1. Analysing Behavioural Responses to Policy Change in Dynamic Decision Environments

2. www.melbourneinstitute.com Outline  Alternative Approaches to Policy Analysis  SIDD: A Life-cycle Model of the Household –Outline of the model –Solution methods –Using the model  On-Going Work  Directions for the Future

3. www.melbourneinstitute.com Analyses of Policy  A continuum of behavioural assumptions very broad behavioural assumptions Back-of-an- envelope analysis detailed statistical analysis no formal model of behaviour detailed statistical analysis formal model of behaviour – uncertainty omitted or poor accounted for detailed numerical analysis formal model of behaviour – uncertainty explicitly accounted for

4. www.melbourneinstitute.com SIDD: a Household Life-cycle Model  Simulator of Individual Dynamic Decisions  The unit of analysis is the household  Households fully described by the following characteristics (state variables): ageeducation status number of adultsnumber and age of children safe liquid assetswage rates pension arrangementstime of death

5. www.melbourneinstitute.com SIDD: Outline  Households make decisions regarding the following (control) variables:  Uncertainty influences the following: consumptionlabour supply pension scheme participationpension contribution rates education participation number of adultsnumber and age of children wage ratespension arrangements time of death

6. www.melbourneinstitute.com SIDD: Outline  Objective function:  where: –intertemporal budget constraint (the entire lifetime) –potential labour income (the working lifetime)

7. www.melbourneinstitute.com SIDD: Solving the Decision Problem hThT wTwT h T-1 w T-1 h T-2 w T-2 h1h1 w1w1

8. www.melbourneinstitute.com SIDD: Solving the Decision Problem  Final period, T: –Death at the end of the period is certain, and the household is assumed to be retired Hence the household consumes all remaining wealth plus any pension income, c T = w T + y T –The values of final period consumption, utility, and marginal utility, are calculated and stored. yTyT wTwT yTyT wTwT yTyT wTwT cTcT VTVT du/dc T

9. www.melbourneinstitute.com SIDD: Solving the Decision Problem  In period T-1: –Two alternative approaches to solution: 1.for any internal solution, the Euler condition must be satisfied: 2.at the solution, expected lifetime utility must be maximised:

10. www.melbourneinstitute.com SIDD: Solving the Decision Problem  Since we know how alternative values of c will affect the state variables in the next period, we can also obtain E t [du/dc t+1 ] from the previously calculated grid at age T. h T-1 w T-1 hThT wTwT du/dc T w T-1,h T-1 c T-1 This is a deterministic solution

11. www.melbourneinstitute.com SIDD: Solving the Decision Problem hThT wTwT h T-1 w T-1 h T-2 w T-2 h1h1 w1w1 time

12. www.melbourneinstitute.com Advantages and Disadvantages  Advantages –Behavioural responses are based on a structural model and therefore free of the Lucas Critique –Fully takes into account endogeneity between alternative decisions (eg fertility and labour) –Explicitly takes into account influence of uncertainty  Disadvantages –Numerically demanding (made less of problem with advances in computing) –Behavioural responses depend upon the validity of assumed model (advantages of bounded rationality)

13. www.melbourneinstitute.com SIDD: Using the Model  The model has been designed so that its parameters can be easily adjusted. These parameters include: –time preference for consumption allowance for myopia –bequest motives –preferences for leisure –attitudes to risk –the tax structure –the pensions structure –the incidence of childcare costs

14. www.melbourneinstitute.com On-going Analyses  Quasi-hyperbolic discounting and retirement behaviour  Training and later-life learning in the UK  The impact of Government plans for pension reform in the UK  Analysis of the influence of decision costs on responses to tax incentives to save  Extending the model to project population cross-sections forward through time

15. www.melbourneinstitute.com Directions for the Future  Specific topics of short-term interest: –Introduce the model to Treasury in Australia –Training and education in Australia –Fertility and female labour supply decisions in Australia  Medium term objectives: –Build in additional behavioural rigidities –Start building in macro-linkages  Others?