Research

Home Country Effects of Multinational Network Restructuring in Times of Deglobalization - Evidence from European MNEs (with B. Michel)

Under review - revisions requested

This paper documents a trend towards deglobalization in European-based multinational networks for the period 2010-2019. In the second half of the decade, the number of foreign contraction episodes shows an increasing trend, while the number of foreign expansions decreased substantially. Foreign expansions also increasingly resulted in a reduced geographic scope of networks (nearshoring) and a higher concentration of activity in geopolitically aligned host countries (friendshoring). We estimate home-country effects of foreign network restructuring by analyzing the number of domestic affiliates and different outcomes for parents and domestic affiliates. We find no evidence of increased home country activity in the wake of foreign contraction episodes, but foreign expansion yields benefits for the domestic economy both along the domestic extensive and intensive margins. A reduced geographic scope and geopolitical reorientation do not induce systematic differences in the home-country effects neither for expansion nor for contraction episodes.

April 2025 version
NBB Working Paper No 465- slides NBB colloquium
SUERF Policy Brief

Gentse Economische Inzichten No. 21 (in dutch) 

State-Owned Enterprises in Europe - Firm Performance and Aggregate Effect (with P. Muylle)

Under review - revisions requested

Since the late 2000s, shocks and crises of various types have led to a revival of state intervention around the world. This paper builds a large firm-level dataset to analyze state ownership of firms in Europe for the period 2002-18.  We confirm the underperformance of state-owned enterprises (SOEs) relative to privately-owned enterprises (POEs) found in earlier literature for this recent period for a range of firm-level performance indicators. We also examine the impact of SOEs on private firms. We find that larger SOE presence in an industry is associated with lower productivity growth and lower productivity levels among private firms in that industry, but does not affect industry dynamics in terms of entry and exit. This suggests potential aggregate productivity gains from reallocating resources from SOEs to POEs. Further, we show that employment is more stable and crisis-resistant at SOEs, and that SOEs are a more stable source of downstream input demand for other firms. Leveraging our dataset's cross-country nature, we find that SOEs are complements to, rather than substitutes for, lower quality institutions.

 UGent Department of Economics Working Paper No 25/1105 

European Integration, Border Regimes, and Cross-Border Technology Diffusion Through FDI Spillovers (with V. Purice)

Under review
We analyze how EU integration has contributed to regional cross-border technology transfers. We modify the typical FDI spillover framework to analyze cross-border spillovers in regions close to the border. We then use a multi-country firm-level panel to analyse the impact of border regimes on the existence and size of cross-border spillovers, exploiting variation in the pace and extent of European integration of seven Eastern European countries between 2000 and 2010. We find EU membership to be a necessary condition for cross-border productivity spillovers. Schengen area participation further magnifies these spillovers. Our results warn about productivity costs of reinstating border restrictions.(previously circulated as "Border regimes and indrect productivity effects from foreign direct investment") 

version February 2025 - Department of Economics Working Paper 2019/965 - Coverage at voxEU

Intangibles within firm boundaries  (with A. Theodorakopoulos)

This paper extends a production function estimator to test whether intangible transfers within firm boundaries lead to overall efficiency gains. Using a panel of European majority owned parent-affiliate relationships, we present novel evidence that parent firms strongly benefit from such transfers alongside productivity enhancements for affiliates. In relative terms, affiliates’ long-run efficiency improvements are twice those of the parent. Such gains appear to be induced by synergies for the affiliate but not for the parent, supporting theories on the existence of common ownership. A falsification exercise suggests that only 2/3 of these gains are actually internalised within firm boundaries.

version December 2020

How domestic firms' absorptive capacity shapes productivity spillovers from (non-)R&D-active foreign affiliates (with Y. Wei and A. Spithoven)

Under review
Foreign direct investment (FDI) and research and development (R&D) are both key drivers of firm productivity and a potential source of spillovers to other firms. Existing spillover research typically examines FDI and R&D separately. This study integrates them by (i) analysing how productivity spillovers from foreign to domestic firms depend on foreign affiliates’ R&D activity and by (ii) examining the role of domestic firms’ absorptive capacity in terms of R&D activity and R&D cooperation to capture spillovers from foreign firms. Using a panel of firms in Belgium, we first document the productivity premia of foreign ownership and R&D activity. Foreign affiliates are generally more productive than domestic firms, and within these two groups, those with R&D activities are even more productive. R&D cooperation entails a further significant productivity premium for domestic firms, but not for foreign affiliates. Positive productivity spillovers to domestic firms are only generated by foreign affiliates that are R&D-active. Domestic firms that are R&D-active experience larger spillovers, and those that are not R&D-active experience negative productivity spillovers from non-R&D-active foreign affiliates. R&D activity thus shields domestic firms from such negative spillovers. R&D cooperation by domestic firms has no significant additional beneficial effect on spillovers.

version July 2025

DATA PAPERS

Female Managers and Firm Performance in Europe

This paper builds a large pan-European panel dataset of firm-level senior management gender composition. We focus the dataset on firms in the business economy that file unconsolidated accounts and report both total assets and strict positive employment throughout their existence. We have management information for 9 million firms for the period 2005-2020 resulting in 60 million firm-year observations. Overall 40% of observations concerns firms with at least one female manager and 60% are led only by male managers. For (predominantly micro) firms with a single manager that account for 53.5% of observations, we find that only 23% are female-managed. 59.5% of firms with two or more managers have at least one female manager. Across countries between 14% and 66% of observations refer to firms where at least half of the managers are female, across industries variation is more limited and ranges between 20% and 53%. We find that within tight country-industry-year cells women-led SMEs are smaller and less productive and show lower leverage. Real performance differences are sustained in an event study analysis of switching firms, financial performance differences are not. Female-managed firms show lower short and medium-run growth rates. These effects are small and remain unchanged (also in magnitude) when controlling for leverage. 

FEB UGent Working Paper 2025/1110

The Margins of Ownership Structures - Insights from 150,000 networks over 25 years (with A. Theodorakopoulos)

This paper uses a newly build dataset of multinational networks with European parents and worldwide affiliates to uncover a set of stylized facts on MNE performance, structure, and extensive and intensive margins. The dataset contains 5,749,201 observations between 1995 and 2020 on 183,938 unique parents (and thus networks) and 1,033,447 unique affiliates. We find that MNE parents are stars compared to parents of domestic networks, but we also detect superstars among the stars with a small number of larger and more efficient MNE parents having larger MNE networks. We find that most networks start small. Most networks are stable and thus stay small. Larger and superstar networks have more churning of affiliates. We find that networks do exit and that these are mainly small and young networks. Full ownership of affiliates is normal as two thirds of affiliates are fully-owned by the paretn. We document that networks cover all industries and thus are not limited to manufacturing and sales activities. Affiliates of more efficient parents are more efficient themselves. Minority-owned, less integrated affiliates, and affiliates with less network maturity are more likely to be dropped from the network. Finally we show that short-lived ownership of affiliates is not uncommon.

(updating, mail me for a pdf)

The role of services sectors for aggregate productivity: A firm-level anatomy of a large panel of European firms (with A. Van Maele)

This paper documents how the composition of value added per worker in Europe is distributed over manufacturing, services, and other industries based on a large panel of firm-level data. We show that a non-negligible part of value added is accounted for by services industries. We then explore how micro-data at the firm level can be used to analyse this important component of aggregate productivity growth. We further discuss and explore using our data whether semi-parametric estimators of total factor productivity that are commonly found in the literature and typically tailored towards manufacturing are fit for analysing firms in services sectors. 

FEB UGent Working Paper 2025/1109

Multinational Networks, Foreign, and Domestic Firms in Europe (with M. de Zwaan, K. Lenaerts, and V. Purice)

This paper introduces two datasets, AUGAMA, a panel of European firms for the period 1996-2011, and EUMULNET, a European Multinational Network data set. These datasets are constructed on the basis of the Amadeus database issued by Bureau Van Dijk Electronic Publishing. We document the process of building these data sets from the raw Amadeus data for 26 European countries. We show that the data sets adequately approximate the structure of the European economy across countries, regions, and industries as portrayed by data from Eurostat (Structural Business Statistics) and Cambridge Econometrics. As an illustration of possible application, we use the datasets to test a number of results from the theoretical literature regarding the productivity of multinational firms vis-a-vis domestic firms. 

FEB UGent Working Paper 15/900

DORMANT

Within-firm Pay Inequality and State Ownership: Evidence from Europe  (with B. De Lange)

This paper examines how political motives and state ownership affect within-firm pay inequality in European listed firms. The ratio of CEO pay to the average wage cost of non-executive employees in state-owned enterprises (SOEs) is 42% lower than in private firms. We provide empirical evidence consistent with political motives as an explanation for the observed differences. We only find significant differences in environments where we expect governments to experience more electoral pressure to be tough on CEO pay and within firm pay inequality. Only countries characterized by high inequality aversion, high redistribution, high transfers, and countries with a left wing government show significant differences.

version September 2021

OLDER WORKING PAPER VERSIONS OF PUBLISHED ARTICLES

Financial Crises and the Global Supply Network - Evidence from Multinational Enterprises (with S. Basco, G. Felice, and M. Mestieri)

Published at Journal of International Economics, NBER working paper 31216 and CEPR DP18163
This paper empirically examines the effects of financial crises on the organization of production of multinational enterprises. We construct a panel of European multinational networks from 2003 through 2015. We use as a financial shock the increase in risk premia between August 2007 and July 2012 and build a multinational-specific shock based on the network structure before the shock. Multinationals facing a larger financial shock perform worse in terms of revenue, employment, and growth in the number of affiliates. Lower growth in the number of affiliates operates through a negative effect on domestic and foreign affiliates, and is concentrated in affiliates in a vertical relationship with the parent. These effects built up slowly over time. Negative effects are driven by multinationals with initially more leveraged parents, who reduce relatively more the number of foreign affiliates. These findings lend support to the hypothesis of financial frictions shaping multinational activity.

Optimal transfer prices and technology in decentralized business groups (with S. Nasini and M. Verschelde)

Published at European Journal of Operational Research , earlier version at SSRN
Decentralized (parent-affiliates) business groups encompass independent firms sharing production resources under common administrative or financial control. Due to the complexity of integrated decision-making, business groups often delegate pricing and technology decisions to separated units. With the growing prevalence of transfer technology in multi-plant operations, there arises a demand for modeling frameworks capturing the joint impact of transfer prices and technology on affiliates’ production. In this paper, we propose an operational methodology to integrate transfer prices and technology, as incentive mechanisms driving affiliates’ production decisions. This consists of a single-leader-multi-follower game where a parent firm (the leader) operates in a single-product market and delegates production to profit-maximizing affiliate firms (the followers). In addition to providing data-driven support for integrated decision-making, our framework reveals how managerial structures respond to the growing significance of intangible technology. In this line, based on the parent’s control of the value chain, we present two specifications and solution approaches: full delegation and partial delegation. For the former, we provide an exact characterization of the equilibrium solution; whereas for business groups operating under full delegation, we provide tight lower and upper bounds, computable by state-of-the-art optimization solvers. Our empirical tests rely on a customized version of the Orbis data set. Figures show that our methodology not only integrates the substitutability between transfer prices and technology, as well as their dependency on the parent’s control of the value chain, but also gives rise to a flexible computational approach that is applicable to a wide range of decentralized business groups.

Gains from trade: Demand, Supply, and Idiosyncratic Uncertainty  (with R. Dewitte and G. Rayp)

Published at Journal of Applied Econometrics, version March 2021

Firm-level sales is often used as a proxy for productivity to quantify welfare Gains from Trade (GFT) using firm-level data. This ignores the existence of uncertainty and heterogeneity other than productivity in firm-level sales. We demonstrate, theoretically and empirically, that the existence of idiosyncratic uncertainty in firm-level sales results in a sizeable bias in GFT. Conflating uncertainty with productivity, as proxied by firm-level sales, results in an over-dispersed distribution of productivity. Assigning this uncertainty-inflated productivity to the modeled economy’s supply-side results in overestimated aggregate trade elasticities and GFT. We show the possibility to obtain unbiased productivity, aggregate trade elasticities, and GFT estimates by relying on the revenue production function.

Network control by a constrained external agent as a continuous optimization problem (with J. Nys, M. van den Heuvel, K. Schoors)

Published at Scientific Reports

Social science studies dealing with control in networks typically resort to heuristics or describing the control distribution as is. Optimal policies, however, require interventions that optimize control over a socioeconomic network subject to real-world constraints. We integrate optimisation tools from deep-learning with network science into a framework that is able to optimize such interventions in real-world networks. We demonstrate the framework in the context of corporate control, where it allows to characterize the vulnerability of strategically important corporate networks to sensitive takeovers, an important contemporaneous policy challenge. The framework produces insights that are relevant for governing real-world socioeconomic networks, and opens up new research avenues for improving our understanding and control of such complex systems.

The productivity impact of R&D and FDI spillovers  (with A. Spithoven)

Published at The Journal of Technology Transfer

R&D activities by indigenous firms create new knowledge and technology in a region, while foreign-owned firms bring advanced knowledge and technologies from their home countries. This paper analyses the spillover effects of R&D-active or foreign-owned firms on total factor productivity of domestic non-R&D active firms. Focusing on productivity spillover effects on domestic non-R&D active firms in three regions in Belgium between 2000 and 2017, we conclude that R&D spillovers generally occur more frequently than FDI spillovers. By focussing on two sources of technological spillovers, the paper offers a more nuanced picture of industrial transformation and regional path development. Region-specific analyses show that the effects on the productivity of domestic non-R&D active firms originate from different spillover sources. Linkages of R&D active or foreign-owned firms with domestic non-R&D active firms exert a heterogeneous impact on their productivity. 

Does a Tax Credit Matter for Job Creation by Multinational Enterprises (with J. Konings, and C. Lecocq)

Published at Canadian Journal of Economics, CEPR Discussion Paper 13105  -  Coverage at voxEU

We analyze the impact of a tax credit on jobs in multinational enterprises. In particular, we exploit the introduction of the 'notional interest deduction' regime in Belgium in 2006, an 'allowance for corporate equity', which aimed to provide an attractive tax system for multinational enterprises active in Belgium. We study employment growth in foreign affiliates of MNEs in Belgium and use as a control group the affiliates of the same MNEs in France. We find that the tax credit has increased employment in Belgian affiliates by 6 to 8 percent over the period 2006-2008.

Productivity Effects of Internationalisation Through the Domestic Supply Chain: Evidence from Europe (with A. Theodorakopoulos)

Published at Journal of Applied Econometrics, version July 2019 - Coverage at voxEU

This paper analyses whether indirect effects of internationalisation occur through the domestic supply chain. We investigate productivity effects for a given firm resulting from the import or export of intermediate inputs by domestic upstream and downstream industries. Using a rich sample of manufacturing firms in 19 EU countries, we find evidence that domestic access to intermediate inputs that are also destined to foreign countries is associated with higher levels of revenue productivity. Further, our results highlight two common, but important, misspecification biases: ignoring the dynamic nature of productivity and estimating a value-added instead of a gross-output production function. 

Downstream Offshoring and Firm-level Employment (with B. Michel)

Published at Canadian Journal of Economics, version May 2016; FEB UGent Working Paper 14/880

When engaging in offshoring, firms do not only import intermediates they used to produce in-house, but also intermediates previously sourced from non-affiliated domestic suppliers. This leads to a negative demand shock for the latter. Prior empirical research has so far neglected this channel through which offshoring may affect employment. We label this demand shock `downstream offshoring' and develop a novel measure capturing its extent for a firm in a given upstream industry. According to our instrumental variables estimations for a representative sample of Belgian manufacturing firms over 1997-2007, downstream offshoring has a robust negative effect on employment.

A constrained nonparametric regression analysis of factor-biased technical change and TFP growth at the firm-level (with Verschelde M., Dewitte R., Dumont M., and Rayp G.) 

Published European Journal of Operational Research entitled Firm-Heterogeneous Biased Technological Change: A Nonparametric Approach under Endogeneity, NBB Working Paper 266

Using firm-level data for Belgium, we study the validity of Hicks neutrality in several sectors that cover the spectrum of knowledge intensity. We find that Hicks neutrality is clearly not supported by the data in different sectors. The results are not sensitive to altering the specification of the technology by including firm age and R&D into the analysis. We also reject Hicks neutrality for a balanced sample, pointing to ‘within-firm’ factor-biased technical change and we also find factor-biased technical change in the pre-crisis era, indicating that unobserved heterogeneity in demand does not drive the results. Overall, our results point towards low-skilled labour- saving and materials-using technical change. So far, this has received little attention and may be linked to offshoring and global value chain networks. Finally, we show that nonparametric estimates of TFP change that allow for factor biases support the evidence of the recent slowdown in TFP growth in many manufacturing sectors in Belgium. Estimations of TFP and technical change are shown to be sensitive to the estimation method and the specification of the factor bias of technical change. 

MNE Heterogeneity and FDI Spillovers (with K. Lenaerts)

published  Review of World Economics, June 2016 version; FEB UGent Working Paper 14/879

This paper analyzes for a panel of Romanian manufacturing firms whether the quality of foreign firms, measured by their productivity level, affects their potential as a source of indirect productivity effects on domestic firms. We find that only sufficiently productive foreign firms generate positive productivity effects on domestic supplier firms. The most productive foreign firms are the main source of productivity effects. Domestic firms with higher productivity levels also enjoy larger total positive productivity effects. When supplying foreign firms that are less productive than themselves, domestic firms experience zero to negative effects. 

Investment-Cash Flow Sensitivity and the Cost of External Finance (with K. Mulier and K. Schoors)

published Journal of Banking and Finance, FEB UGent Working Paper 14/890

We contribute to the investment-cash flow sensitivity debate by creating a new index to identify the supply of finance to firms. We find that firms that are considered constrained according to our index pay a higher interest rate on their debt, and display the highest investment-cash flow sensitivities. Moreover, these findings are not driven by the possible information content of cash flow regarding investment opportunities as we control for oppor- tunities by augmenting our empirical model with firm-level employment growth. We thus provide new evidence consistent with Campbell et al. (2012) that the cost of capital is the driving force behind investment-cash flow sensitivity.

The contribution of start-ups and young firms to industry–level efficiency growth (with Verschelde M., Dumont M., and Rayp G.)

published Applied Economics

This paper examines the impact of start-ups (1 up to 5 years active) and young firms (6 up to 10 years active) on industry-level efficiency growth in six EU countries. Evidence indicates that entrants gradually raise their efficiency level through a phase of learning and adaptation. Starting firms have a high probability to be forced to exit at an early stage but a small share of surviving start-ups explain a disproportionate part of industry-level efficiency dynamics. Results show that efficiency growth of young firms is more important for industry-level efficiency than entry and exit. The relative contribution of efficiency growth of young firms is the only component that is found to have a statistically significant positive impact on industry-level efficency. 

Does it Take Time to Travel Distance?  Geography, Entry Timing and Knowledge Spillovers (with V. Purice)

published Papers in Regional Science, FEB UGent Working Paper 14/896 

This paper investigates the effect of foreign direct investment on the productivity of local firms. We decompose traditional country-wide spillover measures in different components according to both distance between foreign and domestic firms and time- since-foreign-entry. We find larger and faster spillover effects for local suppliers of foreign firms at shorter distance, driven mainly by recent foreign entrants. Irrespective of distance, foreign firms of medium maturity generate backward spillover effects that fade away with longer presence. A positive effect on local competitors is not significantly affected by distance and requires the presence of mature foreign firms.

Supply Chain Fragmentation and Spillovers from Foreign Direct Investment (with K. Lenaerts)

published Economic Systems Research, FEB UGent Working Paper 12/822

The literature on FDI spillovers to domestic firm productivity increasingly points to supply chain linkages with multinational firms as the main channel for positive effects. To determine local and multinational firms’ relative position in the supply chain, the literature relies on input-output tables. For a panel of Romanian firms we show that the level of industry ag- gregation in these tables and the commonly applied definitions for vertical spillovers bear an important impact on results. The use of aggregated input-output tables gives rise to significant and large horizontal spillover effects, whereas backward spillovers tend to be small and only marginally significant. Using detailed input-output tables, backward spillovers become highly significant and dominate horizontal spillover effects whose impact is considerably reduced. Assuming that the true nature of the backward spillover is to be found in a supplier-customer relationship, we show that -for the detailed IO-tables- including within-industry intermediate supply (excluded in the commonly used definition) results in a larger impact of the backward spillover, whereas the horizontal spillovers disappear.

European competitiveness: A semiparametric stochastic metafrontier analysis at the firm level (with Verschelde M., Dumont M., and Rayp G.)

published Journal of Productivity Analysis, ECB Working Paper 1701

In this paper a semiparametric stochastic metafrontier approach is used to ob- tain insight into firm-level competitiveness in Europe. We differ from standard TFP studies at the firm level as we simultaneously allow for inefficiency, noise and do not impose a functional form on the input-output relation. Using AMADEUS firm-level data covering 10 manufacturing sectors from seven EU15 countries, (i) we document substantial, persistent differences in competitiveness (with Belgium and Germany as benchmark countries and Spain lagging behind) and a wide technology gap, (ii) we confirm the absence of convergence in TFP between the seven selected countries, (iii) we confirm that the technology gap is more pronounced for smaller firms, (iv) we highlight the role of post-entry growth for competitiveness.