RESEARCH MASTER THESIS

(1)

0

______________________________________________________

RESEARCH MASTER THESIS

The role of host governments in humanitarian logistics

Impact on international NGO performance in complex

emergencies

______________________________________________

Student name

Nonhlanhla Dube

Student I.D.

S2082578

E-mail

n.dube@rug.nl

Supervisors

Dr. J.T. van der Vaart

Dr. R.H. Teunter

In partial fulfilment of the requirements of Rijksuniversiteit Groningen, Faculty of

Economics and Business, for the MSc Degree Economics and Business (research)

(2)

1

1. Introduction

An “overwhelming majority” of disasters occurring worldwide are complex emergencies, i.e., man-made disasters that tend to propagate long-lasting violence and/or political instability (Albala-Betrand, 2000). International humanitarian non-governmental organisations (IHNGOs) have assumed a first responder role in complex emergencies because the governments of an affected country (host governments) tend to be either nonexistent or ineffective in mounting a response (Natsios, 1995). By the late 1990s, about US$5 billion in donor funding for responding to complex emergencies was largely channelled through IHNGOs (Natsios, 1995; Albala-Betrand, 2000). Over the years, IHNGOs have come under mounting pressure to be more efficient and effective in conducting relief operations due to the escalating frequency and scale of disasters coupled with increasingly cost conscious donors (Van Wassenhove, 2006). Improving logistics is crucial in this respect as humanitarian logistics (HL) can account for as much as 80% of relief activities and costs (Van Wassenhove, 2006). However, enormous challenges are faced in HL (Maon et al., 2009) that lead to contextual constraints which are yet to be well understood in research (Kovacs and Spens, 2011). For instance, one of the unique features of HL compared to traditional logistics is that it takes place in a context where political complexities prevail and is directly affected by them (Tomasini and Van Wassenhove, 2009a; Van Wassenhove, 2006). This is especially true for complex emergencies which inherently have “complex social, political, and economic origins” and can lead to the “disputed legitimacy of host authorities, the abuse of human rights and possibly armed conflict” (World Health Organisation1

).

In the growing field of HL research (Natarajarathinam et al., 2009), complex emergencies have not received much attention (Adivar et al., 2010). One potential explanation is that it has been argued that in the long run, HL (humanitarian supply chains) starts to resemble traditional logistics (supply chains) (Maon et al., 2009; Tomasini and Van Wassenhove, 2009b). However, the long term orientation of complex emergencies does not necessarily erase the political nature of complex emergencies. This in turn implies that the long term orientation of complex emergencies does not necessarily resolve the political complexities that lead to HL challenges in disaster relief. For example, although Somalia and Sudan have been embroiled in internal conflict for more than 3 decades and have had IHNGO intervention, they are in the top 3 most dangerous contexts for aid work in the world (Stoddard et al., 2009). Some of the HL challenges the governments of these countries have caused include the diversion of relief supplies intended for beneficiaries (Somalia) (Menkhaus, 2010) and denying IHNGOs access to affected populations (Sudan) (Pettit and Beresford, 2005). Although these examples may not reflect host government behaviour in all countries experiencing complex emergencies, they illustrate two things. First, the aforementioned argument for the lack of attention to complex emergencies in HL research is not necessarily a valid one. Second, even though political complexities cannot be directly observed, host government behaviour/ actions bring(s) them to the fore.

The observation that host government behaviour brings political complexities to the fore, coupled with the fact that host governments can directly affect HL with their decisions (Ergun et al., 2009) necessitate an investigation into how and why host governments affect HL in complex emergencies. This research seeks to do both. Such an investigation is made

1

(4)

3

all the more relevant by the fact that HL challenges arising from political complexities are not well understood, yet they affect HL performance.

The research design is a multiple case study of 6 countries in long term complex emergency situations. The research follows a single focal IHNGO that has conducted humanitarian work for at least 10 years in each of these countries. Since HL is defined as “the process of planning, implementing and controlling the efficient, cost-effective flow and storage of goods and materials, from point of origin to point of consumption for the purpose of meeting the end beneficiary’s requirements” (Thomas and Mizushima, 2005, p. 2), HL can be said to be primarily concerned with supplies for humanitarian relief. Facility location (FL) decisions of where they are stored; warehousing and inventory management (WIM) decisions of how they are procured/ sourced and stored (WIM); and transport, routing and distribution (TRD) decisions of how they are transported and distributed are key. Therefore the logistics functions of FL, WIM, and TRD are considered in this research.

The logic of this research is grounded theory building. That is, the research seeks to identify key variables (core challenges from host governments affecting HL performance), the patterns of linkages between the variables (how the core challenges affect HL performance), and why these relationships exist (underlying reasons that explain why the host government behaves as it does) (Voss et al., 2002). Since the research is about a rarely explored phenomenon, grounded theory building is suitable (Eisenhardt, 1989). This approach has the potential to lead to novel and accurate insights into the phenomenon being studied (Glaser and Strauss, 1967).

The rest of the paper is organised as follows. In section 2, a brief review of the literature is conducted. Focus is on the impact of host governments on HL performance in complex emergency situations. The research methodology is presented in section 3, followed by a presentation and analysis of the research results in section 4. A discussion follows in section 5 and conclusions about the research are drawn in section 6.

2. Literature review

In this section, a review of the literature on the various ways in which host governments affect HL in complex emergency settings is conducted. Host governments are intrinsically linked to the setting in which humanitarian relief takes place (Kovacs and Spens, 2007; 2008). On one hand, it has been noted that in complex emergencies host governments either do not exist or exist in “such anaemic form” that they are ineffective, rendering IHNGO intervention necessary (Natsios, 1995). On the other hand, it has been argued that host governments can directly impact HL in IHNGO led interventions for example through their control over political and economic conditions (Ergun et al., 2009) or sovereign considerations like the declaration of a state of emergency (Albala-Betrand, 2000; Altay and Green, 2006; Kovacs and Spens, 2009).

(5)

4

al., 2009; Olson, 2006; Robbins, 2007; Slim, 2000). The disruption of relief activities due to security constraints imposed (at least in part) by host governments have been widely reported by IHNGOs and the media. For example, in 2009, the IHNGO Médecins Sans Frontières (MSF) reported that it was forced to halt operations in at least 3 countries due to government related security incidents2. In Yemen, fighting between the government and the rebels forced MSF to close the only hospital serving an entire district after it was shelled; in Pakistan, hospitals were struck by mortar fire and two MSF workers were killed, leading to the suspension of operations; and in the Democratic Republic of Congo, government forces attacked civilians at MSF vaccination sites making it difficult for MSF to work. In the first two instances there was extensive damage/ destruction to essential facilities.

Host governments can also cause disruptions that are not related to security issues. For example, during the Sudan conflict in 2003 the Sudanese government officials began blocking aid from reaching civilians in Darfur by denying IHNGO aid workers travel permits to the affected areas (Pettit and Beresford, 2005). Again in Sudan, the issuing of a warrant of arrest for the Sudanese president by the International Criminal Court in 2009 led to the expulsion of 13 IHNGOs from the country3. In the same year (2009), the government of Sri Lanka were fighting with rebels and banned aid agencies from entering the conflict zone where tens of thousands of civilians were trapped2.

Restricted movements have an obvious bearing on transport, routing and distribution (TRD). An example pertaining to transport and routing is how fighting between Israeli and Lebanese armies in 2006 imposed severe security constraints that made it extremely difficult for IHNGOs to transport goods into Lebannon and forced IHNGOs to take longer routes and use more expensive modes of transport (Tomasini and Van Wassenhove, 2008). With regards to distribution, in July 1994, leaders of the then Hutu controlled Rwandan government assumed control of the distribution system and diverted relief supplies intended for refugees to members of the Rwandan Army (Toole and Waldman, 1997: p. 302). A similar situation has prevailed in the longstanding civil war in Somalia where the (regional) government(s) has (have) diverted relief supplies for refugees (Menkhaus, 2010).

Host governments have been argued to affect the facility location (FL) and warehousing and inventory management (WIM) aspects of HL as well. For example, because of their control over critical assets like warehouses (Kovacs and Spens, 2008), host governments may impact FL decisions made by IHNGOs. Unlike the two examples earlier where MSF’s facilities were destroyed, this impact does not entail security related issues. In terms of WIM, the use of customs regulations as barriers (Long and Wood, 1995) implies longer lead times and illustrates one of the ways in which inventory management issues can be affected by host governments.

The preceding examples are indicative of governments who act to deliberately hinder humanitarian operations by blocking access to, or diverting supplies of, intended

2_{Médecins Sans Frontières, 2009. “Top ten” humanitarian crises: aid blocked and diseases neglected. Press} Release. Accessed on 02 July 2013.

http://www.doctorswithoutborders.org/publications/topten/2009/article.cfm?id=4127

(6)

5

beneficiaries. However, there are also situations that have been reported whereby an “anaemic” host government impacted HL due to a lack of means to facilitate IHNGOs rather than an intention to hinder them. An example with could be found goes as far back as 1983 when OXFAM delivered food in Cambodia during the civil war. The port was reported to have had “no equipment whatsoever—no forklifts, cranes, conveyors, not a thing…” and “it took nearly five days for 1,500 tons to be unloaded, even though the organization was being personally directed by the third most important minister in the government...” (Whitaker 1983, in: Long and Wood, 1995, p. 223). In this case host government willingness to receive IHNGO assistance was overshadowed by capability constraints which in turn led to similar outcomes from a government that is unwilling to accommodate IHNGOs: delays in the delivery of supplies to the intended beneficiaries. On the basis of the above review, it can be argued that host governments can have a significant impact on HL performance in complex emergency settings. Furthermore, this impact is not infrequent but recurrent. Given the long term orientation of complex emergencies (Albala-Betrand, 2000; Pettit and Beresford, 2005), IHNGOs are likely to own facilities, be regularly engaged in stock-replenishment and management activities, as well as frequent transportation and distribution of supplies in the affected areas among others. In order to get supplies to the intended beneficiaries on time and at low costs (effectively and efficiently), most activities in the logistics functions of FL, WIM, and TRD have to be planned and implemented in advance. Supplies have to be ordered, shipped, and stored before they are actually required for consumption. So HL does not differ from traditional logistics in this respect, the difference comes about in the considerations that have to be made in HL due to host government behaviour. In particular, important logistical questions arise that pertain to advance planning when access is denied by host governments or disruptions occur. For example, if there are supplies at hand (inventories) or supplies that need to be delivered when an IHNGO’s operations are disrupted indefinitely due to security or other issues, what happens to those inventories/ supplies? How does this denied access or these disruptions affect the timely delivery of supplies and the overall costs of delivering supplies to affected populations? Basically, how does host government behaviour affect HL performance in complex emergencies? Furthermore, what are the reasons for host government behaviour/ actions? To the extent that the (causes of) government behaviour affecting HL in complex emergencies can be determined or understood, this can reveal issues that can be addressed within HL and those that fall outside the scope of HL. Regarding the latter, it is conceivable for example that no amount of good logistics could have overcome the Sudanese government’s decision to order 13 IHNGOs to leave the country leaving hundreds of thousands of civilians without access to humanitarian assistance.

Although the preceding review shows evidence of some of the ways in which host governments affect HL performance, the evidence remains largely anecdotal. No systematic assessment of how and why host governments impact HL performance has been conducted so far. As a consequence, most of the above questions are inadequately addressed or not addressed at all in existing literature. This study has been designed to address the above questions. In particular, the following research questions are explored:

1. How do different host governments affect the performance of IHNGOs within the logistics functions of FL, WIM, and TRD in complex emergency settings?

(7)

6

3. Methodology

3.1 Research setting and design

The design of the study was a comparative multiple case study. The exploratory nature of this research warranted the use of a case study design because: it is suitable for studying phenomena in highly complex settings (Stuart et al., 2002); puts emphasis on the real-world context in which the phenomena under investigation occur (Eisenhardt and Graebner, 2007); is useful for reliably answering how and why questions in unexplored research (Edmondson & McManus, 2007; Yin, 1994); and the “resultant theory is often novel, testable, and empirically valid” (Eisenhardt, 1989).

The use of multiple cases has multiple benefits as well. First, multiple cases provide a strong base for theory building (Yin, 1994) due to more grounded, more accurate, and more generalisable findings (compared to single case studies) (Eisenhardt and Graebner, 2007). Each case is like a distinct analytic unit/experiment that serves as a replication, contrast, and extension to emerging theory (Yin, 1994). Second, comparisons across cases make it possible to ascertain whether a finding is constantly replicated by several cases or if it is idiosyncratic to a specific case (Eisenhardt, 1991). Furthermore, spurious effects, random patterns, and researcher bias in analysing the data can be exposed by deeper probing into observed conflicting evidence across cases (Eisenhardt, 1989). Finally, research questions and theoretical elaboration can be explored more broadly.

In terms of the research setting, the study focused on countries that have been in complex emergency situations for at least a decade. The long duration of these emergencies makes it possible to ascertain the consistency of host government behaviour over time and distinguish between random and non-random host government actions. An IHNGO with a long history of working in complex emergency settings was also chosen because such an organisation would have the benefits of the long history of exposure to, and understanding of, host government behaviour.

3.2 Sampling

The unit of analysis were countries which are in on-going complex emergencies. Countries were chosen first on the condition that the complex emergency situation has been on-going for the long term (at least 10 years) and that the focal IHNGO has worked in that country for at least a decade. Next to this, theoretical sampling was conducted to ensure variation in the sample (Eisenhardt and Graebner, 2007; Eisenhardt, 1989). Theoretical sampling leads to unambiguous pattern recognition of the main constructs, relationships, and the focal phenomenon logic (Eisenhardt and Graebner, 2007). In order to achieve this variation in the sample, “polar types” that have sharply contrasting characteristics (Eisnehardt and Graebner, 2007; Miles and Huberman, 1994) were selected.

(8)

7

The state fragility index (SFI) developed by Marshall and Cole (2008) and validated by Ziaja (2012) was used as a measure of state fragility. The SFI measures a country’s effectiveness and legitimacy along four performance dimensions: security, political, economic, and social performance (Marshall and Cole 2008) and ranges between 0 (no fragility) and 25 (extreme fragility). Because complex emergencies tend to occur in countries that have high fragility, only countries that are highly fragile (SFI scores from 16 – 19) and those that are extremely fragile (SFI scores from 20 – 25) were considered. The former represented the most fragile countries in a complex emergency situation whereas the later represented fragile but comparably better off countries also going through a complex emergency. The differing degrees of fragility would make it possible to get a wider representation of different host countries.

The economic state of a country is a useful indicator of the ability of a host government to address matters arising from the complex emergency situation. It has been noted in the discussion about state fragility that sometimes a government’s willingness and capacity are conflated (Ziaja, 2012). Should there be differences in the impact of host governments in HL, the economic state could be useful for distinguishing between an incapable host government and one that is capable but lacks the will to act in the interest of its citizens. In this research, the economic state of a host country was measured using the Gross Domestic product (GDP) estimates from the World Bank (2012). In the sampling, countries going through a complex emergency were also selected on the basis of either having a relatively high GDP value or a relatively low GDP value. Assuming that there are 195 countries in the world, this distinction was made by considering countries that are roughly in the top 90 in terms GDP worldwide as high GDP countries and those in the bottom 90 (i.e. ranked from 106 to 195) in terms of GDP as low GDP countries. The stability of both measures (SFI and GDP) over time was assessed by looking at estimates from previous years. To ensure that there was no multicollinearity between the GDP value (which is used as a measure of economic effectiveness in the SFI score calculation) and the SFI score, the data that was used to generate the SFI scores that we adopted was used to calculate the Pearson correlation value between the two. The two were found to be highly correlated (Pearson correlation value = .807, p<.01) but there was no evidence of multicollinearity since the Pearson correlations value was below .9.

On the basis of the above mentioned criteria, 6 countries were selected for the multiple case study with at least one country drawn from one of the four categories resulting from these criteria. The sampled countries are shown in Figure 1. Given the sensitivity of the collected data the actual names of the case countries are not provided. It is to be noted that two of the sampled countries (X and Z) made the list partially because they had been flagged as interesting countries by practitioners in the past.

(9)

8

3.3 Data collection

Data was collected primarily through semi-structured interviews. An interview protocol with semi-structured questions was developed which covered questions about general aspects of the complex emergency; specific questions about how the host government affects the logistics functions of FL, WIM, and TRD; and any other questions about the role/ influence of host governments in HL performance. Regarding the general aspects of the complex emergency, questions focused on the history of the complex emergency; how, when, and why the IHNGO got involved; the nature of the operations that the IHNGO held in the country; and the general relationship between the IHNGO and the host government. Interviews were conducted between December 2012 and March 2013. The bulk of the interviews were conducted in March 2013 during a week-long gathering of logistics personnel working both in the field and at headquarters (HQ). During this time, informal conversations and observations added to the information gathered through the interviews.

Twenty-two highly knowledgeable and experienced respondents working for the IHNGO in various logistics related functions were interviewed. The respondents had work experience ranging between around 6 and 25 years (mean 13.3 years). The mean duration of each of the interviews was about 47 minutes. Although respondents generally dealt with all HL functions in one way or another in their work, some respondents were highly specialised in specific functions. There were 4 respondents focusing on issues related to international sourcing and 1 respondent focusing on facility construction and/ or rehabilitation. The respondents with highly specialised functions were asked questions that were relevant to their functions. Seven of the respondents work in the HQ and the rest either work in the sampled countries or have worked there in the past. Most of the respondents with field postings had coordination roles. Of the 22 respondent, 7 had a non-Western background. Some respondents were able to provide information on more than one country resulting in an equivalent of 50 interviews being conducted with the 22 respondents. Respondents who provided information for more than one country had either worked in one or more of the sampled countries in the last 5 years in the field (in addition to their current posting) or managed a portfolio that contained more than one of the sampled countries at HQ. The choice to collect retrospective data from the informants is explained shortly. Figure 2 provides some basic information about the distribution of the informants, the countries for which they provided information, and the duration of each of the interviews.

(10)

9

Leonard-Barton, 1990). When coupled with observations and informal conversations from the week-long event, it was possible to validate the data collected from the interviews to some extent.

Figure 2: Respondent information- posting, countries covered, and interview duration

Given the importance of situating the researcher, i.e. stating the background of the researcher and their relationship with informants for the purpose of assessment of validity (Myers and Newman, 2007), it is worth mentioning that 13 of the respondents were known to the researcher either through working with them in the past or having met them in other organisational settings. This made the interviews easier to conduct and informants were open in providing information. For informants that the researcher met for the first time, an effort was made to first have a casual meeting with them over lunch or dinner in order to address any potential concerns/ questions they had about the research before conducting the interviews.

3.4 Data analysis

(11)

10

analysis provides the depth of knowledge that is required for cross-case analysis (Voss et al., 2002). The within cases analyses were followed by cross-case analysis which relied on methods suggested by Eisenhardt (1989) and Miles and Huberman (1994). The systematic search for patterns through cross-case analysis makes it possible to review the evidence from single cases through multiple lenses, forcing investigators to look beyond initial impressions (Eisenhardt, 1989). This has the benefits of: mitigating the risks of exaggerating the meaning of findings from the single case analysis; leading to better and stronger grounded findings if they are corroborated between or among cases; and is essential for enhancing generalisability of results obtained from single cases (Voss et al., 2002; Eisenhardt, 1989).

No a priori hypotheses were made. The cases were first compared to identify common challenges faced in HL due to government behaviour. Tables and diagrams were created to facilitate further comparisons between and among cases to identify similarities and differences between them. These were then used to develop emerging constructs and theoretical logic. This led to insights into how and why different host governments affect HL.

4. Results and analysis

In this section the emerging insights that were obtained from the data are presented. Data were analysed iteratively. The first round of analysis entailed the identification of HL challenges faced that are attributable to host governments. This was followed by a second round of revisiting the data to establish the linkages of the identified challenges with performance. In some cases, challenges that had been missed in the first round were identified in the second round. Results obtained from the overall analysis are presented as follows. First the identified specific challenges faced in each of the 6 countries that are attributed to host government behaviour are presented. This is followed by narrowing down the specific challenges to 3 core challenges that constitute host government impacts on HL and their subsequent effect on HL decisions pertaining to FL, WIM, and TRD. The section ends with a presentation of the reasons for host government actions in each of the sampled countries that were identified by the respondents.

4.1 HL challenges and implications for effectiveness and efficiency 4.1.1 Specific challenges identified in the research

(12)

11

Table 1: Specific challenges faced in different countries due to host government actions

When the specific challenges per host country were compiled into a single table (Table 1) it emerged that certain challenges are unique to certain countries. Some of the challenges that were identified for one country but none of the others include the following.

In terms of FL, the IHNGO experienced difficulties in getting land for construction in country W. It was stated that there was clear disinterest to allocate land for construction by the host government authorities and when they eventually made land available for the IHNGO, it was not an ideal location:

(13)

12

For WIM, one of the exceptions was in country Y where the IHNGO had to collect statements of approval from local vendors each time they wanted to internationally source for certain supplies rather than purchase them locally. The reason for the IHNGO’s choice to engage in international sourcing was related to quality concerns but this was not something they could discuss openly:

“So can you imagine you are running a business in country Y and I come to you, you have this stuff in your shop and I say well actually it's nothing, quality wise it's nothing. But please sign here a statement that says that you don't object that I go and get it somewhere else because the authorities need it otherwise I cannot import.” (R#2, country Y)

With respect to TRD, one of the exceptions was in country V where the IHNGO lost supplies in transit to the army:

“… we lost a truck or two filled with fuel to the [army] who thought it is a good idea to stop it and take it off…” (R#9, country V)

Turning to similarities across cases, it can already be seen from Table 1 that most of the specific challenges overlap among the different countries. Insecurity, for example, was one of the challenges faced in most of the case countries. However, even though some specific challenges are unique to single countries as stated above, the nature of the challenges themselves is quite similar. For instance, in the above example of the confiscation of supplies (fuel), the challenge was largely attributed to the security situation in country Y. The next step in the research was then to focus on the nature of these specific challenges, in order to narrow them down to a limited set of core challenges and to establish the impact they had on HL performance.

4.1.2 Core challenges

When focus was shifted to categorising the specific challenges presented in Table 1 in terms of their nature it became apparent that they could all be attributed to three core aspects: regulations, security, and issues of sovereignty. The results are presented in Table 2. The effect on performance is presented in Table 2 as well but a more detailed account of the pathways through which these core challenges eventually lead to impact on HL performance is presented in the next subsection. From Table 2 it can be seen that the challenges that remain exceptional, i.e., they apply to a single case, diminish to no more than 2. These are namely the complicated procedures for land allocation in country W and high import taxes for vehicles in country X.

The regulatory aspect accounts for the majority of the challenges faced in HL that are due to host government behaviour/ actions. The term ‘regulations’ in this study includes such things as policies, procedures, rules, and laws in place that have a bearing on HL. Interestingly enough, all regulations that pose challenges in WIM relate to international sourcing in all countries where they are experienced as a challenge (Table 2). Some of the challenges encountered with regulations have been found to relate more to information sharing about the regulations in place than the regulations themselves, i.e., the accessibility of information, timely sharing of information, and the clarity of available information.

(14)

13

Table 2: Core HL challenges faced in different countries due to host government action

Other challenges are directly related to the regulations themselves, for example the level of complexity of the regulations.

“Now we are at a stage for example that we have to provide that document [certificate of pharmaceutical product] for any drug, if it's not there then it depends on the status of that drug if we can import it. If it's a WHO prequalified product then the [country] says it's okay, but if it's not, and we don't have the document then they don't accept it. And if we do have a document it's in French or in German it has to be translated into English because they want something they can read as well. So this is getting more and more complicated.” (R#8, country Z)

(15)

14

completely safe while others (the majority) mentioned that depending on where projects are located in country X, there are occasional security incidents.

A single challenge was found to affect 3 of the countries, which were all high GDP countries (W, X, and Z), and this was the imposition of travel restrictions – a sovereign consideration - to certain areas at certain times. This differs from regulated movements whereby there are rules and procedures in place for seeking permission to go to a place. With imposed travel restrictions, the host government could order IHNGO personnel to leave an area and/ or order them not to go to a specific area. This would override any regulations in place.

Overall, there are two most interesting results in the cross-case analyses with respect to challenges faced. The first is that it was not just the number of issues that a host government presents to the IHNGO that mattered for performance, but rather the nature of the challenges was more meaningful. An unprompted outcome was that the respondents were able to rank the countries in terms of level of difficulty due to host government behaviour (see ranking on Table 2). Country Z was considered the most difficult, closely followed by X and W. These three countries all had highly regulated international sourcing procedures and had introduced TRD regulations that stipulated that travel permits were required in order to access certain areas. Country Y had the benefit of having no security related challenges but was nevertheless deemed more challenging logistically than countries U and V where security concerns prevailed and this was due to Y’s regulations regarding international sourcing. Disregarding challenges that are experienced only occasionally or at specific regions of a host country (those marked with x** in Table 2), it can be seen that the most challenging countries do not necessarily have the most challenges. The country with the most prevailing challenges is W with 8 such challenges yet it is ranked third. Countries Z and W are ranked first and second most challenging with 5 and 4 prevailing challenges respectively. In addition, even though countries X and Y have the same number of prevailing challenges (4) they are not deemed comparable in terms of the level of difficulty of the issues to be dealt with. It follows from these observations and the unprompted rankings that it was not the number of issues that a host government presents to the IHNGO that mattered the most (although there appears to be some correlation especially looking at the lower ranked countries), but rather the nature of the challenges was more meaningful.

The second interesting result (which is related to the first) is that it emerged from the data that WIM was considered to be by far the most challenging. The unprompted rankings presented in Table 2 provide evidence of the magnitude of the challenges faced in relation to international sourcing. All countries ranked in the top four in terms of difficulty share international sourcing challenges in common. Most respondents stated that they experienced the biggest challenges in international sourcing, a function categorised under WIM in this study, and most of the interview time was spent on discussing international sourcing challenges.

“The biggest for us in logistics is in the importing, you know, that is where it starts.” (R#6, general remark)

(16)

15

4.1.3 Mapping core challenges to HL decisions and outcomes

Having established that the core challenges faced in HL are related to regulations, security, and sovereign aspects, the next step was to ascertain how this affected decisions and outcomes in FL, WIM, and TRD. Figure 3 depicts the path from the 3 core challenges that are imposed by host governments to the decisions they affect and the subsequent impact they have on HL performance. As mentioned earlier, the linkage between the identified challenges and performance was established through first identifying the challenges (Table 1 and table 2) and then revisiting the data to identify the reported effects of the challenges. This was done in terms of the HL decisions that need to be made and the outcome in terms of performance. Although Figure 3 is an abstract representation of what goes on and does not represent an exhaustive list of scenarios and decisions to be made, it is nonetheless useful for visualising the link between host government behaviour and HL performance. The coloured arrows show the decisions affected by specific core challenges and the problems they lead to. Even though different paths lead to different problems, it all culminates in negative impacts on timeliness of providing humanitarian supplies/ relief and costs incurred in the process. Figure 3 also seems to suggest a one-on-one or decomposable relationship between the core challenges that stem from host government behaviour and the logistics functions of FL, WIM, and TRD. However, as is demonstrated later, the results showed interconnectedness across logistics functions and across core challenges. Before evidence of these connections is presented, the results regarding the more straightforward relationships are presented for FL, WIM, and TRD.

4.1.3.1 Facility location

Regulations and security were found to be the problematic issues when it comes to FL. For most of the part, the regulations were encountered when there was a need to construct facilities or rehabilitate existing facilities leading to delays of up to a year to set up urgently needed facilities. The main reasons for challenges faced with respect to regulations in FL were the lack of familiarity with procedures (which can be attributed to the fact that facility construction is an activity that the IHNGOdoes not frequently engage in) and/ or complicated procedures that had to be followed. In the case of country W, it was both.

“... We [had to] design and build a new clinic within a few months. The problem was that who gives authorisation? Where can we go? First of all, where? Is there a plot [of land] available where it is allowed to build a clinic? We have the [township medical official] who is guarding the medical programme, but you also have (...) the boss of the area, you have the city council, you have a regional medical [official] and all of them in the end have to agree. (…) So the first plot which we found in the end we could not build there because there is also ownership of the land. Sometimes ownership of the medical programme is not ownership of the city [council]... So we had to move to another area.” (R#14, country W)

Security considerations in FL were found to lead to one of two things. On one hand, if facilities are intended for direct service provision to the affected population (primarily clinics for patients in the case of this IHNGO) then such facilities are located as close as possible to the affected population to minimise the security risk for the affected population. However, this inherently carries the risk that the facilities are damaged or destroyed in the event of a major security incident. This has indeed happened in countries U and V. In country V for example, one respondent stated:

(17)

16

Figure 3: Mapping core challenges imposed by host governments to HL performance

On the other hand, if facilities are intended for bulk storage, they tend to be located much farther from the location that they are intended to serve. In the case of country V, due to security and other reasons, they had the major warehouse serving country V located in another country.

“Nobody knew what will be the rules, how it will work, if the system is stable enough to put a logistical base there. (…) So because of all this, the logistics base is [in a neighbouring country].” (R#13, country V)

4.1.3.2 Warehousing and inventory management

(18)

17

regulatory component on WIM affecting HL performance was the predictability of customs clearance times. When clearance time was predictable but lengthy, this led to the decision to keep large buffer stocks. One respondent noted during a prior informal conversation (R#3) that keeping large buffer stocks was much less cumbersome than opting for shorter ordering cycles because going through the formalities of getting approval to source internationally and clearing the goods when they physically arrived in the country was a lengthy and labour intensive process. For example, it takes 6 months for the government of W to approve a request for an import licence and when the licence is issued it is valid for 3 months (with the possibility of extension).

When clearance times were unpredictable for example due to lack of clarity in procedures, delays for one were a major issue leading to stock ruptures:

“We had ruptures (…) if you cannot import for one year you have to find different ways of using your drugs or using what is available in the country.” (R#1, country X)

“The projects who want to do interventions, who need to treat people, who have nurses and medical staff standing by waiting for their stethoscopes and their pills and they are waiting, waiting, waiting.” (R#3, country X)

Also for WIM, security concerns led to lower than desirable stock levels on site.

“In general stocks are close to the project but we made an exception recently. [We] moved stocks to the capital for security reasons.” (R#15, country Z)

In country W, however, it was also reported that they were forced to keep stock levels that did not necessarily match demand because they had quantity restrictions on how much they could source internationally in a year. This applied only for drugs and it was not perceived that the intention was to force the IHNGO to engage in local purchase in the case of drugs because there was no local market for drugs.

4.1.3.3 Transport, routing, and distribution

TRD was found to be affected by all three core challenges of regulations, security, and sovereign aspects. Regulations related to taxes make it expensive to bring in vehicles but these are not captured in Figure 3 in part because most respondents were generally of the opinion that taxation is a normal aspect, especially in countries where IHNGOs have a long term presence, although it was sometimes considered extremely costly. Regulations that affect TRD in all cases in which they applied involved the issue of travel permits which were temporary and not always easy to renew.

“We were not allowed to most areas but over time that changes… you have a special visa for inside country for certain areas so maybe they say we don't renew it anymore and maybe for half a year you don't have access.” (R#6, country Z)

However, in some cases the regulations were related to controlled or monitored movement. This was considered acceptable under certain circumstances.

“If you are operating in [a] region where insurgency is going on and where transport is actually banned and the government is very much concerned that the goods are not falling into the hands of, well let's call it the rebels, then of course you have to register each transport and you have to submit a packing list and get security clearance and all these things. I would assume that it is standard operating procedure in a warzone.” (R#7, country X)

(19)

18

“You can't move from the capital to any of our projects by road, they are highly insecure. The chances of you surviving those roads are quite low.” (R#12, country Z)

The sovereign aspect in TRD differs from the regulatory aspect in that in case of the latter, the IHNGO can be ordered to evacuate an area despite having valid permits for operating in that area or forbidden from going there. The sovereign aspect therefore overrides any regulations that may be in place and tends to be unpredictable. In country Z for instance, it was stated that the IHNGO’s expatriate aid workers have not been allowed to a specific area in the country for over a year and no one knows when it will be possible for them to travel to those areas again. It has been possible to send supplies to local aid workers there but this has not been without challenges.

4.1.4 Interconnectedness

The preceding subsections describe the impact of host governments on HL on a one-on-one basis. However, they were found to be interconnected within logistics functions and across logistics functions. Examples of this interconnectedness are presented next.

4.1.4.1 Interconnectedness within logistics functions

Multiple issues were found to be interrelated within logistics functions. Two examples related to WIM are provided here. First, when the unpredictable lead times prevail and stock ruptures are frequently experienced, this eventually leads to projects ordering much more than they need in case of future problems and this has negative cost impacts.

“Another side effect is once the people on the ground are not sure that they get the drugs then they just order twice [the required amount] and we get huge overstock.” (R#14, general remark) Second, although Figure 3captures stock ruptures due to unpredictable lead times, it does not reveal that sometimes the high costs associated with unpredictable lead times lead to stock ruptures. An example is an incident in country Z where due to an unannounced change in regulations, the IHNGO had goods held at the airport by customs officials for an extended period. By the time they were cleared for release, they had incurred such high costs that the IHNGO was forced to abandon the shipment and re-order all over again: “(…) They could not manage to clear it on time. There were some difficulties with some papers. And they just said okay, the cost is higher than the value of the thing, so there is no way I am going to pay for this cost. It's far much higher than the cost of what I am importing.” (R#12, country Z)

4.1.4.2 Interconnectedness across logistics functions

Figure 3 shows how FL decisions influenced by security concerns inevitably affect WIM decisions and outcomes. If a warehouse (facility) is not close to the projects, then it means that supplies are not close to the projects and so they cannot be accessed on time. What this in turn implies is that within such a set-up, there is a need for more frequent transportation to replenish stocks on location. Since TRD would also be affected by security issues transportation could be infrequent and costly, or when there is frequent use of airplanes it will be frequent but extremely costly. Frequent use of air transport due to security reasons was reported for countries U and V. Another issue is that for facilities that are close to the affected population when facilities are damaged or destroyed, the supplies that are kept in those facilities suffer the same fate as well. The following quote captures how TRD, FL, and WIM can be simultaneously affected by security:

(20)

19

stock as well. One-and-a-half weeks ago in one project the whole town burned down so we lost a lot of stock that was there of course.” (R#6, country U)

In addition, the incident in country V where a truckload of fuel was lost in transit to the army also demonstrates this interconnectedness. Because goods were lost in transit, TRD and WIM were coupled in this case. These are probably always tightly coupled since TRD entails the movement of supplies. So an incident during transportation tends to have repercussions for WIM. Another example of the coupling between TRD and WIM is when denied access or security threats render movement to and from an area temporarily impossible leading to difficulties in managing supplies. This was noted to be the case for certain areas in country Z for example.

“We work often by remote management but you're not on ground. Soyou're not able to control the stock, how the stocks are moving. Stocks are missing either because they [people on site] don't know how to manage the stock or they are stolen.” (R#12, country Z)

4.1.4.3 Interconnectedness of core challenges

There is evidence from the data that the core challenges are connected to each other and simultaneously affect HL functions. This sometimes happens in seemingly opposing ways. An example is how regulations about routes (which are restrictive for the IHNGO) are actually for safety reasons, the specified routes are like humanitarian corridors where safety of the IHNGO staff and supplies can be better guaranteed. This for example was noted in countries X and Z.

"It depends very much on the area… We are free to move in [country X] but then only bound to our own security route. But then in the [insurgency] region of course it was only one road which we were allowed to use and then we were using another road which we negotiated access to and then after the security incident they closed [the road]. But then this is a highly militarised area with full blown insurgency so once again I would consider access control as a military standard operation..." (R#7, country X)

In some cases, core challenges can only be dealt with in sequence. For instance, the first step in highly insecure areas is to first get past sovereign considerations about granting the IHNGO access. Only when this is resolved can the IHNGO make decisions about where to locate facilities for example.

An example of security considerations that only follow after sovereign issues have been dealt with in WIM and FL is captured in the following quote:

“Where the security is very bad in general we have less stock than in areas where security is more predictable. But we're … more influenced by where you are allowed to work first and where you think there is reasonable security so that is where you locate your services.” (R#6, country Z)

4.2 Lack of challenges

It is equally important to pay attention to situations in which (almost) no challenges were faced by the IHNGO. Regarding FL, there were generally no problems encountered when facilities were rented in all countries. In country W it was stated that renting facilities had become very expensive over time but this was attributed to market forces of supply and demand and had nothing to do with host government behaviour. However, as mentioned earlier, when there was a need to construct facilities challenges were faced in acquiring land and/ or in establishing what the construction rules are (Table 2). The only exception in the case of construction was country V:

(21)

20

For WIM, once supplies were in-country there were also no challenges noted. It was stated for countries U, X, and Y that there were regulations and warehouse inspections to check that storage conditions are up to standard and/ or ensure environmental safety. However, they were not complicated and were easy to adhere to. In country U this was even perceived as something that the responsible authorities did not take seriously but rather exploited as an opportunity to make money. On occasion they had made up the rule that if they came for an inspection, the IHNGO had to pay for the costs incurred by the inspectors.

“There is an environmental office that is checking, well they are supposed to be checking, whether the warehouse is posing a health risk to the public with regards to drinking water. Our warehouse is really nice but what the environmental office is not checking is that everybody is just washing their cars in front of our office all the oil is flowing into the river and into the lake, people throw their trash inside and so on. So it's slightly ridiculous you see that it's really more money making business rather than really an environmental objective.” (R#16, country U)

Only countries U and V were found to have no challenges regarding international sourcing. In fact, country V was described by one respondent as a logistician’s paradise when it comes to the international sourcing:

“Compared to my previous missions I should say country V is a paradise. The difficulties we are facing are much more linked to ignorance. Like you present something to an officer in charge and he doesn't want to block you but he just doesn't know what to do. So it will be delayed a bit. I have never been in a country like that where almost all my requests for customs are agreed, I have all my tax exemptions.” (R#13, country V)

The only country where no challenges were reported for TRD was country Y. This is in part because country Y was the safest of the sampled countries. However, the IHNGO had been temporarily subjected to travel restrictions and ordered to cease operations in the distant past during election periods. Even then, security was more a perceived potential risk than a real risk.

Main conclusions on challenges Before proceeding to present the results on the causes of

host government behaviour, a concluding summary of the results presented so far is provided. There are three main conclusions concerning the challenges faced in HL. Namely that there is interconnectedness among the logistics functions and also among the core challenges; when host governments behaviour is predictable and requirements are well communicated, their level of perceived impact is generally considered less severe than when it is unpredictable; and respondents consistently noted that the system (nature and impact of host government behaviour) changes over time. Regarding interconnectedness, as demonstrated earlier, a single incident at one logistics function can cause a ripple effect that cuts across other logistics functions while sovereign core challenges, particularly restricted access/ movement, generally need to be dealt with first before the IHNGO could proceed with operations. As for predictability, its importance is captured in the following statement made about international sourcing:

“Predictability, repeatability of these processes is what allows you to prepare and to really make things happen one shipment after the other. But if it is always well depends on who we speak to and maybe we can convince them to allow us this time then you are actually not getting anywhere. Then it's not predictable anymore.” (R#3, country X)

(22)

21

respondents. For example, one of country Z’s regulations implied a major administrative burden which was considered acceptable and necessary:

“And okay sometimes maybe good restrictions that, you know, for the use of psychotropic drugs for example. You have full guidelines… There is a good follow up. Let's say I import one vial of valium, in a lot of countries okay that's it. But there you have to have a registration for example when that vial is administered to the patient. It goes far but I think it's good practice and probably what happens here (in western countries).” (R#6, country Z)

Despite a long term presence in all the 6 sampled countries, changes over time were also deemed inevitable. They could be triggered by various issues like a host government becoming more democratic (country W) to election periods (country Y) and/or outcomes of elections (country Z).

“But then when the elections have been then it could be that the whole playing field changes, that there are new people in the ministry and that it will be easier or more difficult. So how can you influence it? You can't. You have to wait and anticipate on the change... it's something you can't influence but it's something that you are confronted with and you have to deal with it.” (R#8, country Z)

Indeed, the importance of anticipating change and understanding where it originates from was emphasised by most respondents. For example, the following was said of country X in which the IHNGO had experienced multiple (perceived) changes in regulations:

“I think that if you would analyse what was really happening and also their intentions of these legislations and that kind of thing, it all makes perfect sense. And if you would have anticipated, without even compromising yourself, then at least we could have understood much better what was going on ….” (R#4, country X)

Given the importance of understanding the causes of host government behaviour in order to adequately anticipate and prepare for changes where possible, the next step in the research was to seek to understand why the host governments in the sampled countries behave as they do.

4.3 Causes of host government behaviour

Respondents were asked if they could explain the underlying reasons for host government behaviour, since it was not possible to gather this information from host government officials themselves. Figure 4 shows some of the identified causes that the respondents came up with and the countries to which those causes correspond. A few elements of figure 4 are expanded upon next.

In general, respondents noted that host governments are increasingly becoming stricter in dealing with IHNGOs.

“… It is mostly by your restrictions in the government. What is the trend? It's that it gets more restricted, that is very clear. I think 20 years ago you could do much more and more free...” (R#6, general remark)

(23)

22

However, this positive development went undetected for a while in certain countries and took the IHNGO by surprise. For example, when the government of country X developed the capacity to enforce regulations that had been in existence for decades but were never enforced this was first met with confusion.

“Not clear to us was that regulations had always been there but the authorities didn't have the means [to enforce them]. But the regulatory authority now had means of supervising the implementation of these rules... That confused not only [this supply centre] but also the other supply centres to the extent that at a certain moment we thought that the [country X] government was all the time changing their requirements and then we were faced with these new requirements, like we were played with.” (R#2, country X)

(24)

23

Country X also started building its own list of approved drugs for import as part of a large scale quality scheme and this conflicted with the IHNGO’s standard list. Country X is not an exception in this case, and it has been noted by respondents that one of the main issues in international sourcing is this discrepancy between what the IHNGO has always sourced for its operations worldwide, and what the different host governments are now willing to accept into their territories. And this is increasingly becoming non-negotiable.

“So protocols differ, and when protocols differ, government will block you.” (R#22, country Y) In some cases even though market protection is the intention behind restricting the international sourcing of certain supplies, the necessary supplies cannot be sourced in-country and sometimes even when they can, the quality is questionable. Country W was noted for making this difficult for all but drug items.

“It's a protectionist country, [well] protectionist I don't know because there's not so much local industry..., but they don't allow import of many things: computers, cars… it's changing by the way, fridges, food… there's lots of rules that make it difficult or impossible to import things. What can you do? ... Drugs is easy, easier than other items." (R#17, country W)

Although some of the reasons for host government behaviour have been perceived as being legitimate, an issue that is often encountered is that a choice made to act in the interests of the public (like protecting the local job market) can be used as a political tool to block IHNGOs.

“Governments should check what medical materials or what medical services are brought into their country by foreigners and we have the obligation to be transparent with that. Unfortunately it's easy to mix that up with objectives of ‘we don't want their actions in the first place so we just use it to obstruct them’. So the principled question is a valid one, but I don't think it justifies all the obstruction we've seen.” (R#19, country X)

In general, however, political motives were perceived to be dominant, and were generally visible in controlled movement. For example through visa restrictions (regulatory aspect) or restricted access (sovereign aspects)

“If they want to limit you they will just not give you a visa. They know how to do it in clever ways. I hardly remember that they pushed us or they took our drugs or didn't allow us to do… in general they don't do all these small things… if they want to block us they can do it in 5 minutes.” (R#15, country Z)

“It's very easy, if a government has a clear interest on isolating certain populations then you will face logistical problems to reach them. That is how it works.” (R#7, country X)

Finally, there appears to be a sharp contrast between the high GDP countries and low GDP countries in terms of the receptiveness to IHNGO presence. The former has been generally noted by the respondents to lack the political will to address the complex emergency situation, and therefore tends to restrict IHNGOs more in their operations.

(25)

24

“They try to play a strong state but they aren't because they don't have infrastructure and they know that without [this IHNGO] they can't do anything. We are not a drop in the ocean, we are a serious player.” (R#15, country V)

While it was stated that it is not always possible to know the motives for host government behaviour (Figure 3), and though limited to the perspective of IHNGO staff, the preceding discussion reveals that there is a general trend towards more regulations in complex emergency settings; not all reasons for host government behaviour are politically motivated; and that there are differences in host government capabilities to enforce regulations- be they genuine or politically motivated.

5. Discussion

This research explores how and why host governments affect HL performance in IHNGO led relief operations in complex emergency settings. As indicated at the onset, host governments greatly influence HL performance (c.f. Ergun et al., 2009; Kovacs and Spens, 2008; Long and Wood, 1995) but how and why they do so has not been systematically studied. Furthermore, most HL research to date has focused on sudden onset natural disasters (Natarajarathinam et al., 2009).

In order to address this gap in the literature, two main research questions have been posed. The first research question is concerned with how host governments affect HL performance. The key finding is that host governments present IHNGOs with 3 core challenges: regulations, security, and sovereign considerations. These 3 core challenges have generally been found to be interconnected; affect HL performance the most when they lead to unpredictable outcomes; and consistently change over time. The second research question seeks to establish why host governments behave as they do. Here there are two key findings. First, even though complex emergencies are of a political nature, some of the reasons for host government actions signal positive developments in those countries. Host governments do not always act to deliberately hinder IHNGO efforts. Second, certain challenges are faced in HL because host governments lack the capabilities to facilitate the smooth functioning of processes that affect HL, e.g. timely customs clearance. Here the lack of capabilities also negatively impacts HL performance.

The insights obtained from exploring each of the research questions are now discussed in depth and contrasted against existing literature followed by a discussion about the implications of the findings for research and practice.

5.1 Challenges faced in HL

The three core challenges identified in this research, i.e., regulations, security, and sovereign considerations are in line with the specific challenges that have been noted in the existing literature. However, the results of this study further reveal interconnectedness and changes over time in these challenges, something that has not been identified in existing literature. In addition, this research provides further insights into the nature of, and underlying reasons for, these challenges. The three core challenges are now discussed in turn with emphasis on findings that are contrary to existing literature and those that have not been identified in the existing literature.

Regulations Results of this research indicate that although there tend to be communication

(26)

25

modified by the host government throughout the course of IHNGO operations, the regulations generally gravitate towards stability over time. As one respondent stated, regulations “become clearer” making it easier for the IHNGO to plan and ensure the timely and cost-effective delivery of supplies. The stability and clarity of regulations does not necessarily imply that HL performance is enhanced over time, however. In certain instances outcomes either remain unpredictable or requirements are nearly impossible to meet, leading to HL ineffectiveness and inefficiency. In general, high GDP countries (W, X, and Z) were found to have regulations (or at least enforce regulations) that directly impact HL performance across all logistics functions of FL, WIM, and TRD, whereas regulations were used (or enforced) much less (frequently) in low GDP countries. A notable exception within the low GDP case countries is country Y which was found to have regulations for FL and WIM that were actually enforced and directly impacted on HL performance. Country Y was described by one respondent as being “a very sophisticated bureaucracy” with a lot of regulations in place that are actually followed through.

Unpredictable outcomes in relation to well established regulations were reported in case countries W and Z where it could never be known in advance if travel permits would be granted when the IHNGO applied for them according to existing regulations. One respondent had this to say about the issuance of travel permits in country W for example: “[If you want to visit a project you have to apply] 5 weeks in advance and they don't have a guarantee on that... Like one of our projects we have 8 locations where we work, if I mention that I want to travel to 3 places and they don't want me to go to one of those three places they will reject the whole application then they will ask you to apply again and it will take again another 5 [weeks]... Then we say okay, let me go to the other two, they say no you have to apply again.... You cannot have concurrent applications to visit...”(R#5, country W)

Regarding clear regulations where it is understood what needs to be done in order to get a desired outcome yet the requirements are nearly impossible to meet, one issue stood out across multiple case countries (notably countries X and Z). The issue concerns internationally sourced supplies where the misalignment between paperwork requirements set by host governments and prevailing practice in countries where supplies are sourced complicates matters for the IHNGO. This is currently the biggest challenge faced in country Z:

“… So with [country Z] the paperwork is a nightmare, also because they want things that we can't get here [in Europe]. Because the registration rules in the European community are such that, for example, companies don't need to have that documentation, and sometimes they just don't want to give it. And then you have to explain why you need it... so that's a difficult one.” (R#8, country Z) The requirement for additional paperwork/ other requirements was perceived to be a challenge that is increasingly faced in most countries. In the words of a respondent, it is anticipated that this is “going to be the system of the future”. Signs of this issue were reported already for country V which was the least regulated of all.

“Not long ago I could ship almost everything I wanted in country V without any paper. Now I need papers because all flights have to go to the capital [city] and all trucks have to pass by the capital [city].” (R#13, country V)

RESEARCH MASTER THESIS

______________________________________________________