MEIDE Conference, HSE, Moscow, June 16-17, 2016 FIRM INNOVATION AND PRODUCTIVITY IN LATIN AMERICA AND THE CARIBBEAN The Engine of Economic Development Carlo Pietrobelli IDB - Competitiveness and Innovation Division OPEN ACCESS carlop@iadb.org https://publications.iadb.org/handle/11319/7690 www.firmsinlatinamerica.com
What is the problem? GDP pc growth is not fast 35000 Latin America & Caribbean China 30000 Korea, Rep. East Asia & Pacific 25000 OECD members 20000 15000 10000 5000 0 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014
Why? Low Productivity Growth Table 1: Growth Accounting: LAC vs Comparison Countries (1960-2011) (%) D GDP per D Factor % Share Country/ Region D TFP (c) capita (a) Accumulation (b) (c) / (a) LAC 1.79 1.80 -0.01 -0.6% East Asia &Pac. 3.69 2.85 0.83 22.5% United States 1.99 1.21 0.78 39.2% China 6.04 4.21 1.83 30.3% Finland 2.74 1.44 1.30 47.4%
Low Productivity Growth TFP Relative to the US, 1960 – 2013 (in percent)
How can we explain this? Two microeconomic sources of productivity growth drive aggregate efficiency over time: Efficiency gains that occur within 1 firms , due to better production methods, organization, innovation at the plant-level, learning and Focusing on aggregate capability development, etc. data, we miss the Efficiency gains derived from the 2 effects of resource reallocation of resources from less reallocation and productive firms to more productive within firms dynamics ones - due to competition and processes of (Schumpeterian) creation and destruction - entry (exit) of more (less) efficient firms.
Why we focus on 1 Syverson 2011 Hsieh and Klenow 2009 Grazzi & Pietrobelli, 2016, Busso et al., 2012 Huge Heterogeneity among firms in productivity and innovation (also within sectors)
Why we focus on 1 – heterogeneity in productivity in LA Manufacturing Services Many low productive firms Few highly productive firms 0 Labor Productivity Huge Heterogeneity among firms in productivity and innovation (also within sectors)
Data We use World Bank Enterprise Survey data to empirically analyze the determinants of productivity at the firm level. Together with other databases such as EPFE (Productivity and Human Resources Training Establishments Survey), TIVA, Foreign banks’ international presence , PROTEQIN (firm-level innovation datasets for 14 Caribbean countries) WBES: firm-level data on 135 countries , 2006, 2010, with collaboration WB-IDB http://www.enterprisesurveys.org It covers 17 countries from Latin America and 14 countries from the Caribbean from manufacturing and service sector ES in LAC includes questions on innovation, and access to policies
1. Determinants of Enterprise Performance in Latin America and the Caribbean: what does the micro- evidence tell us? Carlo Pietrobelli, Matteo Grazzi, Eddy Szirmai 2. Innovation Dynamics and Productivity: Evidence for Latin America Gustavo Crespi, Fernando Vargas, Ezequiel Tacsir 3. Innovative Activity in the Caribbean: Drivers, Benefits, and Obstacles Patrick K Watson, Eric Strobl, Preeya Mohan 4. ICT, Innovation and Productivity: Evidence from Latin American and the Caribbean Firms Matteo Grazzi, Juan Jung 5. On-the-Job Training in Latin America and the Caribbean: Recent Evidence David Rosas Shady, Carolina González Velosa, Roberto Flores Lima 6. Business Performance in Young Latin American Firms Hugo Kantis, Juan Federico, Pablo Angelelli, Sabrina Ibarra Garcia 7. Different Obstacles for Different Productivity Levels? An Analysis of Caribbean Firms Alison Cathles, Siobhan Pangerl 8. Credit Access in Latin American Enterprises Roberta Rabellotti, Andrea F. Presbitero 9. International Linkages, Value Added Trade and LAC Firms' Productivity Pierluigi Montalbano, Silvia Nenci, Carlo Pietrobelli 10. Innovation and Productivity in Latin America and the Caribbean Firms: Conclusions Carlo Pietrobelli, Matteo Grazzi, Eddy Szirmai
RESULTS
1 Innovation improves productivity 1. First, LA firms are more likely to introduce product or process innovation if they spend more on innovation . +10% in R&D spending = +1.7 percent increase in the probability to innovate and an increase of innovative sales of 1.3%. 2. Second, labor productivity of innovative firms in LA is on average 50% higher than that of non-innovative firms. 3. These mechanisms vary largely depending on firm capabilities and characteristics. e.g. size, product diversification and fixed investment, the quality of human capital.
Ln(Q/L) Ln(Q/L) Ln(Q/L) Ln(Q/L) Positive (1) (2) (3) (4) significant 0.5025 *** 0.5028 *** 0.5028 *** 0.5070 *** Materials (0.0208) (0.0191) (0.0190) (0.0174) impacts of 0.0919 *** 0.0914 *** 0.0918 *** 0.0903 *** Capital (0.0075) (0.0078) (0.0089) (0.0080) innovation on 0.4821 *** 0.4915 *** 0.5170 *** 0.4957 *** Human capital productivity (0.0557) (0.0548) (0.0556) (0.0637) 0.0777 *** 0.0783 *** 0.0909 *** 0.0766 *** Employment (0.0110) (0.0099) (0.0093) (0.0112) Manager experience −0.0003 −0.0005 −0.0007 −0.0004 (0.0007) (0.0006) (0.0006) (0.0007) 0.5543 *** Innovation — — — (0.0879) 0.3635 *** Product innovation — — — (0.1195) Process innovation — 0.1860 0.0636 — (0.1307) (0.1746) 0.5225 ** Innovative sales — — — (0.2113) 0.3477 *** IPRs — — — (0.0865) N 4376 4376 4376 4376 Ll −3596.6234 −3596.8416 −3597.4046 −3607.396 chi2 14787.2106 9124.5645 13287.8438 17278.7706 P 0.0000 0.0000 0.0000 0.0000 Source: Authors’ elaboration using WBES. Notes: Coefficients reported are marginal effects. Bootstrapped standard errors in parentheses.
Caribbean firms: 1 Are they special? .8 Spending on innovation .6 has a higher return than .4 in most LA. A unit increase in logged .2 innovation expenditure per employee increases the 0 probability of an 0 1 2 3 4 Productivity innovation by 56%. Innovative Non-Innovative Innovation improves productivity performance. Labor productivity of innovative firms is on average 63 percent higher than that of non-innovative firms.
2 The productivity benefits of innovation and human capital are greater for firms that are already productive
3 Broadband matters both for innovation and productivity, but only if is utilized well Broadband adoption Use of Internet has a positive and also affects significant effect on probability to innovation (both in innovate, but… products and Only if used for Research processes) and the broader the variety of activities productivity for which broadband is used, the greater the impact on innovation
3 Broadband matters both for innovation and productivity, but only if is utilized well! Product innovation Process innovation Variables (1) (2) (3) (4) Broadband adoption 0.214*** 0.064 0.255*** 0.094** (0.036) (0.044) (0.039) (0.047) Internet use for purchases 0.016 0.019 (0.019) (0.020) Internet use to deliver 0.013 0.038* services (0.020) (0.020) Internet use for research 0.112*** 0.105*** (0.020) (0.021) Internet for purchases + 0.060** 0.048* deliver services + research (0.024) (0.025) Log Likelihood -4929,68 -4868,86 -5017,95 -4961,54 Rho -0.170** -0.145** -0.269*** -0.242*** (0.067) (0.067) (0.071) (0.072) /athrho -0.172** -0.146** -0.276*** -0.247*** (0.069) (0.069) (0.076) (0.077) Observations 5930 5930 5926 5926 Source: Grazzi and Jung (2016)
4 Firms decide to invest in training only if they also spend in R&D, innovate, have credit, face demand for (higher) skills Small Medium Large Has ISO certificate % Expenditure in R&D Improved processes Credit with financial institition Introduced new products Lack of skills is a major obstable Age of the firm Number of employees Number of competitors Fraction of domestic sales Fraction of temporary workers Experience of highest manager Fraction of skilled workers Fraction of domestic property -.2 0 .2 .4 -.2 0 .2 .4 -.2 0 .2 .4 Marginal Effects
5 Training in firms has effect on productivity only for large firms EFFECT OF TRAINING ON PRODUCTIVITY (1) (2) (3) (4) Intensity −0.001 −0.001 −0.001 −0.001 (0.001) (0.001) (0.001) (0.001) Intensity * Large Firm 0.006* 0.007* n.a. n.a. (0.004) (0.004) Intensity * Improved Processes n.a. n.a. -0.003 0.002 (0.002) (0.002) Observations 1479 1461 1479 1461 Firm-level fixed effects Yes Yes Yes Yes Time varying controls No Yes No Yes Source: Authors ʼ calculations based on WBES. Notes: This table presents estimates of the effect of increasing the proportion of trained workers. Standard errors are in parentheses. * Coefficient is statistically significant at the 10% level, ** at the 5% level, *** at the 1% level; no asterisk means the coefficient is not different from zero with statistical significance. n.a. = not applicable.
Recommend
More recommend
