Key Takeaways
- The Phillips Curve describes the empirical relationship between labour-market tightness and inflation — but it is a framework, not a fixed law.
- The expectations-augmented Phillips Curve shifts when inflation expectations change: the curve is stable only when expectations are well-anchored.
- The short-run Phillips Curve can trade off unemployment and inflation; the long-run curve is vertical at the natural rate — there is no permanent trade-off.
- The relationship has flattened in recent decades: falling unemployment has not always produced proportionate inflation, and supply shocks can shift the entire curve.
- Central banks still monitor labour-market tightness because wage-sensitive services inflation remains the most persistent component of overall inflation.
- Falling unemployment is not automatically currency-positive — what matters is how it changes the inflation outlook, central-bank expectations, and relative yields.
What Is the Phillips Curve?
The Phillips Curve is an empirical relationship first documented by economist A.W. Phillips, describing an inverse correlation between unemployment and wage inflation. When unemployment is low, workers have more bargaining power and wages rise faster. When unemployment is high, bargaining power is weak and wage growth is subdued. The original Phillips Curve related unemployment to wage growth; the modern version relates unemployment to price inflation. For the broader framework, see Labour Market Slack and Wage Growth for Forex Traders.
The Phillips Curve is not a theoretical law — it is an empirical regularity that has held in some periods and broken down in others. The relationship is not stable: it shifts with inflation expectations, supply shocks, and structural changes in the labour market. Understanding the Phillips Curve as a framework — not a formula — is essential for interpreting labour-market data and its inflation implications.
The Wage Phillips Curve
The original wage Phillips Curve relates unemployment to wage growth. The intuition is straightforward: when the labour market is tight (low unemployment), employers compete for workers by offering higher wages. When the labour market is loose (high unemployment), workers have little bargaining power and wage growth is subdued. The wage Phillips Curve is the microeconomic foundation of the inflation Phillips Curve — wages are the largest cost for most firms, and wage growth feeds into price inflation. For the framework, see Wage Growth for Forex Traders.
The wage Phillips Curve has been more stable than the price Phillips Curve, but it too has flattened. Globalisation, technology, and the decline of unionisation have weakened the wage-bargaining channel. A tight labour market no longer guarantees strong wage growth to the extent it once did. This is why central banks look at wage growth directly, not just at the unemployment rate as a proxy.
Labour-Market Tightness and Bargaining Power
The Phillips Curve relationship depends on labour-market tightness — the balance between labour demand and labour supply. When demand exceeds supply (vacancies are high relative to unemployment), the market is tight and wage pressure builds. When supply exceeds demand (unemployment is high relative to vacancies), the market is loose and wage pressure is subdued. The vacancies-to-unemployment ratio is one of the most reliable real-time measures of tightness. For the framework, see JOLTS Explained and Labour Market Slack.
Bargaining power is the transmission mechanism. Tight labour markets give workers the ability to demand higher wages, switch jobs for better pay, and resist real wage cuts. Loose labour markets reduce this power. The relationship is not just about the level of unemployment — it is about the balance of power between workers and employers, which depends on vacancies, turnover, and the threat of unemployment.
The Expectations-Augmented Phillips Curve
The original Phillips Curve implied a stable trade-off between unemployment and inflation. The expectations-augmented Phillips Curve, developed by Milton Friedman and Edmund Phelps, modified this: the trade-off exists in the short run, but only when inflation expectations are stable. When expectations shift, the entire curve shifts. If workers expect higher inflation, they demand higher wages to compensate, which raises actual inflation at any given unemployment rate. The curve shifts upward.
This is why inflation expectations matter more than actual inflation for the Phillips Curve relationship. If expectations are well-anchored, the curve is stable and the trade-off is predictable. If expectations become unanchored, the curve shifts and the trade-off disappears. Central banks spend enormous effort anchoring expectations because a stable Phillips Curve gives them more policy flexibility. For the framework, see Inflation Expectations Explained.
Short-Run vs Long-Run Phillips Curve
The short-run Phillips Curve slopes downward: lower unemployment is associated with higher inflation, and higher unemployment with lower inflation. This is the trade-off that policymakers can exploit in the short run. The long-run Phillips Curve, however, is vertical at the natural rate of unemployment (NAIRU). There is no permanent trade-off: attempting to push unemployment below the natural rate generates accelerating inflation, not a permanently lower unemployment rate.
The implication is that monetary policy can reduce unemployment below the natural rate temporarily, but only by generating inflation that will eventually require a recession to correct. The long-run vertical curve is why central banks target inflation rather than unemployment — they cannot permanently reduce unemployment through monetary policy, but they can prevent the inflation that would eventually require costly disinflation.
NAIRU and the Natural Rate
The Non-Accelerating Inflation Rate of Unemployment (NAIRU) is the unemployment rate at which inflation is stable — neither accelerating nor decelerating. Below NAIRU, inflation tends to accelerate; above NAIRU, inflation tends to decelerate. NAIRU is the anchor of the long-run Phillips Curve. For the framework, see Labour Market Slack.
NAIRU is not observable and is estimated with significant uncertainty. It changes over time with demographics, productivity, labour-market institutions, and globalisation. A central bank that believes NAIRU is 5% will interpret a 4% unemployment rate as inflationary; one that believes NAIRU is 4% will interpret the same rate as neutral. The uncertainty around NAIRU is one reason the Phillips Curve relationship is difficult to use in practice — policymakers do not know exactly where the threshold is.
Why the Phillips Curve Flattened
The Phillips Curve relationship has flattened in recent decades — the inverse correlation between unemployment and inflation has weakened. Several factors explain this. Globalisation has expanded the effective labour supply, reducing wage pressure even when domestic unemployment is low. Technology has automated routine work and increased productivity, reducing the wage-cost pass-through. The decline of unionisation has weakened collective bargaining power. Inflation expectations have become better anchored, reducing the expectations-shift channel. For the framework, see How Macro Regimes Change Forex Relationships.
The flattening means that a given change in unemployment produces a smaller change in inflation than it once did. This has important implications for monetary policy: central banks can run the economy "hotter" (lower unemployment) without generating as much inflation as the traditional curve would predict. But it also means that when inflation does accelerate, reducing it may require a larger increase in unemployment than the traditional curve would suggest — the cost of disinflation has risen.
Why the Relationship Can Break Down
The Phillips Curve relationship can break down entirely during supply shocks. A supply shock (oil price spike, supply-chain disruption, pandemic) raises inflation while reducing output and employment — a stagflationary combination that the traditional Phillips Curve cannot explain. The curve shifts upward: the same unemployment rate is associated with higher inflation because the supply shock has raised costs independently of demand. For the framework, see CPI Explained.
The relationship also breaks down when inflation expectations are shifting. If expectations are rising, inflation can accelerate even with high unemployment — the curve shifts upward. If expectations are falling, inflation can decelerate even with low unemployment — the curve shifts downward. The expectations channel is why central banks monitor breakeven inflation rates and survey-based expectations so closely.
Services Inflation and the Wage Channel
The Phillips Curve relationship is strongest in the services sector, where labour is the dominant cost and pass-through to prices is more direct. Services inflation is less volatile than goods inflation and tends to be more persistent — it is the "sticky" component of overall inflation. This is why central banks pay particular attention to services inflation when assessing labour-market tightness. For the framework, see Goods Inflation vs Services Inflation.
The wage-to-services-inflation channel is the core of the modern Phillips Curve. When wages rise, service-sector firms pass through the higher costs to prices. The pass-through is more reliable in services than in goods because services are less tradable and face less international competition. This is why a tight labour market can generate persistent services inflation even if goods inflation is subdued.
Productivity and the Wage-Price Link
Productivity modulates the wage-to-inflation link. If wages rise but productivity also rises, unit labour costs are stable and there is no inflation pressure. If wages rise faster than productivity, unit labour costs increase and firms must raise prices to maintain margins. The Phillips Curve relationship is strongest when productivity growth is low — because wage growth translates directly into unit labour cost growth and price inflation. For the framework, see Economic Growth Differentials.
This is why the productivity backdrop matters for the inflation interpretation of labour-market data. A tight labour market with strong productivity may not generate inflation, because unit labour costs are contained. The same tightness with weak productivity is more inflationary, because wage growth is not offset by output growth. The Phillips Curve is not just about unemployment — it is about the interaction of unemployment, wages, and productivity.
Central-Bank Reaction Functions
Central banks use the Phillips Curve framework to assess inflation pressure, but they do not treat it as a mechanical formula. The reaction function is: labour-market tightness → wage pressure → services inflation → overall inflation → policy response. But each link is assessed in context — with productivity, expectations, and supply shocks all modulating the relationship. For the framework, see Central Bank Reaction Functions and Hawkish vs Dovish.
When the Phillips Curve is flat, central banks can tolerate lower unemployment without tightening aggressively. When the curve is steep (or when it has shifted upward due to a supply shock), the same unemployment rate requires a more forceful response. The central bank's assessment of the curve's slope and position determines how it reacts to labour-market data.
FX Implications
The Phillips Curve matters for FX through the central-bank transmission. If labour-market tightness is generating wage pressure and inflation, the central bank may need to tighten, pushing yields higher and strengthening the currency. If the Phillips Curve is flat and tightness is not generating inflation, the central bank can hold, and the currency impact is muted. For the framework, see Interest Rates and Forex.
The transmission is: labour-market tightness → Phillips Curve assessment → inflation outlook → central-bank expectations → yields → currency. But each link is conditional. If the curve is flat, tightness does not translate into inflation. If expectations are well-anchored, the curve does not shift. If productivity is strong, wage pressure does not translate into unit labour cost growth. The Phillips Curve is a framework for assessing these links, not a mechanical formula for predicting them.
Why Falling Unemployment Is Not Automatically Bullish for a Currency
Falling unemployment is not automatically currency-positive. If the Phillips Curve is flat, falling unemployment does not generate inflation, so the central bank does not need to tighten, and the yield support for the currency is absent. If falling unemployment is driven by participation changes rather than hiring, the wage-pressure implication is muted. If the central bank is already at the peak of its cycle, further labour-market strength may not translate into more tightening.
The relationship also breaks when falling unemployment implies overheating that requires aggressive tightening — which can slow growth and eventually weaken the currency. The net FX impact depends on whether the labour-market strength translates into inflation (via the Phillips Curve), whether the central bank responds, and whether the yield support offsets the growth-drag from tighter policy.
Relative Labour-Market Pressure Between Economies
For FX, the Phillips Curve matters most when it changes the relative labour-market pressure between two economies. If the US labour market is tighter than the eurozone labour market, and the US Phillips Curve is steeper, US inflation pressure is greater and the Fed may need to tighten more than the ECB — supporting USD. If both economies have similar labour-market tightness and similar curve slopes, the FX impact is muted. The pair-specific impact depends on the relative labour-market trajectory and the relative curve characteristics.
Regime Dependency
The Phillips Curve's relevance depends on the macro regime. When inflation is the dominant central-bank concern, the curve is the key framework for assessing labour-market-driven inflation pressure. When growth or financial stability is the dominant concern, the curve may be secondary — the central bank may tolerate labour-market tightness without tightening. When a supply shock has shifted the curve, the relationship is particularly difficult to interpret because the same unemployment rate is associated with different inflation. For the framework, see How Macro Regimes Change Forex Relationships.
Common Analytical Mistakes
- Treating the Phillips Curve as a fixed law: It is an empirical regularity that shifts with expectations, supply shocks, and structural change.
- Assuming low unemployment always generates inflation: The curve has flattened. Productivity, globalisation, and anchored expectations can offset tightness.
- Forgetting inflation expectations: The expectations-augmented curve shifts when expectations change. Anchored expectations are the key to stability.
- Ignoring the short-run vs long-run distinction: There is no permanent trade-off. The long-run curve is vertical at NAIRU.
- Overlooking productivity: Wage growth is inflationary only if it exceeds productivity growth. Unit labour costs are what matter.
- Assuming falling unemployment is always currency-positive: If the curve is flat, tightness does not generate inflation, and the yield support may be absent.
Practical Framework for Traders
- Assess labour-market tightness: Is the market tight (low unemployment, high vacancies) or loose? Check the vacancies-to-unemployment ratio.
- Monitor wage growth: Is wage growth accelerating? Is it exceeding productivity growth (raising unit labour costs)?
- Check inflation expectations: Are expectations anchored? Check breakeven rates and survey-based measures.
- Watch for supply shocks: Has a supply shock shifted the curve? Is inflation rising independently of demand?
- Assess the services-inflation channel: Is services inflation accelerating? This is where the wage-to-price link is strongest.
- Consider the central-bank context: Is the central bank focused on inflation? Does it believe the curve is flat or steep? Is it already at the peak of its cycle?
- Check the relative dimension: Does the labour-market pressure differential between two economies change the expected policy divergence?