Skip to main content
SpringerLink
Log in

Globalisation and technological convergence in the EU

  • Published:
Journal of Productivity Analysis Aims and scope Submit manuscript

Abstract

We employ a two-step approach in investigating the dynamic transmission channels under which globalization factors foster technical efficiency by combining a dynamic efficiency analysis in the stochastic frontier framework, and a time series approach based on VAR and spectral analysis. Using the dataset of the 18 EU countries over 1970–2004, we find that both import and FDI are significant factors in spreading efficiency externalities and thus accelerating technology catch-up in the EU. In particular, the impacts of the import are more prominent in the short-run while those of FDI play a more important role over the longer-run. Furthermore, the impacts of the import are pro-cyclical only in the short-run whereas those of FDI are pro-cyclical mostly over the medium- to the long-run. This evidence is broadly consistent with the sample observation that the recent slowdown of the EU productivity has been closely related to the corresponding FDI decline especially after 2000. Hence, any protection-oriented policy will be likely to be more detrimental for the EU.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Notes

  1. In an earlier version of the paper, we have also considered an alternative specification in which the global variables enter the regression in (1) as the observed factors such as \({\bf \delta }_{i}^{\prime }{\bf \bar{s}}_{t}\) or \(\mathbf{\delta }^{\prime }{\bf \bar{s}}_{t}\) where \({\bf \bar{s}}_{t}=N^{-1}\sum\nolimits_{i=1}^{N}{\bf s}_{it}. \) In both cases, however, there is an identification problem such that technical inefficiency (e it ) in (3) cannot be uniquely estimated. We are grateful to an anonymous referee for pointing this issue.

  2. Notice under the current setup that an exogenous technological change, usually assumed to be common to every individual and specified as γt for example, cannot be uniquely identified from b i t in (2), where b i measures an individual diffusion response to common (exogenous) technological change. We are grateful to an anonymous referee for pointing this issue. Furthermore, the use of a linear time trend cannot track the non-monotonic productivity decline as we observed from the mid 90s in the EU, see Sect. 3 below.

  3. Factor models are shown to exhibit the strong cross-sectional dependence since the maximum eigenvalue of the covariance matrix for \(\varepsilon _{it}\) tends to infinity at rate N as \(N\rightarrow \infty , \) whilst other models such as spatial AR or MA models display much lower degrees of cross-sectional dependence, see Chudik et al. (2011) for details.

  4. Bai (2009) examines the cross-section dependence in panels more extensively, and allows regressors to be correlated with both factors and loadings by including both additive and interactive fixed effects. He then proposes an estimation method in which the unobservable common factors can be consistently estimated by the principal components. In this regard, the extension of the current study using the Bai’s approach will be an interesting future study since the successful implementation will shed further lights on capturing further dynamics even in the presence of weakly correlated idiosyncratic errors, v it .

  5. Monte Carlos studies by Serlenga and Shin (2007b) confirm that the biases of the PCCE estimators of slope parameters (i.e. β and \(\delta \)) decrease with N and T. We have also conducted further Monte Carlo studies, showing that the biases of \(\alpha _{i}^{\ast }\) also tend to decrease with N and T. For example, we find that even when (N, T) = (25, 25) , the biases of all the estimators used in the current study are reasonably small. These results are available upon request.

  6. Barro and Sala-i-Martin (2004) identify the Solow residuals as technical changes, though they represent a total factor productivity capturing both technological and efficiency changes. Alternatively, Caves et al. (1982) propose the Malmquist productivity index, defined as a distance function in order to distinguish between technological changes and efficiency changes.

  7. PIM is necessitated by the lack of capital stock data across all the countries. For an individual country, the capital stock is constructed as \(K_{t}=K_{t-1}\left( 1-\theta \right) +I_{t}, \) where I t is investment and θ the rate of depreciation assumed to be 6 % (e.g., Hall and Jones 1999; Iyer et al. 2008). Repair and maintenance are assumed to keep the physical production capabilities of an asset constant during its lifetime. Initial capital stocks are constructed, assuming that capital and output grow at the same rate. Specifically, for country with investment data beginning in 1970, we set the initial stock, \(K_{1970}=I_{1969}/\left( g+\theta \right) , \) where g is the average 10-year output growth rate from 1970 to 1980 and I is investment in gross physical capital stock. Estimated capital stock includes both residential and nonresidential capital.

  8. Luxembourg is both the largest recipient of FDI and largest trade oriented due to ‘trans-shipped’ FDI, i.e. how companies transfer funds between affiliates within the same group located in different countries, or channel funds to acquire companies in different countries through a holding company. For these reasons, Luxembourg is often treated as an outlier, e.g. Edwards (1998) and Daniels et al. (2005). We have also estimated the model without including Luxembourg but obtained qualitatively similar results.

  9. The peak in 2000 was argued to be mainly driven by a large amount of shares and acquisition operations spearheaded by rapid information technological changes during the 1990s mostly through the regulatory relaxation, as also reported in official statistics in World Bank, Eurostat and Unctad.

  10. Staring from the full set of the augmented factors, \(\mathbf{w}_{t}=\left( \bar{y}_{t},\bar{l}_{t},\bar{k}_{t},\overline{FDI}_{t},\overline{M}_{t},t\right) ^{\prime }\) in (6), we have estimated the model with various combinations of augmented factors, and then selected the final empirical specification (their estimations results reported in Table 2) on the basis of overall statistical significance and parsimony. Overall estimation results are qualitatively similar across different specifications.

  11. The hypothesis of the constant return to scale turns out to be suitable for industrialised countries (e.g. Malley et al. 2005; Iyer et al. 2008). Further, we cannot reject the null hypothesis that the production function is of the Cobb-Douglas form against the translog form.

  12. See the “Appendix” for detailed results of individual efficiencies for the eighteen EU countries.

  13. To save space we do not report the estimation results of (11), which do not suffer from any serious misspecification. Though we mainly focus on modelling the dynamic interactions between efficiency and openness factors simultaneously, we have also conducted the Granger causality tests for the following null hypotheses: (1) efficiency does not Granger-cause imports; (2) efficiency does not Granger-cause FDI; (3) imports do not Granger-cause efficiency; (4) FDI does not Granger cause efficiency. We find that the p values of the χ2 tests are 0.079, 0.334, 0.001, and 0.029, respectively. Combined together, we may conclude that the causation is likely to run from imports and FDI to efficiency. This evidence will provide further support for our focus on the impulse responses of efficiency with respect to imports and FDI shocks below.

  14. Notice that the spectrum is expressed as a function of radians, ω = 2πλ with \(\lambda \in \left[ -0.5,0.5\right] \) rather than time units, see Priestley (1981). T = 1/ω, and, hence, T = 1/0.27 = 3.7.

  15. The partial squared coherency can be used to calculate the proportion of variance in a frequency band due to a specific variable, with the influence of other variables removed, see Koopmans (1974). This enables us to consider isolated effects of fluctuations in imports and FDI on common efficiency.

  16. Given the small time period observations of our data set (35 annual observations), we note in passing that cycles of 10 years or more cannot be accurately identified.

References

  • Aghion P, Howitt P (1992) A model of growth through creative destruction. Econometrica 60:351–423

    Article  Google Scholar 

  • Ahn SC, Good DH, Sickles RC (2000) Estimation of long-run inefficiency levels: a dynamic frontier approach. Econometr Rev 19:461–492

    Article  Google Scholar 

  • Ahn SC, Lee YH, Schmidt P (2007) Stochastic frontier models with multiple time-varying individual effects. J Product Anal 27:1–12

    Article  Google Scholar 

  • Aigner DJ, Lovell CAK, Schmidt P (1977) Formulation and estimation of stochastic frontier production function models. J Econometr 6:21–37

    Article  Google Scholar 

  • Alcalà F, Ciccone A (2004) Trade and productivity. Q J Econ 119:613–646

    Article  Google Scholar 

  • Artis MJ, Galvão ABC, Marcellino M (2007) The transmission mechanism in a changing world. J Appl Econometr 22:39–61

    Article  Google Scholar 

  • Bai J (2009) Panel data models with interactive fixed effects. Econometrica 77:1229–1279

    Article  Google Scholar 

  • Balk BM (2003) The residual: on monitoring and benchmarking firms, industries, and economies with respect to productivity. J Product Anal 20:5–47

    Article  Google Scholar 

  • Barro R (2001) Human capital and growth. Am Econ Rev 91:12–17

    Article  Google Scholar 

  • Barro RJ, Sala-i-Martin X (2004) Economic growth, 2nd edn. The MIT Press, Massachusetts

    Google Scholar 

  • Battese GE, Coelli TJ (1995) A model for technical inefficiency effects in a stochastic frontier production function for panel data. Empiric Econ 20:325–332

    Article  Google Scholar 

  • Bernard AB, Jones CI (1996) Productivity across industries and countries: time series theory and evidence. Rev Econ Stat 78:135–146

    Article  Google Scholar 

  • Borensztein E, De Gregorio J, Lee LW (1998) How does foreign direct investment affect economic growth? J Int Econ 45:115–135

    Article  Google Scholar 

  • Busse M, Groizard JL (2006) Foreign direct investment, regulations and growth. World Bank Policy, Research Working Paper, no. WPS3882

  • Cameron G, Proudman J, Redding S (2005) Technological convergence, R&D, trade and productivity growth. Eur Econ Rev 49:775–807

    Article  Google Scholar 

  • Canova F, Dellas H (1993) Trade interdependence and the international business cycle. J Int Econ 34:23–47

    Article  Google Scholar 

  • Caves DW, Christensen LR, Diewert EW (1982) The economic theory of index numbers and the measurement of input, output, and productivity. Econometrica 50:1393–1414

    Article  Google Scholar 

  • Chudik A, Pesaran MH, Tosetti E (2011) Weak and strong cross-section dependence and estimation of large panels. Econometr J 14:45–90

    Article  Google Scholar 

  • Coe DT, Helpman E (1995) International R&D spillovers. Eur Econ Rev 39:859–887

    Article  Google Scholar 

  • Cohen WM, Levinthal DM (1989) Innovation and learning: the two faces of R&D. Econ J 99:96–569

    Article  Google Scholar 

  • Cornwell C, Schmidt P, Sickles R (1990) Production frontiers with cross-sectional and time-series variation in efficiency levels. J Econometr 46:185–200

    Article  Google Scholar 

  • Croux C, Forni M, Reichlin L (2001) A measure of comovement for economic variables: theory and empirics. Rev Econ Stat 83:232–241

    Article  Google Scholar 

  • Daniels JP, Nourzad F, VanHoose DD (2005) Openness, central bank independence, and the sacrifice ratio. J Money Credit Bank 37:371–379

    Article  Google Scholar 

  • ECB (2009) Assessing global trends in protectionism. European Central Bank Monthly Bulletin

  • Edwards S (1998) Openness, productivity and growth: what we really know? Econ J 108:383–398

    Article  Google Scholar 

  • Eichengreen B (2004) Productivity growth, the new economy, and catching up. Rev Int Econ 12:243–245

    Article  Google Scholar 

  • Frankel J, Romer D (1999) Does trade cause growth? Am Econ Rev 89:373–399

    Article  Google Scholar 

  • Garrett TA (2003) Aggregated versus disaggregated data in regression analysis: implications for inference. Econ Lett 81:61–65

    Article  Google Scholar 

  • Greenwood-Nimmo M, Nguyen VH, Shin Y (2011) Probabilistic forecasting of output growth, inflation and the balance of trade in a GVAR. J Appl Econometr (forthcoming)

  • Grossman GM, Helpman E (1991) Innovation and growth in a global economy. MIT Press, Cambridge

    Google Scholar 

  • Grossman GM, Helpman E (1996) Technology and trade. In: Grossman G, Rogoff K (eds) Handbook of international economics. North-Holland, Amsterdam, vol 3

    Google Scholar 

  • Hall RE, Jones CI (1999) Why do some countries produce so much more output per worker than others? Q J Econ 114:83–116

    Article  Google Scholar 

  • Hodrick R, Prescott EC (1996) Postwar U.S. business cycles: an empirical investigation. J Money Credit Bank 29:1–16

    Article  Google Scholar 

  • Iyer KG, Rambaldi AN, Tang KK (2008) Efficiency externalities of trade and alternative forms of foreign investment in OECD countries. J Appl Econometr 23:749–766

    Article  Google Scholar 

  • Jovanovic B, Rob R (1989) The growth and diffusion of knowledge. Rev Econ Stud 56:569–582

    Article  Google Scholar 

  • Kapetanios G, Pesaran MH (2007) Alternative approaches to estimation and inference in large multifactor panels: small sample results with an application to modelling of asset returns. In: Phillips G, Tzavalis E (eds) The refinement of econometric estimation and test procedures: finite sample and asymptotic analysis. Cambridge University Press, Cambridge

    Google Scholar 

  • Kneip A, Sickles R, Song W (2011) A new panel data treatment for heterogeneity in time trends. Econometr Theory (forthcoming)

  • Kose MA, Prasad EE, Terrones ME (2003) How does globalization affect the synchronization of business cycles? Am Econ Rev 93:57–62

    Article  Google Scholar 

  • Koopmans L (1974) The spectral analysis of time series. Academic Press, New York

    Google Scholar 

  • Krugman P (1991) Increasing returns and economic geography. J Polit Econ 99:99–483

    Article  Google Scholar 

  • Kumbhakar S (1990) Production frontiers, panel data, and time-varying technical inefficiency. J Econometr 46:201–212

    Article  Google Scholar 

  • Kumbhakar S, Lovell CAK (2000) Stochastic frontier analysis. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Lee HL (1991) Panel data models with multiplicative individual and time effects: applications to compensation and frontier production functions, Ph.D. Dissertation

  • Lee HL, Schmidt P (1993) A production frontier model with flexible temporal variation in technical inefficiency in the measurement of productive efficiency: techniques and applications, In: Fried H, Lovell CAK, Schmidt S (eds) Oxford University Press, Oxford

  • Malley JR, Muscatelli VA, Woitek U (2005) Real business cycles or sticky prices? The impact of technology shocks on US manufacturing. Eur Econ Rev 107:745–760

    Article  Google Scholar 

  • Mankiw G, Romer D, Weil D (1992) A contribution to the empirics of economic growth. Q J Econ 107:407–437

    Article  Google Scholar 

  • Mastromarco C, Woitek U (2007) Regional business cycles in Italy. Comput Stat Data Anal 52:907–918

    Article  Google Scholar 

  • Meeusen W, van den Broeck J (1977) Efficiency estimation from Cobb-Douglas production functions with composed error. Int Econ Rev 18:435–444

    Article  Google Scholar 

  • Nelson R, Phelps E (1996) Investment in humans, technological diffusion, and economic growth. Am Econ Rev Pap Proc 51:69–75

    Google Scholar 

  • Pesaran MH (2004) General diagnostic tests for cross section dependence in panels. Cambridge Working Paper in Economics 0435

  • Pesaran MH (2006) Estimation and inference in large heterogeneous panels with a multifactor error structure. Econometrica 74:967–1012

    Article  Google Scholar 

  • Priestley M (1981) Spectral analysis and time series. Academic Press, London

    Google Scholar 

  • Romer PM (1990) Endogenous technological change. J Polit Econ 98:71–102

    Article  Google Scholar 

  • Schmidt P, Sickles RC (1984) Production frontiers and panel data. J Bus Econ Stat 2:74–367

    Google Scholar 

  • Segerstrom P (1991) Innovation, imitation and economic growth. J Polit Econ 99:807–827

    Article  Google Scholar 

  • Serlenga L, Shin Y (2007a) Gravity models of intra-EU trade: application of the CCEP-HT estimation in heterogeneous panels with unobserved common time-specific factors. J Appl Econometr 22:361–381

    Article  Google Scholar 

  • Serlenga L, Shin Y (2007b) Monte Carlo studies on finite sample performance of the CCEP-HT estimator in panels with heterogeneous unobserved common factor. J Appl Econometr (The online supplement)

  • Taylor C (2008) Foreign direct investment and the Euro: the first five years. Cambridge J Econ 32:1–28

    Article  Google Scholar 

  • Tybout JR (1992) Linking trade and productivity: new research directions. World Bank Econ Rev 6:189–211

    Article  Google Scholar 

  • Van Ark BO, Mahony M, Ypma G (2007) The EU KLEMS productivity report: technical report, European Commission. Policy Support and Anticipating Scientific and Technological Needs

  • Wacziarg R, Welch KH (2008) Trade liberalization and growth: new evidence. World Bank Econ Rev 22:187–231

    Article  Google Scholar 

  • Wang H, Schmidt P (2002) One-step and two-step estimation of the effects of exogenous variables on technical efficiency levels. J Product Anal 18:129–144

    Article  Google Scholar 

Download references

Acknowledgments

We owe a debt of gratitude to the Editor and to two anonymous referees for their constructive feedbacks on an earlier draft of the manuscript. We are also grateful to Badi Baltagi, Charlie Cai, Kausik Chaudhuri, Ana Galvão, George Kapetanios, Junsoo Lee, Kevin Reilly, Ulrich Woitek, the session participants at the 2nd Italian Congress of Econometrics and Applied Economics at Rimini, 2007, and at the 15th Panel Data Conference at Bonn, 2009, and the seminar participants at Universities of Zurich and Leeds for their helpful comments. The third author acknowledges partial financial support from the ESRC (Grant No. RES-000-22-3161). The usual disclaimer applies.

Author information

Authors and Affiliations

  1. Department of Management and Economics, University of Salento, CESIfo and RCEA, 73100, Lecce, Italy

    Mastromarco Camilla

  2. Department of Economics, University of Bari, IZA, 70124, Bari, Italy

    Laura Serlenga

  3. Department of Economics and Related Studies, University of York, Heslington, York, YO10 5DD, UK

    Yongcheol Shin

Authors
  1. Mastromarco Camilla
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Laura Serlenga
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yongcheol Shin
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mastromarco Camilla.

Appendix

Appendix

See Table 5.

Table 5 Detailed results for individual efficiency for 18 EU countries
Full size table

Rights and permissions

Reprints and permissions

About this article

Cite this article

Camilla, M., Serlenga, L. & Shin, Y. Globalisation and technological convergence in the EU. J Prod Anal 40, 15–29 (2013). https://doi.org/10.1007/s11123-012-0308-9

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11123-012-0308-9

Keywords

JEL

Search

Navigation