Current public policy does not create the conditions people need to be healthy. In this context, social prescribing, which identifies unmet needs in clinical care and links individuals to community-based resources that help meet those needs, can help mitigate the health consequences of poor public policy. This is important, and social prescribing programs should be expanded. However, social prescribing does not fundamentally reconfigure the power relations and distributive institutions that create the problems in the first place. On the other hand, the welfare state, by publicly providing necessary goods and services that are not well allocated by markets and distributing income to those unable to engage in paid labour, can do this. Thus, the welfare state has a central role in improving population health, and efforts to enhance the welfare state should occur alongside efforts to boost social prescribing.

Introduction

In her recent paper, Social Prescribing as Strategic Infrastructure: Unlocking Community Capacity for Sustainable Healthcare When Institutions Reach Their Limits, Mulligan advances the thesis that increased funding for social prescribing programs will both improve population health and catalyze a needed transition from the current “welfare state” paradigm to an “enabling state” paradigm that focuses on public finance of local, community-based, service delivery infrastructure (Mulligan 2026). In this invited rejoinder, I want to respond to two central arguments in Mulligan's paper, one that I wholeheartedly agree with, and one that I think it would be fruitful to push back on. The argument that I agree with is that public financing of social prescribing should be expanded. The argument I would like to push back on is that to improve population health, the welfare state should be transcended to become an enabling state.

Social Prescribing to Mitigate the Consequences of Poor Public Policy

Current public policy in both Canada and in the US (where I am based) does not do a good job of creating the conditions necessary for health (Beckfield and Bambra 2016; Berkowitz 2026; Bor et al. 2026; OECD 2025; Public Health Agency of Canada 2024; Quick et al. 2025; Tarasuk and McIntyre 2025). Furthermore, the roots of poor health largely lie outside of clinical medicine, and clinicians currently have few tools to address many of the factors that harm people's health. In this setting, social prescribing can be an important intervention to mitigate the health consequences of poor public policy.

Social prescribing refers to structured programs that identify unmet social needs (typically in a clinical context) and use “link workers” to connect people to community resources that help meet these needs (Canadian Institute for Social Prescribing 2026; The Lancet 2025). For example, a social prescribing model might screen for food insecurity in clinics, and match those who screen positive with link workers such that they together find community-based resources to address the issue (e.g., a food pantry). The set of needs that might be addressed is broad, including both material needs, such as food, housing and transportation, and psychosocial needs, such as human connection and creative expression. If the “resource landscape” is such that there are effective community-based resources for individuals to receive assistance from, social prescribing can improve clinical outcomes. Indeed, there is growing evidence that healthcare-involved interventions to address factors like food insecurity, housing instability and transportation barriers can improve chronic condition management, mental health, healthcare utilization and healthcare spending (Berkowitz et al. 2025; Centers for Medicare & Medicaid Services n.d.; Hager et al. 2025; Renaud et al. 2023). Moreover, in the absence of social prescribing interventions, clinicians typically have few tools to address social circumstances that undermine what would otherwise be effective clinical care. As a primary care physician, I know first-hand that what I can offer in clinic often has little effect on health outcomes unless adverse social conditions change.

Having tools to address the clinical consequences of health-harming social conditions is important. However, it is also important to remember that such interventions do not address the fundamental causes of those conditions (Berkowitz 2023; Link and Phelan 1995; Lynch 2021; Raphael 2015). Thus, while there is much to be said for expanding social prescribing, we should not let the quest to do so take away from efforts to create the social conditions needed for better population health.

The Institutions Necessary for Population Health

If social prescribing does not address the fundamental causes of health-harming social circumstances, a reasonable next question might be, “what would address those causes?” The answer to this question lies in both social epidemiology and public policy (Beckfield et al. 2015). Social epidemiology has documented that relational inequality (i.e., social conditions characterized by hierarchies and relationships of domination and subordination) results in poor health through inegalitarian distributions of power and resources (Berkman et al. 2014; Diez Roux 2022; Link and Phelan 1995; Marmot and Bell 2012). The key way to improve population health, then, is public policy directed at situating individuals in relationships of equality with each other, ensuring both the social conditions needed for health and the capabilities needed to participate as free and equal members of society (Berkowitz 2024a). This means creating and maintaining institutions that make substantive the inherent moral equality of all individuals.

There are several spheres for such public policy. In the political and civil sphere, egalitarian social movements have rightly insisted upon establishing a full complement of political and civil liberties for all (such as rights to vote, run for and hold office, freedom of conscience and association, due process and equal protection under the law) (Anderson 1999). Though I will not discuss these more here, such rights are of foundational importance to relational equality.

Another important area is the economic sphere. Here, the goal of public policy is to make arrangement for public provision of necessary goods and services that are not well-distributed by neoclassical markets (particularly education, healthcare and personal care), and to ensure adequate income distribution such that all individuals can purchase those goods and services needed for health that are well-distributed by markets (e.g., food) (Berkowitz 2024a).

Some goods and services should be allocated publicly. For example, education (especially primary and secondary education) and healthcare are characterized by pervasive information imperfections, incomplete contracting, failures of the price mechanism and difficulties in judging quality, among other issues. This means that these services, when considered as economic commodities, do not fit neatly into expected utility theory, which is what underlies the idea that neoclassical markets distribute goods and services (allocatively) efficiently (Arrow 1963; Barr 2020; Henderson et al. 2000). This, in general, means that public provision will be preferable to market provision (Barr 2020; Henderson et al. 2000). Canada has, notably, made great strides in public provision for these services.

For other goods and services, the issue is not so much that markets cannot be made to work well for their distribution, but rather that public policy must ensure people have the necessary purchasing power to purchase them. The income distribution policies needed to provide that purchasing power differ for those engaging in paid labour and those who are unable to. In Canada, as in the US, about 50% of the population in any given year engages in paid labour, and about 50% does not (Berkowitz 2022; Statistics Canada 2022).

For those engaging in paid labour, a key public policy goal is to establish a balance of power between employers and employees. There are many ways to accomplish this, but approaches that have worked historically have included facilitating collective and sectoral bargaining, workplace health and safety standards and minimum wage regulations (Berkowitz 2024a). Moreover, it is important to establish robust social insurance systems (discussed in more detail below) so that workers have real negotiating power. Finally, full employment macroeconomic policy can help overcome the monopsony power of employers (Autor et al. 2023; Berkowitz 2024b).

For those unable to engage in paid labour, an effective way to ensure needed income is to establish a robust system of social insurance (Berkowitz 2025). Here, social insurance can be thought of as public programs that protect against “impairment in the capacity to work” (Rubinow 1913) or, in other words, the risk of having an insufficient income on account of being in a situation in which one cannot engage in paid labour.

It may have been surprising to read, a few paragraphs back, that approximately 50% of the population does not engage in paid labour in any given year. This is not because people are lazy or because “nobody wants to work anymore.” Instead, there are six key reasons people do not engage in paid labour, which fit naturally into the human life course (Berkowitz 2022, 2024a). The two most common reasons are age-related: either being below (roughly, age 18) or above (roughly, age 65) the age in which people are commonly expected to engage in paid labour. This accounts for about 40% of the population (Berkowitz 2022; Statistics Canada 2026). Furthermore, about 10% of the population is “working age” but is not currently engaging in paid labour because they are (1) a full-time student, (2) have a work-limiting disability, (3) are temporarily unemployed (i.e., actively looking for work) and/or (4) are engaging in socially necessary but unpaid caregiving, such as raising children or caring for loved ones (Berkowitz 2022). Notably, these reasons people are unable to engage in paid labour mirror situations that Canada already has some social insurance for: child allowances for younger age, pensions for older age, disability income for work-limiting disabilities, unemployment insurance, student grants and caregiving leave. However, the existing programs all have limitations such that they do not meet the needs of all individuals in these situations, and thus reform is warranted. Putting together a robust package of policies that can cover all the situations individuals may find themselves in across the life course would be an important way to help ensure the conditions needed for health (Bambra 2013).

The Welfare State and the Enabling State

The combination of public provision and social insurance described above can be characterized as the “welfare state” (and I very much mean the term non-pejoratively!). In my opinion, the welfare state is the central technology for improving population health. Reforms to the welfare state are what could reconfigure the distribution of power and resources that underlies so many health problems. It is important to note that what the welfare state concerns itself with is what any of us would spend most of our income on, whether provided publicly or purchased privately (Glennerster 2017). To see this, simply think of the proportion of one's monthly budget spent (whether that spending occurs as tax payments or private purchases) on essentials like food, housing, healthcare, education for ourselves or our dependents, and saving for retirement or a rainy day. Moreover, the economics are such that public approaches are typically more efficient than private ones, and thus can leave more discretionary income available even after these needs are met (Barr 2020; Glennerster 2017).

I bring this up because of the argument made in Mulligan's paper that it would be desirable to catalyze a transition from the welfare state to an enabling state, with the enabling state conceptualized as shifting government's role from providing services to financing community-based organizations to provide social prescribing services (Mulligan 2026). As noted before, I certainly do support more public financing of social prescribing services. But in deciding whether the welfare state should be supplanted by an enabling state, I think it is instructive to return to the two main objectives of the welfare state: public provision of necessary goods and services that are not well allocated by markets, and distributing income to those unable to engage in paid labour. Although an institutional structure focusing primarily on local, community-based organizations providing services to mitigate the consequences of adverse social conditions could be part of what the welfare state finances, it will still be needed to provide education and healthcare, and administer social insurance programs (Berkowitz 2023). And so there is no question, in my opinion, of replacing the welfare state with an enabling state. Although social prescribing can help mitigate the consequences of inequality, I am not sure it can do much to improve equality (Berkowitz 2024a; Raphael 2015). Reforms to the welfare state can, however, position people as equals. Therefore, we should not lose sight of the central role the welfare state plays in population health, and we should focus on strengthening areas of the welfare state that need improvement. Here, we could draw on international evidence regarding what works to improve population health. Key examples include the comprehensive social insurance approaches of the Nordic countries, the Netherlands and Belgium (Bambra 2013; Beckfield and Bambra 2016; Beckfield et al. 2015; Lynch 2021). These systems effectively avert conditions harmful to health, such as childhood poverty, food insecurity and low economic mobility, and which in turn contribute to excellent population health indicators, such as low maternal morbidity and high life expectancy (Berkowitz et al. 2024; OECD n.d., 2023, 2025; Parolin et al. 2024).

Conclusions

Current public policy creates social conditions that harm health, and social prescribing provides an important way to help mitigate those consequences. For that reason, expanding its use is likely to be health-promoting. But at the same time, to really address the root causes of poor population health, we need public policy that situates people in relationships of equality with each other, and provides the material basis for everyone to be healthy. For these reasons, the welfare state, which seeks to achieve these objectives, has a central role in improving population health.

Funding

This work received no specific funding.

Conflicts of Interest

Seth A. Berkowitz reports research grants from NIH, North Carolina Department of Health and Human Services, the American Heart Association, the American Diabetes Association and Blue Cross Blue Shield of North Carolina, consulting fees from the American Diabetes Association for serving as a deputy editor at Diabetes Care and royalties from Johns Hopkins University Press for sales of the book Equal Care: Health Equity, Social Democracy, and the Egalitarian State, outside of the submitted work. The content is solely the responsibility of the author and does not necessarily represent the official views of any organization.

Comments

Be the first to comment on this!