Most of us were taught a tidy story: when unemployment falls, inflation rises, and vice versa. This trade-off — called the Phillips Curve — has anchored central bank policy for decades. But what if that story is not just incomplete, but fundamentally misleading?

A recent paper published in Realidad Económica by José Mauricio Gómez Julián argues exactly that. Using over fifty years of U.S. data (1968–2021), the study finds no significant long-run relationship between inflation and unemployment. Instead, it identifies a surprising positive link between technological change and inflation — and uses that finding to build a Marxist reinterpretation of what inflation actually does inside a capitalist economy.

It’s a paper that challenges both mainstream economics and the popular imagination. Let me walk you through it.

The Phillips Curve: A Love Story with Complications

In 1958, New Zealand economist A.W. Phillips noticed an elegant regularity in British data: wages tended to rise faster when unemployment was low. Later economists generalized this into a policy menu: want less unemployment? Accept a bit more inflation. Want to tame prices? Brace for a recession.

This trade-off became gospel in the 1960s. Central bankers thought they could fine-tune the economy like a thermostat — dial inflation up or down by adjusting demand. But the 1970s shattered that confidence. The U.S. experienced stagflation: high inflation and high unemployment at the same time, something the Phillips Curve said shouldn’t happen.

Since then, economists have debated whether the Phillips Curve is dead, dormant, or merely sleeping. Gómez Julián sides with a more radical verdict: the long-run Phillips Curve doesn’t just flatten — it was never there to begin with.

The Data, the Tools, and What They Found

The study uses three complementary statistical approaches — each chosen for a reason:

Bayesian Correlations

Unlike classical statistics, which gives you a yes-or-no answer (“significant at 5%”), Bayesian analysis lets you say something more nuanced: “Given the data, here is the probability that this relationship is positive, negative, or nonexistent.” Applied to U.S. inflation and unemployment, the Bayesian results show no consistent inverse relationship. The data simply doesn’t support the Phillips Curve story with any confidence.

Granger Causality

This is a standard econometric test that asks: does knowing today’s unemployment help you predict tomorrow’s inflation (or vice versa)? If the Phillips Curve were real, the answer should be yes. Gómez Julián finds that the answer is generally no. Unemployment does not Granger-cause inflation in the U.S. data. What does show predictive power? Research and development spending.

Error Correction Models (ECM)

These models examine whether variables that drift apart over time eventually pull back together — like two dancers who briefly separate but remain on the same floor. The ECM results confirm that inflation and unemployment do not share a stable long-run equilibrium. They are, statistically speaking, dancing to different music.

The Surprising Link: Technology Drives Inflation

Here is the paper’s most provocative finding: R&D expenditure and inflation move together positively. When firms invest more in technology, inflation tends to rise — not fall, as you might expect from a productivity-enhancement standpoint.

Why would better technology lead to higher prices? To answer this, Gómez Julián turns to Marx — specifically, to the distinction between two types of surplus value.

Two Kinds of Surplus Value: A Quick Primer

If you’re not steeped in Marxist theory, don’t worry — the distinction is intuitive.

Absolute surplus value is what a capitalist gets by making workers work longer or harder for the same pay. It’s the old-fashioned squeeze. Relative surplus value, by contrast, comes from making production cheaper — through technology, efficiency, better organization — so that the value of labor-power (i.e., the cost of maintaining a worker) falls, even if wages don’t.

Now imagine a single firm introduces a breakthrough technology. It can produce goods faster and cheaper than its competitors. For a while, it earns extraordinary surplus value — a premium profit that exists only because it’s ahead of the pack. But here’s the catch: once competitors adopt the same technology, that advantage vanishes. The extraordinary surplus value disappears.

Gómez Julián’s argument is that inflation is the mechanism through which this temporary advantage gets converted into a permanent one. How? As the innovating firm’s higher productivity drives down unit costs, prices don’t fall proportionally — instead, the general price level adjusts upward. The gap between the old cost structure and the new one gets absorbed by inflation. What was a one-time windfall for the innovator becomes a structural shift in profitability for the entire capitalist class.

Inflation, in this reading, is not a policy error or a monetary accident. It is a functional mechanism of capitalist accumulation — one that converts technological advantage into lasting class-wide profit.

What This Means for Policy

If the paper is right, the implications are significant:

For central bankers: If inflation isn’t primarily a monetary phenomenon — if it’s rooted in the structural dynamics of production and profit — then raising interest rates to fight inflation is treating the symptom, not the disease. You might cool the economy, but you’re not addressing the engine that generates inflation in the first place.

For mainstream economists: The Phillips Curve may be less a stable empirical law and more a historical coincidence — a relationship that appeared to hold in a particular postwar context and has been propped up by theoretical convenience ever since. The paper adds to a growing body of evidence that the curve has become unreliable as a guide to policy.

For non-economists: This paper reframes inflation as a political question, not just a technical one. If inflation systematically benefits capital at the expense of labor — by preserving the gains of innovation for the capitalist class while workers’ purchasing power erodes — then debates about inflation are, at their core, debates about distribution and power.

A Challenge to Orthodoxy

What makes this paper worth reading — whether you agree with it or not — is that it does something many economists avoid: it takes a heterodox theoretical framework seriously and tests it empirically. This isn’t armchair Marxism. It’s Bayesian statistics, Granger causality, and error correction models applied to five decades of data. The methodology is conventional; the interpretation is not.

The mainstream view treats inflation as essentially a monetary phenomenon — too much money, not enough stuff. Milton Friedman’s famous dictum that “inflation is always and everywhere a monetary phenomenon” still echoes through central banks worldwide. Gómez Julián doesn’t deny that money supply matters. But he argues it’s not the whole story — and may not even be the most important part.

In his framework, the relationship between technology, surplus value, and prices is structural. It doesn’t depend on whether a central bank is dovish or hawkish. It’s embedded in the logic of capitalist production itself.

So, Is the Phillips Curve Dead?

Probably not entirely. There are short-run contexts where demand pressures do push prices up, and the Phillips Curve captures something real about those moments. But the paper pushes us to ask harder questions: What determines the baseline around which those fluctuations occur? Why has inflation behaved the way it has over half a century, regardless of the unemployment rate?

Gómez Julián offers a provocative answer: inflation is the economy’s way of metabolizing technological progress into profit. It’s not a bug in the system. It’s a feature.

Whether you find that convincing depends, in part, on your theoretical priors. But the data doesn’t lie about what it doesn’t show: a reliable Phillips Curve. And that, at minimum, should give everyone — mainstream, heterodox, and curious layperson alike — something to think about.