The Phillips Curve

The Phillips Curve is a relationship between unemployment and inflation discovered by Professor A.W.Phillips. The relationship was based on observations he made of unemployment and changes in wage levels from 1861 to 1957. He found that there appeared to be a trade-off between unemployment and inflation, so that any attempt by governments to reduce unemployment was likely to lead to increased inflation. This relationship was seen by Keynesians as a justification of their policies. However, in the 1970s the curve appeared to break down as the economy suffered from unemployment and inflation rising together (stagflation).

The curve sloped down from left to right and seemed to offer policy makers with a simple choice - you have to accept inflation or unemployment. You can't lower both. Or, of course, accept a level of inflation and unemployment that seemed to be acceptable!

The Phillips Curve

The existence of rising inflation and rising unemployment caused the government many problems and economists struggled to explain the situation. One of the most convincing explanations came from Milton Friedman - a monetarist economist. He developed a variation on the original Phillips Curve called the expectations-augmented Phillips Curve.

Expectations-Augmented Phillips Curve

The Phillips Curve showed a trade-off between unemployment and inflation. However, the problem that emerged with it in the 1970s was its total inability to explain unemployment and inflation going up together - stagflation. According to the Phillips curve they weren't supposed to do that, but throughout the 1970s they did. Friedman then put his mind to whether this could be adapted to show why stagflation was occurring, and the explanation he came up with was to include the role of expectations in the Phillips Curve - hence the name 'expectations-augmented'. Once again the supreme logic of economics comes to the fore!

Friedman argued that there were a series of different Phillips curves for each level of expected inflation. If people expected inflation to occur then they would anticipate and expect a correspondingly higher wage rise. Friedman was therefore assuming no 'money illusion' - people would anticipate inflation and account for it. We therefore got the situation shown below:

Expectations-Augmented Phillips Curve


Say the economy starts at point U with expected inflation at 0%, and the government decide that they want to lower the level of unemployment because it is too high. They therefore decide to boost demand by 5%. The attempt to reduce unemployment would primarily be through boosting aggregate demand (AD) in some way. In the short run, the increase in AD would lead to a rise in national income and subsequently we might expect unemployment to fall. This is represented by a movement along the Philips curve to point V.

However, the adjustment period would also mean that there would be shortages in the economy which would 'pull' prices up. The increase in these prices leads people to seek wage demands that give them a 'real' increase, i.e. is above inflation. Since inflation has risen people could reasonably be expected to build an anticipated inflation rate into their wage demands. If these wage demands were granted (and in the days of powerful trade unions and without the emphasis on global competition as there is now this was very likely), the result would be increased costs for businesses. The increased costs caused the AS curve to shift to the left and the economy would be at point W.

Firms would push up prices to maintain their profit margins or shed labour in response to the additional demand and higher costs and the net result would be that the economy would be back to the unemployment level it started with (U) but with a higher level of inflation (5%). (Note how we have used the framework of AS/AD to explain the Phillips Curve - this shows how closely related the two things are!) The increase in demand for goods and services will fairly soon begin to lead to inflation, and so any increase in employment will quickly be wiped out as people realise that there hasn't been a real increase in demand.

So having moved along the Phillips curve from U to V, the firms now begin to lay people off once again and unemployment moves back to W. Next time around the firms and consumers are ready for this, and anticipate the inflation.

If the government insist on trying again to reduce the unemployment the economy will do the same thing (W to X to Y), but this time at a higher level of inflation. In future wage negotiations they may not push just for a 5% increase in wages but 5% + ! They might think that given that inflation had risen by 5% last year it might rise by 8% next and so put in for a wage rise of 11% to ensure they get a real pay increase plus cover themselves for any anticipated inflation.

Any attempt to reduce inflation below the level U will simply be inflationary. For this reason the rate U is often known as the Natural Rate of Unemployment.



If inflation rises then exports become uncompetitive and so they fall. Exports are injections so AD falls which means unemployment rises.

If there is inflation then demand for Imports will rise and so leakages increase and so...unemployment rises.

If unemployment is low then exchange rates are high which makes imports competitive, exports not unemployment rises.

So, really, inflation is high and unemployment is high; inflation is low and unemployment is low.

The exact opposite.

Or is this wrong...?


  1. It depends on size of inflation and how long it lasts...