Emerging from war: Public policy and patterns of foreign

Emerging from war: Public policy

and patterns of foreign direct investment

recovery in postwar environments

Robert J. Moore

Department of Management and Industrial

Relations, Shidler College of Business, University of

Hawaii at Manoa, 2404 Maile Way, Honolulu,HI 96822, USA; The East-West Center, 1601 East-

West Road, Honolulu, HI 96848, USA

Correspondence:RJ Moore, Department of Management andIndustrial Relations, Shidler College ofBusiness, University of Hawaii at Manoa,2404 Maile Way, Honolulu, HI 96822, USAe-mail: [email protected]

AbstractThe postwar environment is different from those of active war and established

peace, with risks of violence and political volatility existing alongside renewed

commitments to stability and development. International aid organizations joingoverning institutions in guiding policies for postwar growth. Though

investments here are risky, I argue that governments can clarify key

uncertainties and accelerate the process of recovering FDI by strengtheningpolicy in areas of information transparency, governing accountability, and

engagement with international aid. These ideas are tested with a survival

analysis of inbound FDI recovery using a worldwide sample of postwar periodsfrom 1970 to 2008. I find that while transparency and accountability accelerate

FDI recovery as expected, foreign aid tends to be associated with slower rates of

recovery. Rather than encourage postwar FDI with a commitment todevelopment, aid may be an indirect signal that the environment is yet unfit

for private sector investment. Policymakers and aid organizations should not

rely on aid alone to attract foreign investment in postwar environments.

Structures that encourage investment for social responsibility, with a long-termmarket outlook, may be more successful in these contexts.Journal of International Business Policy (2021) 4, 455–475.https://doi.org/10.1057/s42214-020-00084-4

Keywords: emerging economies; foreign direct investment; institutional environment;political risk; foreign aid; survival analysis

INTRODUCTIONInternational companies can be a productive force toward buildingstability and economic growth in countries recovering from war.Investments add resources to the capital base, and operations canbring jobs, knowledge, technology, and new international linkagesto aid the work of returning to normalcy (Bray, 2005; Forrer &Katsos, 2015). Companies may also take active roles in thepeacebuilding process. Some have leveraged their valuable statusas neutral parties to host peace negotiations, as in South Africaduring the transition from the apartheid regime (Oetzel,

The online version of this article is available Open Access

Received: 28 October 2019Revised: 16 October 2020Accepted: 27 October 2020Online publication date: 7 January 2021

Journal of International Business Policy (2021) 4, 455–475ª 2021 The Author(s) All rights reserved 2522-0691/21

www.jibp.net

http://www.jibp.net/

Westerman-Behaylo, Koerber, Fort, & Rivera, 2010).Others have spearheaded initiatives for peacefulelections, as in Kenya in 2008 (Crawford, 2019).Businesses are often regarded as assets in thesecontexts (Oetzel et al., 2010; Westermann-Behaylo,Rehbein, & Fort, 2015).

International business research has made recentadvances in the study of business phenomena incontexts of active war, but studies of the postwartransition remain rare. Driffield, Jones, and Crotty(2013) outline some common characteristics offirms found to be investing in warring environ-ments, and Dai, Eden, and Beamish (2013, 2017)explore dimensions of subsidiary vulnerability towar-related risks. Oetzel and Getz (2012) describestakeholder pressures and firms’ responses whenoperating in contexts of active war.

The postwar transition is more common instudies of political science and economic develop-ment, as the work of postwar recovery is mostdirectly framed as a policy challenge for govern-ments, aid NGOs, and international banks. Thesestudies recognize the potential of internationalcompanies and foreign direct investments to con-tribute to the process, and, indeed, increasing FDIhas been a stated goal (Turner, Aginam, &Popovski, 2008; Boon, 2006). The gap here, though,is that the managerial perspective of postwar FDI isoften overlooked or reduced to simple functions ofincentive response (Rettberg, 2016). Investmentsfrom international business can aid in postwarrecovery, but the environment presents a rich set ofchallenges to those making investment decisions.

The present study is aimed at the intersection ofthese two streams. The central research question isthis: what features of the policy environmentmight be influential in driving the recovery ofinbound foreign direct investment levels in coun-tries recovering from war?

The question has a policy orientation, but itsinvestigation is significant for theory in interna-tional business as well, characteristic of Lundan’s‘double helix’ illustration of IB and public policy(Lundan, 2018). Investing and operating in postwarzones is not ‘business as usual.’ The heightenedrisks perceived of these environments would typi-cally suggest an avoidance strategy for MNEs (Mills& Fan, 2006; Slangen & Beugelsdijk, 2010). Invest-ments are still made, however, and exploring themin this context can extend the scope and situationalnuance of international business theory. This studytakes its cues from the tradition of emergingeconomy business research, viewing policy

structures according to their potential to reducedifferent types of investment uncertainties facingfirms. The emerging economy paradigm is espe-cially valuable for its focus on the interactionsbetween firms and their contexts. The present studydraws on previous findings regarding informationand accountability policy for FDI, and it extendsthe theory with a unique look at the context ofinternational development aid, a prominent realityamong nations rebuilding after war.The study begins with a review of distinctive

features of postwar environments for business andpolicy, heeding recent calls for thicker descriptionsof operational context (e.g., Meyer, 2015). Hypothe-ses are then developed for three policy domains toaid the recovery of FDI inflows: information policy,accountability policy, and engagement with inter-national development aid. An empirical analysisfollows, testing associations between policy featuresand the rates at which countries recover theirinbound FDI to average levels. The sample includes111 postwar periods, worldwide, over the 38-yearspan from 1970 to 2008.Hypotheses relating information and accountabil-

ity policy to FDI recovery are supported, but asurprising result emerges concerning internationaldevelopment aid. Countries receiving aid frominternational development organizations tend totake longer to recover their FDI inflows rather thanshorter, and thepattern is consistentwith the level ofaid received (more aid corresponds with longerrecovery times). Aid organizations typically have apositive influence on the resources and knowledgeavailable in recovery environments, but, asaddressed in the discussion, they also contribute tothe institutional complexity facing firms, and theytend to operate in places that have been especiallyhard-hit (Collier, Hoeffler, & Soderbom, 2008).The study concludes with a discussion of policy

implications and contributions to IB theory. The-ory and policy are closely aligned here in matters ofgovernment transparency and accountability;results add support to well-established calls tostrengthen independent oversight, distribute thepolicy-making process, and bolster market-facilitat-ing institutions (e.g., Biglaiser & Staats, 2010;Henisz & Zelner, 2005). The context of interna-tional aid poses new considerations, as FDI isshown here to avoid locations where greateramounts of aid are received. The indirect message,that aid-heavy environments are poor for invest-ment, may be coming through more strongly thanthe intended message, that the aid is building a

Postwar FDI Recovery Robert J. Moore

456

Journal of International Business Policy

better environment. Policies that encourage MNEsto invest for social responsibility, with potential forlong-term gain, may be more effective in thiscontext than standard appeals to market strategy(Bray, 2005).

BACKGROUND: THE POSTWAR CONTEXTWar is defined by violence. The definition putforward by the Correlates of War project offersthree criteria: ‘‘sustained combat, involving orga-nized armed forces, resulting in a minimum of 1000battle-related fatalities within a 12-month period’’(Small & Singer, 1982). Postwar, then, is the timefollowing its cessation (or reduction). This violence,and the power conflict among organized forcesbehind it, form the immediate backdrop for policyduring the recovery period.

Legacies of violence are felt in the make-up of thepopulation, with groups of people having fled, beendiverted from jobs or education, and fallen victim infighting (Davies, 2004). Violence degrades physicalresources, infrastructure, and human systems, andthis weakens the ability of organizations to imple-ment new initiatives without first committing thetime and investment to rebuild (Mills & Fan, 2006).Recent violence also brings with it the risk of relapseshould the reconciliation effort fail. This is a validfear, as over a third of postwar countries between1960 and 2002 relapsed to renewed fighting within10 years (Collier et al., 2008).

The tasks involved in building stable systems ofpolicy and governance after a period of breakdowncan be enormous. Priorities for peacebuildinginclude re-establishing the legitimacy and effec-tiveness of formal institutions, re-building security,and spurring economic development (Brinkerhoff,2005; Luiz, Ganson, & Wennmann, 2019). Patternsof governance in immediate postwar environmentsoften exhibit just the opposite, however, withentrenched informal power structures, weak over-sight of security forces, broken systems for essentialservices (health, education, and rule of law), andfiscal arrangements that siphon public assets forfavored elites (Alexander, 1997; Rose-Ackerman,2008; Sawyer, 2008).

Despite complex challenges, building peace isnot always a hopeless prospect. Policy and opera-tions here are often characterized by practicalinitiatives for steadily navigating governance gaps:building coalitions around shared priorities, bal-ancing security oversight among different groups,and leveraging basic economic demands to assist in

growth (Del Castillo, 2008). International busi-nesses can contribute in this work as a stabilizingforce, and the employment opportunities theybring can help reduce risks of further violence(Collier, 2009). Though environmental risks arehigh, they may also see strategic benefits in theprocess. Peacebuilding environments offer oppor-tunities for institutional entrepreneurship and co-evolution, and MNEs can build local legitimacy byengaging in this process (Cantwell, Dunning, &Lundan, 2010). Postwar investments can providecompanies access to new marketplaces, govern-ment incentives, preferred terms with suppliers andresources, and the potential to invest in underval-ued firms (Biglaiser & DeRouen, 2007; Hacioglu,Celik, & Dincer, 2012).

THEORY AND HYPOTHESES: POLICY FOR FDIThe postwar context is new to IB, but prior work inthe field of business in emerging economies may berelevant in guiding its study. Emerging economieshave long been recognized as environments withunique constraints and opportunities facing thosewho seek to do business within them. IB scholar-ship in this area has roots in studies of post-Soviettransition economies in the mid-1990s, and,through key works by Hoskisson, Eden, Lau, andWright (2000), Meyer and Peng (2005, 2016), Xuand Meyer (2013), and others, it has developed intoa consistent paradigm for investigating businessphenomena in volatile, high-risk, and high-growthenvironments. Research in this tradition draws on afew different streams of management theory (in-cluding institutional theory, transaction cost the-ory, and resource-based theories), but what holds ittogether, and what makes it relevant here, is anactive focus on the surrounding contexts for busi-ness, questioning assumptions of environmentalstability, market efficiency, institutional capability,and policy efficacy (Peng, Wang, and Jiang, 2008;Wright, Filatotchev, Hoskisson, & Peng, 2005; Xu &Meyer, 2013).Studies of emerging economies link host country

policy environments with different kinds of uncer-tainty faced by foreign investing firms (Hoskissonet al., 2000). Policy structures influence the market-related information asymmetries facing these firms,which may be exacerbated by weak or biasednational reporting systems, poor informationinfrastructures, and weak or absent formal marketinstitutions. Policy structures may also influenceinvestment uncertainties stemming from political


457


volatility, imposing (or failing to impose) account-ability constraints to minimize abrupt or excessiverent-seeking behaviors from governing institutions.In the absence of effective policy systems, theseuncertainties raise the search costs, transactioncosts, and risks of loss facing firms consideringinvestments in these locations (Meyer & Peng,2016).

There are also prominent features of postwarcontexts that have not yet been subject to thor-ough investigation in relation to business out-comes, and it is here where the theory may beextended. One feature concerns the presence ofinternational development assistance. Rebuildingfrom war is often motivation for a concerteddevelopment agenda, with activity from interna-tional banks, foreign governments, and NGOs(Dobbins, Jones, Crane, & DeGrasse, 2007). Firmsconsidering investments in these areas may con-sider the implications of these activities, as theymay contribute to resolving investment-relateduncertainties. Another factor of these contexts thatshould not be ignored is their experience of recentviolence. Though not policy in the direct sense, thewar legacy may still have wide-reaching effects forbusiness and policy.

Information PolicyPolicy structures to make high-quality marketinformation widely available should facilitateinbound investment in postwar environments(Kelly & Souter, 2014). Information can assist MNEsto realize investment ideas, analyze opportunities,and formulate strategies for entry (He, 2002).Postwar environments often face information-re-lated challenges (damaged distribution structures,untrustworthy sources, negative tone), and theavailability of high-quality information from theseplaces can vary widely (Kang &Meernik, 2005). Thebasic point that information is important inaddressing uncertainty is well studied and intuitive(Amiram, 2012; Tan & Meyer, 2011), so here I focusmore concretely on the kinds of information thatmay be important and their mechanisms for facil-itating high-risk investment.

First, information should be relevant for firms toassess their task environment (Alon & Herbert,2009; Hotho, Lyles, & Easterby-Smith, 2015).Beyond top-level information domains (macro-economic figures, government structures, traderegulations, social norms), there are targeted areasthat can facilitate firms’ assessments of investmentopportunities. These include information about the

structure of different industries, including activefirms and industry-specific stakeholders (guilds,ministries, accreditors), processes for acquiringresources (capital markets, labor pools, domesticcommodities), and location-specific operating costs(tax and remittance structures, transportation,communication, and utilities). This strategicallyrelevant information about postwar environmentsreduces asymmetries and allows investing firms toshift focus from the risk-forward initial connota-tions of recent conflict toward more familiarframeworks of strategic analysis.Relatedly, information should be granular to

differences across regions within the postwar coun-try (Ma & Fitza, 2013). National homogeneity is aneasy assumption to make, particularly for outsiders,but this obscures opportune conditions for invest-ment in areas with operating environments fardifferent from the national norm. These differencescan occur in fully- or semi-autonomous regions, asin Kurdistan of Iraq or Somaliland of Somalia, orwithin the borders of special economic zones,found in Cambodia, Iran, Myanmar, Nigeria, andothers (Farole & Akinci, 2011). Regional differencesintroduce more options to the investment decision.They shift the problem frame to the within-countrylevel, and they demonstrate alternatives to thehigh-risk environment that might be perceivedfrom war-related media coverage.Note that information itself does not reduce the

conditions of risk present in a postwar environ-ment (more information will not give an invest-ment in Sudan a similar profile to an investment inFinland). Rather, it should raise the probability forfirm managers of high-risk or emerging marketstrategies to perceive an investment opportunity(through availability and exposure) and facilitatethe process of analysis and communication amongthe decision team (through strategic relevance andgranularity) (Buckley, Chen, Clegg, & Voss, 2016).Hypothesis 1 is thus framed around informationavailability:

Hypothesis 1: Countries with more marketinformation available will take less time torecover their inbound foreign direct investmentlevels in the postwar years.

Accountability PolicyAccountability on governing institutions is anotherpolicy domain affecting inbound FDI (Busse &Hefeker, 2007). Governments influence the con-duct of their economies through taxes, monetary


458


policy, the enforcement of property rights, andother policy levers, and their actions affect thevalue of foreign investments (Ring, Bigley,D’Aunno, & Khanna, 2005). Postwar governmentsare often volatile, having been recently establishedor challenged, and their potential for wide policyswings constitutes an uncertainty to which invest-ing managers are especially sensitive (Henisz &Zelner, 2005). Elements of the policy environmentthat constrain the agency of individual politicalactors and signal a consistent operating environ-ment can help mitigate these concerns for investors(Delios & Henisz, 2003).

Effective policies in this domain should constrainthe ability of government executives to make large,sudden, or self-serving changes to the policy envi-ronment (Henisz, 2000). Policy structures that fillthis role can include veto powers among differentbranches of government, distributed approvalrequirements, obligations for government repre-sentation from diverse parties, mandated trans-parency concerning public budgets and policyagendas, and the empowerment of a free press(Williams, 2015). Foreign investments often havelarge sunk costs and long payback periods, partic-ularly in recovery environments, and firms mustguard against the risks of rent seeking from volatilegoverning bodies (Henisz & Zelner, 2005). Struc-tural accountability constraints are a way for gov-erning institutions to signal credibility in theabsence of long transaction histories.

Promoting strong formal accountability can beparticularly challenging in a postwar context.Governments face legacies of recent organizedviolence, and the governing environment mayinclude leaders accused of war crimes, active rebelgroups, and political opposition who were recentlybattlefield opposition (Fanthorpe, 2006). Theaccountability mechanisms of interest to MNEscan be sidelined to the process of power consolida-tion and domestic postwar response. This trade-offbetween power and constraint is a complex chal-lenge for all stakeholders to postwar recovery, andthe ways in which postwar governments negotiatethis struggle will be of strong interest to managersconsidering FDI decisions. Hypothesis 2 is thusframed around government accountability:

Hypothesis 2: Countries with a greater degreeof accountability on governing institutions willtake less time to recover their inbound foreigndirect investment levels in the postwar years.

International Development AidEngaging with institutions that provide interna-tional development aid is another policy domainthat can have great relevance for postwar foreigndirect investment. Development aid may includemonetary resources (grants, concessional loans),physical resources (food, construction of hospitalsand schools), and services (election oversight,security training, policy consultation) (Dobbinset al., 2007). Aid is often a significant part of thepostwar policy context, but, while it has receivedmuch attention in the scholarship of economicsand international development, its relationshipwith international business has not yet beenexplored in detail (Garriga & Phillips, 2014). Herewe develop the hypothesis that host countryengagement with development aid should raisethe probability of FDI, based on resource increasesand institutional support structures to reduceinvestment-related uncertainties. The reality iscomplicated, however, and a negative trend mightalso be predicted.Economic theories of development aid posit that

countries can be trapped in cycles of poverty andlack the domestic resources to pull themselves out(Sachs, 2005). War, corruption, landlocked borders,and poor neighbors contribute to the cycle, eatingaway the means for economies to reach self-sustaining levels of growth. International aid ismeant to be an infusion of resources (and, increas-ingly, knowledge) to lift an economy past thethreshold of imminent relapse – the first round ofseeds and fertilizer for a farm that can produce itsown in the future. Aid is a cost for those giving it,but its benefits to welfare, stability, and growth areseen as equally or more valuable than its potentialreturn on a competitive market (Degnbol-Marti-nussen & Engberg-Pedersen, 2005).Institutions providing postwar development aid

should be well positioned to address some of theuncertainties facing managers considering FDI, andaid operations may therefore facilitate investmentin host countries. Liquidity from aid funds canprovide greater purchasing power in domesticmarkets, increasing demand and the potential forreturns to the firm (Mills & Fan, 2006). Aid is oftentied to policy frameworks that promote liberaleconomic policy and transparent governance, andhost countries’ engagement with aid agencies cansend the signal that they hold these values (Santiso,2001). Postwar aid services fill voids in institutionalfunctions that may have been weakened through


459


conflict. These have included provisions of publicsecurity, insurance against political risks, infras-tructure support, human capital training, andothers (Crane & Terrill, 2003; World Bank Group,2006). Viewed this way, and absent other contex-tual information, engagement by the host countryin international development aid should be associ-ated with greater inbound FDI. This positive view ofdevelopment aid leads to hypothesis 3a:

Hypothesis 3a: Countries with strongerengagement with international development aidwill take less time to recover their inbound for-eign direct investment levels in the postwar years.

Aid is controversial, however, both in theory andimplementation, and its provision can just as easilysignal caution to investing managers. Not alleconomists agree with the existence of the povertytrap nor the effectiveness of limited-term infusionsof outside resources to spur sustainable growth(Easterly, 2006; Moyo, 2010). From this lens, aidcontributes to cycles of dependence and undermi-nes local efforts to craft solutions for sustainablegrowth for themselves (Djankov, Montalvo, &Reynal-Querol, 2008). Engaging with aid, then,becomes a signal of an impending crash for thehost country rather than a promise of growth.

Efforts to reconcile these different views have ledto rich scholarship on the contexts and conditionsof aid for development, with insights emerging inthe complicated middle (Banerjee & Duflo, 2011;Riddell, 2008). These studies have identified a fewpitfalls in aid implementation that may raise riskperceptions in managers. Conditions of aid fromforeign governments can be tied to self-servingagendas and may exacerbate existing internationalconflicts (Wardak, 2004). Even if offered in goodfaith, aid resources can be co-opted locally afterdisbursement and fuel local structures of corrup-tion (Winters, 2010). Institution-driven develop-ment aid creates interfaces between disparateinstitutional systems – between donors and hostand among different donors – and navigating theresulting complexity of governance can add to thecosts of operations by international firms (Green-wood, Dıaz, Li, & Lorente, 2010; Meyer, Mudambi,& Narula, 2011). Finally, even outside the sphere ofits direct influence, receiving development aid cansend the signal that the surrounding environmentis in bad shape (by war-related destruction ordysfunction), is in need of help, and is unsuitable to

provide returns on investment (Paris, 2010). Thisnegative view of development aid leads to hypoth-esis 3b:

Hypothesis 3b: Countries with strongerengagement with international development aidwill take more time to recover their inboundforeign direct investment levels in the postwaryears.

In the realm of postwar policy, then, develop-ment aid may have large influences on the FDIprocess, but the direction of association may not beas straightforward as that of the other factorsdiscussed above. Here we seek evidence of a dom-inant international trend, positive or negative. Thiscan inform directions for future international busi-ness study.

Controls: Other Factors for Postwar FDIIn cases of postwar countries containing largestocks of valuable natural resources, we mightexpect inbound FDI at higher levels than otherwise(Asiedu, 2006). These resources can be a motivatingfactor, changing the risk and reward calculations atthe firm level. The influence of resource endow-ments on foreign investment has been long studiedfrom the perspective of international trade, and,while the relationship is not always straightforward(see the resource curse; Robinson, Torvik, &Verdier, 2006, Mehlum, Moene, & Torvik, 2006),it is often in evidence. A measure of naturalresource value is thus included in the analysis as acontrolling factor – influential for FDI but separatefrom immediate policy engagement.Two measures of regime fragility are included to

control for the quality of governing institutions.Postwar governments are typically in a fragile state,but it can be important for FDI to distinguish thosethat might offer more longevity than others. Manyindices of institutional quality combine measuresof stability, information transparency, andaccountability together, but these dimensions aredis-aggregated here to allow a closer investigationof the information and accountability policydomains.FDI recovery might be influenced by levels of

economic and humanitarian development. GDP isincluded here as a standard macroeconomic con-trol for market size, and countries’ pre-war levels ofinbound FDI are included to control for general FDIattractiveness. Countries that have always attracted


460


higher levels of FDI may recover in differentpatterns than those that have not. Further, ameasure of humanitarian development is includedin the analysis, represented here by child mortalityrates. This is an important control for assessinglevels of international aid.

No study of war would be complete without asense of the destruction it brings. Battle deaths areincluded as a controlling factor to represent themagnitude of destruction associated with each war.Heavily destructive wars should leave countries in aweaker state, and foreign investing firms’ percep-tions of destruction may independently influencetheir willingness to commit resources in the imme-diate postwar years. War destruction may be acommon contributing factor to the quality of host-country policy structures, with particularly destruc-tive wars resulting in poorer information, loweraccountability, and potentially higher levels ofinternational aid. Including battle deaths in theanalysis helps control for an endogenous compo-nent effect of these institutional factors on postwarFDI, allowing each to be considered more indepen-dently than otherwise.

METHODS

Dependent Variable: Postwar FDI RecoveryHypotheses are framed around the idea of postwarFDI recovery, an upward trajectory of FDI over timefrom depressed wartime levels to those moreindicative of a country’s normal operations. Thedependent variable for this analysis, then, is thetime-to-recovery: the number of years from the endof the war until inbound FDI is observed at or abovethe country’s long-term average rate.

There are a few reasons for framing the analysisthis way. Panel studies that reveal year-on-yearassociations between FDI and policy characteristicshave already been done successfully (e.g., Busse &Hefeker, 2007; Daude & Stein, 2007), and a simplereplication would not offer much contributionhere. Recovery framing allows each country’s mul-ti-year postwar experience to be the level of focus.Policies take time to implement within host coun-tries, to reach foreign investing decision-makers,and to generate successful and measurable inboundFDI. The process may also be influenced by destruc-tion wrought previously during the active waryears. Specifying the multi-year recovery processas the unit of analysis allows these factors to beincorporated more readily, as these do not fall

neatly into year-by-year variations. Postwar circum-stances are temporary, ending either in a new waror an eventual new normal state, and factors thatare associated with an acceleration of the process ofreaching normal levels should be of practicalinterest for policy and governance.What does it mean for a country to have recov-

ered its inbound FDI? Recovery needs a threshold,and the one used in this study is each country’sadjusted 38-year average yearly inbound FDI level,calculated from observations from 1970 to 2008.Recovery is determined when a country’s singleyear FDI meets or exceeds its adjusted long-termaverage. More details about the recovery definitionand calculation are described in ‘‘Appendix’’.Though not as intuitive as a prewar/postwar FDI

comparison, this method has some advantages. Itcan reflect each country’s overall FDI tendencybetter than a single observation would – it guardsagainst the criterion being influenced by a single-year economic shock. It also remains relevant forcountries whose wars span many years. The mag-nitude of changes in the economies of thesecountries, their neighbors, and the rest of theworld can reduce the relevance of any specificearly-year observation.FDI observations were accessed from the World

Bank Indicators portal (data.worldbank.org), withsource data from the IMF Balance of Paymentsdatabase and supplemented by UNCTAD and othernational sources. ‘War’ and ‘postwar’ years for eachcountry are identified using the records at theCorrelates of War project (Sarkees & Wayman,2010).As a final note, meeting this quantitative defini-

tion for FDI recovery does not mean a country has‘recovered’ in the larger sense that it is conductingall business and daily affairs as if the war had neverhappened. FDI is only one dimension of the overallendeavor.

Independent Variables

Information policyThe degree to which relevant market information isavailable for each country is captured in an indexput forth by Williams (2015). It is composed ofthree sub-indicators, together indicating the quan-tity of market information available, the quality ofthat information, and the infrastructure in place forits distribution. Data sources include the


461


International Monetary Fund, the Bank DisclosureIndex, the Telecommunication InfrastructureIndex, and others (see Williams, 2015 for a com-plete list).

Accountability policyGoverning accountability is indicated by an indexof constraints on national governments, also cal-culated by Williams (2015). This index similarlycontains three sub-indicators (political constraints,media freedom, and fiscal transparency), eachaligning with theoretical constraints on the abilityof governments to make sudden, self-serving, orharmful changes to the policy environment. Sourcedata include reports from the International Devel-opment Association, Reporters Sans Frontieres,Freedom House, and others (see Williams, 2015).

There are other prominent accountability indicesavailable from the World Bank (World Bank Gover-nance Indicators, or WGI), and from TransparencyInternational (Corruption Perceptions Index, orCPI), but the present index is better suited for thisstudy design. Its coverage extends earlier in timethan WGI and CPI, with 1980 as its base yearinstead of 1996 (WGI) or 1995 (CPI). This alignsbetter with the FDI sample, and it allows for alonger period of analysis. Further, neither WGI norCPI scores are consistently comparable over time inthe years before 2008: WGI scores are constructedwith a mean of zero for each year, obscuringcountry-level changes over time, and CPI scoreswere constructed using a matching percentilestechnique prior to 2012, with the result that thescores are interpretable as country rankings ratherthan reflections of indicator values (Williams,2015).

Engagement with international development aidTwo indicators are used for engagement withinternational development aid: net official devel-opment assistance (ODA) received from all sources(e.g., governments, banks, NGOs), and net ODAreceived from regional development banks. Regio-nal bank ODA is considered separately because itrepresents a defined set of external developmentinstitutions with operations within and acrossnational postwar settings. Source data for bothcome from the Development Assistance Committeeof the OECD, accessed through the QWIDS inter-face (stats.oecd.org/qwids). Raw data are reportedin current U.S. dollars, and natural log values areused in this analysis due to the wide range of theraw figures.

Control Variables

Natural resourcesNatural resources are indicated by the total naturalresource rents for each country as a percent ofnational GDP. Data were accessed from the WorldBank Indicators portal, with source data also fromThe World Bank. The indicator is defined as ‘‘thesum of oil rents, natural gas rents, coal rents (hardand soft), mineral rents, and forest rents’’ (WorldBank Group, 2018).

State fragilityTwo indicators are used to control for the fragilityof the governments in this sample: the StateFragility Index from the Polity5 Project, and amarker of whether the war resulted in a regimechange (Center for Systemic Peace, 2020a, b).

Pre-War FDIA measure of the countries’ pre-war FDI is includedto control for their previous level of economicdevelopment and FDI attractiveness. Raw data werereported by UNCTAD and were accessed throughthe World Bank Indicators portal(data.worldbank.org).

Under 5 mortality rateThis is included as a measure of the countries’development along humanitarian dimensions. Thisis important in distinguishing the effects of inter-national aid, as it helps tease apart aid for warrecovery from other humanitarian concerns. Datawere reported by the UN Inter-Agency Group forChild Mortality Estimation (www.childmortality.org).

GDPGDP is included here as a macroeconomic controlfor market size (World Bank Group, 2018). Fig-ures are reported in current US dollars (GDP) andtransformed to the natural log.

Battle deathsBattle deaths are defined as the number of deathscaused by the warring parties that can be directlyrelated to combat. The largest source used in thisstudy is the Uppsala Conflict Data Program Battle-Related Deaths dataset (UCDP BRD version 18.1;Pettersson & Eck, 2018), with raw data collectedfrom three sets of sources: global newswire report-ing, global monitoring of local news by the BBC,


462


http://www.childmortality.org

http://www.childmortality.org

and secondary sources like NGO reports, fieldreports, books, and others (Allansson & Croicu,2017). Battle death estimates for wars outside therange of the UCDP dataset were collected by handfrom news and field reports. The raw distribution ofbattle deaths in this sample is highly skewed, andthis is addressed by transforming to natural logvalues.

AnalysisSurvival analysis models can describe the timespanto an event in relation to the factors hypothesizedto be influential in reducing (or extending) it(Kalbfleisch & Prentice, 1980). There are two com-mon options for survival modeling: proportionalhazards and accelerated failure time. Proportionalhazards models estimate effects of explanatoryvariables on the underlying hazard rate (the prob-ability that the outcome of interest will happen ineach successive year, assumed to be proportionalfrom year to year), whereas accelerated failure timemodels estimate the accelerating or deceleratingeffect of the explanatory variables on the fulltimespan. The accelerated failure time specificationis used here, as it does not require the assumptionthat the underlying hazard rate is proportionalfrom year to year. The observed distribution ofrecovery times, displayed in Figure 1 and discussedbelow, suggests a stronger probability of recoveringin the first few years, with the successive probabilitydropping over time.

In the vocabulary of survival analysis, the obser-vation of a successful FDI recovery constitutes a‘failure’, as it signals that the duration of interest(recovery time) fails to extend past the criterionyear (which, in this case, is good). Countries that

relapse back into war before recovering their FDIlevels are treated as right-censored, as are those thathave not yet recovered by 2008 (the last year of dataavailability). The accelerated failure time method isparametric, and the underlying distribution forrecovery time is specified as the log-normal. Thisdistribution matches the observed data well, withmany recoveries in the first few years and fewer asthe years extend, and it provides a better statisticalfit to the data than other distributions according toAIC and log-likelihood values.The accelerated failure time model estimates, via

maximum likelihood, the effects of explanatoryvariables to accelerate or decelerate the time torecovery (‘failure’). Positive coefficients indicatelonger recovery times, and negative shorter. Themagnitude of the effect can be calculated bymultiplying the baseline recovery expectation byexp(v b), where v is the explanatory variable and b isthe coefficient (Mills, 2011). In practice, this alignsclosely to calculations of percent difference: acoefficient of 0.5 corresponds with a recovery timethat is approximately 50% longer than otherwiseexpected.

Distributions and CorrelationsThe distribution of recovery periods, illustrated inFigure 1, offers some insights, as this is the firsttime that patterns of postwar FDI recovery havebeen described in a worldwide sample. The averagetime to FDI recovery after wars between 1970 and2008 is 4.5 years, and experiences range from 1 to19 years. Some of the longer-lasting successfulpostwar recovery periods are those in Ugandafollowing the Ugandan Bush War (1986), and inGeorgia following the Georgian Civil War (1994).Sixteen percent of countries relapse to a new warbefore reaching recovery threshold for FDI. Theincidence of FDI recovery typically starts high, with35% of countries recovering in 1 year, then dropsoff to a steadily lessening pattern of recoverydurations from 2 to 19 years, experienced by theremaining 49% of countries. The large number ofcountries recovering in 1 year is interesting, and aspecific study of this group may yield deeper insightin future research. It is not unexpected, however,and it aligns with postwar recovery patterns of GDPas described by Flores and Nooruddin (2009).Table 1 presents correlations and descriptive

statistics for all variables. Some correlations areinteresting in their direction and significance. Aidfrom development banks is correlated negativelywith accountability policy and GDP, suggestingFigure 1 Distribution of postwar recovery times.


463


that these development institutions have a strongerpresence in countries with poor environments forFDI. These relationships are hidden when com-bined within total development aid. Naturalresources correlate positively with state fragilityand negatively with accountability policy. Theserelationships might follow from environments withsymptoms of a resource curse (more on this in thediscussion).

Correlations are high between information pol-icy and accountability policy, and between statefragility and under 5 mortality. Williams notes thetheoretical connection between information policyand accountability policy in his indices; it tends tobe the case that governments that value account-ability also produce better information (Williams,2015). The multicollinearity of these variable pairscontributes to a variance inflation factor of 4.5when they are included simultaneously in the fullFDI survival model. To address this source of error,the final model is estimated with one variable outof each correlated pair removed.

ResultsTable 2 presents the survival models. Model 0includes control variables only. Models 1–4 presentcoefficients for each independent variable sepa-rately: information policy, accountability policy,then both measures of international aid. Model 5 isthe full model with all variables together, andModel 6 is the final, adjusted to account formulticollinearity among highly correlatedvariables.

Information and accountabilityMeasures of information policy and accountabilitypolicy predict faster FDI recovery, offering supportfor Hypotheses 1 and 2. Coefficients are negative, asexpected, and they suggest a 2–3% reduction in FDIrecovery time per degree of strength in these policydomains. This is observed both in their partialmodels (Models 1 and 2) and when informationpolicy is included with development bank aid inthe final model (Model 6). The effect of account-ability loses significance when information andaccountability are estimated simultaneously inModel 5, but this is likely a result of their highcorrelation. The effects are fairly small, but, withsuch a large potential for noise, it is notable thateffects emerge consistently.

Tab

le1

Descriptive

statisticsandco

rrelations

Variable

Mean

SD

Min

Max

01

23

45

67

89

10

0.Reco

very

duration

4.51

4.24

119

1.Inform

ationavail.

42.09

14.03

580.50

-0.05

2.Accountability

33.75

16.30

10

77.50

-0.03

0.63*

3.Internationalaid,all

sources(log)

19.24

1.53

13.49

23.11

-0.03

0.21*

0.33*

4.Internationalaid,

developmentbanks

(log)

6.88

7.47

018.51

0.23*

-0.13

-0.23*

0.31*

5.Naturalresources

14.92

16.55

082.62

-0.06

-0.24*

-0.36*

-0.10

0.13

6.Regim

ech

ange

0.58

0.50

01

0.03

0.09

0.16

0.01

0.11

-0.05

7.State

fragility

index

15.70

5.43

325

-0.16

-0.46*

-0.27*

0.19*

0.40*

0.25*

0.13

8.Pre-warFD

I(log)

18.30

2.60

9.21

24.46

0.14

0.37*

0.23*

0.08

-0.09

-0.05

-0.07

-0.25*

9.Under5mortality

106.41

70.25

8.94

282.5

-0.23

-0.54*

-0.31*

-0.08

0.31*

0.21*

0.09

0.72*

-0.28*

10.GDP(log)

23.03

1.82

18.88

27.45

0.05

0.47*

0.28*

0.22*

-0.20*

-0.10

-0.29*

-0.40*

0.50*

-0.46*

12.Battle

deaths(log)

8.19

1.48

4.96

13.00

0.02

-0.14

-0.11

0.07

0.12

0.07

0.09

0.16*

-0.15

0.08

0.06

*p\

.05


464


Tab

le2

Acceleratedfailu

retimemodels(lognorm

al)forpostwarFD

Ireco

very

associatedwithdomainsofpublic

policy

Variables

Model0

Model1

Model2

Model3

Model4

Model5

Model6

1.Inform

ationpolicy

(H1)

-0.03**

-0.03*

-0.03**

(0.01)

(0.01)

(0.01)

2.Accountability

policy

(H2)

-0.02*

0.00

(0.01)

(0.01)

3.Internationalaid

(allsources)

(H3)

-0.02

-0.13

-0.17

(0.08)

(0.10)

(0.09)

4.Internationalaid

(developmentbanks)

(H3)

0.05***

0.05**

0.05**

(0.01)

(0.02)

(0.02)

5.Naturalresources

0.00

0.00

-0.01

0.00

0.00

-0.01

-0.01

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

6.Regim

ech

ange

0.05

0.31

0.22

-0.05

0.07

0.14

0.12

(0.24)

(0.24)

(0.24)

(0.25)

(0.22)

(0.24)

(0.24)

7.Fragility

index

-0.01

-0.04

-0.04

-0.01

-0.03

-0.04

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

8.Pre-w

arFD

I0.06

0.06

0.05

0.05

0.04

0.06

0.05

(0.05)

(0.05)

(0.05)

(0.05)

(0.04)

(0.05)

(0.05)

9.Under5mortality

0.00

0.00

0.00

0.00

0.00

0.00

0.00*

(0.00)

(0.00)

(0.00)

(0.00)

(0.00)

(0.00)

(0.00)

10.Lo

gGDP

-0.06

0.01

-0.04

-0.07

-0.05

0.01

0.03

(0.07)

(0.08)

(0.08)

(0.08)

(0.07)

(0.08)

(0.08)

11.Battle

deaths

0.15*

0.06

0.08

0.17*

0.11�

0.06

0.05

(0.07)

(0.07)

(0.07)

(0.07)

(0.06)

(0.07)

(0.07)

Tim

eatrisk

(years)

536

519

519

499

536

482

482

Loglikelih

ood

-149.14

-127.39

-128.49

-142.09

-143.22

-117.27

-117.99

N111

100

100

106

111

95

95

�p\

.1

*p\

.05

**p\

.01

***p

\.001


465


International aidSomewhat surprising here is the result that interna-tional aid from regional development banks predictsa longer time to recovery (Model 4). This offersinitial support for the alternate view of internationalaid, as described in Hypothesis 3b, that aid mightsignal caution more so than opportunity. Thecoefficient is positive, with a magnitude of about5%, and the effect remains significant when esti-mated with other policy factors in the full and finalmodels. An investigation of a binary measure ofregional development bank aid is presented in thenext section to explore the effect of its merepresence. Levels of net international aid aggregatedfrom all donor sources (countries, banks, NGOs)show no significant association with postwar FDIrecovery times. This may indicate a more complexset of relationships among aid agencies, donorgovernments, host governments, and foreign invest-ing firms than this analysis allows.

ControlsNatural resources also fail to consistently predictFDI recovery rates in this analysis. This is surpris-ing, and the result is addressed further in thediscussion. There are likely still effects from naturalresources in practice, but they may not be in thesame direction from country to country. Othercontrolling factors show correlations in theexpected directions (fragility and child mortalityare negatively correlated with information andaccountability policy; GDP and pre-war FDI arepositively correlated), but they fall short of showingconsistent effects on FDI recovery times in themodels. State fragility and child mortality are issuesthat often overlap with policy and aid concerns inpostwar states, and they should not be ignored.Wartime casualty levels predict longer FDI recoveryin some models, and there remains the possibilityfor indirect effects through their influence on otherexplanatory variables. Wars with higher casualtiesmay lead to weaker policy structures and higherlevels of regional aid in the postwar period. This isexplored further in the next section.

Repeat warsA limitation of the survival analysis method is thatit does not account for trends in recovery patternsthat might be present from multiple wars for asingle country; each war is assessed on its own. Afew alternate estimations were conducted to inves-tigate the possibility of within-country trends. As afirst step, the original AFT survival analysis series

was applied to a sample restricted to each country’sfirst war. This was followed by a similar analysis ofeach country’s latest war. The effects of policy andaid variables maintained their direction and signif-icance in these estimations, suggesting that thepolicy effects in the full model were not falselyachieved through within-country trends. P valueswere slightly higher in these models, but stillsignificant, and this was likely due to the necessaryreductions in sample size. As a final step, the samevariables were estimated with a series of multi-levelmixed effects regression models, with a group termadded to account for error correlations at thewithin-country level (Hox, 2010). The limitationof this method, and the reason it was not used forthe main presentation, is that it cannot accommo-date censored observations, which in these datainclude all observations of relapse-before-recovery.Despite this limitation, the models again revealedpolicy and aid effects that were consistent indirection and significance with the main results.More details and full results of these alternateestimations are available from the author.

Further Analyses

Regional development bank aidNot every country receives regional developmentbank aid in its postwar years. Does it matter morefor FDI that they receive a certain level of aid orthat they receive aid at all? I explore this byreplacing the continuous aid variable in Model 4with a binary variable representing aid receipt (1) ornot (0). The result, presented in Table 3, is a muchstronger effect for the aid receipt indicator (binary)than the aid level indicator (continuous). Thissuggests that recovering countries are groupeddistinctively according to whether they receiveregional aid, and those that do receive this aidtend to take about 73% longer to recover FDI levels.

Battle deathsThe wartime destruction associated with humanviolence may have an influence on the policystructures in these countries, and it may justifyhigher levels of regional development bank aid. Asthese circumstances are associated with longer FDIrecovery times, they may facilitate an indirect effectbetween wartime casualties and FDI recovery rates.To explore this possibility, I conduct a multiple

mediation analysis with FDI recovery time as thedependent variable, battle deaths as the indepen-dent variable, information policy and regional


466


development bank aid as mediators, and the full setof controls (Preacher & Hayes, 2008). The first stepis to estimate the relationships between all variablesindividually to determine their path coefficients.The first set of path coefficients (battle deaths toinformation policy and development aid) are esti-mated by OLS regression, since the dependentvariable is not duration dependent. The secondset (to recovery time) are estimated by AFT survivalanalysis – the same as Model 6 in the originalanalysis (presented in Table 2). The path coeffi-cients that precede and follow each mediator arethen combined to determine their indirect effects.The delta method is used for combining the pathcoefficients and determining their statistical signif-icance (Davidson & MacKinnon, 2004).

Results, presented in Figure 2 and Table 4, sug-gest that battle deaths do have an indirect effect onlengthening FDI recovery times through effects onpolicy and aid. The effect underscores the pervasiveand long-lasting nature of legacies of violence inthese places. The magnitude of the effect is fairlysmall, but, when combined with the primaryresults, it contributes to a picture of complexityand long-lasting challenges in these countriesstemming from recent violence.

Table 3 AFT coefficients (lognormal) for postwar FDI recovery associated with international aid receipt (binary)

Variables Control Model 4 (alt)

4. International aid receipt (development banks) 0.73***

(0.21)

5. Natural resources 0.00 0.00

(0.01) (0.01)

6. Regime change 0.05 0.07

(0.24) (0.22)

7. State fragility index - 0.01 - 0.03

(0.03) (0.03)

8. Pre-war FDI 0.06 0.04

(0.05) (0.04)

9. Under 5 mortality 0.00 0.00

(0.00) (0.00)

10. Log GDP - 0.06 - 0.05

(0.07) (0.07)

11. Battle deaths 0.15* 0.12�

(0.07) (0.07)

Time at risk (years) 536 536

Log likelihood - 149.14 - 143.31

N 111 111

�p\ .1

*p\ .05

**p\ .01

Table 4 Indirect effects of battle deaths on FDI recovery rates

Mediator Coefficient, total

indirect effect

Std.

err.

p

value

Regional development

bank aid

0.01 0.01 0.15

Information policy 0.01� 0.00 0.06

Combined effect 0.02* 0.01 0.02

�p\ .1

*p\ .05

Figure 2 Mediated paths between battle deaths and FDI

recovery.


467


DISCUSSIONWe now return to the original research question:what features of the policy environment might beinfluential in driving the recovery of inbound FDIlevels in countries recovering from war? Threepolicy domains were analyzed (information,accountability, and development aid), premisedon their ability to reduce context-specific uncer-tainties for firms considering investment. Positiveeffects of information and accountability policy aresupported in analysis, with consistent associationsbetween these policies and shorter FDI recoverytimes, but evidence of the reverse is found for hostcountry engagement with international develop-ment aid. These findings and their policy implica-tions are discussed below, followed by a discussionof what the study and results contribute to thelarger body of IB theory.

Summary and Policy ImplicationsPostwar environments are often information-poor,and those with policy structures in place that makemore relevant market information available can bemore successful in attracting investment in earlyyears. Market information can increase the boundsof decision-makers’ rationality, reduce firm costs ofanalysis, and allow the focus of assessment to movebeyond war-related headlines (Tan & Meyer, 2011;Mills & Fan, 2006). The pattern is supported here inempirical analysis – countries that made moreinformation available consistently recovered theirFDI levels faster than otherwise.

Similarly, postwar policy structures that signalaccountability – the ability of governing institu-tions to credibly commit to a set of policies – can bemore successful in attracting investment in earlyyears. The potential for wide policy swings consti-tutes an important source of risk and uncertaintyfor investing firms, and the years immediately afterwar are often volatile, with new governmentsestablishing themselves or old ones re-assertingcontrol (Henisz & Zelner, 2005). Those that incor-porate policy controls to constrain the potential forharmful swings may lower an important barrier tofirm resource commitment (Henisz, 2000). This,too, is supported by the present analysis – thosecountries with stronger accountability mechanismsconsistently recovered their FDI levels faster thanotherwise.

Results speak to the importance of transparencyin the internal governance of these countries, bothin the sense of making important information

widely available and in distributing the mecha-nisms of policy change. Policymakers working inthese challenging environments may take heartthat efforts toward internal transparency showconsistent associations with faster FDI recovery.Results are well aligned with the common refrainfrom theory that opacity, corruption, and otherrestrictive patterns are often destructive at themacro level, despite any specific short-term gainthey might provide to the actors involved (Javorcik& Wei, 2009; Le Billon, 2008). To re-affirm a fewpoints collected from prior research, market infor-mation may be especially helpful for the FDIprocess if it is oriented to the strategic environmentfacing firms, specific to industries, supplies,resources, regulations, and other strategic factors(Alon & Herbert, 2009). Systems of information andreporting that are specific across regions withincountries can contribute as well, as the strategicenvironment can vary widely and contain oppor-tunities that are obscured at the national level(Farole & Akinci, 2011; Ma & Fitza, 2013).The Iraq Stock Exchange (ISX) illustrates a policy

initiative that created and published firm-relevantmarket information with some success during Iraq’spostwar transition period. Established in 2004, theISX is self-regulated and operates under the over-sight of the Iraq Securities Commission, an institu-tion independent of direct political influence.When it was established this represented an impor-tant shift away from the structures of strict marketcontrol in place during the reign of Saddam Hus-sein, and it allowed consumers to build confidencethat the information reflected market realities moreso than regime priorities (Hassan, Rankin, & Lu,2014). In its operations, which continue to this day,the ISX offers real-time information about corpo-rate and industry activity, pricing trends, andsector-level market characteristics. The ISX wasone element in a period of strong economic growthin postwar Iraq, reflected in trading volumes,market caps, FDI levels, and GDP per capita (IraqStock Exchange, 2004–2018; World Bank Group,2018).The presence and level of international develop-

ment aid in postwar countries was also investigatedin relation to FDI. Though the mission of interna-tional aid is often to provide resources to helpfacilitate economic recovery, the effectiveness ofthis help is not always apparent, neither in eco-nomic theory nor in practical application (Banerjee& Duflo, 2011). The dominant hypothesis here wasthat international aid should be associated


468


positively with FDI, based on the uncertainty-reducing effect of the resources provided and theconfidence they signal in the potential for thecountry’s recovery (Garriga & Phillips, 2014).

This was not supported in analysis, and in fact anegative association was found between FDI recov-ery and aid from regional development banks (agroup that includes the African Development Bank,Inter-American Development Bank, Arab Bank forEconomic Development, and others). Policy struc-tures based on thepositive viewmaybe ineffective toattract FDI in these postwar contexts. Recommen-dations are tough here, as aid institutions often doprioritize attracting a range of private sector actorstoward postwar development (World Bank Group,2006). It may be the case, however, that the oppor-tunities contained in these efforts are not yet reach-ing their fullest audiences. A case studyof recovery inBosnia suggested that pro-investment MIGA pro-grams may have been targeted too narrowly at largeMNEs, ignoring the smaller but more abundantrange of regional SMEs that can cross nearby bordersto access postwar markets (Bray, 2004). Followingfrom institutional theory in IB, the added opera-tional complexity in environments with a large aidinstitution presence may also be a deterrent factor.These institutions may overlap in function withgoverning institutions, yet they expect differentnorms of interaction (Meyer et al., 2011; Santiso,2001). Aid institutionsworking to attract FDI shouldtake care to reduce policy incongruities that may berisky or costly for firms to negotiate.

Further, the contexts with the poorest economicindicators tend to host more aid. In these places itmay be productive for aid institutions to enlistforeign investment in terms of corporate socialresponsibility, with potential for long-term gain,rather than in standard strategic terms. There issome evidence for the success of this approach.Swedish telecom firm Ericsson operated a post-conflict-focused response program called ‘First onthe Ground’ as part of its CSR portfolio, bringingnetwork connectivity to postwar relief efforts. Thiswas done in partnership with a range of aidorganizations, and it operated in Afghanistan in2002, Liberia in 2003, and others. This work wasdriven to enhance the aid relief effort, but it hadlong-term strategic benefit as well, allowing earlyaccess to growth markets and enhancing its imagewith influential international players (Bray, 2005).Other postwar CSR activities have been reported incase studies from Sierra Leone, Sudan, and Somalia(Bray, 2005; Cronin, 2004).

Natural resources were expected to associate withfaster FDI recovery as a control, but no consistentrelationship was found in this analysis. Priorempirical studies have pointed to both positiveand negative associations between naturalresources and economic growth, and these mayall be at play in recovering countries. On one side,the idea of a ‘‘resource curse’’ has fueled debateamong economists; this is the idea that high levelsof valuable natural resources can encourage cor-ruption and hinder development in countries withlow quality institutions (Robinson et al., 2006). Onthe other side, the same combination of highresources and low-quality institutions has beenshown to be particularly attractive to foreigninvestment from China, and indeed Chinese firmshave been active in many recovering countries inAfrica, Southeast Asia, and the Middle East (Kolstad& Wiig, 2012; Sutherland, Anderson, Baily, & Alon,2020). In aggregate in this sample, however, theseforces and others combine to obscure a consistentpredictive relationship.A limitation of the empirical analysis is that it

can only reveal these patterns at the country level,relying on the investment levels themselves toserve as evidence of firms’ investment decisions.This makes no distinction between the types offirms investing, their comparative resource com-mitments, or their individual decision processes(Beugelsdijk, Hennart, Slangen, & Smeets, 2010).This was a trade-off of using policy data at thenational level, necessary for comparing the experi-ences of different policy environments with FDIacross different post-violent contexts. Firm charac-teristics can be investigated with more detail insingle-country cases, but in this study the broadview was prioritized.Another limitation, and a door for future research,

is that this study leaves open questions of morenuanced qualitative and categorical differencesamong postwar environments. Recovery might takedifferent directions for former colonies of England,France, Spain, or Portugal, for example, or of thosethat had formerly been unified in a larger country,like Yugoslavia or the USSR (Bennett, Faria, Gwart-ney, &Morales, 2017). The Correlates ofWar projectidentifies nine different types of war, depending onthe status of the states involved and the nature of theconflict, and each might set the stage for uniquerecovery concerns (Sarkees & Wayman, 2010). Thestudy as presented focuses onpolicy, aid, andnaturalresource concerns that might be relevant for anystate that has experienced a recent violent conflict,


469


but further studies of the differences among themmay yield new insight.

Contribution and ConclusionTo IB theory, this study extends a framing fromemerging economy business research to postwarrecovery environments, emphasizing the effect ofpolicy to reduce context-related uncertainties facinginvesting firms (Meyer& Peng, 2016). This approachis supported in postwar settings for policy structuresthat address information gaps and political uncer-tainties. Results adds strength to its explanatorypower and extend its relevance to an additional anddistinct set of contextual features. Results also sup-port the broader insight from emerging economyresearch that when policy and institutional envi-ronments areunstable, uncertain, and/or inefficient,these concerns gain prominence as drivers (orrestrainers) of FDI – they have the potential toovercome effects that might be predicted by tradi-tional resource- or market-based logic (Peng, Wang,and Jiang, 2008). The policy variables analyzed inthis study led tomore consistent FDI outcomes thanthe resource and economic control factors.

The results concerning international aid offer anew layer of complexity to this view. Rather thanencourage FDI by addressing uncertainties relatedto resource availability and commitments to stabil-ity, as might be expected, international aid fromregional development banks consistently and pro-portionally associate with weaker FDI recovery inthis postwar sample. The presence of this aid mayindeed be resolving uncertainties for foreign firms,but the message appears to be stronger in itsimplications of poor surroundings than of anyintended additional market strength or stability tobe developed. The relationship between

international aid and FDI is not so straightforward,and it is notable for theory that the indirect signal,based on context, is supported more strongly herethan the direct one, based on intended effect. Thisis worth exploring further, both in theory andpractice. Development aid has been the focus ofmuch study in its capacity to influence economicoutcomes for nations, but its specific interactionswith foreign investment, firm decision-making,and firm-driven development remain open.International businesses can contribute to the

work of postwar rebuilding, and in doing so theypush the bounds of common expectations ofcorporate behavior. It is hoped that the researchpresented here can contribute to the larger effort ofpromoting peaceful recovery by connecting policystructures to FDI recovery trends. The author alsohopes to encourage future research in this impor-tant and (unfortunately) persistent context.

ACKNOWLEDGMENTSMany people lent their considerable expertise andgoodwill toward the development of this research. Ithank Kiyohiko Ito, John Butler, Marjan Houshmand,Wei Huang, Ronald Heck, Charlotte Hildebrand,Teresa Hinnerichs, and Peter Rowan from the Univer-sity of Hawai 0i, and Mary Hammond, Ann Hartman,and Kuhio Vogeler from the East-West Center. I amalso extremely grateful to JIBP Area Editor Hafiz Mirzaand anonymous reviewers for their patient andthoughtful guidance throughout the review process.The study was inspired by management experiences atthe American University of Iraq, Sulaimani, during theinter-war period of 2012–2014.

REFERENCESAlexander, J. 1997. The local state in post-war Mozambique:Political practice and ideas about authority. Africa, 67(1): 1–26.

Allansson, M., & Croicu, M. 2017. UCDP battle-related deathsdataset codebook, version 18.1. Uppsala Conflict DataProgram.

Alon, I., & Herbert, T. T. 2009. A stranger in a strange land:Micro political risk and the multinational firm. BusinessHorizons, 52(2): 127–137.

Amiram, D. 2012. Financial information globalization andforeign investment decisions. Journal of International Account-ing Research, 11(2): 57–81.

Asiedu, E. 2006. Foreign direct investment in Africa: The role ofnatural resources, market size, government policy, institutionsand political instability. The World Economy, 29(1): 63–77.

Banerjee, A. V., & Duflo, E. 2011. Poor economics: A radicalrethinking of the way to fight global poverty. New York:PublicAffairs.

Bennett, D. L., Faria, H. J., Gwartney, J. D., & Morales, D. R.2017. Economic institutions and comparative economicdevelopment: A post-colonial perspective. World Development,96: 503–519.

Beugelsdijk, S., Hennart, J. F., Slangen, A., & Smeets, R. 2010.Why and how FDI stocks are a biased measure of MNE affiliateactivity. Journal of International Business Studies, 41(9): 1444–1459.

Biglaiser, G., & DeRouen, K. 2007. Following the flag: Troopdeployment and US foreign direct investment. InternationalStudies Quarterly, 51(4): 835–854.


470


Biglaiser, G., & Staats, J. L. 2010. Do political institutions affectforeign direct investment? A survey of US corporations in LatinAmerica. Political Research Quarterly, 63(3): 508–522.

Boon, K. 2006. ‘‘Open for Business’’: International financialinstitutions, postconflict economic reform, and rule of law. InProceedings of the ASIL annual meeting (Vol. 100: 142–144).Cambridge: Cambridge University Press.

Bray, J. 2004. MIGA’s Experience in Conflict-Affected Countries: TheCase of Bosnia and Herzegovina. Conflict Prevention & Recon-struction, Environmentally and Socially Sustainable Develop-ment Network, World Bank.

Bray J. 2005. International companies and post-conflict recon-struction: Cross-sectoral comparisons. Social DevelopmentPapers, No. 22. The World Bank.

Brinkerhoff, D. W. 2005. Rebuilding governance in failed statesand post-conflict societies: Core concepts and cross-cuttingthemes. Public Administration and Development, 25(1): 3–14.

Buckley, P. J., Chen, L., Clegg, L. J., & Voss, H. 2016. Experienceand FDI risk-taking: A microfoundational reconceptualization.Journal of International Management, 22(2): 131–146.

Busse, M., & Hefeker, C. 2007. Political risk, institutions andforeign direct investment. European Journal of Political Econ-omy, 23(2): 397–415.

Cantwell, J., Dunning, J. H., & Lundan, S. M. 2010. Anevolutionary approach to understanding international busi-ness activity: The co-evolution of MNEs and the institutionalenvironment. Journal of International Business Studies, 41(4):567–586.

Center for Systemic Peace. 2020a. Polity5 Project. Retrieved July30, 2020, from www.systemicpeace.org.

Center for Systemic Peace. 2020b. State Fragility Index andMatrix. Retrieved July 30, 2020, from www.systemicpeace.org.

Cerra, V., & Saxena, S. C. 2008. Growth dynamics: The myth ofeconomic recovery. American Economic Review, 98(1): 439–457.

Collier, P. 2009. Post-conflict recovery: How should strategies bedistinctive? Journal of African Economies, 18(suppl_1): i99–i131.

Collier, P., Hoeffler, A., & Soderbom, M. 2008. Post-conflictrisks. Journal of Peace Research, 45(4): 461–478.

Crane, C. C., & Terrill, W. A. 2003. Reconstructing Iraq: Insights,challenges, and missions for military forces in a post-conflictscenario. Report, U.S. Army War College, Strategic StudiesInstitute.

Crawford, V. 2019. 7 ways business can be agents for peace.World Economic Forum, 28 May 2019. Retrieved August 2,2020, from https://www.weforum.org/agenda/2019/05/7-ways-business-can-be-agents-for-peace/.

Cronin, J. 2004. A mobile vision for Africa. BBC News Online, 5July, 2004. Retrieved August 2, 2020, from http://news.bbc.co.uk/2/hi/business/3854495.stm.

Dai, L., Eden, L., & Beamish, P. W. 2013. Place, space, andgeographical exposure: Foreign subsidiary survival in conflictzones. Journal of International Business Studies, 44(6): 554–578.

Dai, L., Eden, L., & Beamish, P. W. 2017. Caught in the crossfire:Dimensions of vulnerability and foreign multinationals’ exitfrom war-afflicted countries. Strategic Management Journal,38(7): 1478–1498.

Daude, C., & Stein, E. 2007. The quality of institutions andforeign direct investment. Economics & Politics, 19(3): 317–344.

Davidson, R., & MacKinnon, J. G. 2004. Econometric theory andmethods. New York: Oxford University Press.

Davies, L. 2004. Building a civic culture post-conflict. LondonReview of Education, 2(3): 229–244.

Degnbol-Martinussen, J., & Engberg-Pedersen, P. 2005. Aid:Understanding international development cooperation. Trans:Bille, M. London: Zed Books Ltd.

Del Castillo, G. 2008. Rebuilding war-torn states: The challenge ofpost-conflict economic reconstruction. Oxford: Oxford Univer-sity Press.

Delios, A., & Henisz, W. J. 2003. Political hazards, experience,and sequential entry strategies: The international expansion ofJapanese firms, 1980-1998. Strategic Management Journal,24(11): 1153–1164.

Djankov, S., Montalvo, J. G., & Reynal-Querol, M. 2008. Thecurse of aid. Journal of Economic Growth, 13(3): 169–194.

Dobbins, J., Jones, S. G., Crane, K., & DeGrasse, B. C. 2007. Thebeginner’s guide to nation-building. Santa Monica: The RANDCorporation.

Driffield, N., Jones, C., & Crotty, J. 2013. International businessresearch and risky investments, an analysis of FDI in conflictzones. International Business Review, 22(1): 140–155.

Easterly, W. 2006. The white man’s burden: Why the West’s effortsto aid the rest have done so much ill and so little good. NewYork: Penguin Press.

Fanthorpe, R. 2006. On the limits of liberal peace: Chiefs anddemocratic decentralization in post-war Sierra Leone. Africanaffairs, 105(418): 27–49.

Farole, T., & Akinci, G. 2011. Special economic zones: Progress,emerging challenges, and future directions. Washington, DC:The World Bank.

Flores, T. E., & Nooruddin, I. 2009. Democracy under the gun:Understanding postconflict economic recovery. Journal ofConflict Resolution, 53(1): 3–29.

Forrer, J., & Katsos, J. E. 2015. Business and peace in the buffercondition. Academy of Management Perspectives, 29(4): 438–450.

Garriga, A. C., & Phillips, B. J. 2014. Foreign aid as a signal toinvestors: Predicting FDI in post-conflict countries. Journal ofConflict Resolution, 58(2): 280–306.

Greenwood, R., Dıaz, A. M., Li, S. X., & Lorente, J. C. 2010. Themultiplicity of institutional logics and the heterogeneity oforganizational responses. Organization Science, 21(2): 521–539.

Hacioglu, U., Celik, I. E., & Dincer, H. 2012. Risky business inconflict zones: Opportunities and threats in post conflicteconomies. American Journal of Business and Management,1(2): 76–82.

Hassan, E. A., Rankin, M., & Lu, W. 2014. The development ofaccounting regulation in Iraq and the IFRS adoption decision:An institutional perspective. The International Journal ofAccounting, 49(3): 371–390.

He, C. 2002. Information costs, agglomeration economies andthe location of foreign direct investment in China. RegionalStudies, 36(9): 1029–1036.

Henisz, W. J. 2000. The institutional environment for multina-tional investment. The Journal of Law, Economics, and Organi-zation, 16(2): 334–364.

Henisz, W. J., & Zelner, B. A. 2005. Legitimacy, interest grouppressures, and change in emergent institutions: The case offoreign investors and host country governments. Academy ofManagement Review, 30(2): 361–382.

Hoskisson, R. E., Eden, L., Lau, C. M., & Wright, M. 2000.Strategy in emerging economies. Academy of ManagementJournal, 43(3): 249–267.

Hotho, J. J., Lyles, M. A., & Easterby-Smith, M. 2015. The mutualimpact of global strategy and organizational learning: Currentthemes and future directions. Global Strategy Journal, 5(2):85–112.

Hox, J. J. 2010. Multilevel analysis: Techniques and applications(2nd ed.). New York: Routledge.

Iraq Stock Exchange. 2004-2019. Yearly Reports. RetrievedFebruary 15, 2020, from http://isx-iq.net.

Javorcik, B. S., & Wei, S. J. 2009. Corruption and cross-borderinvestment in emerging markets: Firm-level evidence. Journalof International Money and Finance, 28(4): 605–624.

Kalbfleisch, J., & Prentice, R. 1980. The statistical analysis offailure time data. New York: Wiley.


471


http://www.systemicpeace.org



https://www.weforum.org/agenda/2019/05/7-ways-business-can-be-agents-for-peace/

https://www.weforum.org/agenda/2019/05/7-ways-business-can-be-agents-for-peace/

http://news.bbc.co.uk/2/hi/business/3854495.stm

http://news.bbc.co.uk/2/hi/business/3854495.stm

http://isx-iq.net

Kang, S., & Meernik, J. 2005. Civil war destruction and theprospects for economic growth. The Journal of Politics, 67(1):88–109.

Kelly, T., & Souter, D. 2014. The role of information andcommunication technologies in postconflict reconstruction.Washington, DC: The World Bank.

Kolstad, I., & Wiig, A. 2012. What determines Chinese outwardFDI? Journal of World Business, 47(1): 26–34.

Le Billon, P. 2008. Corrupting peace? Peacebuilding and post-conflict corruption. International Peacekeeping, 15(3): 344–361.

Luiz, J. M., Ganson, B., & Wennmann, A. 2019. Businessenvironment reforms in fragile and conflict-affected states:From a transactions towards a systems approach. Journal ofInternational Business Policy, 2(3): 217–236.

Lundan, S. M. 2018. From the editor: Engaging internationalbusiness scholars with public policy issues. Journal of Interna-tional Business Policy, 1: 1–11.

Ma, X., & Fitza, M. 2013. How much does subnational regionmatter to foreign subsidiary performance? Evidence fromFortune Global 500 corporations’ investment in China. Journalof International Business Studies, 44(1): 66–87.

Mehlum, H., Moene, K., & Torvik, R. 2006. Institutions and theresource curse. The Economic Journal, 116(508): 1–20.

Meyer, K. E. 2015. Context in management research inemerging economies. Management and Organization Review,11(3): 369–377.

Meyer, K. E., & Peng, M. W. 2005. Probing theoretically intoCentral and Eastern Europe: Transactions, resources, andinstitutions. Journal of International Business Studies, 36(6):600–621.

Meyer, K. E., & Peng, M. W. 2016. Theoretical foundations ofemerging economy business research. Journal of InternationalBusiness Studies, 47(1): 3–22.

Meyer, K. E., Mudambi, R., & Narula, R. 2011. Multinationalenterprises and local contexts: The opportunities and chal-lenges of multiple embeddedness. Journal of ManagementStudies, 48(2): 235–252.

Mills R., & Fan, Q. 2006. The investment climate in post-conflictsituations. The World Bank, World Bank Policy ResearchWorking Paper 4055.

Mills, M. 2011. Introducing survival and event history analysis.Thousand Oaks: Sage.

Moyo, D. 2010. Dead aid: Why Aid is not working and how there isa better way for Africa. New York: Farrar, Straus and Giroux.

Oetzel, J., & Getz, K. 2012. Why and how might firms respondstrategically to violent conflict? Journal of International BusinessStudies, 43(2): 166–186.

Oetzel, J., Westerman-Behaylo, M., Koerber, C., Fort, T. L., &Rivera, J. 2010. Business and peace: Sketching the terrain.Journal of Business Ethics, 89: 351–373.

Paris, R. 2010. Saving liberal peacebuilding. Review of Interna-tional Studies, 36(2): 337–365.

Peng, M. W., Wang, D. Y. L., & Jiang, Y. 2008. An institution-based view of international business strategy: A focus onemerging economies. Journal of International Business Studies,39(5): 920–936.

Pettersson, T., & Eck, K. 2018. Organized violence, 1989-2017.Journal of Peace Research, 55(4): 574–587.

Preacher, K. J., & Hayes, A. F. 2008. Asymptotic and resamplingstrategies for assessing and comparing indirect effects inmultiple mediator models. Behavior Research Methods, 40(3):879–891.

Rettberg, A. 2016. Need, creed, and greed: Understanding whybusiness leaders focus on issues of peace. Business Horizons,59: 481–492.

Riddell, R. C. 2008. Does foreign aid really work?. Oxford: OxfordUniversity Press.

Ring, P. S., Bigley, G. A., D’Aunno, T., & Khanna, T. 2005.Perspectives on how governments matter. Academy of Man-agement Review, 30(2): 308–320.

Robinson, J. A., Torvik, R., & Verdier, T. 2006. Politicalfoundations of the resource curse. Journal of DevelopmentEconomics, 79(2): 447–468.

Rose-Ackerman, S. 2008. Corruption and government. Interna-tional Peacekeeping, 15(3): 328–343.

Sachs, J. 2005. The end of poverty. London: Penguin Press.Santiso, C. 2001. Good governance and aid effectiveness: TheWorld Bank and conditionality. The Georgetown Public PolicyReview, 7(1): 1–22.

Sarkees, M. R., & Wayman, F. 2010. Resort to war: 1816–2007.Washington DC: CQ Press.

Sawyer, A. 2008. Emerging patterns in Liberia’s post-conflictpolitics: Observations from the 2005 elections. African Affairs,107(427): 177–199.

Slangen, A. H., & Beugelsdijk, S. 2010. The impact of institu-tional hazards on foreign multinational activity: A contingencyperspective. Journal of International Business Studies, 41(6):980–995.

Small, M., & Singer, J. D. 1982. Resort to arms: International andcivil wars, 1816-1980. Beverly Hills: Sage.

Sutherland, D., Anderson, J., Baily, N., & Alon, I. 2020. Policy,institutional fragility, and Chinese outward foreign directinvestment: An empirical examination of the Belt and RoadInitiative. Journal of International Business Policy, 3(3): 249–272.

Tan, D., & Meyer, K. E. 2011. Country of origin and industry FDIagglomeration of foreign investors in an emerging economy.Journal of International Business Studies, 42(4): 504–520.

Turner, N. Aginam, O., & Popovski, V. 2008. Post-conflictcountries and foreign investment. Policy Brief: United NationsUniversity.

Wardak, A. 2004. Building a post-war justice system inAfghanistan. Crime, Law and Social Change, 41(4): 319–341.

Westermann-Behaylo, M. K., Rehbein, K., & Fort, T. 2015.Enhancing the concept of corporate diplomacy: Encompass-ing political corporate social responsibility, international rela-tions, and peace through commerce. Academy of ManagementPerspectives, 29(4): 387–404.

Williams, A. 2015. A global index of information transparencyand accountability. Journal of Comparative Economics, 43(3):804–824.

Winters, M. S. 2010. Accountability, participation and foreignaid effectiveness. International Studies Review, 12: 218–243.

World Bank Group. 2006. MIGA: Reducing the risk profile offinancial investments. Financial Brief, June 2006.

World Bank Group. 2018. Foreign direct investment, net inflows(BoP, current US$); GDP (current US$); and total naturalresource rents (% of GDP). Databank: The World Bank.Retrieved October 1, 2018.

Wright, M., Filatotchev, I., Hoskisson, R. E., & Peng, M. W.2005. Strategy research in emerging economies: Challengingthe conventional wisdom. Journal of Management Studies,42(1): 1–33.

Xu, D., & Meyer, K. E. 2013. Linking theory and context:‘Strategy research in emerging economies’ after Wright et al.(2005). Journal of Management Studies, 50(7): 1322–1346.


472


APPENDIX: CALCULATION OF FDI RECOVERY

This section begins with a description of theassumptions and calculations behind the definitionof postwar FDI recovery applied in this study. Itconcludes with a discussion of how this definitioncompares with others that are based exclusively onpre-war trends.The definition of recovery employed here beginswith the idea that each country’s long-term averageinbound FDI level can faithfully represent a level ofFDI that is normal for that country. It can thereforemark the threshold by which the country can bedeclared ‘recovered’ in the years after the end of awar. By this definition, a country that is receivingFDI at or above its long-term average level isassumed to have shaken off some of the invest-ment-depressing effects of recent war. The amounttime it takes to achieve this level of investmentafter the end of the war, then, is the recoveryperiod.The long-term average for each country was calcu-lated in the usual way: summing yearly FDI obser-vations for each country from the timespan of focus(1970 to 2008) and dividing by n. A further step wastaken to standardize the yearly observations ofinbound FDI for each country, so that each obser-vation represents its distance from the country’saverage in standard deviations. This was done bysubtracting the mean from each raw observation(centering it on zero) and dividing by the country’sstandard deviation of FDI as calculated from 1970to 2008. Transformed this way, the recovery periodcan be determined by counting the number of yearsfrom the end of the war until a positive FDI figure isobserved.

-200000000

-100000000

0

100000000

200000000

300000000

400000000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

First Liberian Civil War Second Liberian Civil War

Long-term Average

(FDI Recovery)

(FDI Recovery)

Figure 3 Liberia raw inbound

FDI and long-term average.

One challenge to this definition is that it applies astatic threshold for each country across a longperiod of time (1970 to 2008). Even though it iscalculated from each country’s FDI observationsthroughout that time, it may not be realistic tojudge FDI levels in the 1970s by the same staticthreshold as those in the 2000s. Indeed, applyingcountries’ averages this way results in periods thatare consistently and unrealistically long for warsending in the 1970s and 1980s and short for thoseending in the 1990s and 2000s. This definition ofrecovery can be improved by adjusting for time,accounting for the worldwide growth in FDI trendsduring these years.To make this adjustment, a panel regression wasfirst conducted of every country’s standardizedinbound FDI from 1970 to 2008, war or no war.With no independent variables, the constant in thisestimation was zero by design – the averageinbound within-country standardized FDI. A termfor time was then added to this basic estimation tosee how much the worldwide within-country aver-ages tended to change from year to year (andwhether the change was statistically significant). Asexpected, inbound FDI tended to grow each yeararound the world during this period. The averageyearly FDI growth in within-country standarddeviations was represented by the coefficient for‘time’ and was revealed here to be about 0.06 of astandard deviation. The constant in the equationwas used to identify the year during which theadjusted average inbound FDI tends to match theunadjusted figure (zero). This happens between theyears of 1990 and 1991. Terms for quadratic andcubic time were also tried, but they did not revealsignificant effects.


473


The static average for each country was thenadjusted using these estimations to arrive at thefinal recovery threshold for postwar inbound FDI.In 1990, postwar standardized inbound FDI isjudged ‘recovered’ if it is greater than or equal to0. In 1991 the threshold is raised to 0.06, in 1992 to0.12, and so on. In 1989, the threshold is loweredto - 0.06, in 1988 to - 0.12, and so on. The fullrecovery calculation is illustrated in Figures 3, 4,and 5 for Liberia, starting with raw inbound FDI.

Advantages, Limitations and Other DefinitionsThis definition of recovery has a few advantages foranalysis. It can be easily interpreted, being based oncountry averages and linear adjustments, and it canbe applied consistently to all countries with yearlyFDI data during the years in question (1970 to2008). It remains a relevant milestone for inboundinvestment regardless of when the country is at warduring these years or how long the war(s) may last.

It leads to a threshold that should be realisticallyattainable for countries that have gone throughwar-related investment shocks, since it is based on ameasure of center calculated over the entire times-pan. Finally, it is based on recorded observationsfrom each country, rather than counterfactuals, soit avoids making judgments based on assumptionsof what might have been had a war neverhappened.The biggest limitation of this method is that it

does not privilege FDI trends that may have beenpresent in countries’ pre-war years, and thereforediverges from the intuition that recovery shouldmean ‘‘back to how it was before.’’ By includingevery yearly observation in each country’s long-term average calculation, this method has thepotential to set recovery thresholds that are lowerthan might otherwise be expected, being influ-enced downward by wartime trends. These are faircriticisms, and they should be taken as caveats ininterpreting the full results of this study.

-2

-1

0

1

2

3

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008


Standardized

Long-term Average(FDI Recovery)

(FDI Recovery)

Figure 4 Liberia standardized

inbound FDI with static

threshold.

-2

-1

0

1

2

3

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008


Adjusted

Long-term Average

(FDI Recovery)

(FDI Recovery)

Figure 5 Liberia standardized

inbound FDI with time-

adjusted threshold.


474


Methods for defining recovery thresholds basedon countries’ pre-war trends were also considered,but the trade-offs of these were judged to be lessacceptable than the ones presented. One method,which has been applied in prior research of postwarGDP, is to use the pre-war figures themselves asrecovery thresholds (found in Flores & Nooruddin,2009). The five GDP observations leading up to thestart of the war were assessed, and the highest onewas set as the recovery threshold for postwar GDP.This method worked very well within the previousstudy, but it led to some ambiguities when appliedhere. Inbound FDI data are less comprehensive inthe years before 1970, so this method would havenecessitated a reduction in sample size. Complica-tions also arise when considering the threshold forcountries with multiple or relapsed wars (originalthreshold or interim peace?). The single pre-warobservation, up to 5 years’ removed from the startof the war, can lose its significance as a realisticrecovery marker when applied to long-runningwars. Despite these challenges, it speaks well tothe validity of both approaches (the pre-war obser-vation threshold and the long-term average thresh-old) that they result in similar distributions ofworldwide postwar economic recovery periods.

The pre-war observation method described abovemay also be biased downwards – static pre-warfigures do not take into account the expected FDIgrowth that may have happened had the war nevercommenced. A threshold that accounts for lostwartime growth might be constructed by startingwith a country’s pre-war trend and forecasting thisthrough the war and postwar years. A countrywould only be considered recovered when itreaches or exceeds its forecast. This forecast methodis probably closest to a general intuition of what itshould mean for a country to be recovered – to notonly match its previous FDI levels but to have madeup lost ground. The biggest limitation to thismethod is that it often results in thresholds thatare unattainably high. A study by Cerra and Saxena

(2008) tested a similar method to judge GDPrecovery from large economic shocks (includingcivil wars), and they found that countries rarelyrecover to pre-war rates. Even after they stabilize,they remain on trajectories well below their pre-wartrends. This sobering result formed the core mes-sage of their study, titled ‘‘The Myth of EconomicRecovery.’’To conclude, there are many considerations that

go into a quantitative definition of postwar recov-ery. Results of this study are based on a recoverythreshold defined by countries’ adjusted long-termaverage inbound FDI levels, and they should beinterpreted accordingly.

ABOUT THE AUTHORRobert J. Moore is a lecturer at the University ofHawai 0i at Manoa and researcher at the East-WestCenter. His research interests encompass politicaland institutional strategy in environments of rapidchange, including war, new regimes, and interna-tional frontiers like deep oceans and outer space.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutionalaffiliations.

Accepted by Hafiz Mirza, Area Editor, 27 October 2020. This article has been with the author for three revisions.

Open Access This article is licensed under aCreative Commons Attribution 4.0 InternationalLicense, which permits use, sharing, adaptation,distribution and reproduction in any medium orformat, as long as you give appropriate credit to theoriginal author(s) and the source, provide a link tothe Creative Commons licence, and indicate ifchanges were made. The images or other third partymaterial in this article are included in the article’sCreative Commons licence, unless indicated other-wise in a credit line to the material. If material isnot included in the article’s Creative Commonslicence and your intended use is not permitted bystatutory regulation or exceeds the permitted use,you will need to obtain permission directly fromthe copyright holder. To view a copy of this licence,visit http://creativecommons.org/licenses/by/4.0/.


475


http://creativecommons.org/licenses/by/4.0/

Documents

Emerging from war: Public policy and patterns of foreign