The Sydney Olympics, seven years on: an ex-post dynamic CGE assessment

A recent development in ex-ante analysis of mega events is the use of computable general equilibrium (CGE) models. CGE models improve greatly on the input-output model, which they have largely displaced, since they incorporate fixed factors and substitution effects. However, like input-output, the method is still subject to the risk of over-optimistic estimation of benefits. We see three sources of such risk: (i) failure to treat public inputs as costs; (ii) elastic factor supply assumptions; and (iii) overestimation of foreign demand shocks via inclusion of "induced tourism" expenditure. In this paper, we undertake an ex-post analysis of the Olympics that addresses each of these risks. We handle the first two directly: public services used to support the Games (such as security services) are treated as Games-specific inputs, and we model the national labour market in full employment. For the third risk, we undertake an historical simulation to uncover the extent, if any, of induced tourism. We find no evidence of an induced tourism effect, and so exclude it from our analysis. With these assumptions, we find the Sydney Olympics generated a net consumption loss of approximately $2.1 billion.