A food systems approach has the advantage of creating awareness about synergies and trade-offs between policy domains which have historically often been treated in isolation, including interactions which spill across international borders. This chapter shows how policy makers can design coherent policies for food systems through better coordination across policy communities. Documenting and quantifying potential spillover effects is an important first step. Where synergies are found, one policy instrument will rarely be sufficient to meet all objectives; rather, a mix of instruments is usually needed. When trade-offs are found, experience shows that these can often be avoided by a smarter choice of policy instruments. But when trade-offs persist, the question is how society should choose between competing objectives. While those choices need to be based on the best possible evidence, they involve value judgments and need to be made in a way that commands broad support across society, is consistent with international obligations, and effectively addresses the triple challenge.
Making Better Policies for Food Systems
2. Principles for policy coherence
Abstract
Key messages
A food systems approach creates awareness about synergies and trade-offs across the different dimensions of the triple challenge. Synergies occur when making progress on one policy objective makes it easier to make progress on another. Trade-offs occur when making progress in one area leads to worse outcomes in another area.
Coherent policies require increased coordination between different policy making communities (e.g. agriculture, fisheries, environment, public health), so that various policies are aligned to strengthen each other, or at least do not counteract each other.
Documenting and quantifying potential spillover effects is an important first step. Not all hypothesised spillovers are large enough to matter for policy design.
Where synergies are found, one policy instrument will rarely be sufficient to meet all objectives; rather, a mix of instruments is usually needed.
When trade-offs are found, experience shows that these can often be avoided by a smarter choice of policy instruments.
When trade-offs persist, the question is how society should choose between competing objectives. While those choices need to be based on the best possible evidence, they involve value judgments and need to be made in a way that commands broad support across society, is consistent with international obligations, and effectively addresses the triple challenge.
2.1. Introduction
Chapter 1 discussed how food systems around the world are facing a considerable “triple challenge” of simultaneously ensuring food security and nutrition, livelihoods, and environmental sustainability. Crafting effective policy responses is made more difficult by important interactions between the dimensions of the triple challenge: changes on one dimension can lead to unforeseen positive or negative changes on other dimensions, creating complex patterns of synergies and trade-offs. The question for policy makers is thus how to design effective policies in light of these interactions between different policy domains. The overarching ideal is that of coherent policies – where the design of different policies takes into account relevant interactions, and coordination and calibration takes place to achieve a desirable policy mix at the international, national and sub-national levels.1
The triple challenge facing food systems connects to several of the Sustainable Development Goals (SDGs). By their very nature, the SDGs call for coherence, given the broad range of policy domains involved and the transboundary nature of many of the challenges, and insights on policy coherence developed in the context of the SDGs are therefore useful for the triple challenge as well.2 In 2019, the OECD Council adopted the Recommendation on Policy Coherence for Sustainable Development (OECD, 2019[1]), which contains principles to enhance such coherence falling under three pillars:
A strategic vision for implementing the 2030 Agenda and the SDGs in an integrated and coherent manner. Three principles under this pillar are building political commitment and leadership; defining, implementing and communicating a strategic long-term vision; and improving policy integration.
Effective and inclusive institutional and governance mechanisms to address policy interactions across sectors and align actions between levels of government. Three principles under this heading are ensuring whole-of-government coordination; engaging sub-national levels of government; and engaging stakeholders.
A set of responsive and adaptive tools to anticipate, assess and address domestic, transboundary and long-term impacts of policies. This requires analysing and assessing the impact of policies and financing (in particular on developing countries); and strengthening monitoring, reporting and evaluation systems to collect evidence. 3
Building on the OECD Recommendation, this chapter explores key principles for policy coherence in the context of food systems, taking into account the specificities of the triple challenge. Food systems are complex: not only is there a direct connection between food security and nutrition, livelihoods, and environmental sustainability, but many of these connections are global, e.g. through international trade in agricultural products or through global environmental externalities. This global dimension increases the “demand” for policy coherence (as the potential spillovers reach further) but at the same time makes it more difficult to “supply” policy coherence (as coherence needs to be defined across many jurisdictions, and hence is not limited to a single set of decision-makers). The complexity of achieving policy coherence at a global level is illustrated by the difficulties in completing multilateral trade negotiations in the World Trade Organization (WTO).
In addition to being complex, the interactions across the triple challenge are potentially contentious. Two ideas recurring throughout this chapter are therefore that pragmatic approaches are needed (based on the idea of making things “as simple as possible, but not any simpler”)4 and that policy design cannot be reduced to a technical exercise of quantifying interactions: when there are trade-offs, there remains a societal question of preferences in choosing among competing objectives.
2.2. The promise and challenges of policy coherence
When interactions exist between policy objectives, a policy focused only on a single objective might lead to unintended side effects (in the case of trade-offs) or might fail to realise all its possible benefits (in the case of synergies). When these interactions are not taken into account, the resulting policy mix will be less effective and may waste resources. Policymaking should therefore aim for policy coherence, defined for the purposes of this chapter as an outcome where various policies are aligned so that efforts in one policy area do not undermine efforts in another area, and even reinforce those efforts where possible (Parsons and Hawkes, 2019[2]).
Policy incoherence – the absence of policy coherence – can vary in scope and degree. The most serious types of incoherence can occur when general policy goals are misaligned. For example, a government setting ambitious but divergent goals for agricultural production and environmental performance will potentially create more complex problems of policy incoherence compared to a case where goals are aligned but the implementation details of a specific programme inadvertently end up encouraging certain environmentally harmful types of agricultural practices. In the latter case, a change to implementation rules may be sufficient to restore policy coherence; in the former case, a more serious realignment of policy goals and resulting instruments may be needed. Policy incoherence may also vary in degree, e.g. from merely failing to exploit synergies to slowing down or counteracting progress (OECD, 2019[3]). Several cases of policy incoherence of varying scope and degree have been documented in food and agriculture policies, as discussed below.
Ideally, policy coherence avoids misalignments at all levels; it should therefore take into account all relevant synergies and trade-offs between all policy objectives in designing the optimal policy mix across all possible policy instruments. Despite the potential benefits of policy coherence, actually achieving such a level of coherence may prove difficult.
First, achieving coherence is costly for policy makers. Coordinating across a large number of policy areas, and potentially across several levels of government, creates transaction costs. This in turn means that achieving coherence may require a significant amount of time, attention, and energy, or “political capital”, of various policymaking communities. The transaction costs of achieving policy coherence are further increased by a lack of information about all potential interactions. The number of potential interactions between different policy areas is vast, and if information is not readily available, achieving coherence may require extensive research and consultation to identify potential interactions. To be sure, the benefits of improved policy coherence may be worth the higher transaction costs. But there is a risk that striving for perfect coherence might lead to slow decision-making or even paralysis, and it might therefore be more feasible to strive instead for “good enough” coherence (Vanheukelom, Mackie and Ronceray, 2018[4]). Such an approach would imply identifying and dealing with the most important synergies and trade-offs.
A second challenge with achieving coherence is that synergies and trade-offs between policy areas themselves depend on the specific policy instruments chosen by policy makers. For example, policies to reduce greenhouse gas emissions from agriculture could include a carbon tax, an abatement subsidy, policies to encourage shifts in diets away from ruminant livestock products, increased R&D efforts, and so on. Each of these policies will have different implications for food security and nutrition, for livelihoods, and for other dimensions of sustainability (OECD, 2019[5]). The choice of the number and type of instruments thus determines the pattern of synergies and trade-offs. This creates an additional challenge for coherence, as mapping potential interactions depends on all the different instruments that could be considered – thus increasing the analytical requirements. On the other hand, this fact can also present an opportunity, as a smart choice of instruments can help minimise trade-offs and maximise synergies.
A third challenge involves the problem of societal preferences and choice when faced with a trade-off. Designing a coherent policy may force a choice between two or more desirable but competing outcomes. For example, measures to improve animal welfare may raise the price of food, which tends to disproportionately impact poorer households.5 A common trade-off also arises whenever a proposed policy requires public funds, as this typically implies either an increase in the tax burden or a reduction in spending on other programmes. These choices cannot be reduced to technical problems as they depend on a society’s priorities. In democracies, there will rarely be a unanimous view on how such trade-offs should be decided. Moreover, different interests will try to influence decision-makers to tip the scales in their favour. There is thus an important question of how societies make decisions: the processes they use to choose among conflicting objectives, and how this decision process can be well-informed while being insulated from “policy capture” by interest groups. Chapter 3 discusses these issues in more detail.
The above challenges are compounded in the case of transboundary spillovers. As long as spillovers occur within a single jurisdiction, both the costs and benefits of policy coherence are domestic and the choices can be made within one country’s decision-making processes. This typically makes it easier to obtain and exchange information on interactions, and to coordinate different policymaking communities. It also makes it easier to get representation of relevant stakeholders and make a political decision on societal trade-offs. If benefits accrue domestically, it is also easier to justify the transaction costs of achieving coherence. In an international setting, policy coherence becomes a common-pool resource, leading to collective action problems. Climate change mitigation is a case in point. Policy settings in one country may be acceptable from the point of view of domestic interests (e.g. by not imposing costly mitigation efforts) but may indirectly impose costs on stakeholders in other countries (who will suffer the consequences of climate change). These stakeholders are usually not represented in domestic policy-making processes, and at the international level, there is no single last arbiter or decision-maker. As a result, it is harder to address international cases of policy incoherence.
These challenges cannot necessarily be “solved”, but they can be managed more or less effectively. Building on OECD recommendations on policy coherence for sustainable development (OECD, 2019[1]), on regulatory policy and governance (OECD, 2012[6]), on the OECD Best Practice Principles on both Regulatory Impact Assessment (OECD, 2020[7]), on International Regulatory Cooperation (OECD, Forthcoming[8]), as well as on a broad range of OECD work on policy challenges in food and agriculture, the remainder of this chapter proposes principles for a pragmatic approach to ensuring policy coherence for food systems. This approach first seeks to reduce complexity. Where complexity remains, the proposed approach then seeks to rigorously test and quantify potential interactions, carefully calibrating the policy mix, and making societal and transboundary trade-offs explicit to facilitate conscious, transparent decision-making. (The question of how society should choose when faced with those trade-offs is explored in Chapter 3). The principles identified in this chapter (summarised in Figure 2.1 at the end of the chapter) can be translated into different practical approaches. As noted below, the principles could, for instance, be used for an ad hoc assessment of the coherence of existing policies, or could be embedded in routine processes for evaluating new policy initiatives, as is the case in several jurisdictions.
2.3. Reducing complexity
Assessing the need for policy intervention
A proposed first step in reducing the complexity of ensuring coherent policies for food systems is to ask whether and when policy intervention is needed.
The role of public policies in addressing the “triple challenge” can be assessed using the same principles applied to public policy in other policy areas. Communities, markets, and states have different strengths and weaknesses in meeting societal objectives, and, while different societies may opt for different mixes along the continuum, a long tradition of economic and public policy analysis has identified particular roles for government intervention in three areas. A first area is establishing framework conditions such as the rule of law, security, macroeconomic stability, and a rules-based international system; these framework conditions in turn allow markets and help communities to function. A second role for government is to correct specific market failures (including problems resulting from market power and asymmetric information, coordination issues, and externalities) and to provide public goods. Third, public policy can also address notions of social justice and an equitable distribution of incomes and opportunities across society (OECD, 2002[9]). Across all of these areas, it remains an empirical question whether and to what extent policy intervention could help in a given context, and which policy instruments are most fit for purpose, in line with general principles of good regulatory policy (OECD, 2012[6]) – in particular regulatory impact assessment (OECD, 2020[7]) and sound public governance (OECD, 2018[10]).
No policy intervention can be successful without defining the policy context and objectives, in particular the clear and systematic identification of the problem that provides the basis for action by government. In the context of food systems, specific policy interventions may be warranted in order to provide public goods and to address market failures. Agricultural research and development, education on what constitutes a healthy diet and lifestyle, and the provision of food safety regulation and testing are just some examples of investments in public goods with important benefits for society. Policy interventions can also help to correct externalities, where the social costs and benefits of agents’ decisions diverge from the market signals that drive those decisions. Those externalities may be negative, such as GHG emissions and the public health burden of unhealthy food choices. They may also be positive, such as the contribution of agriculture to maintaining rural landscapes. Public policy can help private actors internalise the social costs and benefits of their actions, thus ensuring that private decisions are more aligned with societal objectives. Other policy measures can address problems of market power, asymmetric information, and coordination failures to ensure a better functioning of markets for the benefit of the population.
Market failures and public goods do not automatically imply a specific role for government intervention, however, and the existence of market failures and public goods does not necessarily mean that current levels and instruments of policy intervention are the most efficient in responding to these problems or in providing these benefits (OECD, 2002[9]). There exists a continuum along which some degree of intervention can be warranted – while keeping in mind that different forms of government intervention carry their own costs, including the cost of inaction. In some cases, direct government action is needed (e.g. through regulation or fiscal measures); in other cases, governments can play a coordinating role for initiatives by private stakeholders or can set non-binding standards which can nonetheless become an important reference in guiding others’ action or behaviour, either at the national or international level (Box 2.1). Policy makers should therefore consider all plausible alternatives, from laws and regulations through to non-legislative solutions, and including the possibility of not taking any action.
The social justice argument for income redistribution is regularly conflated with the market failure/public goods rationale for government intervention in food systems, but from a public policy standpoint, it can be important to separate efficiency and equity arguments. Even when markets are functioning well, there is no guarantee that market processes will result in a socially desirable distribution of incomes. Conversely, concerns over income distribution do not necessarily imply that markets are not efficient. Moreover, while both factors could be a rationale for government intervention, the two concerns are not necessarily addressed with the same policies. Mixing the two makes it harder to design effective policies, harder to assess their effectiveness ex post, and reduces political accountability (OECD, 2002[9]). Achieving clarity on policy objectives is therefore an important precondition for developing effective policies (van Tongeren, 2008[14]). If both equity and efficiency are policy objectives, clearly distinguishing them is important to allow a transparent analysis of the different instruments which can be used, and the potential synergies or trade-offs between them.
Box 2.1. The OECD-FAO Guidance for Responsible Agricultural Supply Chains
Businesses can play a major role in meeting the triple challenge. At a minimum, they can minimise any adverse impacts of their operations, supply chains and other business relationships. Governments and international organisations can act as a catalyst or provide incentives of varying strength, e.g. by providing guidance on good practices, by promoting and monitoring their adoption by companies, and by incorporating them into public procurement standards.
The OECD-FAO Guidance for Responsible Agricultural Supply Chains (OECD/FAO, 2016[11]) provides a common framework to help agri-businesses and investors contribute to sustainable development. By implementing the due diligence recommendations of the OECD-FAO Guidance, companies can systematically identify, assess and mitigate potential negative impacts associated with their business. The Guidance is relevant for all enterprises across the entire agricultural supply chain, from the farm to the consumer, across food and non-food commodities. It covers a wide range of topics, including human rights; labour rights; food security and nutrition; health and safety; tenure rights over and access to natural resources; animal welfare; governance; environmental protection and sustainable use of natural resources; and technology and innovation.
In developing this Guidance, OECD and FAO worked closely with supply chain experts, business, civil society, unions and policy makers. The recommendations also incorporate long-standing standards for responsible business conduct (RBC), including the OECD Guidelines for Multinational Enterprises (OECD, 2011[12]) and the UN Committee on World Food Security’s Principles for Responsible Investment in Agriculture and Food Systems (CFS, 2014[13]).
In addition to the standard itself, the OECD provides ongoing support towards the widespread uptake and implementation of the OECD-FAO Guidance with policy makers, business and relevant stakeholders.
The OECD-FAO Guidance demonstrates how the OECD and FAO together with national governments can use their convening power and expertise to encourage and support efforts by the private sector. In 2018-19, the OECD and FAO jointly implemented a pilot project with over 30 companies and industry initiatives. Several companies that took part in this pilot have started exploring how to collaboratively address systemic risks in common supply chains and markets. The OECD is also implementing a regional pilot in 2019-20 in Southeast Asia under the EU-ILO-OECD Responsible Supply Chains in Asia programme, with funding by the European Union.
Furthermore the OECD supports National Contact Points for RBC (NCPs), which are unique agencies for RBC established by governments. Their mandate is twofold: to promote the OECD Guidelines and related due diligence guidance including the OECD-FAO Guidance, and to handle cases of alleged non-observance of the Guidelines as a non-judicial grievance mechanism. To support NCPs in implementing their mandate, the OECD helps facilitate peer learning activities and voluntary peer reviews, and provides capacity building and practical tools to NCPs on promoting RBC and facilitating conciliation and mediation for access to remedy.
As this example shows, public-sector actors such as the OECD and FAO can facilitate a broad coalition of private and public stakeholders, with the support of civil society and worker representatives, to address transboundary impacts on people, the planet and society and meet global commitments on the Sustainable Development Goals (SDGs).
Choosing the right type and number of instruments
A second technique for reducing complexity is through smart choices of the type and number of policy instruments.
Trade-offs tend to be more severe when policy makers try to use a single instrument to achieve multiple policy objectives, or when an inappropriate policy instrument is used. For example, using support coupled to prices or production (e.g. import tariffs, output subsidies) as a means to transfer income to farmers can stimulate overproduction and lead to environmental problems and concerns about unfair competition from other producers. Within the context of this policy instrument, there is a trade-off between supporting farmers’ incomes on the one hand, and economic efficiency, competition and environmental sustainability on the other. By contrast, more decoupled forms of income support (unrelated to current production) are less likely to lead to these problems: they can achieve income support to farmers with considerably smaller economic and environmental costs. Choosing the appropriate type of policy instrument can therefore weaken or even remove trade-offs between different policy objectives, which in turn makes it easier to achieve those different objectives.
When synergies exist between objectives, it will in general not be optimal to rely on a single policy instrument to achieve several objectives (a “silver bullet” approach).6 The type and extent of policy interventions required to obtain one objective is unlikely to be the same as that required to achieve other objectives. For example, the optimal diet for human health is unlikely to coincide exactly with the optimal diet for environmental sustainability, and vice versa. In this example, other policies to improve the sustainability of food production are also likely to be needed.
Using a sufficient number of policy instruments thus reduces the risk of policy-induced trade-offs and increases the likelihood that different objectives can be achieved. As a rule of thumb, policy makers would therefore ideally need as many instruments as objectives. This principle is sometimes referred to as the “Tinbergen rule”, after the Dutch Nobel-winning economist Jan Tinbergen (Tinbergen, 1952[15]).7 When policy makers are faced with a trade-off between different objectives, one possible answer is therefore to reassess whether the existing policy instrument is fit for purpose, and search for new policy instruments which could replace or complement the existing policy instrument and help resolve the trade-off. When synergies are present, policy makers should keep in mind that “silver bullets” rarely exist; multiple objectives usually require multiple policy instruments. Even if one policy instrument has positive effects on a number of different objectives, complementary policy actions are usually needed to fully achieve those objectives.
One way to implement the “Tinbergen rule” is to choose policies which address market failures as directly as possible in order to avoid unintended secondary distortions (van Tongeren, 2008[14]). This may not always prove feasible, but searching for a targeted approach is nonetheless a good starting point. If the goal is to reduce environmental harm, such a targeted approach would seek to link incentives to the environmental outcomes. If a targeted approach is feasible, an additional benefit is that it makes it easier to assess whether the policy is achieving its goal and easier to adjust the policy when needed.8 Indeed, some objectives for the food system could be pursued relatively independently if synergies or trade-offs are minor.
The need to distinguish clearly between different policy objectives, and to choose the appropriate type and number of instruments, is illustrated well by earlier OECD work on multifunctionality in agriculture (OECD (2001[16]), (2003[17]), (2008[18])). The starting point for that work was the observation that agricultural production is in some cases associated with the production of desirable “non-commodity outputs,” including valued positive externalities or public goods (e.g. a pleasing landscape or certain eco-system services). This led to concerns that efforts to reduce production-linked support to agriculture could result in a lower supply of these outputs. OECD work emphasised the importance of first establishing the actual extent to which non-commodity outputs are linked to, or can be dissociated from, commodity production. If the two can be provided separately, a better policy approach is to provide separate incentives for the production of the non-commodity output – an instance of targeting. When a link exists, there may be ways to relax or weaken it, e.g. through changes in farming practices. Multifunctionality only becomes a policy issue when there is a strong link which cannot be altered, and when there is a market failure associated with the production of the non-commodity output. Even then, more targeted policies (rather than relying on production-linked support) are often available.
The principle of targeting can also refer to the level or scale at which government action is taken. For example, measures to reduce nitrogen leakage could in principle target individual farms or even parcels. A study by Konrad et al. (2014[19]) of nitrogen leakage in the Odensje Fjord (Denmark) illustrates how targeting at the parcel level could help to reduce the economic cost of such measures. Using highly disaggregated spatial data on parcels, simulation results show that a targeted approach would involve implementing measures (a mix of reductions in fertiliser application, planting of catch crops, establishing wetlands, and other interventions) on only 15% of all arable land, with no intervention on the remaining parcels. By contrast, policies uniformly applied to all parcels would be considerably more costly. These findings suggest that a more targeted approach can achieve reductions in nitrogen runoff at a lower cost – hence weakening the trade-off between economic costs and environmental benefits in this example. However, this level of targeting itself comes at a cost to policy makers, both in terms of information requirements and in terms of administration. In some cases, the additional cost of targeting may outweigh the benefit. In these cases, policy makers may choose a less targeted approach, while acknowledging that there will be a higher risk of trade-offs.9
One perhaps surprising implication of the “Tinbergen rule” is that the co-existence of policies that seemingly work in different directions is not necessarily incoherent. For example, the same agricultural activity (e.g. keeping ruminant livestock) may create positive externalities in one dimension while creating negative externalities in another. Policy makers might end up simultaneously paying farmers for the positive externalities (e.g. landscape services) while taxing them for the negative externalities (e.g. emissions) of the same activity. From the point of view of an individual farmer, this may seem incoherent. However, this situation might make sense from a policy point of view, as one instrument provides incentives to increase the provision of the positive externalities while the other instrument incentivises farmers to find ways to reduce the negative externalities of that activity. Moreover, using separate policy instruments may make it easier to evaluate the effectiveness of policies ex post.
2.4. Accounting for interaction effects
The techniques described in the previous section can help to reduce complexity and simplify the challenge of coherent policies. At the same time, some complexities will remain. Policy makers need to be aware of the possibility of interactions among policies, and need to have the tools to identify the nature and extent of such interactions. In the case of synergies, even when as many instruments as objectives are used, there is a question of the amount of policy intervention required through each of the different policy instruments. When trade-offs exist, tough choices may exist between different, valued outcomes. Several lessons regarding good practices can help policy makers with these questions of identification, calibration, and mediation.
Identification
Being able to identify the existence, nature and extent of trade-offs or synergies is necessary for coherent policymaking. Identification has two distinct aspects. First, policy makers need to be aware of the possibility that a new policy initiative may have spillover effects, or that an existing set of policies may be incoherent. Various mechanisms such as whole-of-government processes, regulatory impact assessments or stakeholder consultation processes can help in this initial screening process.10 Identifying spillovers requires a broad view: some interactions may affect wellbeing “here and now”, but it is equally important to consider potential interactions “elsewhere” (i.e. transboundary effects) and “later” (i.e. inter-generational effects) (OECD, 2016[20]). For example, in the context of the Sustainable Development Goals, work by the OECD (2016[21]) and (2017[22])) has explored potential positive and negative interactions between SDG 2 (Zero hunger) and other SDGs across a range of policy areas.
Such a screening may uncover potential interactions, but it is important to scrutinise these in a second step to establish where actual interactions exist, and how important these are from a policy perspective. This is an analytical task, which may involve statistical or experimental evidence, simulations, further consultations with experts and stakeholders, and other types of evidence. In an interconnected world, it is easy to hypothesise plausible interactions; but what matters for policy is whether these interactions are important enough to warrant adjustments to policies. Evidence on the extent of interactions is thus an important prerequisite to making transparent policy decisions.
Research by the OECD has uncovered several instances of incoherence between existing policies affecting food systems.11 Chapter 1 reviewed recent OECD work on the environmental impact of agricultural policies, which concluded that policy instruments differ in their environmental impacts. The most negative impacts are generally found for coupled support (e.g. import tariffs, output subsidies), while more decoupled forms of support generally have smaller environmental impacts (OECD, 2019[23]). Work by the OECD on synergies and trade-offs between climate change adaptation, mitigation and agricultural productivity similarly found that existing agricultural and agri-environmental policies may be sending conflicting signals on at least one of those three goals. In addition, the work showed that countries rarely assessed whether existing policies were effective in supporting all three objectives (Lankoski et al., 2018[24]). Despite negative spillovers, much of the support to farmers globally is still provided through coupled support. Agricultural policy around the world thus relies on the potentially most environmentally harmful instruments (OECD, 2019[25]). In many countries, farmers also benefit from fuel tax concessions and other advantages such as lower VAT rates applied to pesticides and fertilisers (OECD, 2020[26]). However, as both the level of agricultural support and the mix of policy instruments used varies around the world, specific analysis is needed to decide on the importance of this type of incoherence in a given country context.
Detailed country reviews by the OECD on policies for innovation, productivity and sustainability in food and agriculture also uncovered several instances of policy incoherence (OECD, 2019[3]). In some cases, potential conflicts could be found at the level of general policy goals (e.g. promoting agricultural production growth but also aiming for more sustainable land use; these goals are not necessarily incoherent but could come into conflict if pursued independently of each other). Numerous instances of potential policy incoherence were found within policy areas (e.g. when public research efforts are not directed towards the needs of the sector) and among policy approaches within an area (e.g. when different commodities receive different degrees of agricultural support). Similarly, recent OECD work on policies for sustainable land use found many instances of policy incoherence, often linked to a lack of institutional coordination and the fragmentation of responsibilities among numerous agencies (OECD, 2020[27]).
It is essential that potential direct and indirect interactions are identified, substantiated by evidence, and where possible quantified. Interactions which are plausible in theory may not materialise; interactions which have been demonstrated in one context may turn out to be less relevant in a different context; and interactions which were demonstrated at some point in the past may have weakened or disappeared since. Rigorous, relevant and up-to-date evidence is thus important.
One example is the changing landscape of agricultural trade and the role of policies (OECD, 2019[28]). Historically, high levels of support and protection in developed countries often led to production surpluses, which were subsequently released on international markets using export subsidies. This resulted in lower international prices, which put pressure on exporters and import-competing producers in developing countries (OECD, 2013[29]). However, important reforms have taken place in the past two decades. Although support for agriculture remains widespread (and, in some countries, very large as a percentage of farm gate receipts) and trade barriers remain more important for agriculture than for industrial goods, the sector overall has seen falling tariffs and reductions in trade-distorting producer support by a number of developed countries (OECD, 2019[28]). In particular, the WTO Ministerial Conference in Nairobi (2015) agreed on the elimination of agricultural export subsidies by developed countries.12 For example, the European Union, which made frequent use of export subsidies in the past, has ceased using this instrument since 2015 (Matthews and Soldi, 2019[30]). In parallel with this development, developing countries have moved away from policies which generally taxed the agricultural sector, with some now also providing support to farmers using distortionary instruments (Anderson, Rausser and Swinnen, 2013[31]) (Bouët and Laborde, 2017[32]) (OECD, 2019[25]). Other important changes have taken place: in the past two decades, agro-food trade has grown strongly, with important flows among emerging and developing countries. Agro-food sectors are also increasingly connected to other sectors, both domestically and through global value chains (OECD, 2019[28]). The international environment thus has changed in important ways over the past two decades, and discussions on potential spill-over effects (e.g. of agricultural and trade policies) should be based on up-to-date evidence and analysis.
An example of how hypothesised interactions may not materialise is the potential link between agricultural support policies and rising obesity. Several commentators have suggested that policies to support agricultural incomes (such as farm subsidies) have been an important contributor to rising obesity (see, for example, Wallinga (2010[33]) and Elinder (2005[34])). However, empirical evidence suggests otherwise. Agricultural support is often provided through trade barriers and other policies which raise domestic prices, thus discouraging consumption. Detailed analyses for the United States find that agricultural support policies have had small and mixed effects on prices of primary agricultural commodities, which in turn only play a small role in influencing the consumer prices of different types of food products. Researchers therefore conclude that agricultural support policies in the United States have a negligible impact on the evolution of obesity (Okrent and Alston, 2012[35]; Rickard, Okrent and Alston, 2013[36]). International evidence also shows that countries with the highest levels of agricultural support tend to have higher agricultural commodity prices which are in turn correlated with lower rates of obesity (Alston, Sumner and Vosti, 2008[37]). Contrary to perception, agricultural support policies in OECD countries are unlikely to be a driver of rising obesity rates.
The implication here is not that agricultural policies never matter for nutritional outcomes: in some contexts, agricultural policies may well be an important contributing factor to negative nutritional outcomes. This could arguably be the case, for example, in developing countries where policies often feature a near-exclusive focus on staple grains such as rice, wheat or maize at the expense of other agricultural products which could contribute to a more diverse diet (Pingali, 2015[38]).
In contrast with agricultural support policies, there is evidence that public investments in agricultural R&D in the United States were a major driver of declining real prices of agricultural commodities and hence contributed to greater calorie intake (Alston, MacEwan and Okrent, 2016[39]). However, even when such interactions can be demonstrated, it may be far from optimal to exploit these in policy design. Estimates by Alston et al. (2016[39]) suggest that if there had been no increase in the public agricultural R&D “knowledge stock” in the United States between 1980 and 2004, agricultural commodity prices and food prices would have been much higher and calorie intake lower. This would indeed imply lower obesity-related public healthcare costs, but those savings would be dwarfed by a large increase in consumer expenditures for food. Alston et al. (2016[39]) conclude that although public R&D spending influences calorie intake (and hence obesity and associated healthcare costs), using this instrument would create a large trade-off in terms of consumer welfare relative to other potential policies. Hence, even if an interaction exists, other potential policies have a better “profile” in terms of effectiveness and associated costs and trade-offs. The identification of interactions is thus only a prelude to a process of calibration (the choice of the optimal combination of instruments) and mediation (the choice between conflicting societal objectives).
At the same time, policy making never happens in a context of full information, whether because scientific knowledge evolves continuously, or because gathering additional evidence is costly and time-consuming so that postponing a policy decision until more information becomes available is itself a potentially costly decision. As discussed in the next chapter, there are important knowledge gaps on food systems, including on the extent and characteristics of policy issues, their synergies and trade-offs, and the costs and benefits of potential policy options. Investing in robust processes to gather the best possible evidence is thus important. But as decisions are often made with incomplete information, it is important to make explicit the uncertainty over possible costs and benefits of policy options.
Calibration
For synergies, the key question is one of calibration – of choosing the optimal combination of policy instruments (and their “settings” or magnitudes) to reach the desired objective, taking into account the empirical evidence on the relative effectiveness and other strengths and weaknesses of different instruments.
This is essentially no different from the task policy makers face in any policy setting: to carefully choose the best policy instrument to reach a given policy objective, including considering the baseline of no policy action (OECD, 2012[6]). Yet it is worth making this principle explicit, as in the context of the food system there is often a tendency to present specific initiatives as “silver bullets” capable of resolving many challenges at the same time. As pointed out earlier, even if synergies exist, it will never be the case that a single policy instrument can achieve multiple objectives perfectly. It is therefore important that policy makers consider different policy instruments alone and in combination, and where possible quantify their effects to determine the best policy mix. It is also important to consider and evaluate the full range of policy instruments, including plausible alternatives to a legislative/regulatory option, which can vary in terms of their stringency.13
One application of this principle regards the use of demand-side measures as a possible instrument to address negative externalities on the supply side (e.g. when a reduction in demand for ruminant livestock products is suggested in order to reduce associated greenhouse gas emissions). A well-known result in economic theory holds that in a closed economy, the effects of a tax do not depend on whether it is levied on producers or consumers: the result is a similar decrease in quantities produced (and consumed). This might suggest that efforts to encourage more “sustainable” consumption patterns could under the right circumstances achieve the same impact as policies that tax production for this purpose – and conversely, that supply-side measures could be used to achieve demand-side goals (e.g. healthier diets). In practical terms, however, there are several difficulties that need to be considered.
First, externalities are rarely linked one-for-one to the level of production or consumption. In the case of greenhouse gas emissions, for example, what matters is not only the volume of production but also the emissions intensity of production. The ideal instrument would tackle emissions itself. A tax on emissions would not only reduce production overall, but would also favour less emissions-intensive production methods, thus reducing the average emissions intensity in the sector. Moreover, intermediate sectors (e.g. food manufacturers) would in turn have an incentive to substitute away from emissions-intensive inputs, including by developing new products and processes. By contrast, an undifferentiated production tax takes away this flexibility. The same is true for a policy which aims to reduce demand (unless the policy can differentiate between producers depending on their emissions intensity, e.g. through a labelling scheme – although this would bring additional issues and costs). These policies will reduce emissions in line with the average emissions intensity of production without having much effect on emissions intensity itself. As a result, these undifferentiated measures achieve a relatively poor targeting of the problem they aim to address. A targeted policy can achieve the same reduction in emissions with a smaller decline in production.14
The bottom line here is that even if in theory demand-side measures could, under the right circumstances, be used to address supply-side problems and vice versa, this may not be optimal in practice. Policies therefore ideally focus on targeting externalities as directly as possible. Thus consumption externalities are best corrected using demand-side policy interventions, while supply-side policies will typically be the most effective approach to address production-related externalities such as emissions. However, if there are synergies between these goals, this could reduce the amount of policy effort needed on the separate targeted policies.
Mediation
For trade-offs, mediation between conflicting objectives is necessary. While identification and calibration have a strong focus on technical analysis, trade-offs between societal objectives necessarily involve value judgments. Identifying trade-offs will often highlight fundamental differences of opinion about how different objectives should be weighed against one another. These differences could reflect conflicting views on what is in society’s wider interests, or narrow resistance from interest groups to policies that disadvantage them. In some cases, interest groups can gain considerable influence on public policy-making processes, leading to policy capture (OECD, 2017[40]). In either case, avoiding gridlock may require making tough choices or striking a grand bargain to overcome opposition.
Trade-offs are not a purely technical matter. For example, Okrent and Alston (2012[35]) consider the potential of taxes and subsidies to combat obesity and find that a tax on calories would be a highly effective obesity policy. The tax raises food prices, which hurts consumers financially, but as this effect is smaller than the savings in public health care costs, the tax has a positive net welfare effect. However, this efficiency gain comes at a potential cost in terms of equity. Food represents a larger share of the household budget for poorer households: in the United States, households in the lowest 20% of the income distribution spend 35% of their income on food, a share which falls to 8% for households in the highest 20% of the distribution (USDA, 2019[41]). While the calorie tax would be an effective way to reduce calorie intake and address obesity, it might also proportionally inflict a greater financial burden on lower-income groups in society in the absence of mechanisms to financially compensate them. At the same time, other research has found that individuals with lower socio-economic status are also likely to see financial benefits from improved diets, including through lower out-of-pocket healthcare spending and improved labour market outcomes (Sassi et al., 2018[42]). 15 The trade-off in terms of equity effects could be weakened or eliminated if those effects are taken into account, again underscoring the importance of empirically assessing and quantifying interactions. But if a trade-off remains between equity and public health, then whether and how to trade off these different outcomes is not a question which can be settled by analysis alone.
In the particular case of the calorie tax, one solution to a potential trade-off is to rely on other instruments, either alone or in combination with a tax. Analysis by the OECD has proposed a “four-track policy approach” to encourage healthier food choices. In addition to fiscal measures, this includes demand-side public interventions (e.g. education programmes or providing dietary information), voluntary collaborations with the food industry at the supply-demand interface (e.g. food reformulation, food labelling), and firmer regulations when public-private incentives are misaligned (e.g. rules on advertising aimed at children) (Giner and Brooks, 2019[43]). Some combination of these instruments may help encourage healthier food choices without necessarily imposing the same financial burden on poorer households. An alternative solution might be to provide offsetting financial compensation to poorer households.
However, such solutions to soften or avoid trade-offs are not always available. In those cases, society faces a choice between two or more valued outcomes. As discussed in more detail in the next chapter, there is no simple decision rule available to unambiguously resolve such trade-offs.16 Democratic deliberation could help, by allowing those involved in the deliberation to reflect on information and arguments as well as on their own preferences; deliberation can also create a context more conducive to compromise (Dryzek and List, 2003[44]).17 But even such a deliberative approach may fail to address transboundary spillovers if foreign stakeholders are not represented, as discussed below.
Policy approaches
There are various ways to translate the principles identified in this chapter into specific policy approaches. In several jurisdictions, the development of new policies routinely requires policy makers to assess issues of policy coherence through regulatory impact assessments. Policy makers can also decide to use a more elaborate multi-stakeholder consultative process for major policy initiatives. Coherence can also be assessed for new policies, but also for existing policies through “stocktaking” exercises. Moreover, various degrees of “policy integration” (coordination between policy-making communities) can also be used to improve coherence, although integration comes at a cost and does not necessarily guarantee better outcomes.
OECD countries are increasingly using Regulatory Impact Assessments (RIA) as a routine “screening” of proposed new laws and regulations (OECD, 2018[45]). Such assessments document the efficiency and distributional effects of proposed policies and can help illuminate trade-offs. The OECD Council Recommendation on Regulatory Policy and Governance calls for RIAs to include economic, social and environmental impacts, including distributional effects over time, and identifying who is likely to benefit and who is likely to bear costs; assessments should also be quantitative whenever possible (OECD, 2012[6]). These principles are increasingly common in OECD countries. In the United States, for example, guidance by the Office of Management and Budget to federal agencies on the development of regulatory analysis explicitly requires such assessments to “look beyond the direct benefits and direct costs (…) and consider any important ancillary benefits and countervailing risks” (Office of Management and Budget, 2003[46]). Similarly, the European Commission’s “Better Regulation” guidelines on impact assessment require a comparison of different policy options on the basis of economic, social and environmental impacts, quantified whenever possible. Impact assessments must also include a description of who will be affected and how, as well as potential impacts on competitiveness, small and medium-sized enterprises (SMEs). Moreover, impact assessments must explain the consultation process used (European Commission, 2017[47]). These and related approaches to RIAs in other countries improve policy coherence by requiring an ex ante evaluation of potential trade-offs and synergies, and allowing a comparison of different policy options taking into account these interaction effects.
Policy makers can go beyond the normal requirements of RIAs in developing major new policies, for example by setting up more elaborate multi-stakeholder consultation mechanisms than would normally be done. To develop its “Food Policy for Canada”, the Canadian government started a process of public consultations in 2017. These consultations included dialogues with Indigenous Peoples and organisations, as well as with people working in the food system, and people and organisations active on issues such as food security and food waste. Close to 45 000 people provided input via online surveys while hundreds participated in face-to-face regional consultation sessions, as well as a National Food Policy Summit held in Ottawa. The consultation process was organised collaboratively by several federal departments and agencies, and the various viewpoints and insights were published in a separate report (Government of Canada, 2018[48]). Based on these extensive consultations, the government announced the first-ever food policy for Canada in 2019, including funding for initiatives such as a “challenge fund” to support innovative food waste reduction ideas, and support for community-led projects to address food insecurity (Agriculture and Agri-Food Canada, 2019[49]). An important component of the Food Policy for Canada will be the creation of a Food Policy Advisory Council, which will explicitly aim to bring together a wide range of perspectives and expertise to help the government tackle complex and systemic issues related to food (Government of Canada, 2019[50]). The development of “A Food Policy for Canada” deliberately sought alignment with the various Sustainable Development Goals related to food, notably SDG 2 (Zero Hunger), SDG 3 (Good Health and Well-Being), SDG 12 (Responsible Production and Consumption) and SDG 13 (Climate Action).
Another multi-stakeholder approach to assess problems of policy coherence related to the food system is proposed by the “Collaborative Framework for Food Systems Transformation” developed by the One Planet Network’s Sustainable Food Systems Programme (UNEP/SFSP, 2019[51]). The framework can be used by national or local governments seeking to develop more coherent policies for their food systems. It emphasises the importance of preparing a holistic food systems assessment, including a stocktaking of existing policies and initiatives and their potential trade-offs and synergies. In addition, the framework calls for robust evidence, active involvement of stakeholders, and dialogue across policy communities throughout the whole policy cycle (including planning, implementation, and evaluation of results).
An example of such a stocktaking are the detailed country reviews by the OECD on policies for innovation, productivity and sustainability in food and agriculture (OECD, 2019[3]), mentioned earlier. Another example is provided by G20 countries’ peer reviews on support to fossil fuels (OECD, 2018[52]). In 2013, G20 Finance Ministers agreed to develop a framework for voluntary peer reviews of inefficient fossil fuel subsidies leading to wasteful consumption. Countries prepare a self-review which is then submitted to a review team consisting of representatives of other countries and international organisations such as the OECD. Preparing for these reviews has helped countries get a better understanding of their own existing policies, and provides a unique opportunity for cross-ministerial coordination and discussion on policy coherence. Fuel tax concessions for agricultural producers are common: a recent OECD review of taxation in agriculture across 35 OECD countries found that such concessions are used in nearly all the countries surveyed (OECD, 2020[53]). Because fuel tax concessions represent revenue foregone, they are less visible than budgetary transfers and not subject to the same scrutiny or frequency of review, underscoring the importance of stock-taking exercises such as the G20 peer reviews.
Responsibility over different policy domains often sits with different policy making communities (e.g. different ministries, departments, specialised agencies) or different levels of government (e.g. federal, provincial/territorial/state, municipal/local). In this case, dealing with both synergies and trade-offs will require some degree of “policy integration”, i.e. coordination between these policy communities (Parsons, 2019[54]). This can vary from ad hoc exchanges of views or collaboration on specific cross-cutting themes to complete functional integration, e.g. by placing all responsibility over a policy area within the same ministry. These mechanisms can help different institutions to align their mandates, objectives and policies, and to take into account synergies and trade-offs in their decision-making process (OECD, 2019[1]).
Since deeper levels of integration come at a cost, the existence of some interactions does not necessarily mean that all policy making related to the triple challenge requires complete functional integration. Among the different costs associated with deeper policy integration is the risk that stronger integration around one issue leads to weaker integration with other areas; the risk that leadership attention may be diverted towards coordination at the expense of programmes at the “subsystem” level; or the risk that integration conflicts with other values such as decentralisation or subsidiarity. In addition, successful policy integration is not automatically guaranteed, but requires leadership and specific analytical, operational, and political capacities (Candel, 2019[55]). A review of empirical evidence shows numerous examples of failed attempts at policy integration, underscoring the challenges involved (Candel, 2017[56]). Given these costs and risks of functional integration, the degree of coordination between policy communities should ideally match the strength of synergies and trade-offs between their respective policy domains.18
2.5. Transboundary effects
Policies in one country can have transboundary effects on one or more countries or on a global public good (OECD, 2019[57]). If countries decide on policies without regard to these spillovers, the likely result is frictions created by incoherent policies across countries and an underprovision of global public goods.19 A lack of international economic cooperation not only creates uncertainty for market participants and imposes costs on those involved in international trade, but also means that important global challenges are not addressed, and that potential mutual gains are not realised.
Achieving international cooperation is far from easy, as countries have different interests, preferences, and policy-making approaches. These differences are unlikely to go away. On the other hand, there are also instances where national policies diverge purely because national governments and regulators were working in isolation (OECD, Forthcoming[8]).The goal of international cooperation is not to eliminate all differences in policies, but to strike a balance between domestic objectives and realising the benefits of international cooperation (von Lampe, Deconinck and Bastien, 2016[58]).
Some key principles for domestic policy-making can help reconcile these objectives. If domestic policy approaches are transparent, non-discriminatory, and not more trade restrictive than necessary to achieve their objective, then this can help avoid unnecessary costs of policy differences. Unnecessary costs can also be avoided by requiring ex ante assessments of transboundary effects of proposed policies, including their coherence with any international standards and other relevant frameworks (in line with the 2012 OECD Council Recommendation on Regulatory Policy and Governance and the forthcoming OECD Best Practice Principles for Regulatory Policy: International Regulatory Coordination). That is, even without explicit coordination, policy makers can improve global coherence by using good regulatory practices in their domestic policy contexts.
Another way in which domestic processes can improve global coherence is by routinely taking into account the potential impact of proposed policies on stakeholders outside of the national boundaries. The RIA process provides an opportunity to do so, in particular through the assessment of trade impacts and of impacts on foreign jurisdictions. This can be facilitated by organising consultation with these external stakeholders, e.g. through business platforms or chambers of commerce. Transparency in terms of compulsory notification of draft regulations to international forums also provides an important means by which to alert and draw inputs from foreign stakeholders. The WTO TBT and SPS Agreements provide such an opportunity through the single central government authority responsible for notifications (OECD, Forthcoming[8]).
There are also several other institutional mechanisms for reducing costs and frictions internationally. These range from dialogue and the exchange of information (which can prevent “accidental” frictions) to joint solutions through international organisations or binding international agreements.20 Dialogue and the exchange of information may help avoid situations where countries end up with different policy approaches which achieve basically similar results domestically but accidentally impose trade costs or create other unwanted transboundary effects. Even when information is available and countries are aware of the actions of others, the result may be policy divergence and unnecessary costs if countries only take into account a short-term view of their own welfare. Negotiations to overcome this collective action problem may lead to a better set of policies across countries – although these do not necessarily imply a uniformity of policies. Other possible approaches might include recognising the equivalence of trading partners’ standards and conformity assessment procedures, or working together to develop international standards (OECD, 2017[59]).
As with other policy issues explored in this chapter, international cooperation is not only a technical question but involves trade-offs. All countries can benefit from international cooperation, but any specific solution adopted will have different patterns of costs and benefits across (and within) countries. The same is true for policy cooperation in other areas. All countries can benefit from improved policy cooperation, but there will inevitably be a debate over the direction in which countries should converge, and over the specific policy packages to be adopted.
2.6. Conclusion
A systematic consideration of synergies and trade-offs across food and agriculture can contribute to more coherent policies. Given the large number of potential interactions, this chapter has proposed some key principles (summarised in Figure 2.1) which can help policy makers reduce complexity and scrutinise potential interactions – while at the same time recognising the difficulty of deciding on trade-offs between competing objectives, and the difficulty of achieving policy coherence at the global level.
While this chapter has emphasised the usefulness of greater policy coherence as a way to improve policy effectiveness, it is important to keep in mind that policy coherence as defined here may not be sufficient by itself, if policies lack the ambition needed to address the triple challenge. For this reason, the OECD Recommendation on Policy Coherence for Sustainable Development identified political commitment, leadership, and a strategic long-term vision as important principles to achieve the Sustainable Development Goals. The same is true for the triple challenge, which is closely related to the SDGs. Ambition and leadership are important because developing better policies for food systems may run into a host of obstacles. Scientific evidence (e.g. about interactions) may not always be available; there may be strong disagreement about which policy objectives deserve to have priority; or interest groups may organise to block policies which would be disadvantageous to them. Robust policy processes will be important in overcoming these obstacles, as explored in the following chapter.
Figure 2.1. Principles for policy coherence
Note: See main text for further information and specific policy approaches.
References
[49] Agriculture and Agri-Food Canada (2019), Everyone at the Table! Government of Canada announces the first-ever Food Policy for Canada, https://www.canada.ca/en/agriculture-agri-food/news/2019/06/everyone-at-the-table-government-of-canada-announces-the-first-ever-food-policy-for-canada.html (accessed on 23 September 2020).
[39] Alston, J., J. MacEwan and A. Okrent (2016), “Effects of U.S. public agricultural R&D on U.S. obesity and its social costs”, Applied Economic Perspectives and Policy, Vol. 38/3, pp. 492-520, http://dx.doi.org/10.1093/aepp/ppw014.
[37] Alston, J., D. Sumner and S. Vosti (2008), “Farm subsidies and obesity in the United States: National evidence and international comparisons”, Food Policy, Vol. 33/6, pp. 470-479, http://dx.doi.org/10.1016/j.foodpol.2008.05.008.
[31] Anderson, K., G. Rausser and J. Swinnen (2013), Political economy of public policies: Insights from distortions to agricultural and food markets, http://dx.doi.org/10.1257/jel.51.2.423.
[74] Angrist, J. and J. Pischke (2009), Mostly Harmless Econometrics: An Empiricist’s Companion, Princeton University Press.
[68] Arrow, K. (1951), Social Choice and Individual Values, John Wiley & Sons.
[69] Arrow, K., A. Sen and K. Suzumura (2002), Handbook of social choice and welfare, Elsevier.
[32] Bouët, A. and D. Laborde (2017), Agriculture, Development and the Global Trading System: 2000 - 2015, http://dx.doi.org/10.2499/9780896292499.
[55] Candel, J. (2019), “The expediency of policy integration”, Policy Studies, pp. 1-16, http://dx.doi.org/10.1080/01442872.2019.1634191.
[56] Candel, J. (2017), “Holy Grail or inflated expectations? The success and failure of integrated policy strategies”, Policy Studies, Vol. 38/6, pp. 519-552, http://dx.doi.org/10.1080/01442872.2017.1337090.
[13] CFS (2014), Principles for Responsible Investment in Agriculture and Food Systems, http://www.fao.org/3/a-au866e.pdf (accessed on 25 September 2020).
[44] Dryzek, J. and C. List (2003), “Social Choice Theory and Deliberative Democracy: A Reconciliation”, British Journal of Political Science, Vol. 33, pp. 1-28, http://dx.doi.org/10.1017/S000712340300001.
[34] Elinder, L. (2005), Obesity, hunger, and agriculture: The damaging role of subsidies, http://dx.doi.org/10.1136/bmj.331.7528.1333.
[47] European Commission (2017), Chapter III Guidelines on impact assessment, European Commission, https://ec.europa.eu/info/sites/info/files/better-regulation-guidelines-impact-assessment.pdf (accessed on 23 September 2020).
[43] Giner, C. and J. Brooks (2019), “Policies for encouraging healthier food choices”, OECD Food, Agriculture and Fisheries Papers, No. 137, OECD, Paris, https://www.oecd-ilibrary.org/agriculture-and-food/policies-for-encouraging-healthier-food-choices_11a42b51-en (accessed on 5 November 2019).
[50] Government of Canada (2019), The Canadian Food Policy Advisory Council - Canada.ca, https://www.canada.ca/en/campaign/food-policy/thecanadianfoodpolicyadvisorycouncil.html (accessed on 23 September 2020).
[48] Government of Canada (2018), What we heard: Consultations on ’A Food Policy for Canada’, https://www.canada.ca/content/dam/aafc-aac/documents/20181025-en.pdf (accessed on 23 September 2020).
[70] Grethe, H. (2017), “The Economics of Farm Animal Welfare”, Annual Review of Resource Economics, Vol. 9/1, pp. 75-94, http://dx.doi.org/10.1146/annurev-resource-100516-053419.
[19] Konrad, M. et al. (2014), “Cost-efficient reductions in nutrient loads; identifying optimal spatially specific policy measures”, Water Resources and Economics, Vol. 7, pp. 39-54, http://dx.doi.org/10.1016/J.WRE.2014.09.001.
[61] Lankoski, J., A. Ignaciuk and F. Jésus (2018), “Synergies and trade-offs between adaptation, mitigation and agricultural productivity: A synthesis report”, OECD Food, Agriculture and Fisheries Papers, No. 110, OECD Publishing, Paris, http://dx.doi.org/10.1787/07dcb05c-en.
[62] Lankoski, J. et al. (2018), “Modelling Policy Coherence Between Adaptation, Mitigation and Agricultural Productivity”, OECD Food, Agriculture and Fisheries Papers, No. 111, OECD Publishing, Paris, http://dx.doi.org/10.1787/ee62a5ae-en.
[24] Lankoski, J. et al. (2018), “Modelling Policy Coherence Between Adaptation, Mitigation and Agricultural Productivity”, OECD Food, Agriculture and Fisheries Papers, No. 111, OECD Publishing, Paris, https://dx.doi.org/10.1787/ee62a5ae-en.
[71] Lusk, J. and F. Norwood (2011), “Animal Welfare Economics”, Applied Economic Perspectives and Policy, Vol. 33/4, pp. 463-483, http://dx.doi.org/10.1093/aepp/ppr036.
[30] Matthews, A. and R. Soldi (2019), Evaluation of the impact of the current CAP on the agriculture of developing countries, http://dx.doi.org/10.2863/81237.
[7] OECD (2020), OECD Best Practice Principles for Regulatory Policy: Regulatory Impact Assessment, OECD Publishing, Paris, https://doi.org/10.1787/7a9638cb-en.
[26] OECD (2020), Taxation in Agriculture, OECD Publishing, Paris, http://www.oecd.org/publications/taxation-in-agriculture-073bdf99-en.htm (accessed on 5 February 2020).
[53] OECD (2020), Taxation in Agriculture, OECD Publishing, Paris, https://dx.doi.org/10.1787/073bdf99-en.
[27] OECD (2020), Towards Sustainable Land Use: Aligning Biodiversity, Climate and Food Policies, OECD Publishing, Paris, https://dx.doi.org/10.1787/3809b6a1-en.
[25] OECD (2019), Agricultural Policy Monitoring and Evaluation 2019, OECD, Paris, https://www.oecd-ilibrary.org/docserver/39bfe6f3-en.pdf?expires=1567006470&id=id&accname=ocid84004878&checksum=E2DDB13143DDDB7027FFCD7BA0843E2B (accessed on 28 August 2019).
[5] OECD (2019), Enhancing Climate Change Mitigation through Agriculture, OECD Publishing, Paris, https://dx.doi.org/10.1787/e9a79226-en.
[23] OECD (2019), Impacts of agricultural policies on productivity and sustainability performance in agriculture: A literature review, http://www.oecd.org/tad/agricultural-policies/innovation-food-agriculture.htm. (accessed on 22 October 2019).
[3] OECD (2019), Innovation, Productivity and Sustainability in Food and Agriculture: Main Findings from Country Reviews and Policy Lessons, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/c9c4ec1d-en.
[57] OECD (2019), Measuring Distance to the SDG Targets 2019: An Assessment of Where OECD Countries Stand, OECD Publishing, Paris, https://dx.doi.org/10.1787/a8caf3fa-en.
[64] OECD (2019), Policy Coherence for Sustainable Development 2019: Empowering People and Ensuring Inclusiveness and Equality, OECD Publishing, Paris, https://dx.doi.org/10.1787/a90f851f-en.
[72] OECD (2019), Principles on Urban Policy and on Rural Policy, https://www.oecd.org/regional/ministerial/documents/urban-rural-Principles.pdf (accessed on 5 February 2020).
[1] OECD (2019), Recommendation of the Council on Policy Coherence for Sustainable Development, OECD, Paris, http://legalinstruments.oecd.org (accessed on 14 January 2020).
[28] OECD (2019), “The changing landscape of agricultural markets and trade: prospects for future reforms”, OECD Food, Agriculture and Fisheries Papers, No. 118, OECD Publishing, Paris, https://dx.doi.org/10.1787/7dec9074-en.
[52] OECD (2018), OECD Companion to the Inventory of Support Measures for Fossil Fuels 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264286061-en.
[10] OECD (2018), OECD Draft Policy Framework on Sound Public Governance, https://www.oecd.org/governance/policy-framework-on-sound-public-governance/ (accessed on 9 December 2019).
[45] OECD (2018), OECD Regulatory Policy Outlook 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264303072-en.
[67] OECD (2017), Draft OECD Best Practice Principles on Stakeholder Engagement in Regulatory Policy, http://www.oecd.org/governance/regulatory-policy/public-consultation-best-practice-principles-on-stakeholder-engagement.htm (accessed on 22 January 2020).
[59] OECD (2017), International Regulatory Co-operation and Trade: Understanding the Trade Costs of Regulatory Divergence and the Remedies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264275942-en.
[22] OECD (2017), Policy Coherence for Sustainable Development 2017: Eradicating Poverty and Promoting Prosperity, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264272576-en.
[40] OECD (2017), Preventing Policy Capture: Integrity in Public Decision Making, OECD Public Governance Reviews, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264065239-en.
[63] OECD (2017), Systems Approaches to Public Sector Challenges: Working with Change, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264279865-en.
[20] OECD (2016), “A new framework for policy coherence for sustainable development”, in Better Policies for Sustainable Development 2016: A New Framework for Policy Coherence, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264256996-6-en.
[21] OECD (2016), Better Policies for Sustainable Development 2016: A New Framework for Policy Coherence, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264256996-en.
[66] OECD (2016), International Regulatory Co-operation: The Role of International Organisations in Fostering Better Rules of Globalisation, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264244047-en.
[29] OECD (2013), Global Food Security: Challenges for the Food and Agricultural System, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264195363-en.
[65] OECD (2013), International Regulatory Co-operation: Addressing Global Challenges, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264200463-en.
[6] OECD (2012), Recommendation of the Council on Regulatory Policy and Governance, http://www.oecd.org/regreform (accessed on 9 December 2019).
[12] OECD (2011), OECD Guidelines for Multinational Enterprises, 2011 Edition, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264115415-en.
[18] OECD (2008), Multifunctionality in Agriculture: Evaluating the degree of jointness, policy implications, OECD Publishing, https://one.oecd.org/document/51200807/en/pdf (accessed on 20 September 2018).
[73] OECD (2008), OECD Ministerial Declaration on Policy Coherence for Development, http://www.oecd.org/dac/parisdeclaration (accessed on 5 February 2020).
[17] OECD (2003), Multifunctionality: The Policy Implications, OECD Publishing, http://www.oecd.org/agriculture/agricultural-policies/40782915.pdf (accessed on 20 September 2018).
[9] OECD (2002), Agricultural Policies in OECD Countries: A Positive Reform Agenda, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264199682-en.
[16] OECD (2001), Multifunctionality: Towards an analytical framework, OECD Publishing, http://www.oecd.org/tad/agricultural-policies/40782727.pdf (accessed on 20 September 2018).
[8] OECD (Forthcoming), OECD Guidance on International Regulatory Co-operation, OECD Publishing, Paris.
[11] OECD/FAO (2016), OECD-FAO Guidance for Responsible Agricultural Supply Chains, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264251052-en.
[46] Office of Management and Budget (2003), Circular A-4, https://obamawhitehouse.archives.gov/omb/circulars_a004_a-4/ (accessed on 23 September 2020).
[35] Okrent, A. and J. Alston (2012), “The effects of farm commodity and retail food policies on obesity and economic welfare in the united states”, American Journal of Agricultural Economics, Vol. 94/3, pp. 611-646, http://dx.doi.org/10.1093/ajae/aar138.
[54] Parsons, K. (2019), Rethinking Food Policy: A Fresh Approach to Policy and Practice - 3: Integrated Food Policy.
[2] Parsons, K. and C. Hawkes (2019), “Rethinking Food Policy: A Fresh Approach to Policy and Practice - Brief 5: Policy Coherence in Food Systems”.
[38] Pingali, P. (2015), “Agricultural policy and nutrition outcomes – getting beyond the preoccupation with staple grains”, Food Security, Vol. 7/3, pp. 583-591, http://dx.doi.org/10.1007/s12571-015-0461-x.
[77] Placzek, O. (2021), Socio-economic and demographic aspects of food security and nutrition.
[76] Quote Investigator (2011), Everything Should Be Made as Simple as Possible, But Not Simpler, https://quoteinvestigator.com/2011/05/13/einstein-simple/ (accessed on 9 December 2019).
[36] Rickard, B., A. Okrent and J. Alston (2013), “How have agricultural policies influenced caloric consumption in the United States?”, Health Economics, Vol. 22/3, pp. 316-339, http://dx.doi.org/10.1002/hec.2799.
[42] Sassi, F. et al. (2018), Equity impacts of price policies to promote healthy behaviours, Lancet Publishing Group, http://dx.doi.org/10.1016/S0140-6736(18)30531-2.
[60] Schaeffer, P. (n.d.), A Note on the Tinbergen Rule, https://www.petervschaeffer.com/uploads/7/4/3/3/74334295/a_note_on_the_relevance_of_tinbergen.pdf (accessed on 2 December 2019).
[15] Tinbergen, J. (1952), On the Theory of Economic Policy., North-Holland Publishing Company, Amsterdam.
[51] UNEP/SFSP (2019), Collaborative Framework for Food Systems Transformation A multi-stakeholder pathway for sustainable food systems, United Nations Environment Programme and One Planet Network Sustainable Food Systems Programme, http://www.oneplanetnetwork.org/initiative/setting-table-our-children-improving-governance-food-systems- (accessed on 5 February 2020).
[41] USDA (2019), Food spending as a share of income declines as income rises, https://www.ers.usda.gov/data-products/chart-gallery/gallery/chart-detail/?chartId=58372 (accessed on 24 January 2020).
[14] van Tongeren, F. (2008), “Agricultural Policy Design and Implementation: A Synthesis”, OECD Food, Agriculture and Fisheries Papers, No. 7, OECD Publishing, Paris, https://dx.doi.org/10.1787/243786286663.
[4] Vanheukelom, J., J. Mackie and M. Ronceray (2018), “Good Enough Coherence”, in Policy Coherence for Sustainable Development 2018, OECD.
[58] von Lampe, M., K. Deconinck and V. Bastien (2016), “Trade-Related International Regulatory Co-operation: A Theoretical Framework”, OECD Trade Policy Papers, No. 195, OECD Publishing, Paris, https://dx.doi.org/10.1787/3fbf60b1-en.
[33] Wallinga, D. (2010), Agricultural policy and childhood obesity: A food systems and public health commentary, http://dx.doi.org/10.1377/hlthaff.2010.0102.
[75] WTO (2020), Export subsidies and other export support measures, https://www.wto.org/english/tratop_e/agric_e/factsheetagric17_e.htm (accessed on 22 September 2020).
Notes
← 1. This chapter builds on insights from earlier work by the OECD, e.g. the OECD Ministerial Declaration on Policy Coherence for Development (OECD, 2008[73]), work on multifunctionality (OECD (2001[16]), (2003[17])) and the links between climate change adaptation, mitigation, and agricultural productivity (Lankoski et al. (2018[61]), Lankoski et al. (2018[62]), as well as between climate change mitigation and other objectives such as food security (OECD, 2019[5]); and on insights from the Collaborative Framework for Food Systems Transformation developed by UNEP (2019[51]).
← 2. The 17 Sustainable Development Goals and 169 associated targets cover a wide range of policy areas. To achieve coherence across these policy areas, SDG Target 17.14 calls on all countries “to enhance policy coherence for sustainable development”. For discussions on policy coherence related to food security and SDG 2 (End hunger, achieve food security and improved nutrition and promote sustainable agriculture), see OECD (2017[22]) and OECD (2016[20]).
← 3. The OECD also supports countries with the practical implementation of these Recommendations. In addition to serving as a forum for the exchange of experiences, the OECD provides a Policy Coherence for Sustainable Development Toolkit, which includes practical guidance, self-assessment checklists, good practice examples and tools to analyse, enhance and track progress on policy coherence in the implementation of the SDGs (see http://www.oecd.org/governance/pcsd/toolkit/). Moreover, an annual report on Policy Coherence for Sustainable Development reviews progress and highlights specific topics (OECD, 2019[64]).
← 4. The quote “everything should be made as simple as possible, but not any simpler” is often attributed to Albert Einstein, although it is doubtful whether he indeed coined this phrase (Quote Investigator, 2011[76]).
← 5. The impact of stricter animal welfare standards on production costs is clearly context-dependent, yet it is not uncommon for additional costs of some animal welfare standards to reach 30-40% of the “standard” production cost (Grethe, 2017[70]). The net impact on consumer prices will generally be smaller, as the final price for consumers includes other costs (e.g. for processing, transport) which are not much affected by stricter animal welfare standards. For a discussion of societal cost-benefit analyses taking into account animal welfare, see Lusk and Norwood (2011[71]).
← 6. The phrase “silver bullet” is defined by Merriam-Webster as “something that acts as a magical weapon, especially one that instantly solves a long-standing problem.” The phrase derives from the folklore belief that only a bullet made of silver could kill werewolves or other supernatural beings.
← 7. For a graphical exposition of the principle, see Schaeffer (n.d.[60]).
← 8. To be sure, assessing the effectiveness of a policy is never straightforward, as it requires not only accurate measurements of the current situation but also an assessment of what would have happened in the absence of the policy (i.e. the “counterfactual”) (Angrist and Pischke, 2009[74]). But it is conceptually and statistically easier to assess whether a single policy instrument has made a positive impact on a single objective than it is to assess the effectiveness of multiple policy instruments in achieving multiple objectives. With multiple objectives, it is possible that some objectives are met while others are not, making it harder to unambiguously define success or failure; with multiple policy instruments, it is less clear which instruments were effective and which ones were not. This is not an argument against the use of multiple instruments for multiple objectives per se; as noted, when synergies or trade-offs exist, this may be unavoidable. However, it does generally complicate the assessment of policy effectiveness.
← 9. This example can be seen as an instance of the more general principle that policies and governance should be organised at the relevant geographic scale, as highlighted by the OECD Principles on Rural Policy (OECD, 2019[72]).
← 10. On whole-of-government processes, see OECD (2018[10]) and OECD (2012[6]); on regulatory impact assessment processes, see OECD (2008[18]); on stakeholder engagement, see OECD (2017[67]).
← 11. Other examples of studies on policy incoherence in food and agriculture are discussed in Parsons and Hawkes (2019[2]).
← 12. The agreement also foresees the elimination of export subsidies by developing countries, albeit with some additional flexibility; for example, developing countries can cover marketing and transport costs for agriculture exports until the end of 2023 (WTO, 2020[75]).
← 13. For example, the OECD Best Practice Principles on Regulatory Impact Assessment stipulate that all plausible alternatives, including non-regulatory solutions, must be taken into account during the policy process. A “do-nothing option” describing the assumed state of the world in the absence of the regulation (the “counterfactual,” or “baseline”) should always be included. RIAs should examine human welfare differences among alternative policies. Thus, RIAs should not just consider the preferred regulatory approach but should consider all plausible alternatives (OECD, 2020[7]).
← 14. As discussed earlier, targeting may itself be costly. A tax on emissions may require detailed monitoring (e.g. measuring methane emissions on-farm). If monitoring activities are very costly, a less-targeted tax may be preferable, although there will be other trade-offs in terms of farm income and potentially nutrition.
← 15. As noted by Sassi et al. (2018[42]), a full accounting of equity effects should include the benefits accruing to different households as a result of the reduction in unhealthy consumption patterns. These benefits may also be concentrated among lower-income households, and may include lower out-of-pocket medical expenses and higher earnings (as a result of longer life expectancy), in addition to the intrinsic value of better health outcomes. As shown in forthcoming OECD work, households with a lower socio-economic status tend to make less healthy food choices for a variety of reasons, including low levels of income and education, stress and time constraints related to being a single parent, and the prevalence and accessibility of unhealthy food options (Placzek, 2021[77]).
← 16. Seminal work by Arrow (1951[68]) demonstrated that aggregating individual preferences to arrive at a consistent set of societal preferences is fraught with difficulties. A simple example can illustrate the problem. Imagine a society with three individuals using majority voting to decide between three alternatives A, B, and C. If individual 1 prefers A over B over C while individual 2 prefers B over C over A, and individual 3 prefers C over A over B, then majority voting will lead to instability: there will be a two-thirds majority choosing A over B, B over C, but C over A. This example (known as the Condorcet Paradox) was generalised by Arrow and gave rise to an extensive literature investigating the problem of “social choice”; see Arrow et al. (2002[69]).
← 17. Dryzek and List (2003[44]) demonstrate formally how these effects of deliberation can soften or overcome the problems identified in the social choice literature following Arrow.
← 18. In the field of complex systems analysis, a similar principle is known as Ashby’s law, which states that a control system must be as complex as the system it is controlling (OECD, 2017[63]).
← 19. For a discussion of the difficulties involved in measuring transboundary effects of policies in the context of the SDGs, see OECD (2019[57]).
← 20. See, for example, OECD (2013[65]) and OECD (2016[66]) for a discussion of different mechanisms.