Frank
Vandenbroucke

Defending the welfare state in hard time

Sidney Ball Lecture, 7 November 2025

The background paper and presentation slides are available at the end of the speech.

Dear President, Fellows, Students, Ladies and Gentlemen,
It is a great honour to be invited for the Sidney Ball Memorial Lecture and to speak in Saint-John’s College. Your invitation triggered me to revisit ideas I have been working on, 30 years ago, here in Oxford: ideas on social justice and solidarity.

One thing is clear today. If we want our European democracies and welfare states to survive and flourish, we need more solidarity. External pressures like Russia’s aggression and the breakdown of the multilateral order, as well as internal pressure from an aging population and rising care needs, all demand greater solidarity. Fortunately, we’ve witnessed, recently, solidarity in action and seen its benefits. During the COVID-19 pandemic, collective action and solidarity —civic and professional—were crucial at both national and EU levels.

It is impossible to think consistently about solidarity without thinking consistently about personal responsibility. That is the thesis I’ve always defended and which I will revisit here. I want to reflect on what exact conception of responsibility best fits contemporary welfare states.

My lecture is based on a longer and more elaborate background paper. The paper is work in progress and I would very much like to receive your critical comments or questions, tonight orally, but also by mail. I want to invite you for an exercise in co-creation.

SLIDE 2: solidarity and responsibility: a long cycle

Since the early 1980s, when I began my career, we witnessed a long cycle in the dominant conception of the benefits and problems of welfare state solidarity. Over time, the emphasis shifted from recognizing the societal benefits of solidarity to worrying about the problem of moral hazard (or reduced personal responsibility). Now, in the wake of COVID-19, it seemed the pendulum was swinging back—less focus on moral hazard and more appreciation of the benefits of solidarity. The pandemic indeed underscored the importance of collective action and mutual support. However, today this hopeful trend appears to have been interrupted. The European Union is now prioritizing industrial competitiveness and defense spending over public health, and many national governments are tightening budgets and retrenching welfare programs. We see a renewed, one-sided focus on moral hazard and punitive conditions on social benefits. Do not misunderstand me. I am not suggesting we should revive some mythical era of unconditional solidarity. Rather, we must revisit the demands of solidarity and tailor them to the challenges of our time.

SLIDE 3: no solidarity without responsibility

Let me first explain how I propose to define and understand ‘welfare state solidarity’. Welfare states provide income security, relieve poverty, organize care, and invest in people’s capabilities, through social benefits, taxation, and public services like health care, childcare, and education. Solidarity is at work in all these areas: we pool resources to help those facing misfortune or disadvantage. I agree with the political philosopher Philippe Van Parijs that a defining feature of this institutionalized solidarity, the reason why we call it ‘solidarity’, is a symmetry. It is a symmetry best captured by the expression ‘mutual responsibility’. Solidarity means that we take responsibility for each other as members of some (more or less imagined) community. Van Parijs traces back this definition the French thinker, Prime Minister and Nobel Peace Prize winner, Léon Bourgeois.

We should also be clear about what we mean by ‘responsibility.’ By the end of this lecture, you will see that the expression ‘responsibility’ returns in four different connotations. The common denominator, the most generic understanding of what ‘responsibility’ is about, is that it refers to a legitimate expectation. We hold Peter responsible for something – say, an outcome, or a readiness to act in a certain way, or a certain type of behaviour – if there is a legitimate expectation viz-à-viz Peter with regard to that outcome, or readiness to act or behaviour.

Welfare state solidarity inherently involves responsibility in two key ways. First, by definition, it entails a mutual duty: I am expected to help you when you encounter bad luck, and you are expected to do the same for me. This responsibility-as-duty is the moral core of solidarity. Second, welfare solidarity usually comes with responsibility-as-accountability: we expect a certain behaviour from people who benefit from it. Our aid can come with conditions – for instance, someone receiving unemployment benefits should make a genuine effort to find work. That expectation is justified by a concern for moral hazard: i.e. our (perfectly rational) propensity to become less prudent and take more risk because we are to some extent protected against it.

Mitigating moral hazard is a pragmatic necessity if solidarity is institutionalized, if this institutional solidarity is to be maintained and, therefore, kept within certain limits. But, on a more fundamental level, there might be an argument of justice about moral hazard – depending on our conception of justice. If an individual’s hardship clearly results from their own choices, many would argue it is not society’s duty to subsidize that choice. This is the intuition behind so-called “luck egalitarian” theories of justice. Luck egalitarians – philosophers like Ronald Dworkin and G.A. Cohen, claim that solidarity should compensate people for (brute) bad luck, not for the (reasonably foreseeable) consequences of deliberate choices. A luck egalitarian, for example, would insist that unemployment benefits be conditional on job-search effort – not only to reduce moral hazard, but as a matter of justice. Without any job search, the unemployment can be considered as voluntary rather than a misfortune for which solidarity is due.

Let me read you what Cohen wrote, in 1989, about Dworkin’s theory of equality of resources, which can be seen as a starting point for luck egalitarianism: “Dworkin has, in effect, performed for egalitarianism the considerable service of incorporating within it the most powerful idea in the arsenal of the anti-egalitarian right: the idea of choice and responsibility.” Indeed, by qualifying the demands of just equality in this way, luck egalitarianism was able to escape familiar anti-egalitarian criticisms like the claim that it panders to undeserving poor at the expense of hardworking citizens, as Serena Olsaretti put is.

Should we apply a luck egalitarian perspective to solidarity? This question has exercised me for thirty years, since I had the privilege of working in the mid-1990s on a doctorate supervised by Jerry Cohen, who played a key role in the development of ‘luck egalitarianism’, and Tony Atkinson, a former Warden of Nuffield College whose work on inequality still inspires us today. That question exercised me when I introduced, for the first time, a comprehensive scheme to activate the unemployed in Belgium in 2003, and it exercises me again today, in view of the massive and growing caseload of people living on incapacity benefits for whom I want to implement a thorough reform of the system.

SLIDE 4: the benefits of solidarity

Before returning to that soul-searching question, I should mention another point. Welfare state solidarity is not purely altruistic; it also reflects enlightened self-interest. In practical terms, welfare state solidarity typically combines elements of insurance and redistribution. Social insurance programs pool risks across the population so that everyone contributes and anyone might benefit. Meanwhile, redistribution – asking the fortunate to support the less fortunate – may not seem to serve one’s own interest at first glance. But, economic theory has shown already a long time ago that without some degree of redistribution, adequate insurance for many risks would simply not be possible. And when a society faces a shock or recession, both insurance and redistributions act as economic stabilizers. In short, solidarity-driven policies not only promote social justice, they also create a more resilient society that benefits everyone.

Van Parijs captures well how solidarity differs from a simple insurance bargain. He contrasts the motives this way: “I help you because I might one day be in your position” (the insurance motive) versus “I help you because I assume I could have been you, even if I know that I shall never be in the sort of trouble in which you are now” (the solidarity motive). It is because of the counterfactual nature of this assumption that solidarity is “not purely selfish.”, unlike insurance. Yet an established practice of solidarity still provides each of us with a reduction of our individual vulnerability: we can trust that others will help if we fall into trouble, and they trust that we will do the same. Solidarity is beneficial for society at large.

Nevertheless, defending the benefits of solidarity has been an uphill battle during much of my professional career, notably in the 1980s.

SLIDE 5: decreasing popularity of insurance

The 1980s indeed marked the heyday of the neoliberal critique of the welfare state. According to neoliberalism welfare state solidarity set wrong incentives, was rife with moral hazard and hence a cost-factor that created a drag on economic performance. By the mid 1990s, neoliberalism was losing momentum, and a reappraisal of the merits of well-organised welfare states emerged. However, in that reappraisal much less emphasis was placed on the merits of insurance and redistribution of incomes. The ‘insurance paradigm’, associated with what we then started to coin as ‘old social risks’ was even seen as something of the past.

In contrast to the ‘old social risks’, such as old age, invalidity, or cyclical unemployment, social policy scholars identified a set of ‘new social risks’, such as obsolete skills, single parenthood or the need to care for a frail parent or partner. What you need foremost, in that new context, are education, training and activation rather than cash unemployment benefits. Childcare rather than child benefits. Elderly care, and not only retirement pensions. Social investment – services and investment in human capital – rather than social insurance.

Alongside this social investment focus came a continued emphasis on personal responsibility. Policymakers worried about “inactivity traps,” where benefits might discourage participation in activation trajectories and work, and so they tightened conditionality. The welfare state was no longer seen purely as a cost-factor, but as a productive factor, but in that ‘social investment welfare state’, benefit recipients were supported but they should seek employment or retrain.

The decreasing popularity of the insurance paradigm can also be explained by the simple fact that between the end of the 1980s and 2008, no major, large-scale economic shock hit Western Europe. The experience of surges in cyclical unemployment receded from memory, whilst structural unemployment and structural lack of opportunity became centre stage. The possible drawbacks of generous insurance systems were much higher on the agenda than their beneficial, stabilizing effects.

SLIDE 6: Covid-19 and the return of solidarity (?)

I’ll skip a part of the story which you may read in the background paper, but, five years ago, the COVID-19 pandemic drove home the benefits of robust social protection more powerfully than anyone could have predicted. The pandemic exposed risks that were clearly not the fault of individuals, and it highlighted how essential collective insurance mechanisms are in a crisis. Inclusive welfare systems moved from being scrutinized to being recognized as indispensable “shock absorbers.” Countries with comprehensive social programs were better able to respond: it’s much easier to tell people to stay home when they are maybe infected, or sick, or when their workplace is shut down if you have universal sick pay, unemployment benefits, or job-retention schemes in place. In 2020, governments across Europe, including my own, expanded income support overnight – covering furloughed employees, self-employed people, and others not traditionally protected – with hardly a mention of moral hazard. Keeping people financially secure during lockdowns wasn’t just about compassion; it helped control the virus’s spread and prevented economic freefall. In short, COVID-19 underscored the message that a degree of moral hazard is a small price to pay for a society that can protect its members and remain stable in the face of massive external shocks. Many observers hoped that this experience would permanently shift the policy narrative away from a punitive focus on individual blame and back toward an appreciation of solidarity.

In reality, the change in mindset after the pandemic has been mixed. As the pandemic subsided, the old narratives reasserted themselves. In my own country, Belgium, the public debate refocused on individual responsibility, especially regarding those on long-term benefits. Politicians drew sharp contrasts between “hard-working taxpayers” and “people living on benefits,” with loud calls to “responsibilize” the latter. Rather than expanding on the promise of solidarity seen during COVID, discussions returned to tightening benefit conditionality and reducing costs.

Obviously, that is not happenstance. It’s not simply the work of badly inspired people. Many European welfare states now face immense fiscal pressure in the wake of, first, COVID, and then the energy crisis from which they wanted also to protect their citizens. They have to invest in defence, there is no doubt about that. But they also have to invest in health and care. Simultaneously, we are confronted in many countries, certainly in Belgium but also in the UK, with a social problem that is in a sense an ‘old social risk’, but one that manifests itself now in a new way and shares many features with the ‘new social risks’ identified in the 1990s: that social problem is a steadily growing caseload of people living on long-term incapacity benefits. This is as much a health and health care problem, as it is a social and labour market problem.

In short, we are confronted with tough choices: how to reinforce solidarity and address emerging needs in “hard times.” This brings us back to the central question I promised to address: what conception of individual responsibility is most appropriate for today’s welfare states, given these challenges?

SLIDE 7: reinventing responsibility and solidarity in hard times (1)

You cannot have welfare state solidarity without responsibility. I defended that thesis for many years. Intellectually, I was influenced by the responsibility-sensitive egalitarianism of John Rawls, by Jerry Cohen’s criticism of Rawls and Cohen’s luck egalitarianism, which you may understand as a specific, rigorous strand of responsibility-sensitive egalitarianism. Very succinctly put: disadvantage that is the consequence of responsible choice is not unjust disadvantage. Simultaneously, I never thought that luck egalitarianism yielded the one and only principle to be applied in the praxis of social policy.

In a thought-provoking book published in 2017, Yascha Mounk questions the legacy of Rawls and challenges in particular the application of luck egalitarianism to social policy. He claims that luck egalitarianism asks the wrong question. Mounk shows how, since the 1970s responsibility – which once meant the moral duty to help and support others – has come to be equated with an obligation to be self-sufficient, how it shifted from ‘responsibility-as-duty’ to a narrow focus on ‘responsibility-as-accountability’. This simply served to justify welfare-state retrenchment. Hence, a fundamental change of course is needed. According to Mounk, left-wing thinkers, such as Cohen, made a fundamental error by tying ‘equality’ too tightly to ‘responsibility’. Returning to the quote on Dworkin by Cohen, one might say that instead of rendering a service to egalitarian thinking, Mounk implies (although he does not say this) that Dworkin surrendered to the anti-egalitarian right.

By asking the wrong question luck egalitarians create an impasse: in order to keep defending redistribution and solidarity within a luck-egalitarian framework, they must develop arguments that minimize the actual responsibility of individuals. But minimizing people’s own responsibility is implausible and contradicts what people themselves believe. Therefore, it is better to stop focusing on responsibility in this way, says Mounk. The responsibilities we attribute to people should flow from our vision of what a just society is. The values that play a role in such a society (such as: no poverty and oppression, equality of status) are the starting point; personal responsibility comes in only insofar as individual behaviour must meet certain expectations in order to sustain such a just society.

There is much to recommend in Mounk’s description of what social policy should do. Rather than punishing people for past choices, social policy should empower them to take responsibility for themselves and others in the future, it should help them to become self-reliant where possible – rather than punishing them for past decisions. This resonates with the philosophy of activation developed by Jean-Michel Bonvin, relying on Amartya Sen’s capability framework. I think that such a forward-looking, capacity-building approach is particularly relevant when dealing with people living on long-term incapacity benefits.

However, that does not set us free from the question of personal responsibility. We cannot avoid that, admittedly very complicated, question, if we care about solidarity and justice. What we must recognize, first of all, is that the answer will rarely be black-and-white. In real life, social disadvantages usually result from a mixture of bad luck and deliberate decisions. That is a first complication.

Another complication is that holding adults strictly accountable for their choices can end up harming people who had no choice at all – notably their children: if parents make bad decisions that lead to poverty, their children suffer the consequences through no fault of their own. You must not harm the children by withholding aid from the parents, even if that would be, as such, legitimate.

Most fundamentally, from an ethical point of view, attention to responsibility does not mean being ruthless towards human failings. Some consequences of individual choices – for which people are indeed responsible – some consequences can be so dramatic, and the vulnerability that results so fundamental, that we still wish to repair or compensate. An emergency ward receiving two gravely injured patients does not ask who respected the traffic rules. That question may arise later, in relation to the financial settlement of the accident, but not in medical care. The same is true in many other social policy contexts, where the question ‘how did you get into this situation?’ is not guiding the decision to help or not. Providing shelter to the homeless is an example. A policy of ‘harm reduction’ for drug addicts by organizing a safe consumption environment – whilst maintaining drug prohibition – might be another example. This intuition is about compassion. Protecting the vulnerable – sometimes regardless of the reason for their vulnerability – is a task that the community must take up as a positive responsibility, as Robert Goodin has argued persuasively 40 years ago, in a book which at that time also influenced me strongly.

One might say that compassion is a compelling duty, also for a policy-maker who values personal responsibility and agency highly, if someone’s vulnerability is so existential that it threatens to wipe out his or her agency completely. There can be no responsisbility without agency. Rescuing someone’s agency is the first task to be accomplished. Therefore, as a policymaker, I’ve always argued that social policy walks on two legs: both compassion and personal responsibility.

SLIDE 8: reinventing responsibility and solidarity in hard times (2)

In fact, my argument used to be ‘compassion’ that compassion limits the role of personal responsibility, or even cancels responsibility. However, that is maybe not the way to understand the philosophical problem at hand. Serena Olsaretti has argued convincingly that it is not necessary to deny someone’s responsibility, or to limit the role of responsibility in our conception justice, if we want to avoid harsh conclusions. Even if we accept the importance of holding individuals responsible for their choices, it is an open question what consequences those choices should have. In addition to a principle concerning the grounds of responsibility, we need a ‘principle of stakes’ (as Olsaretti calls it) to make our judgments of responsibility determinate, rather than to justify the placing of limits on responsibility. The ‘question of stakes’ is a question of what responsibility itself requires: when we defend a particular principle of stakes as superior to alternative ones, we are fleshing out, not constraining, the demands of responsibility.

In a somewhat similar vein, Zofia Stemplowska argues that for responsibility for disadvantage to justify that disadvantage, the options faced by people must be deemed just by independent criteria: there must be a just opportunity structure within which people make responsible choices.

In other words, several values are at play; such pluralism is inherent in philosophical reflection. Yes, a welfare state should encourage and even expect individuals to take responsibility where they can. But it must also guarantee a floor of protection for all, regardless of how or why someone ended up in need. We need to decide, as a matter of principle, how much a person’s behaviour should affect the support and protection they receive.

Returning to Mounk, I would say: it is not a wrong question to ask to what extent solidarity has to be mobilised if someone’s misfortune or disadvantage is, partly of wholly, the result of deliberate choice. However, it is not the only normative question.

Solidarity in a modern welfare state must be responsibility-sensitive without becoming responsibility-obsessed. We should neither ignore the role of personal responsibility nor let a punitive mindset undermine the very purpose of solidarity. Multiple principles are at play: personal accountability, yes, but also preventing poverty, protecting children, maintaining social stability, and upholding basic human agency and dignity. These values need to be reconciled. A pluralistic approach is inherent to any reasonable theory of social justice, including those that emphasize responsibility.

Shying away from ‘personal responsibility’ is not only philosophically unpersuasive; it is also politically unwise. Public support for the welfare state often hinges on perceptions of fairness and reciprocity. People are far more willing to pay into a system that, in their view, does not reward idleness or bad behavior without consequence. Voters ask: Are those who receive benefits doing their part, to the extent they are able? This “deservingness” sentiment is a reality that advocates of solidarity must address. Crucially, we should address it consistently. That means holding not just the poor and vulnerable to appropriate expectations, but also holding the rich and powerful to their responsibilities. If we demand that an unemployed worker actively seek a job as a condition of support, we can likewise demand that a profitable corporation or a wealthy individual pay their fair share in taxes to support the system. Responsibility in a democracy is a two-way street. As citizens, we owe duties to one another, and those duties apply to everyone, not only to those at the bottom.

In fact, appealing to a balanced and comprehensive notion of responsibility strengthens the case for solidarity. Most people accept that someone who genuinely cannot work should be supported—especially if that person has tried to contribute in the past. By the same token, people intuitively feel that those who have benefited most from society (the “winners,” so to speak) should contribute back in proportion. If ordinary hard-working people are expected to play by the rules and put in effort, they will also expect wealthy elites to pay what is due and not game the system. A discourse about responsibility that includes both rich and poor can unite the public around a sense of shared fairness: we all give, and we all take, according to our abilities and needs.

SLIDE 9: Co-responsibility and political responsibility

So far, I’ve argued that solidarity implies responsibility in two senses. The first is mutual responsibility. The most mundane demand of mutual responsibility is that we pay taxes and social security contributions to compensate the victims of misfortune or disadvantage, or, more generally, to support them in coping with misfortune or disadvantage. The second is personal accountability for the beneficiary of that compensation: disadvantage is, inevitably, a responsibility-sensitive concept.

In addition, here is a third sense in which solidarity needs responsibility.

To help people on long-term incapacity benefits rejoin the world of work if they still have that potential, it’s not enough to expect the individuals alone to try harder. Physicians must provide proper care and honest assessments, employers should offer adapted work or retraining opportunities, and public services need to support rehabilitation and job placement. If these actors shirk their part, an individual’s personal responsibility can achieve little. Only when all parties do their part can the fabric of solidarity remain strong and effective. There is a shared obligation, a co-responsibility of many actors and institutions to make solidarity work in practice. Using Stemplowska’s formula, I would say that the co-responsibility for creating and maintaining a just opportunity structure is a critical premise for true welfare state solidarity.

This brings me, finally, to a fourth notion of responsibility: the responsibility of the politician to sustain solidarity. I can tell you that this often requires difficult compromises. Sometimes one must accept a policy that is far less than ideal in order to safeguard the greater system of social support—choosing a lesser evil over allowing opponents of the welfare state to dismantle it entirely. It means fighting, also in the public debate, for a correct and proportionate conception of responsibility, that demands as much in terms of ‘duty’ from wealthy citizens who don’t need social programmes, as it demands in terms of ‘accountability’ from the often relatively poor beneficiaries of such programmes. It also means that one should not shy away from difficult debates about our co-responsibility for the precious good of solidarity – difficult since it involves a very wide range of actors and an ethos of daily conduct.

At the very end of my doctoral dissertation in 1999, I wrote:

“So, invoking responsibility must signal, simultaneously, two normative ideas: personal accountability for personal choice and a sense of social commitment (…). The egalitarian perspective that is the essence of social democracy requires both. This twofold ethos has to shape our society’s institutions and the way agents behave within those institutions. It calls upon the talented and the powerful, as much as it calls on the poor and the powerless.”

All this remains an uphill battle, more than ever. But the difficulty of the task does not release us from it.

SLIDE 10: Background paper

 

Background paper and slides: