John Maynard Keynes Joan Robinson and the prospect

John Maynard Keynes, Joan Robinson and the prospect theory approach to money wage determination Ian M. Mc. Donald University of Melbourne HETSA 2017

Introduction • Paper develops/interprets the ideas of Keynes (1936) and Robinson (1937) on the causes of nominal wage rigidity using prospect theory invented 80 years later by Kahneman and Tversky (1979) • Central idea: Loss aversion explains a strong aversion to wage cuts • Draws on Bhaskar (1990) • Focuses on trade union wage setting

These ideas of Keynes and Robinson from 80 years ago are “new” • Contemporary economics is deficient because it took the wrong road – ‘The road not taken’, Bruni and Sugden – 1968=the disaster year for macroeconomics • Thus for macroeconomics, the ideas of Keynes and Robinson in the 1930 s are innovative in 2017

The flex-fix wage and price sequence • Flex-fix sequence=money wages respond to a change in aggregate demand but then settle at a new level, even if unemployment is high • Observed by Keynes and Robinson • And by Phillips (the loops), speed limits (eg Gordon), and some estimates of the range of equilibria (eg Lye and Mc. Donald (2006)) • A dynamic version=the rate of change settles – Sustained by the influence of the expected rate of inflation

Keynes observed the flex-fix sequence • at high rates of unemployment, prices, and thus wages, “in response to an initiating cause of disturbance, seem to be able to find a level at which they can remain, for the time being, moderately stable”, Keynes (1936, p. 250)

Robinson observed the flex-fix sequence • “(A)n increase in effective demand. . . will be favourable. . . to a rise in money wages” , Robinson (1937, p. 3) • “movements in the level of employment are the chief influence determining movements in the level of money wages”, Robinson (1937, p. 7) • “a community in which money wages fall without limit so long as unemployment exists is very unlike the real world”, Robinson (1937, p. 122).

Phillips’ curve and data from the 1930 s for Australia, UK and US: note flex-fix

Explanation • Standard model does not explain flex-fix – Simple monopoly union model • Adding loss aversion relative to a reference point does – Keynes version – Robinson version

The simple monopoly union model • Union sets the wage, employer sets employment • The union’s wage demand is on the labour demand curve at the point which maximises its objective • Wage determined by a marginal condition – equality between the elasticity of utility gain to the union and the elasticity of the labour demand curve

The wage demand of the SMU Wage wage demand reservation wage LD 1 UIC 1 (union indifference curve) employment

A natural rate theory • Under reasonable assumptions – Marginal condition implies a fixed wage mark-up – Macro equilibrium implies a unique equilibrium rate of unemployment

The natural rate of unemployment inflation unat unemployment

Keynes and relative money wages • “(A)ny individual or group of individuals, who consent to a reduction of money wages relatively to others, will suffer a relative reduction in real wages, which is sufficient justification for them to resist”, Keynes (1936, p. 14)

Keynes and loss aversion • “Every trade union will put up some resistance to a cut in money wages, however small”, Keynes (1936, p. 15, my emphasis) • So the (lost) marginal utility from a wage cut is huge compared with the marginal utility from a wage increase • Bhaskar (1990) expressed this using prospect theory

Within limits, the reference wage determines the desired wage Wage Reference wage LD 1 UIC 1 (union indifference curve) employment

The macroeconomic implications • Loss aversion implies: – two limiting values of the wage mark-up – two limiting values of the equilibrium rate of unemployment – at rates of unemployment between the limiting values, wages are determined by reference wages • “there will be a neutral range within which wages are constant”, Robinson (1937, p. 7)

The range of equilibria inflation constant umin umax unemployment

But what determines inflation? • Keynes’ reference wage is the money wage of other workers – Workers are passive • So within the range any wage is a possible equilibrium if all workers expect that wage • In practice, the reference wage probably is strongly influenced by history – Suggests that within the range inflation will continue at its historic rate, changing only when unemployment is outside the range – This is fix, not flex-fix=reality – Enter Joan Robinson

Joan Robinson’s proactive postulate • unions “demand a rise, and resist a cut [in money wages], whenever they feel strong enough to do so”, Robinson (1937, p. 4) • This suggested to me: – “Resist a cut” implies loss aversion – “demand a rise” implies an upward adjustment of the reference wage • An active approach, rather than the passive approach of Keynes and Bhaskar

The power of the reference wage • Robinson emphasised “the strategic and moral position of Trade Unions”, Robinson (1937, p. 7) • A common reference wage across union members will bind the members together • “Not a penny off the pay, not a minute on the day” • “a line in the sand” • The reference wage has some force of fairness, • emphasised in prospect theory • If TU members think of a wage demand as the reference wage, then they will have more resolve to achieve that wage demand because that would avoid suffering loss aversion if the wage demand fails

Following an increase in demand, where should the union set the reference wage? Wage LD 2 LD 1 WREF 1 UIC 2 UIC 1 L 1 employment

The ex-post optimal constraint • The wage demand if achieved would be regarded as an optimal outcome • A wage demand that violates this constraint may not be credible • Consistent with the rational expectations approach to setting reference levels of Koszegi and Rabin (2006) and Heidhues and Koszegi (2008)

The ex post optimal constraint will put upper and lower limits on the wage demand Wage k-WRES LD 2 LD 1 WREF 1 k+WRES UIC 2 UIC 1 L 1 employment

Robinson’s idea • “to secure the best terms they can for their members”, Robinson (1937, p. 5) • But who are the members? – employed members? – 51% of members?

The solidarity constraint • The union will not countenance a decrease nor value an increase in employment when labour demand has increased

The full-upward adjustment of the reference wage WREF 2 LD 1 WREF 1 UIC 2 UIC 1 L 1 employment

Uncertainty • Where is the labour demand curve? • Heuristic – follow what has been happening – Some argue the productivity slowdown increased unemployment

Robinson’s proactive postulate explains why: 1. Wages can be sensitive to increases in demand at high rates of unemployment 2. With weak union power, wages may lag behind the growth in labour productivity if unemployment exceeds umin - does this explain the weak performance of wages in recent years?

Sep-15 Sep-13 Sep-11 Sep-09 Sep-07 Sep-05 Sep-03 Sep-01 Sep-99 Sep-97 Sep-95 Sep-93 Sep-91 Sep-89 Sep-87 Sep-85 Sep-83 Sep-81 Sep-79 Sep-77 Sep-75 Sep-73 Sep-71 Sep-69 Sep-67 Sep-65 Sep-63 Sep-61 Sep-59 The wage share in factor income, Australia, 1959: 3 to 2017: 1 65 63 61 59 57 55 53 51 49 47 45

Weaknesses of Bhaskar-Robinson model • How is union power determined? • Surely WREF is influenced by the wages of significant others

What if prospect theory had been invented 50 years earlier? • Keynes and Robinson would have adopted prospect theory(? ) • Their observations and ideas on wage determination would have been more influential had they used prospect theory – Phelps’ crime against macroeconomics

Downward flex-fix was very prominent in the 1930 s

But Robinson did not have a theory of downward flex

Downward flex-fix can be explained by extreme duress