a7 1982); namely, the development of the GEMPACK computer modelling environment. In Section 6 some significant applied GE developments with less of an ORANI focus are mentioned. Section 7 contains details of graduate student involvement: whilst viewing their professional formation as part of the "tool-kit" strains the analogy somewhat, this investment in human capital is of importance comparable to the methodological developments surveyed.2. Survey of review papers 2.1 Relationship of applied general equilibrium to macro models By the late 'seventies policy economists had become used to the idea that there was more than one sort of model available. Applied macro modellers and members of the newer school of applied general equilibrium economics behaved as if the other party simply did not exist. At the beginning of the decade under review I explored the scope for reconciliation between the two schools (Powell, 1981), and saw some hope of rapprochement between them. This was evident in the success of Cooper and McLaren (1980, 1982; see also Cooper, 1983, andPowell, 1985) in interfacing a small continuous-time macro model (RBII) with a large applied general equilibrium model (ORANI).Parsell, Powell and Wilcoxen (1989,1991) noted that the developments in theoretical open-economy macro models of the 1980s had been taken up by some applied workers (as in the Murphy Model by Chris Murphy and in the MSG2 Model by Warwick McKibbin and Jeffrey Sachs), and that this had considerably narrowed the gap between applied GE and macm-modelsconsequently, the chances for a synthesis of the two schools had never looked better. This was because the new applied macro modellers were insisting that their models converged over time to neo-classically interpretable steady states or balanced growth paths: this implied, among other things, that agents (including governments) were restrained by intertemporal budget constraints.Cooper and McLaren's work was recently revived using a modern macro model with rational expectations in financial markets (Breece. McLaren, Murphy and Powell, 1991). Although some conflicts in model design remained