Charitable food aid in Finland: from a social issue to an environmental solution

Since the establishment of the first food bank in 1995, charitable food aid (CFA) has become entrenched in Finland as a seemingly irreplaceable solution to food poverty. Further, it has recently been suggested that the focus of food aid activities is shifting from food poverty and temporary hunger alleviation towards environmental sustainability through addressing food waste via organized re-distribution of expiring food from retail to charitable organizations. This potentially creates a mechanism that (1) solidifies food poverty and (2) fortifies the paradoxical situation where charitable organizations delivering food aid are dependent on food waste rather than trying to reduce it. To understand the process that has led to this shift, a longitudinal media data analysis on the evolution of the discussion and the interpretations on CFA is presented. By conducting an inductive frame analysis, the paper answers three key questions: How was CFA framed by and through the media in Finland between 1995 and 2016? Has any single frame dominated the discussion at any given point? Finally, what are the characteristics of the frame that focuses on food surplus redistribution? The results suggest that when the practices are framed as potential receivers and redistributors of surplus, perception of CFA is mainly favourable and the root causes for food insecurity are not addressed. Thus, by focusing on environmental sustainability, food aid practices—hitherto depoliticized as a poverty problem—have gained policy relevance in the discursive space of the circular economy; perhaps at the cost of poverty policy and with unintended consequences.


Introduction
The recession in the 1990s served as a starting point for charitable food aid (CFA) in Finland. As the welfare state was unable to meet or even acknowledge the rising number of food insecure people, it was civil society organisations, most notably the Evangelical Lutheran Church of Finland (ELCF), which took action. These practices were initially thought of as temporary, which would be made redundant as the economy began to grow again. This notion was later proven wrong as recipients continued to rely on charity, despite economic recovery. (Lehtelä and Kestilä 2014;Ohisalo et al. 2014;Silvasti 2011Silvasti , 2014Silvasti , 2015Silvasti and Karjalainen 2014.) More recently, Finnish food safety guidelines were loosened, paving the way for the redistribution of surplus food from the retail and food industry to food charities (EVIRA 2013;Lehtelä and Kestilä 2014). This is in line with the EU lead effort to reduce food waste and promote the circular economy (CE) thinking (European Commission 2017; Sitra 2016). As a result the supply of food for charities increased, and food aid was coupled directly to 1 3 reducing food waste (Kortetmäki and Silvasti 2017;Silvasti and Salonen forthcoming 2019).
The implementation of the redistributive food bank model Shared Table in Vantaa (Yhteinen Pöytä 2016) and the widespread national interest towards the model indicate that the activities-hitherto organized primarily by scattered NGOs, faith-based organisations (FBOs) and ELCF-are expanding into an organized arrangement comprising of multiple actors working together in a circular manner. Simultaneously, the interpretations and meanings associated with CFA are in transition, as the practices are being restructured and organized anew within the context of a Nordic welfare state. To better understand the process that has taken place, a longitudinal analysis on the evolution of the discussion and on the interpretations produced in this discussion on CFA is necessary. Therefore, this article explores the interpretations and relevance given to CFA in the media by analysing the interpretative frames utilized in the largest subscription newspaper in Finland, Helsingin Sanomat (HS), during 1995-2016. For the last two decades CFA in Finland has been the subject of study both during the deep economic recession of the 1990s and the more recent recession of 2009. These studies have ranged from ethnographic studies focusing on the receivers of food aid (e.g., Karjalainen and Järvinen 2000;Salonen 2016a, b) to more theoretical analysis of the ethical connotations embedded in the concept of first world hunger (Kortetmäki 2015). Beyond Finland, studies on the growing dependence on CFA practices in rich countries are plentiful 1 and consist of both critical (e.g., Booth and Whelan 2014;Livingstone 2015;Riches 2002Riches , 2011Wells and Caraher 2014; for a synthesis on critical literature see; McIntyre et al. 2016) and more favourable viewpoints-especially when coupling CFA with waste reduction (e.g., Hebinck and Villareal 2016;Hebinck et al. 2018;Santini and Cavicchi 2014). Looking at the more critical views, an essential aspect of CFA research is the concept of entitlement; food aid is not something one is entitled to-as opposed to official welfare services-but is a gift (Riches and Silvasti 2014). For an individual, being dependant on CFA practices translates to losing "the freedom of choice and inherent human dignity, because they have to accept charity food in spite of their actual needs and preferences" (ibid., p. 9). According to Poppendieck (1998, p. 5) the emergence and prevalence of food aid reveals a larger shift in a society: the underlying abandonment of the fight against poverty-"the end of entitlement," as the book subheading states-and a shift towards "damage control rather than prevention." Silvasti (2014, p. 10) goes on to add that this shift moves the initial aim of poverty reduction to "the margins of social policy because well running successful charity work diminishes the pressure on the political system" (see also Dowler and O'Connor 2012;Riches 2002Riches , 2011Winne 2008).
This paper aims to unfold the ways in which food aid has been framed in Finnish public discourse during its 20-year lifespan and identifies the key stakeholders involved and the temporal and normative dimensions of these framings. This also allows the paper to prove whether a single framing has been dominant at any given time. Finally, as the emerging CFA paradigm seems to be "food waste redistribution as a social innovation" (e.g., Hebinck et al. 2018), the paper especially focuses on this framework. Thus, the research questions are three-fold: firstly, how has CFA been framed during the 20-years of discussion, and by whom?; secondly, has any single frame dominated the discussion at any given point?; and finally, how is the framework focusing on food redistribution utilized in the timeline and what are the specific characteristics of the frame? The article is organized as follows: first, the evolution of Finnish food aid is presented in detail, which is followed by a section focusing on CFA and food surplus; second, the data and methodology, along with a few words on frame and framing as concepts, are presented; third, the findings are structured in three sections (overview of the frames, in-depth look into selected frames and a longitudinal analysis of the evolution of framings); and finally, the results are elaborated on in the contexts of Nordic welfare regime, sustainability and circular economy.

Two decades of food aid in a welfare state
Finland is widely considered a Nordic welfare state, where welfare services are arranged according to the ideal of universalism (e.g., Silvasti 2015; Kortetmäki and Silvasti 2017). As a social-democratic welfare regime, poverty relief should rely first and foremost on the state, as opposed to the market or civil society (e.g., Greve 2015), but the very existence of CFA within this context challenges this notion. In addition to those relying solely on basic social security, one should also consider those living on the edges of being poor-the working poor, students, pensioners etc. As Poppendieck (1998, p. 57;also Riches 1997) reminds us, "food is often the most flexible item in the family budget, the place where you can economize," which implies that in the case of sudden financial turmoil food is the easiest to cut back on.
The first signs of Finnish food insecurity emerged in 1993, when in a survey published by the Ministry of Social Affairs and Health it was estimated that roughly 100,000 Finnish people were hungry during 1992-1993 (Kontula and Koskela 1993). Simultaneously, ELCF arranged hunger relief via "breadlines," which would become the emblem of poverty for years to come (Karjalainen 2008). The first food bank in Finland was established in 1995 in Tampere (Silvasti 2011). Simultaneously, Finland joined the European Union. MDP was applied for in the same year, and the first shipment of aid 2 was received in 1996 (Karjalainen 1997;European Commission 2016a). The global financial crisis initiated a slump in the Finnish economy in 2009 (Tilastokeskus 2009). In 2013, MDP was phased out which led to a temporary disruption in the aid, as the transition period to FEAD lasted roughly a year (Lehtelä and Kestilä 2014;European Commission 2016a;Silvasti and Karjalainen 2014). FEAD supports member countries' actions to provide material assistance to those in need -thus expanding the aid from mere foodstuff to, e.g., clothing. However, the Finnish Government decided to stick to food aid exclusively. (Lehtelä and Kestilä 2014;European Commission 2016a;European Commission 2016b.) In this period, the Huono-osaisin Suomiproject produced the first estimates and reports on modern CFA in Finland (e.g., Ohisalo et al. 2014).
The Finnish Food Safety Authority, EVIRA, loosened the food safety regulations in 2013, which allowed the retail sector and food industry to donate foodstuff to charity more easily (EVIRA 2013;Lehtelä and Kestilä 2014). In 2015, influenced by the Berliner Tafel-foodbank network that incorporates actors from civil society as well as the private and public sectors with the aim of maximizing food waste utilization, the city of Vantaa teamed up with retailers and CFA organizers in creating the Shared Table-model (Yhteinen Pöytä 2016). The emergence of this model coincides not only with other similar practices taking hold within the EU 3 and the discussion of these practices as social innovations (e.g., Baglioni et al. 2017;Hebinck et al. 2018;Tyink 2016), but also with the current Finnish government's wider agenda to promote CE and sustainability (Valtioneuvosto 2015;Sitra 2016). 4 CE is a highly contested concept (e.g., Winans et al. 2017;Korhonen et al. 2018a, b;Prieto-Sandoval et al. 2018). The data utilized in this paper does not specifically mention CE and so does not participate in the debate on the definition of CE. However, CFA has been interpreted as part of food waste redistribution in Finland (see Kortetmäki and Silvasti 2017;Silvasti and Salonen forthcoming 2019), and food waste redistribution can be considered as a resource lifeextending strategy (RLES) (Blomsma and Brennan 2017, p. 606). Also, as CFA is connected to CE through national keywords such as "resource wisdom" (e.g., Linkola 2015; Ratschinskij 2016) the redistributive practices are discussed within the discursive space of CE (see Blomsma and Brennan 2017).

CFA today and food waste
In a recent report Ohisalo et al. (2014) describe Finnish CFA as a "patchwork quilt of aid"; unlike in for example the United States or Canada (Silvasti and Karjalainen 2014), 5 Finland does not have central charitable organizations to coordinate food (re)distribution, but the field of CFA consists of over 400 distributors-including parishes, FBOs, unemployment organizations and other NGOs-operating in various ways (ibid.; Ohisalo et al. 2014). There is no official estimate of the number of recipients frequenting CFA practices, but Ohisalo (2014, p. 40) estimates that over 22,000 people turn to CFA every week. Most CFA organizers hand out primarily foodstuff stemming from FEAD that is supplemented with donations and purchases, though there are organizers that rely solely on donations or FEAD. The origin and level of food donations vary between regions, but the most notable source is the retail sector. Instability and fluctuation of food waste in retail and therefore food donations are especially problematic for those practices that rely heavily on donations and do not possess the necessary financial means to buy food from other sources. .
As Riches and Silvasti (2014, p. 8) state, the "contradiction between unrestrained waste, ecological unsustainability and growing food poverty is ethically intolerable"-a view that has led to a paradigm where CFA is seen as a link between food waste and food insecurity. In a report issued by the Nordic Council of Ministers (Hanssen et al. 2015) the redistribution of food(waste) through food banks was a significant untapped potential in the Nordic region, and in Finland the proposed legislation that would obligate retailers to donate edible foodstuff to charities has widespread support throughout political parties (LA 29/2016). The proposed legislation has met with significant opposition from retailers, who pointed out that voluntary donations are already established and working well-especially as some of the CFA organizers are small and run with volunteers and are thus unable to handle the sudden influx of donations (Kärppä 2016). While it is true that the retail industry in Finland produces a staggering 65-75 million kilos of food waste annually (Silvennoinen et al. 2016), and that food insecurity is clearly present in Finnish society, the redistribution of surplus food from the food supply chain to CFA has been contested as a sustainable solution to both poverty and environmental problems (e.g., Caplan 2017;van der Horst et al. 2014; in Finland see; Kortetmäki and Silvasti 2017; Silvasti and Salonen forthcoming 2019).

Frame package analysis
Framing, as stated by Entman (2004, p. 5), is a process of "selecting and highlighting some facets of events or issues, and making connections among them so as to promote a particular interpretation, evaluation, and/or solution." Furthermore, by highlighting certain aspects, framing is also inherently about shaping what will be left out of the discussion-which, in the case of media frames, leads to a limited set of viewpoints from which the issue may be considered (Altheide 2002;Graber 2010;Park et al. 2013). When analyzing media frames one must also note that although objective and fair reporting on a given subject might be the ultimate goal and key value of the media, journalists cannot not partake in the framing process; writers are always making "subtle alterations in the statement or presentation of judgement and choice problems" (Iyengar 1991, p. 11), and therefore they never just passively relay information (Delli Carpini 2005;Entman 2005).
The aim of the analysis is to identify and reconstruct frame packages: characterized by Van Gorp (2007, p. 64) as "a cluster of logical organized devices that function as an identity kit for the frame." Frame packages are comprised of a core frame, framing devices and reasoning devices (Van Gorp and van der Goot 2012). This paper utilizes these devices in a longitudinal analysis (see Van Gorp 2005). The core frame is the cultural phenomenon that defines the frame package (e.g., "CFA is seen as a national shame" or "CFA signifies a change in welfare responsibilities") which is manifested in media content through the use of framing devices and reasoning devices. Framing devices are conceivable elements, such as word choices, visual imagery, metaphors, arguments and examples. Reasoning devices are the explicit and implicit justifications, assumptions and interpretations that "lie hidden" (Candel et al. 2014, p. 49) within the text. The reasoning devices can form a causal interpretation, treatment recommendations or proposed solutions, but this paper is chiefly interested in the problem definitions and moral implications of CFA (Van Gorp 2007; Van Gorp and van der Goot 2012; Ortiz-Miranda et al. 2016). A slightly lighter take on the reasoning devices is justified here as the media discourse on CFA is often cursory and shallow and therefore ill-suited for such in-depth analysis as suggested by Van Gorp and van der Goot (2012), and the focus here is on a rather particular practice as opposed to a larger theme.

Data and analysis
The data was gathered from the digital archive of Helsingin Sanomat (HS). HS is the largest newspaper in Finland and, more importantly, it is distributed nationwide, thus offering a wide range of discussion, opinions and news on CFA. Data collection and preliminary analysis were exploratory and hence reminiscent of the grounded theory variant of Corbin and Strauss (2008) as interpreted by Reichertz (2009); inductive analysis and coding of the dataset began immediately at the start of the collection of data and were not directly influenced by any specific theory, though one must admit and acknowledge his/her own standpoint-and consequently the biases concerning data collection and analysisand rely on sensitivity rather than objectivity (Corbin and Strauss 2008;Van Gorp and van der Goot 2012).
Sensitivity, in this case, meant that throughout the collection both the preliminary coding and initiative thoughts, as well as the procedure as a whole, i.e., optimal search terms, which sections in which timespan were to be included in the search etc., were under constant reflection. The process was also regularly recorded in written memos; a habit that continued into the analysis phase. This proved vital as the collection progressed, as for example some of the units classified as secondary in the early stages of collection were later re-read and re-coded as primary data. By including secondary units that initially seemed inadequate for analysis, it was possible to backtrack and re-code these units as the process became more focused.
En masse, the search produced 1461 hits with the timespan set between January 1995 and the then present time, June 2016. Out of this exploratory sample, following the exclusion of articles and other texts on foreign food aid or other topics not specifically related to Finnish CFA, 652 units were then downloaded and imported to a qualitative analysis software (Atlas.ti) for analysis. After a preliminary reading of the dataset, a further 123 units were discarded for being ill-suited for analysis-including short news stories with only a brief mention of the topic, causeries and other texts where humour or other forms of expression were used extensively etc.-resulting in a final set of N = 529.
The analysis started off with a simple read-through and preliminary coding of the dataset. One unit of analysis could include multiple codes, but for the unit to be included in further analysis a minimum of one set of the following was required: the tentative frame or framing/reasoning device (code: F_"frame"), the normative perception of food aid (code: Norm_neutral/positive/negative) and the stakeholder(s) utilizing the frame or framing device (code: SH_"stakeholder"). Text units varied from short texts with only one code cluster to longer articles with a maximum of 11 clusters. The preliminary coding produced several tentative frames, which were then combined to prevent unnecessary overlapping, 6 resulting in 11 frameworks for CFA.

Results
Next, the following are presented: a short overview of the 11 identified frames including more in-depth descriptions of the most prominent frame and the food redistribution frame, and then a longitudinal analysis of the evolution of framings in the media.

Overview of the frames utilized 1995-2016
The most prominent frame by far was "CFA as an emblem of poverty," where CFA practices-usually breadlines-were used in a rhetorical way when referring to poverty in Finland in general: We are shown pictures of two kinds of queues: stockbrokers are lining up on the streets, while behind the corner the breadlines of charitable organizations continue to grow. As the dream of striking rich lives on, elsewhere people worry about their daily bread. (Helsingin Sanomat 15 December 1999, Vieraskynä, translated by author.) As the framework was used in a rhetorical way, the text units were usually not concerned about CFA per se but rather about poverty or income differences. The framework addresses the multifaceted phenomenon that is CFA purely as an outcome of poor income, though never delving deeper into what the mechanisms of poverty could be. Since the framework consists mostly of mere mentions of CFA, namely breadlines, it was at first thought best to leave this frame out of the analysis. Eventually, as the coding of the data continued, it was clear that the frame would become the most prominent way of addressing food aid in the data, and therefore in the end it was included in the analysis.
"CFA, surplus and CE" frame-often referred to as the most prevalent frame in recent years (e.g., Ohisalo 2015)sees the potential in CFA inasmuch as the practices redistribute food surplus. Though not explicitly referring to CE, the frame uses similar framing and rhetoric devices-word choices, justifications etc.-as in the national CE discourse. The focus was shifted from the root causes of poverty and food insecurity to the issue of excess in the food supply chain: Ten years ago the Salvation Army observed how edible bread from the shops was either thrown out or given to the pigs to eat. If the demand should cease, fully edible bread would once more become fodder for pigs. This was also the frame utilized prominently by the private sector. These actors framed CFA by saying that by providing the practices with their surplus they were participating in the "common good" and that the arrangement is mutually beneficial: Bread and ready-meals are not allowed to go bad on the shelves but are collected a day before the last selling date, packed in boxes and collected every morning by the parish of Myllypuro to be distributed to those in need. If the [foodstuff] were to be transferred to landfills, it would mean a lot of work. 'The food would have to be removed from their packaging into compostable waste and the packaging sorted separately. Now we are spared this work.' (Helsingin Sanomat 7 February 2005, Kaupunki, translated by author.) As the framework does not address the root causes of food insecurity, the transitions in welfare responsibilities or other criticism of the aid, the normative perception of CFA was predominantly positive (see Table 1). Arguably, even though CFA has been widely used as a rhetorical device and as an emblem of poverty in Finland, it was only when the practices were linked (in a positive manner) to the issue of food waste that CFA gained policy relevance, culminating in the proposed food waste prevention legislation. According to Hajer (1995, p. 22) "[t] he regulation of a problem first and foremost requires forms of discursive closure: the problem needs a definition that gives policy-making a proper target," which suggests that food insecurity was not a sufficient, tangible or distinct issue, whereas food waste is more easily accepted as a policy-making target. Furthermore, food surplus has brought together various actors and concepts, bridging  (96) This framework specifically addresses the causal relationship between CFA and for example housing politics, the deep economic recession and substance abuse. The frame was often used in tandem with the Change in welfare responsibilities frame, as the reason for participating in CFA practices was often associated with the state's inability to provide basic social security Media (86) ELCF (18) Positive (9) Neutral (58) Negative (20) N/A (9) Change in welfare responsibilities (69) The frame considers the relationship between CFA practices and the state, especially through the transition of welfare responsibilities from the state to the church and the civil society. The frame discusses CFA in the context of the welfare state, thus addressing the notion of whether or not CFA is justifiable in a Nordic welfare regime Media (22) ELCF (17) Politicians (10) Positive (4) Neutral (30) Negative (35) CFA as a national shame (65) In this framing CFA was seen as a national disgrace both by those who considered it to prove the existence of food insecure people in Finland and, interestingly, by those who considered that there is no hunger problem in Finland -at least in comparison, for example, to Biafra or North Korea Other (28) Media (16) Positive (4) Neutral (5) Negative (39) N/A (17) "Heart-rending human stories" (50) An example of framing by the media. The title of the framework comes from Jamrozik and Nozella (1998, p. 71) where they describe how "[s]ocial problems are vividly portrayed by presenting them as 'heart-rending human stories' or 'human catastrophes', described in emotive terms --." N/A N/A CFA, surplus and circular economy (46) Described below Media (17) Private sector (10) Positive (26) Neutral (19) Negative (1) The purpose of the aid (46) The question of who is eligible for aid and who is not is at the heart of this framework, as the focus is on the function of the aid. On the one hand, CFA was seen as a socially inclusive practice that could even employ or otherwise keep recipients busy through volunteer work. On the other hand, exclusion of some participants and the recipients' growing reliance on the aid was seen as highly problematic Media (17) ELCF (11) Positive (10) Neutral (25) Negative (11) The purpose of the church ( together the issues of waste management and hunger in a manner that has resonated in the political discourse as well. This frame also coincides with institutional arrangements that suit the framing, as the Shared Table -model is being piloted across Finland .

The evolution of the frames: a longitudinal approach
To see whether a single frame has dominated the discussion, or if there are certain time periods where some frames are  (25) This frame revolved around the negative and positive side-effects of CFA, though chiefly focusing on the negative side. The positive effects overlap with the purpose of the aid framework, but the negative sideeffects (violence and racism in queues, reduced value of estates near the breadlines and the "not in my backyard" phenomenon) were exclusive to this frame Other (10) Media (8) Positive (12) Neutral (11) Negative (2) The agricultural framework (15) During the period of MDP, CFA was also framed within an agricultural framework, as the EU-aid has its roots in EU-level agricultural policy Media (10) Positive (0) Neutral (9) Negative (6) "CFA as an emblem of poverty" did not focus on the CFA per se and is therefore left out of the "Normative perception of CFA" column, whereas it was decided that it would be best to leave "'Heart-rending human stories'" out of the stakeholder and normative perception columns as they were extremely difficult to code Note: Smaller frames ("Side-effects" and "The agricultural frame") have been combined into "Other." more prevalent, a longitudinal presentation of the utilization of frames is presented. Figure 1 aims to show how "rich" the discussion has been, i.e., what variety of frames has been utilized at any given point. This allows not only the presentation of the evolution of specific frameworks, but also the peaks of media coverage are easily and more exactly visualized (see Boydstun and Glazier 2013). As can be seen from the figure, the discussion has encompassed a multitude of frames throughout the timespan, clearly showing that the discussion has not been dominated by a single frame at any point. Regarding the spike in circa 1996, the organization and contents of the aid were the focus of discussion. It is safe to say that this was due to the novelty of CFA practices and the sluggish bureaucracy and subsequent delays in EU food aid. Between the spikes of the mid-1990 s and 2000s, CFA continued to be used in a rhetorical way, though there were fewer text units in the dataset at this point. During the turn of the millennium the discussion focused on the consequences of the economic recession, i.e., CFA was framed as an outcome of the cutbacks and austerity measures implemented during the deep economic recession. This supports the notions that even after the national economy began to grow and the economic recession came to an end, the impact of the period continued to affect social policy and the underprivileged well into the 2000s (e.g., Lehtonen and Aho 2000). This period also marks the first time that the framework "CFA, surplus and CE" was effectively present in the data, though some brief mentions of food waste redistribution date back to the very beginning of the time period.
The next spike coincided with the beginning of the financial crisis of 2007-2008, though the Finnish economy was not hit by it until 2009. During this period, the use of CFA as a rhetorical device was frequent, though there was an increase in other framing as well. From 2013 onwards, "CFA, surplus and CE" became a relatively large framework in the discussion, which suggests that the changes made to the food safety regulations, the subsequent rise of donated foodstuff and the proposed waste prevention legislation have since caused a resurgence of this framework. With this dataset it appears that the discussion on CFA is anchored around economic downtrends and cutbacks in social policy, with media attention given to CFA increasing at periods of economic recession. Especially interesting was the sudden drop in discussion between the two economic recessions, though it was impossible to indicate a cause for this sudden drop in the data. CFA practices certainly did not cease during this period, but they clearly lost some media interest at the time.

Stakeholders and the normative perceptions of the frames
A shortcoming of the data was that it was troublesome to reliably differentiate between journalists' interpretations of interviews and the actual stakeholder interviews, i.e. framing by and through the media. Some tentative considerations could, however, be made, though for the most part the frames were coded under Media or Other (individual writers in Opinions, if the affiliation or role of the writer was not explicitly stated). The overall distribution of frameworks between stakeholders is presented in Fig. 2.
A noteworthy observation was the overall poor representation of the CFA organizers in the data, with the added share of NGOs, FBOs and ELCF not even totalling a fifth of the data. Furthermore, the nigh complete absence of the National Organization for the Unemployed (present in four text units) as well as the overall lack of NGOs in the data were interesting observations. According to the survey by Ohisalo et al. (2014), the NGOs, FBOs and ELCF each roughly amount to a third of the CFA organizers in Finland. Therefore, the NGOs, FBOs and ELCF were somewhat unevenly represented in relation to each other. Whatever the reason for this, the lack of representation from the most invested stakeholders-perhaps apart from recipients, who were also absent from the data-is problematic, as their hands-on knowledge and understanding of CFA is not fully utilized in the current discussion.
Looking at the normative perception of CFA, i.e. whether CFA was seen as positive, neutral or negative within the framework, some intriguing results arise in Table 1. First, observing the frames where the normative perception tended to lean towards negative, two frames stood out: "Change in welfare responsibilities" and "CFA as a national shame." The overarching theme in these frames is that both frameworks view CFA first and foremost within the context of the Finnish welfare state-and therefore discuss it in tandem with universalist social policy. This is also true for the framework "CFA as a result of poverty," though this frame tends to be utilized more neutrally. Second, when discussing CFA within the framework of "CFA, surplus and CE," the normative perception tended to be predominantly positive.

Fig. 2 Utilization of frames by stakeholders
The same applies for framework "Side-effects." Overall, the normative perception was chiefly neutral-which is partially explained by the large proportion of framings by the media, as news stories tend to favour a neutral approach.

Discussion and conclusion
Frame package analysis of the media data associated with CFA in Finland produced 11 interpretational frameworks during the period of analysis . The debate is not dominated by any single frame at any given point in the timeline, but it is clearly centred on the two recent recessions and on the subsequent impacts of these recessions on social and poverty policies in Finland. In most text units CFA is used first and foremost as an emblem of poverty. The focus is not on CFA practices, but rather food aid is used as a symbol or visual cue of poverty. Utilizing CFA as a rhetorical device, the framework usually dismisses the root causes of poverty.
Exploring the timeline through the lens of Nordic welfare regime, where the state is responsible for providing basic security for the citizens, CFA practices can be seen as openly challenging this presumption of state responsibility by providing the basic need for food. This transition in the welfare responsibilities is contested in the 1990s by the church and more widely during the aftermath of the cutbacks of the 1990s recession and during the economic slump of 2009. However, towards the end of the analysis period, the outrage is not as vocal as before. This implies that the transition from the ideals of a Nordic welfare state to the ideal of "neoliberal thinking" that stresses individual responsibility of social security, which is then "supported by charitable work of nongovernmental organizations," as suggested by Ferge et al. (1997, p. 35), has become accepted in the organization of CFA. Eventually, in shifting the state's welfare responsibilities to CFA practices instead of raising the level of basic security, poverty as a phenomenon is narrowed down to just hunger and thus the root causes of food insecurity are dismissed. Leading Finnish researcher on social policy Raija Julkunen (2006) also states that "[a]lthough production of services in NGOs is paid, professional work, they cannot be obliged to provide aid or produce services" (ibid. 217).
There is a noticeable transition to the focus on food waste in the media discussion during the 2010s, though perhaps not as prominent as expected. However, the data presents only a fragment of the public discussion, and hence a broader dataset is needed to elaborate this notion. Nonetheless, according to the analysis there is a small but noteworthy transition in the focus of the discussion from a poverty problem (social problem) to reducing food waste (environmental problem) during the 2010s. The importance of this transition lies in its ability to act as a tentative mechanism that supports transition from state welfare responsibilities to a more liberal welfare regime. In media discussion the use of framings that highlight the positive aspects of CFA, like reducing food waste, promotes dismissing the more critical notions of food aid as a means of poverty policy in a rich Nordic welfare state. Operating within the discursive space of CE, CFA and food waste reduction together construct a framework in which the normative perception of the practices is positive by default as they are seen at the same time as a solution to the shameful overproduction and food waste in the supply chain as well as food poverty. Meanwhile, the underlying social issues that lead people to CFA in the first place are easily ignored. By focusing on food waste management instead of poverty, the frame justifies the institutionalisation of CFA practices from a new perspective, rooting their existence firmly to the end of the food supply chain-as a crucial cog in the CE machine. Indeed, within the data, the role CFA practices play in food waste reduction was used as an argument against the disbanding of an individual CFA practice by the organizers (Laita 2015).
Food waste is unquestionably a challenge that must be addressed post-haste. CFA's redistributive element has been offered as a win-win situation (Hebinck and Villareal 2016;cf. Caplan 2017), as it is mutually beneficial in environmental and economic terms in providing food and nutrition security. But is it truly a win-win(-win) situation after all, i.e. addressing overproduction, waste management and food insecurity? Environmentally speaking, according to Babbitt (2017) by looking at the sustainable food waste solutions the most preferred solution should be source reduction rather than redistribution. Redistribution also creates an environmentally questionable situation, for as Valkonen et al. (2017) point out there is a possibility that this will lead to a paradox where "that which has been too much of, becomes something we have too little of" (ibid., 23); that within the Finnish food system, there operates a scheme that emphasizes reutilization rather than reduction of waste, i.e. one that relies on waste production. Kortetmäki and Silvasti (2017, p. 228) argue that "[t]he unforgivable wastefulness and resulting inefficiency as well as the social and environmental injustice of the food system cannot be solved by organizing a 'secondary food market' or 'charitable food market' to distribute residual food for free to citizens --." The answer to food waste lies thus in the systemic level of the food supply chain, rather than in creating a loophole for the surplus. By promoting food redistribution from the primary market (the retail sector) to the secondary market (CFA), the process is not only marginalizing the root causes of food insecurity (housing policies, income differences etc.) but is also supporting and enabling unsustainable food production (Salonen 2014;Kortetmäki and Silvasti 2017). Kortetmäki and Silvasti (2017, p. 231) conclude that "waste-based [CFA] promotes both social and environmental injustice and therefore cannot have any justified task or position as a part of poverty relief under an eco-social Nordic welfare regime." So, while it is true that in the present situation some overlap is unavoidable, it is a question of how much overlapping, in the context of a Nordic welfare regime, do we deem appropriate and justified-a question to which the answers range from none in Finland (Kortetmäki and Silvasti 2017) to rather more substantial amounts in other regions (e.g., Hebinck et al. 2018). Finally, perhaps the most important element is how irrevocably dependant these practices and the receivers of aid become on the overlap; to not become a sinkhole for surplus, a mechanism that removes the incentive for food system actors to actually look into waste source reduction, rather than just re-routing surplus to charity in a circular manner.
In conclusion, the context in which CFA practices operate is at an interesting stage. The (re)framing of CFA as a CE model or social innovation set to resolve both food insecurity and food waste marks a significant change in the way CFA practices may be organized in Finland. The emerging Shared Table -model incorporates not only the civil society and ELCF, but also actors from the private and public sectors, which can be seen a signal that the practices need no longer operate outside the official social policy; organized in tandem with municipalities, cities etc., the new model of CFA involves an impressive transition in welfare responsibilities. This shift in responsibility from the state to the church and civil society deserves to be scrutinized further. The main limitation of the paper is the narrow scope that media data portray: the individuals working in CFA were underrepresented, and the discussion was often cursory and shallow. A larger dataset, broadening the sample from media to policy documents and further, and the inclusion of interviews from stakeholders would allow the elaboration of the frames presented and could also provide evidence of how these frames have translated into policy actions.