The Swedish industrial support program of the 1970s revisited: Micro to macro analysis

In the wake of the oil crises of the 1970s a large chunk of Swedish manufacturing representing some 35 % of total Swedish exports went bankrupt. In response, the Swedish government implemented a massive industrial support program, notably of the shipyards. A Government Committee on Industrial Subsidies was set up, and two of the authors of the present paper were asked to analyze the macroeconomic impact of the subsidies. Simulations on the Micro to Macro (MM) model were used to quantify the effects. Several alternative policies were investigated. The study concluded that postponing the full impact of the crises on employment involved a significant cost to society in the form of lost output during the 15 years of the simulation through a severe misallocation of resources during the eight years of the subsidy program.

A long time has passed since the subsidies were terminated, and we now know the actual outcome. It is clear that the steep unemployment increase experienced in the rest of the OECD area from 1975 onward was postponed in Sweden for almost fifteen years. The unemployment level rose rapidly to the OECD level after the subsidy program was abandoned and the entire shipyard industry had been shut down. Swedish manufacturing growth lagged significantly behind the OECD development until at least 1995.   So from 1975 until 1995 Swedish manufacturing development was affected by two significant external shocks: the oil crises and the massive industrial subsidy program.

The question that arises is whether the underperformance of the Swedish economy relative to the rest of the OECD between the late 1970s and 1995 was due to the subsidy program which kept wages artificially high in the subsidized firms, and high quality labor away from other expanding firms, or whether there is an alternative explanation.  It has been argued that the stagnation in Swedish manufacturing compared to OECD development at large reflects the “transaction cost” of a normal adjustment to the new post oil crises global competitive situation. This adjustment cost was possibly larger than in other OECD countries because of the greater exposure of Swedish manufacturing to the competitive change caused by the oil price shocks.

The purpose of the present paper is twofold, one methodological and one theoretical. The first is to compare the simulated results of the earlier study with the actual outcome. The other is to examine which of the two theories best explains the facts.