Bending the rules, breaking the rules: How corruption and lobbying affect the investment market selection of Swedish firms

• Author: Susanna Thede,Nils‐Åke Gustafson
• Date: 01 July 2017
• Published date: 01 July 2017
• DOI: http://doi.org/10.1111/twec.12488

ORIGINAL ARTICLE

Bending the rules, breaking the rules: How

corruption and lobbying affect the investment

market selection of Swedish firms

Susanna Thede

|

Nils-Åke Gustafson

Institute for European Studies, University of Malta, Msida, Malta

1

|

INTRODUCTION

The firm’s decision to engage in foreign direct investment (FDI) depends on production conditions

in the prospective host market. These conditions depend on standard economic factors shaping the

firm’s profit opportunities (such as factor prices and sales prospects) as well as business constraints

encountered in complying with local rules of making business (i.e., regulation). In the large litera-

ture that identifies market determinants of FDI, only a small part considers that business con-

straints are regularly influenced via private–public sector links allowing agents to bend the rules

by lobbying or break the rules through corruption. Yet, these alternatives to overcoming compli-

ance with the rules in a prospective host market could play a key role for the firm’s decision to

invest in that location: there is stark evidence that lobbying affect many policies determining firm

gains from FDI (such as tax and trade policies) and that corruption can be as important as corpo-

rate taxes in deterring FDI (Wei, 2000). From this perspective, it is surprising that the importance

of overcoming regulatory constraints has not gained wider recognition in the FDI literature and

likely that many empirical cross-country investigations in the field are subject to omitted variable

bias.

The boundary between lobbying and corruption is defined by legal compliance (see, e.g., Cam-

pos & Giovanni, 2007; Harstad & Svensson, 2011). Prior research examining firm behaviour on

internal market basis shows that corruption and lobbying function as complements (Damania, Fre-

driksson, & Mani, 2004) or substitutes (Bennedsen, Feldmann, & Dreyer Lassen, 2011; Campos &

Giovanni, 2007). Investigations identifying the sole impact of corruption and lobbying are there-

fore likely to capture the interdependent effect of these influence forms and give rise to misleading

results. To our knowledge, no prior study in the FDI literature investigates the simultaneous effects

that corruption and lobbying have on market attractiveness. We remedy this lack by identifying the

simultaneous and interdependent effects that these private–public sector links have on the firm’s

investment decision. To measure the market infiltration of corruption and/or lobbying, we use

modified Kaufmann (2004) indices constructed from corporate responses to the World Economic

Forum’s 2004 Executive Opinion Survey. The indices are categorised based on whether private–

public sector links are within or outside the legal realm, which allows a narrow identification of

lobbying and corruption activities. The main advantage of this categorisation is that it depicts

undue legal activities as lobbying instead of corruption, which brings our empirical investigation

DOI: 10.1111/twec.12488

1266

|

©2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/twec World Econ. 2017;40:1266–1290.

well in parity with the aim of disentangling the investment impact of a market penetration of these

private–public sector links. We address the implications of that it also restricts us to perform cross-

section analysis in investigating the results’robustness.

We use census data for Swedish manufacturing firms from Statistics Sweden. The data include

information on foreign employment, which is used to construct a sample of firm-market observa-

tions. Building on prior research in the field on firms’FDI engagement, we also source data on

firm characteristics and (2-digit level) industry classifications from this data base to control for

home country factors that affect the firm’s market selection. Probit regressions are performed to

investigate the firm’s decision to invest in a particular market, which lets us use the data that con-

tains complete information on firms’market selection to its best advantage. This distinguishes our

work from related research on the firm’s foreign market investment decision, which relies on sur-

vey data excluding information on limited market engagement, in that our results are not subject to

selection bias created by firm-market data sampling restrictions.

In the two strands of the FDI literature that examine lobbying and corruption effects, the former

foremost consists of theoretical models depicting the lobbying impact on a specific policy that

impacts FDI (such as investment regulation, tax and trade policy) and the latter primarily contains

empirical investigations of the general impact of corruption. As this paper relates more to latter

strand, and the former strand does not (to our knowledge) provide any direct empirical evidence

on multinational firms’lobbying efforts to improve market access, prior studies of relevance are

empirical studies investigating corruption effects on firms’investment market selection (Hakkala

Nilsson, Norb€

ack, & Svaleryd, 2008; Henisz, 2000; Javorcik, 2004; Javorcik & Wei, 2009;

Ledyaeva, Karhunen, & Kosonen, 2013).

1

All these contributions except Henisz (2000) find a stark

investment-deterrent effect of corruption, although Ledyaeva et al. (2013) reveal that their effect

stems from the fact that corruption restricts investment of firms from less corrupt countries that are

in majority in their sample. These studies also reveal that the investment-deterrent effect is larger

for firms using more advanced technology (Javorcik, 2004) and that firms that despite this invest

in corrupt markets are more likely to do so through mergers and acquisitions with local producers

(Javorcik & Wei, 2009).

Henisz (2000) reveals that “unexpected”corruption, which goes beyond that interlinked with

political risk, stimulates the FDI engagement of US multinationals. It is difficult to reconcile

this result with prior evidence in the field. One plausible reason for this discrepancy could be

that his corruption parameter captures the interdependent impact of lobbying. At least, this

explanation is consistent with evidence that firms regularly lobby to influence public sector

outcomes in corrupt markets (Campos & Giovanni, 2007). In this paper, we identify the inter-

dependent effects of lobbying and corruption by computing consistent marginal effects and

performing graphical analysis (as advised by Brambor, Clark, & Golder, 2006; and Greene,

2010).

The rest of this paper is structured as follows. In the next section, we provide a background

description on relevant findings from the scarce-related literature on lobbying and corruption. The

empirical approach is introduced in Section 3, and a data description is provided in Section 4. The

empirical results are presented and discussed in Section 5, and their robustness is examined in

Section 6. In the last section, we provide some concluding remarks.

1

These studies investigate multinationals’investment behaviour using survey data on Swedish manufacturers (Hakkala Nils-

son et al., 2008), US manufacturers (Henisz, 2000), foreign multinational engagements in Eastern European, the former

Soviet Union and/or Central European markets (Javorcik, 2004; Javorcik & Wei, 2009) and survey data on foreign multina-

tional engagement in Russian regions (Ledyaeva et al., 2013).

THEDE AND GUSTAFSON

|

1267