Community effects of cash-for-work programmes in Jordan: Supporting social cohesion, more equitable gender roles and local economic development in contexts of flight and migration

Loewe, Markus et al. Research Report Community effects of cash-for-work programmes in Jordan: Supporting social cohesion, more equitable gender roles and local economic development in contexts of flight and migration Studies, No. 103 Provided in Cooperation with: German Institute of Development and Sustainability (IDOS), Bonn Suggested Citation: Loewe, Markus et al. (2020) : Community effects of cash-for-work programmes in Jordan: Supporting social cohesion, more equitable gender roles and local economic development in contexts of flight and migration, Studies, No. 103, ISBN 978-3-96021-135-8, Deutsches Institut für Entwicklungspolitik (DIE), Bonn, https://doi.org/10.23661/s103.2020 This Version is available at: https://hdl.handle.net/10419/227755 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Markus Loewe Tina Zintl Jörn Fritzenkötter Verena Gantner Regina Kaltenbach Lena Pohl Community Effects of Cash-For-Work Programmes in Jordan Supporting Social Cohesion, More Equitable Gender Roles and Local Economic Development in Contexts of Flight and Migration Community effects of cash-for-work programmes in Jordan Supporting social cohesion, more equitable gender roles and local economic development in contexts of flight and migration Markus Loewe Tina Zintl Jörn Fritzenkötter Verena Gantner Regina Kaltenbach Lena Pohl The German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) is a multidisciplinary research, policy advice and training institute for Germany’s bilateral and multilateral development cooperation. On the basis of independent research, it acts as consultant to public institutions in Germany and abroad on current issues of cooperation between developed and developing countries. Through its nine-month training course, the DIE prepares German and European university graduates for careers in the field of development policy. Dr Markus Loewe is a Senior Researcher with the research programme “Transformation of Economic and Social Systems” at the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE). Email: [email protected] Dr Tina Zintl is a Senior Researcher with the research programme “Transformation of Economic and Social Systems” at the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE). Email: [email protected] Studies Deutsches Institut für Entwicklungspolitik 103 Community effects of cash-for-work programmes in Jordan Supporting social cohesion, more equitable gender roles and local economic development in contexts of flight and migration Markus Loewe Tina Zintl Jörn Fritzenkötter Verena Gantner Regina Kaltenbach Lena Pohl Bonn 2020 Studies / Deutsches Institut für Entwicklungspolitik ISSN (Print) 1860-0468 ISSN (Online) 2512-8795 © Deutsches Institut für Entwicklungspolitik gGmbH Tulpenfeld 6, 53113 Bonn +49 (0)228 94927-0 +49 (0)228 94927-130 Email: [email protected] http://www.die-gdi.de Except as otherwise noted this publication is licensed under Creative Commons Attribution (CC BY 4.0). You are free to copy, communicate and adapt this work, as long as you attribute the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) and the author. Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar. The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available on the Internet at http://dnb.d-nb.de. ISBN 978-3-96021-135-8 (printed edition) DOI:10.23661/s103.2020 Printed on eco-friendly, certified paper Foreword This report presents the findings of a DIE research project on the indirect effects of cash-for-work projects in Jordan. It is based on an extensive literature review and empirical research conducted in Jordan by the authors of this study from February to April 2019. Accordingly, the report takes into account all developments that took place up to 2019 but not more recent ones, especially not what happened with the CfW programmes in Jordan during the COVID-19 crisis. Preliminary findings of the research project have been presented and discussed at several workshops and conferences: 17 and 25 April 2019 in Amman; 14 May 2019 in Luxemburg; 27 May 2019 and 4 December 2019 in Bonn; 6 November 2019 in Frankfurt on Main; 31 October 2019 in Cairo; and 25 August 2020 at the ERF Annual Conference Webinar Series. We did our best to take all feedback to the preliminary findings into consideration when writing the final draft of this report between January and May 2020. We should stress that the results, even though preliminary, have received substantial attention in the Jordanian media: for example, on 27 April 2019, the Jordan Times reported in details on the DIE research project (“CfW programmes exhibit”, 2019). In this report, all names of persons are spelled as the individuals wished. The names of towns and other geographical terms have been transliterated into English according to American Library Association and the Library of Congress (ALA-LC) guidelines, the city of Amman being the only exception because its name is regularly cited in press and academic papers in the same non-ALA-LC-conform way. By agreement, quotations by our interviewees cannot be attributed by name, date, or affiliation. We promised them beforehand to treat all information given by them confidentially. Most sincere thanks go to all persons in Jordan, Germany and elsewhere who have supported us in our research. In particular, we express our deep gratitude to all interlocutors for their cooperation and the warm and pleasant atmosphere we were received in. Our very special thanks are addressed to the Jordanian Center for Strategic Studies (CSS) at the University of Jordan, and in particular to Yasmin AlDamen for her excellent and diligent support and feedback to our research. In addition, we are also most grateful to Julie Weltzien and her entire team of the GIZ project Improvement of Green Infrastructure in Jordan, who have supported our empirical research tremendously and allowed us to use the data collected during the first round of their GIZ Post-employment Survey conducted between January and November 2019. In this respect, we are heavily indebted to our student assistant Radwa Hosny, who supported us with the cleaning and statistical analysis of the quantitative data. Finally, we express our most sincere thanks to all our interpreters as well as our bus drivers – without them, our research would not have been possible. We are grateful for all the helpful feedback and comments we received on earlier drafts of this study, from Assia Aldhabbi, Yasmin AlDamen (CSS), Tilman Altenburg (DIE), Ines Dombrowsky (DIE), Charlotte Fiedler (DIE), Lukas Frank (KfW), Anja Gaentzsch (BMZ), Jörn Grävingholt (DIE), Nico Herforth (DEval), Lisa Klinger (GIZ), Alexander Kocks (DEval), Maria Ghauri-van Kruijsdijk (GIZ), Jana Kuhnt (DIE), Kathrin Löber (BMZ), Sarah Christin Meier (KfW), Silvia Morgenroth (BMZ), Karina Mross (DIE), Franke Neumann-Silkow (GIZ), Jakob Rieken (GIZ), Nicole Roy (GIZ), Helge Roxin (DEval), Imme Scholz (DIE), Ralf Senzel (GIZ), Bernhard Trautner (DIE), Helke Wälde (KfW), Ruben Wedel (DEval), Julie Weltzien (GIZ) and Bettina Zoch-Oezel (KfW). All remaining errors are ours. Bonn, November 2020 Markus Loewe Tina Zintl Jörn Fritzenkötter Verena Gantner Regina Kaltenbach Lena Pohl Contents Foreword Abbreviations Executive summary 1 1 Introduction 11 2 The indirect effects of cash-for-work: the analytical framework 12 2.1 Social cohesion 13 2.1.1 Definition and measurement 13 2.1.2 Factors and drivers 14 2.1.3 The context of conflict and forced migration 15 2.2 Gender 16 2.2.1 Definition and measurement 16 2.2.2 Factors and drivers 17 2.2.3 The context of conflict and forced migration 18 2.3 Local economic development (LED) 19 2.3.1 Definition and measurement 19 2.3.2 Factors and drivers 20 2.3.3 Multiplier effects 22 2.4 Cash-for-work programmes 23 2.4.1 Definition 25 2.4.2 Effects 26 3 Country context: Jordan as a haven for refugees 31 3.1 The socio-economic situation of Jordan 31 3.2 Refugees in Jordan 32 3.3 Social cohesion in Jordan 36 3.3.1 Political and national identity in Jordan 36 3.3.2 Government initiatives to promote national identity and social cohesion 37 3.3.3 Effects of refugee presence 38 3.3.4 The effect of external aid 39 GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit (German agency for technical cooperation) GJU German-Jordanian University GPIA Protestant Institute of Archaeology (Jordan) ICDF International Cooperation and Development Fund (Taiwan) ILO International Labour Organization IO international organisation JBW Jordanian Bird Watch JLMPS Jordan Labour Market Panel Survey JOD Jordanian dinar JOHUD Jordan Hashemite Fund for Human Development JRS Jesuit Refugee Service KAS Konrad-Adenauer-Stiftung (Konrad Adenauer Foundation) KfW KfW Development Bank (Germany) LED local economic development LFPR labour force participation rate MENA Middle East and North Africa MoI Ministry of the Interior (Jordan) MoL Ministry of Labour (Jordan) MoMA Minstry of Municipal Affairs (Jordan) MoPIC Ministry of Planning and International Cooperation (Jordan) MU Maastricht University n/a no answer NAF National Aid Fund (Jordan) NARC National Agricultural Research Center (Jordan) NGO non-governmental organisation NORAD Norwegian Agency for Development NRC Norwegian Refugee Council NREGA/NREGS National Rural Employment Guarantee Act/Scheme (India) NZF National Zakat Fund (Jordan) ODA Official development assistance ODI Overseas Development Institute (London) REACH A joint initiative of IMPACT Initiatives, ACTED, and the UN Operational Satellite Applications Programme (UNOSAT), providing data and analysis RSCN Royal Society for the Conservation of Nature (Jordan) SAP structural adjustment programme SEZ special economic zone SOP standard operating procedure UNICEF United Nations Children’s Fund UNOPS United Nations Office for Project Services UNRWA United Nations Relief and Works Agency USAID United States Agency for International Development USD US dollar VNG International Vereniging van Nederlandse Gemeenten / International Activities of the Association of Netherlands Municipalities WANA West Asia-North Africa Institute (Jordan) WASH water, sanitation and hygiene WFP World Food Programme WtPE Waste to (positive) Energy (one of three GIZ CfW programmes) WVI World Vision International German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 1 Community effects of cash-for-work programmes in Jordan Executive summary Interest in cash-for-work (CfW) programmes has increased immensely over recent years because – being both a social transfer scheme and an instrument of passive labour market policies – they contribute to multidimensional poverty reduction in multiple ways. Existing evidence shows a range of positive effects in lowand middle-income countries: CfW programmes are able to provide a double or even triple dividend, if implemented well: they deliver (i) wage employment (that is, work, income and social protection) to vulnerable people; (ii) strongly needed infrastructure such as roads, sanitation, irrigation systems or others, as this is where the labour force is put to use; and sometimes even (iii) skills development among participants if explicitly included in the setting up of the programmes. Yet, there is sparse evidence of how CfW programmes fare in two regards: First, though CfW programmes have recently become a popular instrument in contexts of civil war and forced migration, little is known about how they operate in these contexts. Second, we do not know how they affect the communities in which they are implemented, thus having an indirect effect beyond the direct effects mentioned above. This study examines to what extent and under which circumstances CfW programmes foster (i) local economic development; (ii) more equitable gender roles; and – particularly relevant in contexts affected by flight and migration – (iii) social cohesion for the wider community in which CfW activities take place. At the same time, the report asks under which circumstances – that is, using which kind of CfW project designs – these indirect effects best translate into host communities becoming more resilient and thus contribute to improving fragile contexts. The study builds on empirical research conducted in Jordan which is a relevant country case as it has welcomed very large numbers of Syrian refugees since 2011; the social fabric of host communities has changed a lot and pressures on the local labour and housing markets are high – all this in an already difficult economic situation for most Jordanians. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)2 Markus Loewe / Tina Zintl et al. CfW’s indirect effects: the analytical framework In this study, community effects of CfW programmes are understood as social and economic consequences for the villages where such programmes are implemented. To best grasp these indirect effects on communities we combine the concepts of social cohesion and local economic development while applying a gender-sensitive approach. We rely on a definition of social cohesion specifying four components: (i) social identity, which can be understood as people’s sense of belonging to a community; (ii) horizontal trust (trust between different groups in society); (iii) vertical trust (trust between society and the state); and (iv) willingness to engage in fostering common goods (for example, irrigation channels or clean streets and parks). In our research, however, we focused mainly on social identity and horizontal trust because we expected CfW to affect mainly the relations between Jordanians and Syrians and their sense of belonging to their respective communities. Gender is the second component of our analysis, assuming that CfW programmes entail important elements empowering women socially but especially economically, which in the Jordanian context is all the more important since women are affected disproportionally by flight and displacement. Local economic development as the third dependent variable follows the assumption that CfW programmes (i) build infrastructure that improves the income-generating possibilities within the community; (ii) employ people and provide them with additional income; and (iii) also increase the income of other community members via a multiplier effect, potentially resulting in more investments, business activity and employment opportunities. This aspect is highly important in the Jordanian context because the influx of Syrian refugees has put higher pressure on scarce resources and added to widespread underand unemployment. The country context: Jordan as a haven for refugees Large numbers of Syrian refugees have added to already existing refugee populations in Jordan and triggered a response by the international community which came in multiple forms, but especially as numerous CfW programmes. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 3 Community effects of cash-for-work programmes in Jordan Due to its geographical location in the Middle East, Jordan’s history has been characterised at different points in time by the need to accommodate large numbers of refugees. Refugees and migrant workers have shaped Jordanian history and are still today an important factor of Jordan’s economic and societal life until. In 2018, Jordan hosted around 671,000 Syrian refugees registered with UNHCR (UNHCR [United Nations High Commissioner for Refugees], 2018), which corresponds to approximately 7 per cent of the country’s population. At the same time, the huge number of refugees and migrants poses significant challenges to social cohesion and local economies as Jordan struggles to offer services and employment to the growing population. This is in a context where a difficult economic situation and continuously high unemployment mean that several parts of society are struggling to make ends meet and threaten to strain relations between different societal groups (for instance, tribes; Transjordanians and Palestinian Jordanians) and state legitimacy. CfW and other social transfer programmes in Jordan Jordan spends 12 per cent of GDP on public pension, health and social transfer schemes but most of these schemes only cover formal sector employees, thereby excluding the poor. Furthermore, most of the schemes cover only Jordanians; Syrian refugees only have access to the public health and education system. Foreign donors have therefore set up parallel social transfer programmes explicitly targeting Syrian refugees. Some of them are unconditional cash benefit and voucher schemes but since 2016, as agreed on in the Jordan Compact agreement, a whole range of CfW programmes has been established by various donors, one of the main ones being the German Federal Ministry for Economic Cooperation and Development (Bundesministerium für wirtschaftliche Zusammenarbeit und Entwicklung, BMZ). All employ Jordanians along with Syrians, and women along with men because they are meant to (i) support Syrian refugees as well as vulnerable Jordanians; (ii) strengthen social cohesion between Syrians and Jordanians; (iii) reduce competition on the labour market; and (iv) promote the integration of women into economic life. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)4 Markus Loewe / Tina Zintl et al. The total budget of these CfW programmes over the last five years has been about EUR 300 million and they have employed at least 70,000 workers during this time – typically for a period of 40-90 working days – in the rehabilitation and cleaning of infrastructure, the collection and recycling of waste, the rehabilitation of eco-systems, the creation of municipal parks, and the intensification of agriculture. Research methodology In order to find answers to our research question, we applied a predominantly qualitative research methodology. In a first, step, we formulated 15 hypotheses on the effect of CfW programmes on social cohesion (Hypotheses 1-10), local economic development (Hypotheses 11-15) and gender roles (crosscutting issue and covered in the other 15 hypotheses) – our dependent variables. We intentionally devised such a large set of hypotheses in order to be able to identify (i) which characteristics of the programmes, namely the joint participation of Jordanians and Syrians, the creation of useful assets, the wage payments, or the existence of the programmes as such (independent variables) and (ii) which specific programme design choices, such as targeting, duration or the skills training component (intervening variables), enlarged or minimised the community effects found. In a second step, we spoke to 380 interviewees in 295 semi-structured interviews with experts (national and local experts, mostly representatives of donor and implementing agencies) and community members (CfW participants, non-participants and shopkeepers) during a 3-month field research stay in Jordan. Most interviews (281) took place in ten villages in which CfW activities were being implemented. We selected these villages using two criteria: the village was to be as small as possible and also as far away as possible from other villages so that the community effects of CfW programmes would be noticeable for as many village members as possible and would not unfold mainly outside the respective village. Generally, we gained field access to the sites through international donor agencies and their local implementing partners. Just as the CfW programmes themselves, most of our field sites were situated in northern Jordan, but some were also in central and southern Jordan. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 5 Community effects of cash-for-work programmes in Jordan In addition, we drew on primary data stemming from the Post-employment Survey conducted by GIZ’s Green Infrastructure programme during 2019 among their CfW participants. Findings: community effects of Jordan’s CfW programmes Our research confirms that CfW programmes do not only have positive direct effects at the individual level but also noticeable positive indirect effects at the community level. CfW programmes strengthen social cohesion We found that the participation in CfW programmes strengthens several components of social cohesion. Especially CfW participants and Syrians reported that their sense of belonging and trust in the respective other nationality (horizontal trust) had increased recently and that this was largely due to CfW programmes. The direct effect on CfW participants seemed to be much stronger than the effect on other members of the community. This finding was corroborated by the fact that respondents perceived the joint employment of Jordanians and Syrians in the CfW projects as most beneficial for social cohesion, while the existence of these projects as such and the creation of public infrastructure had only limited effects on the levels of social cohesion felt. The relations between Syrians and Jordanians and their trust towards the respective other group were already good before the introduction of the programmes, so although the positive effect of CfW programmes on social cohesion was noticeable it did not cause a major overhaul of relations. Still, since economic pressures in Jordan are not only high but increasing, heightened trust through CfW programmes could help to prevent potential future tension. The effect on participants’ and other community members’ trust in state authorities (vertical trust) is less clear-cut. While we noticed that CfW programmes frequently had quite positive effects, respondents often saw foreign donors as mainly responsible for the programmes – which was not always the case. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)6 Markus Loewe / Tina Zintl et al. Lastly, some CfW activities were found to affect individuals’ willingness to cooperate for the common good. This was particularly often the case for CfW programmes in the waste sector, which seemed to raise community members’ awareness of the importance of waste collection and recycling in many cases. CfW programmes offer incentives for more equitable gender roles CfW programmes in Jordan contributed to a positive change in the public perception about female labour force participation in several communities covered by our study. The working experience and, often, new skills learnt through CfW programmes were much appreciated by female participants and this was often voiced more clearly than in interviews with male participants. However, these positive effects seemed to depend strongly on two factors: that the work environment was considered suitable for females, and that a woman’s family was in dire need of a second source of family income. So far, working in public places or in mixed-gender teams has been taboo for many Jordanian and even more so for Syrian women. The majority of CfW programmes take this into account, for instance by tasking women with less physically demanding activities, so that many women saw CfW programmes as a suitable and “safe environment” in comparison to other job opportunities. Some programmes, however, did form mixed-gender teams, and their female members reported positive experiences. In general, many female participants planned to look for further CfW opportunities after the programme ended. At the same time, women were not always prepared to accept another, non-CfW job, meaning that their labour market entry may only be temporary. Our evidence also shows that many women worked in the CfW programmes because their families desperately needed the extra income and that these women would not necessarily continue to work, or look for work, if their families’ financial circumstances became better. Thus, increased female labour force participation does not necessarily mean a permanent change in attitude but may represent a temporary exigency. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 7 Community effects of cash-for-work programmes in Jordan CfW programmes support local economic development (LED) As participants indeed spent their income mostly locally and CfW programmes tried to source their building materials locally, increased business activities and a multiplier effect were traceable but difficult to quantify. Local shopkeepers reported higher sales figures and debt repayment rates, but we did not find instances of investments made because of higher revenue. Participants’ consumption patterns focused on basic needs, which included housing (rent, electricity, and water), food, household equipment, and debt repayment. Thus, there was hardly any investment effect as the vast majority of participants were not able to save and invest part of their income (thereby confirming the successful targeting of the programmes). At the same time, we found that female and male participants’ spending patterns differed, so CfW programmes seeking to employ women in particular may result in the multiplier effect being channelled in a slightly different direction, for example, spending related to education or health. The income effect generated by the CfW-built infrastructure was either difficult to gauge or negligible: Some infrastructure, mostly connected to agriculture (rehabilitation of dams, water reservoirs and irrigation systems; intensification of agriculture, slowing soil erosion) does have an effect on local economic development. Other infrastructure, such as parks, playgrounds or school renovations, certainly has a positive effect on the quality of life of residents, yet is unconnected to any income-generating activities. Labour market effects are not clear-cut. In regard to employability, CfW participants improved their soft and – depending of the training component of a particular programme – also their technical skills, yet in most cases this did not translate into good job prospects after the end of the employment in CfW programmes (due to several factors, for instance, the poor general economic situation; skills unneeded on the primary labour market; or – for refugees – skills tied to an economic sector protected against non-national workers). Individual accounts showed that, in a number of cases, the relatively high CfW wage caused crowding-out but that the effect was not strong. On the positive side, CfW also reduced the in Jordan pervasive so-called “shame Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 14 to trust each other. In this study, we call this type horizontal trust. The second type is political trust that measures underlying trust in the “formal, legal organisations of government and state, as distinct from the current incumbents nested within those organisations”. 4 We call this type of trust vertical trust. Inclusive identity (sense of belonging): The concept of an inclusive social identity is analytically distinct from the concept of personal identity. Personal identities are inherently subjective, whereas social identities are grounded in a shared understanding among individuals regarding specific social groups. Individuals can hold multiple social identities at the same time and subjectively ascribe emotional significance and priority to them. The more individuals in a given society agree over the meaning and content of their common social identity, the more cohesive that society is. Within the context of this research, we call this variable sense of belonging since this term better captures whether or not a person’s social identity is related to his/her local community. Cooperation for the common good: DIE understands cooperation for the common good as an individual’s voluntary consent “to take into account interests higher than his own”.5 The concept also includes the willingness of an individual or a group to pay a cost or make a concession in order for the larger community to receive a benefit. Therefore, the concept goes beyond the related notion of social capital that measures the willingness to cooperate for individual and/or mutual benefits. The attributes described above can be evaluated across three dimensions: between the state and individual (vertical); between groups (intergroup, horizontal); and between individuals (interpersonal, horizontal). Indicators are almost always measured by assessing data on public opinion. 2.1.2 Factors and drivers Few studies have produced robust results on what indicators effect a community’s social cohesion. Exceptions include political trust, the legitimacy of government, and social protection, which all seem to have 4 Definition from the documentation of an internal workshop at DIE on 9-10 July 2018 on “Social Cohesion in Africa: Concept and Measurement”. 5 Definition from the documentation of an internal workshop at DIE on 9-10 July 2018 on “Social Cohesion in Africa: Concept and Measurement”. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 15 a positive effect on social cohesion (Burchi et al., 2020). Köhler (in press) argues in the same way that the existence of social protection can have a positive influence on social cohesion. Wietzke (2014) finds that formal employment correlates with social behaviours that are typically associated with higher degrees of social cohesion. Ariely (2014) provides evidence that, in contrast, some forms of diversity (such as ethnic and linguistic fractionalisation) have a negative effect on general trust and solidarity and, hence, on social cohesion. Similarly, Langer, Stewart, Smedts, and Deamrest (2017) show that, in countries where group identities are strongly prioritised over national identities, national social cohesion is more likely to be threatened – and that societies with low levels of social cohesion tend to face more violent conflicts (Langer et al., 2017). Furthermore, various researchers suggest that full integration into a socially cohesive society is more difficult for particular societal groups, including women who tend to face more burdens and challenges than male members of society (Segalo, 2015). As Anzaldua (1999) assumes, this may be the case because the lives of differing societal groups (of different gender, race, religion, and so on) are actually often so interwoven that it becomes difficult to distinguish between “insiders” and “outsiders” despite their differing living experiences. Women, in most societies, are perceived as “insiders” and legally allowed to participate in the community (for instance, by entering the labour force and being household heads); thus, though they face higher burdens in doing so, they become full members of a socially cohesive society (Anzaldua, 1999). 2.1.3 The context of conflict and forced migration Increasingly, researchers, policymakers and the general public around the world are associating the notion of social cohesion with positive and negative changes to society as a result of immigration and refugee movements. However, there are very few studies on the effects of the presence of migrants and refugees on the social cohesion of a host society, especially in lowand middle-income countries (Langer et al., 2017). In addition, limited data sets and the risk of politicisation make it difficult to measure the effects of refugees on social cohesion objectively (Ariely, 2014; Schiefer & van der Noll, 2017). Having said that, some stress that social protection is particularly important in these contexts (Kool & Nimeh, in press). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 16 2.2 Gender Gender was the second dependent variable of our analysis. We wanted to investigate the experiences and specific needs of women and men in the context of CfW programmes and to try to find out to what degree CfW programmes can empower women within the context of crisis and migration. This is particularly important in the Jordanian context because flight and displacement often affect women disproportionately. In the following, we (i) define the term “gender”; (ii) discuss why a gendered perspective is particularly important in the context of flight and migration; and (iii) explain why the empowerment of women is also important for social cohesion and local economic development. 2.2.1 Definition and measurement For our purposes, we adopted Carol Cohn’s definition whereby gender is “a structural power relation […,] a social system which shapes individual identities and lives” (Cohn, 2013, p. 3, emphasis in original). Gender thus constitutes an “organizing principle” (Boyd & Grieco, 2003, p. 2) and refers to the social construct of being male or female and consequences emanating from this categorisation (Cohn, 2013, p. 3). The term “gender” thus differs from the term “sex”, which refers to the biological attributes of men and women (UN DESA [United Nations Department of Economic and Social Affairs], 2004). Gender norms shape roles, expectations, identity constructions, and behaviours associated with masculinity and femininity (Hondagneu-Sotelo, 1994). Since gender is grounded in social interactions, it varies across regions, societies, and time. Existing gender norms can empower or constrain rights and opportunities. For example, in patriarchal societies, the mobility and agency of women tend to be restricted. We understand gender as fluid and non-binary, including identities that go beyond feminine and masculine. In addition, “women” and “men” are not monolithic categories. However, for reasons of practicability, this study still refers predominantly to the categories of male and female, although we do our best to take the diversity of experiences of both – females and males – into account. We have chosen to highlight in particular the experience of women. Of course, a gendered analysis of conflict should look as well at the set of attributes, behaviours, and roles associated with boys and men. However, Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 17 as women often – and especially in Jordan – face more social constraints than men, the participation of women in social, economic and political life is particularly important for development (see, for example, Cuberes & Teignier, 2012; or Sen, 1999). Progress in terms of gender can be measured in many different ways. Ultimately, gender discrimination can exist with regard to every aspect of human, economic, societal and political development. Therefore, the best way for measuring gender equality is to disaggregate all commonly used indicators of development by gender – an approach that has been taken for the Sustainable Development Goals which we also take in our study. Instead of formulating separate indicators for gender inequality, we disaggregate indicators for social cohesion and local economic development by gender. 2.2.2 Factors and drivers Many women are already powerful – or, in the context of flight and migration: resilient – and thus do not need to be empowered. For this reason, we understand the empowerment of women rather as a structural measure to support their own struggles and to open up new opportunities to fulfil their economic and social potential. Categories to define the empowerment of women include […] women’s sense of self-worth and social identity; their willingness and ability to question their subordinate status and identity; their capacity to exercise strategic control over their own lives and to renegotiate their relationships with others who matter to them; and their ability to participate on equal terms with men in reshaping the societies in which they live in ways that contribute to a more just and democratic distribution of power and possibilities. (Kabeer, 2008, p. 27) Empowering women can be crucial for social cohesion and local economic development. A social structure that is marked by the traditional division of labour between men and women (productive versus reproductive) and spaces (public versus private) may hamper the sense of belonging of women and their horizontal trust in other groups (Pateman, 1988). Likewise, an economy in which half of the population is mostly restricted to unpaid housework cannot strive as much as more dynamic and flexible societies. Accordingly, in its “Declaration on Women’s Economic Empowerment for Peacebuilding” the UN Peacebuilding Commission calls upon member states to “take measures to promote sustainable livelihoods for households led by Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 18 women […] in post-conflict societies, including through financial support and access to productive resources and sustainable income-generating activities” and stresses the “importance of assisting post-conflict countries in creating favourable conditions that can generate decent jobs for women” (UN Peacebuilding Commission, 2013, Para. 10). Improving the access of women to social protection is an integral element of such efforts (Jones, in press). 2.2.3 The context of conflict and forced migration A gendered lens on conflict and forced migration uncovers the unequal burden that women are carrying, hampering their access to resources, their participation, or policy changes in favour of the diverse interests of women. In conflicts, women are both a potential force to reduce security threats through their inclusion in social and political affairs, and a population facing specific risks (Bunch, 2008). While men are the majority on the battlefield, they leave their wives behind. Some of them will turn into widows and will hence suddenly be solely responsible for securing their and their children’s livelihood. Others may become victims of sexual abuse (Jacobsen, 2013). In addition, they are often affected by forced displacement. After conflicts, conditions tend to exacerbate women’s already unequal economic and social status relative to men […]. Often, dire economic conditions after conflict foster corruption and criminality, while marginalised groups of women experience extreme income inequality, working in the informal economy and the most precarious employment positions in the labour market. (True, 2013, p. 3) A return to violence can always occur in volatile post-conflict contexts and, in fact, domestic violence increases after conflicts end (Chinkin & Kaldor, 2013). Gender affects all aspects of the migration and refugee experience. It influences the access to resources as well as treatment within economic, social and legal structures. Female migrants and refugees are more vulnerable to physical, sexual, and verbal violence carried out by members of host communities, public officials or by other refugees and migrants. Those who are unaccompanied, pregnant, heads of households, disabled, or elderly are especially vulnerable. This perspective does not negate that male migrants and refugees are exposed to vulnerabilities, too. However, female migrants Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 19 face double discrimination, as women and as migrants. They are generally less mobile when they need to take care of children or live in socio-cultural settings in which they fully depend on their husbands. In addition, more and more crises are protracted, prolonging the plight of displaced persons and forcing them into a state of long-term emergency, often depending on international aid assistance (Giles, 2013). Subsection 2.4.2 shows in what way CfW programmes can help to empower women both socially and economically. 2.3 Local economic development (LED) The term “local economic development” (LED) refers to a concept that shows the complexity and the interplay of the various dimensions of economic processes at the local level. We will use it to understand the economic benefits of CfW for communities (as opposed to economic benefits for individual participants). In the following, we (i) provide a working definition of the term; (ii) elaborate on possible drivers of LED; and (iii) explain the so-called “multiplier effect”, which transforms a singular payment into a repeated benefit. 2.3.1 Definition and measurement The concept of LED describes the sustainability of economic development processes at a local (that is, municipal or quarter of town) level. Just as economic development in general is more than just economic growth, we consider LED to be a multidimensional process, as well. In particular, equity, the inclusion of vulnerable groups and the reduction of multidimensional poverty (see Box 1), are taken into consideration. Local communities’ social, environmental, and political aspects of development are as much foci of analysis as are local labour, commodity and capital markets. This understanding is well reflected in the International Labour Organization’s definition of LED as “promoting participation and local dialogue, connecting people and their resources for better employment and a higher quality of life for both men and women” (ILO [International Labour Organization], 2018). Following this definition, LED entails the two goals: quality of life and employment. While measuring employment is rather straightforward – and commonly done in national statistics collecting unemployment or labour Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 20 market participation figures – quality of life is much more abstract and difficult to measure. People’s subjective perceptions about their material living conditions, their capabilities to access health or education services, as well as their social interactions, the respect of their basic rights and environmental pollution also play a role. In the local context, ways in which material living conditions are affected and leveraged by multiplier effects deserve special attention (see below). Box 1: The conceptual framework of multidimensional poverty This study understands poverty as multiple deprivation of basic capabilities. Capabilities are “the substantive freedoms [a person] enjoys to lead the kind of life he or she has reason to value” (Drèze & Sen, 2013, p. 43). These are determined not only by income and wealth, but also by education, health, social inclusion, political rights and many more factors. The capabilities of any person therefore depend not only on the person’s place of living and working but also on age, gender and social origin (class, family reputation, caste, ethnicity). Thus, we argue – in accordance with the Development Assistance Committee of the Organisation for Economic Cooperation and Development (OECD-DAC, 2001) – that poverty results from deprivation of one or more of the following subsets of capabilities: • Economic capabilities refer to the ability to generate income, consume and have assets. • Human capabilities include health, education, nutrition as well as access to clean water and shelter. • Political capabilities comprise the respect for human rights, opportunities of political participation and having some effect on public policies and political priorities. • Socio-cultural capabilities are the ability to take part as a valued member of a society. • Protective capabilities are resilience, that is, the ability of people to resist economic and external shocks. 2.3.2 Factors and drivers In addition to traditional drivers of economic development, such as institutions, physical and human capital and technology, social or political capacities, locality factors and local business cycles are also important drivers of LED. The combination of those drivers and their resulting augmenting/multiplying effects also play an important role. For example, strong social or political Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 21 capacities can compensate limited resources. However, weak community or political capacities, which could be due to corruption, disorganisation, or cronyism, can hamper LED so that an endowment with natural resources might not necessarily translate into good capacities. Thus, infrastructure, natural resource availability, geographical location, labour markets, capital investment, entrepreneurial climate, transport, communication, industrial composition, technology, size, export market, international economic situation, and national and state government spending can all be considered drivers of LED (Blakely & Leigh, 2017). As CfW programmes are expected to achieve at least a double dividend by offering wage employment and creating infrastructure, this study highlights labour markets and infrastructure as key drivers of LED. Labour markets: Labour markets are an essential element of LED. The skill level of the workforce in a region is an important factor for attracting industries. Thus, the ability to upgrade the skill level of the workforce through training, education and development is crucial for a region to remain competitive and respond to changing labour demand (Pike, RodríguezPose, & Tomaney, 2006). Various models highlight the importance of job creation and retention for LED (Salvini, in press). One example is the exportbase/primary-jobs model that focuses on the effects of creating “primary” jobs producing goods and services for export outside the respective local economy. The generation of income from the sale of these products increases purchasing power and demand for other products, which can be secondary or tertiary and also offer new employment opportunities (Greenwood, Holt, & Power, 2010; see also below). Infrastructure: Functioning infrastructure is essential for any kind of economic activity. In the context of LED, infrastructure subsectors such as energy, water and sanitation, telecommunications and transportation are particularly important. However, access to physical infrastructure alone does not foster gross domestic product (GDP), economic growth, or social returns at a macro-level. Studies find that a high level of poverty and bad governance weaken the effect of infrastructure on economic growth, while a competitive environment and well-made and clear regulations are associated with higher payoffs (Estache & Garsous, 2012). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 22 2.3.3 Multiplier effects In general, a multiplier effect can translate even minor inputs into considerable outputs. This effect stems from different processes, the most famous being the circulation of money within a closed economy: Here, every expenditure raises the income of not only the recipient but of many more people because the recipient of the payment again spends the additional income for her/his own purchases, and then the same money is spent over and over again to other people. So, unless an initial payment is entirely saved or spent outside the community, at least part of it is recycled at least once and thus benefits a second person as well. Unless she/he saves the additional income entirely or spends it entirely outside the community, this second person will pass it on again or at least some portion of it, and so on. In the end, at least parts of the initial payment can thus be passed on infinite times within the community benefitting many people rather than just the primary recipient. In 1936, John Maynard Keynes labelled this phenomenon the “multiplier effect” (Keynes, 2007, p. 117). And the multiplier m can be computed as follows: = (1 ) = 1 + =1 + =1 + =0 Where s is the average share of the income that people save and i is the average share of the income that people spend outside the community. There are two variants of this original multiplier effect: The first variant comes into play when employment is created: Somebody invests in a factory (for instance, a bread factory) and creates employment for several people in the region who achieve higher income and can then buy the products of the investor (that is, the bread). The second variant is the so-called capacity effect, which results from a onetime investment in assets that enable, ease, or cheapen production. Thereby, owners of production facilities enjoy increasing sales figures or falling production costs. The local community also benefits because the additional income or saved spending can be used for other items, again adding to the multiplier effect. For the analysis of LED, the multiplier effect is particularly important because it boosts the effects of one-time expenditures. This applies not least to the wages paid out by CfW programmes (Barrientos, 2008). However, of course, the size of the effect depends totally on the assumption that the Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 23 additional income is indeed spent locally. For example, if businesses are not owned by locals, or if labour is imported from outside the region, or local capital flows out of the region, the local community will benefit less from economic activity (Soifer, 2014). In addition, there is very little empirical evidence so far for the existence of the multiplier effect of CfW but there are a few studies on other social transfer programmes (Bhalla, Kangasniemi, & Winder-Rossi, in press). Barrientos (2008) stressed that most empirical research has focused on the impact of social transfers at the household level and has thereby found multiplier effects mainly at the level of assets and the consumption by beneficiaries. A case study by Robinson & Levy (2014) on Cambodia found that social transfers have more positive effects on economic development if implemented along with productivity-enhancing local policies. A World Bank research project concluded that social cash transfer programmes in Africa “have a nominal income multiplier ranging from USD 1.34 to USD 2.52 for each USD 1.00 transferred” (World Bank, 2015, p. 2). Egger, Haushofer, Miguel, Niehaus and Walker (2019) estimated the multiplier effect of cash transfers in rural Kenya to be approximately USD 2.6 for every US dollar. 2.4 Cash-for-work programmes CfW programmes provide employment and income to people in need. They can hence be seen as an instrument of passive labour market policy as they offer low-wage employment to people who face difficulties finding a job on the primary labour market. But, at the same time, they can also be seen as social transfer schemes providing benefits only on the condition that recipients work for the construction or rehabilitation of public goods such as physical infrastructure (for example, roads, water systems, drains); environmental goods (for instance, municipal parks, river beds, forests); or services enhancing human capital (such as public health, education) (Loewe & Schüring, in press). This dual labelling is not astonishing as there is a huge overlap in any case in the instruments used by social protection and labour market policies. The fact that the receipt of transfers is conditional has a dual purpose: One is targeting, that is, to make sure that only people in need will benefit from the programme – often referred to as “self-targeting” or the “self-selection” mechanism. People who have a job or sufficient assets would not apply for a short-term job that requires hard work for a comparatively low income. The Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 30 transfer scheme in Tanzania increased the vertical trust in local governments. In addition, they revealed that the effect depended to some degree on people being aware of the fact that their local government was in fact involved in the design and establishment of the conditional cash transfer scheme and was larger where community members participated in the design process.9 Adato (2000), in contrast, provided evidence that conditional cash transfers in Mexico had positive effects on community social relationships. However, CfW only fosters political stability when it also considers institutional sustainability. It is important to analyse how CfW fits into the local social protection framework, and whether international aid organisations can build on these structures in order to implement CfW programmes (Gehrke & Hartwig, 2018, p. 121; Zepeda & Alarcón, 2010, p. 3) in order not to harm vertical trust. Gender roles: CfW programmes have a potential to empower women both economically and socially (Jones, in press). They can provide women with access to labour markets in contexts where female labour force participation rates are low. Furthermore, they can enable women to participate in the rehabilitation of public goods, to move and act in public, as well as, through their income, to possibly gain more influence over household decisions. However, most CfW programmes are implemented as short-term measures to ease sudden financial shocks rather than to alleviate poverty in general or to foster social protection in the long term; hence they are unlikely to empower women in a lasting way. In many countries, women tend to enter the labour market only if they have to because of financial stress rather than because they wish to gain equal rights. These women face chronic cycles of impoverishment and it is hard to determine whether they have made an active choice to join the labour market or have been forced into it by the underlying circumstances (Kabeer, 2011). In other words, even if CfW programmes have a positive effect on women’s economic empowerment, such programmes may not change the social roles of women and men as research on the Indian NREGS shows (Sudarshan, 2011). Whether women actually participate in CfW programmes depends on several factors. The willingness of women tends to rise (i) the more equally men and women are paid in the programmes; (ii) the closer CfW activities are to women’s houses; (iii) the more similar the activities are to what 9 Bastagli et al. (2019), Camacho (2014), and Hunter and Sugiyama (2014) provide similar results. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 31 women already do outside the household (for example, if the activities are in agriculture); and (iv) if there are day-care facilities for the children of women next to the CfW sites (Kabeer, 2011). In contrast, if there are not, women may participate but with the unintended negative effect that older daughters are taken out of school in order to look after their younger siblings instead of the mother (Dammert et al., 2017; Sudarshan, 2011). 3 Country context: Jordan as a haven for refugees We now turn to the case of Jordan, which is a very illustrative example of a country that is strongly affected by flight and migration. In fact, refugees and migrant workers have shaped its history since independence in 1946 and they have been an important factor of Jordan’s economic and societal life up to today. At the same time, they create significant challenges for social cohesion and development, as Jordan is a middle-income country struggling in any case to offer social services and employment even to its own growing population. This section describes the socio-economic situation of Jordan (subsection 3.1) and the situation of refugees in Jordan (subsection 3.2). It then reports on social cohesion in Jordan (subsection 3.3), the gender situation (subsection 3.4), and local economic development (subsection 3.5). 3.1 The socio-economic situation of Jordan Jordan is an upper middle-income economy which has achieved a comparatively high level of human development even though it (i) started from a very low level of human development in 1946; (ii) is almost landlocked; (iii) is surrounded mainly by unstable neighbouring countries; (iv) has hardly any natural resources; and (v) has had to integrate huge numbers of migrants throughout its history. This achievement can be attributed to a forward-looking and balanced foreign policy and considerable financial support from both Western and Gulf countries throughout the decades (World Bank, 2016a). The huge majority of its population is urban, with a high concentration in and around the capital city Amman and the north of Jordan. Between 2010 and 2016, the population grew from 6.7 million to 9.8 million (Department of Statistics Jordan, 2016; Krafft, Assaad, & Keo, 2018). One factor for Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 32 this increase was the arrival of large numbers of refugees from Syria, who constitute a considerable additional burden for the development of the country which is already suffering from limited water resources, bottlenecks in infrastructure, and lack of employment (see in more detail below). Yet, the presence of large numbers of refugees is by far not the only challenge for Jordan’s social and economic development. The country is still struggling to attain a coherent national, social, and political identity since its creation by European powers in the first half of the 20th century. Its society and politics are deeply influenced by strong rival tribes and an authoritarian monarchy. Both cooperate to uphold their say in society and polity, thereby limiting the potential for democratisation. This stalemate is nurturing growing discontent within the population and provokes sporadic protests, which culminated in weekly demonstrations in front of the prime minister’s office in Amman between June and December 2018 and a march of people from the south all across Jordan to the Royal Court in February 2019, where they staged a sit-down strike. While in June 2018 people demonstrated mainly against a tax reform (which was then postponed and later brought back onto the table in a slightly different form), at the end of 2018 discontent was growing especially among the youth, who saw no promising perspectives anymore in their country. They felt that nothing had really changed since their protests almost seven years earlier during the so-called Arab Spring 2011 (Ramadan, 2018). 3.2 Refugees in Jordan Due to its geographical location in the Middle East and its relative political stability, Jordan’s history is characterised by the reception of large numbers of refugees throughout the decades. Although this trend had started even before the country became independent, it accelerated thereafter: • In 1948/1949, some 500,000 Palestinians fled from lands that became part of Israel and added to the local population which counted just about 400,000 people itself. • In 1967, some 380,000 Palestinians left the West Bank and the Gaza Strip, which had both been occupied by Israel during the Six-Day War. • Between 1975 and 1990, about one million Lebanese and Palestinians came from Lebanon to escape the Lebanese Civil War. (Most Lebanese returned though during the 1990s.) Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 33 • In 1982, a small group of Syrians came to escape from the massacre of Hama. • In 1991, some 350,000 Jordanian and Palestinian migrant workers returned from the Gulf countries: first from Kuwait because of the country’s occupation by Saddam Hussein’s Iraq, and then from SaudiArabia and other countries, which expelled people with Jordanian passports to sanction the fact that Jordan had remained neutral in the war between the Gulf states and Iraq. • During the 1990s, some 250,000 Iraqis fled to Jordan to escape from persecution, violence, and the poor economic situation in their own country. • During the 2000s, another 750,000 Iraqis followed to escape from the civil war that had started after the US American occupation of their country. (However, most Iraqis have in the meantime returned to their country.) • At the same time, a considerable number of Egyptians came as labour migrants. (Many returned to Egypt again, but some 600,000 Egyptians still reside in Jordan.) • After the uprisings in several Arab countries in 2011, about 20,000 people came from Libya, about 30,000 from Yemen, some from Egypt and Somalia, and up to 1.3 million from Syria (De Bel-Air, 2016). Today, Palestinians represent the largest group of non-natives in Jordan even though it is difficult to estimate their total share of the population: all Palestinians who came until 1949 or resided in the West Bank until 1967 were conceded Jordanian passports, thus they and their descendants are counted as Jordanians in national statistics. Guesses of their number range between 1.6 and 4.5 million (25-70 per cent of the total population of Jordan), which would mean that only between 2 and 5 million (30-75 per cent) of the population of Jordan originate in fact from its current territory. Exact data exist only on the number of Palestinians who came to Jordan from the Gaza Strip or Israel after 1949 and their descendants (634,182) because they were not given a Jordanian passport but only an identity card (Department of Statistics Jordan, 2015; see also Figure 1). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 34 Figure 1: Jordanian population by country of origin Jordanians and Palestinians with Jordanian citizenship: 6,605,389 69% Palestinians without Jordanian citizenship: 634,182 7% Egyptians: 636,270 7% (Number may have decreased since 2015) Iraqis: 130,911 2% Yemenis: 31,163 >1% Libyans: 22,700 >1% Other: 197,385 2% R egistered Syrian refugees: 661,114 7% Syrians having come to Jordan before 2012: 603,886 6% (But many authors argue that this number is highly inflated) Notes: The methodology and results of the census are highly debated in the academic literature. Especially the figures for Syrians having come to Jordan before 2012 and the number of Iraqis living in Jordan are said to have been inflated (Lenner, 2020). In addition, the number of Egyptians living in Jordan may have decreased significantly since 2015 (Krafft, Razzaz, Keo, & Assaad, 2019). Source: Authors, based on data from the Department of Statistics Jordan (2015) publishing results of the Jordan Population and Housing Census of 2015 Apart from the Palestinians, the United Nations High Commissioner for Refugees (UNHCR) in Jordan by 2019 registered about 654,000 Syrians, 67,000 Iraqis, 15,000 Yemenis, 6,000 Sudanese and 2,500 people from 25 other countries (UNHCR, 2020). On top of this, many more unregistered refugees and migrant workers are living in Jordan (see Figure 1). These include, among others, an estimated 17,000 Syrian Palestinians who could not register because the government of Jordan forbid Palestinians from Syria from seeking refuge in Jordan (Grawert, 2019, p. 37). Some estimates set Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 35 the actual number of Syrians living in Jordan to more than 1.2 million (see Figure 1), while other sources argue that their number cannot be much higher than 500,000 (Krafft et al., 2019; Lenner, 2020). Finally, Jordan also hosts a considerable number of migrant workers, some 600,000 alone from Egypt (Schubert & Haase, 2018, p. 102; see also Figure 1). As only about 20 per cent of refugees live in camps, their presence has a strong effect on the economic and social life in their host communities. The refugees need water, food, shelter, clothes and education for their children, as well as medical treatment (Hagen-Zanker & Mallett, 2016; Schubert & Haase, 2018). Despite the fact that the Jordanian-Syrian border has been recently opened again, it is unlikely that many Syrian refugees will return to Syria in the coming years. In general, Syrian refugees only returned infrequently as there are still persistent concerns regarding the security situation in their country and due to a new Syrian property law that effectively expropriates many refugees of their previously owned premises (HRW [Human Rights Watch], 2018). A study on the mobility of Syrian refugees finds that, in addition to the security situation, low provision of education, health and basic services in Syria deters refugees from returning (World Bank, 2019a). In order to gain official recognition and enjoy some social benefits (such as food services and free treatment in Jordan’s public health system, see subsection 4.1), Syrian refugees need to register with an office of the United Nations High Commissioner for Refugees (UNHCR) and then with the Jordan Ministry of the Interior (MoI), which gives them a magnetic “MoI service card” or “security card” (biṭāqa amniyya). By March 2019, 654,266 Syrians had registered with UNRWA (United Nations Relief and Works Agency) (out of whom 299,168 of working age) (UNHCR, 2020) but only about 405,000 with the MoI (Hagen-Zanker, Ulrichs, & Holmes, 2018). On average, the Syrians have just JOD 285 (Jordanian dinar) to spend per household per month, and 69 per cent of their spending is for housing – just 11 per cent for health and food, 9 per cent for education; and 8 per cent for transportation. As a result, 17 per cent of adolescents suffer from hunger, while 35 per cent report chronic illnesses (Jones et al., 2019). Although Syrian children are officially entitled to go to school, 20 per cent of them (and even more than half of those older than 15 years) are still not yet enrolled. Net attendance rates are 47 per cent lower among Syrian Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 36 refugees of secondary school age than among their Jordanian peers. And those who go to school are normally educated in a separate afternoon shift. Of Syrian students, 41 per cent have experienced corporal punishment and discrimination (Jones et al., 2019). 3.3 Social cohesion in Jordan Despite the presence of large numbers of refugees, social cohesion in Jordan is not weak, though it is under pressure. The country benefits from the fact that traditional social norms are still strong, that there is some degree of national identity, and that the dominant group identity is with the tribes (especially in rural regions and among original Jordanians), which stretch across state borders far into neighbouring countries. This subsection provides an assessment of social cohesion and national identity in Jordan (3.3.1); information on national campaigns that have been launched to strengthen it (3.3.2); some preliminary evidence on the effect of refugee and migrant presence (3.3.3); and the effect of international aid (3.3.4) on social cohesion in Jordan. 3.3.1 Political and national identity in Jordan Jordan is still struggling to gain a national political identity due to its origins as a former British colony – after the fall of the Ottoman Empire, Jordan’s borders were effectively drawn by European powers –, its political system headed by the traditional monarchy, and the broader political and social landscape in the Arab world. Since the dissolution of the Ottoman Empire, three different visions of political community – Arab nationalism (qawmiyya); state-centric nationalism (wataniyya); and the Islamic ummah (global community of Muslims) – have struggled for cultural loyalty within the Arab world, thus challenging Jordan’s national identity and social cohesion within the country as well (Lucas, 2008). In addition, and perhaps even more importantly (at least for the original Jordanians but also, perhaps to a lower degree, for many original Palestinians), the tribes claim to be the natural and foremost focus of identity instead of the modern nation state. The tribes have indeed a strong say on political decisions. While the country is by law a constitutional monarchy, parliamentary autonomy has in fact Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 37 been minimal, and the king holds a significant share of the power. Despite elections being held every three years, the process of democratisation in Jordan has frequently been described as a facade. The parallel existence of two systems of political authority – the traditional authority of the monarchy and the state-authority of institutions – has created a situation of incompatibility that leaves the question of national identity to open debate (Lucas, 2008). But even the king’s legitimacy derives from traditional claims of kinship, religion, and Bedouin identification (Al Oudat & Alshboul, 2010; Melián Rodríguez, 2018). As a result, even the king is not free to take decisions on major issues but always has to consider the opinions of the main tribes of the country. At the same time, social cohesion is strong within local communities. Several studies provide evidence of Jordanians having a strong feeling of belonging to their respective local communities and having much trust in their neighbours – even if these neighbours have recently arrived from outside the community or even outside Jordan. Key drivers of mutual trust seem to be age, common interests, and geographic proximity rather than nationality or religion (see, for instance, Kuhnt, Rischke, David, & Lechtenfeld, 2017). 3.3.2 Government initiatives to promote national identity and social cohesion Since the early 2000s, the Government of Jordan has increasingly emphasised the need to build social cohesion and national identity within the country through a variety of campaigns and national plans. In 2002, King Abdallah II launched a national campaign under the slogan “Jordan First”. It aimed at propagating the principles of rule of law, accountability, pluralism, and equality as a means to modernise the Jordanian society and unify it behind a common national goal (Al Oudat & Alshboul, 2010). Nevertheless, state institutions substantially control political and civic associations, and Jordanian law still prohibits critical statements concerning the king, government institutions, and Islam (HRW, 2019). Since the start of the Syrian refugee crisis, the Government of Jordan has also focused more strongly on building social cohesion and resilience within local communities. The three national plans formulated since 2015 to mitigate the refugee crisis – the Regional Refugee Resilience Plan; the Jordan Response Plan; and the three-year National Resilience Plan – all include various social cohesion initiatives, particularly in the labour sector, Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 38 such as the creation of employment opportunities, vocational training offers, apprenticeships, and job-placements (REACH & British Embassy, 2014). Unfortunately, only little research has been conducted on the effects of these efforts on social cohesion within Jordanian host communities (Kuhnt et al., 2017). 3.3.3 Effects of refugee presence In major host communities, social cohesion seems to have suffered from the presence of refugees, though only modestly. REACH found 160 communities in Jordan’s northern governorates (‘Ajlūn, Al-Balqāʾ, Jerash, Irbid, Al-Mafraq and Al-Zarqāʾ) to have experienced a weakening of social cohesion as a result of the Syrian refugee crisis (REACH & British Embassy, 2014). Likewise, Kuhnt et al. (2017) argue that the level of social cohesion in Jordanian communities (measured by mutual trust, sense of belonging, and participation in community events) has decreased slightly on average during recent years – in particular between Jordanians and non-Jordanians (Kuhnt et al., 2017). This trend is mainly due to the fact that the presence of refugees in Jordan is exerting additional pressure on already strained infrastructure, social services, and the labour market (see subsection 3.5.3). A joint report by UNHCR, UNICEF (United Nations Children’s Fund) and the WFP (World Food Programme) (2014, pp. 8, 10) stresses that tensions have erupted several times between different population groups in northern Jordan mainly because of increasing stress on resources such as water, housing and jobs. Reportedly, the situation in Al-Mafraq has deteriorated most significantly. Here, rents skyrocketed and wages fell considerably because large numbers of Syrians in the town were ready to accept any shelter or any job to survive (Grawert, 2019, p. 20; Mercy Corps, 2012, p. 3). Likewise, increasing demand for water has reduced the available amount of water per inhabitant, which has led to tensions in some municipalities. While many schools have adopted double-shifts to also accommodate Syrian children, classrooms are still over-crowded and the quality of education has dropped (Kuhnt et al., 2017). Still, these trends have not put an end to the deep solidarity felt between Jordanians and Syrians. Betts, Ali, and Memişoğlu (2017), for example, acknowledge that tensions and conflicts over scarce resources increased Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 39 when Syrians arrived in large numbers in the Al-Mafraq area. But they also mention that the solidarity between Jordanians and Syrians has endured because of longstanding kinship ties between the Jordanian host communities in Al-Mafraq and the communities of origin of many Syrian refugees. Guay (2015, p. 17) also suggests that different identities should be taken into account beyond the binary setup of host-community versus refugees. Interestingly, many reports highlight the north of Jordan as the place where social cohesion has been particularly under threat because most Syrians found refuge there, while we had the impression that social cohesion between Jordanians and Syrians in the north of Jordan was astonishingly strong (see subsection 6.1). In addition, we often heard that there were much greater tensions in the south of Jordan. Observers explained this by the fact that people in the north of Jordan and southern Syrian belong to the same tribes, while there are quite different tribes in the south of Jordan. Because of this, the initial solidarity might erode faster in the south and centre of Jordan than in the north. Grawert (2019) shows for two “Palestinian” quarters of Amman that the solidarity with Syrians was initially high (perhaps because of the common experience of flight) but has been eroding with time because of conflicts over resources. Apparently, social cohesion has decreased in schools in particular, with worrying consequences for refugee children. UNICEF finds that Syrian refugee children and young people are increasingly becoming isolated and face mounting problems in access to education (Mercy Corps, 2012, p. 5). Other sources confirm that 1,600 Syrian children dropped out of the Jordanian school system in 2016 alone (Matteo Valenza, 2016; Grawert, 2019). Kuhnt et al. (2017) also note that tensions between young men are a growing concern, with Jordanian schools being among the places where social tensions erupt most frequently and openly. 3.3.4 The effect of external aid In some communities, foreign aid has also contributed to weakening social cohesion. For instance, some foreign donors have spurred competition over scarce resources by providing housing subsidies to Syrian refugees: these subsidies have turned out to be goldmines for Jordanian house-owners and severe disadvantages for Jordanian tenants bidding for the same apartments as Syrians (Grawert, 2019, p. 23). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 46 At the same time, numerous jobs have been created recently but have not been filled with Jordanians. Instead, company owners employed migrant workers from Egypt, China or the Philippines or refugees from Syria, Iraq and Yemen claiming that many Jordanians requested higher wages and refused certain kinds of work (for example, in construction, agriculture or in private households), a phenomenon which is often referred to as “shame culture”. Many foreigners, in contrast, are even prepared to work without working contracts, social insurance, and workplace protection (World Bank, 2016a). Some authors (for instance, Abbott & Teti, 2017; and even Jordan’s National Employment Strategy 2011-2020, MoPIC [Ministry of Planning and International Cooperation, Jordan], 2011) suggest that many Jordanians are in fact gambling: they accept unemployment for a while because they still hope to get a well-paid job in the public administration or in the Gulf, in Europe or North America one day. As a result, the Jordanian labour market is still segmented in several ways, and it is difficult to move from one segment to another: Internal versus external: The most important segmentation is that between Jordanians working at home and those who have found a job abroad. The latter account for as much as 15 per cent of the total working population of Jordanian citizenship: around 8 per cent work in the Gulf, 4 per cent in North America, and 2 per cent in Europe. On average, their incomes exceed by far the incomes of those Jordanians who work in their own country. Public versus private sector: In addition, well over 21 per cent of Jordan’s total working population enjoy an employment in the public sector (army, security services, administration, judiciary, higher education, state-owned enterprises). This large number of jobs constitutes an important part of the social contract in Jordan. The total number of public sector jobs has declined in recent years, but the state still uses the allocation of employment to reward certain groups of the population for their loyalty. As a result, most public officials are from originally Jordanian rather than Palestinian families, and the army employs almost exclusively members of the most reliable tribes. Salaries in the public sector tend to be lower than in the private sector but, for many Jordanians, this is more than compensated by job security, pension entitlements, health insurance, and regular working hours (World Bank, 2016a). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 47 Migrants versus Jordanians: As a result of the huge number of refugees and migrant workers, the Jordanian labour market is roughly comprised of three different groups – Jordanians, refugees, and migrant workers (Razzaz, 2017). Around 28.8 per cent of the labour force are non-Jordanians according to the Jordan Labour Market Panel Survey (JLMPS) 2016. Egyptian and other Arab migrants constitute the big majority of non-Jordanian workers, followed by Syrians. Further, the majority of male workers is Arab, while most of the female foreign workers are Asians who come as migrant domestic workers. Many of the foreign workers are either low-skilled or have higher education profiles and compete with their Jordanian peers for jobs that require higher education (World Bank, 2016a). Formal versus informal sector: Within the private sector, there is additional segregation between formal and informal jobs. Two-thirds of all workers in the private sector are informally employed (self-employed or employed without an employment contract) (World Bank, 2016a). However, while half of all Jordanian nationals have a formal job in the public or private sector, the huge majority of the Syrians and other nationalities are in the informal sector (see Figure 2). This shows that Syrians are likely to compete over jobs with other migrants rather than with Jordanians (Krafft, Fallah, & Wahba, 2018). Figure 2: Employment status of workers aged 15-54 by nationality, 2016 42 1 5 5 25 9 18 12 57 52 11 9 7 5 25 12 Jordanians Syrians Other n ationalities Public sector employee Employer Formal private sector employee Informal private sector employee Self-employed Day labourer Source: Authors, based on Krafft et al., 2018 Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 48 3.5.3 Syrians in the Jordanian labour market and the Jordan Compact Syrians face significant barriers to entering the Jordanian labour market.12 Like other non-Jordanians, they must apply for a work permit from the Ministry of Labour. In addition, they are not allowed to work in so-called “closed occupations” (which were designated as such several years ago as an instrument to promote the “Jordanisation” of the labour market), even if there are no qualified Jordanians to fill vacant jobs in these sectors (World Bank, 2019a). And, finally, all migrants are subject to a minimum wage of just JOD 150 per month (EUR 190), which is below the minimum wage rate for Jordanians (JOD 220 per month) (ILO, 2017). For this reason, the Jordan Compact was established in 2016 between Jordan and the international community. In light of the growing urgency to transform humanitarian interventions into livelihood-building development cooperation (Crawford, Cosgrave, Haysom, & Walicki, 2015), it was meant to help Syrians generate their own income and thereby shift the engagement of donors “from short-term humanitarian aid to education, growth, investment and job creation” for Jordanians and Syrians alike (Barbelet, Hagen-Zanker, & Mansour-Ille, 2018, p. 2). As part of the compact, the Government of Jordan promised to issue 200,000 work permits for Syrians and to provide formal jobs in specific economic sectors such as food processing, handicrafts, and tailoring.13 So far, however, the government has tended to formalise existing jobs rather than creating new ones. What is more, most of these jobs are low paid, with poor working conditions and are located only in the special economic zones (SEZs) which are typically far from the places where most refugees live. This forces Syrians to depend on the poor public transportation system to get to the jobs (Staton, 2018), often meaning that Syrian women are not able to take up jobs in the textiles sector because childcare duties render long commutes impossible (Lenner & Turner, 2018). In addition, employers in 12 In the summer of 2020, the government of Jordan indicated in bilateral donor talks that it planned to ease the regulations for attaining work permits and open additional occupations for refugees (internal communication with BMZ country desk officer, September 2020). 13 The main reason for both the earlier closure of the Jordanian labour market and the current sector restrictions were concerns on the part of the Government of Jordan that – like Palestinian refugees before them – Syrians would settle in Jordan permanently (Grawert, 2019, pp. 40-41). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 49 SEZs do not seem to be very interested in employing Syrians as they are content with other migrant workers (mostly Asian workers living on site; Lenner & Turner, 2018) and, in turn, Syrians are often not familiar with the SEZs (Tiltnes et al., 2019, p. 13). As a result, only one-third of all Syrians of working age have a work permit so far (Tiltnes et al., 2019, p. 13). Most try to get a job on the informal labour market, where they compete with other migrants but also with low-skilled Jordanians of Palestinian origin (Grawert, 2019). One reason is that the work permits are associated with too few benefits (Grawert, 2019) and with high costs for social security contributions (World Bank, 2019a). In addition, many Syrians are still not aware of the fact that they can get a work permit: half of the interviewees of a survey conducted in 2019 did not know of this option (Jones et al., 2019). In particular, only very few Syrian women have received a work permit. By mid-2019, 156,761 work permits had been issued but only 7,875 were for women (UNHCR, 2020). One possible factor is that, among Syrian refugees, the sectors that are open for Syrians are not considered suitable for women. Lenner (2020), however, suspects that some officers in the Jordanian administration are also more reluctant to issue work permits to women than to men. 4 CfW and other social transfer programmes in Jordan Jordan has a fairly elaborate public social protection system which includes both contributory social insurance and non-contributory social transfer schemes, but these schemes cover only Jordanian citizens and long-term residents. Because of this, foreign donors have set up in parallel social transfer programmes targeting Syrian refugees (in part along with vulnerable Jordanians), which are entirely financed by multilateral and bilateral foreign donors and run by international or national non-governmental organisations or subordinate public institutions (coordinating only loosely with Jordan’s ministries). The bulk of these latter programmes are CfW schemes, which are again predominantly funded by German development cooperation. This section proceeds as follows: Subsection 4.1 gives an overview of social transfer schemes provided by the Government of Jordan for Jordanian citizens. Subsection 4.2 highlights the fact that there has already been a Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 50 tradition of international donors taking care of refugees in Jordan since Palestinians migrated to Jordan in the wake of the war against Israel in 1948. Subsection 4.3 presents the social transfer schemes that foreign donors have set up, at least in part to support Syrian refugees. 4.1 Social transfer programmes provided by Jordanian institutions The government of Jordan has established numerous social protection schemes but many of them target the urban middle-class rather than the poor. Public spending on social protection and health is high compared to other lowand middle-income countries. In 2012, it accounted for 12 per cent of GDP (Loewe & Jawad, 2018, p. 10). However, the bulk of spending was on health systems for public sector employees and on public pension schemes which covered only 60 per cent of the population. In addition, the membership conditions and benefit levels are different for three groups of people: (i) members of the armed forces; (ii) civil servants; and (iii) private sector employees. Informal sector employees in particular are completely excluded, although they comprise the poorest parts of society (Loewe et al., 2001; Loewe, 2019). At the same time, less than 10 per cent of all public social protection spending in Jordan (that is, less than 1 per cent of GDP) relates to social transfer programmes, including unconditional social assistance, CfW, fee waiver and in-kind programmes – even though they are particularly important for lowincome earners. In fact, all of them together cover 85 per cent of the bottom income quintile of the population although this is mainly due to the health treatment fee waivers that the Royal Court gives on request to households in difficult socio-economic situations. Cash transfers alone reach out to only about 20 per cent of the bottom income quintile while the benefits are too low to close the poverty gap (Loewe, 2019). Another problem is the lack of cooperation and coordination between the different schemes. This results in partial overlaps of provisions and nonharmonisation of targeting rules. Considerable financial means are lost because of these deficiencies and heavy administrative burdens in general (Loewe et al., 2001). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 51 Social transfers are provided by three different institutions: • The National Aid Fund (NAF) administers several social assistance programmes although these only target Jordanian citizens. It is supervised by the Ministry of Social Development with the objective of providing support to poor and vulnerable groups of the population. The main activity of the NAF is the provision of monthly cash transfers to people in need. The transfers are in the range of JOD 40-180 (EUR 50-223) per household, depending on its size and composition. For the selection of beneficiaries, the NAF combines a proxy-means-test with socio-categorical targeting (Zureiqat & Abu Shama, 2015). In theory, only households with an income below the national abject poverty line of JOD 336 per individual and year (Department of Statistics Jordan, 2016) are eligible to receive the benefit (ILO, 2019) – and only if there is no male of working age (15-64 years) without a work-disability in the household (Röth, Nimeh, & Hagen-Zanker, 2017; Loewe et al. 2001, p. 30). Any income received by one of the family members decreases the benefit received by 25 per cent (ILO, 2019). In addition, the NAF considers income-producing property, arable land, or possession of a car (unless used by a disabled member of the family) to be disqualifying factors. In practice, however, with few exceptions, only persons without any income qualify for the programme (ILO, 2019). In the aftermath of the Syrian crisis, the NAF’s budget increased by JOD 3 million annually so that it can now support an additional 20,000 Jordanian families. In 2013, the Ministry of Social Development made an estimate that it would need another EUR 9.96 million to meet the needs of the growing numbers of vulnerable Jordanians in the context of the influx of Syrians. According to the NAF, the fund sought to increase the number of households benefitting from its regular assistance from 75,000 (in 2015) to 98,000 (in 2019) and the number of people receiving emergency assistance from 5,000 (in 2017) to 7,000 (in 2019) (Röth et al., 2017; “Majority of NAF beneficiaries”, 2018). • The National Zakat Fund (NZF) operates a different social transfer scheme on behalf of the Ministry of Awqāf (religious endowments), Islamic Affairs and Holy Places. The scheme provides cash assistance to orphans and very poor households who do not receive social assistance from the NAF (Loewe et al., 2001). In 2015, the NZF extended one-time emergency assistance to more than 30,000 families along with regular Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 52 assistance to 43,000 orphans and 30,000 poor families, mostly through its 210 local zakāt (Islamic religious tax) committees. However, the benefits were very low, ranging from JOD 15 to 30 per household and person (Mechado, Bilo, & Helmy, 2018). In contrast to the NAF, nonJordanians are theoretically eligible for support if they are very poor. In practice, however, the only known case of the NZF ever having provided support to Syrians was a one-time transfer of JOD 200,000 in 2014 to the Za‘atarī Camp, which was meant to finance food support for Syrian refugee families (Zureiqat & Abu Shama, 2015, p. 32). • In 2015, the Ministry of Social Development started to open community centres serving “as a shelter for men, women and children who are victims of human trafficking, covering their basic needs and providing counselling and rehabilitation services” (Hassan, 2015). The Ministry also provides legal counselling services in cooperation with civil society organisations. Three other institutions provide social services in parallel: • The Jordan Hashemite Fund for Human Development (JOHUD) works through a network of 50 community development centres, mainly in rural areas. It is a foundation and considers itself an NGO. Nonetheless, it has been established by the state authorities and gets its core funding from the central government budget. That is why it makes sense to list it next to state institutions carrying out similar activities. These include awareness campaigns, vocational training and skills development, child and family health care, the promotion of productive activities of women, and microenterprise development as well as education and cultural programmes. JOHUD addresses women in particular (Loewe et al., 2001, pp. 38-39). • The Ministry of Education is responsible for Jordan’s public primary and secondary education system. It is free of charge for all Jordanians, and, in 2012, the Government of Jordan extended the free access to Syrian refugees with the financial support of the international donor community (Hagen-Zanker et al., 2018). However, in most schools, Syrian children are in separate classes and their lessons take place in the afternoon, when the Jordanian children have already left their classrooms. Of course, this is partly due to the fact that the schools cannot accommodate all Syrian and Jordanian children in a single shift. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 53 • The Ministry of Health runs Jordan’s public health system. It does not provide free health care but the user fees are highly subsidised, covering presumably just about 10 per cent of actual costs. Very poor Jordanians can apply for a “green card” issued by the Royal Court, which entitles them to health care totally free of charge (Loewe et al., 2001). Since recently, Syrian refugees can get similar waivers (although these exempt parts of secondary and tertiary health care) provided that they are have registered with the Ministry of the Interior (MoI) in order to get a magnetic “MoI service card” or “security card” (biṭāqa amniyya) (Hagen-Zanker et al., 2018). Until the end of 2019, the Jordanian health system dealt with about 328,000 medical consultations of refugees under this programme (UNHCR, 2019). Both Jordan’s public education and its public health system are chronically overstrained and of low quality, so even many lower-income families pay for private schools and private health care instead. 4.2 Social transfer programmes of foreign donors for Palestinian refugees The United Nations Relief and Works Agency for Palestine Refugees in the Near East (UNRWA) runs the oldest social protection scheme for refugees in the Middle East and North Africa (MENA) region. It acts as a quasipublic institution that offers free primary health care and basic education to registered Palestinian refugees living in official refugee camps in Jordan or in other MENA countries. In addition, its Special Hardship Assistance Programme provides regular food assistance to needy Palestinian refugee families as well as occasional cash assistance for shelter rehabilitation (Loewe, 2019; Röth et al., 2017). Since 2018, however, UNRWA has been facing increasing problems in the provision of support because the United States, formerly a major source of funding, has drastically reduced its financial support (UNRWA, 2018). 4.3 Social transfer programmes of foreign donors for Syrian refugees and Jordanians Early on, both biand multilateral donors started up initiatives to support Syrians who had sought refuge in Jordan from the civil war in their country. On the one hand, they wanted to rescue the Syrian refugees Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 54 from the worst forms of poverty; on the other hand, they also wanted to prevent the destabilisation of Jordan and Syrians’ continued flight towards Europe. Their favourite choice was to set up CfW programmes because these provided employment and income to the beneficiaries, could employ Jordanians together with Syrians, and were thereby expected to benefit host communities as well. We can assume that CfW programmes are currently the core of well over 40,000 Syrian refugee families’ livelihood strategies (Röth et al., 2017). In addition, foreign donors are also running unconditional cash transfer, voucher, winterisation, education, vocational training, employment, and empowerment schemes for Syrian refugees (Röth et al., 2017). Up to 2016, international aid agencies mostly provided unconditional shortterm assistance in the form of ad hoc and one-time payments (Röth et al., 2017). Since then, their initiatives have shifted towards a more long-term commitment. Most Syrian households in Jordan are currently benefitting from regular cash assistance provided by UNHCR and UNICEF for an unlimited period of time. However, UNHCR only provides between EUR 65 and 360 per month and household depending on the household size (that is, 30-160 per cent of the official minimum wage of Jordanian workers), while UNICEF grants a monthly cash benefit of EUR 25 for each child (Röth et al., 2017). For more than a half of the Syrians, these grants are their only source of income (Jones et al., 2019). Several other donors have set up additional, but smaller, cash transfer programmes (Jones et al., 2019): • The Norwegian Refugee Council (NRC) administers an emergency cash benefit programme with a fixed benefit of JOD 115 per household. • Oxfam runs a transfer programme supporting Syrian refugees with nonfood purchases. • The World Food Programme (WFP) grants unconditional food vouchers to 95,000 beneficiaries in refugee camps and 430,000 in host communities with co-funding from the German Foreign Office. In addition, it extends school meals in cooperation with the Jordanian Ministry of Finance (298, WFP).14 14 All expert interviews in Amman are listed by number in Appendix A5. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 55 • UNHCR coordinates a large-scale winterisation scheme, to which NRC and Save the Children contribute. It distributes blankets, heating equipment and clothes, and extends cash transfers during cold months. • CARE International runs a cash-for-education-and-protection programme, which supports roughly 3,000 households with co-funding from the German Foreign Office. The programme mainly targets outof-school children and their families with the goal of preventing early marriage. Beneficiaries receive a monthly cash transfer of EUR 90 per household for an initial ten months, provided that all children attend school regularly and do not marry before the age of 16. It is important to note that, even though these programmes were introduced in response to the influx of Syrian refugees, Jordanian residents benefit from most of them, too. This is due to the fact that the Government of Jordan requires 30-50 per cent of the beneficiaries to be vulnerable Jordanians (Röth et al., 2017). CfW programmes became a favourite instrument of international development cooperation in Jordan after the Jordan Compact in 2016. The German Federal Ministry for Economic Cooperation and Development (BMZ) especially has become very active in the use of CfW schemes as a central element of its Beschäftigungsoffensive Nahost (“Partnership for Prospects”, BMZ [Bundesministerium für wirtschaftliche Zusammenarbeit und Entwicklung], 2018). Explicitly, this initiative is meant to ease the financial stress of Syrian refugees and vulnerable Jordanians, to strengthen social cohesion between them, to reduce competition in the labour market and to promote the integration of women into the labour market (BMZ, 2018). At the same time, the WFP, UNHCR and UNICEF receive funds from other bilateral donors as well. The total budget of CfW programmes in Jordan has been about 300 million EUR over the last couple of years (see Table 1). Typical CfW activities in Jordan include the rehabilitation, development and cleaning of “grey” and “green” infrastructure (such as streets, dams, schools and health clinics; and water reservoirs, irrigation systems, municipal parks and ecosystems, respectively), the collection and recycling of waste, and support for the intensification of agriculture (see Figure 3 and Appendix D). Some CfW programmes, such as the Food/Cash-for-Training (FFT/CFT) programmes of the WFP, offer training courses in addition to employment so as to improve participants’ capabilities on the labour market (298, WFP; and 308, AA; and 242, Najmah). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 62 Figure 4: Overview of research hypotheses Design featuresSocial cohesion Sense of belonging (to local community) Horizontal trust Vertical trust Local economic development (LED) Per capita income (multiplied + invested) Employment opportunities Quality of life+ Cash for work (CfW) Participation in project design Targeting procedure of CfW programme Number of working days 15*** Existence of a CfW scheme for Jordanians and Syrians Participation of Jordanians and Syrians in CfW Creation of useful assets by CfW Wages paid by CfW programmes Qualification of participants by CfW programme Cooperation for the common good* 1*** 2*** 3 4 5 10 10 10 9 9 9 8 8 8 14 11** 7 6 12** 13 Gender Notes: *“Cooperation for the common good” was initially not included in our research design (therefore the dashed line) but turned out relevant when we analysed the data collected. **After the pilot phase of our field study we stopped following Hypotheses 11 and 12 related to the dimension “quality of life”. ***For Hypotheses 1, 2 and 15, a gender-differentiated analysis is particularly important. Source: Authors Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 63 Hypothesis 2 was similar to Hypothesis 1 in that it assumed that CfW programmes had a positive effect on people’s sense of belonging. However, it was built on the assumption that the effect goes back to the participation of people in CfW activities rather than the sheer existence of such activities. Evidence confirming the hypothesis would be if only CfW participants attested an increase in their sense of belonging or if their sense of belonging increased substantially more than the sense of belonging of non-participants. Finally Hypothesis 3 purported that the sense of belonging of Jordanians and non-Jordanians alike mainly increased because of the creation of public goods, that is, that they were proud of their community and felt part of it because of improvements in street infrastructure, school buildings, dams, municipal parks, nature reserves, or other public goods. Table 2: Hypotheses related to the sense of belonging to a community Thesis Categories 1. The existence of CfW programmes as such increases people’s sense of belonging to their respective community.* Sense of belonging of participating Syrians and Jordanians Sense of belonging of eligible Syrians and Jordanians who have not yet participated 2. The participation of Jordanians and Syrian refugees in CfW programmes creates a sense of belonging in the respective other group.* Sense of belonging of participating Syrians Sense of belonging of participating Jordanians 3. The creation of useful assets by Jordanians and Syrians creates a sense of belonging to the community for the participants. Sense of belonging of Syrians Sense of belonging of Jordanians Note: *For Hypotheses 1 and 2, a gender-differentiated perspective is particularly relevant. Source: Authors Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 64 5.1.2 Hypotheses 4-5, related to the horizontal trust between people from different social groups Hypotheses 4 and 5 focused on the effects of CfW programmes on horizontal trust. Horizontal trust is the trust of people in people from a different social group in their respective local community. Hypothesis 4 was that the existence of CfW programmes as such increased the horizontal trust between people from different social groups who were eligible for participation in these programmes. In the specific case of Jordan, the assumption would thus be that the provision of decent incomeopportunities to both Syrian refugees and vulnerable Jordanians had a positive effect on the horizontal trust of people from these two groups in each other and no effect, or possibly even a negative effect, on the horizontal trust of people from other groups (who were not eligible for participation in CfW programmes) in people from the two eligible groups. The main information sources to test both hypotheses were ideally CfW-eligible and CfW-participating people, as well as people who were non-eligible, such as non-eligible Syrians, Palestinians without a Jordanian passport, refugees from Iraq, Yemen or Somalia, and other migrant workers. Hypothesis 5 was similar to Hypothesis 4 in the assumption that CfW programmes had a positive effect on the horizontal trust between people who were eligible for participation in CfW programmes and people from other social groups who were also eligible for participation. In contrast to Hypothesis 4, however, the belief was that this effect was due, in the Jordanian case, to the joint work of Jordanians and Syrians in the CfW activities rather than to the existence of the CfW programmes as such. The hypothesis was based on the idea that people who worked together for a number of weeks got to know each other quite well and were thus able to build trust in each other that helped them to also better trust other members of other social groups. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 65 Table 3: Hypotheses related to the horizontal trust between people from different social groups Hypotheses Categories 4. The existence of CfW programmes as such increases horizontal trust among CfW-eligible groups. Horizontal trust of CfW eligible people 5. The participation of Jordanians and Syrian refugees in CfW programmes creates horizontal trust in the respective other group. Horizontal trust of participating Syrians Horizontal trust of participating Jordanians Source: Authors 5.1.3 Hypotheses 6-7, related to people’s vertical trust Hypotheses 6 and 7 deal with the effects of CfW programmes on people’s vertical trust, in other words on their trust in government institutions, such as municipalities. Hypothesis 6 tested to what extent the existence of CfW programmes as such improved the trust of people who were eligible for CfW participation in those institutions that they believed to be responsible for the initiation or implementation of these programmes. In the Jordanian context that would mean that both Jordanians and Syrians perceived increased trust in the staff of their municipality if they believed that this very municipality had initiated or was implementing the CfW programmes – be this in fact the case, or not. Within the context of Hypothesis 6, we also wanted to find out whom CfW participants perceived as being responsible for the programmes. Hypothesis 7 was also an assumption that CfW programmes increased vertical trust – however that they did this because of the public goods that they produced rather than because of their sheer existence. Here, the idea was that people’s satisfaction in their municipality improved because they appreciated the public goods that CfW programmes created – provided that these people believed that their municipality had initiated or was implementing the respective CfW programmes. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 66 Table 4: Hypotheses related to people’s vertical trust Hypotheses Categories 6. The existence of CfW as such leads to higher vertical trust in perceived implementing agencies among eligible Syrians and eligible Jordanians. Vertical trust of eligible Syrians Vertical trust of eligible Jordanians 7. The creation of useful assets has a positive effect on vertical trust. Vertical trust of all community members Source: Authors 5.1.4 Hypotheses 8-10, related to the effect of CfW programme design features on the sense of belonging as well as on horizontal and vertical trust Just like Hypotheses 12 and 15 below, Hypotheses 8, 9 and 10 refer to the possibility that some features in the design of CfW programmes can constitute interfering variables, that is, that they can increase or decrease the possible positive effects of these programmes on communities. These features denoted whether the targeting of CfW programmes (namely the selection of participants) was perceived as fair; whether community members could participate in the design of CfW programmes; and how long participants were employed. Hypothesis 8 assumed that people who perceived the targeting of CfW programmes as unfair experienced less increase in their sense of belonging and in their horizontal and vertical trust than people who had a positive opinion about the targeting. People often perceived targeting as unfair (i) if certain groups were not eligible to participate in the programmes; (ii) if the duration of participation varied between the programmes; and (iii) if the selection process itself was perceived as unfair (with a negative effect especially on vertical trust). Hypothesis 8 can, thus, be seen as an assumption or an interfering variable that may alter or downsize the effects of CfW programmes on social cohesion as identified by Hypotheses 1-7. Interviews with both eligible and non-eligible community members were expected to reveal to what extent the admission of CfW participants was perceived as fair. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 67 Hypothesis 9 assumed that a lack of participation by community members in how CfW programmes functioned and what they were trying to achieve negatively impacts on their sense of belonging as well as on horizontal and vertical trust. If people were not involved in decisions on CfW activities in their respective community this might cause (i) a sense of isolation or of being left out, thus decreasing their feeling of belonging to the community; (ii) a decreased horizontal trust as there was less meaningful interaction between different groups; or (iii) a decrease in vertical trust in the implementing organisations: where project design features were not made transparent participants may feel that the implementing agency does not work efficiently or in the best interest of the beneficiaries. By interviewing local experts as well as eligible and non-eligible community members, we wanted to find out whether, and to what extent, community members were involved in any consultation process prior to the introduction of CfW programmes in their community. Hypothesis 10 took the varying number of working days between various CfW programmes into account (see subsection 4.3), assuming that a higher number of working days raised the positive effects of CfW programmes on participants’ sense of belonging as well as horizontal and vertical trust. It was expected that interviews with participants would provide information about the possible effects on their sense of belonging and horizontal trust, while interviews with local experts and non-eligible community members would serve this purpose for possible effects on vertical trust in particular. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 68 Table 5: Hypotheses related to the effects of design features on the sense of belonging, horizontal and vertical trust Hypotheses Categories 8. Targeting practices of CfW programmes which are perceived as unfair lower the positive effect of CfW programmes on the sense of belonging, and horizontal and vertical trust. Effect of CfW targeting on vertical trust 9. The possibility to participate in project design increases the positive effects of CfW programmes on the sense of belonging, and horizontal and vertical trust. Effect of participation in CfW design on the sense of belonging Effect of participation in CfW design on horizontal trust Effect of participation in CfW design on vertical trust 10. The creation of useful assets by Jordanians and Syrians creates a sense of belonging to the community for the participants. Effect of number of CfW working days on the sense of belonging Effect of number of CfW working days on horizontal trust Effect of number of CfW working days on vertical trust Source: Authors 5.1.5 Hypotheses 11-14, related to per-capita income and other aspects of quality of life Hypotheses 11-14 were the first set of hypotheses related to LED: They dealt with the effects of CfW programmes on the quality of life of community members as the core element of LED (subsection 2.3.1). Originally, we planned to use both subjective and objective indicators for changes in quality of life. However, after a pilot phase, we excluded Hypotheses 11 and 12 – which had primarily been looking at people’s perceptions of their quality of life – and continued only with Hypotheses 13 and 14, which focused on the effect of CfW on per-capita income as a more objective, though very partial, indicator of quality of life. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 69 Hypothesis 11 (eventually not used during our field survey) was the assumption that different types of public goods (created by CfW) led to changes in different dimensions of quality of life. While green infrastructure and public open spaces may not lead to direct economic advantages, they can, for example, significantly improve health, social interactions, or food security. In contrast, grey infrastructure may provide significant new income opportunities and thus lead to improved material living conditions. Hypothesis 12 (also not used) considered the effect of a participatory CfW project design on local economic needs. We had expected different potential effects: A lack of participation might further support local business elite’s choices for specific sites and sectors. Marginalised groups might experience decreased economic outcomes due to crowding out effects. In contrast, an inclusive project design might facilitate stronger connections between various different economic agents and in turn lead to economic benefits. To test this hypothesis, interviews with eligible and non-eligible parts of the community as well as local experts were informative. Hypotheses 13 and 14, in contrast, focused on economic well-being as one major aspect of quality of life. Both dealt with the positive effects that CfW programmes can have on per-capita income in a direct or indirect way. Hypothesis 13 assumed that CfW programmes raised the average percapital income in their vicinity because they extended additional wages to participating households. These wage payments have both a direct and an indirect effect, the latter going through the multiplier as well as investment and employment effects (see subsection 2.3.3). The focus of our own research was on the indirect effect because we expected it to be more important for the community as a whole (while for participants, the direct effect was, of course, more important) and because the direct effects were already quite well covered by another study conducted by DEval (see Roxin et al., 2020). Hypothesis 14 finally looked at income effects generated by the creation and maintenance of public goods. In order to analyse both hypotheses, we interviewed CfW participants, business people (shopkeepers, bakers, butchers, street vendors, taxi/bus drivers) and local experts. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 70 Table 6: Hypotheses related to per-capita income and other aspects of quality of life Hypotheses Categories 11. The creation and maintenance of useful assets increases the quality of life in the host communities.* Perceived change of quality of life 12. Because of a lack of participation in project design, local economic needs are not sufficiently addressed and the quality of life stagnates.* Perceived change of quality of life/ utility of assets Increased economic deprivation among marginalised groups Deterioration of economic opportunities (crowding-out effect) 13. CfW programmes have positive direct (participants) and indirect (community-level) effects on income. Direct effects Investment effects Multiplier effects 14. The creation and maintenance of public goods increases direct (participants) and indirect (community level) income effects. Direct effects Investment effects Multiplier effects Note: *Hypotheses 11 and 12 were not pursued any more during the field research. Source: Authors 5.1.6 Hypothesis 15, related to employment Hypothesis 15 looked at changes in the employability of CfW participants. It was based on the assumption that participants might find it easier to obtain another employment when their contract with the CfW programme ended if they had acquired additional skills: this would be particularly probable if the programme included a training component. In this sense, the design of CfW programmes was decisive for their effect on the likelihood of participants finding further employment. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 71 In addition, the hypothesis invited a particular focus on analysing gender roles because the high share of women employed in CfW programmes could represent a first point of access into employment for women in particular. Table 7: Hypothesis related to employment Hypothesis Categories 15. Skills acquired through participation in CfW lead to better employment opportunities after completion (both for Syrians and Jordanians).* Direct employment effects Indirect labour market effects Note: *For Hypothesis 15, a gender-differentiated perspective is particularly relevant. Source: Authors 5.2 Research design In order to test our hypotheses, we applied a predominantly qualitative research methodology for three reasons: First, we wanted to have enough flexibility and time to discuss specific issues in depth with our interviewees instead of being obliged to ask them all questions in a standard guideline. Second, we also wanted to capture “soft” variables such as opinions and perceptions. Third, we wanted to be as open as possible to new aspects and issues which had not yet been identified by the literature. The two main sources of information for our study are the existing literature and qualitative semi-structured interviews that we conducted ourselves during February, March and April 2019 with a total of 380 people. Most interviews took place in villages where CfW activities were being implemented, others in Amman or in Germany. In addition, we conducted two focus group interviews with Syrian and non-Syrian refugees, as well as carrying out participant observations at two consultation events. All four events took place in Amman in February and provided valuable background information for later interviews at the field sites. Furthermore, we also used the results of a survey that GIZ’s Green Infrastructure programme had conducted during the year 2019 among all of their CfW participants at the end of their engagement. In the remainder of this study, we will refer to this as the “GIZ Post-employment Survey”. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 78 Table10: Characteristicsoffieldinterviewsites Field site Informationaboutfieldsites CfW programmes Implementer Type of work Irbid Highway and Al-Mafraq Interviews were conducted at different sections of the highway from Irbid to Amman, as well as in Al-Mafraq. KfW/ILO with local contractors Waste collection Al-Azraq Remote, small town in Al-Zarqāʾ Governorate, eastern Jordan, located where the only permanent source of fresh water existed in approximately 12,000 square kilometres of desert, now dried out because of groundwater over-use. However, there are still small wetlands, now forming a wellprotected nature reserve under the administration of the Royal Society for the Conservation of Nature. The city of Azraq has a northern part (8,000 Druze, originally from the Syrian-Jordanian Hawran mountain range, and 3,500 Bedouins), and a southern part (6,000 Chechens). Inhabitants complain about a loss of customers due to a new street bypassing the town, as well as border closures following the wars in Iraq and Syria. ICDF with AAH Waste collection and awareness raising GIZ (Green Infrastructure)* Rehabilitation measures in the nature reserve* Deyr ‘Allā A small town in the Jordan Valley in Al-Balqāʾ Governorate with a population of 7,321 in 2015. Largely dependent on agriculture. GIZ (Green Infrastructure) with NARC Agricultural work GIZ (Waste to positive Energy) Waste collection Faqū’a (Al-Qaṣr) A small town in Al-Karak Governorate. Largely dependent on agriculture. The population was 7,300 in 2015. According to local experts, it hosts some 150 Syrian families. The economic situation is dire, especially for young people: The youth unemployment rate is almost 50 per cent. Also, there is no ATM in Faqū’a, which inhibits LED through CfW programmes, as payments by aid agencies are provided via ATM machines. ILO/Norway with Agricultural Directorate Tree planting GIZ (CfW Water) with DRC* Water dam rehabilitation* Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 79 Table10(cont.): Characteristicsoffieldinterviewsites Kafr Ṣawm A small town in Irbid Governorate, less than 15 kilometres south of the Yarmouk river, which constitutes the natural border with Syria. The population was 11,349 in 2015, including roughly 2,500 Syrians. Most people depend on agriculture: the village is well-known for growing olive trees and pomegranates. GIZ (CfW-Water) with World Vision Cooking (women); water dam rehabilitation (men) Kafr Asad (Ḥawfā) A small town very close to Irbid, within Irbid Governorate. The population was 14,232 in 2015. There are roughly 2,500 Syrians living in the area, with 500 living in Kafr Asad itself. Local experts mentioned protests in front of the municipality in order to raise awareness about youth unemployment and rising prices. WFP with Najmah School rehabilitation GIZ (Waste to positive Energy) with NRC* Waste collection* Tal al-Rummān Village with 1,988 inhabitants in 2015 and a high level of unemployment. There are a few shops, but most people work in agriculture. Few Syrians actually live in the village. GIZ (CfW-Water) with NRC Water dam Umm al-Jimāl Village with 4,524 inhabitants in 2015. People depend mainly on employment in agriculture and trade. Plans to invest in the tourism sector are currently being implemented in order to fight high levels of unemployment. Umm al-Jimāl is in the proximity of the Za’atarī refugee camp, the largest camp for Syrian refugees in Jordan. GIZ (Waste to positive Energy) with Oxfam Waste sorting (men); collection of recycling material (women) Note: *Italics denote that we met some participants of a CfW programme active in the area by chance but did not visit their workplace or talk to implementers in the villages. Source: Authors, based on information provided by local experts during field research; population numbers from Department of Statistics Jordan (2016) Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 80 In terms of sampling, we aspired to speak with an approximately even numbers of Jordanians and Syrians, CfW participants and non-participants, men and women at the different field sites. Due to practical constraints, this was not always possible. For instance, some assumed non-participants turned out to be participants of earlier CfW programmes; or some groups – for instance, female non-participants – were more difficult to access; respectively, shopkeepers in rural Jordan are mostly male. Table 8 gives an overview of the gender ratio within our sample for the different interviewee groups. As most Syrians reside in Jordan’s northern region and as, thus, many projects also take place there, we focused on the north. Figure 7 compares the number of interviewees to the number of CfW programmes, total inhabitants, and inhabitants of Syrian origin by region (north, central and south). This demonstrates that our sampling is proportionate to both the number of CfW programmes per region as well as to the population of Syrian origin. Overall – for logistical reasons – accessing the north was easier for us and we also followed the recommendation of a donor agency to not conduct fieldwork in one specific southern site for security reasons. For analysing social cohesion, it might have been interesting to study a higher number of southern communities in depth. 5.2.3 Structure of interviews Interviews with experts on the national level We conducted 54 interviews with “general experts”, that is, people with a broader understanding of CfW and development at a national level such as representatives of biand multilateral donor agencies, academics with a variety of research backgrounds and affiliations (universities, ministries, private research institutions), as well as representatives of non-governmental organisations (NGOs). Most interviews took place in February 2019, during the first phase of our field work; some during a preparation phase in June 2018. Where necessary, additional contacts were obtained through snowballsampling. For this group of interviewees, most questions were formulated in a broad and open manner and the Interview Guidelines (see Appendix B4) was adapted accordingly, so that interviewees could unfold their personal ideas and focus on their respective field of expertise. This approach also helped in order to broach sensitive topics and was combined with closed questions to Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 81 stay focused on relevant research topics and to probe previously expressed content. Interviews with “general experts” were particularly informative in three ways: (i) they helped us to assess the relevance of our study and the interest it might find with different audiences (that is, the Jordanian and international scientific communities; Jordanian civil society actors; and the donor community); (ii) they pointed out how sensitive the topics of social cohesion, gender and LED were in the context of CfW; and (iii) they gave us practical advice on how to conduct interviews with refugees and local community members. Interviews with local experts We interviewed 34 local experts, such as municipality members, local authorities, NGO members and representatives of agencies implementing CfW programmes. The interviews were based on the guideline for expert interviews (see Appendix B3) and, since social cohesion, gender and LED were sensitive topics, we adjusted or skipped questions whenever an interviewee was not ready to talk about a particular issue in an open manner. By using visualisation tools like cards, we attempted to increase interviewees’ readiness to share their opinion about sensitive topics. Interviews with local experts were best suited to gathering information about the communities in question, for instance, community life in general, particular challenges and opportunities for social cohesion, gender equality and LED in the respective community as well as participatory processes involving community members in the project design of a specific CfW programme. Interviews with community members We conducted interviews with 77 CfW participants, 73 non-participants and 97 shopkeepers. In total, we interviewed 247 community members. At the community level, 75 per cent of our interviewees were Jordanians and 25 per cent were Syrian. Two-thirds of our interviewees were male and one-third was female. Furthermore, we tried to interview participants of CfW programmes designed by different implementing agencies (see Appendix A4). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 82 Since the experiences of Syrian and Jordanian CfW participants and nonparticipants differed, we adapted the Interview Guidelines (see Appendix B1 and B2) accordingly. As we got into contact with most of the CfW participants via the implementing agencies, it was important to conduct the interviews in the absence of the responsible field manager so as to create an environment in which the participants were able to speak freely. In order to obtain better – that is, less biased and more spontaneous – responses, we proceeded in two steps: In a first step we posed questions in a broad and open manner, so that interviewees could unfold their personal opinions and perceptions on social cohesion, gender roles and LED in their respective community. Whenever interesting topics and hints came up, we posed an additional follow-up question. Only in a second step were direct questions on the CfW programmes and their possible effects asked. Since CfW participants who had been approached through an implementing agency knew their CfW participation was the reason why they had been asked for an interview, they more often brought up the effects of the programmes themselves. The Interview Guideline for the CfW participants and non-participants was designed to obtain detailed impressions on views and opinions held about the respective other group as well as on the effects of their CfW activities (see Appendix B1). The Interview Guideline for shopkeepers aimed at obtaining detailed information about the influence of CfW on local business activities (see Appendix B2). 5.2.4 Focus group interviews and participant observations At the outset of the field research, we organised two focus group sessions in order to gain additional background information, especially about the topic of social cohesion. The first focus group consisted of six Syrians: three men and three women. All had come to Jordan between 2011 and 2014. In contrast, the second focus group included only non-Syrian refugees who had arrived in Amman between 2012 and 2016: three from Sudan, and one each from Iraq, Ghana and Somalia. Focused and structured discussions with both the all-Syrian and non-Syrian focus groups gave interesting insights about how these groups interacted and how they saw their relationship to one another. We also witnessed two consultation events in East Amman. The aim of this participant observation was to get an impression about the procedure and rationale of participatory consultation in the context of CfW. One was held Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 83 by the Greater Amman Municipality with GIZ support in a quite central quarter of East Amman at the foot of Jabal an-Nasr. This consisted of a very broad and very steep footway connecting two major roads and leading to a number of houses but closed to cars, which could go only part of the way. The organisers had invited all residents to one multi-family house, where all the men gathered in the apartment of one family, while the women met in another apartment. Both groups were asked to discuss how the footway in front of their houses could be improved. Interestingly – but perhaps not surprisingly – they came up with quite distinct results. The ladies sympathised with GIZ’s idea to convert part of the dusty way into a little park with trees providing shadow while at least one group of men suggested making it better accessible for cars. The other consultation event took place at a school in Marka, the most eastern part of Amman, where most of the capital’s manufacturing is located and most people are workers and quite poor. The school had been very nicely renovated and equipped with basic but pleasant furniture by a CfW project implemented by Najmah with funding from the WFP. Parents were invited to inaugurate the school and discuss what they could do to preserve the upgraded school building. It turned out that all stakeholders – parents, teachers and students – were grateful for the support and, particularly, students’ appreciation had become higher, as they either knew the workers personally or had seen them there during the working hours. 5.2.5 The GIZ Post-employment Survey The first phase of the GIZ Post-employment Survey, conducted between January and November 2019, covered 984 people who had completed their employment in GIZ’s Green Infrastructure CfW projects during this time. Interviews were conducted throughout Jordan, roughly a quarter of the interviewees were women (253 interviewees) and almost half were Syrians (471 interviewees).17 17 Most interviewees (43 per cent) were from southern governorates (132 from Al-‘Aqaba, 192 from At-Tawfīla and 103 from Al-Karak); many (39 per cent) from northern governorates (50 from Irbid, 221 from ‘Ajlūn, 113 from Al-Mafraq) and the remaining 17 per cent from central governorates of Greater Amman (51) and Al-Balqāʾ (115). 596 interviewees (61 per cent) were married, 344 (35 per cent) single, 26 (3 per cent) divorced, 13 (1 per cent) widowed, and 5 (0.5 per cent) separated. 60 (6 per cent) of them held a university degree, 19 (2 per cent) had a vocational training certificate, 185 (19 per cent) had completed secondary education, 450 (46 per cent) had completed only primary education, and 227 (23 per cent) had not even completed primary education (GIZ, 2019). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 84 The questionnaire was comparatively short, easily answered within 15 minutes. It included questions on (i) the site of employment (municipality and partner organisation of GIZ); (ii) individual characteristics (nationality, gender, age, marital status, number of dependent family members, education, vocational qualification, and previous employment); (iii) the use of income received from the CfW programme; (iv) satisfaction with the programme (working conditions, payment, employer, workplace, safety at work, treatment by supervisors, working hours, work equipment, feedback mechanisms); (v) lessons learnt on the job; and (vi) future plans (next employment, next schooling or training measure, plans to return to Syria or migrate elsewhere). In the context of our research, the GIZ Post-employment Survey was particularly important for testing Hypothesis 15. 6 Findings: community effects of Jordan’s CfW programmes We now turn to the outcomes of our research. In this section, we discuss first the effect of CfW programmes on social cohesion (subsection 6.1), gender roles (subsection 6.2), and LED (subsection 6.3). After doing that, we report on the various opinions voiced by our interviewees on the way the CfW programmes were designed (subsection 6.4). Within subsections 6.1 and 6.3, we also present findings on two additional effects that we had not included in our research design but found to be too important during our field work to be neglected. Both effects were to be observed within CfW programmes active in the waste sector: first, CfW had a positive effect on community members’ environmental awareness, which indicated a higher willingness to cooperate for the common good, thus strengthening social cohesion (see subsection 6.1.4). Second, CfW activities weakened the so-called shame-culture (that is, the reluctance of people to work in specific sectors considered inferior) with positive effects on LED’s employment effects (see subsection 6.3.4).18 As already explained in 18 Since these two additional effects were researched in an inductive way – in contrast to the otherwise deductive research design with pre-determined dependent variables (subsection 5.1) – the information collected did not provide conclusive evidence. However, we highlight them in order to encourage future, more intense research on these effects. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 85 subsection 5.2.1, we do not mention the names of our interviewees because we have promised them confidentiality about the contents of our talks. The source of direct quotes is identified by the interview number, the nature of the interviewee (for example, CfW participant, shopkeeper, other nonparticipant, local expert) and the location where the interview took place (see Appendix A4). Where relevant, we also specify the gender and nationality of the interviewee. 6.1 Social cohesion Our research results provide evidence that, overall, CfW programmes in Jordan have a moderately positive effect on social cohesion at the community level. In particular, they contribute to the sense of belonging of Syrians to their respective host communities and to the mutual horizontal trust between Jordanians and Syrians. This effect is, however, mainly due to the fact that Jordanians and Jordanians work together on the same activities (Hypotheses 2 and 5) and much less so to the existence of the CfW programmes as such (Hypotheses 1 and 4) or the creation of helpful and enjoyable public goods (Hypothesis 3). Probably, the main reason for the effect being only moderate is that the relations between Syrians and Jordanians are traditionally intense with the effect that their mutual trust was already at quite a high level even before the CfW programmes were launched in Jordan. At the same time, the effect of CfW programmes on vertical trust is much more ambiguous (Hypotheses 6 and 7). Unexpectedly, we also found positive effects on community members’ cooperation for the common good, when it came to the area of the environment (no predefined hypothesis, see Figure 4). In the following, we discuss separately the effects of CfW programmes on the sense of belonging of Jordanians and Syrians to their respective communities (6.1.1); horizontal trust between both groups (6.1.2); and vertical trust in local authorities (6.1.3); along with some observations about cooperation for the common good (6.1.4). 6.1.1 Sense of belonging (Hypotheses 1-3) According to our findings, CfW programmes in Jordan clearly strengthen the sense of belonging of Syrian participants to their host community – but only to a certain degree the sense of belonging of Jordanian participants Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 86 or of people from either group not participating in the programmes. This is presumably due to the fact that Syrians have a quite strong sense of belonging to their host communities anyhow. Sense of belonging irrespective of CfW programmes When asked whether they felt well being part of their respective local community, the vast majority of interviewees stated clearly that they felt comparatively well integrated – both Jordanians and Syrians. Unfortunately, we could not ask all interviewees and some did not answer our question, but a total of 84 people (that is, only a third of all interviewees) responded to this question, including 53 Syrians, 30 Jordanians and 1 Egyptian, respectively 47 males and 37 females. 66 of these respondents (79 per cent) said that their sense of belonging to the local community had always been quite good; 10 interviewees (12 per cent) reported that they had not been well integrated in the past at all but half of them felt much better integrated now. Of course, some Syrians also mentioned that they still did not feel at home as they had in their Syrian home places before their flight: “Yes, I belong here. But there is nothing like home” (22, non-participant, Ḥawfā); “Obviously, people are very friendly here. People are very welcoming. We visit each other on our special occasions. It’s very natural for us to become one. But our heart is in Syria” (73, non-participant, Al-Azraq); “Not 100 per cent, but it is normal, maybe 90 per cent” (46, non-participant, Kafr Assad). One CfW participant in Kafr Assad explicitly made a lack of interaction responsible for her low sense of belonging: “I have no contacts to other people in the village because everybody is at work at day and sleeps at night. Therefore, I do not feel to be part of it. And all my relatives live in other villages” (60, participant, Kafr Asad). Some female interviewees stressed that life in Jordan was not always easy for their children. A Syrian CfW participant said, for example: “My girl had problems in the public school” (149, participant, Kafr Ṣawm) and added that her daughter had to change to a private school because she was discriminated against because she was not a Muslim; she was hit by a teacher and other pupils asked her why her mother was not veiled. Another non-participant mentioned: “My children sometimes get abused because we are Syrians. […] In the schools, or when they go out of them. My older son has got a broken arm because he got into a fight” (135, non-participant, Kafr Ṣawm). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 87 Male interviewees, by contrast, never mentioned problems that other family members had faced. Instead, they highlighted how much they appreciated the high degree of security of life in Jordan, which had helped them to settle in the host country and to feel more at ease. Virtually all interviewees recounted that they had tried to migrate to communities where they knew someone, preferably relatives and preferably Syrians – but also where they knew at least somebody from the community. Likewise, many Syrians mentioned that they felt connected to their place of residence in Jordan because it reminded them of the environment of their previous Syrian home. Most Jordanians stated that they had a very strong sense of belonging to their respective home communities, anyhow. Some also highlighted that many people from their communities did their best to integrate everybody – Jordanians and Syrians: “There are people going to the weddings, funerals, graduation parties, different occasions, so in the beginning they invite the new people that come to the community” (117, non-participants, Umm al-Jimāl); “There is no discrimination [between Syrians and Jordanians]” (224, shopkeeper, Al-Qaṣr/ Faqū’a); “We don’t think of ourselves as Syrians or Jordanians here, we are one and we all face the same challenges” (133, non-participant, Kafr Ṣawm). Effects of the existence of CfW programmes as such (Hypothesis 1) At the same time, a considerable number of interviewees commented on the CfW programmes’ effect on their feeling of belonging, with most pointing towards a positive effect. 30 out of 80 respondents giving any concrete answer to this specific question said that the CfW programmes had eased their integration into local communities: 5 because they had been not so well integrated in the past and 25 even though they had always been quite well integrated. Ten respondents declared explicitly (and another 31 more implicitly) that the CfW programmes had not had any tangible effect on their feeling of belonging because it had always been good. Only 9 said that their feeling of belonging was bad or at least not so good or that it was still the same (Syrian women and men) and no respondent mentioned any negative effect of CfW programmes on their sense of belonging. Almost all of those who confirmed a positive effect were CfW workers. This finding supports the assumption that, in a context like the Jordanian where the sense of belonging of locals and immigrants is already quite strong, only Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 94 In the same way, a Jordanian man argued: We are all humans and all have needs, but the refugees took opportunities and replaced Jordanian labour. The Syrians can live here and have their life, but there are more disadvantages than advantages. Everyone lives their life here, but they took job opportunities from Jordanians. (65, non-participant, Tal al-Rummān) Clearly some Jordanians associate the Syrian crisis with an increase in prices, especially rents, as well as increased competition over jobs. For instance, in a group discussion with CfW participants living in a Palestinian refugee camp, people testified that Syrians work for lower wages, caused an increase in rent from JOD 50 to 120, and a priority shift of international aid from poor Palestinians to Syrians (182, participants, highway). Finally, many Jordanians also stated that the Syrians were better off because they were receiving aid and support from international organisations (77, participant, Al-Azraq; 145, shopkeeper, Kafr Ṣawm; 145, shopkeeper, Kafr Ṣawm; 104, non-participant, Umm al-Jimāl; 276, non-participants, Kafr Ṣawm). “Syrians can work for JOD 4-5 per day. I can’t” (107, nonparticipant, Umm al-Jimāl). These accusations are a risk for Jordanian-Syrian relations and how they may evolve in the future. So far, most Jordanians differentiate between the competition for jobs and accommodation on the one hand, and their personal relations with individual Syrian neighbours, on the other hand: The Syrians get money without any effort. They pay rent without bargaining so the rent is going up. They would pay twice the rent if they could get the house. Now, house owners ask for JOD 150 and there are no more empty houses left. I mean, as a house owner, of course I would give the house to a Syrian. The house owners are clearly the biggest beneficiaries from the Syrian crisis. This situation can also lead to tensions – as a house owner, if I prefer to rent out my place to a Syrian instead of a close relative, this creates tensions. So, are there tensions? No, we are one family. We can separate between work and life. Work is one thing, but our relations in general are very good. (209, shopkeeper, Faqū’a) But this may change. Should the economic situation deteriorate further, the horizontal trust between Jordanians and Syrians could weaken on the personal level as well. Again, our findings are in line with those of the NAMA study mentioned above. It found that almost two-thirds of 1,305 Jordanians interviewed Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 95 believed that the presence of Syrian refugees has a negative effect on their life. In particular, 46 per cent stated that the Syrians’ presence negatively affected the security situation in Jordan, 53 per cent education, 64 per cent healthcare, 67 per cent the water supply, 68 per cent the government budget, 87 per cent the labour market, and 81 per cent the economy as a whole. Even among the Syrians interviewed, 9 per cent admitted that the presence of a large number of compatriots had a negative effect on security in Jordan, 21 per cent on education, 33 per cent on healthcare, 38 per cent on water supply, 23 per cent on the government budget, 40 per cent on the labour market, and 37 per cent on the overall economy. Likewise, 87 per cent of the Jordanians and 73 per cent of the Syrians blamed the immigration of Syrians for causing prices in Jordan to rise, while 92 per cent of Jordanians and 56 per cent of Syrians stated that the immigration has contributed to an increase in unemployment rates (NAMA & KAS, 2018). Several interviews revealed that negative feelings between Syrians and Jordanians prevailed mainly where the two groups did not have any contact. A Syrian woman said that her relationship with the Jordanians was quite bad at the beginning because she did not know any at the local level: “In the beginning, we did not know anyone; no one would come over to the Syrians, now there is interaction” (114, participant, Umm al-Jimāl; similarly 177, Syrian non-participant, Al-Mafraq). Likewise, several Syrian interviewees said that all of their good friends were Syrians and that their interaction with Jordanians was limited to work. Again, the economic situation strongly inhibits forging new relationships and strengthening existing ones: “The social relations have deteriorated because […] they do not have the time anymore to establish relations, because they have to survive the current economic situation” (77, participant, Al-Azraq). Unfortunately, we were not able to collect comprehensive evidence on the situation of other migrant groups such as Iraqis or Egyptians. The reason was that we mainly went to small villages with CfW activities while most Egyptians and Iraqis had settled in larger towns. Nonetheless, we were still able to make some observations and, based on these, it would seem that Egyptians in particular are not well integrated and have only a weak sense of belonging. Only one of our few Egyptian interviewees knew the CfW programmes in Jordan. He stated that he was not jealous because he was not eligible to participate, but was critical about the degree to which these programmes were in fact able to select the most vulnerable Syrians and Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 96 Jordanians: “We are not bothered that we don’t get these opportunities. But we’re confused about the selection criteria” (72, non-participant, Al-Azraq). Effects of the existence of CfW programmes as such (Hypothesis 4) According to our research, the existence of CfW programmes has limited effects on the horizontal trust between Jordanian and Syrian community members. 26 Jordanians and 34 Syrians said that their trust in the respective other group had clearly improved over time but only a quarter of these 60 people were not participating in CfW programmes. The main channel through which CfW impacts social cohesion is therefore, once again, personal participation in CfW programmes. And most of the non-participants who also reported improvements in horizontal trust because of the CfW programmes did so because relatives, friends and neighbours had told them about their positive experience. Evidence for Hypothesis 4 is thus quite weak. Nevertheless, non-participants highly welcome the programmes. Several interviewees highlighted that these programmes were good for both Syrians and Jordanians and that they liked the idea that the two groups worked together. However, only two interviewees – both from Kafr Asad – explicitly confirmed our hypothesis that the existence of CfW as such already has positive effects on horizontal trust even between non-participants: The programmes had a lot of impact on the social relationships between Syrians and Jordanians. First of all, when they came, there were not many connections between the groups. Now, it got better due to the programme. Before, everyone was a bit ignorant of each other. Now the truth was revealed that the Syrians can also work for the community. Both Syrians and Jordanians have the same objective. (230, non-participant, Kafr Asad) As a result [of the CfW programmes], there are a lot of good relationships between people. There was a street-cleaning project. The public would come out and give them tea. So even the people who did not participate were still happy. (229, non-participant, Kafr Asad) Some interviewees, in contrast, were quite critical about the CfW programmes. Some complained that Jordanian participants in general were not selected on the grounds of their socio-economic needs but their wasţa (Arabic for “connections” or “favouritism”; see more on this issue below in subsection 6.4.3). And some disliked the fact that the programmes were employing an equal share of Syrians and Jordanians even though much fewer Syrians lived in their respective communities than Jordanians. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 97 But none of them blamed the Syrians. For instance, when asked “What are the Jordanians thinking about the fact that the Syrians can also work in the CfW programmes?”, a shopkeeper in Al-Azraq summarised the situation as follows: Some voices, of course, say that the jobs should be for Jordanians first. They feel that it is unjust that the Syrians get more money than themselves because the Syrians get also support from international organisations. This means that the Syrians can afford to work at wages far below the levels at which Jordanians would work. This creates quite a bit of frustration in town because there are also very many Jordanian families in dire need. And the Syrians are getting much more help than even the poorest Jordanians get. So, many people in town say that all the support programmes are unfair. (91, shopkeeper, Al-Azraq) A CfW participant from Kafr Ṣawm answered our question “Are nonparticipants jealous of your participation in the GIZ programme?” as follows: “Yes, of course many people wanted the work, Jordanians as well as Syrians, and are jealous of the good chances. Many people compete because they are in need” (149, participant, Kafr Ṣawm). Our findings are backed by the DEval evaluation conducted in 20182019 (Roxin et al., 2020). It also found that Jordanians and Syrians had a comparatively high trust in each other in any case (while the mutual trust of Syrians in Turkey and local Turks was much weaker). Nevertheless, the horizontal trust of the two groups in Jordan increased during the operation of the CfW programmes and, while the initial horizontal trust was stronger among participants than among non-participants, the trust of non-participants increased at least as much as the trust felt by participants (Roxin et al., 2020). This finding would mean that the sheer existence of CfW programmes employing both refugees and locals can already have a positive impact on horizontal trust. For Turkey, however, Roxin et al. (2020) found that horizontal trust had decreased over time among non-participants and had only slightly increased among CfW participants. Effects of participation in CfW programmes (Hypothesis 5) That said however, CfW participants confirmed across the board that CfW programmes tended to have a positive effect on horizontal trust because they brought people from different social groups together and made them work for the same goals. Three-quarters of all CfW participants interviewed stated that the horizontal trust between Syrians and Jordanians was strong Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 98 while only 8 per cent said that it is rather weak. Almost half said that CfW programmes had further strengthened horizontal trust. A fifth said that CfW programmes had not strengthened horizontal trust but with two single exceptions they argued that this was only the case because the horizontal trust between Syrians and Jordanians had always been so strong that it could not be further strengthened. Our evidence for Hypothesis 5 is thus quite strong – in particular, if we consider that many interviewees made their statements spontaneously, that is, even before we had asked a question on the issue. Participants mentioned various different reasons for this positive effect of CfW programmes: conversations during work to learn about each other’s interests and values; the need to collaborate to succeed in the work objectives; shared meals; or leisure activities after work (such as football, invitations to celebrations, and so on). In addition, the relations improved when participants exchanged skills. “People start to exchange experience. For example, Syrians show locals how to plant” (194, participant, Faqū’a). In a cooking project in Kafr Ṣawm, participants exchanged recipes and cooking styles. CfW is sometimes cited as the only means that brings Syrians and Jordanians actively together. “The project is the first way to interact with Syrians. Through the job I felt that I have the chance to choose to interact with Syrians” (27, participant, Kafr Asad). Several times participants pointed out that, prior to the programmes, they had not enjoyed strong interactions with Syrians, despite the fact that they shared many values, customs and traditions. The joint activity made them connect and led to increased daily-life interaction, also beyond the programme. “I worked with Syrians and we built friendships. We were like brothers. When one of us was sick, we took care of him” (271, Jordanian participant, Deyr ‘Allā). In many instances, the relationships seem to continue. One participant from Deyr ‘Allā said that “the relationships between Syrians and Jordanians are very good. We are like brothers. Even when the programme finished” (2, participant, Deyr ‘Allā). Another person recounted: “We are still in contact via WhatsApp. At the end of the project, we made a small celebration, and everybody contributed” (128, participant, Kafr Ṣawm). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 99 These findings are again in line with the results of the DEval study mentioned above. It found that Syrians and Jordanians get to know each other at their workplaces much better than anywhere else (Roxin et al., 2020). Similarly, the results of a workers survey conducted among the participants of the ILO’s employment-intensive infrastructure programme were very clear on this point: 93 per cent of its respondents stated that Jordanians and Syrians were able to work together as one team; just 8 per cent of the Jordanians and 4 per cent of the Syrians disagreed. 90 per cent of its respondents said that Jordanians and Syrians trusted each other, and only 11 per cent of the Jordanians and 4 per cent of the Syrians disagreed. 91 per cent confirmed that they had built a friendship with other workers including people of the respective other nationality, while only 8 per cent of the Jordanians and 7 per cent of the Syrians disagreed. Along with this, 83 per cent of all workers felt that participation in the CfW programme had contributed to a reduction in the tensions between Jordanians and Syrians in Jordan, while 15 per cent of the Jordanians and 13 per cent of the Syrians disagreed (NAMA & ILO, 2019). Finding new friends of the other nationality is a commonly cited outcome by the programme participants, as the GIZ Post-employment Survey (GIZ, 2019) demonstrates: 86 per cent of all respondents stated that they had made new friendships with people from the other nationality group. More Syrians than Jordanians (94 versus 78 per cent) and more women than men (94 versus 83 per cent) felt they had made friends of the other nationality through their CfW participation; also participants in urban project sites were more probable to have built up a friendship (see Appendix E1).19 Our results confirm that CfW programmes connect women to men and fellow women. “I became close friends with my female co-workers and they became like sisters to me. This is due to the site engineer who is making us all feel the same” (25, female participant, Kafr Asad). Further, our results do not significantly vary between project sites. Interestingly, several participants in the ILO Highway project pointed out that their joint activity helped them to learn from each other and eased preexisting tensions related to job competition. 19 However, only the finding on urban population was – at a 99.9 confidence level – statistically significant. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 100 Another finding was that mixed nationality teams strengthened the effects. A number of times, interviewees pointed to the benefits of working in mixed groups. Interestingly, one member of a CfW-team mentioned that this important decision was to be made by the respective supervisor: It depends on the supervisor of the team; some supervisors mix the groups and other supervisors separate the groups, which also manifests hate between the groups. […] It is better to work in a mixed group; it is good to integrate. (53, participant, Umm al-Jimāl) A participant pointed out that separating teams due to tensions exacerbated those tensions further and thus clearly favoured a mixed approach: I love that I work in mixed groups of Syrians and Jordanians. In cities, there are many mixed groups. But not so in the village. Many other workers were sceptical in the beginning about cooperating with Syrians. But the project changed their mind. All of us increased our contacts with the Syrians, and we all became good friends. (59, Jordanian participant, Kafr Asad) Yet, there are also individual critical voices on mixed teams. According to experts, Syrians especially perceived that they had to carry out harder tasks than their Jordanian co-workers (306, CfW Coordination Group meeting). Such unequal treatment within mixed teams can weaken horizontal trust. Getting to know each other is something rare in the context of the economic hardship of many interviewees, who said that it has become difficult to accept invitations for weddings, funerals, and so on as the cost for gifts was exceeding their household’s budgets. When asked “What are other works that should be done in the community?”, one interviewee answered: “There should be more work that supports both Jordanians and Syrians, we should work with them together so we can better understand their situation. Sometimes you feel that Jordanians are like us, they don’t have income either” (114, participant, Umm al-Jimāl). A non-participant from Kafr Ṣawm added: “We don’t think of ourselves as Syrians or Jordanians here; we are one and we all face the same challenges” (133, non-participant, Kafr Ṣawm). 6.1.3 Vertical trust (Hypotheses 6-7) Our findings suggest that CfW programmes can have a positive effect on vertical trust even though they are limited with regard to this aspect of social cohesion. We had difficulty in getting meaningful answers on our questions related to vertical trust for two reasons: First, direct questions Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 101 on this topic triggered a biased response: we had the impression that many interviewees said what they thought we wanted to hear. For that reason, after the first two weeks of research, we stopped asking direct questions. Second, we found that many interviewees indeed knew well which actors (for example, foreign donors, local NGOs) were effectively responsible for the CfW programmes. It was therefore difficult to say whether their opinions about the local authorities had really been influenced by their experiences with CfW programmes. We therefore started asking at least some of our interviewees who, in their perception, was responsible for the programmes (central government, municipality, NGOs or foreign donors) in order to better pin down whom they would see responsible for the good and the bad sides of the programmes. The answers of this small group of people gave a quite mixed picture. Three respondents (all men) thought that their respective municipality had set up the CfW programmes. Eight insisted that a foreign donor (GIZ, ILO or WFP) was responsible for it (including an equal number of women and men and of Jordanians and Syrians). Two said that they had no idea. And even the municipalities considered many CfW projects as GIZ or ILO projects rather than their own ones (5, local expert, Deyr ‘Allā; 228, local expert Kafr Asad). We also asked some interviewees whether the CfW programmes had strengthened their vertical trust in the local authorities and here 20 gave us clear answers: Exactly half of them confirmed this while the other half negated the question. Some even said explicitly that CfW programmes had raised their trust only in international donors. Where local authorities were actively involved in the project design and appeared to be open to the wishes of community members (for instance, when they organised participatory events; see subsection 6.4.4) CfW programmes had a positive effect on the vertical trust in local authorities: The workers believe that the community centre is responsible for the programme [… and] they attribute the programme and its benefits to the Ministry of Agriculture. As a result, the workers are mainly grateful to the government, and of course they feel better connected to it because they now have an open door to it through the programme. (8, local expert, Deyr ‘Allā) Similarly, CfW participants in Kafr Asad saw the municipality as being mainly responsible, indicating that trust in it increased with the creation of CfW job opportunities (17 and 43, participants, as well as 46, nonparticipant) and the local procurement of building materials (27, participant). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 102 A Jordanian woman said: “I feel that I can trust the local authorities more. They do what we asked for immediately, concerning the materials that were needed” (27, participant, Kafr Asad). Reversely, if community members attributed the CfW programmes to foreign agencies or implementers from outside their community, vertical trust was reinforced towards those actors, rather than towards Jordanian authorities: “The people here in Deyr ‘Allā like GIZ very much. There are GIZ stickers on every fridge” (5, local expert, Deyr ‘Allā). I hope that we work [more] together with the supporting organisations, to solve those problems together. It is better to talk to people directly in the communities than talking only to the centre or the government. We trust you [the foreign donors] more than we trust the government. (31, local expert, Kafr Asad) Another factor that undermines vertical trust is the fact that wasţa (connections) often has an effect on decisions taken by the authorities – such as the selection of Jordanian participants in the CfW programmes or the choice of shops where building materials and machines required for the CfW activities are procured. Many community members, Jordanians and Syrians, men and women, disapproved of the fact that local decision-makers employed CfW participants on the basis of their wasţa (connections): “If you don’t know anyone, you can’t find job opportunities. I feel that the municipalities would only announce the [CfW] opportunities, if they already had registered the people” (94, female participant, Umm al-Jimāl); “I am more angry at the municipality. There is wasţa everywhere” (231, male nonparticipant, Kafr Asad). Likewise, interviewees complained that decisions on local procurement were often driven by wasţa (connections): “We need to ensure that all standards are implemented before handing it [that is, the created infrastructure] over [to the municipality]. Otherwise, they will just employ the usual suspects – their cousins etcetera – and do whatever they like” (111, local expert, Umm al-Jimāl). These considerations show that, in all settings, CfW programmes are creating new structures that alter and run the risk of damaging carefully built relations within municipalities. In that vein, if the handing-over of created services and infrastructure is not well planned in advance, the temporary parallel structures of CfW programmes may even weaken local employment and social services (291, GIZ). The projects become part of local politics. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 103 6.1.4 Cooperation for the common good through environmental awareness (no pre-determined hypothesis) Though we had not formulated a hypothesis concerning the possible effects on community members’ cooperation for the common good, we found that CfW programmes in Jordan had an effect on environmental awareness and, thus, fostered a readiness to engage for the common good. This is especially the case for projects in the area of waste. Nine interviewees stressed that they appreciated that the streets in their community were clearer because the CfW projects were leading by example: “Other than myself, others became aware of the environment. They started using the trashcans” (20, participant, Kafr Asad). Many interviewees said that an increasing number of people had become aware of the necessity of recycling waste (for instance, 101, participant, Umm al-Jimāl). “Many of the young men were actually affected, for example you would find someone in the group throwing a cigarette on the floor and then people would tell him to pick it up and throw it in the bin” (54, participant, Kafr Asad). Apparently, this holds true in particular for children: “You can see people working on the streets, children in school observe people cleaning and separating waste and thus also act on it and realise that putting waste in the environment is bad” (102, local expert, Umm al-Jimāl; similar: 70, participant, Al-Azraq). Both the environmental effect and the decrease of shame culture on the labour market (subsection 6.3.4) are related: “Now, the environment is considered, and it is now okay to do cleaning” (28, participant, Kafr Asad). 6.2 Gender roles (especially Hypotheses 1-2 and 15) All CfW programmes in Jordan aim at providing employment opportunities for both women and men. As a result, these programmes have effects on gender roles within communities. Yet, if female labour force participation rates have increased somewhat recently in parts of Jordan, the main reason has been the protracted economic crisis which forces many families to look into new options. CfW programmes play only a minor role because many Jordanians and Syrians still have reservations against women in paid employment. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 110 programme in a given location determined to what degree it could affect gender roles. In our sample of interviewees, the vast majority of women and men considered female labour force participation as acceptable or even positive. Only 3 out of 23 male respondents and 1 out of 20 females said that women should not have paid employment. However, an additional 7 men and 2 women said that women should only work under specific circumstances: for example, if the kind of work was adequate, if the sector of activities was acceptable for women and if the work environment was safe (see also subsections 6.2.1 and 6.2.2). On the other hand, 6 men but only 1 woman insisted that CfW programmes should employ more women. We also found evidence that the acceptance of female labour force participation depended somewhat on the age of a person. Younger people, both women and men, appeared to us more open to women in paid employment than older people. As one woman stated: “Especially old people have difficulties in accepting women in jobs like mine. But all the young people have no problems with this anymore” (93, participant, Al-Azraq). The attitudes of CfW participants and non-participants did not differ substantially, neither did those of Jordanians and Syrians. But both groups stated that Syrians were more conservative, that is, less likely to let a woman work outside the house. As a Jordanian woman put it: It is very normal among Jordanians that girls work outside the house. Syrians, in contrast, think that women cannot work for pay and that they should stay at home. […] When we sit together, I try to convince the other girls that it is normal for us to work. […] I try to convince them that they have to be more educated and self-reliant. (58, participant, Kafr Asad) Many Syrian women confirmed that, in general, women in Syria were only active in certain sectors of the labour market, such as education. But they also stated that it was more common for Syrian women to participate in the labour market in Jordan than used to be the case back in Syria, mainly because of the difficult economic situation of Syrian refugees in Jordan. Two Syrian women told us: “Women work here more than they did in Aleppo. […] I have told him [my husband] that you have to let me help you and work because if I do not help, we will probably [financially] break down” (136, non-participant, Kafr Ṣawm); “This [increase in labour force participation among Syrian women] is also because life, paying rents, etcetera, is Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 111 very expensive in Jordan and it used to be much cheaper in Syria” (149, participant, Kafr Ṣawm). In any case, financial hardship seems to be the main reason for women joining the labour market. Many interviewees acknowledged that it was still “culturally inappropriate” (112, female non-participant, Umm al-Jimāl) for a family to allow a married woman to work for pay and only the last step it should take if it faced a financial crisis. A Syrian woman declared: “Women should not be working outside their homes. It is the last option, really – only allowed if you do not have sons. Our neighbour is in this situation and therefore this is not a shame for him” (112, non-participant, Umm al-Jimāl). Likewise, several interviewees pointed out that the protracted economic crisis has forced many families to break with the traditional model of the single male breadwinner family. They decided that, in order to sustain a family, both men and women had to earn an income. A Jordanian man stressed: “I think the economic situation needs all members of a family to work. Also, the females have to contribute to the family income” (21, nonparticipant, Kafr Asad). This change is not irreversible. Once the economic situation improves, some Jordanian and Syrian women might reconsider their strategy and give up their jobs again. Some interviewees emphasised strongly that they were only working because of the difficult economic circumstances and would rather stay at home and fulfil their role as housewives: “I do not like it because it is forced by the economic situation. I would prefer to stay home and care for my children” (30, female non-participant, Kafr Asad). Longerterm acceptance of female labour force participation can only be achieved if coercion is not the only driving factor. However, some women stated that they had joined CfW programmes also because they enjoyed the work and the interaction with others at the workplace. This once again supports the notion that participation in CfW programmes can especially strengthen female participants’ sense of belonging (Hypothesis 2) and horizontal trust between one another (Hypothesis 5). For many women, CfW programmes are particularly attractive because the joint activity allows them to talk with other members of the community with whom they have little contact in everyday life. A Jordanian woman told us: “I want to increase my communication with the people in the community” (123, participant, Kafr Ṣawm). Similarly, a Syrian woman said: “The programmes have many benefits. They are entertaining. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 112 And they are bringing people together to do more productive things” (279, non-participant, Kafr Ṣawm). Finally, several women called attention to the fact that their participation had also strengthened their self-confidence and thereby their standing within the community. A woman from Kafr Asad said: “It [working in a CfW programme] helped to shape my personality; I became more confident. I can now be an active member of the society” (27, participant, Kafr Asad). 6.3 Local economic development (LED) CfW programmes in Jordan likewise contribute to LED. The direct effect, through the wages paid to CfW participants, is very clear and significant. In addition, the programmes also raise the wages of non-participants through the multiplier effect and local procurement, even if the size of this effect remains difficult to quantify, not least because of the difficult economic situation which is another strong – yet CfW-independent – impacting factor. The same holds true for effects on LED caused by the creation of public goods. However, the effect of CfW programmes on the long-term employment prospects of CfW participants seems to be negligible with the exception of mitigating effects on the so-called “shame culture” in regard to unattractive job opportunities. In the following, we discuss how interviewees in our sample perceived LED in their respective community at the time of the interview (6.3.1); whether the wages paid by CfW programmes had an effect on average per capita income in these communities (6.3.2); whether the public goods created by the programmes had an impact (6.3.3); and whether the programmes had long-term effects on participants through the upgrading of their soft and technical skills (6.3.4). 6.3.1 Perceptions of local economic development Most of our interviewees – regardless of their nationality or employment status – confirmed that the general economic situation in Jordan has been increasingly difficult. Both Syrians and Jordanians mentioned the lack of jobs and the prevalence of poverty as the main problems. In addition, many talked about decreasing wages due to higher competition on the labour market, rising rents, increasing food prices, large numbers of students in school classes, transportation problems, shortages in water and electricity, and long Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 113 waiting hours in health care centres. As mentioned above (subsection 6.1.2), Jordanians often accused the Syrians of these problems, for instance, the following Jordanian woman: There was a great impact [of the arrival of Syrians in the region]. Job opportunities are rare for Jordanians, and […] the Syrians accepted all jobs for a lower wage. Also, there was great impact on the electricity and water, and also in many other aspects. (65, shopkeeper, Tal al-Rummān) A local expert explicated: The schools are very crowded. Before, classrooms […] took 30 students. Now, the classrooms need to fit 45 students. […] Medical centres are prepared to treat 25 cases per day. Because of the […] Syrian refugees, they now have to accept 40-60 people per day. This is a lot of pressure on the health services. (3, local expert, Kafr Asad) Some Jordanians, however, acknowledged that the arrival of the Syrians has also had some positive economic effects on Jordan. Some shopkeepers in the south of Jordan highlighted, for example, that the presence of the Syrians has led to an increase in business activities (201, shopkeeper, Faqū’a; 202, shopkeeper, Faqū’a). Some interviewees stressed that many programmes of international donors would not exist without the presence of thousands of Syrians in Jordan. Likewise, some streets and schools had been built to host the growing population. Furthermore, Syrians had brought with them certain technical skills that were very helpful for the Jordanians as well (78, local expert, Azraq). 6.3.2 Direct and indirect effects of the wages paid by CfW programmes (Hypothesis 13) CfW programmes have raised the income of participant households by almost a quarter. Such households spent most of the additional income on basic needs and repaying debts and only invested a very small share of it. In addition, these programmes also raised the income of non-participant households through the multiplier effect and local procurement – but not the investment effect. However, we cannot say how high these indirect effects were. Below, we discuss (i) to what extent the wages paid to CfW participants increased their income and what the participants did with the extra income; and (ii) whether the broader community had also benefitted (indirectly) from the wages through the multiplier or the investment effect. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 114 Direct effects (income of CfW participants) The most direct economic effect of CfW programmes is that the wages paid to participants increase their total household income during the time of their employment. The study by Roxin et al. (2020) found that participant households had on average a monthly income of JOD 376 in 2017 when they applied for a job in any of the GIZ programmes and a total monthly income of about JOD 460 during their participation. The GIZ programmes would thus raise the household income by just about JOD 85 or 22 per cent on average in net terms. This result is similar to the findings of Jones et al. (2019) that Syrian households, which tend to be poorer than vulnerable Jordanian households, had an average income of JOD 285 per month in 2018 (excluding any CfW wage but including the unconditional cash grant provided by UNHCR and UNICEF; see subsection 4.3). The abject national poverty line of Jordan is currently JOD 28 per person and month, which is – depending on household size – between JOD 100 and 250 per household (ILO, 2019). For all CfW programmes in Jordan, Roxin et al. (2020) found that the participation of women increased the monthly income of households by JOD 77 on average while the participation of men increased the monthly income of households by only JOD 33. The effect is thus clearly positive but much smaller than the wage that CfW programmes typically pay to their workers (JOD 240). Roxin et al. (2020) assumed that this is mainly due to the fact that many CfW workers give up other jobs for their CfW participation and possibly also receive less support from other households. Other studies, however, – as Roxin et al. (2020) admit – have not found crowding-in effects. Whatever the exact amount, the rise in income of CfW participant households is in any case only temporary: with the end of their employment, their income decreases again to the previous level (or even below it) unless some household member finds another job. CfW participants spent the bulk of their wages on consumptive purposes. We asked 64 of 72 CfW participants about their spending patterns. Only a third mentioned investments in human capital (specifically the education of their children: 23 per cent) or small projects (11 per cent) as one of their top-spending priorities. In contrast, 47 per cent mentioned items related to housing (rent, electricity and water), 30 per cent food, 23 per cent household equipment, and 22 per cent debt repayment, and 20 per cent the support for children and other relatives. Other items were clothing, individual needs, health, transportation, weddings, and holidays (see Table 11). Some experts also told us that the CfW participants did not necessarily buy more, but sometimes higher-quality products. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 115 Interestingly, the spending patterns of women and men differed significantly. We have information on the spending patterns of 26 female and 38 male CfW participants and both groups spent in similar ways on household equipment, support for children or other relatives, and investments. But while men often spent more significant parts of their CfW wages on housing (rent, electricity and water), food and debt repayment than women, women spent much more often on education and somewhat more often on health (see Table 12). Possibly, the CfW wage is more often the largest income source if the worker is a male, while, when females work in CfW programmes, their husbands often have another income from which they pay for the most basic items like housing or food. Women can therefore more easily afford to use their wage for “second order” items (health, transportation, education). Table 11: Spending patterns of CfW participants (N=64) Item Number of answers Share of respondents (multiple answers possible) Share of answers given Housing (rent, electricity, water) 30 47% 23% Food 19 30% 15% Household equipment 15 23% 12% Education of children (school items, university, etc.) 15 23% 12% Debt repayment 14 22% 11% Support children or other relatives 13 20% 10% Investment in small projects 7 11% 5% Health 6 9% 5% Personal needs 5 8% 4% Clothes 3 5% 2% Transportation 2 3% 5% Holidays 1 2% 0.5% Sum 130 100% Source: Authors Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 116 Table 12: Spending patterns of CfW participants – gender differences Women Men Items named Count Items named Count Education 13 Housing (rent, electricity, water) 19 Housing (rent, electricity, water) 11 Food 13 Household equipment 7 Debt repayment 10 Support children or other relatives 7 Household equipment 8 Food 6 Support children or other relatives 6 Debt repayment 4 Health 2 Health 4 Investments 2 Investments 3 Education 2 Other 4 Other 6 Source: Authors These divergencies can be due to various factors. One explanation is that men are more interested in hardware (housing, household equipment, food, and the like) while women give the software a higher weight (especially education, but also health). Another, more probable, explanation is that within the families, men’s wages are considered the main source of income and therefore used mainly for core items (housing, debt repayment, food, and so on) while women’s wages are much more a windfall profit which can be used for occasional or additional needs such as health treatments and the costs of education. Some responses indicated that failures may have occurred in the targeting of vulnerable households. For example, a Jordanian man told us “I used the salary from the first month to go to Turkey for holidays” (13, participant, Deyr ‘Allā), and another one said that he would use his wage to buy work gear (26, participant, Kafr Asad). However, only Jordanians made such statements. Some Syrians, by way of contrast, mentioned that they were sending money to their family back home in Syria (149, participant, Kafr Ṣawm; 73, participant, Al-Azraq). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 117 These findings are very similar to those of the GIZ Post-employment Survey conducted in 2019. Here, participants were asked to tick the three main uses of their income and the largest share of respondents ticked house rent (45 per cent of all respondents), followed by debt repay (34 per cent), the paying of open bills (32 per cent), food (31 per cent), health (22 per cent), household items (20 per cent) and transportation (17 per cent). Just 4 per cent of the respondents ticked education and 5 per cent ticked items related to leisure (see Table 13). The GIZ Post-employment Survey also discovered that female and male CfW participants had, on average, different preferences in the use of their wages. Some of these differences were quite significant (some even at the 99 per cent confidence level) but possibly largely due to another factor: The relative share of female participants was much larger in the north of Jordan than in the south, where people had quite different consumption preferences. We ran endogeneity tests alluding to the fact that the impact of the regional factor was by far dominant, and the direct effect of gender became insignificant in regressions with interaction terms. Spending patterns of Syrians and Jordanians also differ according to the GIZ Post-employment Survey from 2019. Syrians more often spend a particularly high share of their CfW wage on house rent and health, while Jordanians more often spend a particularly high share of their income on leisure. Both findings are statistically significant at 1 per cent, respectively even at the 0.1 per cent level according to regressions run with different probit models. Possibly this is due to the fact that a higher share of Jordanians live in their own house (and hence do not have to pay a rent), have access to a form of health insurance (covering all medical treatment costs), have a longer history of making debts (now to be paid back) or extra income (that allows the CfW wage to be used for leisure activities). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 118 Table 13: CfW-income spending and saving patterns (results of GIZ Postemployment Survey, 984 respondents) Number of answers Share of all answers Share of all respondents All All All Females* Males* Jordanians* Syrians* Among top three spending items House rent 446 26% 45% 48% 44% 24% 68% Debt repay 336 19% 34% 45% 30% 38% 30% Paying open bills 319 18% 32% 42% 29% 28% 37% Food 305 18% 31% 28% 32% 30% 31% Health 217 12% 22% 27% 21% 17% 28% Household items 197 11% 20% 24% 19% 29% 10% Transportation 167 10% 17% 21% 15% 17% 17% Education 75 4% 8% 9% 7% 9% 6% Leisure 46 3% 5% 3% 5% 8% 1% Other 65 5% 7% 6% 7% 8% 6% Among top three savings items Personal items 139 13% 14% 18% 13% 18% 10% Health 126 11% 13% 15% 12% 13% 13% Education 89 8% 9% 12% 8% 10% 8% Small business 29 3% 3% 3% 3% 4% 1% Other 36 3% 4% 5% 3% 3% 4% Could not save at all 684 62% 70% 69% 69% 64% 75% Note: *While differences from the average are statistically significant (regressions with different probit model specifications), in some of the cases, this may actually be due to regional factors as endogeneity tests show. Source: Results of GIZ Post-employment Survey (GIZ, 2019) Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 119 Another interesting finding is that CfW workers pay significantly more on house rent, bills and food and less on debt repayment when they are married rather than single (statistically significant at the 95 per cent (food), respectively 1 per cent (house rent, bills) confidence level). Possibly, they live in larger houses and less often with their parents. This would explain why an above average share of married CfW participants stated in the GIZ Post-employment Survey that their CfW wage did not cover all of their daily needs (statistically significant at the 95 per cent confidence level). Statistically more often, CfW participants with a university degree spent a high share of their wages on education and their house rent (statistically significant at the 95 per cent (house rent), respectively 99.9 per cent (education) confidence level). In contrast, participants who had not even completed primary school spent a high share of their wages statistically more often on repaying debts and less often on transportation (both statistically significant at the 95 per cent confidence level). Finally, the GIZ survey also reveals that CfW participants in the south of Jordan more often used a high share of their wages for their house rent, due bills and food than those in the north of Jordan but less often on transportation, health and repaying debt (these regional differences are all statistically significant at the 99.9 per cent confidence level). Participants in urban areas more often used a relatively high share of their CfW wage on transportation and repaying debts (statistically significant at the 95 per cent (transportation), respectively 99.9 per cent (debt repay) confidence level). At the same time, our findings were also in line with the results of a study conducted by the Overseas Development Institute (ODI) on the effect of UNHCR cash assistance extended to Syrian refugees in Jordan. According to this study, Syrian households spent 69 per cent of their total income on housing and just 11 per cent each on health and food. The UNHCR cash assistance allowed them to raise their monthly spending on rent from JOD 120 to 130, utilities from JOD 20 to 35, health from JOD 15 to 29, education from JOD 9 to 25, and transport from JOD 10 to 20. On average, the Syrian households spent JOD 285 per month, which is equal to USD 5-6 per person per day on average in purchasing power parities (Jones et al., 2019). Another study reports on the spending patterns of participants in the KfW and ILO’s CfW programme, with similar figures. 83 per cent of its respondents listed “daily consumption” among the three main targets of use of their CfW wages, while 31 per cent listed repaying debt. 14 per cent mentioned housing Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 126 Donor representatives are aware of the importance of local procurement. A GIZ representative explained: We try to buy all materials and intermediate products locally, which in turn benefits the local economy. [In one of our projects], women are trained to produce tent sheets in the traditional way, which are then set up in the parks as sun sails. Men are also trained to manufacture brick blocks themselves in a specific way. (285, GIZ, Amman) Likewise, the programmes of the WFP and World Vision buy all ingredients that they need for cooking in worksite kitchens from the local market, if at all possible: “The local market of Kafr Ṣawm saw development because they are providing us with the food for the programme” (263, World Vision, Amman). The ILO projects also purchase all materials that they need for road construction and maintenance locally: “This should also have an effect [on local economic development]” (255, ILO, Amman). Other interviewees said it was not always easy to procure locally because the local products sometimes fell short in terms of quality and were sometimes more expensive than those available elsewhere and thus would not meet procurement regulations to choose the best-priced offer (306, CfW Coordination Group meeting). Several shopkeepers complained spontaneously that the CfW programmes had not bought their equipment at a local shop. For example, a shopkeeper from Kafr Ṣawm mentioned that the equipment of the kitchen, where food for the workers was cooked, could have been bought in his shop instead of a shop in Irbid (155, shopkeeper, Kafr Ṣawm). On the other hand, while a shopkeeper in Faqū’a told us that his compressor had been used on the treeplanting side, his profit had been rather small: So, […the CfW implementer] made a contract with me. The compressor was needed for the stony soil. They used it for five days. I got JOD 100 per day. JOD 50 was for the Egyptians [operating the machines] plus JOD 25 for diesel and transport. (196, shopkeeper, Faqū’a) Investment effects The investment effects of the Jordanian CfW programmes appear to be small. Some experts said that the CfW programmes had a positive effect on investment: “Also, there is impact on the local economic market and there is more investment in communities” (264, Oxfam, Amman). Similarly, a Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 127 shopkeeper stated: “I have heard that the people invest more because of the programmes. It has affected our markets” (227, shopkeeper, Kafr Asad). Yet, as CfW wages are too low to be invested rather than consumed, this is most likely a minor spin-off of the multiplier effect, for which we could not, however, find conclusive evidence. During our own field research, only 7 CfW participants told us that they were planning to invest part of their CfW wages. This is understandable because the wages are not very high and are only paid for three months. In addition, many Jordanians and Syrians have debts to be repaid first while others are also struggling with the costs of their everyday needs (162, participant, highway; 27, participant, Kafr Asad). In other words, none of the CfW participants had much financial flexibility for investments. Two out of the 7 participants with concrete investment plans interviewed also admitted that the CfW wages would probably not suffice to finance their investment ideas. If these investments are ever made, they will predominantly be in subsistence farming with some minor additional earnings. Two interviewees had the idea to plant some crops in the garden and sell them later on the local market; two wanted to buy cattle; one planned to keep bees for honey; one intended to open a bakery; and one wanted to set up a household repair shop. These findings are in line with the workers survey conducted in 2019 among the participants of the ILO employment infrastructure programme. Only 0.9 per cent of the households interviewed reported investing part of their CfW wages in private business. Another 0.5 per cent bought animals as an additional source of income while 1.4 per cent planned to save part of their wages (NAMA & ILO, 2019). Likewise, the GIZ’s Post-employment Survey found that only 30 per cent of the respondents were able to make any savings at all (see above) and, of those, most saved for personal items, that is, future consumption. Just 13 per cent reported having made provisions for future health care spending, 9 per cent for future spending on education, and only 3 per cent for small business investment (GIZ, 2019, see also Table 13). Interestingly, the share of people who saved for small business investment was particularly high among divorced people and particularly low among CfW participants in the south of Jordan. CfW participants in urban areas were particularly rarely able to make any savings at all but, quite surprisingly, more prone to save for investment rather than for education, health, or personal items. At the same time, the share of CfW participants who had made savings mainly Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 128 for individual future consumption needs turned out to be particularly high among women and particularly low among married people.21 6.3.3 Effects of the creation of public goods (Hypothesis 14) It remains, however, unclear to what extent the assets created by CfW programmes in Jordan had secondary effects on LED as well (Hypothesis 14). Most infrastructure or services provided by CfW projects are unlikely to have tangible income effects. For example, the collection and recycling of waste, the embellishment of public parks, and the planting of trees and rehabilitation of nature reserves are definitely positive for the quality of life of people in the community but not important for any income-generating activity. Likewise, the renovation of schools and health clinics is very important for the well-being of students and patients but presumably without measurable income effect. Even the construction of rural roads in Jordan has probably no major economic importance because the Jordanian road network does not suffer from major gaps. In contrast, other CfW projects can have a very substantial economic payoff but the size of this effect is difficult to estimate. This applies to the rehabilitation of dams, water reservoirs and irrigation systems, support for the intensification of agriculture, and protective measures against soil erosion. However, our methodology proved inadequate to produce any estimations. Despite that, numerous interviewees stressed how much they appreciated the public goods that CfW programmes had created in their communities. When we asked a young Jordanian man “Has there been an effect on the community?”, he answered: “The infrastructure has been improved; they have computers in the schools now; there have been developments; it has helped the community” (116, non-participant, Umm al-Jimāl). An elderly Jordanian woman noted: “There was an improvement of the streets and 21 Both findings regarding savings for investment are statistically significant at the 95 per cent confidence level and the findings regarding urban participants at the 99 per cent confidence level. Both findings on groups most probably saving for future consumption are statistically significant at the 99.9 per cent (gender) and 99 per cent (marital status) confidence level, respectively. All confidence levels were computed through regressions run with different probit model specifications. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 129 also in electricity. Streetlamps have been installed, also communication techniques, cables and so on improved” (23, non-participant, Hawfa). These statements are in line with the results of qualitative interviews conducted by Roxin et al. (2020), who found that CfW participants, other community members, and the representatives of the municipalities consider the local infrastructure built by the CfW programmes useful to everybody. 6.3.4 Direct and indirect labour market effects It is difficult to draw conclusions on the overall effects of CfW programmes on the labour market from our own research as we did not focus on that topic. Yet, we found indications for both negative and positive effects, corroborated by accounts from the literature. In the following subsections, we will first consider the direct effects of CfW programmes on the employability of participants who go through (i) technical skills upgrading; and (ii) the improvement of soft skills and higher psychological resilience. Second, we will briefly discuss the more ambiguous indirect effects that are due to (iii) an increased willingness to work in less attractive sectors (in other words: a diminished shame culture), and (iv) competition over suitable labour induced by CfW wages and working conditions. Direct effects through technical skills upgrading The most interesting question concerns the long-term employment effect for former CfW participants: When CfW programmes end, will more or less workers have a job than before? Of course, the answer depends on many aspects, but the main one is the direct employment effect, that is, whether CfW programmes have bettered or worsened the employability of their participants. The dominant way to improve the employability of CfW participants is via skills upgrading, which was the focus of the initial Hypothesis 15. As our interviews provided only little evidence on this, we also drew complementary data from other surveys. Several of our interviews revealed that, at the local level, many CfW participants, non-participants and local experts thought that most CfW projects in Jordan did not provide enough training and hence did not prepare participants well in looking for follow-up employment. At the same time, our interviews showed that national and international experts were just as well Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 130 aware of the potential employability effect but also of the related difficulties: “The MOL has understood [...] that CfW programmes are not only crucial for the creation of infrastructure but also for the creation of employment” (256, ILO) but it is problematic for CfW programmes to provide skills development “according to the needs of the labour market” (245, Caritas). Even if skills training provided by CfW programmes opens up access to labour market segments, this may often be in theory only, and not in practice, as several sectors are not open to migrant workers (see subsection 3.5.3). In that case, CfW could raise expectations that cannot be met. The DEval evaluation report (Roxin et al., 2020) suggests that, all in all, the CfW programmes had no positive impact on the employability of their participants. It reports that 40 per cent of the participants had a job before they started working for the programmes while the respective share was only 28 per cent among those who applied for a CfW job but were not accepted. In the course of their employment, of course, most CfW participants give up their previous job. However, even afterwards, only 25 per cent – and hence less than before – had any employment. This finding was similar for Syrians and Jordanians, for women and for men (Roxin et al., 2020). The results of the workers survey conducted in 2019 among participants of the ILO’s Employment-intensive Investment Programme (EIIP) were similar. It revealed that 57 per cent of men and 43 per cent of the women had a job before they started working for the programme but that only 32 per cent, respectively 13 per cent, had a job right afterwards. These shares do not differ substantially between Jordanians and Syrians. However, while 47 per cent of the Syrians had been working in construction, agriculture or basic services before they became engaged in the CfW programme, but only 1 per cent in commerce or manufacturing, the respective shares were 31 per cent against 6 per cent for Jordanians. The majority of those who had had a job before their CfW participation had a wage of between JOD 7 and 14 per day, while the majority of those who had found a job right after their CfW participation received somewhat lower wages (mostly between JOD 5 and 12 per day) (NAMA & ILO, 2019). The employment situation immediately after the CfW placement thus looked dim; unfortunately, no data was available to check what share of former CfW participants found better job opportunities after a search period. Interestingly, 7 per cent of all female participants and 4 per cent of all men had actually worked in a skilled worker’s or skilled employee’s position Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 131 before they were hired by the ILO’s CfW programme (NAMA & ILO, 2019). This can be seen as a sign that the ILO programme in some instances crowds out existing employment, that is, that it attracts people from good jobs but with only a limited chance of getting the same kind of jobs again after the end of the CfW labour contract. Even more so, most CfW participants who found a new job after the end of their respective CfW project did not believe that their CfW employment had helped them obtain the new job. Only 11 per cent of all men and 4 per cent of all women believed that the CfW employment had been helpful while 35 per cent of them said that this was because they had learned a new skill and 2 per cent because it has provided them with the seed capital to start their own project; 63 per cent said that the CfW employment had only been helpful because it built wasţa (relations) with other employers or other workers (NAMA & ILO, 2019). However, the GIZ Post-employment Survey produces a much more positive picture. Possibly, the difference is due to the fact that the ILO and the GIZ survey covered only the participants of the programmes run by the respective organisation. While the GIZ survey included the entirety of participants in the GIZ Green Infrastructure Programme in 2019, the ILO workers survey only covered a sample of those who were working in the organisation’s Employment-Intensive Investment Programme – and any sampling produces biases. In any case, it is possible to say that the GIZ Green Infrastructure Programme did not crowd out workers in large numbers from residual occupations. 60 per cent of the participants of the GIZ Green Infrastructure programme had been without work before the programme started, and just 11 per cent had been in formal sector jobs (GIZ, 2019). In regard to their post-CfW employment, 74 per cent of the participants in the GIZ Green Infrastructure Programme longed for a new CfW job, while 31 per cent were looking for a different job and 12 per cent wanted to participate in a more systematic training programme (whereby double answers were possible). Respondents previously unemployed were likely to look for another CfW opportunity; those previously in informal employment likely to consider another informal job opportunity; while respondents previously employed, whether formally or informally, were likely to seek formal employment.22 Educated respondents were more likely to look for a 22 The first and the last finding at a 99 per cent; the two findings in the middle at 99.9 per cent confidence level. Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 132 formal job afterwards.23 Interestingly, there was also a correlation between respondents’ spending patterns of their CfW income and their post-CfW plans24: those spending their income on rent tended to look for another CfW job; those spending their income on bills tended to look for informal employment; and those spending on debt repayments, as well as those not able to save CfW income, tended to look for formal jobs (GIZ, 2019). Direct effects through soft skills, attitude changes and networking opportunities In addition to technical and vocational skills, participation in CfW programmes can foster time management skills and better discipline through the regularity of the activity – soft skills, which not only affect employability in a positive way but also create networking opportunities. A Syrian women stressed these aspects next to the positive effects on social cohesion: How does it help the community as a whole? They [the people] help each other with their needs; the programme brings people together; it helps them to manage their time more efficiently, and it helps them to regulate spending their income. (119, non-participant, Umm al-Jimāl) “This [additional income] has a good impact and effect and has a good impact on the psychological factors of the person” (31, local expert, Kafr Asad). Likewise, most respondents to the GIZ Post-employment Survey also considered soft skills rather than technical skills as their main takeaway from participation in the GIZ Green Infrastructure Programme: 55 per cent stated that they had learnt to work in teams; 35 per cent said that they had improved their commitment; 25 per cent believed that they had become more patient; while 18 per cent perceived they were now better in their time management. At the same time, 24 per cent thought that they had acquired new technical skills and 1 per cent felt that they had gained entrepreneurial skills (interestingly, all percentages for both hard and soft skills are even much higher for female participants, see Appendix E2). 23 Respondents with a BA degree at 95 per cent confidence level, those with secondary education or vocational training both at a 99 per cent confidence level. 24 These findings are statistically significant at 95 per cent (first two) or 99.9 (last two) confidence levels. Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 133 Another benefit gained from participation in CfW programmes was the opportunity to build networks. Although we found hardly any concrete evidence that this opportunity had been used Roxin et al. (2020) revealed that most CfW participants established new contacts to people who might be helpful for their future employment: 63 per cent of their interviewees mentioned helpful contacts to employees of the municipality and 25 per cent to GIZ staff members but only 0.6 per cent to employees of private firms. Indirect effects through a decreased shame culture CfW programmes have a mitigating effect on what many people in Jordan call “the shame culture” (see subsection 3.5.2). We gathered accounts on this aversion to “disreputable” work in the waste sector but similar effects for CfW programmes in the agricultural or construction sectors are conceivable, too. Eighteen of our interviewees raised the issue (2 participants, 5 local experts, 6 non-participants, 3 shopkeepers, and 2 general experts) stressing that CfW programmes had reduced people’s reluctance to work in the waste sector. In many sites, side-events of the CfW programmes such as lectures and awareness campaigns were mainly responsible for this change in attitudes: Shame culture was eradicated, the municipality terminated the idea that work in waste collection is shameful. How did you achieve that? Many lectures were held. We pointed out that they (waste collectors) are working for the municipality. The head of the municipality held speeches with important stakeholders. By combined efforts of municipalities and the local community itself. (5, local expert, Deyr ‘Allā) In some cases, the staff of the CfW programmes visited private houses one by one in order to talk on the issue of working in the waste sector (29, GIZ, Kafr Asad). What also played a role was that “large numbers of people [were] looking for jobs. They changed their perception about what is shameful” (70, participant, Al-Azraq). Many workers just had to accept working in a sector without a good reputation because of economic pressure and the need to seize any opportunity to generate income: Before this, the idea was that you would work as an engineer. Working with waste would not even be on your radar. People do need the work, the situation is difficult and prices are rising. This helped to get rid of the shame. (106, participant, Umm al-Jimāl) Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 134 While, in the past, it had sometimes been difficult to recruit workers for the work on waste (29, GIZ, Kafr Asad) now there was “a growing trend that people participate in waste collection. Before, people looked down on people collecting waste. Even shopkeepers now tell us that they wished they had a job like ours, where we collect the waste” (106, participant, Umm al-Jimāl). People have got just accustomed to working in the sector: In the beginning, there was this stigma about collecting waste. But now, it got ingrained in their own mentality. What is the reason for this mind shift? The way how they approach people. They came in a different way, they told us about the reasons for collecting waste. Many people shifted from working on farms because they pay more here [but] the people work here not only because of money. The people love working here. […] They tell us that waste in nothing to be ashamed of. They made working with waste as something that is not shameful. (94, participant, Umm al-Jimāl) Yet, among non-participants, the shame culture is sometimes still quite present: “What do you think about the programmes where Syrians and Jordanians work with each other?” “These programmes are run by GIZ or foreigners. Jordanians would not want to work in such occupations” (50, shopkeeper, Kafr Asad). In general, the association of shame culture and waste is nothing to “be solved in one round of CfW” but will take time. Yet, it is worth doing “in a country that has a huge problem with waste” (259, expert interview, AAH). Indirect effects through wages and working conditions There are, however, also signs of distortionary effects. In some areas at least, CfW programmes crowd out private employment because the wages paid by the CfW programmes exceed the average level of wages for unskilled workers (for instance, 256, ILO; 305, GIZ). A government official said that when CfW programmes start, “workers leave their [previous] jobs in order to take part in more lucrative CfW programmes for 4-6 months” (248, Ministry of Labour). And a donor representative seconded: Syrians work for JOD 1 per hour in private farms, which means that they do not earn more than JOD 8 per day. Offering them JOD 12 or 14 [in CfW projects] means thus that crowding-out is not unlikely to happen. (256, ILO) Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 135 In the long run, competition over Syrian workers can of course have positive effects. If the demand for workers by CfW programmes remains large, private employers may feel at some point in time that they have to adjust their wages in order to withstand the competition. But this scenario is not very probable because international donors are likely to reduce rather than increase their spending on CfW programmes in Jordan over the coming years and the offer of cheap workers on the Jordanian labour market will probably continue to increase rather than decrease. A more positive effect of CfW programmes is the establishment of labour standards at least in an “artificial”, that is, donor-funded segment of the Jordanian labour market. The CfW programmes carefully observe ILO norms regarding income levels, safety at the workplace, social protection for families, prospects for personal development, social integration, and freedom to express concerns, to organise and participate in the decisions that affect work. Some experts emphasised that CfW programmes provide formal jobs, namely that they bring with them social security (102, local expert, Umm al-Jimāl; 259, AAH) and an “aware[ness] of specific working standards” (259, AAH). This may, in the long term, have an effect on the attitudes of workers and employers in other parts of the labour market, either improving working conditions in the formal sector or even making the formalisation of informal jobs more likely. 6.4 Effects of the way CfW programmes are designed With regard to the design of CfW programmes, three issues were mainly raised by our interviewees: the duration of employment; skills development during participation; as well as the targeting mechanisms and application procedures. In addition, some CfW participants criticised that their wages were too low or paid too late, that their work equipment or safety provisions were insufficient, that it was difficult for them to commute to the sites of CfW projects, or they complained, more generally, about the CfW setup (as opposed to other supporting mechanisms) and management choices (see Table 14). Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 238 TableE2(cont.): ResultsofthefirstroundoftheGIZPost-employmentSurveyconductedamongallparticipantsof the GIZ Green Infrastructure (GI) Programme during 2019 All Jordanian females Syrian females Jordanian males Syrian males All females All males All Jordanians All Syrians Recommendations Increase project duration 8 1% 1 1% 2 2% 3 1% 2 1% 3 1% 5 1% 4 1% 4 1% Increase salary 5 1% 1 1% 1 1% 0 0% 3 1% 2 1% 3 0% 1 0% 4 1% Decrease working hours 16 2% 2 2% 2 2% 7 2% 5 1% 4 2% 12 2% 9 2% 7 1% Improve meals 19 2% 3 2% 5 4% 7 2% 4 1% 8 3% 11 2% 10 2% 9 2% In need for assistance to start a new project 2 0% 0 0% 0 0% 0 0% 2 1% 0 0% 2 0% 0 0% 2 0% Increase training 2 0% 0 0% 0 0% 2 1% 0 0% 0 0% 2 0% 2 0% 0 0% Expressed wish to join in other projects 1 0% 1 1% 0 0% 0 0% 0 0% 1 0% 0 0% 1 0% 0 0% Provide greater accessibility for agricultural sites 9 1% 5 4% 0 0% 1 0% 3 1% 5 2% 4 1% 6 1% 3 1% Provide work opportunities in other projects 7 1% 2 2% 0 0% 1 0% 4 1% 2 1% 5 1% 3 1% 4 1% Establish literacy centres for workers 4 0% 1 1% 0 0% 3 1% 0 0% 1 0% 3 0% 4 1% 0 0% Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 239 TableE2(cont.): ResultsofthefirstroundoftheGIZPost-employmentSurveyconductedamongallparticipantsof the GIZ Green Infrastructure (GI) Programme during 2019 Improve treatment of supervisors 3 0% 0 0% 0 0% 3 1% 0 0% 0 0% 3 0% 3 1% 0 0% Improve transportation 1 0% 1 1% 0 0% 0 0% 0 0% 1 0% 0 0% 1 0% 0 0% Add more projects 6 1% 2 2% 1 1% 1 0% 2 1% 3 1% 3 0% 3 1% 3 1% Increase application period 2 0% 0 0% 0 0% 1 0% 1 0% 0 0% 2 0% 1 0% 1 0% Provide training certificate 1 0% 0 0% 0 0% 0 0% 1 0% 0 0% 1 0% 0 0% 1 0% Work discrimination 1 0% 0 0% 0 0% 0 0% 1 0% 0 0% 1 0% 0 0% 1 0% Set a work plan 1 0% 0 0% 0 0% 1 0% 0 0% 0 0% 1 0% 1 0% 0 0% Fair distribution of the workload 7 1% 6 5% 0 0% 0 0% 1 0% 6 2% 1 0% 6 1% 1 0% Improve complaint system 6 1% 0 0% 0 0% 6 2% 0 0% 0 0% 6 1% 6 1% 0 0% Set attendance policy 61 6% 9 7% 13 10% 16 4% 23 7% 22 9% 39 5% 25 5% 36 8% Improve workplace 27 3% 6 5% 6 5% 2 1% 13 4% 12 5% 15 2% 8 2% 19 4% Source: Authors, based on the results of the GIZ Post-employment Survey (GIZ 2019), not undergone checks for robustness or endogeneity so far Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 240 Table E3: Made new friendships with the other nationality Probit regression Log likelihood = -239.73979 Number of obs = 778 LR chi2 (9) = 168.81 Proh > chi 2 = 0.0000 Pseudo R2 = 0.2604 new_frnds Coef. Std. Err. z P> |z| [95% Conf. Interval] Gender .7475154 .1944679 3.84 0.000*** .3663653 1.128666 Nationality -1.065718 .1602609 -6.65 0.000*** -1.379824 -.7516128 urban -.3672281 .1595638 -2.30 0.021* -.6799674 -.0544889 South -1.031437 .1341911 -7.69 0.000*** -.1.294447 -.7684275 mstatus_ married -.2363204 .146625 -1.61 0.107 -.5237001 .0510593 mstatus_ divorced .302945 .5800481 0.52 0.601 -.8339285 1.439818 dbts_uses_slry .128012 .1475413 0.87 0.386 -.1611636 .4171877 daily_needs_ uses_slry .0366557 .1577549 0.23 0.816 -.2725381 .3458496 no_savings .4165064 .1380541 3.02 0.003** .1459253 .6870875 _cons 1.990852 .2624538 7.59 0.000 1.476452 2.505252 Notes: All variables (the dependent variable and all independent variables) are binary. “Gender” has been defined as being “1” for females and “0” for males. “Nationality” has been defined as being “1” for females and “0” for males. Marital status indicators compare married participants (“mstatus_married”=1) and divorced participants (“mstatus_divorced”=1) with singles. “dbts_uses salary” means that a participant had debts before starting work in the CfW programme. “daily_needs_uses salary” means that CfW wage covered the daily expenses of the CfW participant. “no_savings” means that the participant could not make any savings during the CfW emplyoment. *means: statistically significant at the 95%-confidence level. **means: statistically significant at the 99%-confidence level. ***means: statistically significant at the 99.9%-confidence level. Source: Authors, based on the results of results of the GIZ post-employment survey (GIZ, 2019). Selected probit regressions testing the statistical significance of differences in the results (performed using STATA). Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 241 Table E4: Could not make any savings during CfW employment Probit regression Log likelihood = -571.80253 Number of obs = 984 LR chi2 (10) = 66.60 Proh > chi 2 = 0.0000 Pseudo R2 = 0.0550 no_savings Coef. Std. Err. z P> |z| [95% Conf. Interval] gender -.2162433 .103765 -2.08 0.037* -.4196189 -.0128676 nationality -.21778 .0956275 -2.28 0.023* -.4052064 -.0303536 urban .7304568 .1322762 5.52 0.000*** .4712002 .9897135 south -.2656011 .096399 -2.76 0.006** -.4545396 -.0766626 mstatus_ married .2454744 .0971645 2.53 0. 012* .0550355 .4359134 mstatus_ divorced .2921801 .2963084 0.99 0.324 -.2885738 .8729339 another_job_ afterCfW -.3433629 .1841814 -1.86 0.062 -.7043518 .0716261 edulvl_ below_sec .348979 .1157886 3.01 0.003** .1220375 .5759204 edulvl_bchlr -.1850794 .1756522 -1.05 0.292 -.5293513 .1591926 edulvl_ vocational .1139042 .1471986 0.77 0.439 -.1745997 .4024081 _cons .4787478 .1161983 4.12 0.000 .2510033 .7064923 Notes: All variables (the dependent variable and all independent variables) are binary. “Gender” has been defined as being “1” for females and “0” for males. “Nationality” has been defined as being “1” for females and “0” for males. “South” refers to the south of Jordan (governorates of Al-Karak, At-Tawfila, Ma’ān and Al-’Aqaba) Marital status indicators compare married participants (“mstatus_married”=1) and divorced participants (“mstatus_divorced”=1) with singles. Education level indicators compare people with less than primary school education (“edulvl_below_sec”=1), university degree (“edulvl_bchlr”=1) or vocational training (“edulvl_vocational”=1) with people who have full but only primary education. “Another_job_afterCfW” means that a participant had a job in parallel to her/his CfW employment. *means: statistically significant at the 95%-confidence level. **means: statistically significant at the 99%-confidence level. ***means: statistically significant at the 99.9%-confidence level. Source: Authors, based on the results of results of the GIZ post-employment survey (GIZ, 2019). Selected probit regressions testing the statistical significance of differences in the results (performed using STATA) Markus Loewe / Tina Zintl et al. German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 242 Table E5: Would advise CfW participants to a friend Probit regression Log likelihood = -46.542915 Number of obs = 924 LR chi2 (22) = 52.01 Proh > chi 2 = 0.0003 Pseudo R2 = 0.3585 advice_friend Coef. Std. Err. z P> |z| [95% Conf. Interval] gender 1.131951 .5546166 2.04 0.041* .0449222 2.218979 nationality .110164 .3345984 0.33 0.742 -.5456369 .7659649 urban -.7515016 .3860939 -1,95 0.052 -1.508232 .0052285 south 1.030569 .4931131 2.09 0.037* -.0640847 1.997052 another_job_ afterCfW 1.490147 .5772357 -0.26 0.796 -1.280376 .9823464 mstatus_ married .9814666 .3937832 2.49 0.013* .2096658 1.753267 mstatus_ divorced -.7601855 .6521055 -1.17 0.244 -2.038289 .5179179 mstatus_ widowed -1.203755 .7369287 -1.63 0.102 -2.648109 .2405984 edulvl_ below_sec .1070208 .4357264 0.25 0.806 -.7469873 .9610289 edulvl_bchlr .1586334 .5348877 0.30 0.767 -.8897272 1.206994 edulvl_ vocational -.1249568 .415215 -0.30 0.763 -.9387633 .6888497 unsatisfied_ pympt -.0131809 .396238 -0.03 0.973 -.789793 .7634313 unsatisfied_er -.5001824 .6619111 -0.76 0.450 -1.797504 .7971395 unsatisfied_ sup -.4130111 .4398527 -0.94 0.348 -1.275107 .4490843 unsatisfied_ WP -.7253904 .4421619 -1.64 0.101 -1.592012 .1412309 unsatisfied_ WHrs -.9614934 .3810765 -2.52 0.012* -1.70839 -.2145973 unsatisfied_ WEq .1949029 .4786466 0.41 0.684 -.7432272 1.133033 unsatisfied_ sfty .5945896 .6985337 0.85 0.395 -.7745112 1.96369 Community effects of cash-for-work programmes in Jordan German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 243 unsatisfied_ meals .0992028 .4044824 0.25 0.806 -.6935681 .8919737 unsatisfied_ trnspt .1307423 .4069397 0.32 0.748 -.6668449 .9283295 unsatisfied_ trng .0161077 .5888063 0.03 0.978 -1.137931 1.170147 unsatisfied_ FBmchsm -.5086811 .4322887 -1.18 0.239 -1.1355951 .3385892 _cons 2.078618 .350615 5.93 0.000 1.476452 2.765811 Notes: All variables (the dependent variable and all independent variables) are binary. “Gender” has been defined as being “1” for females and “0” for males. “Nationality” has been defined as being “1” for females and “0” for males. “South” refers to the south of Jordan (governorates of Al-Karak, At-Tawfila, Ma’ān and Al-’Aqaba). “Another_job_afterCfW” means that a participant had a job in parallel to her/his CfW employment. Marital status indicators compare married participants (“mstatus_married”=1), divorced participants (“mstatus_divorced”=1) and widowed participants (“mstatus_widowed”=1) with singles. Education level indicators compare people with less than primary school education (“edulvl_below_sec”=1), university degree (“edulvl_bchlr”=1) or vocational training (“edulvl_vocational”=1) with people who have full but only primary education. The remainder of the independent variables represent dissatisfaction with single aspects of the CfW programme: the wage (“unsatisfied_pymt”), the employer (“unsatisfied_er”), the supervisor (“unsatisfied_sup”), the workplace (“unsatisfied_WP”), the working hours (“unsatisfied_WHrs”), the work equipment (“unsatisfied_WEq”), safety at the workplace (“unsatisfied_sfty”), the meals provided at the worksites (“unsatisfied_meals”), transportation to the worksites (“unsatisfied_trnspt”), the training (“unsatisfied_trng”) and the feed-back mechanisms (“unsatisfied_FBmchsm”). *means: statistically significant at the 95%-confidence level Source: Authors, based on the results of results of the GIZ post-employment survey (GIZ, 2019). Selected probit regressions testing the statistical significance of differences in the results (performed using STATA) Publications of the German Development Institute/ Deutsches Institut für Entwicklungspolitik (DIE) Studies 102 Vidican Auktor, Georgeta, Tilman Altenburg, & Andreas Stamm. (2020). The transition towards a green economy and its implications for quality infrastructure (218 pp.). ISBN 978-3-96021-131-0. DOI: 10.23661/s102.2020. 101 Silke Weinlich, Max-Otto Baumann, Erik Lundsgaarde, & Peter Wolff. (2020). Earmarking in the multilateral development system: Many shades of grey (347 pp.). ISBN 978-3-96021-112-9. DOI: 10.23661/s101.2020. 100 Stamm, Andreas, Luise Dietrich, Heike Harling, Laura Häußler, Florian Münch, Jana Preiß, & Jan Siebert. (2019). Sustainable public procurement as a tool to foster sustainable development in Costa Rica. Challenges and recommendations for policy implementation (114 pp.). ISBN: 978-3-96021102-0. DOI: 10.23661/s100.2019. 99 Licht, Nathalie. (2019). Wandel von Sozialpolitik im Kontext von Global Social Governance: Das Beispiel Peru (346 pp.). ISBN: 978-3-96021-083-2. DOI: 10.23661/s99.2019. 98 Duguma, Mesay K., Michael Brüntrup, & Daniel Tsegai. (2017). Policy options for improving drought resilience and its implication for food security: The cases of Ethiopia and Kenya (87 pp.). ISBN 978-3-96021-056-6. 97 Reeg, Caraoline. (2017) Spatial development initiatives – potentials, challenges and policy lesson: With a specific outlook for inclusive agrocorridors in Sub-Sahara Africa (176 pp.). ISBN 978-3-96021-048-1. 96 Horstmann, Britta, & Jonas Hein. (2017). Aligning climate change mitigation and sustainable development under the UNFCCC: A critical assessment of the Clean Development Mechanism, the Green Climate Fund and REDD+ (135 pp.). ISBN 978-3-96021-043-6. 95 Hampel-Milagrosa, Aimée, Hauke Brankamp, Thomas Cremer, Alexander Haddad, Katharina Pannwitz, Franziska Wehinger, Sangeeta Agasty, & Tamal Sarkar. (2017). Retail FDI liberalisation and the transformation of agrifood value chains in India (123 pp.). ISBN 978-3-96021-038-2. 94 Altenburg, Tilman, Cecilia Fischer, Kerstin Huck, Anna Kruip, Sören Müller, & Stefanie Sörensen. (2017). Managing coastal ecosystems in the Philippines: What Cash for Work programmes can contribute (108 pp.). ISBN 978-3-96021-033-7. [Price: 10.00 Euro; books may be ordered from the Institute or through bookshops.] Discussion Paper 21/2020 Pahl, Stefan, Clara Brandi, Jakob Schwab, & Frederik Stender. Cling together, swing together: the contagious effects of COVID-19 on developing countries through global value chains (1-29 pp.). ISBN 9783-96021-133-4. DOI: 10.23661/dp21.2020. 20/2020 von Haldenwang, Christian. Digitalising the fiscal contract: an interdisciplinary framework for empirical inquiry (1-28 pp.). ISBN 9783-96021-132-7. DOI: 10.23661/dp20.2020. 19/2020 Fischer, Roger A. How the G7 Reviews its Work on Development. A Case Study of Internal Accountability (45 pp.). ISBN 978-3-96021-130-3. DOI: 10.23661/dp19.2020. 18/2020 Römling, Cornelia, & Sarah Holzapfel. Monitoring in German bilateral development cooperation: A case study of agricultural, rural development and food security projects (49 pp.). ISBN 978-3-96021129-7. DOI: 10.23661/dp18.2020. 17/2020 Laudage, Sabine. Corporate tax revenue and foreign direct investment: potential trade-offs and how to address them (46 pp.). ISBN 978-396021-128-0. DOI: 10.23661/dp17.2020. 16/2020 Cavatorta, Francesco, & Fabio Merone. Never-ending reformism from above and dissatisfaction from below: The paradox of Moroccan post-Spring politics (32 pp.). ISBN 978-3-96021-127-3. DOI: 10.23661/dp16.2020. 15/2020 Eppler, Mirko, Stella Gaetani, Francy Köllner, Jana Kuhnt, Charles Martin-Shields, Nyat Mebrahtu, …Carlotta Preiß. Information and communication technology in the lives of forcibly displaced persons in Kenya (31 pp.). ISBN 978-3-96021-126-6 DOI: 10.23661/dp15.2020. 14/2020 Hackenesch, Christine, Julia Leininger, & Karina Mross. What the EU should do for democracy support in Africa: Ten proposals for a new strategic initiative in times of polarisation (45 pp.). ISBN 978-3-96021125-9. DOI: 10.23661/dp14.2020. 13/2020 Never, Babette, Jose Ramon Albert, Hanna Fuhrmann, Sebastian Gsell, Miguel Jaramillo, Sascha Kuhn, & Bernardin Senadza. Carbon consumption patterns of emerging middle classes (42 pp.). ISBN 978-396021-124-2. DOI: 10.23661/dp13.2020. 12/2020 Wehrmann, Dorothea. Transnational cooperation in times of rapid global changes: The Arctic Council as a success case? (29 pp). ISBN 978-3-96021-123-5. DOI: 10.23661/dp12.2020. [Price: EUR 6.00; publications may be ordered from the DIE or through bookshops.] Analysen und Stellungnahmen [ISSN Print 1434-8934 / Online 2512-9325] 17/2020 Rauch, Theo, & Michael Brüntrup. Ansätze der Kleinbauernförderung im Globalen Süden: Kontroversen, Erfahrungen, Synthesen. DOI: 10.23661/as17.2020. 16/2020 Schütze, Nora, Andreas Thiel, Rodrigo Vidaurre, Pilar Paneque, & Jesús Vargas. Koordination in der Governance von Flussgebieten in Südspanien stärken: Kooperation, Anreize und Überzeugungsarbeit. DOI: 10.23661/as16.2020. 15/2020 Schoderer, Mirja, & Ines Dombrowsky. Foren, Gebühren und Datenflüsse: Koordination der Bergbauund Wasserpolitik in der Mongolei. DOI: 10.23661/as15.2020. 14/2020 Weinlich, Silke, Max-Otto Baumann, & Erik Lundsgaarde. Entwicklungszusammenarbeit im Nahen Osten und in Nordafrika: auch in konfliktbetroffenen Ländern neue Gesellschaftsverträge unterstützen. DOI: 10.23661/as14.2020. 13/2020 Meergans, Franziska, Christina Aue, Christian Knieper, Sascha Kochendörfer, Andrea Lenschow, & Claudia Pahl-Wostl. Im Spannungsfeld von Wasser-, Energieund Landwirtschaftspolitik: Neue Wege für den Wasserschutz in der Weser-Ems-Region. DOI: 10.23661/as13.2020. [Analysen und Stellungnahmen free of charge available from the DIE.] BriefingPaper[ISSNPrint1615-5483/Online2512-9384] 25/2020 Hendra, John, & Max Baumann. Towards more policy advice: maximizing the UN’s assets to build back better. DOI: 10.23661/bp25.2020. 24/2020 Baumann, Max-Otto, Erik Lundsgaarde, & Silke Weinlich. Earmarked funding for multilateral development cooperation: asset and impediment. DOI: 10.23661/bp24.2020. 23/2020 Paczyńska, Agnieszka. Russia in Africa: is great power competition returning to the continent? DOI: 10.23661/bp23.2020. 22/2020 Zoccal, Geovana. Triangular cooperation: broader, more dynamic and flexible. DOI: 10.23661/bp22.2020. 21/2020 Müngersdorff, Maximilian, & Tim Stoffel. Pathways for integrating socially responsible public procurement in municipalities. DOI: 10.23661/bp21.2020. [Briefing Papers free of charge available from the DIE.] For a complete list of DIE publications: http://www.die-gdi.de