Keynes’ contributions to monetary theory were combined with the work of his predecessors to make a working synthesis. For a given supply of money, there is an equilibrium path of the price level. The equilibrium price in the current period, given next period’s price level, is just high enough to reduce the real value of this period’s cash balances down to the quantity demanded. This is figured at the corresponding rate of interest and output level. If people expect that the general level of prices and nominal wages is higher, and we assume the actual price level equals this expected level, the result will be disappointment.
Keynes noticed that the wage rate would not take the jumps sometimes necessary to maintain equilibrium and so the need for a more general theory with the wage off the equilibrium track was needed. It was decided that the wage would move gradually move towards the new equilibrium path but the mismatch would lead to a bulge of unemployment. Likewise if the new path was higher, there would be a drop off in unemployment. They thought this could be curbed with inflation but they learned that over time higher doses would be needed to overcome expectations.
From this came an article by A.W. Phillips in the academic journal Economica in 1957. Phillip’s used ordinary and quantitative terms which were much more accessible to all. The rate at which the nominal wage level is changing is a decreasing function of the level of the unemployment rate. Further, the rate of unemployment required to hold down the rate of wage inflation to the level of normal experience- the average, and accustomed, rate- is certainly positive, perhaps 2 to 3 percent in the UK. He also discovered that among years with the same level of unemployment rate there tended to be a higher rate of wage inflation when the annual unemployment rate was falling than when the unemployment rate was rising. It seemed to say that if there was an aggregate excess supply then nominal wages would be found falling and employment would be depressed. This would be as long as wages remained too high to eliminate the excess supply and both the effects of the excess supply would be larger the greater was the size of excess supply. That is to say that with the sudden appearance of an excess supply that is maintained at a given level for a while would first generate a positive rate of change of unemployment alongside falling wages and only later, a higher level of unemployment rate without a positive rate of change. This part of the Phillip’s curve seemed intuitive to most.
Some problems did lay hidden among these theories. Among them was the question of why nominal wages did not jump down to their new equilibrium level and the problem of determining the pace at which wages fell. More problems came about when they had to explain how the Phillip’s curve was sloping and its rightward position. Money wage rates tended to be rising over a range of positive unemployment rates, including rates exceeding the lower bound obtainable by high-pressure aggregate demand levels. Yet more trouble came along when they had to make sense out of the older Continental doctrine. Below normal unemployment constantly fuelled by a permissive monetary-fiscal policy will soon cause wages and prices to hyper-inflate until collapse or structural change takes place.
In the 1960s, a revelation took place: The higher unemployment level comes about because “expected wages” in the economy as a whole exceed “actual wages,” and as information comes in that actual wages elsewhere are lower than expected the ensuing downward revision of expectations will induce workers to accept still lower actual wages. The news of the initial fall of wages is enough for workers to expect that the general wage level will now fall to exactly the job-preserving level, so that the unemployment rate will return to the equilibrium level; otherwise workers are implied to be repeatedly misforecasting the wage level, contrary to rational expectations. This also shows high unemployment preceding a large wage fall.
Another problem came with explaining the coexistence of rising nominal wages with above-minimum unemployment. If the unemployment rate were driven to zero, quitting would be presumably rampant and so the firm would pay a wage premium. Wages do not rise only when firms want to be more competitive than others. Wages may also rise because the firms believe they must raise their wages just to avoid losing competitiveness. Therefore, nominal wages may not be rising due to labor market disequilibrium, but rather the prospect of productivity growth. Maintaining a steady unemployment rate below that constant equilibrium rate would entail rising inflation without bound.
This was another good article when it comes to getting an overview of a topic. It did a good job of describing how the Phillip’s curve works and some of the problems it faces. I think it was kind of dense at parts and I did not understand all of it, but overall it was pretty good.
“Phillips Curve,” New Palgrave Encyclopedia of Economics, pp. 858-860