RDP 8611: The Effectiveness of Fiscal Policy in an Economy with Anticipatory Wage Contracts 1. Introduction

One of the key tenets of Keynesian fundamentalism is that fiscal pump-priming can stimulate demand-determined output. Resistance to the faith has focussed theoretically on the assumption of non-clearing goods and labour markets, and practically on the fear, hardened by bitter experience, that the gains have always seemed shortlived and that inflation has followed early in the wake.

The simplistic belief in non-clearing markets has been seriously disturbed by the emergence of New Classical macroeconomics but, while concessions may be made about goods markets, there are few who might budge on the issue of non-clearing labour markets. From an empirical viewpoint, Artus (1984) has established that the slowdown of productivity and employment in all major non-American[1] OECD economies in the late 1970s and early 1980s has been due to supply shocks which have not been accommodated by warranted falls in real wages. This malaise would not appear to recommend fiscal pump-priming – indeed quite the opposite. A widespread preference has been for the pursuit of disinflation, hoping that the harsh stick of unemployment will beat real wages down. According to this view, tight demand management and a period of unemployment is vital to weaken inflationary expectations and unrealistic behaviour in labour markets. Labour legislation, particularly in the United Kingdom, has been used to increase wage responsiveness to unemployment.

This paper takes up these issues to establish whether there is a case for fiscal expansion in an economy suffering from real wage stickiness and having clearing/non-clearing goods markets. Real wage stickiness is achieved by assuming perfect indexation of wages to prices. Throughout, output is assumed to be supply-determined and so there is no scope for conventional Keynesian demand management. However fiscal policy can have an influence on current inflation and on core or steady-state inflation.[2] Outside steady state, these two are not necessarily the same. All of the real effectiveness results in this paper come from this wedge. Fiscal policy derives its supply-side leverage, outside steady state, from its ability to influence this “inflation gap”. In this paper the gap only emerges when the goods market always clears and expectations of price inflation are formed rationally. With price inertia, no gap will be created.

The inflation gap is a critical element in the wage adjustment equation. In conventional models with ‘expectations-augmented Phillips curves’, only expected current inflation appears. But it is now well-established that these models have very unsatisfactory properties when perfect foresight or rational expectations is assumed. In particular the stable solutions to these models are typically only backward-looking in time, thus totally ignoring relevant future information which may be foreseen. Mussa (1981) discusses the problem in detail, and develops a microeconomic foundation of price adjustment[3] that rationalises the presence of ‘equilibrium’ rather than current, inflation in the wage adjustment equation. The essence of the analysis is that individual agents suffer a fixed cost for each price adjustment, and yet suffer integral costs that depend upon the deviation of their own price from the ‘equilibrium’ price. Search theory models also justify the use of ‘equilibrium’ inflation in the wage adjustment equation, since the costly gathering of information about the ‘equilibrium’ price is the foundation of that model.

The difficulty is to agree on the meaning of the ‘equilibrium’ price. Mussa (1981) gives a succinct definition that suggests the difficulty: “The ‘equilibrium price’ reflects both the play of market forces and the relative strength and skill of bargainers in any price negotiation”. If only market forces were considered, the equilibrium price would be the rationally expected aggregate price over the life of the contract. The models of Taylor (1979) Phelps (1978) and Calvo (1983) solve this problem giving wage behaviour that is both forward and backward looking, and delivering policy effectiveness. If we permit bargaining forces, an additional source of policy effectiveness may be adduced. For example, political pressures may force wage setters to take account of core or underlying indicators of the rate of inflation (or in an open economy, competitiveness). Anticipatory wage contracts will then optimally apply a discount for core inflation and the ‘equilibrium’ price would be the core price which in this paper grows at the same rate as nominal money per capita. The analysis is particularly relevant for economies in which the trade union movement has a “social contract” with the ruling political party. Such a phenomenon exists or has recently existed in a number of OECD countries, e.g. United Kingdom, Australia, Sweden, Norway.

If there is no government debt, fiscal policy can only have real effects if an output gap exists and prices are flexible. Fiscal pump-priming can speed up the process of adjustment of real wages and output, provided core inflation is below a critical rate defined by the maximum sustainable deficit. On the fiscal effectiveness side of the knife-edge, stagflation is a natural outcome.

The introduction of partial government debt financing allows for fiscal effectiveness, even if the economy is initially in steady state. A debt gap is created which will be matched by an inflation gap provided the goods market clears. Again, this is dependent on core inflation being below a critical rate. However, the issue of debt finance creates problems of future debt service, and stability of this process may require endogeneity of fiscal policy (see Christ (1979) Sheen (1987)). Over-cautious governments will make sure that the fiscal deficit adjusts down by more than the implied debt service. This is an important factor in the behaviour of output and inflation after a shock to the exogenous fiscal component. Below the critical inflation rate, an over-cautious government will create net output gains after a fiscal expansion.

If goods prices are sticky, no inflation gap can exist and there are no forces to reduce Classical Unemployment associated with excessive real wages. With wages fully indexed, and inflation unchanging, there are no incentives through the “social contract” to lower real wages; Classical Unemployment persists. This is consistent with the insider/outsider view of labour markets whereby wage setters do not concern themselves with the disenfranchised unemployed.

In Section 2, the underlying behavioural equations and identities of the paper are discussed. Sections 3 and 4 deal with a model of price flexibility and sticky real wages. In Section 3, fiscal deficits are financed only by money while Section 4 introduces debt. Section 5 discusses the implications of sticky prices and Section 6 presents some conclusions.


Sachs (1983) presents evidence of a real wage gap in the United States and Japan, but cannot attribute much of this to their unemployment. However, he also provides support for the hypothesis that real wage gap is the main problem for unemployment in the European economies. [1]

Buiter and Miller (1981) used core inflation in their Phillips curve. [2]

See also, Sheshinski and Weiss (1977). [3]