The Global Trends Reports have become UNHCR's key annual reporting tool about previous-year developments in forced migration. Their release on World Refugee Day (20 June) demonstrates the relevance that UNHCR ascribes to these reports. They address various subjects—including international displacements, demographics, durable solutions, and, regularly, accommodation distributions—mainly through numbers. The reports also enclose comprehensive statistical annexes, which convey the impression of full transparency, as the reader has access not only to explanations in the reports, but also to what appear to be exact numbers in the annexes. The 2018 report, for example, highlights that “[t]he majority of refugees lived in privately hosted and out-of-camp individual accommodation (60 per cent) at the end of 2018, a proportion that has been stable since 2014 with variation of only a few percentage points” (UNHCR 2019c: 62). A linear graph (Figure 23) visualizes how little the numbers have seemingly changed even since 2011. The precise presentation of ‘big data’ intrigued me; I wondered why the time span was not ten years or more—as is fairly common for such charts?1 What types of accommodation are generally considered in the reports, how are they defined, and how are local sites categorized? More fundamentally, how do accommodation quantifications produce (non)knowledge and link with humanitarian bureaucracies?
Such numbers are neither new nor limited to accommodation statistics and generally reflect a quantification of social phenomena for humanitarian and political purposes (see Urla 1993; Desrosières 1998; Lawson 2021). By creating or referring to statistical data, political and humanitarian actors prove their awareness and expertise in decision-making. Yet resonating with Foucault's intertwinement of knowledge and power, actors mobilize not only knowledge but also ignorance and nonknowledge through quantifications, the latter being understood not as “unknown unknowns” but as “known unknowns” (e.g., Gross 2007). Whereas statistics seem to state “unbiased facts” (Broome and Quirk 2015: 827), depict reality, and claim truth, they only render certain matters and categories meaningful—those that are included (Davis et al. 2012; Merry 2016). Moreover, following the idea of a life cycle, categories and numbers are (made) relevant only when there is demand, communication, and public meaning, among other things (Simpson and Dorling 1999; Lawson 2020).
This is also evident for migration and displacement; ‘migration’ evolved into a quantifiable category over decades (Espahangizi and Mähr 2020; Stricker 2019) but the ‘real-time’ counting of those on the move, and the deaths, enacts migration and makes parts of it politically relevant (Scheel and Ustek-Spilda 2019; Heller and Pécoud 2020) while other parts are rendered invisible or inconsequential, as I argue in my work. While registering ‘refugees’ may appear an ordinary measure in the humanitarian aid system, Crisp (1999: 3) noted already two decades ago that to UNHCR, it is about “‘caseload statistics’ to fulfill its mandatory task of refugee protection, to plan its programmes, to draw up its budgets. … ” Counting ‘the vulnerable’ for project design, or even collecting biometric data, reflects critical links of humanitarian, technological, and funding interests (Glasman 2019; Lemberg-Pedersen and Haioty 2020), but also indicates how central the numbers are in humanitarian bureaucracies.
The Global Trends Reports’ focus on ‘big data’ and the presentation of accommodation quantifications must be seen in this context. UNHCR explains online that the reports serve public, strategic, and political objectives: to “deepen public understanding of ongoing crises,” “to meet the needs of refugees and other populations of concern,” and to “help organizations and States to plan their humanitarian response” (UNHCR 2020b). UNHCR finds accommodation data to be “important for efficient policymaking and programme design” (2019c: 62). The Global Trends Reports can thus be understood as a tool for governing displaced people and a means for UNHCR to represent its position in the political economy of the humanitarian landscape—yet, I argue, based on unclear and inconsistent numbers.
In this article, I explore the quantification of refugee accommodation in 18 Global Trends Reports and their statistical annexes available online,2 published from about 2003 to 2020. I focus on fundamental elements of accommodation categories, their definitions, quantifications,3 and local categorizations in the reports over time. Inspired by Scheel and Ustek-Spilda (2019), I draw on McGoey's concept of strategic ignorance and Bourdieu's field of struggle to study modes of (non)knowledge production and the reports’ role in humanitarian bureaucracies. McGoey's concept addresses how decision-makers produce and use ignorance to “generate support for future political initiatives” and “avoid liability for earlier actions” (McGoey 2019: 3). Despite McGoey's focus on ignorance, I use the term ‘nonknowledge’ as a means of further highlighting the unknown—whether produced knowingly or unknowingly. In turn, Bourdieu's field of struggle captures how, influenced by the structure, actors seek and act to maintain or transform power (Bourdieu and Wacquant 1992: 101). This approach helps position UNHCR's Global Trends Reports in the political economy of the humanitarian landscape and to consider competition with other aid agencies (Scheel and Ustek-Spilda 2019: 666). In particular, I aim to deconstruct what is presented, and thus also what is missing, in these reports to explore meanings and suggest the implications of such processes.
Based on the analysis, my findings reflect both the reports’ critical production of statistical nonknowledge and the latter's roles in the humanitarian landscape. On the one hand, while the accommodation numbers in each report portray a level of ‘hard facts,’ they are, instead, inconsistent and unstable. Accommodation categories lack definition in the reports and change over time; readers are thus left to interpret meanings. The numbers that fill the categories are at times vaguely explained, and calculations are hard to trace. This also applies to local accommodation categorizations; sudden changes from one year to the next happen without explanation as to how or why—or even that they occur. These issues indicate that the reports produce not only fragmented or selective statistical knowledge, but indeed nonknowledge, as of known unknowns. On the other hand, this nonknowledge still facilitates the governance of refugees, and the reports’ tendency toward addressing high numbers of refugees in individual accommodations or urban areas essentially makes them politically and strategically relevant and links with the “urban turn” in policy (Fiddian-Qasmiyeh 2019).
Undefined Accommodation Categories
“Who or what gets counted, by whom, and for what purposes are questions of immediate consequence”—Urla's (1993: 819) statement is as relevant today as it was then, and points to the critical entanglement of indicators, knowledge, and power. As with all statistics, they can only present information captured in categories or indicators, which are thus indispensable for quantifications (see Davis et al. 2012; Cooley and Snyder 2015; Merry 2016: 12ff). So which accommodation categories are included in the Global Trends Reports, and how are they defined?
Table 1 below reflects the categories in each report. Importantly, accommodation categories only emerge in the 2010 report, and none of the Global Trends Reports contains definitions of accommodation or location categories in the text or statistical annexes. Only three reports have footnotes referring to definitions in other sources, namely in UNHCR's statistical yearbook or on its websites (UNHCR 2013a: 34–35; 2014a: 36–37; 2015b: 41–43). Moreover, the reports do not clarify how accommodation and location differ, why changes occur over time, or how the categories are (re)applied locally (more below).
Overview of categories.
|Global Trends Reports||Location categories||Accommodation categories|
|2003–2004||camps/centers, urban, dispersed/various *|
|2005||camps/centers, urban, dispersed/various/unknown*|
|2006||camps/centers, urban, rural/dispersed, various/unknown *|
|2009||camps/centers, urban, rural/dispersed, unknown|
|2010||(-) urban, rural/dispersed, various/unknown *||camps, centers, dispersed, individual accommodation, settlement, undefined/unknown|
|2011||as 2010||camps, centres, (-) individual accommodation (private), settlement, undefined/unknown|
|2012–2013||as 2010||planned/managed camp, collective center, individual accommodation (private), self-settled camp, reception/transit camp, undefined/unknown|
|2014–2018||urban, rural/dispersed, various/unknown (Annex 16)|
urban, rural, unknown/unclear(Annex 19)
|2019–2020||urban, rural, unknown/unclear||as 2012–2013|
Sources: UNHCR (2004: annex 11; 2005: annex 11; 2006: annex 12; 2007: annex 11; 2010: annex 17; 2011a: annex 15, 17; 2012: annex 15, 17; 2013a: annex 15, 17; 2014a: annex 15, 17; 2015b: annex 16, 18, 19; 2016a: annex 16, 18, 19; 2017a: annex 16, 18, 19; 2018: annex 16, 18, 19; 2019c: annex 16, 18, 19; 2020a: annex 17, 18; 2021: annex 17, 18).
Note: Bold text marks change and dash (-) deletion. Asterisk (*) indicates annexes on categories regarding populations of concern, not specified for refugees. Annexes are unavailable for the 2007 and 2008 reports.
Since these categories are not merely narrative elements in the reports, but key factors for measuring and presenting developments, the absence of definitions and explanations for changes is critical. Although definitions attract criticism for many reasons and are only “prototypical” (on state fragility, see Cooley and Snyder 2015: 88ff), and although it is common that authorities turn to new or more refined indicators for detailed projections, these issues lead to a lack of accuracy, transparency, traceability, and validity. Yet each report conveys an impression of knowing exactly what can be known about seemingly concrete categories, and also of knowing the unknown—by including “unknown” and “undefined” as categories. The contradicting vagueness is only evident when questioning the categories’ meanings and acknowledging the changes in them over time. In some cases, categories even vary across annexes of the same reports (e.g., the 2003 report's annex 11 includes “camps/centers” and annex 12 “camps”; see also in Table 1 the locations in the 2014–2018 reports’ annexes). The categories thus turn into a ‘conceptual dustbin,’4 as such vagueness allows for arbitrary applications.
Whereas any quantification “organizes and simplifies knowledge, facilitating decision making in the absence of more detailed, contextual information” (Merry 2016: 1), the missing definitions and explanations contribute to a problematic simplification. It is ultimately up to the reports’ reader to make sense of the categories, as they are not informed about features that identify and distinguish them. The Global Trends Reports therefore produce and mobilize not only fragmented—that is, partial or selective—knowledge, but indeed nonknowledge. Whether this is deliberate or results from unconscious practices cannot be determined, but if it is unclear what the categories are about, the numbers behind them are equally ambiguous.
Unclear Accommodation Numbers
Most Global Trends Reports, and their annexes, indicate high percentages of refugees or populations of concern in individual accommodations—or urban areas, as categorized in earlier ones (UNHCR 2010: 15–16; 2011a: 35; 2012: 35; 2013a: 35; 2014a: 37; 2015b: 43; 2016a: 53; 2017a: 55; 2018: 60; 2019c: 62; 2005: annex 12; 2007: annex 12; 2020a: annex 17; 2021: annex 17). This data must be treated with caution, however. Such numbers do not necessarily reveal a majority; until the 2014 report, the numbers of those in unknown, undefined, or dispersed contexts are higher than in the other categories. The data primarily denotes that more refugees (appear to) live in urban areas or individual accommodations than in encampment settings. Interestingly, the 2010 report—in which new categories emerged—notes a low number of refugees in individual accommodations, at 29 percent, while the 2011 report refers to 56 percent (both excluding those in unknown contexts; UNHCR 2011a: 35; 2012: 35). It is safe to assume that this is the reason why the linear graph in the 2018 report does not cover a 10-year or longer period, as is quite common in such reports (see introduction), but only one since 2011 to visualize the trends (UNHCR 2019c: 62); had a longer period been covered, justifications would have been necessary as to why the variation occurred between 2010 and 2011.
But what do we know about the numbers? To explore the quantifications in the reports, I focus in particular on the 2018 publication as an example, while also addressing others. As a reminder, this report states that most refugees reside “in privately hosted and out-of-camp individual accommodation (60 per cent).” It also refers to some national cases, outlines the scope of available data, and notes that “UNHCR requests geographically disaggregated data … from its office, partners and governments” and “collects data” (UNHCR 2019c: 62).
When examining both the report and its annexes, five inconsistencies emerge. First, annexes 16 and 18, about location and accommodation types of “refugees and people in refugee-like situations,” reflect data for a total of 20,193,884 and 20,360,562 people respectively. Why the numbers differ, despite the mutual focus on refugee(-like) populations, prompts questions that remain unanswered. Second, focusing on annex 16, due to the detailed names of locations therein, the numbers lead to 32.94 percent of refugees being categorized as living in the different encampment settings and 48.89 percent in individual accommodations.5 How, then, does the report arrive at 60 percent? Instead of using the total to determine the percentage, it seems that the number of those in undefined accommodations was deducted in the report's calculations, leading to 59.74 percent. While UNHCR staff can only work with the data gathered or received from states in compiling such statistics, the report does not explain this deduction—an aspect that is stated in some of the previous reports (e.g., UNHCR 2017a: 55) or indicated in tables in several (e.g., UNHCR 2010: 35; 2011a: 35; 2012: 35; 2013a: 35; 2014a: 37; 2015b: 43; 2016a: 53; 2017a: 55; 2018: 60). The issue that I wish to highlight here is the creation of not only an illusion of clarity, but also nonknowledge for readers: almost 49 percent is less than half, and signifies a different meaning to the report's wording of “[t]he majority … (60 per cent).” Moreover, considering the reports’ public, strategic, and political objectives, disregarding 3,669,943 individuals in undefined accommodations shows how these people are rendered inexistent and irrelevant in the chosen quantifications, and thus also for policy-making, humanitarian response, and public understanding.
Third, annex 16 states that “[c]ountries with no information on location or demographics are not included” (UNHCR 2019c: annex 16). How is this different from the “undefined” or “unknown” accommodations included in the annexes? Similarly to the missing category definitions, no information is provided in the report or annexes. This approach prevailed from the 2005 report, when “unknown” was added in location types (except the 2007 and 2008 reports, without annexes) and only changed in the 2019 and 2020 reports, which no longer include this reference (UNHCR 2006: annex 12; 2007: annex 11; 2010: annex 15; 2011a: annex 15; 2012: annex 15; 2013a: annex 15; 2014a: annex 15; 2015b: annex 15, 16; 2016a: annex 15, 16; 2017a: annex 15, 16; 2018: annex 15, 16; 2020a: annex 17, 18; 2021: annex 17, 18). Fourth, the 2018 report claims that the “[a]ccommodation type was known for some 18.1 million refugees, about 89 per cent of the global total in 2018” (UNHCR 2019c: 62). When subtracting the number of those in undefined accommodation from the total in annex 16, an overall 16,523,941 is reached (to further blur the situation, annex 18's figure is 16,582,956). Based on the available data, precisely how this deviation emerged cannot be established by the reader.
Finally, why are the accommodation types of so many refugees marked as undefined? If one compares annex 16, about accommodation, with annex 3, about the total refugee(-like) population, astonishing tendencies emerge: accommodation types are indicated as undefined for the whole refugee(-like) populations in Australia, Denmark, Finland, France, Germany, Greece, the Netherlands, Norway, Poland, Sweden, and Switzerland.6 These are just some examples of states with strong demographic surveys, leading to unresolved questions as to whether some states strategically withhold or categorize data as such, or whether UNHCR does. Yet in 2018, some of these same states pursued restrictive policies against refugees; for example, Germany sought to prolong refugees’ encampment in so-called AnkER-Zentren; Poland increasingly rejected asylum seekers; Australia continued detention politics and Greece hotspots; and so on. Such categorization is not limited to 2018 but prevails in some cases since the 2003 report, and varies in other cases from year to year (more below).
The fact that some of these states are UNHCR's key donors (UNHCR 2019a) should be taken into account, as UNHCR's dependence on particular states’ funding may be related to the portrayal of favorable information in such reports, especially to ensure those states’ continued commitment. The above accommodation categorization as undefined is possible not only because definitions are lacking but also because UNHCR is accountable to donors, not the people themselves. The institutional constraints faced by UNHCR are further complicated because the agency operates in a field characterized by competition over resources with other international organizations (increasingly IOM), NGOs, and think tanks. This pressures UNHCR to not only maintain its position in the political economy of the humanitarian landscape, but to continuously broaden it too. A recent example of this is UNHCR's expansion from a ‘refugee’ to a ‘forced migration’ agency in order to also protect people displaced due to climate change (Hall 2010: 111).
The inconsistencies in the 2018 Global Trends Report (and beyond) demonstrate how vague the quantification and presentation of accommodation categories are. As Glasman (2019: 246) writes, “when we looked closer at the production of humanitarian data, the idea of robustness vanished.” The report's numbers nevertheless convey the impression of hard facts.
Inconsistent Local Recategorizations
To explore local categorizations further, I address the example of accommodation types in Uganda—not a core donor to UNHCR, but a key refugee-hosting state. Examining this country is particularly apt as Uganda has been said to use so-called rural settlements since the 1960s (e.g., UNHCR 1969: para. 147). While this labeling is critical in itself, as the term romanticizes the encampment structures that many of the sites embody, the framing has been institutionalized through the development-oriented approach to refugee aid in the country (Krause 2021: 8–10). Moreover, I have been conducting research with refugees in Uganda since 2008, and can partly contextualize the categorizations over time on this basis.
Table 2 below shows the main tendencies of categorizations in Uganda—and their intense variations. Such local recategorizations are neither mentioned—let alone explained—in any of the reports and annexes, nor consistent with other documents and local developments. For example, while these sites are labeled “planned/managed camp” in the 2012 Global Trends Report, UNHCR's other reporting tool, the Global Report, continues to call them “settlements” (UNHCR 2013b: 1–3). Despite the significant change toward categorizing them as “individual accommodation” in the 2013–2017 Global Trends Reports and then again in 2019 and 2020, other tools, such as the Global Report, global appeals, and local reports, maintain the terminology of settlements (UNHCR 2014b, 2015a, 2016b, 2017b, 2019b). The ambivalent use of categories and sudden recategorizations further contrasts with the practice regarding maps. In 2015, 2017, and 2019, in which the Global Trends Reports framed the sites as “individual accommodation,” maps created by UNHCR refer to them as “refugee settlement” and even “refugee camp” (UNHCR 2015c, 2017c, 2019d).
Main tendencies of categorizations in Uganda.
|Global Trends Reports||Location or accommodation categorizations|
Sources: UNHCR (2004: annex 1, 11; 2005: annex 1, 11; 2006: annex 1, 12; 2007: annex 1, 11; 2010: annex 15, 17; 2011a: annex 15, 17; 2012: annex 15, 17; 2013a: annex 15, 17; 2014a: annex 15, 17; 2015b: annex 16; 2016a: annex 16; 2017a: annex 16; 2018: annex 16; 2019c: annex 16; 2020a: annex 17; 2021: annex 17).
Note: Annexes are unavailable for the 2007 and 2008 reports. The 2019 and 2020 reports’ annexes only provided overviews. Earlier reports’ annexes about populations of concern are compared to annexes 1 to discern the numbers concerning refugees.
When some of these recategorizations occurred in the Global Trends Reports, I was in Uganda for longer and shorter stays, but did not see any local changes in the physical nature of the accommodation types (i.e., a camp suddenly becoming a city, or a private accommodation suddenly becoming a self-settled camp, which might be suggested in the categorization). Neither did I learn about any changes from interlocutors, or read about any in other scholars’ work. Yet from one year to the next, these sites in Uganda are labeled differently—as has happened to sites in other countries in the Global Trends Reports’ annexes too. For instance, in the annexes to the 2011 and 2012 reports, the previously “undefined” sites in Angola were recategorized as “individual accommodation,” while most sites previously labeled “camp” or “settlement” in the Democratic Republic of Congo changed to “individual accommodation”—at a time when the number of refugees in the latter country increased. Moreover, sites labeled “individual accommodation” in Cameroon changed to “self-settled camp” (UNHCR 2012: annex 15; 2013a: annex 15). This categorization prevailed in Cameroon for years, but most sites labeled “self-settled camp” in 2017 turned into “individual accommodation” in 2018 (UNHCR 2018: annex 16; 2019c: annex 16).
These variations weigh heavily. Labeling the same sites with changing categorizations over time reveals the fluid use and interchangeable character of accommodation categories, which is only possible due to the latter being a ‘conceptual dustbin’ in terms of their definition. While each report appears to depict established global knowledge annually, the reports mobilize this knowledge over time and develop the impression of fairly consistent numbers of refugees in the different accommodation types—especially in individual accommodation. This consistency portrays big numbers of refugees not as potential sources of risk and uncertainty, but as constitutive of a predictable, projectable, and thus manageable future, and of UNHCR being in control of the situation. This can be linked to UNHCR having to position—and defend—itself, its work, and its expertise in the humanitarian landscape (see also Lemberg-Pedersen and Haioty 2020: 615).
The local recategorizations expose, however, how this impression of consistency is an illusion. The (re)categorizations and thus also quantifications are based on and produce nonknowledge. It is unclear how categorizations are applied locally and why they change—and more fundamentally, how it is even possible that global quantifications can reflect consistency despite the recategorizations. While categories generally translate “knowledge from the incredible complexity of everyday life to numbers and ranks” (Merry 2016: 33), the recategorizations cloud—or even conceal—the actual living conditions of the people and their knowledge of the sites. Instead of reflecting local realities, as the reader would assume regarding refugees’ accommodation, the recategorizations and global statistics simulate a “precisely quantifiable reality” (Scheel and Ustek-Spilda 2019: 669).
The issues of lacking definitions, untraceable counting, and fluid recategorizations contradict the apparently robust quantifications in the Global Trends Reports and their annexes. While other scholars also address inconsistencies in statistics about refugee (e.g., Adelman 1982; Bakewell 1999; Sigona 2015), we essentially see here how living conditions as social phenomena are simplified and deconstructed into categories, and then statistically reconstructed to develop what are presented as comprehensive overviews and frameworks of global comparability. This results in the construction of new knowledge and nonknowledge, and the reports ascribe relevance to categories through numbers. The categories and quantifications lead to strong degrees of abstraction that portray certain realities—not refugees’ realities, but high-number realities—in even knowing the unknown numbers. This contributes to an idea of controllability and governability of refugees as data objects.
Maintaining control is arguably important for UNHCR to preserve and expand its humanitarian position. Since UNHCR is in a field characterized by institutional competition, maneuvers politically tense developments to fulfill its mandate, and faces chronic underfunding, it may be understandable for the agency to develop, publish, and mobilize such reporting tools as rhetorical and tactical devices. With its dependence on states’ cooperation for protection and funding, UNHCR must be diplomatic when raising issues and attempt to maintain donors’ commitments to the agency. Yet considering the Global Trends Reports’ public, strategic, and political objectives, the process of nonknowledge production is critical and shows how certain information is—consciously or unconsciously—staged for communication with donors, states, NGOs, and the wider public.
Addressing refugees’ distribution across urban areas and in individual accommodation continuously can be linked with the “urban ‘bias’” in the humanitarian landscape (Fiddian-Qasmiyeh 2019: S48). Although individual accommodations can of course be in urban and rural areas, since this category was introduced most Global Trends Reports note strong (quantified) connections between encampment and rural life, as well as individual accommodations and urban life (UNHCR 2011a: 35; 2012: 35; 2013a: 35; 2014a: 37; 2015b: 43; 2016a: 53; 2017a: 55; 2018: 60). Moreover, some earlier reports mention high numbers of refugees in urban areas (UNHCR 2008: 2; 2009a: 2, 4; 2010: 1, 3). Such presentation ascribes political and strategic relevance to refugees in individual and thus also urban accommodation, and supports policies for refugees’ (self-)settlement in urban areas that UNHCR has been advocating for (UNHCR 2009b; 2011b; 2014c). Some reports make this link explicit, explaining the reasons for political attention being paid to refugees in urban areas (UNHCR 2010: 3), refugees’ increasing numbers outside of camps being “fully in line with UNHCR's Policy on Alternatives to Camps” (UNHCR 2015b: 43), or the differing responses required being “fully recognized in the Global Compact on Refugees” and UNHCR having “operationalized innovative and networked approaches” for years (UNHCR 2019c: 57). While the urban focus is not new, it is revived regularly (Fiddian-Qasmiyeh 2019: S48), and such explanations signify how UNHCR navigates the humanitarian landscape through numbers. However, despite the recurring narrative of refugees in individual accommodation and urban areas in the Global Trends Reports, I found no evidence for major changes, especially in funding allocations for urban responses, and Muggah and Abdenur (2018: 7–9) discuss the ineffective implementation of the policies.
Against this background, it may not be surprising that no references to accommodation data occur in the 2019 and 2020 reports, with their annexes being reduced to national overviews similar to early reports. This might indicate the end of the life cycle of accommodation numbers due to limited outcome, or it could also result from the recent need to address effects of the COVID-19 pandemic. Regardless, the reduced overviews certainly refute the idea of annexes giving the reader detailed and transparent insights.
The various issues examined in this article in turn prompt many questions about quantifications in the Global Trends Reports beyond matters of accommodation. The most fundamental issue I see is the way in which quantifications subsume the dynamic social surroundings of so many people worldwide in some numbers, disregarding their experiences, histories, practices, mobilities, and wishes, and leading the reader to believe that authorities are omniscient and in control.
I wish to thank Maja Janmyr, Lucy Earle, Kijan Espahangizi, Stephan Scheel, and Frank Wolff for helpful discussions, and also the reviewer and editors for their valuable recommendations.
Of 23 figures with time frames in the 2018 report, 11 cover 10 years or more, and only four shorter periods.
See https://www.unhcr.org/search?comid=56b079c44&&cid=49aea93aba&tags=globaltrends (accessed 9 February 2022).
In the analysis, I sometimes give percentages with numbers after the decimal point. This is problematic, as the numbers are about people. However, I keep them to trace the reports’ quantifications.
Stephan Scheel, in a discussion about this article.
Collective centers: 40,013 (0.2 percent); planned/managed camps: 4,598,264 (22.77 percent); reception/transit camps: 41,609 (0.21 percent); self-settled camps: 1,971,866 (9.76 percent); individual accommodations: 9,872,189 (48.89 percent); undefined: 3,669,943 (18.17 percent) (own calculations based on UNHCR 2019c: annex 16).
Strikingly, the refugee(-like) population in Italy is greater in annex 16 than in annex 3 (UNHCR 2019c: annex 3, 16).
Bakewell, Oliver. 1999. “Can We Ever Rely on Refugee Statistics?” Radical Statistics 72. https://www.radstats.org.uk/no072/article1.htm (accessed 10 February 2022).
Broome, André, and Joel Quirk. 2015. “Governing the World at a Distance: The Practice of Global Benchmarking.” Review of International Studies 41 (5): 819–841.
Cooley, Alexander, and Jack Snyder. 2015. Ranking the World: Grading States as a Tool of Global Governance. Cambridge: Cambridge University Press.
Davis, Kevin E., Benedict Kingsbury, and Sally E. Merry. 2012. “Indicators as a Technology of Global Governance.” Law & Society Review 46 (1): 71–104.
Desrosières, Alain. 1998. The Politics of Large Numbers: A History of Statistical Reasoning. Cambridge, MA: Harvard University Press.
Espahangizi, Kijan, and Moritz Mähr. 2020. “The Making of a Swiss Migration Regime: Electronic Data Infrastructures and Statistics in the Federal Administration, 1960s–1990s.” Journal of Migration History 6 (3): 379–404.
Gross, Matthias. 2007. “The Unknown in Process: Dynamic Connections of Ignorance, Non-Knowledge and Related Concepts.” Current Sociology 55 (5): 742–759.
Hall, Nina. 2010. “Climate Change and Organizational Change in UNHCR.” In Climate Change and Migration: Rethinking Policies for Adaptation and Disaster Risk Reduction, ed. Michelle Leighton, Xiaomeng Shen, and Koko Warner, 104–116. Bonn: UNU-EHS.
Heller, Charles, and Antoine Pécoud. 2020. “Counting Migrants’ Deaths at the Border: From Civil Society Counterstatistics to (Inter)Governmental Recuperation.” American Behavioral Scientist 64 (4): 480–500.
Krause, Ulrike. 2021. Difficult Life in a Refugee Camp: Violence, Gender, and Coping in Uganda. Cambridge: Cambridge University Press.
Lawson, Brendan T. 2020. “The Use of Numbers by Journalists in the Coverage of Crises: A Study of Seven Humanitarian Emergencies in the UK Press.” PhD thesis, University of Leeds.
Lemberg-Pedersen, Martin, and Eman Haioty. 2020. “Re-Assembling the Surveillable Refugee Body in the Era of Data-Craving.” Citizenship Studies 24 (5): 607–624.
Merry, Sally Engle. 2016. The Seductions of Quantification: Measuring Human Rights, Gender Violence, and Sex Trafficking. Chicago: University of Chicago Press.
Muggah, Robert, and Adriana Erthal Abdenur. 2018. “Refugees and the City: The Twenty-First-Century Front Line.” World Refugee Council Research Paper No. 2.
Scheel, Stephan, and Funda Ustek-Spilda. 2019. “The Politics of Expertise and Ignorance in the Field of Migration Management.” Environment and Planning D: Society and Space 37 (4): 663–681.
Sigona, Nando. 2015. “Seeing Double? How the EU Miscounts Migrants Arriving at Its Borders.” The Conversation, 16 October. http://theconversation.com/seeing-double-how-the-eu-miscounts-migrants-arriving-at-its-borders-49242.
Simpson, Stephen, and Daniel Dorling. 1999. “Conclusion: Statistics and ‘the Truth.’” In Statistics in Society: The Arithmetic of Politics, ed. Daniel Dorling and Stephen Simpson, 414–420. New York: Oxford University Press.
Stricker, Yann. 2019. “‘International Migration’ between Empire and Nation: The Statistical Construction of an Ambiguous Global Category in the International Labour Office in the 1920s.” Ethnicities 19 (3): 469–485.
UNHCR. 1969. Report of the United Nations High Commissioner for Refugees to the General Assembly. Supplement No.11 (A/7211). New York: UN.
UNHCR. 2007. 2006 Global Trends: Refugees, Asylum-Seekers, Returnees, Internally Displaced and Stateless Persons. Geneva: UNHCR.
UNHCR. 2008. 2007 Global Trends: Refugees, Asylum-Seekers, Returnees, Internally Displaced and Stateless Persons. Geneva: UNHCR.
UNHCR. 2009a. 2008 Global Trends: Refugees, Asylum-Seekers, Returnees, Internally Displaced and Stateless Persons. Geneva: UNHCR.
UNHCR. 2010. 2009 Global Trends: Refugees, Asylum-Seekers, Returnees, Internally Displaced and Stateless Persons. Geneva: UNHCR.
UNHCR. 2011b. Promoting Livelihoods and Self-Reliance: Operational Guidance on Refugee Protection and Solutions in Urban Areas. Geneva: UNHCR.
UNHCR. 2016b. “Uganda—Weekly Sector Indicators Matrix for South Sudan Refugees.” 29 October. https://data2.unhcr.org/en/documents/details/52480.
UNHCR. 2017b. “Uganda Refugee Response Monitoring Settlement Fact Sheet: Oruchinga.” 31 December. https://data2.unhcr.org/en/documents/details/63292.
UNHCR. 2019a. “Contributions to UNHCR—2018.” 21 March. https://reporting.unhcr.org/sites/default/files/donor_ranking/2018percent20-percent20UNHCRpercent20Donorpercent20Ranking.pdf?origin = gr2018.
UNHCR. 2019b. “Fact Sheet Uganda, February 2019.” https://reporting.unhcr.org/sites/default/files/UNHCR%20Uganda%20Factsheet%20-%20February%202019.pdf (accessed 9 February 2022).
UNHCR. 2019d. “Refugees and Asylum Seekers Statistics.” 3 January. https://data2.unhcr.org/en/documents/details/73196.
Urla, Jacqueline. 1993. “Cultural Politics in an Age of Statistics: Numbers, Nations, and the Making of Basque Identity.” American Ethnologist 20 (4): 818–843.