theory and evidence from Britain. Journal of Labor Economics, 17(1).
Dolado, J., Kramarz, F., Machin, S., Manning, A., Margolis, D., and Teulings, C. (1996). The economic impact of minimum wages in Europe. Economic Policy, 11(23).
Eaton, J. and Kortum, S. (1996). Trade in ideas: patenting and productivity in the OECD. Journal of International Economics, 40:251â278.
Eaton, J., Kortum, S., and Kramarz, F. (2011). An anatomy of international trade: evidence from French firms. Econometrica, 79(5):1453â1498.
- Figure B.X plots the patent-based weights against the trade-based weights. Panel (b) focuses on a few origin countries while Panel (a) plots all countries together. We find a strong correlation between the two measures with a regression coefficient of 0.94 (when observations are weighted by the trade flow in 1996). 50 To do that we use a fractional approach: each patent is allocated NACE sectoral weights (and machinery weights) depending on the share of IPC codes associated with a NACE sector or machinery. Coefficient: 0.42 2.4 2.6 2.8 3 3.2 3.4 Lowâskill wages (patents based) 2.2 2.4 2.6 2.8 3 3.2 Lowâskill wages (exports based) (a) Low-skill wages.
Paper not yet in RePEc: Add citation now
- Figures For Paper (a) For all C/IPC 6 digit codes (b) For C/IPC 6 digit codes in machinery with at least 100 patents Figure I: Histogram of the prevalence of automation keywords for C/IPC 6 digit codes (a) Example with keywords (b) Example without keywords Figure II: Examples of automation patents from technological code B65G1, which are both automated storage cabinets. 0 .06 .12 .18 .24 .3 1980 1985 1990 1995 2000 2005 2010 2015 year auto95 auto90 (a) Share of automation patents in machinery worldwide. 0 .05 .1 .15 .2 .25 .3 .35 1980 1985 1990 1995 2000 2005 2010 2015 year United States Germany France United Kindom Japan (b) Share of automation patents (auto95) in machinery by applicantâs nationality.
Paper not yet in RePEc: Add citation now
- Fixed effects F + Y F + Y F + Y F + CY F + CY F + CY F + CY F + CY F + CY Note:Marginal effects; P-values in parentheses. The independent variables are lagged by two periods. Estimation is by conditional Poisson regressions fixed-effects (HHG). Columns (1)-(3) include firm fixed effects and year dummies. Columns (4)-(9) include firm and country-year fixed effects. Columns (7)-(9) use the log foreign components of the macro variables interacted with the share of the foreign macro variable in the total macro variable at the beginning of the sample. All regressions include controls for stocks and spillovers. P-values are computed by sampling with replacement the entire path of macroeconomic variables for each firm with 1000 draws.* p < 0.1; ** p < 0.05; *** p < 0.01 1
Paper not yet in RePEc: Add citation now
- Fixed effects F + Y F + Y F + Y F + CY F + CY F + CY F + CY F + CY F + CY Panel B: skill premium Low-skill/ High-skill wages 1.9423* 2.0420* 1.9000* 2.1995* 2.0520* 2.2870** 3.5089*** 3.4205*** 3.5000*** [0.074] [0.059] [0.06] [0.055] [0.063] [0.048] [0.004] [0.005] [0.004] GDP gap Y Y Y Y Y Y Y Y Y Labor productivity N Y N N Y N N Y N GDP per capita N N Y N N Y N N Y Control variables stocks stocks stocks stocks stocks stocks stocks stocks stocks + spill. + spill. + spill. + spill. + spill. + spill. + spill. + spill. + spill.
Paper not yet in RePEc: Add citation now
- Further, workers have 6 weeks off and the standard work week is 38 hours. Consequently we calculate the hourly minimum wages as monthly minimum wageÃ14/ [(52 â 6) Ã 38], which in the case of 2009 is 5.83 euros per hour. We perform similar calculations, depending on individual work conditions, for other countries with minimum wages that are not stated per hour: Belgium, Brazil, Israel, Mexico, Netherlands, Poland and Portugal. For the US, we use data from FRED for state minimum wages and calculate the nation-level minimum wage as the weighted average of the state-by-state maximum of state minimum and federal minimum wages, where the weight is the manufacturing employment in a given state.
Paper not yet in RePEc: Add citation now
Gaggl, P. and Eden, M. (2018). On the welfare implications of automation. Review of Economic Dynamics, 29:15â43.
Gaggl, P. and Wright, G. (2017). A short-run view of what computers do: Evidence from a UK tax incentive. American Economic Journal: Applied Economics, 9(3):262â94.
Gentzkow, M. and Shapiro, J. (2010). What drives media slant? evidence from U.S. daily newspapers. Econometrica, 78(1):35â71.
Graetz, G. and Michaels, G. (2018). Robots at work. Review of Economics and Statistics, 100(5):753â768.
- Habakkuk, J. (1962). American and British Technology in the Nineteenth Century. Cambridge University Press.
Paper not yet in RePEc: Add citation now
Hanlon, W. (2015). Necessity is the mother of invention: Input supplies and directed technical change. Econometrica, 83(1):67â100.
Harhoff, D., Scherer, F., and Vopel, K. (2003). Citations, family size, opposition and the value of patent rights. Research Policy, 32:1343â1363.
Hausman, J., Hall, B., and Griliches, Z. (1984). Econometric models for count data with an application to the patents-R&D relationship. Econometrica, 52(4):909â938.
- HeÌmous, D. and Olsen, M. (2018). The rise of the machines: Automation, horizontal innovation and income inequality. CEPR dp 10244.
Paper not yet in RePEc: Add citation now
Henderson, R., Jaffe, A., and Trajtenberg, M. (1993). Geographic localization of knowledge spillovers as evidenced by patent citations. Quarterly Journal of Economics, 108(3):578â98.
Hornbeck, R. and Naidu, S. (2014). When the levee breaks: Black migration and economic development in the American South. American Economic Review, 104(3):963â 990.
Jacobi, L. and Kluve, J. (2007). Before and after the Hartz reforms: The performance of active labour market policy in Germany. Journal for Labour Market Research, 40(1):45â64.
Kogan, L., Papanikolaou, D., Seru, A., and Stoffman, N. (2017). Technological innovation, resource allocation, and growth. The Quarterly Journal of Economics, 132(2):665â712.
Krause, M. and Uhlig, H. (2012). Transitions in the German labor market: Structure and crisis. Journal of Monetary Economics, 59(1):64â79.
Krebs, T. and Scheffel, M. (2013). Macroeconomic evaluation of labor market reform in Germany. IMF Economic Review, 61(4).
Krebs, T. and Scheffel, M. (2017). Labor market institutions and the cost of recessions. IMF working paper WP/17/97.
Krusell, P., Ohanian, L., RıÌos-Rull, J.-V., and Violante, G. (2000). Capital-Skill Complementarity and Inequality: A Macroeconomic Analysis. Econometrica, 68(5):1029â 1053.
- Labor productivity 0.6678 0.7340 0.7724 0.1980 0.6519 1 Note: Correlation of residuals for the auto95 sample controlling for year and firm fixed effects. 1
Paper not yet in RePEc: Add citation now
Lewis, E. (2011). Immigration, skill mix and capital skill complementarity. Quarterly Journal of Economics, 126 (2):1029â1069.
- Log patents based weights â12 â8 â4 0 Log exports based weights IT (b) Trade from the 6 largest countries Figure B.X: Bilateral patent flows and trade flows in machinery. Panel (a) plots log patent based weights, which are a weighted average of the destination countryâs weights in the (foreign) patent portfolio of firms from the origin country, against export shares in machinery over the years 1995-2009. The size of each circle represents the product of the GDP of both countries, which is used as a weight in the regression. Panel (b) focuses on the weights from the listed countries and observations are weighted by the GDP of the partner country.
Paper not yet in RePEc: Add citation now
Lordan, G. and Neumark, D. (2018). People versus machines: The impact of minimum wages on automatable jobs. Labour Economics, 52(40-53).
Lybbert, T. and Zolas, N. (2014). Getting patents and economic data to speak to each other: An âAlgorithmic Links with Probabilitiesâ approach for joint analyses of patenting and economic activity. Research Policy, 43:530â542.
- Mann, K. and Puettmann, L. (2018). Benign effects of automation: New evidence from patent texts.
Paper not yet in RePEc: Add citation now
Manuelli, R. and Seshadri, A. (2014). Frictionless technology diffusion: The case of tractors. American Economic Review, 104(4):1368â1391.
Martinez, J. (2018). Automation, growth and factor shares. mimeo.
Newell, R. G., Jaffe, A. B., and Stavins, R. N. (1999). The induced innovation hypothesis and energy-saving technological change. The Quarterly Journal of Economics, 114(3):941â975.
Noailly, J. and Smeets, R. (2015). Directing technical change from fossil-fuel to renewable energy innovation: An application using firm-level patent data. Journal of Environmental Economics and Management, 72:15â37.
- Panel (a) plots log foreign low-skill wages using either patent-based weights or trade-based weights. Panel (b) plots the residuals of foreign wages according to both methods controlling for country and year fixed effects. Observations are weighted by the number of biadic machinery patents by firms from the the country over the years 1997-2011. Another way to summarize how close the two distributions are is to compute what low-skill wages would be according to either sets of weights. We do this in Figure B.XI.
Paper not yet in RePEc: Add citation now
Popp, D. (2002). Induced innovation and energy prices. The American Economic Review, 92(1):160â180.
- sectoral bilateral trade flow from UN Comtrade data between between 1995 and 2009 for 40 countries (Taiwan is not included in the data). To obtain trade flows in machinery, we use the Eurostat concordance table between 4 digit IPC codes and 2 or 3 digits NACE Rev 2 codes (van Looy, Vereyen, and Schmoch, 2014), this concordance table matches IPC codes to the industry of manufacturing. The concordance table assigns a unique industry to each IPC code. Then, for each industry, we compute the share of biadic patents over the period 1995-2009 which are in machinery according to our definition.50 This gives us a machinery weight for each industry code and each country.
Paper not yet in RePEc: Add citation now
- Statistics computed on biadic patents from 1997-2011. 1
Paper not yet in RePEc: Add citation now
- Table VII: Descriptive statistics for firms in our baseline regression Variable Auto95 Auto90 Auto95 Auto90 Automation patents per year 1997-2011 per year 1997-2011 weights Mean 0.7 11.22 0.84 13.24 Largest country 0.47 0.46 Standard deviation 3.46 48.71 4.04 56.76 Second largest 0.17 0.18 p50 0 2 0 3 US 0.21 0.21 p75 0.27 6 0.33 7 Japan 0.17 0.15 p90 1.4 19 1.6 22 Germany 0.2 0.21 p95 3 41 3.27 50 France 0.09 0.09 p99 12 173 13.73 194 UK 0.09 0.09 Number of firms 3341 4903 Note: Summary statistics for the firms used in our baseline regression. 1
Paper not yet in RePEc: Add citation now
- The primary data source for the hourly minimum wage data is OECD Statistics. Not all countries have government-imposed hourly minimum wages. Spain, for instance, had a monthly minimum wage of 728 euros in 2009. To convert this into hourly wage we note that Spain has 14 monthly payments a year (+1 payments in December and July).
Paper not yet in RePEc: Add citation now
- There for each country, we compute âforeign low-skill wagesâ as a weighted average of foreign wages where the weights are either the patent-based weights or the trade-based weights derived above. Foreign wages are deflated with the local PPI and converted in USD in 1995 as in our main analysis. Panel (a) then reports foreign log low-skill wages according to both types of weights in 1995-2009, we find that they are strongly correlated.
Paper not yet in RePEc: Add citation now
Timmer, M., Dietzenbacher, E., Los, B., Stehrer, R., and de Vries, G. J. (2015). An illustrated user guide to the world input-output database: the case of global automotive production. Review of International Economics, 23(3).
- To allocate patents according to their industry of manufacturing (which we use for Table V), we proceed as follows. First, we use the Eurostat concordance table (van Looy, Vereyen and Schmoch, 2014) which maps 4-digit IPC codes to 2 or 3 digit NACE rev 2 sectors to allocate all US machinery patents to sectors fractionally according their C/IPC 49 To interpret the effect of the automation variable, note that the means are 0.13, 0.15 and 0.14 in the 70s, 80s and 90s, and the standard deviations are 0.10, 0.12 and 0.11 with the auto90 definition.
Paper not yet in RePEc: Add citation now
- van Looy, B., Vereyen, C., and Schmoch, U. (2014). Patent statistics: Concordance IPC V8 - NACE REV.2. Technical report, Eurostat.
Paper not yet in RePEc: Add citation now
van Pottelsberghe de la Potterie, B. and van Zeebroeck, N. (2008). A brief history of space and time: The scope-year index as a patent value indicator based on families and renewals. Scientometrics, 75(2):319â338.
- We supplement this data with data from UNSTAT on exchange rates and GDP (and add Taiwan separately from the Taiwanese Statistical office). We calculate the GDP gap as the deviations of log GDP from HP-filtered log GDP using a smoothing parameter of 6.25. Table B.VII provides summary statistics for low-skill and high-skill wages for all our countries for our baseline measure (i.e. manufacturing labor costs deflated by the manufacturing PPI and converted in USD in 1995).
Paper not yet in RePEc: Add citation now
- We then multiply sectoral trade flows (after having aggregated the original data to the NACE Rev 2 codes used in the concordance table) by this weight to get bilateral trade in machinery. We then compute the export share in machinery across destinations. We compute trade based weights for each year in 1995-2009 and take the average (there are a few missing observations for 1995).
Paper not yet in RePEc: Add citation now