Macroeconomic models play a crucial role in policymaking

Macroeconomic models play a crucial role in policymaking. They combine macroeconomic theories with data to estimate the system-wide impact of changes in the economy. Although models can be very insightful, they are only as strong as their underlying assumptions. Computable general equilibrium (CGE) are one class of widely used macroeconomic models which are extremely flexible and able to accommodate different assumptions about agents' behaviours. Recent developments in macroeconomic modelling have used insights from behavioural economics to explain anomalies in economic models. Specifically, the theories of time inconsistency relevant in saving and investment behaviours and reference dependence relevant to labour market and consumption behaviour have been very impactful. These are crucial to the results of CGE simulations. Unfortunately, CGE models still rely on neoclassical assumptions. This project will seek to rethink some crucial assumptions presented below and will adapt the University of Strathclyde's CGE model AMOS to improve the economic toolset available to evaluate and respond to macroeconomic shocks and to tailor optimal policies. This model is particularly relevant as it is used by the Scottish government (SG) and shares key characteristics with models used by the HMRC, Northern Irish government (NIG) and the EU Commission.

The first contribution to this PhD will be to incorporate ideas of present-biased preferences. 'Present-biased preferences capture the idea that when considering decisions that have consequences in the future, the present is particularly salient and has a large impact on the decision at hand'. These preferences capture that behaviour is time inconsistent (decision makers make future plans but over time these change), whereas current modelling approaches assume time consistent preferences.

The second contribution will be the adoption of reference dependence in consumption decisions. This refers to the fact that 'outcomes are evaluated using a value function defined over departures from a reference point'. This will allow agent behaviour to be defined differently depending on price and income changes.

The third contribution will be the incorporation of reference dependence in labour markets. Similar to consumption decisions, individuals' employment decisions and motivation to work rely on whether their wage lies above or below a reference point.
The central aim is to incorporate well-known behavioural ideas developed to better capture how people actually behave into CGE modelling approaches, to improve the accuracy of these models and to test the sensitivity of results to different assumptions. In addition, it will offer guidance to governments using CGE models.
If the accuracy of the results of CGE simulations are greatly improved by this project, then policymakers will have an improved toolset for their policy making. This would have positive societal implications. These implications would not be limited to CGE models in Scotland but could also be used elsewhere given the flexibility and relative ease in changing CGE models.
This project would also benefit behavioural economists as it would provide quantifiable forecasts and simulations to their theories and economic modelers as the changes in CGE assumptions made could be transferable to other classes of models.