1 Exports and Productivity Link in Manufacturing: Microeconomic Evidence from Croatia Gorana Lukinić Čardić Dubrovnik, June 23, 2010

2 Outline  Introduction  Croatian exports: macroeconomic view  Structural features of Croatian exporters  Methodology  Results  Conclusion

3 Introduction  "Productivity isn't everything, but in the long run it is almost everything. A country’s ability to improve its standard of living over time depends almost entirely on its ability to raise its output per worker.“ Paul Krugman  Croatian economy characterized by macroeconomic stability but deteriorating external position  Paper presents microeconomic data on Croatian exports, focusing on the link between productivity and exports  Significant evidence on exporter premium: exporters are on average larger, more productive, pay higher wages, etc.  Self-selection vs. learning-by-exporting hypothesis

4 Low competitiveness of Croatian merchandise exports Annual growth rate of exports (%) Exports per capita (EUR) Source: Eurostat

5 Factors behindweak results  lack of investment in the creation of advanced products, low level of competition in innovation and investment in R&D, the lack of more “demanding” domestic market (Mikić, 2002)  slow implementation of reforms, bureaucratic obstacles, problems with the judiciary, unregulated land registry (Teodorovic and Buturac, 2006)  delay in the process of Croatian accession to the EU (Škuflić, 2005)  ….

6 Who exports in Croatia? (i) Exporter participation rate and export intensity Source: own calculations based on FINA database Exporter participation rate is the ratio between the number of exporters and the total number of enterprises. Export intensity is the average share of exports in total sales of exporters.

7 Who exports in Croatia? (ii)  Majority of Croatian exporters are small firms – they accounted for almost 70 percent of exporters in 2007  But the bulk of total exports is realized by large firms that have the highest export intensity, which was on average around 40 percent  According to ownership structure, privately owned enterprises dominate  Most of the exporters (around 80 percent) belong to low and medium-low technology industries

8 Who exports in Croatia? (iii) Concentration indicators Source: own calculations based on FINA database

9 Transition matrix Values show percentages. N stands for non-exporter and E for exporter.

10 Estimation methodology  Exporter productivity premium: LP is labour productivity EDUM is a dummy variable for the exporter status CONTROL is a vector of control variables: dummy variables for each year and industry at the two-letter level, firm size (assets or dummy for firm size according to official classification)  OLS regression on panel data: without and with fixed effects

11 Estimation methodology (ii)  Self-selection hypothesis:  Learning-by-exporting hypothesis:

12 Results: exporter productivity premium All coefficients are statistically significant. Robustness checks: sample of firms with more than 20 employees; sample without small exporters; sample without occasional exporters

13 Results: self-selection hypothesis No coefficient is statistically significant. The dependent variable is labour productivity defined as the ratio of sales and employment. Firm size is measured by their assets.

14 Results: learning-by-exporting hypothesis The dependent variable is labour productivity defined as the ratio of sales and employment. Firm size is measured by their assets.

15 Conclusion  Rich structure of exports is hidden behind aggregate data  Exports are more concentrated than sales and employments: within-sector concentration in line with the new new trade theory  Robust estimates of exporter productivity premium  Ambiguous evidence on the self-selection and learning- by-exporting hypothesis  Many possibilities for improving the estimates and for further research

16 Thank you for your attention!