I received this article from the group that describes itself as MMT UK yesterday. It was written by Patricia Pino, a recognised expert on modern monetary theory, in response to my suggestions on the job guarantee.

It has been suggested by those in MMT UK that we are going to need to agree to disagree on this issue, but that nonetheless it is appropriate for both sides' arguments to be aired, and because this blog has such an impact on MMT debate, they did wonder whether I might share the post here, which they have already published on their own website. I have agreed to do that, in the spirit of what they note.

I will be writing a separate response to what Patricia had to say, but this is the article that she has written:

PATRICIA PINO

MMTUK.ORG 31 MARCH 2026

THE JOB GUARANTEE: A RESPONSE TO RICHARD MURPHY

Richard Murphy recently published four connected blog posts (one, two, three, four) critiquing the Job Guarantee. His engagement with these questions is welcome, given his prominence in bringing MMT ideas to a wider audience in the UK, and a conversation we had recently in Edinburgh suggested a genuine willingness to work through the disagreements. This is my attempt to continue that conversation. The concerns he raises span administrative feasibility, inflation, stigma, the proper role of fiscal policy, and the psychology of meaning. Taken together, they reveal a deeper disagreement about the nature of unemployment itself.

One question must be addressed first. Richard treats the Job Guarantee as a separable policy proposal: one favoured by some MMT proponents, but not intrinsic to MMT as such. This view is understandable, and many would prefer it to be correct. But it is difficult to sustain once MMT's monetary framework is followed through to its logical conclusion.

MMT holds that the requirement to meet tax obligations in the state's currency creates the structural conditions under which paid employment becomes necessary, not as individual choice, but as a system-wide condition. The state establishes, by design, demand for its currency, and therefore for the work required to earn it, regardless of whether any particular person is liable. In that context, a government that understands its own monetary capacity and then leaves people involuntarily unemployed is not simply failing on welfare grounds; it is withdrawing the means of meeting an obligation it has itself imposed.

The Job Guarantee completes that circuit and, crucially, anchors macroeconomic stabilisation in the labour market rather than the money market. Without the Job Guarantee, MMT is left endorsing the very mechanism it explicitly rejects: price stabilisation secured through money-market management with unemployment operating as the buffer stock, or assuming a level of fiscal fine-tuning that no government has ever demonstrated in practice: spending the right amount, of the right kind, at the right time, to perpetually maintain full employment.

This does not mean that MMT's insights about money creation and fiscal space depend on accepting the Job Guarantee. They do not. But for as long as the state imposes its currency through taxation, the Job Guarantee follows from that logic, regardless of the preferences of its advocates.

With that established, Richard's specific objections deserve direct engagement.

Richard's administrative objection is straightforward: the continuous identification, design, and supervision of meaningful, non-displacing work in ever-evolving conditions represents a challenge of considerable scale. This is true, but it is not a decisive objection. The NHS is complex. The education system is complex. Complexity has never been a sufficient reason to reject an institution whose function is considered essential. The relevant question is whether guaranteeing access to work is deemed sufficiently important to justify organisation at that scale.

Nor is the Job Guarantee conceived as a centrally micromanaged system. The model proposed in the UK, and reflected in international experience, is deliberately decentralised: national funding combined with local delivery. 

Communities, local authorities, and civil society organisations design roles around unmet needs. Rather than technocratic job creation from above, it is a framework that enables community established socially useful activity, much of it already taking place informally and without recognition, to be formalised and supported. The Argentine Jefes y jefas experience demonstrates that such programmes can operate at scale, and that participation naturally expands and contracts with broader economic conditions. The Job Guarantee does not need to place every worker in the perfect role; it needs a standing institutional mechanism that can generate and adapt work to local needs.

On inflation, Richard's concern is that a “uniform labour buffer” cannot stabilise prices given the diversity of the workforce, and that inflation is often driven by factors unrelated to wages. It is true that the workforce is diverse and that inflation is not limited to wage pressures, but neither addresses what the Job Guarantee actually claims.

The Job Guarantee function is narrower than Richard suggests: to stabilise demand and provide a nominal anchor through a publicly defined wage floor, replacing unemployment with employment at a fixed rate. Labour markets are diverse, but diversity does not undermine the anchor; the Job Guarantee is the benchmark against which all other wages are set, because employers must offer above it to attract workers out of the public programme. Inflation generated by supply shocks, geopolitical disruptions, industry bottlenecks, or market concentration requires other instruments. The Job Guarantee addresses a specific problem, not every problem.

Richard's claim that Job Guarantee participants would be trapped in a lower-status parallel labour market is difficult to reconcile with how the programme functions. Participation rises during downturns and falls as private-sector employment recovers. Employers tend to prefer hiring from those already in work, which means that Job Guarantee participants are better placed than the unemployed when private sector opportunities do arise. The relevant comparison is not between the Job Guarantee and some idealised labour market, but between the Job Guarantee and the current system, where the floor is unemployment itself, a condition defined by income loss, exclusion, stigma and deskilling. For those individuals who remain in the programme in the long term, guaranteed employment still provides income, participation, and purpose. That is a better outcome than a system that accepts permanent unemployment as normal.

Richard's preferred alternative, centred on investment, industrial strategy, and broad fiscal expansion, is not wrong. It is genuinely important. But it is not a substitute for the Job Guarantee. Public investment operates at the project and sector levels. It cannot guarantee that every individual will be matched to employment. Differences in skills, geography, timing, and personal circumstance ensure that some people remain outside even the best-designed investment programmes. Broader fiscal policy can reduce this mismatch. It has never eliminated it.

This brings us to the central disagreement.

Richard's framework is fundamentally project-led. It begins with programmes and asks how labour can be matched to them. The Job Guarantee, by contrast, is person-led. It begins with people and asks how the economy can be organised to include them. This is not a disagreement about the purpose of public policy, but about how access to work is organised.

In a project-led system, every programme faces a dual constraint: it must deliver its intended social purpose while also matching the workers available to it. These objectives do not always point in the same direction, and when they conflict, some mismatch is structurally inevitable. Richard's position is that broader fiscal policy can, in practice, reduce this mismatch to negligible levels. But the complexity does not disappear, it is simply left unaddressed. People are not static. Their circumstances, health, capabilities, and needs evolve continuously, and no programme designed around fixed project requirements can fully respond to such variation. The same ever-evolving conditions Richard cites as a reason to doubt the Job Guarantee apply equally to his own preferred approach.

The difference is that the Job Guarantee is explicitly designed to catch those who fall through the gaps. It does not assume mismatch can be engineered away. It ensures that where alignment between projects and people fails, access to work does not.

There is a striking asymmetry in this debate. The evidence for the Job Guarantee is incomplete, but it exists. Programmes in Argentina, India, Austria and elsewhere demonstrate that employment guarantees can function in practice, expand and contract with economic conditions, and deliver measurable improvements in wellbeing and participation. By contrast, the claim that a sufficiently broad set of other fiscal tools can eliminate unemployment altogether remains largely theoretical. Richard calls for “a better approach” but does not specify how the residual unemployment would be handled. The implication is that it can be reduced to a tolerable level. But tolerable to whom?

Richard's final and most philosophically ambitious objection draws on Viktor Frankl: namely, that work created outside existing structures risks being experienced as meaningless because it may lack the forms of social validation that those structures normally provide. This is a serious point, but the premise needs to be examined.

Richard's own approach rightly seeks to realise socially necessary work through improved public programmes and investment. But the boundary of what gets included is shaped by institutional design and practical constraints. Some of the most necessary forms of work do not map neatly onto planned projects. The Job Guarantee does not create meaning from nothing. It confers institutional legitimacy to contributions that already matter to communities, but which existing institutions have not been able to reach. Social recognition matters, but it is not fixed. It is shaped by institutions and institutional capacity. The Job Guarantee extends that capacity.

The empirical evidence also cuts against Richard's concern. Participation in public employment programmes is consistently associated with improved wellbeing and a stronger sense of purpose. A recent Job Guarantee trial in Austria (one of the most rigorously evaluated experiments of its kind) found participants reporting meaningful work and measurable improvements in mental health. The psychological harm in unemployment lies in exclusion, inactivity, and loss of contribution. The Job Guarantee addresses all three.

Richard is right that the meaning of work matters. He is right that social recognition matters too. But recognition is not the preserve of markets or conventional public investment alone. The Job Guarantee extends it to those from whom it is otherwise withheld, and the evidence suggests that participants are fully capable of recognising the difference.

PDF of article