![Sample our Global Development journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5929/TOC-Subject-Sample-2021-Global-Development.jpg"})
Abstract
Since the birth of the formal Swedish development policy in 1962, the overarching goal of Swedish aid has been poverty reduction. While the goal has been up for revision four times (under two social democratic and two conservative governments) it has only been rephrased rather than reformed. In addition, Sweden’s ambition to devote one per cent of its gross national income to development aid each year has been largely upheld since it was first achieved in 1975, save a period of 13 years (1993–2005, under the social democratic rule) where it was abandoned with reference to budgetary pressures. Aid was, however, never less than 0.7 per cent of GNI. Goals and volumes not being altered over time, even under government changes, suggest Swedish development cooperation policy has been formed under consensus. This apparent continuity, however, masks some important differences between political parties’ policy preferences, and the stability may be better explained by large majorities being in favour of current policies than by consensus. I argue that the preconditions for those differing ideational positions to be translated into policy change are today quite favourable, suggesting that a policy paradigm shift, a ‘radical policy change’, may be in the making.
Acknowledgements
I would like to thank Bertil Odén, Mats Hårsmar, the authors of the companion papers in this volume, and in particular one of this journal’s peer reviewers for valuable comments on previous versions of this paper. All remaining ambiguities, misconceptions and errors are my own.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 The social learning process and the three orders of change as described by Hall (Citation1993) can be linked to various dimensions of organisational learning. First-order changes are similar to single-loop learning, the everyday business using existing instruments, making changes when and where needed. When instruments are found to work less effectively or efficiently, changes are searched for in processes that may be based on learning from what has worked or not, or on altered ambitions or priorities. This second-order change resembles double loop learning. When reflections on the deeper determinants, conceptual learning occur, when rules of the game are redrawn and the frames of reference are changed we observe third-order changes, reminiscent of triple-loop learning (i.e. ‘double loop learning about double loop learning’).
2 Hall (Citation1993) does recognise the importance of agency. However, Woods (Citation2015, p. 7) argues that ‘little room is left within this puzzling-powering account for “powering” that is ideational/discursive in nature, […] this potential is not explicitly recognized’.
3 First, documentation for the budget years 1996, 2001, 2006, 2011, 2015 and 2021 was consulted. Thereafter the years in between (and before 1996) were rapidly reviewed to find and date alterations in positions. Special attention was given to election years and years following changes in government.
4 The number of recipient countries in 1991 is stated to have been 91 in the budget bill for 1993/94 (prop 1993/94:100, bil 4: 120). However, in the OECD/DAC (CRS) data, there are only 63 countries reported to have had an inflow of Swedish aid in 1991.
5 SIDA (the Swedish International Development Authority) and three other agencies were in 1995 merged into a new agency, named Sida (the Swedish International Development Cooperation Agency).
6 However, the Moderate Party, in 2018 and 2019, argued for reaching the one per cent not before 2022.
7 To complicate things even further, there are some different accounting principles between DAC and Sweden leading to differences between actual amounts reported to DAC and domestic statistics. For example, replenishments to multilateral institutions are recorded in DAC at full value when the promissory note is issued, whereas Swedish budget outcomes use annual payments, see EBA (Citation2018).
8 It should be noted, however, that the crowding out mainly took place in the form of a non-expansion of the aid budget (an automatic result of the one per cent target and a growing economy) rather than in reductions of the budget. A particular problem is amendments to the central government budget during the financial year in order to meet higher or lowered costs for asylum seekers in Sweden due to differences in forecasted versus realised numbers of asylum seekers.
9 Multi-bi aid may be allocated to specific countries or for global purposes (public goods) but is defined as bilateral aid in the ODA-statistics.
10 An overview and discussion of the funding of UN funds and programmes, with specific focus on Sweden, is provided in Browne et al (Citation2017).
11 Most multi-bi supports (around 60 per cent each year) have been disbursed to individual countries, between 10 and 20 per cent has been disbursed as regional aid, and the remaining 20–30 per cent per year as global (‘Bilateral, unspecified’ in data).
12 While these figures suggest direct state to state cooperation to have decreased, cooperation with states can also occur through multi-bi support as well as through some channels in the ‘other’ group (including public-private partnerships, networks, universities, think-tanks and the private sector).
13 Of this, the largest channel is international NGOs, followed by NGOs in Sweden (around 30% in 2019), and in developing countries (less than 10%).
14 Note that these are shares of country allocated bilateral aid. Thus, the denominator does not include all bilateral aid, such as multi-bi aid for global purposes.
15 Descriptions of Sweden’s strategies and how they were/are constructed are given in Government Offices of Sweden (Citation2017; Citation2013).
16 Nicaragua 2003 and Honduras 2004 (included in the regional strategy for Latin America); Pakistan and Haiti in 2010 (humanitarian assistance).
17 If aid through NGOs is excluded (not being formal cooperation countries), the concentration policy appears to have had some effect on proliferation. Still, even in this case proliferation increased after 2009 but the composition of Sweden’s bilateral aid did change
18 ‘The [previous] six subgoals … tended to overshadow the primary goal of poverty reduction. The effects of this were that poverty reduction was not able to serve adequately as either the focus or the goal of development cooperation … , despite the fact that this was supposed to be the primary goal. In order to emphasize that poverty reduction is the main goal of development cooperation, the Government considers a poverty-oriented goal to be a more effective policy instrument than a division into several subgoals’ (Gov. Bill. 2002/03:122, Sect. 6.2).
19 See Niño-Zarazúa et al (Citation2020) for a recent study on the effects of democracy aid.
20 New Democracy (Nyd) was a right-wing populist party that had parliamentary representation during one term, 1991–1994, proposing large reductions in the aid budget.
21 The counter proposal gained majority so the social democratic minority government will govern on that through 2022. The aid budget however did not contain any amendments from the government, except for a point stating that ‘One billion SEK of aid is earmarked for reconstruction of war-thorned countries to enable people to again reside and live in safe societies and contribute to its growth’. The three parties have different interpretations as to which this means an earmarking to already existing priorities or if it’s a ‘return-billion’ to facilitate return migration from Sweden.
22 Each party’s deviation from the aid frame is not reported, so the figure shows the counter proposal for the aid budget (expenditure area 7). In some instances, the proposal may appear counter intuitive. For example, in 2016–2018, a period when large shares of ODA were allocated to migration expenditures in Sweden, the Sweden Democrats, still proposing a lower total aid frame, proposed a higher aid budget than the government.
23 In a similar vein, Larue (Citation2022) uses the largest positive and negative counter proposals (as weighted by the number of seats in the Riksdag) to create a measure of the ‘conflict intensity’ in the parliament since 1971.
24 See Bolin and Larue (Citation2016) for an analysis of the use and relevance of such resolutions in a wider context.
25 A partial explanation for this, as noted in Larue (Citation2022) may be that changes in the budgetary process have moved some issues of potential conflict from the parliament to the government. An example (as noted in the section on first-order changes) is the replacement of yearly country frames (decided by the parliament) to multiannual country strategies (decided by the government) in the late 1990s.
26 Of the eight coalition governments since 1956, in only two cases have the aid minister been from the same party as the prime minister, in 1954–1966 (Ulla Lindström, in a Social Democratic-led coalition with the Farmers’ League, now the Centre Party) and in 2006–2014 (Gunilla Carlsson and Hillevi Engström, in the centre-right coalition led by the Moderate Party).
27 It may be discussed whether the ‘business as usual’-answer in respective series of polls (in particular the answer in the SOM-data) should be coded as a positive position. However, available data from Sida’s polls before 2001 have merged the ‘about right’ answer and ‘should increase’, so comparability over time and between series are arguably less compromised by this aggregation. Also, the shapes of the SOM-series excluding or including the ‘neutral’ response are similar.
28 ‘Preserve’ in 2018 counts the Moderate Party as negative to the one percent target, which is the position held from 2020.
29 I consult the Media archive of the media monitoring company Retriever to search for the yearly number of articles from Sweden’s largest news agency (Tidningarnas Telegrambyrå, TT) related to Sweden’s development cooperation during the years 1990–2020.
30 According to Grafström and Windell (Citation2019), Swedish media reporting on foreign aid are predominantly neutral and driven by events. Even without a thorough analysis, it would therefore not be too daring to guess that the local maxima can be explained by the abandonment of the one per cent target (1992), the Paris agenda and discussions on general budget support (2005) and related to migration (2015).
Additional information
Notes on contributors
Jan Pettersson
Jan Pettersson is the Manager Director of the Expert Group for Aid Studies, EBA (www.eba.se). He holds a Phd in Economics from Stockholm University and has been a research fellow at Uppsala University and Stockholm University.